PATHOLOGY OF THE NERVOUS SYSTEM PATHOLOGY OF THE NERVOUS SYSTEM E. FARQUHAR ^UZZARD M.A., M.D., F.R.C.P. PHYSICIAN TO ST. THOMAS'S HOSPITAL AND PHYSICIAN TO OUTPATIENTS AT THE NATIONAL HOSPITAL FOR THE PARALYSED AND EPILEPTIC AND J. GODWIN GREENFIELD B.Sc, M.D., M.R.C.P. i^VVTHOLOGIST TO THE NATIONAL HOSPITAL FOR THE PARALYSED AND EPILEPTIC PAUL B. HOEBER 67 & 69 EAST 59TH STREET NEW YORK 1922 » c • e . • . r .^ • e^ PRINTED IN GREAT BRITAIN BY ILLING AND SONS, LTD., GUILDFORD AND ESHER PREFACE In offering this volume to medical students and practitioners the authors believe that they are meeting a real need. They have attempted to describe clearly the anatomical changes which are associated with disorders of nervous function, to discuss briefly questions of pathogenesis, and to indicate in a few words, where it is possible, the relationship between structural alterations and clinical signs and symptoms. It may be true that further advances in our knowledge of nervous diseases must depend on the study of function and disorders of function more than on the consideration of advanced structural defects, but scientific speculation in regard to the former must always be tempered by acquaintance with the latter. No student of neurology or of psychiatry can be fully equipped for his work unless he has spent time and energy in a neuro-pathological laboratory, and the object of this book will be achieved if it serves to economise his time and to guide his energy. The authors are conscious of the debt they owe to the Board of Management and Medical Staff of the National Hospital for the Paralysed and Epileptic for the opportunities of working in the laboratory of that institution. 480661 CONTENTS CHAPTER I GENERAL PATHOLOGY PAGE Introduction ... . . . i The Neuron ... . . .2 {a) The Nerve Cell . . . " . .4 (6) The Nerve Fibre . . . . .14 The Neuroglia . . . . . .18 Paths of Infection in the Central Nervous System . 24 The Cerebro-spinal Fluid . . . • 30 CHAPTER II DEVELOPMENTAL AND FAMILIAL DISEASES 1. Developmental Diseases .... 2. Birth Injuries ..... 3. Familial and Congenital Diseases . . (i.) Cerebral and Myelopathic . . . [a] Amaurotic family idiocy (6) Werdnig-Hoffraann paralysis . (c) Friedreich's ataxia {d) Progressive lenticular degeneration {e) Huntington's chorea (ii.) Neural . . (/) Peroneal atrophy [g) Progressive hypertrophic interstitial neuritis children . . . . . . (iii.) Myopathic . . . . . [h) Family periodic paralysis . . [i) Myotonia .... (/) Myotonia atrophica . . . =, ' {k) Myopathy * E (/) Amyotonia congenita . . . ol 50 58 58 58 58 62 63 66 68 70 70 71 73 73 74 75 75 78 VIU CONTENTS CHAPTER III INJURIES TO THE NERVOUS SYSTEM 1. Injuries to the Brain and its Coverings 2. Injuries to the Spinal Cord — Caisson Disease . 3. Injuries to Nerves .... 81 87 94 CHAPTER IV CIRCULATORY DISTURBANCES OF THE BRAIN AND SPINAL CORD IscHAEMic Softening of the Brain Cerebral Haemorrhage Aneurysms of the Cerebral Arteries False Porencephaly Meningeal Haemorrhage . (a) Subarachnoid haemorrhage . {b) Subdural haemorrhage. rhagica interna (c) Extradural haemorrhage 6. Sinus Thrombosis . 7. Haematomyelia 8. Haematorrhachis . Pachymeningitis haemor 105 117 120 121 121 122 122 123 124 125 127 CHAPTER V SYPHILIS OF THE NERVOUS SYSTEM GUMMATA ....... Gummatous Meningitis. Pachymeningitis Cervicalis Hypertrophica Gummatous Arteritis Syphilitic Myelitis Tabes Dorsalis General Paralysis of the Insane 133 135 138 141 145 154 CHAPTER VI OTHER INFECTIVE DISEASES 1. Leprosy ....... 160 2. Streptothrix Infection . . . . .163 3. Tuberculosis ...... 163 {a) Tuberculosis of the cranium, vertebrae, and dura mater ...... 164 {b) Tubercular leptomeningitis .... 167 (c) Tuberculomata of the brain and spinal cord , .169 CONTENTS IX lO. II. 12. 13- 14. 15- Acute Leptomeningitis .... (a) Meningococcal meningitis (b) Meningitis due to other pyogenic micro-organisms (c) Serous meningitis and meningism Pyogenic Pachymeningitis Suppurative Encephalitis and Myelitis (Abscess Brain and Spinal Cord) Acute Myelitis Tetanus OF Acute Poliomyelitis and Lethargic Encephalitis Landry's Paralysis Herpes Zoster Chorea Trypanosomiasis . Hydrophobia or Rabies Polio-encephalitis PAGE 176 177 179 184 188 193 202 209 212 214 216 217 CHAPTER VII EFFECTS OF POISONS Neuritis ...... Toxic Myelitis ..... Encephalopathy (Effects of Poisons on the Brain) Ergotism . . . . Pellagra ...... Lathyrism ...... Beri-beri . 222 226 227 229 230 232 234 CHAPTER VIII TUMOURS OF THE BRAIN AND SPINAL CORD Tumours of the Brain and its Envelopes Glioma — Ependymal glioma Neuroblastoma . Neuro-fibroma Carcinoma Angeioma Sarcoma — Perivascular sarcoma . Cylindroma Endothelioma Psammoma Cholesteatoma Pituitary tumours Pineal tumours . Dermoid, parasitic and other cjrsts General pathology of intracranial timiours 239 240 246 247 249 249 250 252 253 256 256 256 258 258 260 CONTENTS B. Spinal Tumours .... 1. Tumours of the Envelopes of the Cord (a) Vertebral (6) Intra vertebral . 2. Intramedullary Tumours . 3. The Pathology of Spinal Compression 264 264 264 266 267 269 CHAPTER IX DISEASES OF OBSCURE ORIGIN 1. Motor Neuron Disease . . 2. Subacute Combined Degeneration of the Spinal Cord 3. Disseminated Sclerosis .... 4. Syringomyelia 5. Paralysis Agitans . . , . . 6. Myasthenia Gravis. 271 277 283 290 298 299 Staining Methods APPENDIX I 305 APPENDIX II Methods of Examination of the Cerebro-Spinal Fluid Index . . . . . 322 327 LIST OF ILLUSTRATIONS FIG. 1. (a) Normal cells from hypoglossal nucleus . . -5 {b) Normal cell from ventral horn of lumbar enlargement of spinal cord ... . . .5 (c) Pigmentation of ventral horn cells . . • 5 2. Ventral horn cells showing changes resulting from destruction of ventral roots (" reaction a distance ") . . 7 3. Ventral horn cells from a patient dying from alcoholic neuritis showing much vacuolation . . .8 4. A single nerve fibre undergoing Wallerian degeneration stained by the Marchi method . . . .16 5. Enlarged neuroglia cells in the lateral columns of the cord underlying a tumour . . . . .21 6. Neuroglial thickening in an old focus of softening in the cere- bellum. (From preparation stained by Professor Carl Weigert, through the kindness of Dr. Gordon Holmes) . . . . . .22 7. Section from spinal cord showing many corpora amylacea . 23 8. Normal choroid plexus from fourth ventricle . . . 30 9. Retarded development of brain from a diplegic infant eighteen months old . . . . '53 10. Cerebral asymmetry of congenital origin . . .54 11. Brain from case of diplegia showing shrunken convolutions and patches of cortical degeneration . . • 55 12. Hydromyelus of cervical cord . . . -57 13. {a) Cells from dorsal horn and Clarke's column in a case of amaurotic family idiocy . . . .60 (6) Pur kin je cells from the same case . . .60 14. Three sections from a case of Friedreich's ataxy stained by the Weigert -Pal method . . . .65 15. Section of muscle from a case of peroneal atrophy . .71 16. Muscle in the pseudo -hypertrophic form of myopathy . 77 17. Sections stained by the Marchi method illustrating the ascend- ing degeneration in the spinal cord following a fracture- dislocation in the mid-thoracic region . . .92 18. A peripheral nerve undergoing degeneration as the result of pressure, stained by the Marchi method . '95 xii LIST OF ILLUSTRATIONS FIG PAGE 19. {a) A low-power photograph of an excised portion of nerve which had undergone changes of a fibrotic character following the passage of a bullet through the tissues in its immediate vicinity. Photograph of sections taken {b) just above and (c) below the site of injury stained by the Bielschowsky method . . . .96 20. Sections of ulnar nerve seven months after a wound that severed it. Stained by the Weigert-Pal method . 98 2 1 . Sections of an ulnar nerve five months after section by wound, stained by Bielschowsky's method . . .99 22. Cerebral softening, the result of embolism of the left middle cerebral artery, in a case of aortic valvular disease ...... 106 23. Thrombosis of the left posterior cerebral artery . .107 24. Softening in the region of the lenticular nucleus . . 108 25. Pontine thrombosis due to disease of the basilar artery or its branches . . . . . . 108 26. Cerebral thrombosis . . . . . .109 27. Photographs of a brain illustrating the appearances pro- duced by vascular lesions of varying severity . .111 28. Softened cortex resulting from arterial thrombosis, showing granular corpuscles and necrotic pyramidal cells. (Stained by haematoxylin and van Gieson) . .112 29. Fat-laden granular corpuscles in an area of cerebral soften- ing, the result of embolism of a cerebral artery (Haematoxylin and van Gieson) . . .112 30. Softened brain tissue with fat-laden granular corpuscles stained by the Marchi method . . • 113 31. Degeneration of the pyramidal tract due to softening of the internal capsule; stained by the Marchi method: {a) decussation of pyramid ; (6) thoracic cord . • 115 32. Four sections from a case of pontine thrombosis illustrating the secondary degeneration in the pyramidal tract: {a) decussation of pyramid; [b] cervical enlargement; (c) thoracic region; {d) lumbo-sacral enlargement (Weigert-Pal) . . . . . .116 33. Haemorrhage from the anterior cerebral artery ploughing up the frontal lobe . . . . .117 34. Three photographs from a case of cerebral haemorrhage with extravasation of blood into the ventricles and minor haemorrhages in the pons . . . .119 35. Section of optic nerve in secondary syphilis, showing enormous infiltration of Virchow-Robin space with mononuclear cells, and glial overgrowth in the optic nerve . . . . . . . 131 36. Drawing of a section of the lumbar enlargement, showing an intramedullary gumma undergoing central caseation . 132 37. Two sections through the brain of a patient with extensive gummatous meningitis and meningo-encephalitis . 134 38. A section from sclerosed area of brain underlying the gum- matous meningitis shown in fig. 37, and showing proliferation and enlargement of neuroglial cells . 135 LIST OF ILLUSTRATIONS xiii FIG PAGE 39. {a) Gummatous arteritis of the right middle cerebral artery with organisation of the central clot . . -139 (b) The area of softening in the right hemisphere resulting from the arterial thrombosis . . . .139 40. {a) Gummatous arteritis of the anterior spinal artery in a case of syphilitic meningo-myelitis .... 140 (6) Section of cord from the same case . . .140 41. Two sections illustrating the degeneration in the dorsal columns in tabes dorsalis (Weigert-Pal) . • 151 42. Section from prae-Rolandic cortex in a case of general paralysis of the insane . . . .156 43. Section of first lumbar segment in a case of general paralysis of the insane, showing degeneration in lateral columns 157 44. Section from lumbar cord in a case of taboparesis . .157 45. Leprosy bacilli in a dorsal root ganglion cell . . . 162 46. Caries of lumbar vertebrae . . . . .166 47. Section of a vessel on the cerebral cortex the seat of tuber- cular arteritis . . . . . .169 48. Photograph of a section from the medulla oblongata, showing a tuberculoma lying dorsal to one olive . 170 49. Post-basic meningitis . . , . .174 50. Right temporo-sphenoidal abscess secondary to middle-ear disease ...... 181 51. Bilateral abscesses in frontal lobes secondary to sphenoidal sinusitis ...... 181 52. Cavity in left cerebellar lobe, the result of an abscess secondary to middle-ear disease . . . . .181 53. Section from the wall of a cerebral abscess, showing the formation of fibrous tissue and numerous granular corpuscles . . . . . .183 54. Acute poliomyelitis: {a) Section from high thoracic region stained by haematoxylin and van Gieson to show the cellular infiltration of the perivascular spaces and of the grey matter; {b) a higher power photograph of one side of the same section . . . . • 195 55. Acute poliomyelitis. Photograph showing the meningeal and perivascular cellular infiltration in the ventral fissure . 196 56. Acute poliomyelitis. Ventral horn cell preserving a fairly healthy appearance and surrounded by intense small- celled infiltration . . . . .197 57. Acute poliomyelitis. Two ventral horn cells undergoing destruction in the midst of serous and cellular exudation 198 58. Acute poliomyelitis. Changes in the cells of Clarke's column and surrounding cellular infiltration . . -199 59. Acute poliomyelitis. Section of spinal cord showing softened area with granular corpuscles, perivascular infiltration and haemorrhage ..... 200 60. Lethargic encephalitis. Cortex .... 204 61. Lethargic encephalitis. Cortex .... 204 62. Lethargic encephalitis. View of longitudinal section of a small blood vessel in the cortex . • . . 205 xiv LIST OF ILLUSTRATIONS PAGE 63. Lethargic encephalitis. Cortex . , . . 206 64. Lethargic encephalitis. View of cortex and meninges . 206 65. Lethargic encephalitis. Blood vessel in medulla . . 208 66. Lethargic encephalitis. Medium-sized artery on the cortex partially obstructed with haemorrhage in its walls, and leading to small wedge-shaped area of infarction . 208 67. Landry's paralysis . . . . . .211 68. The cord in a case of lathyrism stained by the Weigert-Pal method ...... 233 69. Glioma in parieto-occipital region seen on the mesial surface of the left hemisphere . . . . .241 70. Glioma of corpus callosum with numerous haemorrhages . 241 71. Cystic glioma of left frontal lobe .... 242 72. Glioma of a rather fibrous type (microscopic section) . 243 73. Diffuse glioma of pons (hypertrophy of pons) . . 243 74. Diffuse glioma of brain-stem (sagittal section) . . 244 75. Glioma of pons invading fourth ventricle . . . 244 76. Glioma pontis (microscopic section) .... 245 77. Ganglio -neuroma. High-power view of ganglion cell area (Bielschowsky's stain) .... 247 78. Neuro-fibroma of acoustic nerve .... 248 79. Acoustic nerve tumour, showing " palisade " arrangement of nuclei (microscopic section) .... 249 80. Secondary carcinoma in the cerebral peduncles . .250 81. Perivascular sarcoma. High-power view of small vessel . 251 82. Cylindroma (microscopic section) .... 253 83. Endothelioma growing from falx cerebri . . . 254 84. Nodular endothelioma removed by operation . . . 254 85. Endothelioma of dura (microscopic section) . . . 255 86. Pituitary tumour ...... 257 87. Dermoid cyst ...... 258 88. Colloid tumour of third ventricle, possibly derived from the pituitary ...... 259 89. Hydrocephalus produced by pontine tumour . . . 261 90. A pressure-cone . . . . . . 262 91. Sarcoma of cord ...... 267 92. Neuro -fibromatosis of cauda equina .... 268 93. Amyotrophic lateral sclerosis. Section from the cervical en- largement . . . . . '273 94. Muscles in amyotrophic lateral sclerosis : [a) Early degenera- tion of muscles; {b) advanced atrophy of muscle, the muscle spindles escaping . . . . 2.j^ 95. Amyotrophic lateral sclerosis. Longitudinal section of atro- phied muscle fibre . . . . .276 96. Subacute combined degeneration of cord. Three sections representing the changes seen in the cervical, thoracic and lumbar regions stained by the Weigert-Pal method ...... 279 LIST OF ILLUSTRATIONS xv FIG. 97. Two transverse sections and one longitudinal section from a case of subacute combined degeneration, stained by the Marchi method . . . . .281 98. Disseminated sclerosis. Sections of pons and medulla . 286 99. Disseminated sclerosis. Sections of cerebellar cortex and of cord (one longitudinal and several transverse at various levels in the same cord) . . . . 287 100. Syringomyelia ...... 294 10 1. Syringobulbia ...... 295 102. Myasthenia gravis: {a) Drawing of transverse section of an ocular muscle showing a " lymphorrhage "; (b) photo- graph of transverse section of a skeletal muscle . 302 103. Diagram of Lange's colloidal gold reaction . . . 325 PATHOLOGY OF THE NERVOUS SYSTEM CHAPTER I GENERAL PATHOLOGY I. Introduction. The science of neuro-pathology includes the study not only of pathological processes limited to nervous structures — the nerve cell and its processes — but of changes in other structures with which the nervous system is intimately associated both in an anatomical and physiological sense. The nerve cells and their processes, in other words the neurons, may exhibit states of health which are at variance with the normal for a number of different reasons. In the first place, they may never reach maturity, or may lag behind the rest of the body in the process of development. In the second place, they may reach maturity with an inherent lack of endurance which prevents them from completing the full course of life. Thirdly, they may suffer from over-activity or unnatural idleness. Fourthly, they may undergo biochemical or nutritional dis- turbance owing to changes of a qualitative or quantitative character in the fluids on which they are dependent for their food. Fifthly, and finally, they may be the victims of patho- logical changes in neighbouring tissues. Each of these unhealthy states will be illustrated in some detail in the pages of this volume, but examples may be cited here in order that the principle of this classification may display at once its practical bearing. 1. Agenesis : — the undeveloped state of the higher cerebral centres in congenital idiots. 2. Abiotrophy : — the degeneration in early adult life of certain spinal tracts in Friedreich's ataxy. 2 GENERAL PATHOLOGY 3. (a) Hyperactivity : — the cortical cell changes found after death from status epilepticus. {b) Disuse : — the atrophic changes in the ventral horn cells of the spinal cord after amputation of a limb. 4. (a) Intoxication : — the cell changes in the medulla and spinal cord associated with diphtheritic paralysis or with chronic alcoholism. (b) Ischaemia : — the cell changes resulting from disturbance of the blood supply to any part of the nervous system, as in cerebral thrombosis. 5. {a) Neuroglial overgrowth : — syringomyelia. {b) Meningeal disease : — lepto- or pachy-meningitis. (c) Compression : — tumours or diseases of the cranium or vertebral column. {d) Vascular disturbance : — cerebral or spinal haemorrhage. Our knowledge concerning these pathological processes has grown hand-in-hand with the advance of nervous physiology during the last seventy years, and so closely related and inter- dependent are neuro-pathology and neuro-physiology that it is impossible to discuss the one without constant reference to the other. For this reason it is desirable to refer briefly and immediately to some of the well-established physiological facts which have intimate pathological bearings, and which have added so much to our acquaintance with the histological anatomy of the nervous system. 2. The Neuron. The whole of present-day neuro-histology and neuro-patho- logy has been built up on the basis of Waller's experiments in 1850, by means of which he demonstrated that degenerative changes take place throughout the whole of a frog's nerve below the point at which it has been divided. This experiment showed not only that degeneration commences at once and simultaneously in the whole of that part of the nerve fibre which is separated from the nerve cell, but, indirectly, that the cell is the central organ for the formation and nutrition of all its processes. This discovery was supplemented later by Forel, who observed that such secondary degeneration does not extend through the cell-station forming the link between two physio- logically connected fibres. For example, the degeneration THE NEURON 3 of pyramidal tract fibres cut across in the thoracic region of the cord can be traced into the lumbo-sacral enlargement, but does not extend through the ventral horn cells of that region into their root fibres and the peripheral nerves. The relation- ship of the cell to its chief process was still further illustrated by the experimental work of von Gudden, who first noted the cellular changes which follow injury or disease of its axis cylinder process. In these simple facts lie substantially the foundation of the neuron doctrine which holds that the nervous system consists of a large number of units, each composed of a cell and its processes, which are genetically and nutritionally indepen- dent of one another, although closely related in function and variously grouped to form physiological systems and arcs. The early popularity of the neuron doctrine twenty years ago was followed by a period of considerable unrest and doubt, during which the chief bone of contention was the primary origin of the nerve fibre and the nature of its regenera- tion after injury. Most of those who were ready to give the doctrine an indecently hasty burial because researches seemed to show that nerve fibres could be laid down by sheath cells in segmental fashion, have lived to see its prosperous revival on a firmer basis. The investigations of His, Ramon-y-Cajal, Held and Ross Harrison have all contributed to this end by proving conclusively that the axis cylinder process is an out- growth of the cell protoplasm and is not dependent on sheath cells for its origin or its existence. On the other hand, it is generally agreed that sheath cells play a considerable, if only accessory, part in bringing about the regeneration of peripheral nerve fibres, and the success of that part is emphasised by the paucity of regenerative power in the fibres of the central nervous system where sheath cells do not exist. Although there must remain some difference of opinion with regard to minor points, there is at the present time general agreement in favour of the view that the neuron is a distinct unit, especially in respect to its reaction to pathological processes. 4 GENERAL PATHOLOGY (a) The Nerve Cell. Although the ii^pregnation methods of Golgi, Ramon-y-Caj al and Bielschowsky have been responsible for much valuable information concerning the anatomy of the nervous system and the external form of the nerve cells and their processes, our knowledge of pathological changes occurring within the cells owes more to the staining method invented by Nissl. The Nissl method and its modifications depend for their value on the fact that certain constituents of nerve cells are readily stained by basic aniline dyes, while other parts remain un- coloured. Thus the coloured or chromatophile structures include the nucleolus, the nuclear membrane, and certain bodies found in the cell protoplasm. To the latter have been applied a variety of names, such as Nissl bodies, chromatin granules, tigroid bodies, etc. The value of the Nissl method lies in the fact that we are enabled by its employment to detect, in some cases with ease, any abnormalities with regard to the size and shape of the cell body, the size, shape and position of the nucleus, and the disposition and appearance of the chromatin granules. The methods of Ramon-y-Caj al and Bielschowsky reveal the neuro- fibrils of the cell body and its processes, and bring into prom- inence the fibrillar network imbedded in the homogeneous protoplasm. While some authors believe that the distribution of the Nissl bodies is determined by the fibrillar network, and that the former occupy and fill the meshes of the latter, others find that the chromophile substance is encrusted upon knots of the network at the points where one fibril joins another. In this way the Nissl bodies may be star-shaped, rod-shaped, or polyhedral in form. Other authorities dispute the presence of an intracellular network at all, and only describe neuro- fibrils passing independently through the cell protoplasms from one process into another. The disadvantage of the impregna- tion methods for pathological investigation is brought home to us by these conflicting observations. On the other hand, the Nissl method has its limitations. While a large number of the more important ganglion cells of the nervous system are eminently suited for investigation by this method, many others are so deficient in chromophile THE NERVE CELL a. Normal cells from hypoglossal nucleus. b, Normal cell from ventral horn of lumbar enlargement of spinal cord. Q, Pigmentation of ventral horn cells without -pathological significance. 6 GENERAL PATHOLOGY substance and display such variation in their size and shape and in the position of their nuclei that very special knowledge and experience are required before an accurate estimate can be formed with regard to the presence or absence of abnormal features. In spite of this drawback, the Nissl method has given us a vast amount of information, and is, at the present time, our mainstay in the study of nerve cell changes. It has been employed, in both experimental and human pathology, by many skilled observers, and results have been obtained not only in connection with the chief cells of the brain and spinal cord, but also in observations upon the peripheral ganglia and the cells of the sympathetic system. Cell changes after lesions of axis cylinders. — It has been shown that, speaking generally, lesions of axis cylinders are followed by certain changes in the cells of their origin. This phenomenon has been widely studied in many neurons and in many animals; the changes in the hypoglossal nucleus follow- ing division of the hypoglossal nerve of a rabbit may be taken as an example. The changes may be divided into two stages, the reaction stage and the reparation stage. The reaction stage begins within forty-eight hours after the nerve division and reaches its zenith at the end of fifteen to twenty days. During this period the following phenomena may be observed in the majority of the cells of the nucleus, (i) Breaking down of the chromatin granules into a fine dust which still retains the colour of the stain. This change begins in the perinuclear zone and spreads towards the periphery of the cell and into the bases of the cell processes; it has received the name of chromatolysis or chromolysis. Occasionally the dust loses its chromophilic tendency, the whole cell becomes almost colourless, and a condition of " achromatosis " is estab- lished. (2) The cell increases in volume, loses its polygonal form, and becomes rounded and swollen. (3) The nucleus is also increased in size, and becomes displaced towards the cell margin or into the base of one of its processes. As a rule the nucleus retains its shape, but when it reaches the surface of the cell it may become more oval, with its long axis parallel to the cell margin, and may even show some invagination on its inner aspect. The nucleus may project from the surface of the cell, and in some instances is completely extruded. THE NERVE CELL 7 This event is followed by atrophy and disappearance of the cell. The reparation stage is of longer duration, beginning about the twentieth day after nerve division and attaining completion about eighty days later, by which time the cell has resumed its normal appearance. During this period the chromatin granules are gradually reformed and retake their place in the Fig. 2. Ventral horn cellsjshowing changes resulting from destruction of ventral roots (" reaction d, distance "). picture. Meanwhile the cell slowly diminishes in size and the nucleus regains its central position. Characteristic of the stage of repair is the appearance produced by a somewhat enlarged cell filled with granules which are bigger and more numerous than those seen in the normal state. This is the ** pyknomorphic " condition of Nissl. The above description of cellular changes cannot be univer- sally applied to all cases of nerve division. Variations are 8 GENERAL PATHOLOGY common and occur in two directions. In the first place, spinal nerves may be divided in peripheral parts of the limbs, and a search for cell changes may give very meagre or no results. According to some authors, the amount of reaction varies directly with the proximity of the axis cylinder lesion to the cell of origin. In the second place, the character of the injury to the axis cylinder appears to determine the amount of cell Fig. 3. Ventral horn cells from a patient dying from alcoholic neuritis showing much vacuolation. change. On the one hand, a slight lesion, such as the temporary application of a ligature to a nerve, may evoke a very poor and transient reaction which rapidly passes into the stage of repara- tion, the whole process lasting a comparatively short time. On the other hand, when nerves are roughly treated, for instance when they are torn apart, the cell reaction is rapid in onset,fulminant in character, and frequently fails to be followed by reparation. Thus the forcible rupture of one hypoglossal THE NERVE CELL 9 nerve in a rabbit may result in the complete disappearance of all cells in the corresponding nucleus after the lapse of thirty-five days. As far as human pathology is concerned, cell reaction to axonal injuries or diseases appears to be fairly constant, and it is the rule to find cell changes and a more or less complete disappearance of cells in the corresponding segments of the spinal cord after limb amputation. According to some authors, there are exceptions even to this rule. The question as to the exact nature of the process underlying the cell reaction has not been finally settled, but it appears probable that chromatolysis depends upon an increase in the fluid content of the cell. The factors on which the process of repair depends have also given rise to a conflict of views. According to some observers, the reparation stage can only be entered upon when regenera- tion of the peripheral portion of the divided axon takes place. According to others it is equally certain that complete reparation of the cell can obtain, under some circumstances, in the entire absence of axonal regeneration. The circum- stances determining these results, however, have not yet been established. The peripheral sensory neurons have been investigated in the same way, and with results which present many similarities as well as some differences. The ganglion cells of the peri- pheral sensory neurons lie outside the central nervous system and are bipolar in form ; that is to say, they possess two axonal prolongations, central and peripheral, cellulifugal and cellu- lipetal. Division of the peripheral or cellulipetal axons is generally followed by a stage of reaction in the cells of origin, which show changes resembling those described in connection with the motor neurons. Perinuclear chromolysis, enlarge- ment of the cell body and lateral displacement of the nucleus, are more or less constantly observed. The onset of those changes is detected twenty-five to thirty hours after the nerve division and reaches its height at the end of seven days. In the majority of cases the reaction is followed by repair, which, again, is of shorter duration than the corresponding stage in motor neurons. In other instances repair does not take place, and a larger or smaller number of cells undergo atrophy and 10 GENERAL PATHOLOGY destruction. The factors determining the presence or absence of the reparatory stage are still unknown. Division of cellulipetal fibres, therefore, produces cell changes which only differ materially from those of peripheral motor neuronic cells in their precocity. When, however, central or cellulifugal axons are divided, for instance by section of the dorsal spinal roots, cell reaction is either entirely wanting or occurs in doubtful and meagre quantity at a very much later period. If it occurs at all, it takes the form of slight cellular atrophy associated with some interstitial proliferation in ganglia which are examined many months after the operation. This is rather in accordance with the general experience in human pathology that degeneration of the dorsal roots and sclerosis of the dorsal columns is often unaccompanied by any changes in the cells of the dorsal root ganglia. It is worth while noting in this connection that three months after division of the dorsal roots some myelin degeneration may be found in the terminal parts of the peripheral fibres. Similar but less marked degeneration may be found in the dorsal roots after section of the peripheral sensory axons. The reaction of cells which lie within the central nervous system and whose axons are distributed within the brain or spinal cord now requires our attention. There appears to be little or no doubt that division of these axons is generally followed by changes of cellular reaction similar to those de- scribed in relation to peripheral neurons, and it is equally certain that the reaction stage is never followed by any attempt at repair. It is usual to find that a large proportion of the cells of origin proceed to atrophy and final destruction. This rule holds good at any rate for the pyramidal tract and the Betz cells of the motor cortex, for the direct cerebellar tract and the cells of Clarke's column, and for Monakow's bundle and the cells of the red nucleus. According to some authors, the prox- imity of the axonal lesion to the cells of origin determines to some extent the intensity of cellular reaction. It has been found, for instance in human pathology, that the Betz cells present more intense and more rapid signs of reaction after lesions of the internal capsule than they do after similar affections of the pyramidal path in the medulla or spinal cord. The absence of cell reparation in the central neurons is of THE NERVE CELL ii interest in connection with the well-known failure of central axons to regenerate after a breach of continuity, and may be compared to the more general facility towards cellular repara- tion and axonal regeneration which is possessed by peripheral neurons. In the sympathetic system there is considerable difficulty in determining the presence of chromolysis and nuclear dis- placement in the ganglion cells on account of their scanty chromatophile substance and variously situated nuclei under normal circumstances; it has, however, been demonstrated that cell reaction does take place after division of sympathetic cords. Section of the cervical sympathetic above the superior cervical ganglion leads to a variable amount of cell enlarge- ment, chromolysis, and nuclear displacement in the latter, and resection of the hypogastric plexus has been shown to produce a typical axonal reaction in the intermedio-lateral column of cells in the lumbo-sacral region of the cord. Further details of the axonal reaction of cells are observed by means of neuro-fibril stains, such as those of Ramon-y-Cajal and Bielschowsky. The first change corresponds in time to that of the perinuclear chromolysis, and consists of some disintegration of the central part of the intracellular fibrillar network. Meanwhile, the more peripheral fibrils, namely those passing through the cell protoplasm from one process to another, remain intact. The central part of the neuro-fibril plexus which is the site of the earliest change represents the origin of the axonal process. With the displacement of the nucleus the network breaks up into fragments which are diffi- cult to detect, and, if the reaction is intense, the peripheral fibrils and those of the processes also undergo granular dis- integration. In the stage of repair it is not unusual to find cells having a striated appearance due to the absence of fibrillar network and the preservation of the long fibrils traversing the cytoplasm. Finally, a reconstruction of the plexus takes place, and the normal appearances are regained. In conclusion, emphasis may be laid on the fact that while cellular reaction to axonal lesions is very general in human and experimental pathology, there appears to be a considerable variabiHty in the amount of resistance offered to this process by individual cells. It is the rule to find that the cells of any 12 GENERAL PATHOLOGY system, although all exposed to the same insult through their axons, do not all respond in equal fashion. Some do not react at all, others react but recover, while the remainder pass through the stage of reaction only to undergo atrophy and final destruction. Cellular reaction to physiological and pathological influences. — While the axonal reaction of nerve cells described above may be regarded as a more or less specific phenomenon, the same can hardly be claimed for all forms of cellular changes associated with physiological or pathological states. It is not unusual to describe axonal reaction as a secondary cell change, and those cell changes dependent on chemical or physical alterations in the cellular environment as primary. Rest. — Resting cells present what is regarded as a normal appearance, but there is some reason to suppose that absolute cellular rest, that is to say a condition in which there is an unusually large reserve of energy, is characterised by the presence of numerous and large compact masses of Nissl chromatin (Pyknomorphic state). Activity. — Normal activity produces a diminution in the amount of chromophilic substance together with a slight increase of the volume of the cell and its nucleus {parapykno- morphic state). Exhaustion. — When activity goes on to prolonged exhaustion the chromophilic substance becomes broken up into smaller particles, and both the cell and its nucleus may tend to shrink. In extreme . degrees of exhaustion the cell protoplasm may show some vacuolisation, and it is often surrounded by parasitic cells of neuroglial or vascular origin {neuronophagy) . Intoxication. — Cellular reactions occur to various forms of poisoning, organic and inorganic, exogenous and autogenous. There is nothing characteristic of any individual poison; in other words, the cellular appearances give no certain clue as to the nature of the toxic agent in any particular case. Chromo- lysis, sometimes achromatosis, together with swelling and vacuolisation of the cell body and displacement, or even extru- sion of the nucleus, are features common to many forms of intoxication Chromolysis may begin either in the peripheral or the central zone of the cell protoplasm. An attempt to show that perinuclear chromolysis is always due to a lesion of the THE NERVE CELL 13 axis cylinder, while peripheral chromolysis is the result of primary cell affections, has not been substantiated. While it is safe to say that axonal reaction is always associated with perinuclear chromolysis, it must be allowed that the loss of chromolytic substance in the primary affections may begin either centrally or peripherally. According to some authors, the central form of chromolysis is an indication that the process is acute and virulent in type. Speaking generally, it may be stated that the effect of a large number of poisons is to produce cell changes which are not always to be distinguished from those of axonal reaction. Circulatory disorders. — Partial interference with the circula- tion may bring about cell changes similar in type to those described as due to the influence of various poisons. On the other hand, sudden and complete cutting off of the blood supply from any part of the central nervous system leads to a rapid cell change which is characterised by the fact that the cell body is reduced in size, and that the whole of the protoplasm and nucleus is stained more or less homogeneously by the basic dye. This has been well illustrated by the results of experimental ligature of the carotid and vertebral arteries or of the abdominal aorta in animals, and by the results of local thrombotic lesions in human brains. There is evidence to show that the chromo- lysis, together with some swelling of the cell and displacement of the nucleus, may be recovered from if and when the cir- culation is re-established, but the state of cell shrinkage asso- ciated with uniform deep staining indicates a biochemical change in the protoplasm from which recovery is impossible. Hyperpyrexia. — It has been ascertained experimentally that the exposure of nerve cells to a temperature of 109° F. (43° C.) rapidly brings about a change which is characterised by the loss of differentiation between the chromatic and achromatic constituents of the cytoplasm. The cell volume is somewhat increased, and the whole protoplasm, as well as the- nucleus, becomes more or less uniformly stained by the aniline dye. A somewhat vitreous appearance is given to the cell, and the contour of the nucleus tends to lose its definition. In some cases there is evidence to show that this chemical change, amounting to a coagulative necrosis, begins in the processes and peripheral parts of the cell and spreads towards the centre. 14 GENERAL PATHOLOGY A prolonged exposure to a temperature between 107° and 108° F. has a similar effect. Identical changes have been found in the brain and spinal cord of persons who have displayed a high temperature immediately before death or who have suffered for some time from considerable fever. The common association of high fever in human beings with infective processes makes it difficult to discriminate between the results of pyrexia and those of circulating toxins, but there is little doubt that the condition of coagulative necrosis, especially when all the nerve cells are more or less uniformly affected, is characteristic of hyperpyrexia. It is probable, from experimental results, that a very slight degree of this coagulative change in the cell periphery is not incompatible with life and even with recovery of cell function. It is interesting to note that the cell changes of hyperpyrexia have been found also in animals which have died as the result of exposure to powerful sun-rays, as this may throw light on some cases of sunstroke or insolation in human beings. Pigmentary changes in the nerve cell. — Nerve cells may normally contain two varieties of pigment — melanin and lipo- chromes. (i) Melanin is normally present in some of the cells of the basal ganglia and brain-stem, especially in certain nuclei, such as the substantia nigra and the substantia ferru- ginea, where the amount of melanin pigment in the nerve cells gives a black tinge to the cut surface of the brain-stem. Apart from these localities, melanin pigment does not occur in nerve cells even in pathological conditions. (2) Lipochromes are found in all varieties of nerve cells in adult and advanced life, usually as a small mass of yellowish pigment lying between the nucleus and the origin of the axon. The pigment stains fairly intensely with Marchi's method and with Scharlach R. In many degenerative conditions it increases in amount, and may almost entirely fill the nerve cell, displacing the Nissl granules. It is never seen in a part of the nerve cell where Nissl granules are also present. (b) The Nerve Fibre. When a peripheral nerve is divided certain changes take place almost immediately both in the proximal and distal portions. THE NERVE FIBRE 15 Changes in the proximal portion. — The myelin sheath undergoes fatty degeneration as far as the nearest node of Ranvier or a segment or two higher, and proUferation of the cells of the neurolemma sheath does not extend above this portion of the fibre. Within a few hours of the nerve division the terminal portions of the axis cylinders are seen to swell. The interfibrillar substance increases in such a way that the fibrils of the axon become separated and, as it were, combed out into loose strands. The latter do not remain tidy, but tend to interlace and to curl up into complicated and plexiform structures, to which various descriptive names have been applied, such as button, ring, plaquette, spiral, etc. From these tangled masses soon arise a number of young and delicate axis cylinders which grow forward in all directions to penetrate the cicatrix uniting the cut ends of the nerve, and tend generally to follow the lines of least resistance. In the course of their forward growth many divide, and some develop bulbous ex- tremities. These bulbs are cone-shaped or olive-shaped, and may be compared to the pseudopodial cone of growth which is observed in the developing axon of the embryo. According to other authors, the bulb is the entangled end of an axon which has met with a temporary obstacle in the way of its progress. It should be pointed out that not all the newly formed fibres pass straight through the cicatricial tissue. Many stray in lateral and reverse directions, forming a plexus of much intricacy. The proportion of fibres passing through the cica- trix into the distal end of the nerve probably depends largely upon the amount of connective tissue which intervenes and the^ degree of resistance which it offers to penetration. This explains the more rapid regeneration of a peripheral nerve after early surgical union of the cut ends, as compared to the slower and less complete process which obtains when the cut ends are not brought into contact and remain separated by scar tissue. Along with this outgrowth from the axon, there is an outgrowth of the sheath cells in the form of finger-like processes. These enclose the neuro-fibrils in all their ramifica- tions, and some eventually unite with similar processes arising from the sheath cells of the peripheral cut end of the nerve. Changes in the distal portion, — In that part of the nerve which is separated from its nutritional centre changes take place i6 GENERAL PATHOLOGY rapidly throughout its length, affecting the axis cylinders, the medullary sheath, and the sheath of Schwann. The axis cylinder swells and becomes more obviously fibrillated. In the second place, it becomes varicose or irregular in contour. Vacuolation and granular disintegra- tion (axolysis) follow, and finally the debris is removed by wandering cells. The myelin exhibits a tendency to break up into droplets which stain deeply in specimens treated by the Marchi method. These droplets undergo further division and are finally absorbed or removed by phagocytic cells. There is some reason to sup- FlG. 4. A single nerve fibrejunder going Wallerian degeneration' stained by the Marchi method. pose that with the rupture of continuity between the nerve cell and the distal portion of its axon digestive ferments are set free within the neurolemma of the latter, and that these fer- ments excite the breaking up of the myelin and axis cylinder. The nuclei of the sheath of Schwann multiply by direct and mitotic division with such rapidity that in three or four days' time there are many cells, richly provided with protoplasm, scattered about between the myelin droplets. They penetrate the disintegrated interior of the fibre in all directions, and some of them, at any rate, assume the role of scavengers. Others are represented in the final stage of degeneration as spindle-shaped THE NERVE FIBRE 17 structures occupying the site from which the axis cyUnders have disappeared, and so maintaining a path for the newly formed axis cyHnders to use when they have traversed the intervening scar tissue. There is also an outgrowth from the sheath cells nearest the cut end of the nerve towards the proxi- mal portion, and if it does not reach this, it forms an " end bulb " similar to that formed on the proximal cut end of the nerve in every respect, except that it contains no nerve fibres. In this way the process of degeneration passes imperceptibly into one which may be regarded as preparatory to regeneration. Owing to the activity of the sheath cells from the moment when degeneration begins, dead material is rapidly removed and all necessary measures are taken for promoting the antici- pated regeneration. The door, as it were, is left open for the reconstruction of a nervous pathway. The next step in the process of regeneration is an interaction between the axons and the sheath cells, which results in the formation of a new myelin covering for the former. Finally, the sheath cells, having completed the numerous and important duties allotted to them, retire again into comparative obscurity by reforming a normal neurolemma and losing the greater part of their protoplasmic structure. The reconstruction of the complete nerve fibre takes place earlier at the central end of the distal portion than it does in the periphery, and the whole process of degeneration and regeneration is one which occupies some months even under the most favourable circumstances. If, instead of a simple division, a length of nerve is resected, regeneration is either much retarded or rendered impossible. Even in these conditions some regeneration may be promoted by the artificial interposition of a graft between the separated ends of the nerve. The graft directs the growth of the re- generating fibres towards the peripheral end, and, as far as is known, has no vital influence on regeneration. So far we have described the phenomena attending a complete rupture or resection of a nerve. But other pathological processes are common which tend to injure or compress nerve fibres without severing their continuity. In proportion to the severity of the lesion changes are produced in the sheaths and axis cylinders of the fibres, but it does not follow that loss of conductivity from the clinical point of view is necessarily i8 GENERAL PATHOLOGY associated with destruction of the axis cyUnders. On the contrary, there may be local changes in the myelin and neuro- lemma sheaths and even some structural alteration of the axis cylinders, with loss of function in the latter, without degenera- tion of the fibre beyond the site of the lesion. In this way cases of rapid return of function, following modification of the pathological process producing paralysis or anaesthesia in the distribution of a nerve, are easily explained. Some cases of Bell's palsy, for instance, recover rapidly in a few days even though the initial lesion has been severe enough to cause complete facial paralysis. In other instances, nine to twelve months elapse before the process of regeneration permits of the return of function. 3. The Neuroglia. The neuroglia has a common origin with the nerve cells and their processes from the cells of the neural groove. It appears, however, normally to act purely as a supporting structure. In the early months of extra-uterine existence it may serve to some extent as an insulator for nerve fibres in the unmyelin- ated tracts of the cord. The ependyma lining the ventricles, the iter of Sylvius and the central canal of the cord, as well as the cells covering the choroid plexus, have a similar origin to the neuroglia, and may be considered as specialised glial structures. Of their changes in pathological conditions we know very little, but they resemble the neuroglia in a tendency to degenerate and disappear in various morbid conditions. The central canal of the cord is not usually preserved in adult life, but an irregular clump of ependymal cells can always be seen lying in the middle of the grey commissure, and in destructive lesions of the cord these cells may show evidence of proliferation. The neuroglia cells as seen in the brain of a healthy young adult are of three types. (i) The commonest form which is found everywhere in the brain and cord is composed of a small cell body and long, un- branched, filamentous processes. It has a small, round, eccen- trically placed nucleus which almost fills the cell body, and in which the chromatin is disposed chiefly as fine dots on the inner surface of the nuclear membrane. (2) Another form, seen in the grey matter, has relatively THE NEUROGLIA 19 short branching processes, one of which may be attached to the wall of a blood vessel. (3) The third type, seen typically in the cerebellar cortex, has three or four thick processes which branch into leashes of fine parallel fibres covered with small swellings or spines. These run direct to the surface of the cerebellum and attach themselves to the pia mater. This type of cell is sometimes known as a " Deiter's cell." When stained by Weigert's neuroglia stain it is seen that glial fibres pass through the cell body and do not blend with it, but in immature cells the peculiar staining of the fibres becomes lost, and the fibres broaden out as they approach the cell body. Glial fibres run for considerable distances in all directions, mainly parallel to the nerve tracts, but may pass across the fibres of several different tracts; this may explain the fact to which Eurich has drawn attention that a neuroglial sclerosis secondary to degeneration of one tract may produce glial thickening among the fibres of the neighbouring tracts. The walls of the ventricles, the iter of Sylvius, and the central canal of the cord, are supported by an increase in the glial tissue, and on the surface of the brain and cord, in the subpial layer, there is also a condensation of glial fibres. Here, and around the walls of the blood vessels, the glial fibres form loose connections with the connective-tissue fibres by means of small thickenings or foot-plates attached to the outer surface of the connective tissue. When degeneration of nerve elements takes place from any cause the neuroglia reacts, and by multiplication of cells and fibres replaces the disintegrated parenchymatous tissue. It also takes on a phagocytic or " scavenging " function, and helps in the removal of debris. Glial cells also join with cells of vascular origin in the process of ** neuronophagy." This term does not mean that glial and other cells directly attack degenerated nerve cells, but rather that they absorb from them and the fluids surrounding them the products of degeneration. This process has also been termed " necrophagy," but as cells are often seen surrounding nerve cells which are obviously not dead, and which probably are able to recover, it is no more exact than the better-known term. The cells of the sheath of Schwann have a similar origin to the neuroglia cells, and are 20 GENERAL PATHOLOGY analogous to them, and it has been seen that cells derived from them may exercise similar phagocytic powers. NeurogHal overgrowth may take place without any obvious primary nerve degeneration, as in syringomyelia and in tuberous hypertrophic sclerosis. These conditions form a connecting-link between reactionary glial proliferation and tumour formation. In gliomata are found all the forms of glia cells which are usually seen in neuroglial reaction, those which represent earlier and less specialised stages being most numerous. Abnormal forms with ring- or horseshoe-shaped nuclei may also appear. The various forms of reactionary glia cell have been investi- gated by special staining methods by Alzheimer, who attri- buted different functions to the different forms. Of these we may note the following : (i) Miniature glia cells, which occur both in the grey and white matter, are small cells resembling lymphocytes, with comparatively little cytoplasm. Their nuclei, which are usually single, are similar to those of the larger form, the chro- matin being scanty and in the form of granules, chiefly around the periphery of the cell. These cells may become enlarged and loaded with degeneration products. They probably arise directly from the pre-existing glial cells by a process of enlarge- ment and division of their cell bodies. They may go on to form either type (2) or (3) . (2) Amoeboid glia cells : Large variety. — These occur in the white matter of the brain and cord at a later stage than type (i) in all forms of degeneration. They are large balloon-like cells of the average size of a ganglion nerve cell, with one or more nuclei, which are usually eccentrically placed. Their cytoplasm is clear or finely granular, and may contain various particles which stain differently with aniline dyes, some colour- ing with acid fuchsin, some with methylene blue, etc. Lipoid substances may also be present in the cytoplasm. Their inclusions seem to arise from degeneration products of myelin and nerve cells, and to be taken up in solution by the glial cells and by them transformed into simpler products which can be ingested by the mesoblastic cells. These amoeboid glia cells seem to have considerable migratory powers, and they are frequently found, loaded with granular inclusions and, it may THE NEUROGLIA 21 be, themselves undergoing degeneration, in close relation to the vessels. From this situation their work is continued by the mesoblastic cells, which appear to have more powers of fat metabolism than the glial elements (fig. 30). (3) Fibre-forming glia cells. — These cells are similar to the large amoeboid glia cells, but send out shoots of cytoplasm in various directions. These processes are considerably thicker and much more irregular than glial fibres, some being short like pseudopodia, others longer and approximating more closely to the processes of spider cells. This type of cell frequently :^ t^^■ V -■■; 4>j^ ^Yk^ Fig. 8. Normal choroid plexus from fourth ventricle. other epithelial structures. In favour of this view have been cited certain changes seen in the cells covering the plexus during active formation of the fluid, as under the action of ether, and also the increase of the amount of fluid which results from the injection of various substances into the circulation. There have been many fallacies in connection with this work, as the experiments have not always eliminated factors such as increase of arterial pressure or venous stasis, which, by increasing the intracranial contents, would force more cerebro-spinal fluid out THE CEREBRO-SPINAL FLUID 31 of the skull. The results of Dixon and Halliburton, however, appear to be sufficiently established. These authors found that extract of dried choroid plexus had a very definite action in increasing the amount of fluid produced, and that other cholesterin-containing substances, such as extracts of brain tissue, had a similar but weaker effect. A different theory of the action of the choroid plexuses was advanced by Mestrezat. He considered the cerebro-spinal fluid to be a product of dialysis from blood serum, the choroid plexuses performing the role of a colloid membrane. This theory is based on the similarity in composition of the cerebro- spinal fluid to a dialysate from blood serum, and on the readiness with which any increase from the normal in the percentage of glucose, chloride or urea in the blood shows itself in an increased percentage of these elements in the cerebro-spinal fluid. He himself noted, however, that abnormal salts, even those of small molecule, such as nitrates and iodides, cannot be made to pass through into the cerebro-spinal fluid, even though they are administered in doses large enough to make them appear in the blood in considerable amounts. His theory must there- fore be qualified to the extent of allowing the cells covering the choroid plexus a certain selective action on the constituents of the blood plasma ; and with this reservation it appears to be in accordance with the observed facts. It is certain that the choroid plexus has no power of elaborating special enzymes or protein substances such as that possessed by most secreting glands. (2) Circulation. — The cerebro-spinal fluid formed in the lateral and third ventricles passes by the iter of Sylvius to the fourth ventricle, where it mixes with the fluid formed by the small "plexus quarti ventriculi." Thence it escapes through the ependymal roof of the ventricle by the central foramen of Magendie and the two lateral foramina of Luschka, all of which are usually present. It passes directly into the cisterna magna, and thence diffuses both downwards in the subarach- noid space surrounding the cord, and upwards by the pontine and basal cisterns and the cisterns of the Sylvian fissure. Over the cortex of the brain the arachnoid membrane is fairly closely applied to the pia mater, but it bridges over the sulci, and along the channels thus formed the cerebro-spinal fluid gains access 32 GENERAL PATHOLOGY to the large venous sinuses and lacunae, which form its chief means of escape into the blood stream. In relation to these there are developed arachnoid villi, the larger of which in relation to the superior longitudinal sinus have been termed Pacchionian granulations. In addition to the superior longi- tudinal sinus and its lateral lacunae, Key and Retzius found such villi in relation to the transverse sinus, the cavernous sinus, the superior petrosal sinus and the middle meningeal veins. The conception of their structures formed by these authors is that they are composed of an inner core consisting of a meshwork of arachnoid tissue, both lined and covered by endothelium. This is separated from the wall of the sinus into which it projects by a continuation of the subdural space, so that the villus can be dislodged from its nest in the wall of the sinus by injection of fluids into the subdural space, without a hole being made in the wall of the sinus. According to this view, between the cerebro-spinal fluid and the blood in the sinus there are four thin endothelial layers. Key and Retzius considered that stomata were present between the cells, allowing for the free passage of the cerebro-spinal fluid into the sinus, but of these there is no evidence. There are several important prolongations of the subarach- noid space, such as the funnel-shaped sheaths round the cranial nerves, which are especially important in relation to the optic nerves and the auditory nerves. By means of the latter the cerebro-spinal fluid communicates freely with the perilymph of the internal ear. Scarcely less important is the Virchow- Robin space, which surrounds both arteries and veins as they enter and leave the brain, and is continued along them right up to the capillaries. Its walls appear to be derived from the pia mater, and it communicates directly with the lymph chan- nels in the pia, and through these with the subarachnoid space. It may thus be injected from the subarachnoid space, and virtually forms a prolongation of the space into the substance of the brain and cord. The arachnoid is covered with endothelium, both internally and externally, so that between the subarachnoid and subdural spaces there are two distinct layers of endothelium which normally form an efficient barrier against the passage of fluids from one to the other. It must be remembered that in the CEREBRO-SPINAL FLUID 33 normal condition there is no cerebro-spinal fluid in the subdural space, which is, in fact, only a potential space containing a very thin film of lymph. The amount of cerebro-spinal fluid which can be formed in twenty-four hours may rise to one or two litres when there is free escape. In the human subject under normal conditions there are probably from 60 to 80 c.c. of cerebro-spinal fluid, of which only a very small proportion is present in the ventricles, the larger part lying in the subarachnoid space around the cord and in the cisterns at the base of the brain. It has been calculated that it is renewed completely six or seven times a day. The pressure of the cerebro-spinal fluid in the lumbar theca in man varies normally from 60 to 120 mm. of water in the recumbent position. It may be raised much above this in pathological conditions. It seems to correspond very closely to the pressure in the large venous sinuses of the brain, especially the torcular Herophili. Both the arterial pulse and the altera- tion in venous pressure due to respiration affect the cerebro- spinal fluid, so that it undergoes a perpetual to and fro movement. This probably has a great effect in hastening its absorption and in promoting free renewal of the fluid which surrounds the spinal cord. It is a common enough observa- tion that a forcible expiratory effort such as coughing, or, in infants, crying, raises the pressure of the cerebro-spinal fluid remarkably. (3) Mode of absorption. — Absorption of the fluid into the circulation takes place by two main routes. First aind more important is the venous route, especially by means of the arachnoid villi. It is not unlikely that there is also a consider- able absorption by the adventitial sheaths of the veins, but this is probably very small as compared with the escape into the larger venous sinuses. The other route is along the peri- neural lymph spaces, and in the spinal theca this probably constitutes the chief means of escape. Lymph spaces are more than usually abundant around the olfactory nerves, and accord- ing to Mestrezat and others the olfactory mucous membrane is freely bathed with cerebro-spinal fluid. But it is probable that under normal conditions the lymph current runs as much up as down the nerves, and that what cerebro-spinal 3 34 GENERAL PATHOLOGY fluid escapes by this route is replaced, at least partially, by lymph. Various experiments have been done to find out the relative importance of these two routes. Leonard Hill found that methylene-blue injected into the theca appeared in the urine in from ten to twenty minutes, but was not visible in the lymphatics of the neck till after an hour. Even more direct are the observations of Weed that dye-stuffs introduced into the spinal theca when this was ligated at its upper end did not appear in the urine until after seventy-five minutes, and then in minute traces only, whereas, with the theca intact, they appeared in the urine in twenty minutes. In another series of experiments he injected dye-stuffs into the cerebellar cistern, and found that there was no difference in the proportion of the dye recovered from the urine in two hours, whether the spinal theca was ligated off or left intact. This experiment shows that under normal conditions of pressure the escape of cerebro- spinal fluid along the spinal nerve roots is minimal. (4) Relation of the cerebrospinal fluid with the nerve centres. — The free communication which exists between the subarachnoid space and the adventitial (Virchow-Robin) space of the cerebral and spinal vessels has already been mentioned. In all cases of meningitis and meningeal infiltration this space is found dis- tended with leucocytes or small round cells for a considerable distance into the substance of the brain, especially along the larger vessels. Mott has suggested that the relation of the cerebro-spinal fluid with the nerve cells does not end here, but that by means of the so-called perivascular space of His it gains access directly through fine lymph spaces to the space of Obersteiner surrounding the nerve cells so that these are actually bathed in cerebro-spinal fluid. There is general agreement that the original description of these spaces as lined with endothelium is wrong, and doubts have been cast on their existence under normal conditions. But whether they exist or not it is clear that some fluid, either lymph or cerebro-spinal fluid, can find its way through the meshes of the glia to the nerve cells. Weed, using ferro-cyanide injections into the subarachnoid space, found that some of the nerve cells were ringed round with granules of Prussian blue. It seems therefore likely that, under some conditions, either by increase of the CEREBRO-SPINAL FLUID 35 cerebro-spinal fluid pressure, or by diminution of the capillary pressure (as in Mott's experiments, where the carotid artery was tied), the cerebro-spinal fluid can penetrate pretty freely into the nervous tissue. There are, however, several arguments against its doing so, at any rate, to any considerable degree, under normal conditions. In the first place, the Virchow- Robin space seems in normal histological preparations to be empty; and Bruce's work on degenerative conditions of the nerve elements, in which this space becomes filled with scavenger cells, indicates that the current is chiefly from within outwards — that is, from the smaller to the larger vessels. Teale and Embleton have found that tetanus antitoxin introduced into the subarachnoid space, after tetanic symptoms had been produced by the injection of tetanus toxin, has no effect in reducing the spasms. It appears therefore that foreign proteids, at any rate, cannot pass from the subarachnoid space to the nerve cells. They have also shown that horse serum does not pass out into the brain tissue from the capillaries, although it passes out into the tissues of the liver, spleen and omentum. This may be due to the walls of the brain capillaries having a power of selective filtration similar to that exercised by the choroid plexus. Various mechanisms may thus be at work to prevent the sensitive nerve cells from attack by proteid and other poisons. (5) Function. — The chief functions of the cerebro-spinal fluid are: (i.) To form a water cushion protecting the brain and cord from any sudden shock. The spinal cord is so suspended in this fluid by its ligaments that in health it probably never comes in contact with its bony envelope. (ii.) As a reservoir to regulate the contents of the cranial cavity. When the brain substance increases or vascular congestion is present the cerebro-spinal fluid ebbs away. When the brain shrinks, more cerebro-spinal fluid is formed to fill the vacant space. This is well seen in general paralysis of the insane. (iii.) It serves to wash away the products of tissue metabo- lism, and it may provide pabulum for the nourishment of the nerve cells. (6) Normal composition. — The cerebro-spinal fluid is a clear 36 GENERAL PATHOLOGY watery fluid of low specific gravity (about 1007-5), and con- taining only a very few lymphocytes in suspension (less than 5 per c.mm.). It contains a small quantity of albumen, which normally does not exceed 0*03 per cent. This is made up of both serum albumen and serum globulin, with a preponderance of globulin. Glucose is present to the extent of about o 05 per cent., and is fairly constant at this level. The chlorides, which are mainly sodium and potassium salts, form a remarkably constant ingredient. Their proportion of 073 per cent, is extraordinarily steady in health, and variations below 070 per cent., or above 076 per cent., may be regarded as indications of disease. Urea is normally present, but varies considerably in quantity. The normal percentage is given as from 0*003 per cent, to o 01 per cent. In addition, the fluid contains traces of bicarbonates, phosphates, sulphates, calcium, and magnesium, and minute traces of nitrates. Its alkalinity is similar to that of blood serum both as regards H-ion concentra- tion and alkaline reserve. It is normally alkaline to litmus, and acid to phenolphthalein (Levinson) . The following table is taken from Mestrezat, who compiled it from an analysis of fluids removed from healthy patients prior to stovaine anaesthesia: Normal Mean. Normal Variations. Specific gravity . . 1007-6 1007-3-1008 Freezing-point . . -0-576° c. -o-57°-o-59°C. Alkalinity of the ash (in Na2C03 percentage) 0-125 0-120-0-137 Albumens (percentage) . . . . 0-018 0-013-0-030 Glucose . . 0-053 0-048-0-058 Chlorides . . 0.732 0-725-0-740 Urea .. o-oo6 0-003-0-010 (7) Pathological alterations in the cerebrospinal fluid. — Although the most striking changes met with in the cerebro- spinal fluid are due to disease of the cerebro-spinal nervous system or its coverings, it must not be forgotten that certain general diseases may affect the composition of the fluid. These are the more important as in them the fluid is altered at its source, and that obtained by lumbar puncture is almost exactly similar to the ventricular fluid. CEREBRO-SPINAL FLUID 37 The chief general disorders which produce definite effects on the cerebro-spinal fluid are those where the normal propor- tion of the constituents of the blood plasma are changed, as in diabetes mellitus and the various forms of nephritis. The cerebro-spinal fluid is extremely sensitive to changes in the chemical composition of the blood plasma, and a raised per- centage of sugar, chloride, or urea in the blood will at once show itself in the cerebro-spinal fluid. The ease with which the exact percentages of these substances may be estimated make examination of the cerebro-spinal fluid of such cases a valuable aid in diagnosis and prognosis. Other diffusible substances, such as alcohol, acetone and formaldehyde, pass readily into the cerebro-spinal fluid. Apart from these diseases and the cases where arterio- sclerotic changes have impaired the efliciency of the choroid plexus, it may be assumed that the fluid as secreted has a fairly constant chemical composition. Whether it actually contains albumen and cells is a matter of conjecture, but other- wise it probably differs little from that obtained by lumbar puncture. Alterations from the normal may take place in its course from the choroid plexus to the lumbar theca, and are therefore most likely to occur when either the walls of the ventricles, or the subarachnoid space surrounding the base of the brain and cord, are the seat of disease. (a) Appearance. — Normal cerebro-spinal fluid is perfectly clear and colourless, like water. This appearance may be preserved in spite of considerable pathological increase both in the number of lymphocytes and the albumen content. Where a very considerable lymphocytosis is present the fluid may present to the naked eye a very slight turbidity or opales- cence, which can only be seen on examining the fluid in a good light against a black background. It is sometimes impossible to distinguish this turbidity from that produced by a very slight blood admixture, but the difference is at once ap- parent on centrifugalisation, as in the latter case the deposit is definitely red. An apparently clear fluid which on standing yields a fine web of coagulum is characteristic of tubercular meningitis. A similar appearance may be given in post-basic meningitis, in poliomyelitis and in some acute cases of syphilitic meningitis. 38 GENERAL PATHOLOGY In these conditions the fluid may show a very faint yellow tint, only seen on looking down the long axis of the tube; this is most frequently seen in tubercular meningitis. In more acute forms of meningitis the turbidity is much more obvious, and the fluid may become completely opaque with pus. Such fluids also coagulate into a finer or heavier coagulum after being withdrawn, and often show a yellowish or greenish dis- coloration. Blood admixture may be due to damage to a vessel at the time of the lumbar puncture, or to intraventricular or sub- arachnoid haemorrhage. In the former case the amount of blood admixture is not constant, and if the fluid is received into a series of tubes some will be more and others less blood-stained. Centrifugalisation in this case will yield a colourless fluid unless the blood is present in large amounts. When the blood was present in the subarachnoid space before lumbar puncture the fluid is evenly tinged, and if the blood has been present for more than one or two days there will be a yellow coloration of the fluid due to disintegration of the red blood corpuscles and alteration of the haemoglobin. For the first two or three days after such a haemorrhage many red blood corpuscles are still present in a yellowish fluid. A few days later the fluid becomes clear, but of a darker yellow colour, which gradually disappears in the course of one or two weeks. An alteration in the appearance of the fluid is also frequently present in the loculation syndrome first described by Froin (p. 43) . The colour may vary from a pale straw to a deep greenish tan. It may also coagulate either spontaneously or after the addition of a drop of fresh blood. In the more pronounced cases the fluid may coagulate en masse so that on inversion of the tube no fluid escapes, and such fluids by coagulating in the needle may completely arrest the flow of cerebro-spinal fluid. Frequently, however, only a fine web-like coagulum results. The actual colour is usually identical with that resulting from haemorrhage into the subarachnoid space, and it is only by a further examination of the fluid that its cause can be identified. [h) Cells. — So long as all the factors influencing the passage of cells into the cerebro-spinal fluid are not definitely known, it is impossible to dogmatise as to the number of cells normally present, especially as the majority of cerebro-spinal fluids CEREBRO-SPINAL FLUID 39 examined come from patients with some disease either of a degenerative or inflammatory nature in their brain or cord. The examination of a very large number of fluids by many competent observers has given rise to a general consensus of opinion that, apart from disease of the cerebro-spinal axis, the number of cells present never exceeds 5 per c.mm.* Except for the occasional presence of one of the endothelial cells lining the arachnoid, these are all of the small lymphocyte type. It is of as great diagnostic importance to find abnormal types such as large mononuclear or polymorphonuclear cells as to find an excess in the total number of cells. Such abnormal cells, even though they do not exceed 5 per c.mm., may have great diagnostic significance in cases of intracranial disease, as a few large mononuclear cells are often found in cases of tumour involving the meninges or invading the walls of the ventricles, whereas a few polymorphonuclear and phago- cytic cells may be present when a brain abscess lies in close relation to the meninges. A'^lymphocytic increase is usually present in all forms of syphilitic disease of the meninges, as also in tabes and general paralysis. Any figure up to 200 or 500 per c.mm. may be reached, but those between 10 and 50 are most common. In these conditions larger forms of mononuclear cells are con- stantly present in smaller or larger percentages, and plasma cells are not uncommon. A lymphocytic increase is also present in certain other acute or subacute diseases, notably in herpes zoster and disseminated sclerosis. In lethargic encephalitis the fluid may contain no increase of cells, or a considerable mononuclear increase may be present. Poliomyelitis and polio-encephalitis, during the acute stage, usually give rise to a considerable cell increase up to 40 or 50 per c.mm., consisting chiefly of mononuclear with a small proportion of polymorphonuclear cells. Tubercular meningitis is characterised by a very considerable rise in cells, which may be almost wholly lymphocytes, or may * It is sometimes stated in textbooks that the number of cells may normally rise to 10 or even 20 per c.mm. This is almost certainly an error and prob- ably arises either from faulty technique, e.g. estimating from the number of cells present in ^ c.mm., or from films made after centrifugalising the fluid, or from neglecting the presence of some inflammatory or degenerative disease of the cerebro-spinaFsystem. 40 PROTEINS IN show a considerable proportion, up to 50 per cent., of poly- morphonuclear cells. This proportion does not seem to bear any relation either to the acuteness or to the stage of the disease, but is usually higher in children than in adults. Acute meningitis due to pyogenetic organisms always leads to an enormous increase of cells, the great majority of which are polymorphonuclear. Along with these there is usually a small proportion of large phagocytic mononuclear cells. In this as in the tubercular form the causal organism may be seen in stained films either free or within the cells. Finally, in all degenerative conditions of the brain or cord, compound granular corpuscles may make their way into the cerebro-spinal fluid along the Virchow-Robin space of the veins. (c) The albuminous substances are subject to great variations. As a general rule, in diseases other than syphilitic disease of the nervous system the chief increase is in the serum albumen. In syphilitic disease the serum globulin also rises in percentage to a considerable degree, though never to the same extent as the serum albumen. Herein lies the advantage of com- bining an estimation of the total albumen with the Nonne- Apelt or Noguchi reaction. If this is not done the observer may be led astray by various non-syphilitic conditions which give a positive globulin reaction, along with a considerable increase in the total albumens and in the cellular elements. If, on the other hand, a positive globulin reaction is obtained from a fluid with a total albumen content of less than 0*04 per cent., there is a strong presumption that the case is syphilitic. In acute tubercular meningitis the albumen may rise to 0*2 per cent, or even higher. It is apt to be rather higher in adults than in children. Meningococcal meningitis gives especially high albumen percentages, 0*3 per cent, being common and 0*6 to 0*8 per cent, not very unusual. In meningeal inflamma- tion due to other organisms the albumen content usually does not rise above 0*2 per cent. In all these acuter forms of meningitis fibrinogen is present, so that a coagulum forms in the tube soon after the fluid is withdrawn. This constitutes a very valuable diagnostic sign in tubercular meningitis, and should always be examined for when the fluid appears to be almost clear. It may be missed if the fluid is shaken about after withdrawal. CEREBRO-SPINAL FLUID 41 In most cases of subacute meningitis, syphilitic or otherwise, the percentage of albumen does not rise above o-i per cent. In syphilitic meningitis, quite apart from the " syndrome of Froin," fibrinogen may sometimes be found in the acuter stages of the disease. Fibrinogen may be present in sufficient quantity to give rise to a fine coagulum without any very great rise in the quantity of albumen. This is most frequently the case in poliomyelitis and polio-encephalitis, in which fibrinogen may be present for several weeks after the onset of the disease. In acute polyneuritis the albumen content may be raised to O'l per cent, or over. This may be due, as Mestrezat considers, to an accompanying inflammation of the nerve cells of the ventral cornua and dorsal root ganglia — a so-called neuronitis. It is also probably due, in part at least, to the addition of lymph which has passed up the lymphatics of the inflamed nerves into the subarachnoid space. This addition of albumen will be the more evident in the cerebro-spinal fluid drawn off from the lumbar theca in that it reaches the sub- arachnoid space chiefly along the large roots of the lumbo- sacral plexus. In certain cases of cerebral oedema, as in uraemia, there is a definite rise in the albumen content of the cerebro-spinal fluid, but except for this it is seldom increased by diseases which do not primarily affect the cerebro-spinal nervous system. The greatest increase in the percentage of albumen is seen in Froin's loculation syndrome, where readings above i per cent, are by no means infrequent. Kafka has applied to the examination of the cerebro-spinal fluid the method of salting out the different globulin fractions by different degrees of saturation with ammonium sulphate. He finds that a precipitation of globulin is given by 28 per cent, saturation in acute meningitis, by 33 per cent, saturation in general paralysis, and by 40 per cent, in lues cerebri. These percentages indicate the presence of fibrinogen in acute menin- gitis (and also occasionally in the more acute forms of syphilitic meningitis), of euglobulin in general paralysis, and of pseudo- globulin in lues cerebri. According to Mestrezat, the normal cerebro-spinal fluid contains rather more globulin than serum albumen. But in all forms of inflammation of the meninges the serum albumen 42 CHEMICAL CHANGES IN appears in the cerebro-spinal fluid in far greater quantity than the serum globuHn, the proportions ranging from i: 6 to i: 8. When, however, the choroid plexus becomes diseased and its function of selective filtration is lost, globulin may pass into the cerebro-spinal fluid in much greater amounts, and the proportion of globulin to albumen may rise to i : 2. Similarly in the increase of proteid shown by loculated fluids, globulin may bear a proportion to albumen of 1^3. Eskuchen gives the proportion of globulin to albumen in cerebro-spinal lues as 1 : 12, and in general paralysis 3:7. (d) Glucose. — The percentage of glucose in the cerebro-spinal fluid may be increased or reduced. Of the factors leading to increased percentages the most important is excess of glucose in the blood (hyperglycaemia) . In diabetes mellitus the amount in the cerebro-spinal fluid is constantly above o*i per cent. It is also above normal in all acute infective diseases or intoxications associated with a general polymorphonuclear leucocytosis. This is probably due to the power which these cells possess of carrying glycogen to the tissues. An increase is usually found in cases of increased intracranial pressure, even during the initial stages of tubercular meningitis. The cause of this is obscure, but it may be due to the effects of pressure on the pituitary gland. The percentage of glucose is reduced in all cases of meningeal inflammation, whether acute, subacute or chronic. In acute meningitis the small amount of glucose that may be present when the fluid is newly drawn may disappear from the fluid in a few hours. This is due chiefly to the avidity with which bacteria use glucose as a food and also to some extent to the power which polymorphonuclear leucocytes possess of absorbing it. Glucose is also reduced in cases of subarachnoid haemorrhage from any cause. As will be seen later in dealing with the loculation syndrome, a fluid which contains no bacteria and only an extremely small number of polymorphonuclear cells may have its glucose reduced if it lies close to an area of meningeal inflammation. In such cases a process of diffusion must be supposed to take place between the loculated sugar-containing fluid and the purulent sugar- free cerebro-spinal fluid above it. In all cases of meningitis there is, along with the diminution CEREBRO-SPINAL FLUID 43 of glucose, a diminution of the alkalinity of the fluid due to the formation of organic acids. (e) Chlorides. — The constant level (0*70 to 0*76 per cent.) to which the chlorides in the cerebro-spinal fluid normally keep makes their estimation one of the most useful parts of the examination of the fluid. The indications which ab- normal percentages give are also very plain and straight- forward. A raised percentage of chlorides is always an indi- cation of renal inadequacy ; a slightly lowered percentage (from 0*68 to 0'70 per cent.) is an indication either of an acute general infection or of slight meningitis. Figures below 0*68 per cent, are indicative of grave meningeal infection. The only condition in which percentages below 0*6 per cent, are found is tubercular meningitis. This point is emphasised by Mestrezat, who regards it as of a pathognomonic significance second only to the discovery of the tubercle bacillus in the cerebro-spinal fluid. It is due in part to the reduction of the chlorides in the blood, which occurs in acute general tuber- culosis, and partly to a local action of the bacilli in the sub- arachnoid space on the chlorides in the cerebro-spinal fluid. The chloride percentage may not, however, be reduced in the early stages of the disease. (/) Urea. — The chief importance of the urea in the cerebro- spinal fluid is in relation to uraemia. But in meningitis or wherever there is an abnormally free passage of lymph into the cerebro-spinal fluid the urea is increased, as also in many degenerative conditions of the cerebro-spinal axis. Any reading above 0*05 per cent, with a normal albumen content is indicative of renal inadequacy. In uraemia the percentages run higher. Mestrezat states that cases with a percentage of 0*3 are uniformly fatal. (g) Loculation syndrome (Syndrome of Froin). — When for any reason the cerebro-spinal fluid is shut off completely, or almost completely, from access to the lower part of the spinal theca, the fluid obtained by lumbar puncture is altered in certain definite ways. The chief is a great increase in the proteid con- tent, both albumen and globulin being represented in propor- tions approximating to those in which they occur in the blood serum. Frequently, and in well-established cases constantly, the fluid acquires a yellow tint, varying from the palest straw 44 CEREBRO-SPINAL FLUID colour, only visible on looking down the long axis of the tube, to a dark greenish-tan colour resembling that of the urine in jaundice. Along with these changes fibrinogen is almost always present, and sometimes causes the fluid to coagulate spontaneously in the test-tube as soon as it is withdrawn, but if cellular elements are scanty it may be necessary to add a drop of fresh blood, thus providing fibrin ferment, in order to produce this coagulation. A yellow fluid coagulating spontaneously within a few seconds of withdrawal from the body gives a typical picture of the syndrome described by Froin in 1903. In addition to the above-mentioned changes which give it its characteristic appearance, it may show other abnormalities of which the most important is the presence of albumoses and peptones due to the autolysis of albumens, so that the fluid may give a typical biuret reaction after removal of the albumen. This is only found in the more pronounced cases. The percentage of glucose is usually in- creased slightly, and the urea percentage may be considerably increased. The cell content and the chloride content are frequently normal, but these depend entirely on the cause of the loculation of the fluid. Where it is due to syphilitic menin- gitis the cells may be very greatly increased and a considerable percentage of large mononuclear cells may be present. Usually the chlorides are slightly reduced (in the neighbourhood of 0*7 per cent.), but when the loculation is due to some acute inflammation, e.g. acute myelitis or meningitis, they may be considerably reduced. The most important part of the syndrome undoubtedly is the high albumen content, and it may be laid down as a rule that any cerebro-spinal fluid containing i per cent, albumen or over is shut off from communication with the ventricular system. But with very much lower albumen percentages, and in the absence of any yellow coloration, there may be a strong presumption that we are dealing with the loculation syndrome if the number of cells and the percentages of the other constituents of the fluid are normal. Such fluids are often found in cases of spinal tumour. Pott's disease or arach- noid cyst where the communication between the ventricles and the lumbar theca is not completely obliterated. The highest albumen percentages we have personally encountered, e.g. SYNDROME OF FROIN 45 2 per cent., 2*5 per cent., and 3 per cent., have been in cases of acute myelitis or meningitis. In all of these the number of cells would have given no indication of the inflammatory nature of the disease, but it is possible that the vascular congestion was responsible for producing the very high percentages of albumen which were found. In one of these the fluid was only very slightly tinted with yellow, and gave a good reduction of Fehling's solution. In the other two the glucose was diminished or absent, and both were associated with purulent meningitis above the level of the meningeal adhesion. From the point of view of the cell content, fluids of this character separate themselves into two classes: (i) those with a high cell content (which provide the majority of the fluids which coagulate spontaneously after withdrawal, i.e. which contain fibrin ferment) are almost constantly due to syphilitic meningitis. (2) Fluids with low (normal) cell content may be due to a variety of causes, e.g. spinal tumour, arachnoid cyst. Pott's disease, extradural tumour or abscess, chronic basal meningitis, acute myelitis and, rarely, acute meningitis. In these the nature of the cells (i.e. the presence of cells other than small lymphocytes) and the percentages of glucose and of chlorides are of great diagnostic value. Pathology of the syndrome. — Assuming that the fluid in the lumbar theca is constantly renewed from above by being pumped to and fro by the arterial and venous pulses, then, if for any reason the ventricular fluid fails to get free access to the lumbar theca, the fluid there must either remain stagnant or be absorbed along the perineural lymphatics and possibly also along the perivascular sheaths of the spinal veins. In either case there will be a constant admixture of lymph by way of the perineural lymphatics, the spinal capillaries and the Virchow-Robin spaces of the cord. Gradually, therefore, the fluid in the loculated area of the subarachnoid space comes to correspond more and more closely in con- stitution to blood plasma. In addition, the albumens may become autolysed, giving rise to albumoses and peptones, and if any red blood corpuscles find their way into the fluid by rupture of congested vessels, these will be broken up into pigments derived from the haemoglobin. It is probable also that venous congestion below the level of the lesion plays, a considerable 46 WASSERMANN REACTION part in the production of the syndrome, but it is doubtful whether this is a constant or necessary factor. (h) Wassermann reaction. — The examination of the cerebro- spinal fluid as well as the blood for the Wassermann reaction is of such importance that it has become a routine measure in cases of suspected syphilis of the nervous system. In progres- sive general paralysis the reaction is positive in both cerebro- spinal fluid and blood in practically every case, and is slightly, if at all, influenced by treatment. This rule also holds good, although not to the same extent, for tabes dorsalis, but in this disease from 20 to 30 per cent, of cases give a negative reaction both in the cerebro-spinal fluid and blood. It may be assumed that some such cases have become stationary. In tabes it is not uncommon to find the cerebro-spinal fluid giving a positive reaction when the blood gives a negative, especially in treated cases. The reverse condition, a positive reaction in the blood with a negative reaction in the fluid, may also be found in tabes, but is much more common in syphilitic meningitis, especially when the disease is wholly or chiefly confined to the vertex of the brain. When the spinal membranes are involved either alone or in conjunction with disease of the brain, the Wasser- mann reaction is usually positive both in blood and fluid. This anomaly may be understood on the assumption that the current of the cerebro-spinal fluid over the cortex is towards the dural venous sinuses, and that any antibodies elaborated as a result of the presence of the Spirochaeta pallida in the cerebral meninges are rapidly washed out into the general circulation. On the other hand, a considerable proportion of the antibodies formed in relation to the membranes covering the cord would be found in the fluid removed by lumbar punc- ture. In these cases of syphilitic meningitis the Wassermann reaction can be altered by treatment, and may be rendered completely negative in the fluid as well as in the blood. In the meningitis of secondary syphilis the reaction of the cerebro- spinal fluid is much less frequently positive than in the later stages, in spite of the presence of large numbers of lymphocytes and an increased globulin and albumen percentage. The Wassermann reactions of the blood and cerebro-spinal fluid, the presence of a lymphocytosis of greater or less degree, and a positive globulin reaction constitute the " four reactions " GOLD-SOL REACTION 47 which are of chief value in the diagnosis of syphiUtic disease of the nervous system. By means of these reactions it is possible to diagnose definitely syphilitic disease of the nervous system, but it is not always possible to tell what form of disease is present. In a doubtful case the re-examination of the cerebro-spinal fluid after a few months of energetic treatment will usually decide whether the disease is of the " parenchyma- tous " or " interstitial " form. The effects of treatment are usually first evident in regard to the cells and albumen content. The disappearance of abnormal cells and a gradual diminution of the lymphocytosis are favourable prognostic signs. (i) Lange's colloidal gold reaction. — According to Cruickshank, the reaction depends on the power which the globulin of human serum possesses of precipitating metallic gold from a colloidal solution. This precipitation is checked by the protective action of serum albumen. In the test a series of increasing dilutions of cerebro-spinal fluid from i: 10 to 1:5120 is added to a fixed quantity of colloidal gold, and the alteration of colour is noted in each tube in order. Various curves are thus given which indicate the balance of the precipitating and protective factors. Fig. 103 indicates the classes of curve which are obtained in this way. Experience has shown that the results of the reaction can only be interpreted in relation to the clinical findings. Thus, while the type of curve indicated for general paralysis is almost constantly found in that condition, it may also occur in acute forms of tabes and in disseminated sclerosis. Similarly, the so-called " syphilitic curve " may be given by a variety of conditions, among which may be mentioned subacute combined degeneration, and all gradations between this curve and that of acute meningitis may be given by many of the acute or chronic forms of nervous disease, e.g. poliomyelitis, lethargic encepha- litis, disseminated sclerosis, etc. While it may be of some use in distinguishing between functional and organic disease of the central nervous system, the reaction has probably more theoretical than practical value, and is apt to lead to a con- fusion of diagnosis if too much reliance is placed on it. (/) Relationship of anatomical with clinical phenomena. — That hydrocephalus is due to a disorder of the circulation 48 CAUSATION OF HYDROCEPHALUS of the cerebro-spinal fluid has long been known. Theoretically it may arise either (i) because more cerebro-spinal fluid is secreted than can be absorbed, or (2) because the escape of the cerebro-spinal fluid from the ventricles is blocked. The former is probably the cause of many cases of congenital hydrocephalus, and must be the cause when there is ** external " as well as " internal " hydrocephalus; whereas cases of hydrocephalus following post-basic meningitis or occurring with tumours in the region of the pons are to be explained by the latter cause. It is commonly observed that a tumour in one centrum ovale or on the cortex of the parietal lobe leads to hydrocephalus of the opposite ventricle. For this there may be several reasons, (i) The increase in the size of one cerebral hemisphere may compress the third ventricle laterally, or (2) may press on the roof of the iter of Sylvius ; (3) the flattening of the convolu- tions due to the increased volumxC of the brain may to some extent dam back the cerebro-spinal fluid from reaching the superior longitudinal sinus by which a large part of it is nor- mally absorbed ; or (4) there may be delay in the escape of fluid from the fourth ventricle due to the formation of a " pressure cone " consisting of the amygdaloid lobules of the cerebellum, which are pressed down into the foramen magnum behind the medulla, taking the place normally occupied by the cisterna magna. A history of a blow on the head or a sudden jar to the cranial contents, such as is caused by falling in a sitting position, is sometimes elicited in cases of hydrocephalus in the adult. Trotter has put forward a very suggestive theory to account for such cases. By the jar the dorsal surface of the mid-brain is brought sharply against the sharp anterior edge of the tentorium cerebelli. The resulting bruising and oedema in this part of the brain rapidly closes, temporarily at all events, the iter of Sylvius, and hydrocephalus results. Once it is present a vicious circle is established in which factors 3 and 4 play the chief part, and the condition may remain permanently or progress and lead to the death of the patient. GENERAL PATHOLOGY 49 REFERENCES The Neuron. BIELSCHOWSKY, M. I Histologic und Histopathologie des New en- systems. Berlin, 1910. GiERLicH, N., AND Hexheimer, G. : Studien eber die Neurofibrillen im Zentral- nervensystem. Wiesbaden, 1907. Homen: Changes in Nervous System after Amputation, Ziegler's Beitrdge, vol. vii. Marinesco: La cellule new euse. Paris, 1909. MoTT, F. W. : An Introduction to Neuropathology. Allbutt and RoUeston's System of Medicine, 1910, vol. vii., pp. 173-236. Ramon-y-Cajal, S. : Nouvelles observations sur revolution des neuroblastes, etc., Anat. Anzeigev, Jena, 1908, pp. 1-25 and 65-87, also pp. 418-448 and 468-493 (also previous papers). Ross Harrison : The Outgrowth of the Nerve Fibre, etc., and previous papers, Journal of Exper. Zoology, 1910, vol. ix., p. 787. The Neuroglia. Alzheimer, A.: Beitrdge zur pathol. neuroglia, etc. Jena, 1910. Volland: Zeitschrift f. d. ges. Neurologic und Psychiatric, 1914, Bd. 21, z. 194. Paths of Infection. Orr, David, and Rows, R. G. : Subacute and Acute Inflammatory Reactions produced in the Spinal Cord by Infection of its Lymph Stream, etc., Brain, 191 8, vol. xli., p. i, and references to earlier papers. Teale and Embleton: Studies in Infection — II., Journ. of Path, and Bad., 1919, vol. xxiii., No. i. McIntosh, J., AND FiLDES, P.: The Factors which Govern the Penetration of Arsenic, etc., into the Brain, Brain, 1916, vol. xxxix., p. 478. Cerebro-spinal Fluid. Bruce, A., and Dawson, J. W.: On the Relation of the Lymphatics of the Spinal Cord, Rev. of N cur. and Psych.. 1912. Blancheti£:re and Lejonne: Gaz. des Hopit., 1909, Ixxxii., 1303. Cruickshank, J.: The Value and Mechanism of the Colloidal Gold Test, Brit. Journ. of Exper. Path., 1920, vol. i., p. 71. EsKUCHEN, Karl: Die Lumbalpunktion. Berlin, 1919. Froin: Inflammations meningees avec reactions chromatiques, etc., Gaz. des Hopitaux, Sept., 1903. Head, H., and Fearnsides, E. G.: Syphilis of the Nervous System, Brain, 1 91 5, vol. xxxvii. Hill, Leonard : Physiology and Pathology of the Cerebral Circulation. London, 1896. Key, G., and Retzius, A. : Anatomic des Ncwensy stems und des Bindesgewebcs. Stockholm, 1876. Levinson, a.: The Cerebro-spinal Fluid. New York, 1919. Majendie, F. : RSchcrches smv Ic Liquide Cephalo-Rachidien. Paris, 1825 and 1842. Mestrezat, W. : Lc liquidc cephalo-rachidien. Paris, 191 2. Miller, Brush, Hammers and Felton: Bulletin of the Johns Hopkins Hospital, 191 5, vol. xxvi., No. 298, p. 391. MoTT, F. W.: The Oliver- Sharpey Lectures on the Cerebro-spinal Fluid, Lancet, 1910, vol. ii. Quincke, H.: Ueber Lumbalpunktion, Berlin, klin. Woch., 1895, No. 41. Robin: Recherches sur les capillaires de I'encephale, Journ. dc Physiologic, 1899. SiCARD AND Descomps: Syndrome de Coagulation Massive, etc., Gaz. des Hopit., 1908, Ixxxi., 1431. Weed, L. H.: Studies on the Cerebro-spinal Fluid, Journ. Med. Research, 1 91 4-15, vol. xxxi. 4 CHAPTER II DEVELOPMENTAL AND FAMILIAL DISEASES Certain abnormalities of the central nervous system show themselves at birth either as results of defective development or of intra-uterine disease. These will be dealt with under the term " developmental disease." Along with these must be classed injuries received during birth and certain cases of idiocy, in which the defect appears to date from birth, although in some cases it may be the result of a degeneration starting in the early months of life. Closely related to the latter are those forms of disease which tend to attack several members of one family, and usually commence about the same age in each member. These diseases may start at any stage of the period of growth, though many of them have a preference for a certain definite age period. Their tendency to occur in families separates them from other diseases, and classes them aetio- logically as due either to an inherited tendency to degenera- tion or abnormal function of certain tissues, or to an inherited lack of vitality in these tissues. I. Developmental Diseases. (a) Defective development. — In these cases, from a certain period of intra-uterine life, development seems to have pro- ceeded along perverted lines. The commonest of these are associated with defective closure of the dorsal groove (cranio- rhachischisis). The most complete example of this is the anencephalic monster; less complete are the cases of exencephaly, in which there is a larger or smaller defect in the cranial vault. In the lesser degrees of this we have simply a meningocele, in which a pouch of meninges, containing simply fluid or a varying amount of cerebral matter, extrudes through the small defect in the skull. This usually takes place in the middle line, 50 SPINA BIFIDA 51 and is most common either at the junction of the frontal and nasal bones or in the occipital region. When it affects the spine alone, the condition is termed rhachischisis or spina bifida. This, again, may exhibit all degrees from those in- compatible with life to conditions accidentally discovered in apparently healthy people. (i) In spina bifida completa the cord is exposed in a greater or less extent with no covering at all. This is due to the failure of the medullary groove to close in and form the neural canal, and in consequence the ependyma, which normally lines the central canal of the cord, here covers its dorsal surface, and the central canal opens into the defect above and below. (2) In spina bifida incompleta the medullary groove is closed, but the bony and connective tissues have failed to cover it normally. There results an opening in the posterior wall of the spinal canal, through which there may protrude a pouch of dura mater. This may contain simply fluid or undeveloped medullary substance. Frequently a pad of fat lies over the meningeal pouch or covers the opening in the spinal canal when no pouch is protruded, and may press on the cord through the dura mater. The term spina bifida occulta is given to the cases where there is no meningeal protrusion, and where the only indication of the abnormality is a tuft of hairs over- lying the bony defect. This defect is usually in the lumbar region, rarely in the cervical or thoracic: in many cases the spinal cord beneath it is imperfectly developed. Cyclencephaly . — In some cases there is absence of the normal division of the brain into two hemispheres, and this is usually associated with the presence of a single eye in the middle of the forehead or of a double eye in a single socket. Porencephaly. — Of this there are two forms, the true and the false. True porencephaly is an abnormal development of the brain in which certain parts of the cortex of the brain, both- white and grey matter, fail to be developed, and in consequence we have a direct communication between the lateral ventricle and the surface of the brain. Usually this occurs on the sur- > face of the hemisphere, but a smaller defect may exceptionally be found in the corpus callosum or the cerebellum. When it occurs on both hemispheres the lesions are symmetrical. 52 PORENCEPHALY In its greatest degree the two hemispheres are almost non- existent, except for remains of the cortex at the lower part of the anterior pole. In these cases the lenticular and caudate nuclei and the optic thalami persist. The usual form is a crateriform cavity on the surface of one or both hemispheres communicating with the lateral ventricle. Its sides are formed by complete convolutions which dip down smoothly into it. These, again, are covered by a vascular membrane continuous with the pia-arachnoid. These characteristics distinguish true poren- cephaly from pseudo-porencephaly , which is described below (pp. 56 and 121). Another type of developmental defect is shown by those brains which present the features associated with the early months of foetal existence, and which do not show the division of the cranial hemispheres into numerous convolutions {v. Fig. 9). Cranio-cleido-dysostosis is sometimes associated with defects of the central nervous system, particularly with hydrocephalus and hydromyelus. (b) Other congenital abnormalities are due to prenatal diseases or to degenerations occurring in a central nervous system which, up to that point, has been developing normally. In some of these the disease or morbid process is progressive; in others the abnormality produced gives rise to disease in later life. One form of morbid process which is often concerned in producing such congenital abnormalities has been termed by the French atrophic sclerosis. This may give rise to many forms of idiocy, imbecility and infantile diplegia, and is probably at the bottom of many cases of microcephaly, as well as of microgyria, cerebral asymmetry and external hydro- cephalus. Macroscopically, the chief appearance is that of atrophy, either localised or affecting to a greater or less extent the whole brain substance. In slight cases the convolutions when stripped of their meninges appear normal, but examina- tion with a lens reveals here and there depressions or scar-like puckerings. The surface may have a finely worm-eaten appearance. In some advanced cases the convolutions may be irregular, atrophied, firmer than normal, and unlike healthy brain substance. In extreme degrees of the process the con- volution is represented by a thin leaf of fibrous substance, the so-called " parchment-like convolution." Sections of such DEVELOPMENTAL AND FAMILIAL DISEASES 53 FiG: 9. Retarded development of brain from a diplegia infant eighteen months old. Primitive convoluting is apparent on the surface. 54 ATROPHIC SCLEROSIS areas show numerous small cavities in the grey matter. A slight degree of this process affecting both hemispheres fairly uniformly may be one of the causes of microcephaly. Where it affects one hemisphere more than the other we get the rare Fig. lo. Cerebral asymmetry of congenital origin. cases of inequality of the two hemispheres; in both cases the parts of the mid- and hind-brain associated with the sclerosed areas fail to develop, and in the latter case the cerebral peduncles are smaller on the affected side. Where the disease DEVELOPMENTAL AND FAMILIAL DISEASES 55 is more localised and patchy in distribution, one or more lobes or convolutions may be chiefly affected. Microscopically the condition is one of neuroglial overgrowth associated with degeneration of the neuron substance. The process appears to affect primarily the deeper layers of the cortex, and spread Fig. II. Brain from case of diplegia showing shrunken convolutions and patches of cortical degeneration. thence to the underlying white matter and to the superficial layers of the cortex. The vessels are affected, showing — (i) Perivascular neuroglial sclerosis. (2) A dilatation of the adventitial lymph spaces with gran- ular corpuscles. (3) Some degree of periarteritis. 56 HYPERTROPHIC TUBEROUS SCLEROSIS The ependyma is involved, and may proliferate and produce glandular masses which penetrate more or less deeply into the tissue round the ventricle. This tissue usually shows a great degree of sclerosis, and may narrow the outlet from the ventricles, thus producing internal hydrocephalus. In the more extreme degrees of the process the sclerosed tissue tends towards cavity formation. This may happen in the following ways : (i) The smallest cavities are simply a dilatation of the lymphatic sheaths of the vessels. (2) Others are areas of vascular softening which have become cystic, as evidenced by yellow pigment due to altered blood either in the fluid or in the walls. (3) The most common form of origin is the molecular dis- integration of the neuroglial masses. Apparently the essential element in the disease is the neurc- glial overgrowth. The glial cells are from the first greatly multiplied, but giant spider cells (" fibre-forming glial cells ") are rare. Hypertrophic tuberous sclerosis. — In this disease the brain appears sprinkled with whitish nodules of various sizes, from that of a pea to that of a walnut. They may be rounded or elongated, and may occur both in white and in grey matter, but mainly in the latter. Cavities may be formed between them. The brain substance as a whole shows little alteration, and the lesions are not so diffuse as in the atrophic form. Microscopically these nodules are characterised by a large number of " fibre-forming glial cells." These are, for the most part, mononuclear and fusiform. They are similar to cells seen in other types of gliomatosis, but in this disease, especially in the nodules of the white matter, they dominate the picture. The nodule is mainly composed of glial fibres with comparatively few cells. Its structure blends insensibly with the surrounding nervous tissue. Pseudo-porencephaly . — This differs from the true form (p. 51, fig. 27) in that it is the result of a destructive process, such as haemorrhage, thrombosis or encephalitis in a develop- ing brain. In this form the cavity is lined by a thin cyst wall, and does not usually communicate with the lateral ventricle. The fluid in the cavity may be clear or yellowish from blood HYDROCEPHALUS 57 pigment. The cavity cuts across the convolutions irregularly, and its walls are formed of white matter to which the cyst wall is firmly adherent. Microscopically there is evidence of the primary lesion. Hydrocephalus. — By external hydrocephalus we mean the presence of spaces between the dura mater and the surface of the brain filled with cerebro-spinal fluid. This form is always congenital, and of doubtful aetiology, but in some cases it is associated with atrophic sclerosis. Fig. 12. Hydromyelus of cervical cord found in a child who died of acute poliomyelitis. Internal hydrocephalus may be congenital or acquired. In both cases it may be due to closure of the foramina in the ependymal covering of the fourth ventricle, or to other mechanical blockage of the outlet of cerebro-spinal fluid. In most congenital forms it is due to stenosis or obstruction of the iter of Sylvius, and only the lateral and third ventricles are distended. When it results from prenatal inflammation of the meninges round the fourth ventricle, or from closure of the foramina of Majendie and Luschka, the fourth ventricle 58 HYDROMYELUS may be very greatly distended, leading to the condition called "hydrocele of the fourth ventricle." The distension of the ventricles may be enormous, and the brain tissue covering them may be extremely thin. Hydromyelus. — This in its pure form is a simple dilatation of the central canal of the cord. The dilatation may affect the canal equally throughout its length or may be greater in certain areas. It is probably in most cases an error of develop- ment, but it may be associated with hydrocephalus. It may give rise to a progressive symptomatology as the dilated canal becomes surrounded with gliomatous tissue, and in this way many cases of syringomyelia undoubtedly arise. The two con- ditions, in fact, insensibly merge into one another (v. Fig. 12) . 2. Birth Injuries. Traumatism during birth may give rise to unilateral or bilateral defects in the hemispheres, the most common of which are due to haemorrhages either within the brain substance or upon its surface. These haemorrhages may be arterial in origin, or may be the result of rupture of the venous sinuses, due to over-riding of the cranial bones during the passage of the skull through the pelvic outlet. 3. Familial and Congenital Diseases. (i.) Cerebral and Myelopathic. (a) Amaurotic Family Idiocy {Tay-Sachs Disease). Aetiology. — This disease was first observed in 1881 by Waren Tay, who described fully the characteristic changes at the macula. Sachs, in 1887, investigated it from the neurological point of view and showed its familial character, as twenty- eight of his cases occurred in fifteen families. Since that time a large number of cases has been recorded, all of which have been the children of Jewish parents. Sachs called the disease " amaurotic family idiocy," but as the disease is not present at birth, but develops after the first few months, it is not pro- perly classed as " idiocy." The first-born are rarely affected, but when one child has had the disease subsequent children AMAUROTIC FAMILY IDIOCY 59 rarely escape. It is invariably fatal, usually before the end of the third year. Macroscopic. — The cherry-red spot at the macula surrounded by a pale halo is clearly seen in eyes examined post-mortem. The brains are of normal development, but the convolutions are shrunken and the sulci widened. There is an abnormal amount of fluid in the subarachnoid space. Microscopic. — (i) Tract lesions. — The pyramidal tracts throughout their whole course are smaller than normal. This is best seen in the pes pedunculi. But the tract degenera- tion is not confined to the pyramidal system, as in all tracts of the cord Marchi's stain shows evidence of degeneration. Wei- gert's method shows signs of atrophy in the tangential fibres of the cortex and in the fibres of Gennari in the visual cortex, but the optic radiations and the medullated fibres of the centrum ovale are relatively little affected. (2) The chief morbid changes are seen in the ganglion cells. These are swollen, some globular in form, others showing balloon-like swellings at the base of the dendrites. The nucleus usually stains well, and shows relatively slight changes, but is eccentrically placed, being apparently pushed towards the apical portion of the cell by a swelling of the cell sub- stance at the opposite pole. It is surrounded by a zone of granular matter which takes the stain by Nissl's method, although healthy Nissl granules are rarely seen. Elsewhere the cell protoplasm is clear and stains poorly, so that it is diffi- cult to distinguish the cell margin from the surrounding tissues, and vacuoles of varying size may be present. The swelling may be very great and the pyramidal cells may measure from 30 [jl to 60 jj,. A balloon-like swelling on the proximal portion of one of the dendrites, separated from the cell by a narrow neck, is sometimes seen, or there may be varicose swellings in the dendrites a short distance from the cell body. By Bielschowsky's neuro-fibrillar stain it is evident that the fibrils in the dendrites are relatively little affected. They do not, however, run through the cell body in a normal manner, but are pushed aside by the globular swelling or vacuolation, and are often seen lying bunched closely together at the side of the cell. In a similar manner they run round and not through the swellings in the dendrites. 6o DEVELOPMENTAL AND FAMILIAL DISEASES r -■"'f^-p'^rrmfk JillilSfiisH*^ V, - \i # :~ %. ■ m ' ' \ 1 b Fig. 13. a, Cells from dorsal horn and Clarke's column in a case of amauroHc family idiocy. b, Purkinje cells from the same case. AMAUROTIC FAMILY IDIOCY 6i Mott has shown that these degenerated nerve cells contain an abnormally large amount of lipochrome granules which stain by Scharlach R. or Sudan III, and also by Marchi's method. The ganglion cells are surrounded by " parasitic " glia cells, in which the lipochrome granules are rather larger, and granular corpuscles filled with fatty substances are numerous. These changes in the nerve cells are seen throughout the central nervous system. They are, perhaps, most marked in the cells of the hippocampal region and frontal lobes, but are present in the pyramidal cells, and in the cells of the ventral horns and Clarke's column. The cerebellum is less affected, but certain of the cells of Purkinje show similar changes. Neuroglial overgrowth is slight, and seems to be secondary to the degenera- tion of neurons. Apart from some degree of congestion the blood vessels are normal. The changes in the retina are essentially of the same nature. The ganglion cells show lipochrome granules, but there is little change in the nerve fibrils. The fovea centralis is abnormally thin, and is surrounded by a zone of slight oedema. The optic nerves show some degree of atrophy, and this is also seen in the optic tracts. From the microscopic appearances it seems that the " hyalo- plasm " of the nerve cells is primarily affected, becoming swollen, and undergoing degeneration into various lipoid substances. The changes in the Nissl granules follow rapidly, but the neuro-fibrils continue for some time comparatively immune. The cells may die or may enter a state of necrobiosis in which they continue until the death of the patient. From the pathological standpoint this disease is scarcely distinguishable from certain cases oi familial cerebral degenera- tion with macular changes occurring at a later period of child- hood, and not confined to the Jewish race. Although the clinical picture varies in some ways, and the typical macular changes of Tay-Sachs disease are not present, the classical description of the pathological findings, both macroscopic and microscopic, applies with few differences to the two forms. 62 WERDNIG-HOFFMANN PARALYSIS {b) Progressive Spinal Muscular Atrophy of Infants (Werdnig- Hoffmann Paralysis). This rare disease was described first by Werdnig in 189 1 and by Hoffmann in 1894. The total number of recorded cases is still small (between twenty and thirty), but sufficient have been examined to establish the specificity of the disease and the nature of the pathological changes special to it. Aetiology. — The patients are usually well developed at birth and appear normal, but after a few weeks or months muscular weakness sets in, commencing in the muscles of the back and the proximal parts of the limbs, and spreading to the inter- costal and abdominal muscles, the neck muscles and the muscles controlling the knee and elbow. The diaphragm and the muscles supplied by the cranial nerves are spared, and the movements of the fingers and toes are usually affected late in the disease. The paralysis steadily increases and leads to the death of the patients at ages varying from a few months to six years. In several instances more than one member of a family has been affected, but in no case has any evidence of hereditary transmission been recorded; nor does the disease affect particularly the last members of families. From the point of view of pathology it is related most closely to the classical types of progressive muscular atrophy (motor neuron disease, p. 271), but has an inverse distribution of the paralysis. In both cases we are absolutely in the dark as to the cause of the disease. Histology. — The chief changes are those found in the ventral horn cells and their nerve roots. These cells are greatly diminished in number, especially in the older and more advanced cases, and some of them show chromolysis of varying severity. In some cases very little chromolysis has been present, but the cells are definitely smaller, and much less numerous than normal. The ventral nerve roots are thin and their myelin stains poorly as compared with the dorsal nerve roots. Marchi's method shows recent myelin degeneration in some cases. T-he affected muscles also show definite changes, which appear to be secondary to the atrophy of the neurons. A varying proportion of the muscle fibres are of normal size or slightly larger, and either retain their usual polygonal outline or appear FRIEDREIGH'S ATAXIA 63 rounded on section, some of them showing hyaUne changes. Their nuclei are not increased in number, and are present only in relation to the sarcolemma. On the other hand, a large number of the muscle fibres are extremely small, measuring 10 fjL or less in diameter. Sometimes it is possible to see a large fibre breaking up into a number of smaller fibres. The nuclei of the small fibres appear greatly increased, but both large and small fibres usually preserve their transverse striation. Marchi's method shows evidence of fatty degeneration in diffuse stippling of the muscle fibres with black dots. The intermuscular fibrous tissue is increased and numerous fat cells are present between the fibres, but these changes do not approach the degree seen in myopathic muscles. The appear- ances, on the whole, resemble those seen in the wasting of muscles following disease of their motor nerves, although certain features of this are wanting. This may, perhaps, be explained by the age of the patient and the relatively undeveloped state of the muscles when they become affected. (c) Friedreich' s Disease. Aetiology. — The disease tends to attack more than one member of the same family, but it is not often directly inherited, and cases with regard to which no family history can be obtained are by no means uncommon. It affects both sexes, and its onset is generally noticed either in childhood or in early adoles- cence. The occurrence of the disease in several members of the same family, and the fact that some portions of the central nervous system have often been found to be unusually small, lead to the conclusion that the disease is really due to some inherent congenital defect. It may be cited as a good example of a nervous abiotrophy, or in other words of a tendency to early decay and subsequent death of certain tracts in the central nervous system for no apparent reason other than their lack of vitality. Morbid anatomy. — Gross examination of the central nervous system often reveals a fairly well-developed brain with a some- what abnormally slender spinal cord. In only a few instances has the cerebellum been small in proportion to the cerebrum or shown any naked-eye changes. 64 DEVELOPMENTAL AND FAMILIAL DISEASES On cross section through the spinal cord sclerosis of the dorsal and lateral columns is readily distinguished, and the definite limitations of the sclerotic process can be best studied by the Weigert-Pal method. Such sections show that a primary degeneration of the dorsal columns, of the pyramidal and of the spino-cerebellar tracts is the essential feature of the disease. If the dorsal columns are examined more minutely it will be seen that the endogenous fibres suffer less than those which enter the spinal cord through the dorsal roots. Goll's column is generally more completely degenerated than that of Burdach. Apparently all the dorsal root fibres suffer to some extent, as there is degeneration not only of those which ascend in the dorsal columns, but also of those which pass into the central grey matter to end either in the region of Clarke's column or in the ventral horns. The degeneration in all the dorsal root fibres becomes less marked as we pass from the spinal cord towards the dorsal root ganglia. These ganglia themselves are usually healthy or show slight cellular changes which are probably secondary to the degeneration in the axis cylinder process. In the lateral columns is found symmetrical degeneration of the pyramidal tracts throughout their length, the area of sclerosis being better marked in sections taken from lower levels. The direct cerebellar tract is also constantly degenerated and certainly more often affected than Gowers' tract, which sometimes escapes in cases which have not survived very long. Atrophic changes are well marked in the cells of Clarke's column, and the latter are often much reduced in number. Along with the degeneration of the various tracts there is the usual overgrowth of glial tissue, which must be regarded as a secondary phenomenon in spite of the fact that its excessive proliferation and curious irregular arrangement in so-called whorls at one time led to the belief that it was the essential pathological factor in the disease. Microscopical examination of the brain shows that the Betz cells of the Rolandic area undergo atrophy and some diminution in number, which is exactly what one would expect, considering the degeneration of the pyramidal fibres. In the cerebellum microscopical changes are by no means constant, and cannot be considered typical of Friedreich's disease. The peripheral FRIEDREICH'S ATAXIA 65 Fig. 14. Three sections from a case of Friedreich's ataxy stained by the Weigert-Pal method. 66. DEVELOPMENTAL AND FAMILIAL DISEASES nerves have sometimes been reported to be healthy, but in other cases have shown degenerative changes. If the Marchi method is used for the examination of sections, the degeneration can be traced in the nerve fibres considerably nearer to their cells of origin. For instance, degeneration can be traced in the pyramidal fibres much higher in the brain- stem and mid-brain than is shown in the Weigert-Pal sections. There is abundant evidence to show that the neuronic decay is more marked the greater the distance from the ganglion cell. On the other hand, the Marchi stain is not adapted to the display of all degenerated fibres because only those which have recently succumbed are brought into prominence by this method. The relationship of anatomical to clinical phenomena. — The symptoms and physical signs in Friedreich's disease can be closely correlated with the anatomical findings. The marked ataxy is the result of degeneration in the dorsal columns and in the cerebellar tracts. The weakness of the limbs and the trunk muscles, the absence of the abdominal reflexes, and the presence of extensor responses, are to be associated with the degeneration of the pyramidal tracts. The absence of tendon jerks is the result of decay in those fibres of the dorsal roots which form the afferent part of the reflex arc, upon which the maintenance of those jerks depends. The nystagmus is not quite so easily explained, as the mechanism of that physical sign is not yet fully understood. At the same time, we recog- nise that nystagmus is a common sign of disease of the cerebellar system. The marked scoliosis in this disease is probably de- pendent upon a certain amount of asymmetry in the progressive loss of power in the trunk muscles and on a loss of tone in the spinal muscles during the period of skeletal growth, as evi- denced by its frequency in juvenile tabes. The pes cavus is probably associated with degeneration of the pyramidal tracts, as it is also found in many cases of early amyotrophic lateral sclerosis. {d) Progressive Lenticular Degeneration. This rare and interesting disease has recently been thoroughly investigated by Kinnier Wilson, who has analysed all the cases published up till 1910, and has collected, with his own cases, twelve instances of the disease. Its characteristics, from the PROGRESSIVE LENTICULAR DEGENERATION 67 pathological standpoint, are its familial incidence and the constancy of coarse cirrhosis of the liver with degeneration of the lenticular nuclei. Of his collected cases, three occurred in one family, three in a second, two in a third, and four occurred sporadically. Aetiology. — The age of commencement of the disease varied from ten years in the youngest to twenty-six in the oldest, the average being fifteen. It is thus a disease of adolescence. It attacks either sex indiscriminately, and may run its fatal course in any period from four months to seven years. In some cases there had been an attack of jaundice before the onset of the disease, but otherwise no predisposing causes were found. Neither alcohol nor syphilis plays any role in its aetiology. Macroscopic appearances. — The only changes in the nervous system were those of primary symmetrical degeneration of the lenticular nucleus and secondary degeneration in the tracts passing from it. The degeneration chiefly affects the put amen, to a less extent the globus pallidus. The caudate nucleus and the internal capsule may be more slightly affected. All stages, from a spongy worm-eaten appearance, with some discoloration of the nucleus, to complete cavitation are found. Microscopically, the chief change is an overgrowth of neuro- glia with a multiplication of neuroglia cells which may be extremely numerous. This overgrowth of glia tissue goes on to disintegration and cavity formation, the nerve cells and fibres and the blood vessels being affected by the gliosis. The few remaining nerve cells are atrophic and stain deeply; the myelin sheaths are broken up, and numerous granular corpuscles scattered through the degenerated area are full of fatty globules. The vessels share passively in the process of disintegration, being broken up and disappearing; in the cases which Wilson examined personally, he found no evidence of endarteritis or thrombosis, the only changes being hyaline degeneration of the middle coat, dilatation of the adventitial lymphatics, and sometimes perivascular gliosis. He concludes that blockage of the arteries plays no part in the aetiology of the disease. The nerve cells of the cortex, cranial motor nuclei and anterior horns were found to have undergone only the changes due to prolonged muscular 68 DEVELOPMENTAL AND FAMILIAL DISEASES inactivity, some of them undergoing pigmentary degeneration and early chromolysis, and others staining more darkly than normal. Sometimes they showed evidence of reaction to a terminal toxaemia. The changes in the liver were those of cirrhosis, similar to the " hob-nailed " form, but even coarser. The pro- jections on the surface are described as varying in size from a threepenny-bit to a shilling in some cases, and as rather smaller in others. The amount of fibrous tissue in the portal tracts may be as great as is ever se-en in cirrhosis ; and there was evi- dence of proliferation of the liver cells and formation of new bile ducts in the cirrhotic tissue. The whole organ was yellow- ish in colour and usually somewhat diminished in size, but no ascites was found in any case. The spleen was usually firmer than normal, and sometimes enlarged. This may have been secondary to the disease of the liver. Wilson considers it probable that the changes in the brain are due to the selective action of a toxin formed in the liver or associated with the hepatic disease, the cirrhosis thus being the primary affection and the lenticular degeneration secondary. He notes that there is a familial form of " icterus neonatorum " in which yellow pigment is found to stain the lenticular nucleus and corpus Luysii deeply, the dentate and olivary nuclei and the sensory nuclei of the medulla and pons less intensely, and the cortex cerebri, caudate nuclei and optic thalami faintly, if at all. In these cases the pigment is found in the bodies of the nerve cells. It is well known that in other forms of jaundice the nervous system shows little or no pigmentation and that there is no evidence of selective action when staining does occur. (e) Huntington's Chorea. Aetiology. — Little is known about the true cause of this rare disease, which is characterised by its hereditary tendencies and the onset of symptoms in the middle period of life. Several members of a family may be affected, and direct transmission from parent to child is by no means uncommon. The disease may be inherited through either parent, and if a child escapes the next generation will probably remain unaffected. The two sexes are equally liable to inherit the disease. Sporadic cases HUNTINGDON'S CHOREA 69 are not very uncommon. A great many theories have been advanced to account for the pathological process at work, but few have any substantial foundation. As it is evident that the morbid influence at work must be of prenatal origin, some observers have assumed that the victims of the disease are born with hereditary malformations either in the nerve cells, the neuroglia, or the interstitial tissues of the cerebral hemi- spheres. On this malformation may be grafted the gross changes which are usually observed in fatal cases. Morbid anatomy. — The gross changes found in the brain are variable and include pachymeningitis, simple or haemor- rhagic, chronic leptomeningitis with adhesion of the meninges to the cortex, atrophy of the convolutions with a compensatory deepening of the sulci and increase of the cerebro-spinal fluid. Under the microscope there is always found some increase of the neuroglia and degeneration of nerve cells. Small areas of softening and disseminated foci of encephalitis are also described. Special importance has been attached to the atrophy of the cells between the first and second cortical layers. On the other hand, some authors consider that vascular changes are of primary significance and that the disease of the blood vessels results in neuroglial proliferation and other tissue changes. Perhaps the most constant changes and those which many pathologists consider to be primary are the degeneration of pyramidal cells, most marked in the psycho-motor areas, and of the smaller ganglion cells of the corpus striatum, especially in the putamen and caudate nucleus. There is corresponding degeneration of the pyramidal tracts and glial increase in the areas affected. It is possible that the variety of morbid changes described is partly due to the confusion between true hereditary chorea and cases of senile dementia associated with choreiform move- ments, in which vascular and meningeal disease would readily account for the clinical symptoms. Relationship of anatomical to clinical phenomena. — It is not very difficult to correlate the degenerative and other changes in the psycho-motor cortex with the progressive dementia which forms one of the chief features of the disease. On the other hand, the pathological basis of choreiform movements 70 DEVELOPMENTAL AND FAMILIAL DISEASES is a subject concerning which there has been much discussion, and the changes found both in this disease and in Sydenham's chorea do not afford sufficient grounds for the solution of this problem. (ii.) Neural. (/) Peroneal Atrophy (Char cot-Marie-Tooth Paralysis). History and aetiology. — This disease was first described by Charcot and Pierre Marie in 1886, and by Tooth independently in the same year. Two years later Herringham had collected a pedigree comprising twenty cases in five generations. In this family the transmission resembled that of haemophilia, in that only males were affected, and the disease was transmitted by the females. But in other families both sexes are affected, and the disease is transmitted by either parent. As a rule, males are much more frequent sufferers than females. It is essentially a hereditary familial disease, and usually starts in the first decade, but the onset may be delayed till the second or third decade. Its course is very slow and, as the paralysis remains limited to certain muscles, life is not endangered. Consequently pathological material has been very scanty, and there is no unanimity of opinion as to the essential pathological process. It is undoubtedly a disease su' generis, and has no relation to the myopathies. . The muscles affected are, first, those supplied by the peroneal nerves, especially the peroneal group; later the calf muscles and the small muscles of the hands. In some reported cases the disease has been first noticed in the hands. Atrophy does not spread to other muscles, except in a limited degree to the thighs and forearms. There is always club-foot of the pes equinus or equinovarus type, and the claw hand typical of Aran-Duchenne paralysis. Except for these there is no de- formity. Fibrillary tremor in the muscles resembling that seen in motor neuron disease and some sensory disturbances, are often present. Histology. — The nervous system is affected both on its motor and sensory side. Degeneration of the dorsal columns, especi- ally in the tract of Goll, advances pari passu with atrophy of the ventral horn cells. The spinal ganglion cells and the cells of Clarke's column also show atrophic changes. Later PERONEAL ATROPHY 71 the pyramidal tracts may become degenerated. On the other hand, in certain cases no demonstrable lesion of the spinal cord has been found. The ventral nerve roots are usually atrophied, and there is some degree of interstitial neuritis in the branches of the peroneal nerves. This appears to be the most constant and, probably, the primary morbid process of the disease. Fig. 15. Section of muscle from a case of peroneal atrophy. The change in the muscles is similar to that seen in lesions of the motor nerves. It is essentially secondary to the changes in the nerves and spinal cord. (g) Progressive Hypertrophic Interstitial Neuritis of Children. This disease was described in 1893 by Dejerine and Sottas as occurring in a brother and sister, and the details of the pathological examination of one case were given. Since then a small number of cases have been observed in which the same clinical and pathological features were present. Some of these have been isolated cases, in others another member of the 72 PROGRESSIVE HYPERTROPHIC NEURITIS family has been affected, but there has been no evidence of hereditary transmission. The disease usually commences in the first decade, sometimes in the second. It is characterised (i) by muscular wasting, starting in the muscles of the leg and spreading to those of the hands and the mouth, and later affecting the limbs more generally; (2) by sensory disorders of the polyneuritic type, along with inco-ordination in both upper and lower limbs; (3) by Argyll-Robertson pupil, mypsis and nystagmus; (4) by kyphoscoliosis and talipes equinovarus; and (5) by thickening of all the peripheral nerves, which can be felt and sometimes seen as firm cords under the skin. This last feature is most characteristic of the disease, but the clinical picture is other- wise quite constant. In spite of the Argyll-Robertson pupil, syphilis appears to play no part in the aetiology. Morbid anatomy and histology. — The swelling of the nerves is diffuse and uniform, the consistency being firmer in Iheir peripheral parts. Both ventral and dorsal spinal nerve roots are usually thickened, but in some cases only the dorsal roots are affected. Histologically, there is found a diffuse interstitial neuritis which in the more peripheral parts is characterised by dense fibrous tissue surrounding the individual nerve fibres. Nearer to the central nervous system and especially in the spinal nerve roots the interstitial tissue is more cellular, sometimes myxomatous, but otherwise presents the same arrangement. In the peripheral parts of the nerves both axon and myelin sheath have usually disappeared completely, but closer to the cord there may be some irregular remnants of myelin sheaths and degenerated axons, and it may be possible to observe skeins of new fibrils running out in the old nerve tubes. There is great increase of the cells of the sheath of Schwann and those of the interfibrillar connective tissue. Changes are also found in the spinal cord, especially degeneration of the dorsal columns affecting the dorsal root zones and the columns of Goll and Burdach. The cells of the spinal ganglia and of the ventral cornua appear normal. FAMILY PERIODIC PARALYSIS 73 (iii.) Myopathic. (h) Family Periodic Paralysis. Aetiology. — The famUial character is well established, and the disease seems to fall with equal frequency on the two sexes. The disease may be transmitted either through the father or the mother. In some cases it is associated with migraine either in the patient himself or in other members of the family. Its onset is not usually observed until after the age of six, but in certain cases the attacks undoubtedly have been present from infancy. They are very irregular, and true periodicity is seldom observed. They come on either during a period of rest after severe exercise or during a prolonged period of rest. In some cases they appear to have some relation to errors in diet. The paralysis affects chiefly the proximal muscles, especially in the lower limbs. The peripheral parts of the limb are affected along with the trunk in more severe attacks, and the muscles of the neck are sometimes paralysed. Usually the cranial nerves escape, but ptosis of the upper eyelid is not very infrequent. Although the intercostals may be paralysed, the diaphragm seems to escape, but several cases have been reported where the patient died during an attack, presumably from respiratory failure. During attacks the muscular response diminishes progres- sively and equally to the various forms of stimulation, e.g. volitional, reflex, galvanic, faradic, mechanical. During the paralysis there is neither spasticity nor loss of tone, and there are no sensory changes. The attacks pass off, leaving no im- pairment of muscular power, but when they continue for long periods some contracture may result. Certain observations have been put forward to account for the disease. Goldflam found that the urine of patients just before and during an attack had toxic properties, but this has not been confirmed by other observers. Others have found that the occurrence of vomiting in the early stages of the attack has caused the paralysis to pass off rapidly. Again, by careful dieting, and avoiding special articles of food, such as pork, cheese, beer, etc., some sufferers from this disease have been able to take violent exercise without an attack occurring. 74 MYOTONIA During an attack arterial tension is usually raised, and there may be an increase in the cardiac dulness. The absence of electrical response during the attacks indi- cates that the seat of the paralysis lies in the muscles them- selves. The nature of this paralysis is uncertain, but from the histological appearances of muscles excised during an attack, and from other facts already mentioned, it seems probable that it is associated with lymph stasis, oedema and the accumulation of toxic products in and around the muscle fibres. (i) Myotonia {Thomsen's Disease). Aetiology. — Nothing is known of the aetiology of this disease except the well-established facts of its familial character, its appearance in the earliest years of life suggesting that it is really congenital, and its tendency to affect the males of the family more than the females. Anatomy. — The general musculature is usually well developed and sometimes gives the appearance of hypertrophy. Macro- scopically, the muscles have perhaps a paler and clearer appearance than normal, but are not fatty. Microscopically, in pieces removed during life there is a very definite enlargement of the muscular fibres, which average from 50^ to 100 jLi (normal 12 jut to yo ju). The fibres are rounder than normal, the transverse striations are poorly marked and the sarcolemma nuclei are increased in number. In muscles examined after death various other changes have been described which may be due to other influences, but when removed during life the above appearances are constant. No constant changes have been found in the nerves or spinal cord, which are usually healthy. Physico-pathology . — Much work has been done on the reac- tions of myotonic muscles to electrical and other stimuli. It has long been known that the contraction produced by electrical stimulation is slower, and relaxation very much slower; the latent period, however, is not prolonged. The myotonic re- laxation is most marked after very strong, and especially after strong and prolonged, contractions, tending to pass off as the contraction is repeated. Findlay has shown that certain stimuli, such as a tendon jerk or a minimal and momentary MYOTONIA ATROPHICA 75 faradic or galvanic stimulus, produce a normal response, whereas direct percussion or a stronger or longer electrical stimulus produce a myotonic response. Voluntary movements were found to be at first slow, then gradually increasing in rate till they were quicker than those of the examiner, while involun- tary movements were never myotonic. These facts do not militate against the hypothesis that the morbid process lies in the muscle fibres, and not in the nervous mechanism of contraction. (j) Myotonia Atrophica. Aetiology. — This disease links myotonia with the myopathies. It has the same familial characters as both. The myotonia is usually slight in degree and sometimes has not been noted till late adolescence, but in most cases has been present since infancy. The association with hereditary cataract in some cases seems to be more than a fortuitous one. The disease corresponds with both myotonia and myopathy in attacking males more frequently than females. Histology. — In cases where pieces of the affected muscles have been examined during life the changes described under myotonia and under myopathy (q.v.) are present in varying degree. The muscle fibres tend to be larger and more rounded than normal, with an increase in the number of sarcolemma nuclei and of the interstitial connective tissue. In one case some degeneration of the dorsal columns of the cord in the lumbar region was found. (k) Myopathy. Aetiology. — In this disease the familial character is very well marked, and hereditary transmission has been frequently noted. Males are more frequently attacked than females, especially in the pseudo-hypertrophic form. In this the trans- mission may be through unaffected females, as in haemophilia and congenital night blindness. In certain cases the disease seems to have been started by some trauma or febrile illness, but as a rule there is no obvious cause for its onset, which takes place at about the same age in each member of a family who is affected. Babinski and Onanoff have found that the muscles most differentiated in a five-month foetus are those in which myopathic change is most liable to occur, and suggest that 76 DEVELOPMENTAL AND FAMILIAL DISEASES these muscles have less vitality than those which are laid down later. Macroscopic. — The disease is always symmetrical in its attack on muscles, and tends to affect the muscles at the roots of the limbs more than those more distal. Pseudo-hypertrophy is most commonly seen in the deltoid, supra- and infra-spinati, and in the gastrocnemii. The hyper trophied muscles are firmer than normal, but less hard on contraction. On section they appear yellowish, from increase of fatty tissue, or they may give the appearance of a mass of fat, with little or no suggestion of their original muscular character. Histology. — Although divided into several clinical types, myopathy is a pathological entity, and the microscopical appearances are the same in all forms; in some one change predominates, in others another, but all varieties of the appearances to be described are seen in almost any case of the disease. Apparently the earliest change, and that which is most commonly found in pieces of muscle excised during the earliest stages of the disease, is swelling of some of the muscle fibres and increase in the sarcolemma nuclei. The fibres may measure from 150 /u to 230 ^ across, and are more rounded in outline and more hyaline in appearance, with less well- marked striation. This appearance has been ascribed to swelling of the sarcoplasm. At a somewhat later stage in the disease there is an increase in thickness of the connective- tissue septa between the various fibres and a tendency for fat to be laid down here. In the pseudo- hypertrophic form the swollen muscles show a very great deposition of fat between the fibres, which are often themselves hypertrophied. Except in the very earliest stages, in addition to large fibres, one finds a greater or smaller number of very small ones, and it has been suggested by Marinesco and others that these often arise from a splitting up of the hypertrophied fibres into smaller bundles of " sarcostyles." Thus, in longitudinal sections a muscle fibre is sometimes seen to break up into two or three segments, each of which may be surrounded by a large number of sarcolemma nuclei. Hyaline degeneration and vacuolation of the hypertrophied muscle fibres are occasionally seen. It is thus clear that this hypertrophy is a morbid change with a MYOPATHY 77 p^^i^p^- 4, '^^: ^- - ' Fig. i6. Muscle in the pseudo-hypertrophic form of myopathy. y% AMYOTONIA CONGENITA great tendency to further degeneration. The muscle fibres are also seen to undergo a retrogressive change into fibrous tissue, especially in the neighbourhood of the tendinous attachments. It is doubtful whether the encroachment of tendinous fibrous tissue which undoubtedly takes place at both ends of an affected muscle is due to this retrograde metaplasia of muscle fibres or to their atrophy and a simultaneous increase of the interstitial connective tissue. Probably both these factors are at work. Eventually the picture presented is that of a few scattered fibres, some larger than normal, others small, atrophic, and irregularly shrunken, separated by a large amount of fibrous and fatty tissue. Morbid changes have been described in the motor nerve endings, but the muscle spindles have been found intact even in advanced stages of the disease. Changes in the intra- muscular blood vessels are constant. Their walls are thickened and their lumen narrowed by connective-tissue proliferation around them. Thromboses have been described, and the adventitial lymph spaces may be packed with newly-formed cells. Changes in the sympathetic nerves have also been noted. It is thus possible that the disease is primarily one of the sympathetic system, and that inadequacy of blood supply to a muscular area may determine the changes in the muscle fibres. But the bulk of opinion favours the hypothesis that the muscles themselves are primarily affected, and that the interstitial and vascular changes are secondary. (/) Amyotonia Congenita. This rare disease was first described by Oppenheim, who gave it the name of myatonia, but as this name is apt to lead to confusion, English writers have adopted the name amyotonia. The disease is congenital, but seems to have no tendency to run in families. Sylvestre in 1909 reported the occurrence of two cases in a family, one of which developed Erb's type of myopathy at the age of sixteen, but this appears to be the only recorded instance of the kind, and a sufficiently large number of cases of the disease have been seen to make it fairly certain that it has no familial characters. On the other hand, a considerable number of the patients develop definite myopathy DEVELOPMENTAL AND FAMILIAL DISEASES 79 at a later period of childhood, and both on this account and owing to the similarity in the histological appearances the disease is usually classed along with the myopathies. Macroscopically, the limbs are slender and the musculature very soft. The hands and feet are peculiarly long and delicate. On section the muscles are yellowish, and are distinguished with difficulty from the fat which lies around and between the muscle bundles. Microscopically, the most obvious change consists in the small size of the majority of the fibres, which only measure from 7 /^ to 12 /^ in diameter. But among them there are seen a few fibres of giant size which measure up to 140 /n. These usually lie singly, and are of more rounded shape than normal. They often show central vacuolation, and some of them tend to split longitudinally. They resemble in every way the large fibres seen in myopathic muscles, and like them are less numerous in cases of longer standing. The changes in the central nervous system are equally definite, but do not appear to be primary. These consist in extreme thinness of the ventral nerve roots and diminution in the number of motor cells in the ventral horns. The nerve roots contain fewer fibres than normal, and these are thin and poorly myelinated. These appearances seem to be accounted for by the presence of disease of the muscles both before birth and during the first few months of extra- uterine life. REFERENCES Prenatal Diseases and Abnormalities. Anton, G. : In Flatau, Jacobsohn and Minor, Handbuch der path. Anat. des Nervensy stems, 1904. Ballantyne, J. W.: Manual of Antenatal Pathology and Hygiene: The Embryo. Edinburgh, 1904. Gombault and Riche: Manuel d' histologie pathologique (Cornil and Ranvier), vol. iii., 1907. Birth Injuries. Taylor, James: Allbutt and Rolleston, System of Medicine, vol. vii., 1910. Familial and Congenital Diseases. (a) Amaurotic Family Idiocy. Carlyll, H., and Mott, F. W. : Proc. Roy. Soc. of Med., 1911. Path., pp. 147-198. Hume, W. E.: Rev. Neur. and Psych., 1914, vol. xii., p. 281. Russell, J. S. Risien: Allbutt and Rolleston, System of Medicine, 1910, vol. viii., p. 468. 8o DEVELOPMENTAL AND FAMILIAL DISEASES {b) Werdnig -Hoffmann Paralysis. Batten, F. E.: Brain. 1911, vol. xxxiii., p. 433. Batten, F. E., and Holmes, G. : Brain, 1912-13, vol. xxxv., p. 38. Krabbe, K. H.: Brain, 1920, vol. xliii., p. 166. (c) Friedreich's Ataxia. Holmes, Gordon: Allbutt and RoUeston, System of Medicine, 1910, vol. vii., P- 770- {d) Progressive Lenticular Degeneration. Wilson, S. A. Kinnier: Brain, 1912, vol. xxxiv., p. 296. {e) Huntington's Chorea. Marie, P., and Lhermitte, J.: Rev. Neur., 1912, vol. xxiv., pp. 40-45. Russell, J. S. Risien: Allbutt and Rolleston, System of Medicine, 1910, vol. viii., p. 548. (/) Peroneal Atrophy. Batten, F. E.: Allbutt and Rolleston, System of Medicine, 1910, vol. vii. p. 71. Charcot et Marie: Rev. de MSd., 1886, vol. vi., p. 97. Tooth, H. H.: Brain, 1888, vol. x., p. 243. (g) Progressive Hypertrophic Interstitial Neuritis of Children. Dejerine et Thomas: Nouv. Icon, de la SalpSt., 1906, vol. xix. p. 477. Durante, G. : Cornil et Ranvier, Manuel d' histologic pathologique, vol. iii., Paris, 1907. ScHALLER, W. F.: Arch. int. Med., 1912, vol. x., p. 399. {h) Family Periodic Paralysis. Buzzard, E. F.: Lancet. 1901, II., 1565. Clarke J. Michell: Allbutt and Rolleston, System of Medicine, 1910, vol. vii., p. 65. {i, j, k) Myotonia, Myotonia Atrophica, Myopathy. Batten, F. E.: Quarterly Journal of Med., 1910, vol. iii., p. 313. Collier, J.: Allbutt and Rolleston, System of Medicine, 1910, vol. vii., p. 19. Findlay: Qtiart. Journ. of Med.. 1912, p. 495. Oppenheim": Textbook of Nerv. Disease (trans. A. Bruce), 191 1, p. 255. White, W. H.: Allbutt and Rolleston, System of Medicine, 1910, vol. vii., P- 25- CHAPTER III INJURIES TO THE NERVOUS SYSTEM I. Injuries to the Brain and its Coverings. Anatomical considerations. — The skull, after infancy, may be regarded as an almost rigid bony box, with one large opening, the foramen magnum, and several smaller ones, of which the most important are the jugular foramen, the sphenoidal fissure, and the frontal, occipital, and mastoid openings for emissary veins. Any increase of the intracranial contents is therefore impossible, but the bulk of any one constituent may increase if at the same time there is a diminution of another. Thus, an increase of the size of the brain by tumour, abscess, oedema, etc., may be compensated for by a diminution of the amount of cerebro-spinal fluid or of blood. The amount of blood sup- plied by the arteries cannot vary to any great extent, as the vasomotor centre in the medulla immediately reacts to any diminution in the blood supply to the brain by raising the general arterial tension and thus causing the intracranial vessels to be supplied at a higher pressure. (The action of the vasomotor centre appears to be regulated by the reaction of the fluids in which it is bathed, a very slight rise in H-ion concentration sufficing to bring it into activity. It may thus be stimulated by (i) a diminution of the blood supply to the medulla, (2) a deficiency in oxygen or increase in carbon dioxide in the blood, or (3) by a general acidosis.) It follows that an increase in the size of the brain cannot be compensated for by diminution in the amount of blood supplied to the inside of the cranium. The cerebro-spinal fluid is therefore driven out, the ventricles become more slit-like, the sulci of the brain narrowed, and the cortex of the brain more closely applied to the cranial vault. There is also an increase in the amount of fluid in the meshes of the spinal 8i , 6 82 INJURIES TO THE NERVOUS SYSTEM arachnoid, and probably some slight escape along the lymphatics of the sheaths of the spinal nerves. But, unfortunately, two vicious circles are liable to occur in this connection, to which attention has already been drawn (p. 48). With regard to the anatomical relationship between the brain and the skull, it must be remembered that the brain proper is nowhere in contact with the dura mater, but is separated from it everywhere by the water cushion formed by the pia-arachnoid. The large cisternae at the base of the brain form a water- bed on which the brain rests, and protect the vulnerable cranial nerves and vessels. The brain is also protected from undue movement inside the cranium by the falx cerebri and tentorium cerebelli, which prevent lateral and vertical movement. Traumata produce lesions of the brain either by causing fractures of the skull or by jarring the brain inside the skull and thereby damaging its tissues; the former gives rise to wounds or lacerations of the brain, the latter to con- cussion. Concussion may result either from direct or indirect violence. In the first case, it is most often due to blows from clubs, falling stones, bricks, timber, etc., or to falls or kicks on the head. As a result of indirect violence it may be due to jars transmitted from the spine, as in landing on the heels in the erect position with the knees straight, or landing in a sitting position from a height; or to blows on the chin whereby the shock is trans- mitted to the middle cranial fossa through the ramus of the lower jaw. It may also be caused by explosions. It varies greatly in degree from that which follows but a momentary loss of consciousness to that associated with prolonged uncon- sciousness, retrograde amnesia, and mental confusion. True concussion may be associated with fractures of the skull, especially fractures of the base of the skull, but although it is often associated with some subarachnoid haemorrhage, it should be carefully distinguished from the results of com- pression due to bleeding from meningeal vessels into the epidural or subdural space. The effects on the brain may be minimal: either no macro- scopic lesion is observed, or there may be slight bruising of the cortex over a larger or smaller area at the site of the CONCUSSION OF THE BRAIN 83 contusion, or on the opposite surface of the brain. This is spoken of as the contrecoup effect. It is probably most often seen at the frontal poles from bruises on the occiput, or vice versa, and when we consider the way in which the cranial cavity is divided up by dural partitions, this is not difficult to under- stand. Subarachnoid haemorrhage is usually of small amount unless the large veins and venous lacunae near the vertex are torn, in which case the escape of blood may be large enough to give rise to serious pressure effects. The arteries of the base are seldom ruptured, but the subcortical vessels may rupture and give rise to haemorrhages of varying size. These are most frequently found in the frontal region, and are most likely to occur in subjects with degenerated arteries. Later effects are the formation of areas of softening either cortical or subcortical, or around the iter of Sylvius. The latter is a common situation, probably because this part of the brain is often bruised by the sharp edge of the tentorium cerebelli. A curious sequel is the occurrence of haemorrhages in the areas of softening days, weeks, or it may be months after the original injury. Such cases are not very uncommon in the literature, and in a large proportion of them the haemorrhage is found in the region of the iter of Sylvius. Another unusual result of concussion is internal hydro- cephalus, the causation of which has already been discussed (p. 47). Cases are occasionally seen where such hydro- cephalus is associated with a cortical softening extending through to the ventricle, and where the ventricular fluid under increasing pressure ruptures the softened brain substance and escapes into the subdural space. Damage to cranial nerves often results from concussion (p. 94). Healing of the bruised area in the cortex and meninges takes place later with the formation of a scar in which an area of thickened meninges is more or less closely bound down to sclerosed cortex. This may be a focus of increased irritability, and may give rise to fits of the Jacksonian type. But it is probable that another factor than the presence of the scar is required to cause these, as only a certain percentage of patients with scarred brains develop epilepsy. The extent of the local softenings and the amount of damage which the brain suffers 84 INJURIES TO THE NERVOUS SYSTEM from any traumatism is to a great degree dependent on the condition of the arteries, and injuries which in youth would have left no ill effects may, in later life, lead to gross loss of function. Wounds of the Brain. In civil life wounds of the brain are most often caused by blows with pointed implements, such as sharp stones; more rarely they are due to falls on the head or blows from large, heavy objects. When such blows fracture and indent the skull, the shock is usually sufficient to cause instant death. There are, however, many cases where, without any fracture of the outer table, a small splinter breaks off from the inner table and becomes imbedded in the brain. In such cases the general concussion effects usually overbalance those of the local lesion. The chief points wherein fractures of the skull differ from concussions are their tendency to cause meningeal haemorrhage at the site of the fracture and the frequency with which they are followed by septic infection. The latter may occur either directly where the track of the wound passes straight through the skull, or indirectly, as in fractures of the nasal and ethmoid bones, or fractures of the base of the skull, when a channel may be established between the outside air and the subdural space through the bony air sinuses. Of wounds of the brain in war, the majority are either immediately fatal or cause death within a few hours. But even when there has been a large fracture of the skull and very considerable damage to the brain tissue, a surprisingly large number of patients live for days, weeks or months, and either survive with some loss of cerebral faculties or ultimately succumb to septic infection. The majority of the wounds which were seen at the base and home hospitals were either those caused by small fragments of shrapnel, which had penetrated the skull, or by bullets which, glancing off the skull, had caused a " gutter " fracture. In the former cases there was more or less bruising of brain tissue all along the track of the missile, and sometimes areas of haemorrhage where smaller or larger vessels had been cut. The area of bruising was usually not very extensive, and might be considerably larger on the cortex, where fragments of inner table had been driven WOUNDS OF THE BRAIN 85 in, than deeper in the brain. As a rule bruising did not extend more than i cm. from the track of the missile. In gutter fractures of the skull the effects are similar to those caused by blows on the head, with certain differences consequent on the greater velocity of the missile, which causes the fragments of the inner table to be driven more deeply into the brain, and on the greater liability to septic infection which is shown by the wounds of war. Not only may the brain be injured by fragments of bone, but, even in the absence of splintering of bone, both direct and contrecoup bruising may be very considerable. As has already been noted, such areas of bruised and softened brain may extend down to the lateral ventricles. They form funnel-shaped areas of i to 3 cm. in diameter, extending for a similar depth from the cortex. In every case within twenty-four hours of the wound some oedema of the brain tissues around the damaged area occurs and leads to enlargement of the cerebral substance. As a consequence the intracranial pressure rises, and if there is any opening in the skull, softened brain matter is extruded through it. This increase in intracranial pressure after a wound of the brain has been thought by some to be chiefly due to haemorrhage, but although bleeding may be a pre- dominant factor, it is by no means necessarily so, and many wounded brains have been examined post mortem in which there was no evidence of any but petechial haemorrhages. This oedema may bring about a fatal result or it may subside, leaving no trace, but when it occurs concurrently with septic infection it is more easy for organisms to penetrate through the brain tissue and cause encephalitis. Septic infection of wounds of the brain may start (i) at the skin edges, (2) in the middle of the track, where, perhaps, some hair or dirt carried in with the missile has been left behind, or (3) around the missile, when this remains embedded in the brain. The nature of the reaction varies with the type of organism present. Streptococci and pneumococci are most liable to cause rapidly spreading meningitis, or diffuse encepha- litis; infection with the gas gangrene bacilli usually causes a very acute and rapidly fatal encephalitis, whereas other organisms may cause merely a local infection, resulting eventually in abscess or a slowly progressive meningitis. 86 INJURIES TO THE NERVOUS SYSTEM Streptothrix infections are usually mild. When the ventricles are infected the fatal course is usually rapid, as also when a purulent meningitis occurs in the posterior cranial fossa, whereas a superficial abscess, or area of meningitis over the vertex, may remain circumscribed and give rise to no dangerous symptoms for weeks or months after the wound. Infection rarely passes through an intact dura; but this accident may happen, especially in the neighbourhood of the large venous sinuses, where the emissary veins are more numerous. In this case the sequence of events is probably first an infiltration of the dura mater, starting at the mouth of a venous channel, then a purulent collection in the subdural space, and lastly a purulent leptomeningitis. It is probable that in the absence of sepsis the presence of a foreign body in the brain causes only a minimal irritation, as it becomes surrounded with a layer of neuroglial sclerosis and is thus kept from direct contact with neurons. An exception to this rule is afforded by metal bodies of considerable size and weight which are apt to shift their position and to set up an aseptic form of encephalitis, giving rise to clinical signs and symptoms identical with those of cerebral abscess. Wounds may injure the brain or its coverings indirectly. Thus, a wound involving the common carotid or internal carotid artery may obstruct the blood supply to the parts of the brain supplied by the middle and anterior cerebral arteries. This may be followed by hemiplegia or hemiparesis, though not necessarily. In such cases the anterior two-thirds of the cerebral cortex of the affected side have been found congested, and pinker than on the normal side, a condition of " diffuse infarction." Wounds over the superior longitudinal sinus frequently lead to sinus thrombosis, and by obstructing the venous return cause oedema and temporary loss of function of the cortex. Severe wounds involving the nasal cavities in which there is a considerable degree of sepsis not infrequently give rise to generalised cerebro-spinal meningitis. This is usually due to the streptococcus or pneumococcus, which probably pass through the cribriform plate along the sheaths of the olfactory nerves. INJURIES TO THE NERVOUS SYSTEM 87 2. Injuries to the Spinal Cord. Anatomical considerations. — The spinal cord is suspended within the bony walls of the spinal canal by very loose and indirect attachments. The spinal dura mater, which is thinner than that which lines the cranial vault, is not firmly applied to the bony canal, but is separated from it by the epidural space, which is filled with loose areolar tissue and a plexus of veins. The dura is fixed to the spinal canal above at the foramen magnum. It extends down as a loose sheath to the second sacral vertebra, whence it is prolonged as a fibrous band, the filum terminale externum, to the periosteum lining the lower end of the sacral canal. It is loosely attached along all its extent to the posterior common ligament of the vertebrae, this connection being firmest in the upper cervical region, especially opposite the body of the axis, and also in the lower lumbar and sacral region. The sheath of dura mater is also loosely anchored in the centre of the vertebral canal by the spinal nerve roots, around which it sends prolongations which are attached by areolar tissue to the walls of the inter- vertebral foramina. It is thus evident that in the lower cervical and thoracic region the connections between the dural sheath and the bony canal in which it lies are by no means firm. The cord is suspended within this sheath both by the spinal nerves and by the ligamentum denticulatum, which may be regarded as a thickening of the arachnoid membrane along the sides of the cord, with attachments to the dura mater between the points of exit of the spinal nerves. Between the dura and the cord lies the cushion of the subarachnoid space, normally filled with cerebro-spinal fluid. The subdural space is normally non-existent, as the membranes which form its outer and inner boundaries are closely applied to one another. It. must also be remembered that the pia mater over the cord is thicker and firmer than that over the brain, and more adherent to the nervous tissue. It is strengthened by longitudinal strands of fibrous tissue, which are specially firm over the ventral median fissure. The cord is subject both to concussion and to wounding or laceration. It may also be compressed in fracture-disloca- 88 INJURIES TO THE NJERVOUS SYSTEM tion of the spine, and is specially liable to injury in caisson disease. These various lesions will be considered in order. Concussion. — The forms of trauma which most frequently cause spinal concussion are (i) blows on the back, especially over the thoracic region, causing abnormal straightening of the spine; (2) falls on the head, which cause sudden extreme backward flexion of the neck ; and (3) wounds by bullets and shell fragments which strike or pierce laterally the spinous processes. In the first two cases the normal thoracic curve of the spine is straightened out, the spinous processes and laminae forming a fulcrum, and thus the vertebral bodies are drawn away from each other and the intervertebral discs stretched or torn across. The articular processes are sometimes found fractured in such cases, with little or no displacement. In case (3) the momentary distortion is a lateral one, with rotation around a fixed point supplied by the vertebral bodies and intervertebral discs. If any bruising happens in this case it affects the side of the cord from which the missile came, whereas in the first two cases the dorsal surface of the cord is primarily affected. Macroscopically, the cord, a few days after such an injury, is swollen at the level of the lesion and for a varying distance, usually not exceeding a couple of segments above and below it. The pia-arachnoid may show some evidence of bruising, or, in cases of severe lesion, may be distended by yellowish or blood- stained fluid. Unless the dura mater is pierced by bony fragments, the pia is usually intact. On section at the level of the lesion the cord may present all degrees of change, from complete disintegration into a pultaceous custard-like or sanious material, to slight oedema, with small scattered haemorrhages; these may, however, be completely wanting. A not uncommon, and very striking, change is the formation of cylindrical cavities, of one to two millimetres in diameter, most usually in the ventral part of one or both dorsal columns or in a dorsal horn. They may in slighter cases extend longitudinally upwards and downwards from the level of the lesion for a distance of two or three segments in either direction ; and in severe lesions of the cord, with disintegration of its substance at the level of the lesion, they may be found either CONCUSSION OF THE SPINAL CORD 89 below this or both above and below. Usually they do not commence in the softened area, but in healthier tissue a few millimetres away from it. They are filled with a greyish or brownish gelatinous material, and are thus easily visible to the naked eye. Examined microscopically their contents are seen to be largely blood serum, with an admixture of granular detritus which is greatest near to the site of the chief lesion, and of granular " scavenging " cells, which also line their walls. Blood cells may be found in them, but it is evident that they are not caused by haemorrhage. That they are formed under pressure is clear from the concentric arrangement of the tissues immediately round them, and from the way in which they separate the nervous tissues without destroying them. Various theories have been put forward to account for these cavities, such as the stasis of lymph currents, the formation of exudates with a poisonous action on the nervous elements, or vascular thrombosis with resulting necrosis. When we consider that the cord is always swollen and oedematous at the level of the lesion, and that the pia mater is a firm membrane and is almost always intact in such cases, it is not difficult to imagine that the serum may, in trying to find a way of escape, track up and down the cord in the planes of least resistance: it is more difficult to formulate a reason why the situation is so constantly in the ventral portions of the dorsal columns, but this appears to be an area of comparatively poor blood supply, and thus may form a locus minor is resistantiae. Microscopically, it is found that the chief incidence of the lesion falls on the axons and their myelin sheaths. In the more softened areas both of these are completely disintegrated, leaving a glial network containing nothing but granules of fat contained in granular corpuscles. Where there is less destruc- tion of tissue the axons show a considerable degree of swelling which may be diffuse, fusiform or moniliform, so that they approach the size of a normal myelin sheath. Sometimes they break up into their constituent fibrils. Their sheaths may be thinned out or broken up into droplets, and the glial space containing the neuron may be distended. Small areas with these appearances are scattered irregularly in the neigh- bourhood of the lesion, often affecting the outer layers of the cord, but their position and the fact that they are separated 90 INJURIES TO THE NERVOUS SYSTEM by more or less normal tissue suggest that they are not due to the direct influence of the trauma, but more probably to the effect of vibration in the cord. The nerve cells show a varying amount of change of a chromolytic type, which is probably secondary to the lesions of the axons. In addition to the changes in the neurons, oedema of the tissues is a constant feature. It is seen as a distension of the perivascular spaces and a general separation of the neurons, with, at times, the formation of small cylindrical cavities which run longitudinally, usually near the surface of the cord and especially in the dorsal columns. A glial reaction follows at a later period, but probably does not commence until after the first week or two. Haemorrhages are frequently seen, usually petechial in size, and more often in the grey than in the white matter. When they are larger and dominate the picture the condition is described as " haematomyelia," which will be described in the next chapter. As seen in the wounds of war, concussion of the cord is not often associated with gross intramedullary haemorrhage, probably because the majority of soldiers are young men with healthy arteries. Suppuration of the cord at the site of the lesion is infrequent, even when septic meningitis is present, so long as the pia mater is intact. It is unusual in such cases to find polymorphonuclear leucocytes anywhere except in the sheaths of the vessels in the dorso- median fissure. Wounds and injuries of the spinal cord may be caused in a great variety of ways. First, the cord may be cut by knife stabs, or wounded by bullets or fragments of shell-casing, or by spicules of bone broken off in their passage. Secondly, and more frequently, it is injured by fractures and dislocations of the vertebrae caused by falls or severe blows on the back, or by the weight of fallen masonry, stones, etc., under which the spine bends and breaks. Thirdly, a very slight trauma is sufficient to produce dislocation of the vertebrae in diseases or malforma- tions of the spine. Thus, in Pott's disease or neoplasm of the vertebral bodies, the sudden onset of paraplegia maybe brought about by some small accident. There are cases of imperfect ossification of the odontoid process of the axis where this has become broken off by a very slight jar or sudden voluntary FRACTURE-DISLOCATION OF THE SPINE 91 movement of the head, and the atlas has slipped forward on the axis, bruising and compressing the cord. The first of these varieties of trauma differs from the others in causing a wound which lacerates both dura and pia mater, and has a channel leading from the cord to the outside air. Owing to this, septic myelitis is far more common in this than in any of the other varieties of trauma. Also because of the drainage through the torn pia mater oedema of the cord is not so frequent. Lateral hemisection of the cord (producing the Brown-Sequard syndrome) is more common in knife stabs than in any other form of injury, but it is also frequently caused by gunshot wounds. In knife stabs, as has been pointed out by Head, it is most usual for both dorsal columns to be divided, with otherwise a fairly exact hemisection. Dislocations and fractures of the vertebrae, on the other hand, usually leave the dura and pia mater intact, and the condition then resembles the results of the severer forms of concussion, with the difference that the nerve roots are more apt to be compressed either directly by bony pressure in the intervertebral foramina or by the clotting of effused blood. The haemorrhage into the subdural space may be considerable, and the resulting paralysis fairly extensive. Haemorrhage outside the dura is less likely to produce of itself paralytic symptoms, as the intervertebral foramina provide a ready means of escape for blood. In the majority of cases of fracture or dislocation of the spine the cord is injured at the time of the accident, but oedema and haemorrhage coming on within the next few days may aggravate the damage done. Probably after this period, in the majority of cases, the condition of the cord is not aggra- vated by the bony compression, but, on the other hand, the canal may be so definitely narrowed at the level of the lesion as to hinder what repair of the cord is possible. In these cases rapid amelioration of symptoms results from the removal of the pressure by surgical interference. In the present state of our knowledge it is a little difficult to account satisfactorily for this. Several factors are involved, to all of which in greater or less degree the delay in the return of function is attributable. There may be obstruction to the blood supply of the cord at 'the level of the lesion, and this is the more serious owing to the 92 INJURIES TO THE SPINAL CORD ^rSp /■# N P'iG. 17. Sections stained by the Marchi method illustrating the ascending degeneration the spinal cord following a fracture-dislocation in the mid-thoracic region. CAISSON DISEASE 93 fact that longitudinal anastomosis between the arteries of the cord is completely absent within the substance of the cord, although it exists to some extent in the pia mater. Further, in all cases of compression of the cord there exists some oedema of the cord, both above and below the lesion. In Weigert-Pal stained sections of the cord in such cases, we are struck with the small degree of ascending and descending degeneration above and below the lesion, as compared with the almost complete absence of healthy myelin in the compressed segment. It appears, therefore, that many of the swollen and fragmented axis cylinders may be capable of preserving the nutrition of the distal parts of the neuron, and presumably may also be capable of resuming their function when the con- ditions at the site of the lesion are improved. Histologically , there is little to be added to the description of the severer degrees of concussion changes. Where sepsis and septic myelitis occur, there is an invasion of the cord by polymorpho- nuclear cells and mononuclear cells of endothelial and con- nective-tissue origin. These are at first seen in the vicinity of the larger blood vessels, but when the myelitis is more diffuse they are scattered everywhere in the cord. In the slighter septic cases when recovery takes place, the repair is largely accomplished by connective tissue, whereas in the aseptic cases the new-formed tissue is chiefly of neuroglial origin. Caisson Disease {Diver's Paralysis). Aetiology. — This disease has been proved, both clinically and experimentally, to be due to a too rapid change from the air pressure in the caisson to that of the outside atmosphere. The pressure in caissons or inside a diving dress may be any- thing from one to five additional atmospheres, and at such pressures a considerable quantity of nitrogen is absorbed both by the blood and by the tissues, and is released as bubbles of gas when the pressure is reduced too quickly. In most parts of the body this does no permanent harm, as the gas is rapidly absorbed and passed out of the lungs, but in the nervous system, especially in the spinal cord, the effects are severe and may be more or less permanent. Schrotter's view of the pathology of the disease is that the bubbles are set free in the blood vessels, and thus produce gas 94 INJURIES TO THE NERVOUS SYSTEM embolism and infarction. The reason why the white matter of the cord is more affected than the grey matter Hes, according to this view, in its lower vascularity, and, in consequence, its greater vulnerability by vascular obstruction. On the other hand, Vernon has shown that myelin in common with other fatty tissues in the body has a much greater power of absorbing nitrogen than the body fluids generally. As a consequence, the brain and spinal cord would absorb very much more nitrogen in virtue of their high myelin content, and when the pressure is lowered the gas would be set free, not in the blood vessels, but in and around the neurons, thus directly breaking up the myelin sheaths and causing pressure on the axons. This view gives an explanation why the white matter of the cord is especially affected, and reconciles the changes in the cord with the absence of infarctions in other organs. Histologically , the disease is characterised by multiple small softenings in the cord, chiefly in the white matter. At the centre of the softened area the tissues are broken up, and very little of the original structure remains. Towards the outer parts the myelin is fragmented and degenerated, and the axis cylinders are often greatly swollen, sometimes to twenty times their natural size. There is little vascular reaction, and the processes of repair are mainly undertaken by neuroglial elements. 3. Injuries to Nerves. The peripheral nerves are less protected from injury than the brain and spinal cord — in fact, some of them lie in very exposed positions; but the limited paralysis caused, and the probability of recovery, make lesions to nerves much less im- portant than lesions of similar extent in the spinal cord. In a large proportion of cases the function of the nerve is not com- pletely lost, and even where complete severance of the nerve has occurred some degree of function may be restored within a year if the conditions are favourable. This does not hold good for cases of injury of a sensory nerve between its ganglion and the central nervous system, as the fibres in that case are centripetal, and degenerate into the tracts of the cord or brain where no regeneration is known to take place. Thus, deafness due to lesion of the auditory nerve in the internal auditory canal cannot be cured. The olfactory and optic nerves hold a similar INJURIES TO NERVES 95 relation to their cells of origin, though they are not correctly spoken of as peripheral nerves, as in structure and development they are very different. In certain birth paralyses, which will be mentioned later, the lesion may affect the spinal nerve roots between the dorsal root ganglion and the cord, and in these cases, although theoretically the motor root may recover, the sensory cannot ; as a matter of fact, recovery of motor power is exceptional. X '"^^op Fig. 18. Jj peripheral nerve undergoing degeneration as the result of pressure, stained by the Marchi method. Injuries to peripheral nerves are most usefully classified from a pathological standpoint into those in which the nerve is injured by blow^ on it or by damage to neighbouring tissues in such a way that its sheath is not ruptured, and those in which the whole nerve or a part of it is cut or torn across. In the former case there will be some effusion of blood or lymph into and around the sheath of the nerve. The nerve fibres may be injured to a greater or less degree. In the slighter cases the axis cylinders are relatively intact, although there Fig. 19. {a) A low-power photograph of an excised portion of nerve which had undergone changes of a fibrotic character following the passage of a bullet through the tissues in its immediate vicinity. Photograph of sections taken (6) just above and (c) below the site of injury stained by the Bielschowsky method. In (c) there are very few axis cyUnders remaining. INJURIES TO NERVES 97 may be a considerable degree of damage to the myelin sheaths, many of which break up into fatty substances and are absorbed. The resulting paralysis will then pass off completely in the course of days or weeks, unless the effusion of blood or lymph within the sheath of the nerve or the resulting fibrous cicatrix interferes with the reparative processes. In more severe cases some of the axis cylinders may be so damaged that they become unable to carry on the nutrition of the peripheral part of the nerve, and Wallerian degeneration results. In some cases the whole nerve is affected in this way, in others only a smaller or larger proportion of the fibres. In the latter case, return of function takes place in two stages, between which there is an interval of months or years, as the less damaged fibres resume function rapidly, whereas those more severely injured give rise to new fibres which must grow from the level of the lesion to the nerve endings in skin, muscle and tendon before function can be restored. When the nerve and its sheath are cut or torn across, there must be rupture of the axis cylinders with consequent Wallerian degeneration in the peripheral segment of the nerve. The pro- cesses of nerve regeneration then take place with the formation of end bulbs on the upper and lower cut ends, which fuse to- gether if the ends are kept in contact, and conduct the young neuro-fibrils to the old neurolemma sheaths of the peripheral segment. When the ends are separated by a gap, some of the out- growing nerve-fibre processes may leave the bulb on the proximal end and pass along the planes of the intervening fibrous tissue to reach the bulb on the lower end, but the number of fibres which pass across in this way is rarely large enough to lead to useful return of function. Although the formation of end bulbs usually occurs, they may be absent in cases where sepsis interferes with the normal processes of recovery and checks the proliferation of the nervous elements. Sepsis may even cause an ascending neuritis in the proximal end of the nerve, which may extend up the nerve for several inches from the level of the lesion. Evidence of this is afforded histologically by degeneration of the axis cylinder and myelin sheath, and by inflammatory changes in the peri- and endo-neurium. 7 '^^^ Fig. 20. Sections of ulnar nerve seven months after a wound that severed it. a, A longitudinal section immediately above the wound, showing a few of the original well-myelinated fibres, with many finely myelinated new fibres running in all directions, b, Centre of neuroma bulb separating ends. c. Lower end. A very few myelinated fibres have regained the nerve tracts (neurolemma channels) of the lower end. Stained by the Weigert-Pal method. INJURIES TO NERVES 99 Fig. 21. Sections of an ulnar nerve five months after section by wound, stained by Biel- schowsky's method, a, at upper end of neuroma just above level of injury. b, in the middle of the neuroma, c. Lower end, a few fibres having regained the neurolemmal tracts of the nerve beyond the injury. 100 INJURIES TO CRANIAL NERVES Where the nerve is liable to long-continued pressure or irritation, a condition of interstitial neuritis may be produced. In this case the degeneration in the nerve fibres is secondary to overgrowth of the fibrous tissue forming the peri- and endo- neurium. In the earlier stages the condition is one of loss of function without destruction of the axis cylinders, and recovery is usually rapid and complete. But in some cases the fibrous tissue becomes so dense and firm that it is very slowly absorbed, even under the most favourable conditions, and the functional restoration of the nerve is correspondingly slow. The cranial nerves are protected from direct injury in their intracranial course, but are liable to injury both in their bony canals and in their extracranial portion. They may also be pressed on by effused blood in the meningeal haemorrhage resulting from injuries. The nerves of special sense (I., 11. , VIII., and the pars intermedia of VII.) are particularly liable to such injuries, as is also the facial nerve, both in its bony canal and outside the skull. The olfactory nerves are often torn across in severe blows or falls on the head, resulting in complete loss of the sense of smell. The optic nerves may be directly injured by puncture wounds of the orbit or by frag- ments of metal in steel works, or pieces of shell or bomb-casing. The chiasma or the optic nerve or tract may also be pressed on by meningeal haemorrhages. The facial and auditory nerves and the pars intermedia are very frequently involved in fractures of the base of the brain which pass through the petrous portion of the temporal bone. Fortunately, this usually happens on one side only, as it results in complete loss of hearing and facial paralysis. Tho^ facial is also liable to injury in its course through the parotid gland, either by severe blows of the fist or by sabre wounds (which is a fairly common in- cident in duelling in Germany). The other cranial nerves are less liable to injury, but the spinal accessory in the neck is not infrequently injured in surgical operations. Injuries to the brachial plexus or its roots are not uncommon, and two forms are sufficiently classical to be called after the neurologists who first investigated them. These are the Erh- Duchenne and the Klumpke types of paralysis. The former name is given to lesions of the upper trunk of the brachial plexus, or to the fifth and sixth cervical roots which form it. INJURIES TO THE BRACHIAL PI;feXlJS i [Vx^yJ^,] This may be produced in a variety of ways. The commonest and classic cause is a fall on the shoulder of such a kind that the latter is forcibly separated from the head (Erb), or it may occur as a birth injury in breech presentations from traction on the shoulders or arms before the head is born (Duchenne). It may also be caused by pressure on the shoulder, as by a haversack or pack strap. In this case it is probably due to the nerve being nipped between the clavicle and the first rib. In the Klumpke type of paralysis the inferior trunk of the plexus is injured, or the eighth cervical and first thoracic roots torn across close to their point of emergence from the cord. It is usually produced by traction on the arm in an upward and outward direction, as when the hand is caught in machinery. When the roots are torn close to the cord the fibres going to the cervical sympathetic are destroyed. Paralysis of the eighth cervical root alone is often associated with the presence of cervical ribs, and may come on after some injury or strain. In this condition either the cervical rib, or the fibrous band which joins it to the first rib, rubs against the nerve root. It is to be noted that the longer the rib is, the less likely is it to produce paralysis. Paralysis of the plexus as a whole, or the greater part of it, may result from crushing, associated with fracture of the clavicle. Apart from penetrating wounds of the arm, certain forms of injuries to the brachial nerves are fairly common. Such are the circumflex nerve paralysis resulting from dislocation of the shoulder joint or severe wrenching of the arm at the joint, and the muscuh-spiral palsy which may result when the arm hangs over the side of a bench or table during sleep, the so-called " drunkard's palsy," or which may be caused by long-continued pressure in " crutch paralysis." The musculo-spiral is also very liable to injury by the blow of a stick on the outside of the arm, or by sword wounds. The ulnar nerve, owing to its intimate connection with the capsule of the elbow joint, is often injured in dislocations or fractures of the joint, and its exposed position renders it specially liable to develop a chronic interstitial neuritis. In the lower limb traumatic paralysis is uncommon, and when it does occur it is usually the external popliteal nerve which is affected. This branch of the sciatic seems particularly liable 102 INJURIES TO THE NERVOUS SYSTEM to injury even in the pelvis, as it is chiefly affected in the paralysis of childbirth, which is produced by the pressure of the foetal head in the pelvis. It may also be paralysed by injuries to the head or neck of the fibula, or by the prolonged kneeling necessitated by certain occupations such as asphalter's drop-foot. REFERENCES Babonneix et Voisin: Paralysie radiculaire type Erb, d'origine obst6tricale. Gaz. des Hop. de Paris, 1909, vol. Ixxxii., p. 719. Boycott, A. E., Damant, G. C. C, and Haldane, J. S.: The Prevention of Compressed-aii Illness, Journal of Hygiene, vol. viii., 1908, p. 342; Quart. Journ. of Med., vol. i., 1908, p. 348; Journ. of Path, and Bact., vol. xii., 1908, p. 507. D^jerine-Klumpke: Paralysie radiculaire des plexus brachial. Rev. prat. de trav. de Mid., 1898, vol. Iv., p. 23. Heller Mager und v. Schrotter: Luftdruckerkrangungen. Vienna, 1900. Hill, Leonard: Caisson Sickness. London, 1912. Holmes, Gordon: The Spinal Injuries of Warfare, B.M.J., 1915, vol. ii., pp. 769, 815, 855. Holmes, G., and Sargent, P.: Injuries of the Superior Longitudinal Sinus, B.M.J.. 1915, vol. ii., p. 493. MoTT. F. W. : Effects of High Explosives on the Central Nervous System, (Lettsonian Lectures, 1916), Archives of Neurology, 1918, vol. vii. CHAPTER IV CIRCULATORY DISTURBANCES OF THE BRAIN AND SPINAL CORD The physiology of the cerebral circulation has been the subject of much discussion for many years, and there are yet many questions in connection with it which have not received a final answer. The pathology of the cerebral circulation is still more complex, but recent work has cleared up some of the earlier misconceptions. For instance, it is no longer permissible to take the post-mortem appearance of the cerebral and menin- geal vessels after the skull cap has been removed as indicative of the circulatory condition when the brain was still pulsating in the closed cranium. Little importance is, therefore, attached to hyperaemia or anaemia of the brain tissue after death, and " cerebral congestion " as a diagnosis has ceased to satisfy the scientific pathologist. According to modern teaching the amount of blood con- tained in the cranial cavity can vary but little, and arterial hyperaemia has no significance as a morbid state. When the current of arterial blood in the brain is increased the amount of venous blood is proportionately diminished. In the same way, venous congestion of the brain is associated with a diminished supply of arterial blood. Cerebral hyperaemia depends, therefore, on a general rise in arterial pressure, and exerts its influence, not so much by increasing the blood content of the brain, as by sending blood more rapidly through the cerebral vessels. Cerebral hyperaemia is associated with an improved quality of blood in the brain, and, therefore, with greater cerebral activity. Probably the initiation of cerebral activity under normal conditions brings about the rise in the arterial pressure through stimulation of the bulbar vasomotor centres. The results of cerebral hyperaemia are increased mental energy combined with a sense of well-being. 103 104 CEREBRAL ANAEMIA Cerebral anaemia may be a pathological condition, and may be either local or general. The results are very similar, whether they arise from a diminution of the arterial blood supply or from venous congestion produced by obstruction to the venous outflow. General anaemia of the brain is brought about by cardiac failure, great loss of blood, respiratory embarrassment, or determination of blood to the abdomen ; on the other hand, local anaemia may result from embolism or thrombosis of a cerebral artery, meningeal or cerebral haemorrhage, com- pression by tumours, etc. The effects of cerebral anaemia vary according to whether it is local or general, partial or complete, slow or sudden in onset. Local anaemia is considered below. When general cerebral anaemia is rapidly or suddenly produced loss of consciousness is the first result, and this may be followed by epileptic convulsions, dilatation of pupils, slowing of pulse and respiration and rise of arterial pressure. If the anaemia continues the pulse becomes rapid, arterial pressure falls, and respiration ceases. It seems, therefore, that cerebral anaemia in the first instance increases the excitability of the bulbar centres, while it rapidly destroys that of the cortex. When ligatures are placed simultaneously on both carotid and both vertebral arteries of a dog the animal survives, but a condition of idiocy is established similar to that of the de- cerebrate animal. Histological examination of the cortex twenty-four hours after the operation reveals definite changes in the nerve cells. They are swollen, their protoplasm is diffusely stained with methylene blue, and the Nissl granules have disappeared (p. 13). In higher animals, such as the monkey, the sudden ligature of all four arteries is followed by death, and this would doubtless obtain in human beings. As a matter of fact, it is unsafe to tie one carotid artery in man unless the process is carried out slowly. The results of a slowly progressive cerebral anaemia in human pathology are illustrated by what occurs in extensive atheroma of the cerebral arteries. The cortical cells become starved and atrophic, the nervous tracts degenerate, and minute areas of liquefying necrosis may be seen in the central brain substance surrounding the diseased arterioles. Such cases display progressive signs of dementia and impairment of motor CIRCULATORY DISTURBANCES OF THE BRAIN 105 and sensory function. They may be classed under senile dementia or chronic progressive double hemiplegia, according to the pre- dominance of mental or physical deterioration. I. Ischaemic Softening of the Brain. This form of softening is limited to the area of distribution of a particular artery, and is dependent on the obliteration of the circulation through that artery. It is identical with the infarcts of other organs, such as the spleen, the kidney, the lung, etc., and differs only from the latter on account of the peculiar structure of the brain tissue. The obliteration of the arterial lumen is brought about in the majority of instances either by the lodgment of an embolus or by thrombosis taking place as the result of disease in the vessel wall. The changes in the vessel wall may be of an atheromatous character or due to arteritis, generally syphilitic in origin. It should not be forgotten that other infections — those of tuberculosis and the acute specific fevers, for instance — may occasionally cause arteritis. Sometimes, though comparatively rarely, the inter- ference with the circulation is brought about by pressure on the vessel from without — for instance, by the presence of a tumour. Thrombosis is naturally favoured by any increase in the coagulability of the blood or by diminution in the rate of blood flow. In the case of an embolus, which may be a small mass of fibrin, a portion of a diseased cardiac valve, or a fragment of diseased intima from the aorta or one of its large branches, the arrest of circulation is generally brought about by the lodgment of the offending particle at or near an arterial bifurcation. The source of the embolus must be sought for either in the pul- monary circulation, the left side of the heart, the ascending aorta or the branches arising from the aortic arch. It appears to be more frequent for the embolus to pass into the left carotid artery, and therefore to cause ischaemic softening more often on the left than on the right side of the brain. As a matter of experience, the middle cerebral artery on the left side is more often affected in this way than any other vessel. The position of an area of ischaemic softening depends upon the artery obliterated, and its size is determined by the calibre of the artery. Thus, if the main trunk of the middle cerebral io6 CIRCULATORY DISTURBANCES OF THE BRAIN Fig. 22. Cerebral softening, the result of embolism of the left middle cerebral artery, in a case of aortic valvular disease. THROMBOSIS OF CEREBRAL ARTERIES 107 artery is blocked before the perforating branches are given off, softening will ensue in the basal ganglia and internal capsule, as well as in parts of the cortex around the fissures of Rolando and Sylvius. On the other hand, if the interference with circulation is situated beyond the origin of the perforating arteries, only the cortical substance will be affected. Some- times, when smaller branches are the seat of thrombosis or embolism, only limited areas of the cortex undergo the softening process. Atheroma of the basilar artery is common in advanced life, with the result that pontine thrombosis due to blocking of one of its branches is of frequent occurrence. Fig. 23. Thrombosis of the left posterior cerebral artery. As a matter of fact, necrosis does not take place throughout the whole of the territory supplied by the blocked artery, owing to the fact that, although the cerebral arteries are to some extent terminal vessels, there is a certain amount of over- lapping of one arterial territory with another. The greatest degree of destruction of brain tissue from the obliteration of one middle cerebral artery seems to occur in cases of gross aortic incompetence, and is probably due to the low dia- stolic blood pressure in this condition, and the resulting poverty of the capillary blood supply which reaches the damaged area from the anterior and posterior cerebral arteries. io8 THROMBOSIS OF CEREBRAL ARTERIES Fig. 2^. Softening in the region of the lenticular nucleus. Fig. 25. Pontine thrombosis due to disease of the basilar artery or its branches. CIRCULATORY DISTURBANCES OF THE BRAIN 109 Morbid anatomy. — Recent softening cannot be satisfactorily investigated if the brain is examined in a fresh condition. In fact, it may very easily escape detection if the organ is cut up on the post-mortem table. On the other hand, after hardening in 10 per cent, formalin, even recent patches of softening are readily traced. The appearances of a softened area vary with the age of the lesion, and with more minute differences dependent on the blood supply of the region affected. Recent softening. — That part of the cerebral hemisphere from which the blood supply has been recently cut off is generally somewhat swollen, owing to the fact that it is the seat of a serous infiltration. If the patch is on the surface, the con- FlG. 26. Cerebral thrombosis. The shrunken convolutions have a yellowish tinge. volutions are enlarged; if in the deeper parts, the increase in size is demonstrated by a flattening of the overlying convolu- tions. In all cases the necrotic area is characterised by its soft consistence, and in most cases by a change in colour from that of the surrounding parts. The amount of softening varies with the age of the lesion. When death has ensued almost immediately upon the obliteration of a large artery, the amount of softening may be difficult to detect. A little later the tissue, although permitting section, is obviously moist and crumbly. Within a few days the central parts of the necrosed area may be reduced to the consistence of creamy milk. The necrotic tissue may be white, red or yellow. The absence no ISCHAEMIC SOFTENING OF THE BRAIN of colour in the white patches is due to the ischaemia. The red tint is the result of the general capillary congestion re- sulting from infarction and the escape of blood from these vessels by the rupture of their walls. Yellow softening is usually of older standing, and its colour depends on the presence of altered blood pigment and of the fatty products of myelin disintegration. Yellow softening of the surface convolutions is generally associated with a diminution in their size. In such an area the fissures are widened, and the leptomeninges cannot be peeled from the surface without disintegration of the latter. Old softenings are often represented by cystic cavities which may or may not be loculated, and which are filled with clear or slightly turbid fluid. In other cases, when the destruction of tissue has not been so complete, the convolutions may be represented by narrow strips of a substance resembling wash- leather. A still less severe form of necrosis is represented by convolutions which are somewhat atrophied, and marked on their surface by indentations giving an appearance similar to that of beaten silver. Microscopical appearances. — In a very recent case the only manifestation of the arrested circulation may be the coagula- tion of the blood in the distended vessels. Necrosis of the tissues follows rapidly, and the first evidence of this change is afforded by the altered reaction of the various elements to artificial stains used in the preparation of microscopical sections. When stained by haematoxylin the nuclei are paler than those in healthy parts. If the Weigert-Pal method is employed, the myelin substance fails to take on the normal dark blue coloration, and if Marchi preparations are made, the medullary sheaths appear dark, swollen and partly pigmented. Nerve cells in the affected area may be either swollen and chromolytic or shrunken, broken and homogeneously stained. After the lapse of two or three days the whole appearance of the softened part is altered. In addition to the normal con- stituents of the tissues, there are to be seen numbers of large round cells containing one or two nuclei. These are the so- called compound granular or fat granule cells which by suitable methods may be shown to contain numerous fat droplets, the products of tissue degeneration. Their origin is not CIRCULATORY DISTURBANCES OF THE BRAIN iii Fig. 27. Photographs of a brain illustrating the appearances produced by vascular lesions of varying severity. The somewhat battered aspect of some convolu- tions is the result of moderate interference with the circulation. The poren- cephaly in the left post-central region indicates a considerable loss of tissue. 112 ISCHAEMIC SOFTENING OF THE BRAIN Fig. 28. Softened cortex resulting from arterial thrombosis, showing granular corpuscles and necrotic pyramidal cells. (Stained by haematoxylin and van Gieson.) f .*••-%.' ^••c#^' .-.*'•'■ ••-■•'■■■ :'•■' ^ r " I \. * » -. f • • #.•!■'•■ - g, •^: Fig. 29. Fat-laden granular corpuscles in an area of cerebral softening, the result of embolism of a cerebral artery. (Haematoxylin and van Gieson.) CIRCULATORY DISTURBANCES OF THE BRAIN 113 absolutely determined, but many of them probably represent the proliferation of neuroglial cells, while others may be derived from fibroblasts. Whatever their derivation may be, they act as scavengers, and may be found chiefly in lymphatic spaces around blood vessels, and in the scars of all recent and long-standing softenings of the central nervous system. The fate of the neuroglia varies with the region examined. In the central parts of the softened area the neuroglia is destroyed along with the more specialised nervous structures. •^ijHk Fig. 30. Softened brain tissue with fat-laden granular corpuscles stained by the Marchi method. but in the peripheral zones — that is to say, in the parts where the ischaemia is only relative, sufficiently severe to bring about necrosis of nerve cells and nerve fibres, but insufiicient to destroy the more resistant supporting structures — evidence of neuroglial reaction is readily detected. The neuroglial cells are increased in size, their processes are more conspicuous and more numerous, and their nuclei undergo division and multiplication. At a somewhat later period the neuroglial fibrils become more and more prominent, and form a dense network to which the term sclerosis i» properly applied. From 114 CIRCULATORY DISTURBANCES OF THE BRAIN each peripheral zone of neuroghal activity the process of repair extends inwards towards the centre of the focus of softening, and an effort is made to replace the nervous elements which have disappeared by new-formed glial tissue. There are, no doubt, small patches of softening in which the neuroglia is. nowhere completely destroyed, and in which its reaction is so early and complete that the parts retain a good deal of their former consistence and shape, although they have been de- prived of their nervous constituents. Such is the case in some of the yellow, shrunken, but fairly tough, convolutions which have been described above. There is another form of scar tissue to which the term " lacunar " is applied owing to its appearance. It is charac- terised by a number of spaces separated from each other by strips of tissue composed of neuroglial fibres with perhaps a few cells surrounding a thickened blood vessel. These small cavities contain a liquid^in which are to be found drops of fat, compound granular cells, crystals derived from the blood pigment, and particles of cholesterin. Some of the blood vessels within the necrosed area, which no longer contain circulating blood, disappear with the other tissue elements. Others remain as strands of connective tissue, with obliterated lumen, to take part in the formation of the scar tissue. The majority, however, appear to regain their function; at any rate, in the softened areas, it is usual to find a large number of small vessels in which circulation is re- established. Newly formed vessels and fibroblasts also appear and gradually replace the products of disintegration, with the result that the margins of the necrosed area may be largely composed of young granulation tissue. The more remote effects of the destruction of nervous tissue are to be found in the changes which take place in the affected neurons. For instance, a patch of softening in the posterior part of the internal capsule produces atrophy and disappearance of the Betz cells in the corresponding motor cortex, as well as degeneration of the pyramidal fibres throughout their course in the mid-brain, medulla, and spinal cord. This is a common occur- rence in cases of hemiplegia. Similarly, pontine thrombosis leads to secondary degeneration of the ascending and descend- ing tracts passing through that region, {v. Figs. 31 and 32.) THROMBOSIS OF CEREBRAL ARTERIES 115 Degeneration of the pyramidal tract due to softening of the internal capsule, stained by the Marchi method: a, decussation of pyramid; b, thoracic cord. ii6 PONTINE THROMBOSIS Fig. 32. Fouf sections from a case of pontine thrombosis illustrating the secondary degeneration in the pyramidal tract: a, decussation of pyramid; h, cervical enlargement; c, thoracic region; d, lumbo-sacral enlargement (Weigert-Pal). CIRCULATORY DISTURBANCES OF THE BRAIN 117 2. Cerebral Haemorrhage. Haemorrhage on the surface or within the substance of the brain is, in the great majority of instances, arterial, and rarely venous, in origin. The immediate cause of haemorrhage is the rupture of some artery which has undergone pathological changes, such as arterio-sclerosis or atheroma. In a certain proportion of cases the blood escapes into the tissues from a miliary aneurysm, which has resulted from hyaline degeneration in the vessel wall. While disease of the arterial wall is the most important factor in cerebral haemorrhage, an associated rise in blood pressure and hypertrophy of Fig. 33. Haemorrhage from the anterior cerebral artery ploughing up the frontal lobe. the left ventricle of the heart are others which play no insignificant part. Haemorrhage may take place from any cerebral artery, but there are certain localities in which the event is more frequently observed than in others. Haemorrhages are more common within the brain substance than on the surface, and they are most frequent in the central grey matter of the basal ganglia, and especially in the region of the external capsule and lenticular nucleus. In this situation are found the lenticulo-optic and lenticulo-striate arteries, and one of the latter received the name of " the artery of cerebral haemorrhage " from Charcot. Extravasations of blood may be of any size and any shape, ii8 CIRCULATORY DISTURBANCES OF THE BRAIN, and they make room for themselves partly by pressing back and partly by tearing up the substance of the brain. Cerebral haemorrhages may be limited in their extent or may penetrate as far as the lateral ventricle on the one side or the surface of the cortex on the other. A ventricular haemorrhage is the result in one case and a subarachnoid extravasation in the other. A ventricular haemorrhage may extend through the aqueduct of Sylvius and the fourth ventricle into the sub- arachnoid cistern in the posterior fossa of the skull, whence the blood may escape into the spinal subarachnoid space, usually tracking along the dorsal surface of the cord. Gross appearances. — A large recent haemorrhage into the substance of one hemisphere may give rise to obvious changes on the surface of the brain. The affected hemisphere is more voluminous than its fellow, and the convolutions on its surface are flattened and sometimes anaemic. On section the seat of haemorrhage is occupied by a red clot, which is easily separ- able from the adjacent cerebral substance. The latter is infiltrated and discoloured to some extent by the blood, and small haemorrhages are frequently found in the neighbourhood. The surrounding tissues may also be softer than normal, owing to the presence of oedema. At a later period the clot and the walls of the haemorrhagic cavity are found to have undergone certain changes. The clot tends to shrink, remaining red in the centre and yellowish in its peripheral parts. At the edges nervous elements may be mixed up with crystals of blood pigment and the debris of disintegrated brain substance. After the lapse of a certain time the coagulum becomes completely absorbed, and its place may be taken either by proliferated scar tissue or by a quantity of more or less blood-stained fluid. While these changes are going on in the blood clot, there are others taking place in the walls of the cavity. The latter comprise the destruction of nervous elements and the pro- liferation of the neuroglial tissue. The tendency of the new- formed glial substance is either in the direction of forming a thin lining membrane to the cavity, or, in other instances, towards the creation of a network which binds the walls together. The final results may be either a large single cavity con- CEREBRAL HAEMORRHAGE 119 Fig. 34- Three photographs from a case of cerebral haemorrhage with extravasation of blood into the ventricles and minor haemorrhages in the pons. 120 ANEURYSMS OF THE CEREBRAL ARTERIES taining serous fluid, a multilocular cavity with similar contents, or a linear scar obliterating the site of the previous extrava- sation. It may be readily understood that these relics of haemorrhagic catastrophes can scarcely be distinguished from those of other vascular or inflammatory lesions of similar antiquity. Microscopical changes. — Owing to the presence of a foreign body in the form of a blood clot, the neighbouring brain matter is rendered anaemic and oedematous. Consequently the nerve fibres and nerve cells undergo degenerative changes, such as have been described under ischaemic softening of the brain (p. 105). The cells tend to become swollen and chromolytic. The medullary sheaths of the axons swell up, become fragmented, and undergo fatty changes. The axis cylinders become varicose and may break up into short frag- ments. As early as the second or third day numerous com- pound granular cells appear on the scene, and these take up into their interior detritus of all kinds, including fat droplets and pigment granules. Finally, these scavenger cells wander into the lymphatic spaces of the vessel walls, and so enlarge the meshes of their adventitial sheaths. In the meantime, the apoplectic focus is invaded by young vascular sprouts and spindle-shaped connective-tissue cells with large vesicular nuclei. These form a granulation tissue which gradually tends to fill the haemorrhagic cavity. Simultaneously the neuroglial tissue proliferates and produces a network of fibres which shares in the formation of the permanent scar. 3. Aneurysms of the Cerebral Arteries. The arteries at the base of the brain may be the seat of aneurysmal dilatations which differ in no way from similar structures found in the arterial system of the body generally. These do not require any further description here. On the other hand, the cerebral arteries are sometimes the seat of miliary aneurysms which were first described by Charcot and Bouchard, who regarded them as secondary to arterio-capillary fibrosis. These appear in the form of small round bodies about the size of a pin's head, red or greyish-red in colour, and either imbedded in the substance of the brain or situated on its surface They may be scarce or numerous. They are CIRCULATORY DISTURBANCES OF THE BRAIN 121 often difficult to detect, unless, in the case of a recent haemor- rhage, the affected region is more or less broken up in water, so that the brain substance separates from the finer arterial and capillary vessels and enables the observer to examine the latter in detail. If this course is adopted, it is customary to find that the vessels which are the seat of one or more aneurysmal dilatations show marked sclerotic changes in their walls. A section through the aneurysm itself shows that its wall consists only of connective tissue, the muscular fibres and inner coat having disappeared. 4. False Porencephaly. False porencephaly (p. 56) results from a softening of the brain due to some vascular lesion. In contradistinction to true porencephaly it has no communication with the lateral ventricle, although there are exceptional cases with extensive excavation in which this occurs. The wall of the cavity has no respect for the architecture of neighbouring structures, and convolutions may be half destroyed and cut across irregularly. The crater-like excavation is usually of considerable size, and its walls may be lined by a membrane formed partly of glial and partly of connective tissue. It may be traversed by bands of similar structure. Microscopical examination of the neigh- bouring brain substance generally reveals a diminution in the quantity of nervous elements. The fluid in the cavity may contain remnants of brain substance and a certain amount of colouring Tffilfter derived from the blood. It is not unusual to find in the brain which is the seat of a false porencephaly other patches of softening or sclerosis. The majority of these cases result from pathological lesions occurring in early life, and may in some instances be antenatal in origin. 5. Meningeal Haemorrhage. This term may be conveniently applied to all conditions under which blood escapes from the intracranial arteries or veins, and accumulates somewhere between the surface of the brain and the inner aspect of the cranium. The extravasation may be either subarachnoid, subdural or extradural. Haemor- rhages, generally of small dimensions, in any of these situations 122 MENINGEAL HAEMORRHAGE may be due to certain obscure causes, which are embraced under the term haemorrhagic diathesis, and which prevail in purpura, haemophiha, pernicious anaemia, scurvy, the maUgnant examples of the acute specific fevers, alcoholism, etc. Such accidental complications of diseases, sufficiently serious in themselves, have not the same clinical importance as cases of meningeal haemorrhage in which a profuse escape of blood, with its power of irritating and compressing the brain tissue, is responsible for grave symptoms. The pathology of the latter group, therefore, must attract most of our attention. (a) Subarachnoid haemorrhage. — Extravasations of blood of varying size in the subarachnoid space are found in all cases of severe injury to the surface of the brain, especially in bruises and lacerations resulting from blows on the skull, with or without fracture of the latter. The haemorrhage is not con- fined to that part of the cerebral surface which corresponds to the cranial injury, but, through the influence of " contrecoup," may be equally well marked in distant regions, especially at the opposite pole. Blood often finds its way into the subarachnoid space after rupture of a cerebral artery, either in a sulcus or in the substance of the brain. In the latter case, the blood forces its way to the surface through the nervous tissues. Similarly an aneurysm, most frequently in the circle of Willis, may rupture into the subarachnoid space and cause an extensive haemorrhage at the base of the brain. Such an accident leads to a rapidly fatal issue in most cases, and the extravasated blood may often be traced through the foramen magnum and along the surface of the cord to its lower extremity. (b) Subdural haemorrhage. — This is usually the result of trauma, and due to the rupture of a meningeal artery or venous sinus. Haemorrhages in this situation are not uncommon in newly born infants, in consequence of injury to the skull and dura mater during birth. Such extravasations are found on the surface of one or both cerebral hemispheres, and are, according to some observers, the cause of some cases of infantile hemiplegia or diplegia. More often they are rapidly fatal. In adults, a blow on the skull, with or without fracture of the latter, may cause rupture of the superior longitudinal sinus or one of its tributaries. Slow oozing of blood takes place, PACHYMENINGITIS HAEMORRHAGICA INTERNA 123 and, accumulating between the dura and the arachnoid membranes, brings about gradual compression of one cerebral hemisphere. Hemiplegia, hemianaesthesia, hemianopia, coma and death may follow, unless the condition is relieved by operative interference. Arterial haemorrhage within the subdural space may follow severe injuries to the skull, and compression symptoms arise much more rapidly than when the pressure is due to venous bleeding. Pachymeningitis haemorrhagica interna is a term applied to a somewhat rare condition, found either alone or in association with other morbid states. In a mild form it is present in a few cases of pulmonary, cardiac or renal disease, but more frequently it accompanies chronic brain affections with atrophy of the convolutions, such as is met with in general paralysis of the insane, Huntington's chorea, and long-standing alcoholism. The pathological appearances are characteristic. On the inner surface of the dura there is deposited a pale greyish-red or yellowish-red material, which, by its laminated composition, suggests a series of haemorrhages, some recent and some old and organised. The dura mater may be thickened in some cases. In others it is of normal density but tightly stretched. The haemorrhagic membrane may be easily stripped off the surface of the brain, or may be firmly adherent to the arachnoid membrane. The condition may be limited to the outer aspect of one hemisphere, or it may affect both, and even in exceptional cases extend to the base of the brain. According to some authorities the condition is primarily haemor- rhagic, according to others, primarily inflammatory in origin, but its exact pathogenesis is little understood. Microscopical examination reveals the presence of new fibrous tissue and new blood vessels, as well as the remnants of old haemorrhages in the form of pigment composed of haematoidin and haemosiderin. {c) Extradural haemorrhage is nearly always the result of an injury to the bones of the skull. Although usually ascribed to rupture of one of the branches of the meningeal arteries, it is probable, according to the observations of Wood-Jones, that the bleeding is of venous origin in many if not the majority of cases. The arteries are lodged in thin-walled venous sinuses. 124 EXTRADURAL HAEMORRHAGE which are much more easily damaged than the former. The blood collects between the dura mater and the cranium, and by so doing produces a haematoma which, playing the part of a foreign body, compresses the subjacent brain tissue. It should not be forgotten that a severe injury may produce at once an extradural and a subdural haemorrhage, as well as bruising and laceration of the brain. The results of intracranial haemorrhage, whether extra- dural, subdural or subarachnoid, in cases which survive and in which measures are not taken to remove the clot, are very similar. The coagulum is partly absorbed and partly organised so that scar tissue, with or without the presence of cysts, forms the permanent remains. 6. Sinus Thrombosis. Thrombosis of the cerebral veins is a rare event compared to thrombosis of the cerebral arteries. Two varieties are usually described: (i) primary or marantic sinus thrombosis; (2) secondary infective or phlebitic sinus thrombosis. Primary sinus thrombosis is due to diminished blood pressure, to changes in the quality, and especially in the coagulability, of the blood or to a combination of these factors. It may be favoured in some instances by fatty degeneration of the endothelial lining of the veins and their trabeculae. It occurs chiefly in infancy and old age, especially when the victim has been greatly weakened by diarrhoea, pulmonary tuberculosis or malignant disease. ^ Occasionally it is met with in the course of one of the acute specific fevers, such as typhoid variola or pneumonia, but in such cases some phlebitis may play an important part. Severe cases of chlorosis have also been reported in which sinus thrombosis has been a serious complication. The thrombosis may be general or confined to one or two sinuses ; in the latter case, the superior longitudinal and the lateral sinuses are those most commonly affected. In recent cases the clot is of a dark greyish-red colour. The longer it remains the paler is its colour and the firmer is its adherence to the venous wall. The coagulum usually extends into the contributory veins, and when the longitudinal sinus is the seat of the thrombosis. SINUS THROMBOSIS 125 the veins of the cerebral convexity are converted into firm, dark purpHsh tubes. In contrast to these clotted vessels are the distended tortuous veins containing fluid blood. The brain tissue from which the thrombosed vessels draw their blood supply is usually oedematous, congested, and perhaps haemor- rhagic and softened. Secondary sinus thrombosis is the result of some inflammatory process involving the cranial bones, the cranial contents or the tissues and cavities on the external surface of the skull. Otitis media, necrosis of the temporal bone and suppuration in the nasal cavities, must be regarded as the commonest causes of this variety, and the otitic cases form the large majority. Consequently the lateral sinus is the favourite seat of trouble. From it the internal jugular vein is frequently infected. Caries of the cranial bones, meningitis, facial erysipelas, orbital suppuration, septic wounds of the scalp, and even carbuncles and parotitis, may be the cause of phlebitic sinus thrombosis. Purulent meningitis, cerebral or cerebellar abscesses, and general pyaemia are the most frequent complications. The wall of the sinus is greenish-yellow in colour, and the clot within has a dirty greyish-red appearance and necrotic or purulent characters. Suppuration may extend into the layers of the dura mater or to its internal surface. Secondary abscesses, especially in the lungs, are not uncommon. 7. Haematomyelia. Although'^haemorrhages are found not infrequently in the spinal cord of patients who have suffered from acute inflam- matory or acute vascular conditions of that organ, and are quite common in patients who have died with urgent dyspnoea, the term haematomyelia is generally reserved for cases in which a haemorrhage is mainly responsible for the clinical symptoms. Although cerebral haemorrhage is chiefly deter- mined by sclerotic changes in the cerebral vessels, the aetiology of spinal haemorrhage appears to be of an entirely different nature. Sclerotic degeneration of spinal arteries is by no means uncommon, but the rarity of haematomyelia, and the fact that it is not by any means clearly associated with arterial changes, is sufficient evidence that we must look elsewhere for its principal cause. This is not far to seek, as haematomyelia 126 HAEMATOMYELIA nearly always follows upon some definite injury affecting the spinal column. It is associated frequently with fracture- dislocation, but may also arise when there has been no such result of the injury. Blows upon the spinal column, falls on the head or the feet or the sacral region, are the common immediate precursors of the condition. Obstetrical injuries have also resulted in haemorrhage into the child's spinal cord. Excessive and constant muscular exertion is probably respon- sible in some instances, and such conditions as haemophilia, congenital or acquired fragility of the blood vessels and purpura, may sometimes be regarded as predisposing factors. Haemato- myelia may occur at any age, but is most common between those of twenty and forty, the period of greatest physical exertion and exposure to injury. As might be expected, men are more liable than women to the condition. ■ Pathogenesis. — The grey matter of the spinal cord is much richer in vessels than the white matter, and the tissue being of a looser character affords less support to the vessel walls. This appears to explain the common incidence of haemorrhage into the central grey matter, and also the tendency for the blood to track its way in a longitudinal direction, that is to say, in the path of least resistance. Morbid anatomy. — In recent cases nothing may be detected on the surface of the cord and its meninges, although the latter may present evidence of bruising. Palpation with the finger often detects a soft fluctuating swelling, and the eye may sometimes be attracted to the dark bluish-red hue of the central blood clot, visible through the surrounding white matter. The most common site of the haemorrhage is in the cervico-thoracic enlargement ; its occurrence in the thoracic or lumbo-sacral region is comparatively rare. A series of trans- verse sections shows that the extravasation may be limited to one or two segments when it is round or oval in shape, or it may extend through many segments in the form of tapering prolongations upwards and downwards, in which case it has a more spindle-shaped contour. The blood is limited to the grey matter at most levels, although at the site of the original leakage the white matter may be seriously encroached upon. The track pursued by the haemorrhage usually involves the bases of the dorsal horns, but extends also in some regions HAEMATORRHACHIS 127 into the ventral and lateral grey substance. Multiple haemorrhagic foci are fairly common. The colour depends on the age of the haemorrhage, being red in the early cases and brown or deep yellow in those of longer standing. In very old cases the track of the extravasation may be represented by a kind of cyst containing clear fluid, or by narrow cracks or fissures with fairly well-defined walls. Under the microscope the substance of the cord is seen to be partially disintegrated around the blood clot, and is often somewlmt oedematous. In cases of a few days' standing, evidence of neuroglial pro- liferation is usually observed, and large granular cells may be seen in considerable numbers. In later cases these changes are replaced by the appearance of neuroglial sclerosis which has arisen in the process of repair. Secondary changes in the nervous elements comprise disappearance or atrophy of the ventral and dorsal horn cells, degenerations in the ascending or descending spinal tracts, and atrophy of the ventral root fibres. Relationship of anatomical to clinical phenomena. — The common clinical picture of haematomyelia is that presented by a man who has received a serious injury to the cervical cord. The origin of the haemorrhage into the grey matter of the cervical enlargement and the consequent destruction of ventral horn cells is responsible for the atrophic palsy usually found in the muscles of the arms and hands. At the same time pressure is exerted upon the pyramidal tracts, which may be actually invaded by the extravasation, with the result that the lower extremities are affected by a spastic paraplegia. The characteristic dissociative anaesthesia is brought about by the incidence of haemorrhage upon the central parts of the cord, whereby the fibres carrying painful and thermal impulses are involved as they cross from one side to the other. The oculo-pupillary symptoms so oftien observed in these cases are the natural consequence of the involvement of the eighth cervical and first thoracic segments in the morbid process. 8. Haematorrhachis. This somewhat inelegant term is applied to conditions in which blood is found extravasated within the vertebral canal. It embraces at least two or three different types of haemor- 128 HAEMATORRHACHIS rhage. Haemorrhage between the dura mater and the bony walls of the vertebral canal or extrameningeal haemorrhage is usually the result of some injury to the spinal column with or without definite fracture or dislocation. It may also be brought about by the rupture of an aneurysm into that space, and more rarely may be produced as the result of severe con- vulsive attacks in patients dying from eclampsia, tetanus or the status epilepticus. Diseases of the heart or lungs in which there is generally congestion of the venous system may have a predisposing influence. Subdural or intrameningeal haemorrhage may also result from injury either to the spinal column or to the head. In the latter case, blood which is extravasated into the posterior fossa readily finds its way into the spinal subdural space. Obstetric injuries are well recognised as a cause of haemorrhage in this situation, and the convulsive affections which have just been mentioned in connection with extradural haemorrhage may have their influence in bringing about the intrameningeal form. Similar haemorrhages are common enough in associa- tion with the haemorrhagic forms of infective fever, and also in connection with various types of meningitis, both before and at the same time as the serous or purulent exudations become prominent features. Morbid anatomy. — The extradural space contains a large number of veins separated by a loose, fatty connective tissue, and the usual supine position of the body allows these vessels to become easily engorged. Post-mortem haemorrhages, therefore, are not uncommonly produced in the process of exposing the spinal cord, and these must not be confused with extradural extravasations originating during life. It is very rarely that haemorrhage in this situation is sufficiently extensive to cause pressure upon the spinal cord. The opportunities for escape upwards and downwards are plentiful, and in large extravasations the blood may even be found extending along the course of the spinal nerves through the intervertebral foramina. An intrameningeal haemorrhage of greater or less degree is much more common than the extradural variety. It is by no means uncommon to find the greater part of the subarachnoid space throughout the length of the spinal cord more or less CIRCULATORY DISTURBANCES OF THE BRAIN 129 filled with blood which has found its way from the cranial cavity. Sometimes, but much more uncommonly, a sub- arachnoid haemorrhage in the spinal canal may extend up- wards and reach the ventricles of the brain. Intrameningeal haemorrhage rarely, if ever, produces compression of the spinal cord, but, of course, it is often the result of a trauma which has also severely injured the latter organ. This is a point worthy of remembrance in connection with clinical work, because operations performed light-heartedly with a view to remove pressure by blood clot on the spinal cord after injuries to the vertebral column may not afford the relief of symptoms which was expected. The symptoms of a transverse lesion of the cord after spinal injuries are in the great majority of cases due either to intramedullary haemorrhage, to actual laceration of the spinal tissues or to pressure exerted by dis- placed bone, and are not the result of accumulated blood clot within the dura mater. REFERENCES Batten, F. E. : Haematomyelia. Allbutt and Rolleston, System of Medicine, vol. vii., 1910, p. 680. Fearnsides, E. G. : Intracranial Aneurysms. Brain, 1914, vol. xxxix., p. 224. Russell, J. S. Risien: Haematorrhachis. Allbutt and Rolleston, System of Medicine, vol. vii., 1910, p. 587. Taylor, James : Thrombosis of Cerebral Vessels. Allbutt and Rolleston. System of Medicine, vol. viii., 1910, p. 290. Wood- Jones, F.: The Vascular Lesion in Middle Meningeal Haemorrhage. Lancet, 1912, vol. ii., p. 7, CHAPTER V SYPHILIS OF THE NERVOUS SYSTEM The discovery by Schaudinn of the specific micro-organism of syphilis, the Spirochaeta pallida, inaugurated a new era in our knowledge of the pathology of syphilis of the nervous system. To Metchnikoff and Roux we owe the additional knowledge derived from observations on the successful inoculation of animals with the syphilitic virus, and to Wassermann we are indebted for the invention of his invaluable, if empirical, test. The interval which elapses between the appearance of a primary chancre and the onset of symptoms suggesting cerebro- spinal syphilis is very variable. Headache is not uncommonly associated with the constitutional disturbances and cutaneous rashes of the secondary period, and it has been shown that even at this time the cerebro-spinal fluid may contain the virus. A monkey has been successfully inoculated with syphilis from the cerebro-spinal fluid of a man suffering from a papular syphilide, and spirochaetes have been demonstrated in the fluid under similar conditions. The more notorious evidences of nervous syphilis are usually delayed for a year or more after infection. They may appear within a month, are fairly common in the first, and most frequent in the third or fourth years. On the other hand, ten, fifteen or even twenty years may elapse before the nervous system exhibits signs of being implicated. Although it is much more difficult to detect spirochaetes in the later manifestations of syphilis than in earlier lesions, it is interesting to remember that an ape has been successfully inoculated from a human gumma appearing three and a half years after infection. It is further of im- portance to remember that the cerebro-spinal fluid may give a positive Wassermann reaction in the absence of any evidence pointing to syphilis of the brain or spinal cord. There is reason to suppose that the virus of syphilis may lie latent in the 130 SYPHILIS OF THE NERVOUS SYSTEM ^31 lymphatic organs of the body and that, in the majority of instances, attacks on the central nervous system are delivered via lymphatic routes. Although the morbid tissue changes of a primary sore are practically identical with those of a gumma, the spirochaete is rarely demonstrated in the latter lesion. This may possibly be explained by one or other of the following assumptions, (i) Atypical forms of the spirochaete remain latent in the lymphatic system and, when the opportunity arises, attack Fig. 35. Section of optic nerve in secondary syphilis, showing enormous infiltration of Virchow-Robin space with mononuclear cells, and glial overgrowth in the optic some part of the nervous system where resistance is lowered. (2) Atypical forms of the spirochaete may be lodged in the nervous system and are only roused into activity by changes in the surrounding tissues, such as those produced by injury. (3) One race only of spirochaete has a special predilection for multiplication and activity in nervous tissues. The essential lesion of syphilis in the nervous system may be described as a more or less limited perivascular lymphangitis. It is, therefore, not a disease of nervous tissue proper, but an inflammatory process which may affect the nervous elements 132 SYPHILIS OF THE NERVOUS SYSTEM either by direct pressure and encroachment, by interfering with their supply of nutrition or by poisoning the fluids on which they depend for nourishment. Although it is convenient to describe several varieties of lesion in connection with syphilis of the nervous system, the microscopical characters of each are virtually the same. Such differences as exist depend probably upon degrees of virulence, degrees of resist- ance, and purely anatomical differences in the locality of the Fig. 36. Drawing of a section of the lumbar enlargement, showing an intramedullary gumma undergoing central caseation. initial infection. The perivascular lymphangitis is an in- flammatory process characterised by a tendency to " gumma- tosity." The more active and virulent the inflammation the less productive of gummatous tissue ; the slower and quieter the inflammation the more likely is the formation of circumscribed gummata. We may interpret the microscopical features in the following way. The virus multiplies in the perivascular lymph spaces and incites a proliferative hyperplasia of neighbouring cells, endothelial, conjunctival and perhaps epithelial. The result GUMMATA 133 is an abundant mass of cells chiefly of the plasma-cell and lymphocytic type filling the interstices of the connective tissue. The proliferative process spreads along adjacent lymphatic channels, fresh cellular masses are added in succession, until the early and more central parts of the newly-formed substance suffer from deficiency of nutrition. Depending upon this cutting off of nutrient supplies, the central parts become either necrotic or fibrotic, frequently a mixture of both. The mass is now caseous or fibrous, generally caseous in one part and fibrotic in another. If the tendency to spread dies out or is combated by local resistance or treatment, the fibrous process preponderates in the peripheral parts in such a way as to produce a tough wall around the granuloma. Under other circumstances the inflammatory reaction tends to diffuse much more rapidly and widely, with the result that the forma- tion of gummata is less prominent, at any rate to the naked eye. A richly cellular and vascular granulation tissue is then formed with widespread adhesions to neighbouring structures. With this description of the fundamental change occurring in syphilitic inflammation to help us, we may consider the three varieties into which these lesions are conveniently divided. At the same time, it must not be understood that these varieties usually occur singly ; on the contrary, it is the rule to find two or more present in every case of cerebro-spinal syphilis. I. Gummata. — These gummata, or syphilomata, are more often multiple than single, and may be found almost anywhere in relation to the brain or spinal cord, although always con- nected in some way with the meninges. On the convexity of the brain they may spring from the dura mater or the leptomeninges and, when deep-seated, arise from the pial trabeculae which penetrate between the convolutions and follow the course of the vessels. At the base of the brain and in the spinal cord they usually originate from the pia- arachnoid, and the same may be said of those which are found on the cranial nerves or the spinal roots. Their size is ex- tremely variable and their shape irregularly spherical, often with nodular projections. The main colour is greyish or greenish-yellow, but the circumference is often tinged with pink owing to its greater vascularity. The consistence varies with the relative amount of tough fibrous tissue and soft 134 SYPHILIS OF THE NERVOUS SYSTEM Fig. 37. Two sections through the brain of a patient with extensive gummatous menin- gitis and meningo-encephaWis. This process was fairly general over the surface of both hemispheres, but was most advanced in the right fronto-parietal region. GUMMATOUS MENINGITIS 135 necrotic material. They may be definitely encapsulated, but the capsule is always more or less adherent to adjacent tissues. Under the microscope the peripheral parts of the gumma are seen to be richly cellular, and sometimes to contain numerous new vessels. The cells are chiefly of the plasma and lympho- cytic type, but spindle and stellate cells of connective-tissue origin may be abundant. In the more central zones there is often much amorphous material which represents the necrosed cells and which may contain fatty globules. Bands of fibrous :, : . : ' ■ vfv':- ••,•.:'. ■'.■ft.,--- ■:■; «?..•«„■«■ /i-nA--: . •.^-■:''* Fig. 38. A section from sclerosed area of brain underlying the gummatous meningitis shown in Fig. 37, and showing proliferation and enlargement of neuroglial cells. tissue divide the masses of caseous substance. Around the gumma there is often an area of meningitis on the one hand, and encephalitis or myelitis on the other. In addition, the effect of pressure on the adjacent tissue may be observed in the form of oedema, neuroglial proliferation or softening and rarefaction. 2 . Gummatous meningitis. — Syphilitic meningitis may involve the hard and soft membranes together, and occasionally the soft membranes alone. The best examples of pachymeningitis, i.e. inflammation of the dura and pia-arachnoid, are seen on the convexity of the hemispheres and in the cervical region of 136 SYPHILIS OF THE NERVOUS SYSTEM the cord. The whole of one or both hemispheres may be covered with a dense adherent membrane as much as an eighth of an inch thick. A similar tough sheath may tightly embrace considerable areas of the spinal cord, and lead to softening of the latter by interference with its blood supply and by direct pressure on its substance. Leptomeningitis, with no involvement of the dura, is ex- ceedingly common at the base of the brain and over the lower parts of the spinal cord, although in the latter situation the theca is generally adherent to the inflamed pia-arachnoid. The interpeduncular space and optic chiasma is perhaps the most frequent site of gummatous meningitis, and in this situation may involve the third nerves and also the various branches of the circle of Willis. Remote effects on the cerebral circulation may be produced in the latter case. Both in the brain and spinal cord the meningitis is always associated with more or less inflammation of the subjacent nervous tissue. Sections of the latter show infiltration of the perivascular sheaths with lymphocytes and plasma cells, thickening of the pial trabeculae, and sometimes neuroglial hyperplasia. Nerve cells and fibres near the surface are often observed in various stages of disintegration. An interesting secondary result of basal meningitis in the posterior fossa of the skull is found occasionally in the form of an hydrocephalus, due to the blocking of the posterior outlet of the fourth ventricle. This occurs in adults as well as children, although the cranial deformities are not so prominent in the former and the fatal issue is much more speedy. The microscopical appearances of gummatous meningitis can be readily inferred from what has been already stated. The meninges are densely infiltrated with cells, and scattered about are nodules which may be caseous or fibrotic. There is always a certain amount of arteritis and phlebitis, often leading to great thickening of the vessel walls and sometimes to oblitera- tion of their lumina. Pachymeningitis cervicalis hypertrophica consists of a primary thickening of the dura mater in the cervical region of the spine, leading to secondary changes in the spinal cord and nerve roots. The majority of cases undoubtedly owe their origin to syphilis. Whether this is a constant aetiological PACHYMENINGITIS CERVICALIS 137 factor or not will probably be settled in the course of the next few years, when it will be possible to correlate the results of the serological examination of a number of cases of this some- what rare condition. In a considerable proportion of the cases described the symptoms have commenced after a trauma to the spine, or after some infectious disease. Some of the slighter cases appear to be attributable to chronic alcoholism. On removal of the cord in its envelopes from the spinal canal, the cervical region is seen to be swollen in an elongated spindle, and to be extremely firm and hard. The swollen cord may completely fill the spinal canal, and may be abnormally ad- herent to the posterior common ligament of the vertebrae. Apart from that, the outer surface of the dura mater shows little abnormality. The nerve roots may show evidence of atrophy due to pressure, but are often apparently normal. It is extremely difficult, and often impossible, to separate the thickened dura mater from the pia-arachnoid to which it is closely adherent ; and the cord is thus enveloped by a compact fibrous ring, which may obliterate all trace of a subarachnoid space. Calcareous plaques are often present in this dense fibrous tissue. Some thickening of the pia-arachnoid is also usually present. The cord itself is more or less sclerosed, but sometimes it is swollen and oedematous. Very often it contains a central cavity which appears to be due to the constriction of its nutrient blood vessels by the thickened membranes. This extends for a greater or less distance up and down the cord from the seat of the lesion. Along with this there is some degeneration in the longer ascending and descending tracts. An analogous process is sometimes found at the base of the brain, and may be continuous with the disease in the cervical region. The controversy which existed for many years as to whether the thickening of the membranes was primary or secondary to a condition of syringomyelia may now be considered settled. There is no doubt that constriction of the vessels supplying the cord may produce a central cavitation which may extend a considerable distance from the lesion. For this reason it is often difficult clinically to distinguish between primary syringomyelia and pachymeningitis cervicalis hypertrophica. The root pains, anaesthesia, and muscular atrophy are due 138 SYPHILIS OF THE NERVOUS SYSTEM partly to direct nipping of the nerve roots in their passage through the thickened dura mater, and partly also to the diminished blood supply to the grey matter of the cord. 3. Gummatous arteritis. — It is an almost invariable rule that cases of cerebro-spinal syphilis show more or less arterial change, in the form of thickening of the vessel walls, even when the most prominent lesions are circumscribed gummata or patches of meningitis. This obtains not only with regard to the arteries in the immediate neighbourhood of the gummatous lesion, but to many of the vessels throughout the central nervous system. On the other hand, there are cases in which the chief brunt of the disease falls on the arteries, with the result that aneurysms, haemorrhages and thromboses play the principal part in the production of symptoms. Syphilitic arteritis may affect vessels in either a diffuse or nodular manner, the affected parts being rigid, thickened, and of a yellow or greenish-yellow colour. The arteries composing the circle of Willis are a favourite site for these changes, but vessels of much smaller calibre may be equally involved. Much controversy has arisen concerning the exact course of the pathological changes in this form of arteritis. Into the details of this we will not enter, but will adopt the view of Mott, which is most in accord with our own experience. The virus attacks the vessel from its outer aspect — that is to say, it initiates active changes in the lymphatic spaces of the adventitial sheath, and so starts a gummatous process identical with that which we have already described in connection with the meninges. The result of this change is to interfere with the vasa vasorum, and so with the circulation in the muscular coats. If the degeneration of the latter is rapidly effected, the wall of the vessel is likely to give way. In this case aneurysmal dilatation will take place. More frequently the process is slower, and a compensatory thickening of the intima is brought about by hyperplasia of the connective-tissue elements of that coat. The endarteritis so occasioned leads to encroachment on the lumen of the vessel, and in some instances to such a degree that complete obliteration or thrombosis follows. The pathological process just described is by no means uncommon in relation to the middle cerebral artery and its GUMMATOUS ARTERITIS 139 j^^^w^w-'^:ww' Fig. 39. a, Gummatous arteritis of the right middle cerebral artery with organisation of the central clot, b, The area of softening in the right hemisphere resulting from the arterial thrombosis. i4o^ SYPHILIS OF THE NERVOUS SYSTEM (, \ .-.J I b Fig, 40. a, Gummatous arteritis of the anterior spinal artery in a case of syphilitic meningo-myelitis. b, Section of cord from the same case. SYPHILITIC MYELITIS 141 branches, and herein lies the explanation of many cases of hemiplegia due to syphilitic arteritis and thrombosis which occur in young adults before the age of ordinary arterio- sclerosis. It is a disputed point whether syphilitic arteritis produces calcareous changes such as are frequent in arterio- sclerosis, but it is generally agreed that the syphilitic diathesis predisposes towards early arterial degeneration. This is exemplified by many victims of congenital syphilis, whose arteries in early life resemble those commonly met with in people over fifty years of age. In describing the morbid anatomy of syphilis of the nervous system we have purposely avoided any attempt to distinguish between cerebral and spinal lesions. The processes in both regions are identical, and it is common to find them coexistent or developing first in one and then in the other. From the clinical standpoint, it is well known that the subject of cerebral syphilis may at any time develop spinal symptoms and vice versa unless thorough treatment has been enforced. Similarly, the post-mortem examination of a fatal case of nervous syphilis nearly always reveals morbid changes in parts of the brain and spinal cord, other than that to which attention has been directed by the clinical symptoms. Further, the mixture of coarse syphilitic processes, such as gummatous arteritis and gummatous meningitis, with the characteristic degenerative changes of tabes, is by no means rare. 4. Syphilitic myelitis. — This is by far the most common form of myelitis, and appears to affect males much more frequently than females. It may occur at almost any age, but is more frequent between twenty and forty. It has been known to be the result of congenital syphilis. The interval between the primary infection and the spinal disease varies from a few months to a great many years, but the majority of cases develop before five years have elapsed. Pathogenesis. — The term syphilitic myelitis usefully describes a process which is not altogether a simple one, because the influence of the syphilitic virus upon the spinal disease is exerted in a number of different ways. The resulting process consists partly of changes produced by interference with the local circulation, partly of changes due to . the diffusion of toxic substances, and partly of an inflammatory process 142 SYPHILIS OF THE NERVOUS SYSTEM leading to the formation of granulomatous or gummatous material. Any or all of these results may be concerned in any particular case, but as a rule one or other preponderates. It is the rule to find evidence of syphilitic changes in the blood vessels and meninges over a much wider area of the cord than the clinical symptoms have suggested, and it is probably often the case that the latter have been determined by an acute or subacute interference with the circulation brought about by vascular thrombosis of one or more important vessels. The incidence of syphilitic myelitis is much more frequent in the thoracic region than elsewhere, probably owing to the fact that this part of the cord is not so well supplied with blood as the cervical and lumbo-sacral enlargements. Morbid anatomy. — The lesion in syphilitic myelitis is usually single and generally limited to a few segments, so that it may be generally described as a transverse myelitis. The diseased area is readily recognised by its soft consistence and often by its creamy colour. The overlying membranes may be some what opaque and sometimes firmly adherent to the dura mater. Difficulty may be experienced in separating the latter from the leptomeninges and cord itself. Having regard to the variety of the processes involved, the changes in each kind of tissue are best considered separately. (a) The blood vessels. — The vascular changes are of three chief types. (i) Moderate cellular infiltration of the adventitial sheaths with thickening of the intima sufficient to produce diminution or complete occlusion of the lumen. (2) Excessive perivascular cellular infiltration with little or no endarteritis. In such cases there may be no evidence of actual obstructive thrombosis, although slowing of the circula- tion and blood stasis is suggested by the excess of white corpuscles. Around such vessels there may be evidence of considerable toxic degeneration, or an extension of the cellular proliferation into the neighbouring tissues. (3) Paralytic vaso-dilatation with very moderate peri- vascular infiltration and with no obvious thickening of the vessel walls. This is associated with capillary thrombosis and profound alterations in the nervous elements. The appearance SYPHILITIC MYELITIS 143 produced is one of general hyperaemia involving the white and grey matter, and accompanied by degenerative changes in the myelin sheaths and the ganglion cells. (b) The neuroglia offers much more resistance than do the nervous elements to disturbances in the circulation. In areas of complete necrosis it may succumb with everything else, the neuroglial cells rapidly losing their staining reaction and with the fibres taking part in the general liquefaction. In other regions where the vascular changes have not been so profound the neuroglial cells survive when more specialised structures are destroyed. Under these circumstances they tend to increase in size and to become multinuclear. Later on they proliferate and take a leading part in the process of repair, sending out long interlacing fibres in all directions. They probably give origin to some, at any rate, of the large granular cells which act as scavengers, and which, at first distributed in the tissues, are later collected in the perivascular lymph channels. With the neuroglial changes must also be mentioned the gummatous infiltration which occasionally goes hand in hand with the circulatory disturbances. Sometimes it is a pre- dominant feature of the process, and a true gumma, with the characteristic central degeneration and caseation, may occupy a considerable part of the transverse area of the cord. On the other hand, there may be a more diffuse condition starting near the surface and spreading in from the vessels and meninges, to which the term gummatous meningo-myelitis is quite appropriate. (c) The nervous elements. — Degeneration of the myelin sheaths and retrograde changes in the ganglion cells result either from the circulatory disturbances or from the diffusion of toxic substances. The myelin sheaths swell, disintegrate and disappear, leaving axis cylinders which may subsequently undergo degeneration or destruction. The ganglion cells in the affected areas display the usual chromatolytic changes and are not infrequently completely destroyed. Secondary de- generations both in the ventral roots and in the long tracts of the cord are common enough. (d) The leptomeninges. — Infiltration of the pia- arachnoid with cells chiefly of the mononuclear type is almost of universal 144 SYPHILIS OF THE NERVOUS SYSTEM occurrence in this disease, and is usually followed in later stages by well-marked thickening and adhesions of the membrane. Hypertrophy of the arterial muscular coats is frequently seen, and even the veins may display an obliterative phlebitis. Some- times the latter present a laminated appearance, layers of round cells being separated from each other by strands of connective tissue, with the result that their bulk is often as great as that of the arteries. Processes of repair are essentially the same as those described in other forms of myelitis, and the examination of a focal lesion a year or more after the acute stage may reveal nothing characteristic of its original nature. It is worth while remembering that more than one syphilitic lesion maybe found in the central nervous system, and, further, that a cord may at the same time be the seat of a localised syphilitic lesion and of a parasyphilitic process, such as tabes dorsalis. We have even met with a cerebral syphilitic throm- bosis, a syphilitic myelitis, and the appearances produced by tabes in the same subject. The relationship of the anatomical to the clinical phenomena. — The clinical phenomena may be divided into those which are local and those which are remote. The local phenomena include atrophic palsy of the muscles innervated by the segments which are the seat of the disease, and also sensory and reflex changes corresponding to them. The remote phenomena depend upon the loss of conduction of impulses through the diseased areas, and are illustrated by the spastic forms of paraplegia and loss of sensation in those parts of the body innervated from the cord below the level of the lesion. In the infective form of myelitis it is quite common for one of the spinal enlargements to be the seat of the disease, and an atrophic palsy of the arms with a spastic paralysis affecting the trunk and limbs is by no means rare. Owing to the special incidence of the syphilitic process on the thoracic region of the cord, a spastic paraplegia is the most prominent clinical feature, evidences of atrophic palsy in the trunk muscles being less easily detected than they are in the limbs. When a transverse lesion is very complete in the early stages of myelitis, whether it be of the infective or syphilitic type, the resulting paraplegia may at any rate for a time be flaccid in character. TABES DORSALIS 145 spasticity supervening as time goes on. Both forms are associated with disturbances of the sphincters, either because of the incidence of the disease on the lumbo-sacral centres or because of the interference in the conducting paths when the lesion lies above those centres. A syphilitic lesion limited to one-half of the spinal cord may produce the clinical pheno- mena known as Brown-Sequard's paralysis. The condition usually described as Erh's syphilitic paraplegia may be briefly referred to here. Little is known about the morbid anatomy of these cases because many of them which have corresponded to the clinical diagnosis have been shown post mortem to be really the subjects of a transverse syphilitic myelitis in the dorsal region of the cord with the usual ascending and descending tract degenerations. On the other hand, there is some evidence to show that syphilitic toxins may produce a degeneration of the long tracts in the posterior and lateral columns without the intervention of any specific changes in the blood vessels or meninges of the spinal cord. 5 . Tahes dorsalis. — So far as is known, the only essential factor in the aetiology of tabes is syphilitic infection, either congenital or acquired. Probably the majority of physicians would agree that tabes does not occur without syphilis, although evidence sufficient for proof of such a view is not, and of course never can be, forthcoming. Originally, the connection between tabes and syphilis was inferred merely from clinical observations ; recently laboratory methods have shown that a positive Wasser- mann reaction can be obtained in a large majority of tabetic patients in both the blood and cerebro-spinal fluid. Tabes is far more common in the male than in the female sex, an indication perhaps that other factors, such as prolonged and severe physical exertion, may play an important additional part in producing tabes in syphilised individuals. The in- fluence of age is not important in itself, although there appears to be a definite relationship between the age at which syphilis is contracted and that at which tabetic symptoms commence. Speaking roughly, there is an interval averaging from eight to twelve years between these two dates. On the other hand, there are exceptional cases in which tabes follows syphilis after the lapse of only two or three years. Syphilitic infection is most common between twenty and twenty-five years of age, and it 146 SYPHILIS OF THE NERVOUS SYSTEM is not surprising, therefore, to find that the earUest signs of tabes are most frequently noticed in the fourth decade of Hfe. Cases resembUng tabes among children are very uncommon, but there are a few tabetic patients whose symptoms can be traced back to early childhood, and who may have been the victims of congenital syphilis or of an infection acquired in infancy. So small a proportion of individuals who acquire syphilis become tabetic that it is natural to look for other factors in- fluencing the incidence of the disease. Prolonged over-exertion has already been mentioned as a possible explanation of the peculiar susceptibility of the male sex, and many observations are on record showing that constant exercise of particular parts of the body has appeared to determine the first symptoms of the malady. An artist may begin with optic atrophy, a postman with ataxic gait, or a wood-turner with tabetic symptoms limited to the upper extremities, the so-called cervical form of tabes. This line of argument, however, cannot be pushed too far as many cases present contradictory features. We are so accustomed to hear that exposure to cold, various excesses and abuses, and especially injuries, are determining causes of disease that a feeling of disappointment might be engendered if they were not mentioned in this connection. From a pathological point of view, it is impossible or very difficult to believe that trauma can materially help in the pro- duction of such a disease as tabes, although it cannot be denied that a shock may reveal or bring into prominence certain symptoms which the patient might otherwise have continued to ignore — at any rate, for a time. The mere fact that it is within our experience that one or two cases of tabes have realised their first symptoms in the sequel of an accident is no proof whatever that trauma can take any part in the production of this disease. It is sometimes taught that although a patient may become ataxic after an injury, it must not be assumed that his disease originated in this way until it is ascertained that there is no previous history of lightning pains, squint, etc. Even then such an assumption is not justified, as we have had personal experience of detecting undoubted signs of tabes in several persons who have not sought medical advice, and who, in answer TABES DORSALIS 147 to questions, have assured us that they have nothing whatever to complain of. It is inaccurate to say that a person has not been ataxic because neither he nor his friends have noticed any abnormahty of gait. Many tabetics can walk naturally and easily, but come to hopeless grief when asked to toe and heel a line. They, too, suffer from ataxy, although they may not know it. Alcoholism does not appear to be closely related to tabes, although of course it would be easy to cite numerous cases in which alcoholic excess was admitted. We are unlikely to be far wrong if we conclude that uncured syphilis (and our older views on the curability of syphilis have received rude shocks in recent years) is the one essential aetiological factor in the production of locomotor ataxy. Pathogenesis. — From a clinical standpoint, tabes is a disease characterised by a more or less symmetrical affection of the lower afferent neurons with the occasional addition of motor palsies. The affection of the afferent neurons is peculiar in that some afferent impulses are destroyed while others remain intact. Those impulses connected with sense of position, and those associated with deep and superficial pain, are apt ta succumb before those which are of tactile origin. When we consider that all the various forms of afferent impulses must reach the central nervous system through the posterior roots or analogous cranial nerves, it is difficult to believe that any gross lesion, such as is produced by meningeal inflammation, can so affect the posterior roots that only fibres of particular function are picked out for destruction. For this reason the view that the essential lesion in tabes is a syphilitic meningitis involving the posterior roots, either on the surface of the cord or at the point where they pass through the dura mater, is one which is difficult to support. Moreover, it has not been con- clusively demonstrated that meningeal changes are constant and early phenomena in the pathology of the disease. The French school which favours the view that there exists a radiculitis at the point where the roots pierce the dural sheath has not explained satisfactorily the escape of the anterior root fibres. The older conception of the disease, which assumed a primary sclerosis of the posterior columns, failed to hold its place when it 148 SYPHILIS OF THE NERVOUS SYSTEM was shown that the distribution of degeneration within the cord corresponded closely to the distribution of exogenous fibres. No primary sclerotic process of the cord itself could be credited with the power of discrimination between fibres of intra- and extra-medullary origin. Modern pathologists have tended to look upon neuroglial proliferation and meningeal inflammations as secondary or accidental phenomena, and to regard the pro- gressive degeneration of the lower afferent neurons as the primary and fundamental morbid process. It is reasonable to regard this destruction of afferent neurons as thejresult of a degeneration brought about by syphilitic toxins and to put tabes in the same class as other toxic conditions of the nervous system, from which it differs chiefly in the choice of the neurons affected. We are familiar with the effects of alcohol, diphtheria, etc., on peripheral neurons, and we see exceptional cases in which those poisons produce clinical phenomena closely resembling those of tabes. The chief difference between a diphtheritic infection and a syphilitic infection lies in the difliculty in eradicating the latter and, consequently, in its profound and enduring effects on those neurons which are specifically susceptible to its influence. It has been argued that the long interval between syphilitic infection and the onset of tabes makes the analogy between the latter and such a toxic process as post-diphtheritic palsy untenable. On the other hand, we do not expect alcoholic neuritis to follow the first indulgence in alcohol, but regard it as the result of the long-continued exposure of the peripheral neurons to the influence of that poison. Many cases of alcoholic neuritis recover if the patient is prevented from drinking, but others die or remain permanently paralysed even after indulgence in the drug has ceased. It must be remembered that although the lower afferent neurons are the site of most changes in tabes, other neurons, afferent and efferent, may also be affected. Optic atrophy and oculo-motor palsies are common examples of this observation. The frequent presence of a positive Wassermann reaction in cases of tabes has already been noted, and, as this reaction is an indication of the presence of an active virus rather than a clue to past syphilis, there is strong reason for regarding tabes as a definitely syphilitic process. TABES DORSALIS 149 Morbid anatomy. — We are no longer satisfied with sclerosis of the posterior spinal columns as a. description of the morbid anatomy of tabes, and we owe our more exact knowledge of the process to observation on the development and on the minute composition of the posterior columns. A true apprecia- tion of the pathological changes in tabes has been derived from the study of many cases, especially of early cases, and the examination of one long-standing example of the disease may easily give rise to erroneous conclusions. The essential lesion is a slow, progressive degeneration of the lower afferent neurons of the spinal cord and brain-stem, and to this must be added the results of accidental and secondary processes. In order to appreciate the essential lesion of tabes it is desirable to recall certain anatomical facts about the com- position of the posterior columns of the cord. The dorsal columns are composed of medullated nerve fibres, some of which are derived from spinal cells (endogenous), and the remainder of which are the central prolongations of the dorsal root ganglion cells (exogenous). The endogenous fibres probably serve to connect the cells of one spinal segment with those of other spinal segments above or below. They occupy chiefly the ventro-lateral margin of the columns, and form a band known as the cornu-commissural zone. Degeneration of this zone is not an essential part of the morbid anatomy of tabes, but it occurs in long-standing cases — in some parts of the cord at least — owing to the large amount of neuroglial over- growth which is a direct result of the destruction of neigh- bouring nerve fibres. It may be assumed that this loss of endogenous fibres is brought about by their strangulation in the contracting glial tissue and perhaps by reason of inter- ference with their vascular and lymphatic supply. There is a system of descending fibres in the posterior columns which may be endogenous in origin, but which more probably re- presents the descending branches of exogenous root fibres. It goes by the name of the " comma tract " in the cervical and upper thoracic region, the " septo-marginal tract " in the lower thoracic region, the " oval field of Flechsig " in the lumbar region, and the " median triangle " in the sacral region. Surviving fibres of this system are to be found in the lower segments of 150 SYPHILIS OF THE NERVOUS SYSTEM the cord in cases of lumbo-sacral tabes even when the greater part of the posterior columns are sclerosed. The exogenous fibres are those which enter by the dorsal roots at different levels. These may be divided into short, medium, and long fibres. The short fibres pass straight into the grey matter of the corresponding segment, and are dis- tributed to form connections of some kind with the ventral and dorsal cornual cells. The medium fibres pass upwards to other levels, but eventually turn inwards to arborise around the cells of Clarke's column. In their passage upwards they become displaced from their position on the edge of the dorsal horn to form a band of fibres called the middle root zone or " bandelet te " of Pierret. This displacement is brought about by the successive entry of higher dorsal roots, and if the latter chance to be healthy, while the lower fibres are atrophied, the bandelette becomes a conspicuous object between two masses of healthy tissue. The long fibres are those which are destined for the nuclei gracilis and cuneatus at the upper extremity of the cord. These, too, are gradually displaced backwards and towards the dorsal median septum, those belonging to the lowest roots eventually occupying the column of Goll, which becomes present as a distinct band of fibres about the mid- thoracic region. The essential lesion of tabes is the degeneration of these central prolongations of the dorsal root ganglion cells, and it is important to note that the process begins earlier in that portion of the fibres which lies within the cord than in the dorsal roots themselves. It is probable, although difficult of demonstration, that the descending branches of the exogenous fibres also undergo degenerative changes. What has been described in regard to the afferent root fibres applies, in some cases at least, to the analogous fibres of certain cranial nerves such as the glosso-pharyngeal, the auditory, and the trigeminal. In addition to the destruction of fibres in the dorsal columns, there are other demonstrable changes probably of secondary origin. The neuroglial tissue proliferates, the vessel walls become thickened and hyaline, and the pial trabeculae are distinctly more substantial and prominent. Thickening of the pia-arachnoid membrane on the dorsal surface of the TABES DORSALIS 151 Fig. 41. Two sections illustrating the degeneration in the dorsal columns in tabes dorsalis (Weigert-Pal). 15^ SYPHILIS OF THE NERVOUS SYSTEM cord is also a conspicuous feature of cases of any standing. The dorsal root ganglion cells are usually intact. In rare cases they may present atrophic changes, and not uncommonly the connective-tissue elements by which they are surrounded show signs of proliferation. The peripheral prolongations of the root ganglion cells are occasionally observed to be atrophied, especially towards their extremities. Rarely there is found a localised or widespread degeneration of the lower motor neurons of cranial or spinal origin. In such instances the central cells may be seen to be decreased in number, while those which survive display simple atrophic or chromo- lytic changes. The corresponding efferent roots, peripheral nerves, and muscle fibres present evidence of the secondary atrophy which one would naturally expect. Recent investigation has confirmed the older theory that the sympathetic system is involved, and Roux has found definite degeneration of the afferent myelinated fibres, to which he attributes many of the visceral symptoms characteristic of the disease. Many attempts have been made to discover the anatomical basis of the Argyll-Robertson pupil, but the problem has not yet been solved in such a way as to satisfy all critics. De- generation of the optic nerve begins near the disc and spreads towards the chiasma. Some doubt has been expressed as to whether this process is a primary one or an atrophy secondary to a chronic interstitial change in the nerve. The ganglion cells of the retina are not usually affected. The gross abnormalities of the central nervous system in cases of tabes are generally striking and characteristic. The dorsal roots, most frequently those of the lower thoracic and lumbo- sacral regions, are reduced in calibre, and, compared with the opaque white ventral roots, present a greyish semi-translucent appearance. The surface of the dorsal columns, instead of being convex, is often flattened and sometimes even concave, owing to the contraction of the glial tissue and the destruction of nerve fibres. On section they present the same contrast to the ventro-lateral columns as do the dorsal to the ventral roots. In the brain, atrophy of the optic and perhaps of other afferent cranial nerves may be detected by the naked eye, while there is frequently some shrinking of the precentral TABES DORSALIS 153 convolutions, especially in cases complicated by paralytic dementia. Some thickening of the meninges, particularly the pia-arachnoid on the dorsal surface of the cord, is usually to be noticed. In addition to what we have termed the essential lesions of tabes, evidence of tertiary syphilitic processes in the central nervous system is not so very uncommon. In one case there was a patch of softening in the pons due to syphilitic arteritis of the basilar artery; in another the tabetic changes in the spinal cord were complicated by an extensive gummatous pachymeningitis. Syphilitic aortitis, with or without aneu- rysm, is another complication of tabes which is sufficiently frequent to deserve notice. The osseous and arthritic changes, so well described by Charcot, are interesting. The bones tend to become brittle owing to the dilatation of the Haversian canals and the absorption of phosphates. Spontaneous fractures which fail to unite properly are by no means uncommon in the long bones. The capsules and synovial membranes of the Charcot joints are thickened. The cavities contain an excess of fluid, and the cartilages are ulcerated or entirely destroyed. Erosion of the epiphysial end of the bone may take place in advanced cases, and along with the destructive process there is often a crop of bony growths in connection with the synovial membranes. The knee, hip and tarsal joints are favourite sites for this change, and the deformity is often increased by the strain on the ligaments and tendons being more than they can bear. The relation of the anatomical to the clinical phenomena. — The fundamental symptoms of tabes are due to disturbance of afferent impulses, and how the latter come to be disturbed has been already demonstrated. We have to remember, however, that the process is a gradual one, and that before impulses are cut off entirely there are stages in which they are modified, impaired or delayed in transmission. In this way we may understand the various forms of paraesthesiae, of hyper- aesthesia, and of delayed sensation so common in the history of the disease. Similarly, we may correlate the disturbance of non-sensory afferent impulses with our anatomical data. The paths conveying impulses necessary for co-ordinate move- 154 SYPHILIS OF THE NERVOUS SYSTEM ment from the structures of the periphery, especially from the feet and legs, are involved in the decay of afferent neurons, with the result that there is an imperfect transport service both to the cerebrum via the dorsal columns and dorsal column nuclei, as well as to the cerebellum via the medium fibres, Clarke's column, and the direct cerebellar tract. In either case disturbance of equilibrium must result, which may or may not be compensated for by visual and vestibular impulses when the paths of the latter remain intact. The explanation of the severe lightning pains in cases in which pain sensibility is profoundly impaired is a difficult one, especially if we regard these active phenomena as the result of a slow katabolic process of degeneration, rather than as the result of an irritating toxic influence. The diminution of tendon jerks is only to be expected in view of the involvement of the afferent part of the reflex arc represented by the short fibres entering with the dorsal roots and proceeding direct to the ventral cornual cells. The problem of the Argyll- Robertson pupil still awaits solution. 6. General paralysis of the insane — Aetiology. — A history of syphilis is obtained in from 75 to 85 per cent, of cases of this disease, and this percentage may be regarded as sufficient to prove the dependence of general paralysis upon luetic infec- tion : certainly no higher percentage can be obtained in cases of gumma in various parts of the body. Juvenile general paralysis occurs in the victims of congenital syphilis. A positive Wassermann reaction is obtained from the blood serum and cerebro-spinal fluid in at least 90 per cent, of cases. An attempt to infect general paralytics with syphilis has proved a failure. Tabes and general paralysis are not in- frequently associated in the same patient, and there are many instances on record in which a husband and wife have both suffered from one or other of these so-called parasyphilitic diseases. Recently conclusive proof of the aetiology of the disease has been afforded by the discovery of the Spirochaeta pallida in the cortex of general paralytics in a very large proportion of cases, and the occasional discovery of this organism in the cerebro-spinal fluid. Although syphilis must be the essential cause of the malady GENERAL PARALYSIS 155 under consideration, it is open to discussion whether other factors may play less important parts in determining the onset of symptoms. It is usual and perhaps justifiable to refer to excessive mental activity and prolonged anxiety and strain in this connection, but the exact value of such influences is largely a matter of guesswork. Alcoholic and venereal ex- cesses are as commonly early symptoms of the disease as predisposing causes. There is no satisfactory evidence that injuries can initiate the morbid process. The onset of general paralysis is commonly during the fourth and fifth decades of life, and this corresponds closely to the age incidence of tabes. Juvenile general paralysis generally displays itself between the ages of eight and eighteen. As in tabes, so in general paralysis, the male sex is more often attacked than the female. Pathogenesis. — Although the histological changes in the central nervous system are characteristic and well known, it is still a matter of debate as to how far the process may be regarded as primarily degenerative, or how far the degeneration of neurons is dependent upon toxic and vascular influences. It must be admitted, in any case, that many of the classical phenomena of the disease are directly due to circulatory and inflammatory disturbances. Probably the neuronic decay is the initial lesion, although its progress may be accelerated by venous stasis, changes in the arterial walls, neuroglial prolifera- tion, meningeal thickening, and the influence of poisons re- sulting from tissue disintegration. Morbid anatomy. — The gross changes observed on the post- mortem table are somewhat various and dependent upon the stage of the disease at which death takes place. In long- standing cases terminated by exhaustion or some low form of pneumonia or broncho-pneumonia, it is usual to find con- siderable general wasting of the central nervous system. In others, which have ended with some sudden seizure, the atrophy may be less general and less remarkable. The skull bones are usually denser than normal and more closely adherent to the subjacent dura. Beneath the latter there may be old or recent extravasations of blood, giving rise to the condition known as pachymeningitis interna haemor- rhagica. The weight of the brain is diminished as a whole, 156 SYPHILIS OF THE NERVOUS SYSTEM and that of the cerebrum proportionately more so than that of the brain-stem and cerebellum. The pia-arachnoid mem- brane is thickened and often opalescent in appearance, especially over the fronto-parietal region. . In the same area the convo- lutions are markedly atrophic, and the intervening sulci are filled with slightly turbid fluid. Sections through the hemi- sphere disclose narrowing and vascularity of the grey matter, and obscuration of the lines of Gennari and Baillarger. The white matter is soft and oedematous. The ventricles may be 1 ^ c f ''. I • • _- '1*; -r . '^ ,j^ ♦. T, ,' > - ■'' -. , i' ■ . , *. " • • " *• " • .'h' ' ' . . ' - * ^ " ' • . '*•«»*' ' •• * * . • * . ■'^i • ,-.•* . \ .:••••• >" : ■■^.m,: . ,. " . « / . ' ;. 1 >:\-\*''-: ''■ ■:^- ..* • -x.y ,• • . . *- /, / '^ ' ^ . % , '" i •• " '■"* v>^ >'•.'. •s, '. ^,. •' • * ' '* , ^ •'•.-.;•-"'>' ;/' , - "•• . /■'• ' '. - . ' • . '"' ' ' " • .' '■-' '- ' f • . . . ' ■ - »•■•.'' *, " * .. ■'. ' ' . .1 .. • ■ \* ' / -" " " ■$ ■■./-■••■ ] ' •; _ ■ ," J Fig. 42. Section from pvae-Rolandic cortex in a case of general paralysis of the insane, showing the presence of large numbers of small round cells in the neighbourhood of a vessel, not confined to the perivascular sheath. dilated, and their ependymal lining is practically always somewhat granular : that of the fourth ventricle is often described as " frosted." Atrophic changes in the optic nerves and some- times in the long tracts of the spinal cord may be visible to the naked eye. Under the microscope the most characteristic changes are to be found in the fronto-central parts of the cerebral hemispheres. Here the tissue is seen to be highly vascular, owing partly to the dilatation of vessels, and partly to the presence of sprouting new capillaries. The vessel walls are rendered GENERAL PARALYSIS 157 Fig. 43. Section of first lumbar segment in a case of general paralysis of the insane, showing degeneration in lateral columns. Fig. 44. Section from lumbar cord in a case of taboparesis. 158 GENERAL PARALYSIS thicker than normal by the proUferation of endotheUal cells, and by the presence of numerous lymphocytes and plasma cells crowding the perivascular spaces. Here and there may be seen isolated elongated cells resembling collapsed capil- laries. The neuroglial cells of all kinds are increased in number, and many spider cells may be detected in the vicinity of vessels with their processes impinging on the latter. The neuroglial fibrils beneath the pia are abundant and form a kind of felt work in that situation. The natural arrangement of nerve cells is upset, and difficulty is experienced in recognising the different layers. Pyramidal cells are destroyed, altered, or displaced; their processes are often ruptured and their dendrites decayed. Chromatolysis, vacuolation and eccentric position of nuclei are prominent features in many cases. The higher association systems of neurons suffer the most, and the supragranular pyramidal cells are especially involved. Sections stained by the Weigert-Pal method reveal atrophy of the tangential, supra- and intra-radial fibres. With the Marchi stain, scattered degeneration may be seen in many systems throughout the central nervous system. Degeneration may be traced in the pyramidal tracts of the brain-stem and spinal cord, and in some cases in which tabetic processes have been present, characteristic sclerosis may be found in the dorsal columns. Occasionally the cells of the ventral horns and of the dorsal root ganglia show chromolytic changes. Even the Purkinje cells of the cerebellum may be involved in a similar manner. The relation of the anatomical to the clinical phenomena. — There are many clinical varieties of paralytic dementia; some are characterised by profound and early mental changes, while others are equally remarkable for their physical disabilities. Most, however, exhibit at an early period some impairment of the highest intellectual and moral qualities, a defect which may be correlated with the changes we have described in the as- sociation systems of the cerebral cortex. Motor paralysis is dependent on the involvement of the Betz cells of the pre- Rolandic area, and on the consequent degeneration of the pyramidal tracts. Various transitory phenomena, such as congestive attacks, epileptiform seizures, hemiplegias, aphasia, etc., depend on irregularities of circulation, and on the toxic or irritative products of the degenerative process. SYPHILIS OF THE NERVOUS SYSTEM 159 REFERENCES Good descriptions of the morbid anatomy of syphilitic diseases of the nervous system are to be found in most standard textbooks. The following are recommended, and contain full bibliographies: Allbutt and Rolleston: System of Medicine, vols. vi. and vii., 1910. CoRNiL ET Ranvier: MuHuel d'Histologie Pathologique, vol. iii. Paris, 1907, pp. 263-356. Flatau, Minor and Jacobson: Handbuch der path. Anat. des Nervensy stems, 1904. Oppenheim: Textbook of Nervous Diseases: trans, by A. Bruce, 1911. Recent Articles. Fearnsides, E. G., Head, H., McIntosh, J., and Fildes, P.: Brain, 1914, vol. xxxvi., p. I. Fildes, P., and McIntosh, J.: Brain, 1915, vol. xxxvii., pp. 141 and 401. CHAPTER VI OTHER INFECTIVE DISEASES I. Leprosy. Aetiology. — Leprosy is due to infection of the tissues of the body with the Bacillus Icp.a^. How this infection occurs is uncertain, but there is some evidence which suggests that it comes through the bite of bugs. The incubation period is a long one, probably some months at least. Having established a foothold under the skin or mucous membranes, the lepra bacilli may attack almost all the tissues of the body, but it is only necessary here to describe their effects on the nervous system. Morbid anatomy. — The bacilli are found in the lepra cells of leprous nodules. These are large connective-tissue cells, varying from thrice the size of a leucocyte to cells of giant size. The smaller varieties have one or more nuclei, usually eccentric ; the larger are always multinuclear. Their cytoplasm is homogeneous, and contains vacuoles in which the bacilli lie packed in sheaves, or arranged radially. The lepra cells usually occupy the centre of a leprous nodule, which is com- posed of epithelioid and connective-tissue cells, with plasma and mast cells in varying number. These nodules may be discrete or form a confluent fibrous mass of firm consistency. In addition to being present in lepra cells, the bacilli may occur in rounded clumps suggesting bacillary thrombi in the lymphatics or the remains of disintegrated lepra cells. The number of bacilli in a nodule is enormous, and gives it, at first glance, a diffuse red stain in sections stained by the Ziehl Neelsen process. The nerve trunks are invaded, usually in their peripheral portions, by leprous nodules, which give rise to an inter- stitial neuritis producing great thickening of the nerve trunks with irregularly spaced fusiform swellings along their course. The amount of paralysis produced is notoriously at variance with the amount of this thickening. In the i6o LEPROSY i6i nodular form the paralysis and anaesthesia are minimal in spite of nerve trunks which are palpable as hard irregular cords, whereas in the anaesthetic form an advanced degree of anaesthesia and trophic changes in the limbs may be present before any obvious nervous swelling. Neisser attributes this to a variation in the bacillus, which, in the latter form, seems to have a specially toxic action on nervous tissue. It is probable that in the majority of cases the bacilli gain access to the nerve fibre at or near its termination, and infiltrate its sheath from the point of access upwards. But in some cases they must be carried either by the blood or lymph channels direct to the upper parts of the nerve, as nodules are often found there independently of disease lower down the nerve, in which the secondary degeneration resulting from the patch of interstitial neuritis is the only obvious change. Cross section of the nerves shows that the different bundles of which the nerve is composed are attacked in varying degree. Some appear as cords of fibrous tissue in which run denuded axis cylinders. Others are intact except for a thick collar of connective tissue, which exercises a certain amount of pressure. The myelin tends to break up, but may persist in spite of much perineural infiltration. In the last degree, after the leprous nodules have invaded the interfibrillar substance, lepra bacilli attack the neurolemma sheath, and form colonies in the migratory cells given off from its internal surface (p. i6). The spinal ganglia rank next to the peripheral nerves as a favourite site of leprous invasion. They may appear normal, or may be enlarged and sclerosed. The bacilli are most commonly found in the nerve cells, which may show little reaction or may be enlarged and deformed or smaller than normal. The bacilli occupy the cytoplasm, and the chromo- phil substance tends to diminish as the bacilli multiply. In the cord the bacilli are most frequently found in ventral horn cells, although not so commonly as in the spinal ganglion cells. The changes produced are similar in the two forms of cell. Bacilli have also been described in the peri-ependymal tissue. Their presence in the cord seems to give rise to very little reaction. Bacilli are rarely found in the brain, but their presence in the Gasserian ganglion is not uncommon. They have also been described in the sympathetic ganglion cells. i62 INFECTIVE DISEASES Degenerative lesions of the cord, although not constant, are present in a large proportion of cases. These are confined to the dorsal columns, where the fibres arising in the root ganglia of the limb plexuses, lumbo-sacral and cervical, are chiefly affected. This gives the dorsal columns in the upper cervical region a tigroid appearance in sections stained by Weigert's method, as degenerated bands of fibres alternate FIG. 45. Leprosy bacilli in a dorsal root ganglion cell. with healthy bands. This degeneration may be due to in- vasion of the spinal ganglia by the bacilli or to toxins ascending in the nerve sheaths from affected nerve trunks. Its relative frequency and its histological characteristics seem to favour the latter hypothesis. Affected tracts are thinner and firmer than normal, and show some meningeal thickening over them, especially near the entrance of the dorsal roots. Similar meningeal reaction may be seen around the cranial nerves. ACTINOMYCOSIS 163 2. Streptothrix Infection (Actinomycosis). Various members of the genus Streptothrix attack the meninges, producing a suppurative pachymeningitis, a diffuse leptomeningitis, or abscess of the brain or spinal cord. Usually the latter are secondary to disease of the lungs or other organs, but cases have been described in which no focus of disease outside the nervous system could be found. In the brain the abscess may arise in any situation. The case where a small tumour of the size of a hazel-nut in the third ventricle proved to be a thin-walled actinomycotic abscess is probably the earliest in the literature. Usually the abscesses are found in the substance of the cerebral hemispheres. They may be single or multiple, and their walls may be definite and firm, or irregular and necrotic, in which case the surrounding brain substance may be unhealthy and gelatinous. The contents of the abscess are described as thick, viscous, mucoid, or oily, and resembling in colour the contents of a compound cystic ovarian tumour. Sometimes the contents are more greyish-yellow in colour; they never resemble ordinary pus. The odour is usually foul and rancid. Histologically the con- tents of the abscess are found to consist mainly of degenerated polymorphonuclear cells, with an admixture of large and small mononuclears. Clumps of the streptothrix are found towards the centre of the abscess. The walls consist of brain tissue, more or less necrotic, infiltrated with polymorpho- and mono- nuclear cells. Around this is a zone of oedema, in which haemorrhages may have occurred. 3. Tuberculosis. The nervous system is attacked by tubercular disease with a frequency surpassed only by syphilis. In fact, apart from severe epidemics of cerebro-spinal meningitis or poliomyelitis, there is no doubt that tuberculosis is the most common infective disease of the nervous system during childhood and adolescence. There is no need to enter here into the aetiology of the disease or to discuss the relative importance of heredity and environ- ment in determining its onset; suffice it to say that recent work is bringing more and more into light the important part played by the bovine bacillus in causing disease in the earlier years of life. i64 INFECTIVE DISEASES From the clinical as well as from the pathological standpoint tuberculosis of the central nervous system divides itself naturally into three classes, according as it attacks (i) the cranial bones and vertebrae, and the dura mater lining them, (2) the soft meninges, (3) the substance of the brain and spinal cord. Not only does the course of the disease differ in the three cases, but the differences in the tissues, both as regards their texture and vascularity, and the facilities which they afford for the spread of the disease, produce variations in the minute pathology of the process. It must be remembered, however, that in its essentials tuberculosis as it occurs in the central nervous system differs in no respect from the forms assumed elsewhere in the body. Here, as elsewhere, we have the acute miliary, the chronic tubercular, and the caseous forms, and all possible combinations of these. (a) Tuberculosis of the cranium, vertebrae, and dura mater.— Tubercular disease of the dura mater must be considered along with disease of the bony coverings of the brain and spinal cord, as this membrane is necessarily affected before the morbid process can injure the central nervous system. Primary disease of the dura mater is rare; in almost every case it is possible to trace the disease to a focus of osteitis in the neighbouring bones. Tubercular disease of the cranium occurs in the mastoid cells as a sequel of tubercular otitis media, but rarely elsewhere. Here there is usually mixed infection with septic bacteria, and in default of surgical intervention the disease may spread rapidly inwards, pierce the dura mater, and produce either generalised leptomeningitis or abscess of the brain. Unless the dura mater is passed the disease does not affect the central nervous system, although it may produce paralysis of the seventh and eighth cranial nerves. Tubercular disease of the vertebrae (Pott's disease) is common, and is a frequent cause of paraplegia. It is usually secondary to disease elsewhere in the body, but apparently may be primary. It occurs as an osteomyelitis of the vertebrae, commencing either in the cancellous tissue of the body close to an inter- vertebral disc, or at the root of the laminae. The vertebral body may be considerably eroded before there is any evidence of disease outside it, and in such cases we have well-marked TUBERCULOSIS 165 bony deformity with no sign of abscess formation. On the other hand, an abscess, originating in a small focus of bone disease, may track in various directions and attain a great size. All stages between these two extremes may be present, and in any of them the spinal cord may be implicated; but it is not often that bony deformity alone produces paraplegia, nor is the presence of a psoas or lumbar abscess often associated with nervous symptoms. The most usual cause of these is either an abscess pressing backwards into the vertebral canal from a focus of caries, or a tubercular pachymeningitis. (i) An abscess of the vertebral body may burst through the posterior common ligament and press directly against the anterior surface of the dura mater. More frequently it spreads round the ligament in a horseshoe shape, and presses on the cord on its ventro-lateral surfaces. In these cases the pus is thick, and the abscess is surrounded by a firm wall. (2) More commonly without much liquefaction the tubercular granula- tions fill up the epidural space and encircle the dura mater more or less extensively. The outer layers of the membrane become infiltrated, and the disease spreads up and down and around it, taking the form of a signet ring or a cuff of thickening. The subarachnoid space may be completely obliterated and the cord may be slowly compressed, or the vascular supply of one segment may be interfered with, leading to thrombotic or ischaemic softening. It is seldom that the soft meninges are invaded and there is usually no excess of lymphocytes in the cerebro-spinal fluid. On the other hand, it is not uncommon for layers of false membrane to be deposited on the inner surface of the dura mater in the subdural space. The cord itself is rarely invaded, but tubercles may spread along the walls of the vessels into the substance of the cord without causing a generalised lepto- meningitis. The latter accident may occur, however, in which case it runs its usual rapid course. Occasionally the displacement of a sequestrum backwards from a diseased vertebral body may cause direct pressure on the cord, or a sharp kyphosis may produce such a degree of flexion of the cord that the segments at the bend are damaged, and cease to transmit nervous impulses. Sometimes the amount of paralysis is out of all proportion to any apparent cause, either i66 SPINAL CARIES in the bone or dura mater. Such cases may be due to occlusion of vessels by local endarteritis. The nerve roots are often implicated, either by tubercular infiltration of their substance, or of the membranes immediately surrounding them, or by Fig. 46. Caries of lumbar vertebrae. Almost complete destruction of one vertebral body, with granulations invading the spinal canal. being compressed between the laminae when there is sub- sidence of the vertebral bodies. Tubercular disease of the vertebrae usually progresses favourably and results in cure, and there may be a corre- sponding improvement in the nervous symptoms, especially TUBERCULAR LEPTOMENINGITIS 167 when they have been of short duration. But when, in the process of heahng, the kyphosis becomes more marked, a sharp angle may be left against which the cord is rubbed and lacerated by the movements of the body. Again, the cicatricial fibrosis of the peridural tissues may leave a tight sleeve around the cord which prevents its recovery. Lastly, in long-standing cases there may be such a degree of neuroglial sclerosis that only a very imperfect restoration of the nerve tracts is possible. The condition of the cord is similar to that found in com- pression from other causes. According to the rapidity of the process there is more or less oedema, myelin destruction and neuroglial overgrowth. The axis cylinders frequently persist in a remarkable manner, and when the compression is relieved in the process of healing the tracts may resume good functional activity. (b) Tubercular leptomeningitis. — This is probably always secondary to a focus of tubercular disease somewhere in the body, and is often a part of generalised acute miliary tuber- culosis. In young children the primary focus is commonly a caseating mediastinal gland. Later in life it may be a tubercular joint lesion of long standing, especially where sinuses are present, or a tubercular focus of greater or less activity in the lungs or abdomen. In a certain number of cases the meninges are affected by a direct spread from caries of the vertebral bodies. Although tubercular meningitis is pre-eminently a disease of young children, and shows its greatest incidence during the second year of life, it does not seem to be due at all frequently to the bovine type of bacillus ; but more ample statistics on this point are wanted. Morbid anatomj^. — On removing the skull cap and dura mater one is struck by the extreme softness of the brain, which tears with the utmost readiness. The surface is pale; the arachnoid is slightly greasy to the touch, and may be more opaque than normal, and along the vessels in the sulci a few whitish tubercles may be seen. The veins over the cortex are not usually dis- tended. The convolutions may be flattened and the sulci somewhat less obvious than normal. At the base of the brain a much more evident exudation is seen. This may appear as a thickening of the arachnoid of a greenish-grey colour, showing 168 INFECTIVE DISEASES numerous tubercles scattered over it and spread along the vessels, or the base may be covered with a greenish gelatinous exudate which spreads forwards to cover the tips of the temporo-sphenoidal lobes. Or, again, the membranes at the base may be matted into a dense cream-coloured mass, com- posed of caseating fibrous tissue. The Sylvian fissures are usually firmly tacked down by adhesions, and on separating the lobes of the brain the Sylvian arteries are found to be covered with miliary tubercles. On section of the brain more or less hydrocephalus may be apparent, but the extreme softness of the brain makes it very difiicult to gauge the amount of this. The choroid plexus and velum interpositum may be covered with tubercles, and the latter is often bound down by tubercular adhesions. The walls of the ventricles, even where the ependyma is intact, are often extremely soft and friable. Small areas of capillary haemorrhages and yellow patches of softening may be found in the brain substance, caused by invasion of the vessel walls by the tubercular process. The dura mater is usually healthy, but may contain a few small miliary tubercular nodules, or one or more larger caseous masses. Microscopically the pathological changes are mainly confined to the meninges, where a varying amount of exudation is present. This may take the form of definite tubercles, most of which lack giant cells, or it may be more or less diffusely spread over the membranes forming nests of epithelioid cells. These usually surround the smaller vessels and penetrate through their walls, eventually filHng up and blocking their lumen. The cells composing these are mononuclear cells of all kinds; many plasma cells and large macrophages are present, and connective-tissue cells are seen proliferating in many parts. Polymorphonuclear cells often form a ring round these nests, and as they degenerate are ingested by the macrophages. Caseation usually takes place first in the centre of the nodule, but rarely reaches a very extreme degree. The substance of the brain and cord shows little or no in- flammatory reaction. The walls of the vessels show some proliferation of the connective tissue and endothelial cells, and there may be an accumulation of small cells in the TUBERCULOMATA 169 adventitial lymph spaces. On the other hand, evidences of toxic degeneration of the nervous tissues are everywhere seen. The nerve cells show a varying degree of degenerative change and may be greatly affected. The axis cylinders also break up, and the brain tissue is oedematous. This feature of Fig. 47. Section of a vessel on the cerebral cortex the seat of tubercular arteritis. the disease is so marked that many writers have termed it tubercular meningo-encephalitis. {c) Tuberculomata are among the commonest tumours of the central nervous system. In the hrain they may occur in any situation, being probably more common in those parts which, fill the posterior cranial 170 INFECTIVE DISEASES fossa. They are frequently multiple and of small size, varying from I to 5 mm. in diameter. In other cases solitary tubercles of a larger size, upwards of i cm. in diameter, are found. On section they may appear to blend intimately with the sur- rounding nervous tissue, or may be definitely encapsuled. They are usually spherical in shape, but where several small tubercles have grown into one another the appearance may be more irregular. The outer zone is usually of a pinkish-grey colour, and merges gradually into the yellow caseous centre. Fig. 48. Photograph of a section from the medulla oblongata, showing a tuberculoma lying dorsal to one olive. The surrounding nervous substance may be obviously softened or apparently normal, but the ease with which tuberculomata may be enucleated suggests that some degree of softening of the surrounding tissues is always present. It is unusual for tuberculomata in the central nervous system to break down and form abscesses ; usually they do not go beyond the stage of caseation, but except in the very smallest nodules a caseous centre is always present. In the cord also they may be multiple or single. The latter ACUTE LEPTOMENINGITIS 171 are necessarily of smaller size than in the brain and tend to a more elongated shape. They are most frequent in the cervical enlargement. Localised tubercular meningo-myelitis may occur. In this case the tubercles are found in the pia-arachnoid, and to some extent invading the cord along the vessels. For the most part, however, the softening of the cord is due, not to tubercular infiltration, but to vascular occlusion and the effects of tuber- cular toxins. Tuberculomata of varying size may also be found in relation to the soft meninges of the brain without giving rise to a generalised tubercular leptomeningitis. Their size and appearance is similar to that of tubercles in the brain substance. Microscopically tuberculomata of the central nervous system usually show a caseous centre, which occupies the greater part of the tumour. This is surrounded by a zone either of giant- celled nodules, or of infiltration with epithelioid cells, along with destruction of the nervous elements. This forms the outer layer of the tuberculoma. Around it is a zone which shows evidence of toxic degeneration of nerve cells and fibres with little or no phagocytic infiltration, unless we class under that heading the large number of granular corpuscles which is present. The vessels are congested, and run through this area into the outer layer of the tuberculoma, where they become thrombosed. It is sometimes difficult to define exactly the boundaries of the nodule, as the passage from the zone of epithelioid cells to that of degenerated nervous substance may be gradual and its contour irregular. When, however, giant cells are present they form a useful indication of the outer strata of the tuberculoma. 4. Acute Leptomeningitis. According as the inflammatory process attacks chiefly the pia-arachnoid membranes or the dura mater, the terms lepto- and pachy-meningitis are used to denote it. It is at first sight surprising that in acute leptomeningitis there is little tendency for the disease to spread outwards and involve the dura mater. But it must be remembered that though the membranes are in contact along their endothelial surfaces, there is no communica- tion between the subdural and subarachnoid spaces. On the other hand, infection readily spreads from without inwards. 172 INFECTIVE DISEASES by perforation of the dura mater, and thus collections of pus forming in the bones of the skull or the vertebral column may burst through into the subarachnoid space, and give rise to a generalised leptomeningitis. There are two main factors which render leptomeningitis from any cause one of the most acute and fatal forms of infection. 1. We have to deal with inflammation in a space wherein there is continual movement and circulation of a fluid which has minimal antitoxic and bactericidal powers and yet affords just sufficient pabulum for the growth of organisms in it. Owing to the absence of lymph and the continual movement of the fluid, there is very little tendency for inflamed areas to be shut off by adhesions; rather the tendency is for the infection to be washed from one place to another. Conse- quently infection introduced into any part of the subarachnoid area rapidly becomes diffused generally throughout it. This is especially true of parts below the tentorium cerebelli, as over the cerebrum inflammatory foci are frequently limited by adhesions in the pia-arachnoid. 2. The central nervous system is enclosed in a rigid bony box which does not allow of more than a very limited enlarge- ment of the brain, and checks any increase in the cranial contents. Therefore, when meningeal inflammation raises the intracranial pressure, the blood supply to the brain is diminished, and the vital centres, already poisoned by bacterial toxins, are deprived of nutrition. This rise of intracranial pressure may be due to one or more of the following factors: (i.) Adhesions forming round the base of the brain may shut off the escape of the cerebro-spinal fluid from the cisterna magna and lead to an internal hydrocephalus. This is seen typically in the more chronic forms of meningococcal meningitis. (ii.) Obstruction to the exit of cerebro-spinal fluid into the general circulation may be caused by inflammation and oedema of the walls of the cerebral veins, and of the Pacchionian bodies, and by proliferation of tissues and cells in the meshes of the subarachnoid space. (iii.) The brain substance itself often becomes oedematous owing to the influence of bacterial poisons. MENINGOCOCCAL MENINGITIS 173 The causes of acute leptomeningitis may be classified as follows: ^ Primary r Sporadic type. Meningococcic \ Epidemic cerebro - spinal Secondary Pneumococcic Traumatic Tuberculous Pneumococcic Pyogenetic Other forms \ menmgitis. Injuries of scalp, skull, etc. Tubercular infection. Infection of ear, nose, orbit, etc. Pneumonia, empyema, en- docarditis, peritonitis, etc. Infection by various forms of staphylococci or strepto- cocci derived from the skin or internal organs. Infection by B. typhosus, coli, enteritidis{G2ieTtneT),dysen- teriae, influenzae, Gonococ- cus, Streptothrix, Lepto- thrix, and B. anthracis. {a) Meningococcal meningitis — Aetiology. — The meningo- coccus of Weichselbaum is a kidney-shaped coccus, occurring in pairs with the flattened surfaces opposed. In morphological characters it closely resembles the gonococcus, and like the latter is not stained by Gram's method. Various strains of the organism react differently in agglutinative reactions, and these differences are most marked as distinguishing the epidemic form of the disease from the chronic posterior basic type of Still. Usually, also, the organism of the latter disease shows a hardier growth in culture. Undoubtedly the organism is subject to many variations, both in virulence and in character. In the tissues and exudate it occurs typically as an enclosure in the polymorphonuclear cells, and resembles the gonococcus in occurring packed in large numbers in single cells, while the great majority of leucocytes remain free from any organisms. As many as fifty cocci may be counted in a cell in the more acute cases, but often several fields must be searched before any 174 INFECTIVE DISEASES organisms are found. Cocci are also found extracellularly both in acute and chronic cases. The mode of entrance of the meningococcus to the subarach- noid space has been disputed. Some observers favour the naso- pharyngeal route, and consider that the organism finds its way Fig. 49. Post-basic meningitis. along the lymphatics surrounding the nasal and olfactory nerves. Others state that the symptoms are usually at first abdominal, and that the organism gains entrance to the spinal area from the abdominal lymphatics. Morbid anatomy. — By whichever route the organism gains the meninges the effects are similar. On removing the cover- POST-BASIC MENINGITIS 175 ings of the brain the cerebral hemispheres are seen to be covered with a layer of greenish-yellow pus, which may be confined to the sulci or spread in a more even layer all over the cortex. The cortical veins are greatly distended and dark in colour. The lymph is entirely in the meshes of the arachnoid mem- brane, and so is not easily washed away by a stream of water. When the exudation is more confined to the base it fills the basal cistern, and extends downwards over the pons and medulla and forwards over the poles of the temporal lobes, but does not tend to spread round the Sylvian fissure in the same way as that of tuberculous meningitis. On stripping the exudate from the brain cortex, this is seen to be firm and of almost normal colour. The convolutions are often flattened and the sulci filled up owing to internal hydrocephalus, and in cases of long duration, especially where the intracranial pressure has been greatly increased, the cortex may appear pale. Microscopically the infiltration of the meninges is found to be practically confined to the subarachnoid space. Here, according to the acuteness of the infection, we find an exudate consisting wholly of polymorphonuclear cells, or of these with a greater or smaller number of mononuclear cells of various types — lymphocytes, plasma cells, and macrophages. Phago- cytosis is commonly observed. The polymorphonuclear forms show vacuoles containing organisms or other debris, and as they degenerate are themselves ingested by the larger mono- nuclear cells. The cortex shows very little inflammatory reaction except congestion of vessels and accumulation of small cells in the adventitial lymph spaces. In chronic posterior basic meningitis the exudate is confined entirely to the base, and consists merely of a chronic thickening of the pia-arachnoid membranes in this area, with little or no lymph exudation. In these cases hydrocephalus forms the outstanding pathological feature. (b) Meningitis due to other pyogenic organisms. — Menin- gitis due to other pyogenic organisms gives pathological appearances similar to the above — the amount of exudate varying with the acuteness and severity of the inflammation. The form due to the Bacillus anthracis gives a very characteristic 176 SEROUS MENINGITIS picture at the autopsy, as the surface of the brain is covered with a bright red stain due to multiple haemorrhages into the subarachnoid space. The characters of the cerebro-spinal fluid in meningitis have already been discussed (pp. 37-48). The pressure is always raised, but the flow may be retarded by the thick consistency of the fluid. (c) Serous meningitis and meningism. — These terms are mainly clinical in significance. They have been applied to cases pre- senting meningeal symptoms which pass off either rapidly or gradually, sometimes leaving hydrocephalus in their train. Aetiology. — Almost all the organisms which have been mentioned as causing meningitis may give rise to the conditions described under the term " serous meningitis," if they occur in attenuated form or if the natural resistance of the patient is raised. One cause may be the tubercle bacillus; it was con- sidered until recently that tubercular meningitis was always fatal, but the discovery of the bacillus in the spinal fluid of cases of mild meningitis which recover has proved that this is not the case. A more common cause is disease of the middle or internal ear, and in this connection it must be remembered that the perilymph of the internal ear is in communication with the cerebro-spinal fluid in the subarachnoid space. The specific fevers and acute pneumonia (especially when the apices of the lungs are infected) may be accompanied by signs of meningitis. This is readily understood on the theory of absorption of toxic substances along the spinal nerves, or by assuming a meningitis of low virulence. Morbid anatomy. — The post-mortem appearances which have been described as " serous meningitis " are, for the most part, those of internal hydrocephalus, due to closure of the foramina round the fourth ventricle by meningeal thickening or by ependymal adhesions. One ventricle alone may be dilated: the lateral when one foramen of Munro becomes closed, or the third or fourth when the Sylvian iter is blocked. Perhaps the term most appropriately describes the rare cases of " localised serous meningitis " where meningeal adhesions have led to accumulations of fluid over a limited area of the cortex. In one such case adhesions between the upper surface of the cerebellum and the tentorium had dammed back the PYOGENIC PACHYMENINGITIS 177 fluid in the posterior cranial fossa and led to signs of cerebellar disease. In other cases the Rolandic area of the cortex has been compressed by a cyst of similar origin. Thrombosis of the longitudinal sinus may give rise to a clinical picture indistinguishable from hydrocephalus, but on pathological grounds the term " serous meningitis " is here quite inadmissible. 5. Pyogenic Pachymeningitis. Aetiology. — Pyogenic infection of the dura mater surrounding the brain or spinal cord usually occurs secondarily to disease in the overlying bones. In the skull this commonly occurs from infection of the middle ear or of the air sinuses, especially the mastoid cells and frontal sinus, more rarely the ethmoidal or sphenoidal cells. It may also result from compound fractures of the skull. The causative organisms most frequently found are staphylo-, pneumo-, and strepto- cocci, and members of the B. coli group, but any of the pyogenic organisms may be responsible either singly or in combination. The dura mater lining the skull is extremely resistant to septic infection, and, were it not for two anatomical char- acteristics, would prove an impassable barrier. These are (i) the presence of venous sinuses. These communicate on one side with emissary veins running through the cranium, and on the other with the veins and venous lacunae of the arachnoid membrane, and thus provide channels whereby septic processes may pass through the dura mater to the lepto- meninges and thence to the brain itself. In the case of the larger sinuses septic clots may become detached and be carried to the heart and lungs. Again, thrombosis of the superior longitudinal sinus, which may result from inflammation of the frontal air sinus or from fractures of the vertex of the skull by wounds or otherwise, causes congestion and oedema of the cortex by blocking the normal venous return from the cortical vessels. (2) The dura mater lies between the resistant skull and the less resistant brain tissue. Consequently, when pus appears on the inner side of the cranium, it tends to bulge the dura mater inwards, and strip it from its attachments to the cranial bones. When adhesions form the tension of the pus on the inflamed dura mater may cause it to give way, and a 12 178 INFECTIVE DISEASES communication is thus formed through it between the skull and the subdural and subarachnoid spaces. Once the barrier of the dura mater is passed the inflammation usually finds its way rapidly through the outer layer of the arachnoid and leads to a generalised leptomeningitis, to purulent cerebritis or to abscess of the brain. Usually, unless both these factors come into play, i.e. unless it takes place in a region of the dura mater whence emissary veins are given off and where the overlying bone is more or less intact, the inflammatory process fails to penetrate the dura mater. Thus, large abscesses may form between the dura and the parietal bones ; even over the frontal and mastoid regions removal of the overlying bone is usually suflicient to arrest the progress of the inflammation. The most vulnerable region seems to be the neighbourhood of the superior longi- tudinal sinus. It must be remembered that here the meningeal and cortical veins, running in the dura and the arachnoid respectively, pour their blood into the sinus, which also receives lymph from the subdural space by short channels running in its walls. The latter are in all probability the most common paths whereby infection passes through the dura mater in this region, but it may also pass along the cortical veins from an infected and thrombosed sinus. In regard to the dura covering the spinal cord neither of these anatomical factors is present. Except with the emerging nerve roots no veins traverse the membrane, and, laterally and posteriorly at any rate, there is little continuity in the bony covering. Even in the sacrum there are large foramina whereby pus surrounding the theca can escape, and although inflammation starting in the bodies of the vertebrae may penetrate the dura and lead to leptomeningitis, such an event is rare even in the sacral region and very rare elsewhere in the spine. On the oth-er hand, an abscess starting in the bodies of the vertebrae may press the dura backwards against the cord, or may produce a circular thickening of the theca, and in either case the function of the cord may be interfered with (p. 164). Infection through the spinal theca occurs most frequently in bed-sores following lesions of the spinal cord, as the cutting off of trophic influences renders the tissues more vulnerable to septic processes. ABSCESS OF THE BRAIN 179 Pyogenic inflammation of the dura mater differs in no respect from inflammation of a connective-tissue membrane elsewhere in the body. There may be diffuse infiltration with pus cells but no macroscopic abscess, or the layers of the dura may be separated by a stratum of pus cells; or, again, the membrane may be greatly thickened and contain small abscesses varying from I to 5 mm. in diameter. Where a tunnel has been formed through the dura its walls are formed by pus cells and granulation tissue with degenerated fibrous tissue elements. 6. Suppurative Encephalitis (Abscess of the Brain). Aetiology. — Suppurative encephalitis is due to the ordinary pyogenic organisms, most frequently the streptococci and pneumococci, less often Staphylococcus aureus, Bacillus coli and Friedlander's bacillus: anaerobic bacilli may also be present. Abscesses due to the tubercle bacillus (p. 169) and streptothrix (p. 163) have already been described. Abscess of the brain has four main causes : (i) It may arise as a result of compound fracture of the skull, especially when accompanied by some laceration of the brain. This is a direct infection of the brain substance from the overlying tissues. (2) It is frequently seen as a result of acute or chronic suppuration in the bones of the cranium, especially in otitis media and mastoiditis. (3) It is not uncommon in cases of purulent bronchitis or bronchiectasis. (4) It may be due to organisms circulating in the blood. Thus multiple abscesses may occur in pyaemia, and, less frequently, so-called " idiopathic " abscess of the brain may be due to invasions of the brain substance by an organism derived from some hidden focus of inflammation or suppuration. Such abscesses may arise after some contusion or concussion of the brain, where there is no possibility that the brain tissue has been invaded directly from the skin. Macroscopic appearances. — Abscesses of the brain may present different features according to their cause. Thus, abscesses arising from wounds of the skull and brain usually i8o ABSCESS OF THE BRAIN show a track of red, softened brain tissue leading to the fracture of the cranial vault. When they occur around some foreign body, such as a splinter of bone, a fragment of shell or bomb- casing, or a bullet, there may be an open track surrounded by bruised brain tissue. In some cases the foreign body, as it passes through the brain, leaves septic material behind it, and an abscess results in the middle of the track, while the foreign body itself lies at the end of the track surrounded by healthy brain tissue. In wounds of the brain it is common enough to find a chain of abscesses of diminishing size apparently separate from one another, but connected by a zone of hyperaemic brain tissue, and the difference in thickness and firmness of the capsules of these abscesses gives an indication as to their relative age. Such abscesses may be of any shape : when formed in the track of a foreign body they tend to run along the track; not in- frequently they are rounded, especially in the more chronic cases, but they may have numerous finger-like extensions separated by spurs or bands of firm tissue. The limiting wall also varies considerably. In some, especially in the earlier stages, it is irregular and indefinite, apparently consisting merely of haemorrhagic and softened brain tissue. In others, which are older, the wall is thick and firm, and yellowish in colour, and may strip with the greatest ease from the softened brain tissue surrounding it, so that the abscess may be enucleated without rupture. Abscesses resulting from mastoiditis are usually found in the temporal lobe or in the cerebellar hemisphere on the same side as the ear disease ; but they may occur on the opposite side of the brain, probably as a result of the infection having spread along the petrosal sinuses or the lymphatics. Usually there is evidence of disease of the dura mater, and a track of hyperaemic brain tissue joins the abscess with the meninges. But in other cases it is not possible to trace any connection of the abscess with the surface of the brain. These abscesses may become chronic and attain a large size: they are usually single and rounded, but may be multiple. Abscesses of the brain arising in the course of chronic lung disease are usually single and are most often found in the left cerebral hemisphere, but they may occur anywhere Fig. 50. Right tempovo- sphenoidal abscess secondary to middle-ear disease. Fig. 51. Bilateral abscesses in frontal lobes secondary to sphenoidal sinusitis. Fig. 52. Cavity in left cerebellar lobe, the result of an abscess secondary to middle. ear disease. i82 INFECTIVE DISEASES in the brain, cerebellum or brain-stem. The pus is frequently foetid. The evolution of abscess from acute encephalitis may be divided into three stages : (i) A stage resembling acute encephalitis, in which there appear areas of softening and liquefaction. (2) These areas enlarge, merge into one another, and tend to lose their red colour; fully formed pus makes its appearance, at first in minute drops. (3) In the third stage the reaction in the surrounding brain tissue limits the abscess, a definite wall is formed, and the surrounding encephalitis subsides. This stage may never be reached, and the encephalitis may continue its spread until the death of the patient. The pus of a cerebral abscess is usually yellowish and creamy, but may have a reddish or greenish tinge. Although in most cases it is odourless it may be extremely offensive, and where anaerobic bacilli are present it has a characteristic acrid smell. Microscopic appearances. — From within outwards the walls of a brain abscess consist of : (i) A layer of pus cells, mainly degenerated polymorpho- nuclear leucocytes, with an admixture of larger cells with pyknotic nuclei and granular cytoplasm which may contain fatty substances. Calcification may be present in this layer, sometimes forming a ring near its junction with the layer next outside it. (2) Granulation tissue in process of organi- sation. Connective-tissue fibres form a large part of its structure, and small blood vessels with relatively abundant fibrous tissue in their walls are plentiful, both these structures tending to lie parallel to the wall. It is obvious that the fibrous tissue arises from the walls of the vessels, around which it is especially dense. Few glia cells or fibres are present in this layer, but in the earlier stages there are numerous fat- containing phagocytic cells (compound granular corpuscles). It is by this layer that the abscess is limited, and it eventually forms the firm wall of a chronic abscess. (3) The zone of glial reaction. Here the fibrous tissue is confined to the walls of the blood vessels, round which it forms a much thicker sheath than in healthy brain tissue. In recent abscesses there is dilatation of the vessels and proliferative activity of ABSCESS OF THE BRAIN 183 their walls, with swelling and multiplication of the lining endothelial cells and the formation of numerous new capillaries. The adventitial sheaths of the larger vessels for a considerable distance from the abscess wall are distended with small round cells, plasma cells, and polymorphonuclear leucocytes. The glia cells show the various changes common to all forms of glial reaction. Everywhere they are increased in number, and there may be many large, globular " amoeboid " forms and elongated giant cells, with terminal fibres arranged parallel Ai A*v^ . •*•-*•: ,Y, *-T >.» « » J •■.•■- /■>> .'.••'• •:".■ •-" " '../rr^^-tr^'MP Fig. 53. Section from the wall of a cerebral abscess, showing the formation of fibrous tissue and numerous granular corpuscles. to the abscess wall. The amount and character of the glial reaction varies according to the state of encapsulation and age of the abscess. It does not begin to be apparent until the third week. Eventually the reaction settles down into a gliosis of which fine glial fibres form the basis, with numerous small glial nuclei scattered along them. The neighbouring nerve cells are always damaged to some extent. There is a powdery condition of the Nissl granules or a diffuse staining of the cell body. i84 ABSCESS OF THE SPINAL CORD Suppurative Myelitis (Abscess of the Cord). Aetiology. — This rare condition may be due to causes similar to those of abscess of the brain, and most often arises in pyaemia or in the course of chronic suppurative disease of the lungs or of the genito-urinary tract. It may also be due to direct spread of infection in wounds of the spine. Abscesses of the cord are usually solitary, but may be multiple. They tend to affect the centre of the cord, especially the grey matter, and to run up and down the cord in the form of a spindle, often causing much distension. This appears to be due to the firmness of the surrounding pia mater, which limits the outward growth of the abscess and makes it seek a line of less resistance in the long axis of the cord. The abscess may have definite walls, or it may consist merely of a pale purulent centre in a red area of acute myelitis. It is often surrounded by a zone of suppurative meningitis. The microscopic appearances differ little from those described above in reference to abscess of the brain. Owing to the tendency to spread longitudinally there is less encapsulation by fibrous tissue. 7. Acute Myelitis. The only scientific classification of various forms of acute myelitis would depend upon their bacteriology. In the present state of our knowledge, however, the bacteriology of acute myelitis is not sufficiently advanced to allow of a definite classification on that basis, and it is wiser to divide the forms of myelitis into those which are due to organisms of a non- syphilitic nature and those which are brought about by the action of the Treponema pallidum. It is convenient to call the former class acute infective myelitis and the latter syphilitic myelitis (p. 141), and it should be realised at once that the latter includes probably 80 per cent, of all cases. Infective myelitis. — This is a comparatively rare disease having no definite relation to age or sex, although young adults are probably its most common victims. It differs from acute poliomyelitis in presenting no epidemic form and having no associations with special climates or special seasons. The aetiological value of trauma, chills, strains, or various excesses ACUTE MYELITIS 185 cannot be highly estimated, although some of these factors may have a predisposing influence. The acute specific fevers appear to have a more intimate relationship with the disease, and a number of cases have been recorded in which the onset of the spinal symptoms has occurred in the course or in the sequel of measles, smallpox, typhoid fever, and gonorrhoea. It must not be too hastily assumed that in such cases the infective agent of the fever and that of the myelitis has always been identical, but some observers hold the view that a specific typhoid myelitis for which the typhoid bacilli are actively responsible should be recognised. No other aetiological factor has attracted special attention, but the association of certain forms of myelitis with pregnancy or the puerperium has been noted in several instances, and a toxic myelitis with relapsing features has been described in this connection. Pathogenesis. — It is to be presumed that an infective myelitis can arise in one or other of the following ways : (i) As an extension from inflammatory processes involving neighbouring tissues, such as the meninges or the vertebral column. An example of this process is afforded by the tuber- cular form of myelitis which occurs as a secondary complica- tion in cases of spinal caries or of tubercular meningitis. Similar forms of myelitis are often present in some degree in cases of meningitis due to various other organisms, such as the pneumococcus, staphylococcus, streptococcus, etc. This may not be evident on the clinical side, as the symptoms of myelitis are often overshadowed by those of the meningeal affection. (2) As the result of an infection of the spinal marrow through the blood stream. This may occur in the course of a general pyaemia, or possibly as a purely local phenomenon when the circulating virus affects the spinal cord alone. (3) As the result of an infection through the lymphatic system. It has been shown experimentally that infection of the spinal cord may take place along the course of the lymphatic vessels which accompany the spinal nerves and spinal roots, and that such an infection may give rise to an infective myelitis similar to the disease which is met with in human beings. It is probable that some cases of infective myelitis which occur in association with septic processes of the pelvic, abdominal or thoracic organs are to be explained on these lines. i86 INFECTIVE DISEASES The bacteriology of infective myelitis is only in its in- fancy, its progress being hampered by the fact that many organisms appear to be very short-lived in the spinal tissues, and also by the fact that secondary infections may sometimes lead to wrong conclusions. The ordinary pyogenic organisms have been cultivated from the spinal cord in fatal cases of the disease, and occasionally organisms which are less familiar have been recovered in a similar way. Morbid anatomy. — It has been customary to describe various forms of myelitis as transverse, diffuse or disseminated, accord- ing to the distribution of the morbid process in the cord, although the actual character of the lesion is similar in most cases. For instance, a myelitis which is limited to one, two, three, or at most four segments, and which affects the transverse area of the cord within these limits more or less completely, is usually called a transverse myelitis. A diffuse myelitis involves a considerable length of the spinal marrow without definite interruption, and is often found when the clinical symptoms have suggested an ascending process. A case of this kind is clinically termed an acute ascending myelitis. Disseminated myelitis implies the presence of two or more foci of the disease separated by comparatively healthy tissue. The macroscopical appearances vary, of course, with the length of time which has elapsed since the onset of the disease. In early cases the areas involved are softer than normal, sometimes almost diffluent, swollen, oedematous and generally hyperaemic. Haemorrhages may be present and give a dark red or brown colour to the tissues. The most noticeable feature upon making a transverse section through an area of myelitis is the absence of any definition between the white and grey substances. The soft meninges usually present dilated blood vessels and sometimes afford evidence of serous or purulent inflammation. The exudation is generally more pro- fuse on the posterior than on the anterior aspects of the cord. In cases which are of longer standing the tissues are more shrunken and less vascular, although they may present some mottling with old blood pigment. Sections through the cord at levels distant from the seat of inflammation may show secondary degenerations in the ascending and descending tracts of the white columns. At a still later period the diseased area is of ACUTE MYELITIS 187 firmer consistence, and transverse sections may reveal the presence of one or more cysts. When sections of the cord have been cut and stained by suitable methods, the morbid changes are seen under the microscope to involve the blood vessels, the neuroglia, the nervous tissues and the meninges. All the hlood vessels of the affected area, small and large, show marked engorgement, and many of them are surrounded by large masses of nucleated cells partly crowded in the meshes of their adventitial sheaths and partly in the surrounding tissues. These nucleated cells include lymphocytes, plasma cells, mast cells and polymorphonuclear leucocytes in varying numbers. The proportion of leucocytes to other cells appears to vary considerably and to depend upon factors which are still obscure. In Marchi-stained sections many cells filled with fatty granules may be detected in and around the walls of the blood vessels. Thrombosis is occasionally met with, and haemorrhages into the tissues are frequent. The neuroglia shows evidence of oedema in the shape of spaces which are either empty or filled with an amorphous granular material, deposited in the process of fixation from the albuminous fluid in which the tissues have been bathed. The neuroglial cells are often swollen and sometimes vacuolated. Their nuclei have undergone multiplication and their processes are often lost. They are probably much increased in number, especially in the neighbourhood of the dilated vessels and their lymphatics. The nerve cells of the grey matter become swollen, rounded and homogeneous in appearance. The chromatin granules disappear, and the nuclei are displaced from their central position and may often protrude from the surface of the cell. The axis cylinder processes are swollen and sometimes broken. In some parts the nerve cells may have disappeared altogether. The chief changes in the white matter are those which affect the myelin sheaths. These become swollen and varicose, and rapidly lose their power of staining by the Weigert-Pal method. The Marchi method reveals the presence of fatty changes in the form of darkly stained globules of irregular size and shape. The axis cylinders are often swollen and irregular in outline. When the myelin has disappeared the axis cylinders may remain unsupported, with the result that a rarefied i88 INFECTIVE DISEASES appearance is given to some areas. The meninges show dilated blood vessels and a cellular infiltration of the pial tissues and of the vessel walls. If the spinal cord is examined, by the Marchi method, at a period of two or three weeks after the onset of the disease, it will be seen that much of the fat which results from myelin degeneration has been taken up by large granular cells which are probably of neuroglial origin. Sections through other levels of the cord display in a striking manner the ascending and descending degeneration in the white columns. Still later these degenerations are only visible when the Weigert-Pal method is used. By this time the vessels have ceased to show such marked cellular infiltration, and their walls are thickened and perhaps hyaline in appearance. Neuroglial sclerosis is evident in parts from which the nervous elements have disap- peared, and such areas may show an increase in the number of blood vessels. Secondary degeneration may also be found in the ventral roots and in the efferent fibres of the peripheral nerves coming from levels at which the grey matter has been seriously affected. 8. Tetanus. Tetanus is a disease produced by the exotoxins of a specific micro-organism, the tetanus bacillus, which have a powerful action on the motor cells of the central nervous system. Tetanus bacilli usually gain access to the body through a wound or scratch of the skin, especially when this is contaminated with street dust or highly manured soil. The tetanus bacillus is a common saprophyte of the intestines of horses, and is therefore spread mostly through the medium of road sweepings and horse manure. It is very pathogenic to horses as well as to man, and causes many deaths among new-born foals. Experimental work has shown that if washed spore-free tetanus bacilli are injected under the skin no tetanic symptoms result, but if suppuration is brought about by injecting other pyogenetic organisms along with them, or if their entrance is accompanied by mechanical irritants, such as a splinter of wood, the bacilli multiply in the wound and give rise to the disease. This has borne out the clinical facts that tetanus rarely results from a clean wound, but if suppuration ensues, and still more TETANUS 189 if there is any splintering of neighbouring bones, tetanus is very liable to occur. The bacilli themselves remain localised to the site of infection, and seem to have no power of migration, but the exotoxins formed by them diffuse out into the tissues and reach the central nervous system, on which they have a selective action. Much work has been done on the paths by which this is reached. It was shown in 1902 by Marie and Morax, and a year later by Meyer and Ransome, that the chief path is along the nerves of the affected limb, and it was thought by them that the toxins passed chiefly up the axis cylinders, which were reached by the motor end-plates of the muscles of the limb. Passing thence to the motor cells of the ventral horn of the same side, the toxins were considered to spread up and down the cord either by lymphatic channels or along axis cylinders to the cells supplying the other limbs and the cranial nerve nuclei. They showed that tetanus toxin injected into the muscles of a limb passed rapidly into the nerve, so that an hour after the injection it could already be demonstrated by injection of emulsions of the nerve into smaller animals. Section of the nerve of the injected limb prevented the toxin spreading directly to the nervous system and made its action weaker and more diffuse. Section of the cord after injection of toxin into a hind-limb prevented the spread of symptoms to the fore-limb and head. They also showed that the latent period between the injection of the toxin and the resulting symptoms varied directly with the size of the animal — that is, with the distance which had to be travelled by the toxin. They found that whereas maximal effects are obtained by injection of tetanus toxin into a motor nerve, minimal effects result when it is injected into a purely sensory nerve, such as the infra-orbital. This anomaly has been ex- plained by the more recent work of Teale and Embleton. When toxin is injected into the spinal cord or into a dorsal root between the ganglion and the cord there is evidence of hyperaesthesia and paroxysmal hyperalgesia in the correspond- ing segmental area. This has been called " tetanus dolorosus." Direct injection of the toxin into the brain does not produce the ordinary tetanic symptoms, but causes psychical irritability and epileptiform convulsions and is rapidly fatal (cerebral tetanus). The minimal fatal dose in this case is very much 190 INFECTIVE DISEASES smaller and the incubation period much shorter than with subcutaneous injection. It has been shown by Wassermann and Takaki that emulsion of brain of a susceptible animal can neutralise tetanus toxin so that injection of the supernatant fluid from a mixture of these two produces no evil effects. It was suggested at first that this was due to the formation of antibodies by the cells of the central nervous system, but it seems more likely that, as Roux and Borrell suggested, this is an adsorption phenomenon or loose chemical union between the brain substance and the toxin. The experiments on the effects of antitoxin have afforded confirmatory evidence of the selective affinity of the toxin for the nervous system; thus death results after the injection of a fatal dose of toxin directly into the sciatic nerve of an animal, even when a dose of antitoxin, sufficient to neutralise it, has previously been injected by the intravenous or subcutaneous route. Further, the injection of antitoxin into the main nerve of the limb will prevent the effects of a previous dose of toxin spreading to the cord at a period after the injection when a similar dose injected intravenously would have had little or no effect. It seems likely that antitoxin circulating in the blood, whether introduced by injection or formed by the cells of the animal, has only power to neutralise that part of the tetanus toxin which is not already fixed by the tissues of the nervous system. Additional light has recently been shed on the paths of spread of tetanus toxin and the action of antitoxin by the work of Teale and Embleton. These authors confirmed the earlier work, which showed that tetanus toxin spread directly up the motor nerves to the cells of the ventral cornua. They found, however, that the injection into the nerve of foreign colloids, such as horse serum and egg albumen, have an exactly similar action to tetanus antitoxin in blocking the spread of toxin up the nerve. They concluded, therefore, that the toxin spread chiefly up the neural lymphatics, and that this action of tetanus antitoxin is not wholly specific, but due partly to its colloidal nature. Other methods of temporarily closing the neural lymphatics, such as injecting iodine into the nerve ten days before the injection of toxin, had the same blocking effect. They found that the reason why tetanus toxin does not TETANUS 191 spread to the cord by the dorsal roots and so produce tetanus dolorosus was that the dorsal root ganglion has a definite filtering or blocking action on toxins, as well as on other colloid substances, such as trypan blue. When the latter was injected into a nerve it could be traced as far as the dorsal root ganglion, but not beyond it to the dorsal root, whereas it was seen to spread to the ventral surface of the cord along the ventral root. It was also found that when all the ventral roots of a hind-limb were cut before tetanus toxin was injected into the corresponding sciatic nerve no symptoms resulted from the injection. Meyer and Ransome had previously shown that the injection of tetanus antitoxin into a sciatic nerve protected that limb from involvement in the general spasms produced by intra- venous injection of tetanus toxin. This was thought to be due to spread of the antitoxin to the cells of the ventral horn. Teale and Embleton, however, showed that the same pro- tective action is exerted by egg albumen, and must be due in that case simply to a blocking of the neural lymphatics. This, in their view, is strong evidence that although some of the toxin may spread in the cord, the greater part of that which affects other parts of the cord and brain-stem reaches them along the motor nerves arising there. They also came to the conclusion as a result of their experiments that a foreign protein, such as tetanus antitoxin, could not pass into the tissues of the brain and cord if injected either into the general circulation or into the subarachnoid space, a conclusion which, if confirmed, is of very far-reaching importance. The clinical features of the disease as observed in man bear out these conclusions. In the first place, tetanus fre- quently remains limited to the affected limb when a prophylactic dose of antitoxin has been given, and as a general rule the onset of tetanus is heralded by spasms limited to the wounded limb. In the natural disease the conditions are somewhat different from those produced by the injection of tetanus toxin, as the tetanus bacilli in the wound continue to elaborate toxins, and therefore the intravenous injection of antitoxin will be of some use in all stages of the disease, but it has been shown that it has little effect in modifying the course of the disease once tetanic symptoms have shown themselves. .Intrathecal 192 INFECTIVE DISEASES injection has been thought to be more efficacious; but if the rapidity with which foreign substances disappear from the subarachnoid space, and the hmited relations between this space and the lymphatics of the cord are taken into account, it seems unlikely that the passage of the toxin up the cord would be arrested by such measures. A form of tetanus has been observed resulting from wounds of the face, in which there is more or less complete paralysis of the facial nerve. A similar paralysis may affect the ocular muscles in wounds of the orbit and globe. While the paralysis is usually unilateral, it may be bilateral or affect only the opposite side to the lesion. The pathology of such cases is doubtful, and it is possible that a peripheral neuritis of septic origin plays some part in them, though their absence in ordinary septic wounds of the face renders this very doubtful. It is more likely that they are direct effects of the tetanus toxin which, passing up the motor filaments, comes very rapidly into contact with the cells of the seventh and third cranial nuclei. It is likely that a similar paralytic effect occurs sometimes in the limbs, but as only a small proportion of the motor nerve cells are affected, the resulting paralysis is apt to be obscured by the tetanic spasm of other muscles. Splanchnic tetanus affecting the involuntary muscles has been described as occurring after wounds of the thorax and abdomen, and is supposed to be due to absorption of tetanus toxins along the sympathetic nerves. The histological changes in the nervous system in tetanus have no definite characteristics. Some degree of interstitial neuritis in the nerves connected with the wound is frequently found, but this has more relationship to the presence of pyogenic organisms than to the tetanus bacillus. Various changes have been described in the cells of the nervous system, particularly those of the motor cortex. These consist in perinuclear chromatolysis and pallor of the cell body at the origin of the axon, going on to swelling of the cell body and alteration of its contour. It has been shown, however, that these appearances have no definite relation to the disease, as they occur on both sides of the cord when the symptoms are unilateral, and may be more intense when the severity of the symptoms has diminished. There is usually hyperaemia of various areas in ACUTE POLIOMYELITIS 193 the central nervous system, in both the white and grey matter. The muscles show no characteristic change, but there may be rupture of some of the fibres and small haemorrhages into the tissues due to the excessive strain thrown upon them. 9. Acute Poliomyelitis and Polio-encephalitis. Acute poliomyelitis must now be regarded as one of the acute specific fevers, the specific lesion being an acute inflammation of the central nervous system affecting the spinal cord more frequently and more severely than the brain. Although the disease is sometimes endemic and sometimes epidemic, these two forms show no difference from one another in regard to their pathology. Aetiology. — Within recent years it has been shown that a filterable virus is responsible for the disease, and that it can be obtained from the spinal cord of acute cases. Monkeys inoculated with the virus become paralysed after an interval of a few days and present pathological lesions identical with those found in the human cases. The disease has also been transmitted from one animal to another by the inoculation of spinal cord emulsions. The virus can be obtained from the naso-pharyngeal mucosa and intestinal excreta of infected animals, and there is some reason to believe that these sources are responsible for the undoubted contagiousness of the disease. The virus is destroyed by half an hour's exposure to a tempera- ture of 55° Centigrade, but survives exposure to temperatures below freezing-point. Certain animals can be immunised by inoculation with the virus, and the serum of such animals as well as the serum of human beings who have survived an attack of the disease is capable of modifying or neutralising the active properties of the virus. Under normal conditions the incuba- tion period is not more constant than that of most specific fevers, and probably varies between three and eight days. There is no doubt from clinical evidence that the disease can be conveyed from one person to another, even by persons who are not themselves subjects of an attack. In connection with this it is important to note that while the virus may live in an active state for many weeks and perhaps months in the 13 194 INFECTIVE DISEASES nasal mucosa, its life in the spinal tissues does not appear to be prolonged beyond a few days, and it disappears from the spinal fluid before symptoms of paralysis appear. Morphological and cultural characteristics. — The organism has now been grown on artificial culture media. For this purpose pieces of sterile fresh monkey's brain yield the highest proportion of successes, but cultures have also been obtained from filtered and glycerinated tissues. The method employed is similar to that used by Noguchi for cultivation of the Spirochaeta pallida. Pieces of infected brain are put into tubes of ascitic fluid along with a small piece of fresh rabbit kidney, and the fluid covered with a layer of sterile paraflin oil to exclude air. Growth takes place in from five to seven days, occurring first in the brain tissue, and diffusing later through the fluid, causing a slight turbidity or opalescence. The organism stains with the Giemsa stain, appearing as minute purplish globoid bodies in clusters or short chains, and varying from oa5 jbt to 0*3 ju in size. It behaves variously with Gram's stain. Smears and sections of infected brain show the organism readily, and it has once been found in the heart's blood of an infected monkey. Morbid anatomy. — The morbid changes which are present in the acute stage of this disease differ entirely from those found months or years after the onset of symptoms. The former illustrate the nature of the inflammatory process; the latter represent merely the resulting scars. I. Acute stage. — The nature of the acute changes is the same in all cases, whether they occur in epidemics or sporadically, and whether the patients are children or adults, but their extent and intensity may vary considerably. It is an invariable rule to find that the extent of the disease as shown by a post-mortem examination of the tissues is considerably greater than could be inferred from the clinical symptoms. To the naked eye the changes in the central nervous system are not very striking. The meninges are often hyperaemic, but present no obvious exudation, although, as will be seen later, they are sometimes the seat of a considerable cellular infiltration. The substance of the cord and brain, and particularly the grey matter, appears hyperaemic and oedema- ACUTE POLIOMYELITIS 195 Acute poliomyelitis, a. Section from high thoracic region stained by haema- toxyUn and van Gieson to show the cellular infiltration of the perivascular spaces and of the grey matter, b, A higher power photograph of one side of the same section. 196 INFECTIVE DISEASES tous In some cases haemorrhagic and necrotic areas may be recognised. A section of the spinal cord simply stained with haematoxylin and eosin and examined under a low power presents features which are practically characteristic. The grey matter is darkly stained and stands out unusually well from the white columns, and the blood vessels both in the white and grey matter are conspicuously prominent owing to the fact that their adventitial sheaths are filled with cells containing darkly stained nuclei. A glance at such a section is instructive because '^ *^ t Fig. 55. Acute poliomyelitis. Photograph showing the meningeal and perivascular cellular infiltration in the ventral fissure. it shows at once that the process is really a general one, and that it is only the great vascularity of the grey matter as compared with the white, and especially that of the ventral horns, which suggests at first sight an inflammation limited to a particular region. The microscopic appearances may be described under the following heads : {a) Meninges. — The vessels of the soft meninges are full of blood, and their adventitial sheaths often contain an excess of cells chiefly of the small round type, with a variable proportion ACUTE POLIOMYELITIS 197 of polymorphonuclear leucocytes. Similar cells are seen scattered about in the meshes of the arachnoid, especially in the neighbourhood of the vessels. Sometimes this meningeal infiltration is only found in the lower parts of the cord ; in other cases it has been observed in the higher parts, and even in the basal and vertical meninges of the brain. It shows up more Fig. 56. A cute poliomyelitis. Ventral horn cell preserving a fairly healthy appear- ance and surrounded by intense small-celled infiltration. prominently on the ventral than on the dorsal aspect of the cord. {b) Grey matter. — The vessels of the grey matter, whether they are derived from the anterior or the posterior systems, and whether they are small or large, all present the same cellular infiltration of their adventitial sheaths, which has already been described in reference to the meningeal vessels. The cells lie chiefly in the spaces between the media and the adventitia, and igS INFECTIVE DISEASES also in the meshes of the latter coat. They consist of small cells similar to those seen in the meningeal exudation along with a few plasma cells. This cellular infiltration affects arteries, veins and capillaries alike, and may be traced in many cases to the smallest capillary branches. The vessels themselves are generally engorged with blood which may show signs of thrombosis. In addition to the vessel-changes, the grey matter itself is the seat of great cellular proliferation, which gives it the dark appearance in sections stained with haematoxylin. The Fig. 57. Acute poliomyelitis. Two ventral horn cells undergoing destruction in the midst of serous and cellular exudation. cells taking part in this proliferation are probably various, some being of neuroglial, some of endothelial, and some of blood origin. The grey matter can also be seen to be oedematous, and sometimes to contain capillary haemorrhages. The ganglion cells of the grey matter, which include the large ventral horn cells, the cells of Clarke's column, and those of the dorsal horn, are found to suffer to a greater or less extent. Owing to the richer blood supply of the ventral horn, the cells of that region are generally most affected; but definite changes are ACUTE POLIOMYELITIS 199 frequently observed in the other parts. Speaking generally, those ganglion cells undergo the greatest change which are most closely imbedded in the masses of inflammatory cells, but this rule has its exceptions, and occasionally healthy-looking cells are met with in areas of intense inflammatory reaction. On the other hand, it is very unusual to meet with definite changes in ganglion cells which are far removed from the Fig. 58. Acute poliomyelitis. Changes in the cells of Clarke's column and surrounding cellular infiltration. inflammatory centres. Evidence of neuronophagia is provided by the presence of neuroglial and other cells within the peri- cellular spaces or actually invading the protoplasm of the ganglion cells. All trace of myelinated fibres and cell processes is apt to be lost in the inflamed areas. (c) White matter. — The white columns differ from the grey matter in the fact that they are not nearly so richly supplied with blood, but it is the rule to find that all vessels passing 200 INFECTIVE DISEASES through them from the periphery of the cord towards the central grey matter present the same cellular infiltration of their adventitial sheaths as is found in other parts. On the other hand, it is rare to find cell masses in the white matter itself, although occasionally with suitable staining the neuroglial cells can be shown to have increased in size and to have become rounded in shape, sometimes presenting two or more nuclei. The columns of nerve fibres suffer little, but a certain amount of degeneration is occasionally met with, especially in those parts which border upon the ventral and dorsal horns. Fig, 59. Acute poliomyelitis. Section of spinal cord showing softened area with granular corpuscles, perivascular infiltration and haemorrhage. 2. Chronic stage. — When the acute changes have passed away their results are seen in various forms. In some parts resolu- tion may have taken place with the loss of a few ganglion cells; in other parts necrosis has led to disappearance of the cellular elements ; in other cases the necrosis may have produced actual fluid-containing cavities in the central parts of the cord which are pathologically similar in origin to the porencephalic cavities found in the brain. Sections stained by the Weigert- Pal method may show some diffuse degeneration of the white ACUTE POLIOMYELITIS 201 columns, particularly in the neighbourhood of the grey matter. Examination of the nerves and muscles connected with dis- eased parts of the cord shows at this stage signs of secondary degeneration. The muscular change is that of a simple atrophy. The connective tissue seems to be proportionately increased with the shrinkage of the muscle fibres, and at later periods a certain amount of lipomatosis or fatty infiltration may be observed in some muscles. In the acute stage of the disease the morbid process can often be traced into the medulla oblongata, and sometimes into still higher parts of the central nervous system. The changes are similar to those described in the cord, but, owing to the fact that the white and the grey matter are not so sharply defined above the spinal cord, the distribution of inflammation has the appearance of being less limited and more irregular. This extension of inflammatory changes into the brain is not infrequently found in cases in which the clinical symptoms have not suggested their presence. On the other hand, there are numerous instances, especially in epidemics of the disease, in which the inflammatory process is confined to the brain. The hemispheres, the brain-stem, or even the cerebellum may be the chief site of the morbid changes. It has not yet been decided whether the spinal inflammation is the result of an infection through the blood vessels or through the lymphatics, and opinions are more or less divided on this point. But however the infection reaches the cord, there seems little doubt that the inflammation spreads in it by the lym- phatics, as it always shows a gradual diminution on passing from the level of greatest involvement to levels that are not at all affected. The spread in the cord seems to be mainly upwards and downwards. Thus one ventral horn may be apparently intact, although the other is completely destroyed over several segments. Similarly, it is not uncommon to find very extensive paralysis of one upper or lower limb, with little or no paralysis of its fellow. In such cases, however, the history usually suggests that both ventral horns have been attacked in the first instance, but whereas one has recovered rapidly and completely, the other has undergone much more severe injury. That the cellular infiltration of the meninges which is found in fatal cases is also present in non-fatal cases is proved by the 202 INFECTIVE DISEASES fact that lumbar puncture usually demonstrates a lympho- cytosis of the cerebro-spinal fluid. The relation of the anatomical to the clinical phenomena. — A consideration of the pathological process just described explains the clinical features of this disease. In the first place, it is clear that the virus has no specific action upon the nerve cells, and that the latter suffer irregularly from the inflammatory exudation in which they become submerged. In some regions the ventral cornual cells disappear rapidly, and these regions correspond to the muscular territories in which atrophy is rapid, complete and permanent. In other spinal segments the ventral horn cells undergo a process of partial degeneration with chromatolysis and displacement of their nuclei, and these areas find their clinical counterpart in the muscles which under- go atrophy, but which, after a considerable interval of time, regenerate either completely or in part. Finally, there are the cells which retain a more or less healthy-looking appearance in spite of surrounding inflammation. Such cells may, and doubtless often do, suffer from some temporary functional disturbance, and they are represented clinically by those muscles which are paralysed for a few hours or a few days only, and which rapidly recover their normal activity. Evidence of the occasional involvement of the white matter is afforded clinically by the presence of reflex changes, increased knee jerks, ankle clonus and extensor responses, in a small minority of cases. 10. Lethargic Encephalitis. Definition. — This is a rather rare disease which may occur in epidemic form. It is characterised by a tendency to somno- lence and slight delirium from which the patient is easily roused, and by various nervous manifestations of which the commonest is paralysis of the cranial nerve nuclei, especially the oculo- motor. It may, however, affect chiefly the cerebral hemi- spheres or the basal ganglia. When it occurs in epidemic form the cases are comparatively few in number, and are scattered irregularly through a district. There is seldom any evidence of direct infection, although in a few instances two members of a family have been affected at an interval of a few days to a fortnight. LETHARGIC ENCEPHALITIS 203 Aetiology. — Many observers with long experience in the transmission of poUomyehtis have failed to transmit the virus of lethargic encephalitis, but recently Mcintosh appears to have been successful in infecting monkeys and passing the disease from them to rabbits. He states that it always breeds true, constantly producing cerebral symptoms which may progress through many days, whereas poliomyelitis usually affects the limbs of infected animals. Lethargic encephalitis seems, therefore, to be of different aetiology from poliomyelitis and polio-encephalitis. It differs from this disease also in seasonal incidence, occurring mostly in the winter and spring months, whereas poliomyelitis is more common in the summer and autumn months. Its clinical history is also different, as many cases go on for weeks with periodic exacerbations of nervous symptoms. It affects all ages indifferently, and affects children less commonly than adults. Clinical pathology. — The fluid obtained by lumbar puncture varies in different cases. In about 50 per cent, of cases it shows no abnormality, either as regards cells or albumen. In other cases a diffuse blood admixture, with a corresponding albumen increase, and slight lymphocytosis indicates subarachnoid haemorrhage. In still others a cell increase is present, with or without a noticeable increase in albumen. The cells are entirely of a mononuclear type, many large forms being present in addition to the small lymphocyte. In no cases which we have examined has there been the fibrin coagulum which is so commonly found in poliomyelitis, nor have poly- morphonuclear cells been present. Morbid anatomy. — The macroscopic appearances during the early stages are often striking and almost pathognomonic, the surface of the brain being tinged a deep cherry red, and showing numerous small subarachnoid haemorrhages. On section the grey matter everywhere is of a purplish red colour, and the larger blood vessels stand out prominently. Smaller or larger areas of haemorrhage into the grey or white matter may be seen either in the cerebrum, or more commonly, in the mid-brain and pons. In some cases there is evidence of thrombosis of some of the larger cortical veins or arteries, and, in connection with the latter, areas of infarction of the brain substance may be seen. Where the brain-stem is more affected 204 INFECTIVE* DISEASES Fig. 6o. Lethargic encephalitis . A vessel is seen with a fibrinous clot in its lumen, and surrounded by a thick ring of small round cells. Great glial overgrowth is seen in the surrounding nervous tissue. r^. t -v.: '^^\ 5ri.'3 ^' #--*'*A.,^v-"L.>-»r.-7.::.;. ■■ Fig. 6i. Lethargic encephalitis. Cortex. The typical " cuffing " of the medium and small-sized veins of the cortex is shown, as well as the great congestion of the capillaries. LETHARGIC ENCEPHALITIS 205 than the fore-brain, the only obvious appearance may be a basal haemorrhage in the region of the oculomotor sulcus, spreading thence over the cerebellum and pons. It is usual to find that the grey matter of the cord, at any rate in the cervical region, also shows some hyperaemia. Microscopically, in the early stages the chief changes are the great congestion of all the blood vessels, even of the smallest capillaries, and a diffuse alteration in the nerve cells, most of Fig. 62. Lethargic encephalitis. View of longitudinal section of a small blood vessel in the cortex. The adventitial sheath is seen to contain numerous small round cells. Two petechial haemorrhages are seen close to the vessel. which show a smaller or greater degree of chromatolysis and increase of pigment. While some nerve cells in any section are more affected than others, it is usual at this stage to find some change in all. The more affected cells are often ringed with three or four round cells, which appear to be chiefly of vascular origin, and similar small round cells can be traced emerging from the walls of the capillaries and permeating the brain tissue. It is, however, the exception to find such clumps of round cells as are found in the grey matter of the cord in 206 INFECTIVE DISEASES Fig. 63. Lethargic encephalitis. Cortex. A capillary vessel is seen in the centre of the field, showing short chains of small round cells along its walls. Fig. 64. Lethargic encephalitis. View of cortex and meninges to show the patchy infiltration of the meninges with small cells. LETHARGIC ENCEPHALITIS 207 poliomyelitis, and polymorphonuclear cells are practically never found outside the vessels. Another striking appearance, athough by no means a constant one, is the infiltration of the Virchow-Robin space with small round cells, among which a small proportion of plasma cells may be found. This appears to occur at a slightly later stage of the disease than the changes previously mentioned. The greatest degree of infiltration is seen in the walls of the venules, and in some cases can be traced along them into the subarachnoid space, where it is not uncommon to find collections of round cells close to the medium-sized veins. Small haemor- rhages are often seen distending the so-called perivascular space of His, and tearing away and pressing back the neuroglia from the vessel wall. Smaller or larger haemorrhages into the tissues may also occur. Thrombosis of venules and even of arterioles is not uncommon, and the majority of the vessels in a section may be so affected, but as a rule it is patchy in distribution and only affects one or two venules in each section. At a later stage there may be evidence of softening of small areas in the brain-stem or cerebrum. The perivascular infiltra- tion is usually more pronounced. A large proportion of the nerve cells have returned to their normal appearance, but those more severely damaged by the disease are being attacked by neuronophages, and are in all stages of dissolution. Correlation of anatomical and clinical phenomena. — The general lethargic or somnolent condition of the patient may be partly accounted for by the general intoxication, but its striking remissions suggest that it may be due to some degree of hydrocephalus resulting from intermittent blocking of the aqueduct of Sylvius. The more sudden nervous manifestations, such as ocular palsies, facial palsy or hemiplegia, appear to be due to thrombosis of the vessels supplying the affected area or to haemorrhagic destruction of this part of the brain. Many of the more intractable sequelae are also probably due to vascular changes, but some, such as tremors and athetoid or choreiform movements, may be due to an extensive destruction of the nerve cells of the cortex or the basal ganglia. 208 INFECTIVE DISEASES i \ A.*1 Fig. 65. Lethargic encephalitis. Blood vessel in medulla filled with fibrinous clot, which is being absorbed by phagocytic mononuclear cells. Fig. 66. Lethargic encephalitis. Medium-sized artery on the cortex partially obstructed with haemorrhage in its walls, and leading to small wedge-shaped area of infarction, a corner of which is seen in the right lower corner of the figure. LANDRY'S PARALYSIS 209 II. Landry's Paralysis. The term Landry's paralysis has been used by different observers to describe a number of different pathological conditions, although in the first place it was applied to a rapidly fatal disease of which the pathology and morbid anatomy were quite unknown. Some of the cases which have gone by this name have been instances of multiple neuritis, others of acute poliomyelitis. For the present purpose the term Landry's paralysis will be applied to a morbid condition which corresponds closely in its clinical features with the cases described by Landry, and which displays anatomical changes of so slight a character that they would not have been recognisable by the methods employed fifty years ago. Using the name in this sense we know practically nothing about the aetiology of the disease beyond the fact that it attacks, as a rule, healthy adults, chiefly males between twenty and forty years of age. In some instances the patients have suffered previously from some infective disorder such as gonorrhoea, influenza, or typhoid fever, and in other instances their work has exposed them to extreme variations of temperature, but there is no evidence sufficiently strong to regard these historical facts as really important from an aetiological point of view. The same obscurity may be said to involve the question of bacteriology, and it is not possible to say definitely whether the disease has a specific cause or whether it may result from exposure to various poisons of bacterial or other origin. Many cases have been examined from a bacteriological point of view with negative results. On the other hand, a few have provided discoveries of some interest. In one instance a diplococcus resembling the pneumococcus was cultivated from the patient's blood, and this organism gave rise to symptoms of paraplegia when injected into a rabbit. In another case an organism was isolated from the blood of the patient and also seen under the microscope in the loose vascular tissue forming the external layer of the spinal theca after the patient's death. A subdural injection of this organism into a rabbit produced after some days a rapidly spreading paralysis, and the same coccus was discovered in the theca of the rabbit and isolated from its blood. A somewhat similar microbe has been found in the blood and 14 210 INFECTIVE DISEASES cerebro-spinal fluid of a non-fatal case, but in that instance no results were obtained from experimental inoculation. Pathogenesis. — There are not sufficient data for constructing a definite theory for the origin of this disease, and we must be content for the present to assume that it results from some form of infection, although we are not fully acquainted with the nature of the responsible virus nor of its mode or path of entry. For some reasons it seems likely that the bacteria do not themselves infest the spinal cord, but exert their toxic influence upon that organ from a distance, possibly through the medium of the cerebro-spinal and lymphatic fluids. Morbid anatomy. — To the naked eye the central and peri- pheral nervous systems present nothing remarkable, with the exception of some general hyperaemia of the cord. The latter, however, has a firm consistence before it is cut, provided that post-mortem decomposition has not commenced. After section the vascularity of the grey matter may be noticeable, and may stand out somewhat conspicuously from the white columns. It is sometimes even possible to detect small haemorrhages in the ventral horns. There is no exudation on the surface. The degree of microscopic change varies to some extent with the length of the clinical symptoms, and in some rapidly fatal cases it may be impossible to detect any definite abnormalities. In the majority of cases, however, a careful examination of the spinal cord by means of the Nissl and Marchi methods discovers the following changes. 1. Cells. — A certain number of the ganglion cells, especially those of the ventral horns and of Clarke's column, present either early pericentral chromatolysis or a more complete loss of the chromatin granules, together with some displacement of the nucleus. The more marked changes are usually found in the lumbo-sacral enlargement, and this is not surprising when it is remembered that the lower limbs are usually the first to be paralysed. 2. Myelin. — The medullary sheaths of the spinal cord nerve fibres, and to a less extent those of the peripheral nerve fibres, may show some diffuse fatty change quite unlike that of secondary degenerations. In longitudinal sections small droplets of altered myelin are seen lying singly or two or three together within or between the fibres, but not filling the whole LANDRY'S PARALYSIS 211 area of the medullary sheaths. This appearance may be found in many toxic states unassociated with definite paralysis, and does not therefore necessarily indicate a loss of function on the part of the nerve fibres. Neuroglia and blood vessels. — There is no evidence of neuro- glial proliferation, although some of the cells may appear to be slightly swollen. The blood vessels are somewhat engorged, but otherwise healthy in appearance. Very rarely a Fig. 67. Landry's paralysis. One ventral horn cell shows chromatolysis with displacement of nucleus. slight excess of small round cells may be seen in the immediate neighbourhood of one or two vessels, but the paucity of any such cellular infiltration is in striking contrast to the anatomical changes seen in acute poliomyelitis. The above are the only morbid changes found in the majority of cases at the time of death. Occasionally, when the fatal termination has been postponed for a longer period, the Marchi method will reveal the presence of true Wallerian degeneration which may be 212 INFECTIVE DISEASES secondary to the cell changes. If this exists, some atrophy may also be present in some of the skeletal muscles. No constant changes have been found in other organs, although an enlarged spleen and enlarged mesenteric glands are by no means uncommon. 12. Herpes Zoster. Aetiology. — Herpes zoster is included in this chapter as, for various reasons, there seems little doubt that it is caused by a micro-organism of either bacterial or protozoal nature. The pathology is characterised by acute inflammation of a dorsal root ganglion; and although there is practically no difference between the histological appearances of this disease and those of epidemic poliomyelitis, allowance being made for the part of the nervous system attacked, there is no evidence that the same organism is responsible for both. The observed facts seem, indeed, to point in the opposite direction, as we do not see an increased number of cases of herpes during an epidemic of poliomyelitis, nor do analogous lesions occur in the ex- perimental transmission of the virus of poliomyelitis. On the other hand, the similarity between the two diseases is incon- testable. Both are characterised by an acute inflammation of the nervous system with special incidence on one locality, by the presence of a great degree of small-celled infiltration and vascular congestion with haemorrhages at the site of the lesion, and by lymphocytosis of moderate degree in the cerebro-spinal fluid. The general febrile symptoms are not dissimilar, and both diseases occur either as sporadic cases or in epidemics of less or greater extent, in which there is no direct transmission of infection, and both protect the patients from subsequent attacks. The immunity in herpes is not, however, so absolute as in poliomyelitis, as Head and Campbell in 400 cases found four instances of a second attack. Herpes zoster may occur at any period of life, although it tends to be more severe as age advances. It occurs in the course of degenerative diseases of the central nervous system, such a;s tabes, general paresis, and disseminated sclerosis; and in some of these may be symptomatic. But its course and pathology in such cases, and the rarity of its occurrence, suggest that here, too, we are dealing with true idiopathic herpes HERPES ZOSTER 213 zoster. On the other hand, a rash similar in appearance and distribution may occur in the course of destructive lesions of the cord and its envelopes which affect the dorsal root ganglia. This may be seen in the course of Pott's disease or cancer of the vertebrae, or after severe traumatism affecting the spine. These cases are probably due to a lesion of the ganglion either directly or by interference with its blood supply. Herpes lahialis, or facialis, may occur in the course of acute infective diseases. Examination of the Gasserian ganglion in such cases has shown inflammatory lesions probably due to the attacking organism. There is thus some pathological relationship between the two varieties of herpes. Macroscopic appearances. — In the early stages the affected ganglion is swollen and hyperaemic. On section it may have a uniform pinkish tint, or there may be haemorrhages visible to the naked eye. The nerve in relation to it may also be swollen. Microscopically the chief changes seen in the ganglion are congestion of all the vessels, numerous small haemorrhages around them, and infiltration of the tissues with small round cells. These seem to be more numerous in relation to some arterioles than others, especially towards the dorsal side of the ganglion. Some of the nerve cells are completely destroyed, others show various degrees of degeneration, being swollen and staining diffusely, but chromolytic changes are not seen. Many of them are surrounded by round cells which have invaded their capsule, and seem to be acting as phagocytes by a process of neuronophagy. Others escape completely and show no altera- tion in their Nissl granules. The sheath of the ganglion is infiltrated with round cells and its vessels engorged, and there may be some oedema between its layers. Similar changes are seen in the nerve close to its emergence from the ganglion. After eleven days degeneration is found by Marchi's method both in the dorsal nerve roots and in the peripheral nerves. It has been traced into the cord and up the dorsal columns for a varying distance, and also down the peripheral nerves as far as the terminal twigs to the skin. It is similar in character to Wallerian degeneration. At a later stage the changes seen in the ganglia and peripheral nerves are of a cicatricial nature, and may not be obvious unless the original 214 INFECTIVE DISEASES inflammation has been severe and fairly extensive, but some degree of sclerosis can usually be found both in the ganglion and its capsule and in the peripheral nerves. Head and Campbell in 392 cases found that certain ganglia were more often affected than others ; thus, those on the dorsal roots from the fifth cervical to the second thoracic and from the third lumbar downwards were particularly exempt — that is, those in the areas corresponding to the nerve supply of the limbs. The Gasserian ganglion is frequently affected, and the changes in it are similar to those in the spinal root ganglia. 13. Chorea. Aetiology. — This disease is common in childhood and adolescence, and rare after twenty years of age. It has no direct hereditary basis, but more than one member of a family of the same generation is often affected. Nervous, excitable children, very often those who are intellectually well equipped, appear to be specially disposed to the disease, and girls are more frequently affected than boys. Climatic influences are not marked, but there are certainly seasonal variations in the prevalence of the disease. The most important aetiological factor is the relationship of chorea to acute rheumatism, and many authorities regard chorea, rheumatic arthritis, rheumatic endocarditis, and pericarditis as manifestations of the same virus affecting different tissues. On this theory there is no difficulty in understanding why arthritis precedes chorea in the majority of instances, and why less commonly chorea may appear before attention has been drawn to the rheumatic poison by the occurrence of joint trouble. The frequent association of chorea with endo- carditis and myocarditis is further evidence of the affinity between these various manifestations of some specific virus. In recent years Poynton and Payne have succeeded in cultivat- ing from the cerebro-spinal fluid of choreic patients a diplo- coccus identical with that which they have described as the Diphcoccus rheiimaticiis, and they have also produced move- ments in rabbits which bear a resemblance to the movements in human chorea. The diplococcus has also been found in the pia mater and cerebral tissues, and has been isolated from the blood of patients suffering from chorea. CHOREA 215 Pregnancy appears to have a special influence in evoking an attack of chorea, most commonly in the case of women who have had previous attacks or who give a history of the rheumatic diathesis. The relationship of emotion, and particularly of fright, with the onset of symptoms has often been remarked, but the observation of parents with regard to such coincidences is not very often entirely trustworthy. It would not, however, be very surprising if fright were responsible for exag- gerating slight symptoms of chorea which may have been overlooked, or even for evoking them in a predisposed subject. Pathogenesis. — The accumulation of evidence supports the view that chorea is an infective disease of microbic origin, although there may be room for doubt with regard to the specificity of the virus. The old belief that numerous emboli of the cerebral vessels were responsible for the clinical symptoms of the disease has not been supported by the pathological examination of recent cases, and the absence in many cases of any source of the emboli in the shape of valvular vegetations further discredits this theory. No agreement has been arrived at with regard to the actual site of the morbid changes which are responsible for the clinical manifestations of the disease. The dendrites, the cortical cells, the cells of the basal ganglia and of the brain-stem have all been indicted by different authorities, and blame has been attached to those parts of the nervous system which bring the influence of the cerebellum to bear upon the cortico-spinal and rubro- spinal systems. Morbid anatomy. — Macroscopical changes in the central nervous system in cases of chorea have usually been conspicuous by their absence, but attention has been called to the occasional presence of hyperaemia and of small thrombotic lesions which, however, are by no means constant. The same remark applies to the occurrence of small perivascular haemorrhages and slight neuroglial proliferations. According to some authors, evidences of slight inflammation and small foci of necrosis with exudation are more common in the basal ganglia than elsewhere. Under the microscope similar changes have often been described, as well as chromatolysis of ganglion cells in the cortex and 2i6 TRYPANOSOMIASIS basal ganglia. Micro-organisms of various kinds have been detected in addition to the Diplococcus rheumaticus found by Poynton and Payne and Gordon Holmes. At the present time it is hardly justifiable to do more than mention the various microscopical findings which have been referred to, and it is too early to express a decided opinion as to the significance of each or to characterise any of them as essential factors in the morbid anatomy of chorea. Relationship of anatomical to clinical phenomena. — We are not in a position to correlate the movements so typical of chorea with any particular lesion in any particular part of the brain, but there is a general tendency to regard choreiform movements as a manifestation of disturbance in the functions of the basal gangliaralher than in those of the motor cortex. The asthenia, which may sometimes be very profound in choreic patients, may be looked upon as evidence of the influence exerted by an infective virus upon the nerve cells of the brain, and the oc- casional presence of reflex changes, such as an extensor plantar response, points to the conclusion that the pyramidal tracts may at times suffer severely from toxic effects. Hyperpyrexia and mania are occasionally serious complications in the course of chorea, and are generally attributed to the effects of poison and exhaustion on particular regions of the central nervous system. 14. Trypanosomiasis. This is a chronic infective disease affecting the nervous system in the later stages, when it produces the clinical condition called " negro lethargy " or " sleeping sickness." It occurs in endemic and epidemic form in most parts of Central Africa, and has been found as far south as Rhodesia. The infective agent is one of two species of trypanosome, T. gambiense and T. rhodesiense, which are transmitted to man by the bites of the tsetse flies, Glossina palpalis and G. morsita' s respectively. When transferred to the human subject, the trypanosomes enter the blood stream and multiply. They never, however, become very numerous in human blood, and are not easily found in direct blood smears. It is better to search for them in the centrifugalised deposit of citrated blood, They also RABIES 217 enter the lymphatics, and are readily found at any stage of the disease in the fluid from enlarged glands. They are rarely present in the cerebro-spinal fluid in the earlier stages of the disease, but may be found there in almost every case after the characteristic symptoms of " sleeping sickness " have made their appearance. They may also be found there during attacks of pyrexia. The chief change in the gross appearance of the brain and spinal cord is a diffuse opacity of the pia-arachnoid, which may be abnormally adherent to the cortex of the brain. There may also be a moderate degree of hydrocephalus. The cerebro-spinal fluid is sometimes slightly yellow in colour. Microscopically the characteristic feature of the disease is a crowding of the perivascular lymphatics and of the meshes of the pia-arachnoid by small round cells. This affects different parts of the brain and cord in varying degree, but is present to some extent all over. A similar condition is present throughout the lymphatic channels of the body, and is specially noticeable in the intestinal lymphatics. The other changes in the central nervous system appear either to be secondary to this (Mott) or due to toxins derived from the parasite. In the cortex of the brain the pyramidal cells may show loss of their Nissl granules, displacement of their nuclei or diffuse staining of the'r protoplasm. There is also a diminution of the fibres of the supraradial and tangential layers. In the cord there is a diffuse sclerosis. In contrast to the appearances observed in general paralysis of the insane, there are no changes in the vessels. 15. Hydrophobia or Rabies. Definition. — Rabies is an acute infective disease of the nervous system communicated from animal to animal, or from animal to man through the saliva, either by biting the healthy skin or by licking an abraded surface. The incubation period varies with the size of the infected animal and with the distance of the site of inoculation from the central nervous system and especially from the brain. Thus, in the human subject, infection through the skin of the face or neck brings on the disease after a much shorter incubation period than when the hands or legs 2i8 INFECTIVE DISEASES are bitten. Usually, in the human subject, it varies from twenty to fifty days, but cases have been reported in which it has extended to two or even five years. When the disease is established it is uniformly fatal. Causative organism. — In 1903 Negri described certain minute bodies in the nerve cells of the central nervous system in dogs affected with rabies. When appropriately stained these appear as small bodies of from 0*5 to 18 //- in diameter, composed of an oxyphil cytoplasm, with a basophil " central body " and basophil granules, and surrounded by a definite limiting membrane. In shape these bodies are rounded, oval, triangular or spindle-shaped. They are found in the substance or the pro- cesses of the ganglion cells of the brain, spinal cord and spinal ganglia, and in the cells of Purkinje in the cerebellum. Some- times they lie immediately around the cell. They are most constantly found in the large cells of the cornu ammonis in dogs. In these they are found in almost 100 per cent, of cases of rabies. They have also been found in the human subject in a very large percentage of cases, and have not been identified in any other condition. But although their presence was considered pathognomonic of the disease, their aetiological relationship to it was not established until in 1912 Noguchi, using similar cultural methods to those found successful for the spirochaetes of syphilis and relapsing fever, cultivated from the brains and spinal cords of infected animals " very minute granular and coarser pleomorphic chromatoid bodies transferable through many generations." On four occasions in these cultures nucleated round or oval bodies surrounded with membranes, similar in appearance to the Negri bodies of the brain, appeared suddenly in the cultures and lasted for from four to five days. He also produced rabies experimentally by inoculation of either form of culture into dogs, rabbits and guinea-pigs. It was known previously that the virus could pass through coarse Berkefeld filters, and this militated against the view that the Negri bodies were the cause of the disease. Recent workers consider, however, that these are either a phase in the life cycle of a filter-passing organism, or are, in part, constituted by the reaction of the tissues to the presence of the organism. RABIES 219 Noguchi's results suggest that the former view is the more correct. Pathology. — The virus of the disease seems to reach the nervous system along the nerve trunks after being absorbed by the fine terminal filaments of the nerves, and for this reason the incubation period varies with the distance of the point of inoculation from the nervous centres. Thus, inoculations into the anterior chamber of the eye or under the dura mater bring on a very acute attack of the disease after a minimal incubation period. Infection may also take place from a healthy mucous membrane, such as the conjunctival or nasal. As regards the path whereby the poison reaches the higher nervous centres, experiments seem to prove that it travels up the cord in the same way as tetanus. It seems also to extend into the nerves of the opposite limb, but this may be part of the general diffusion of the virus throughout the body. The virus is excreted by most of the glands of the body, perhaps chiefly by the parotid, and it has been found in the secretions from the lachrymal and mammary glands and the pancreas. It is found in all parts of the nervous system, in- cluding the cerebro-spinal fluid, the sympathetic nervous system and the suprarenal glands. A great deal of experimental work has been done on the variations in the toxicity of the virus. Thus, the passage of cerebro-spinal fluid through a series of monkeys attenuates the virus, whereas passage through rabbits increases the toxicity. With dogs it remains fairly constant. Exposure to light and air diminishes its strength, and it is destroyed by heating to 50° C. for an hour, or to 60° C. for half an hour, and by most antiseptics in weak solutions. Pasteur passed the virus through rabbits until the incubation period was reduced to six or seven days. Beyond this he could not raise the toxicity, and he called this the " virus fixe." Starting then with the cords of rabbits of this maximum toxicity, he found that storage in sterile flasks over some hygroscopic material gradually reduced the toxicity until after fourteen days of storage inoculations no longer produced the disease. From this basis he started preventive inoculations, giving injections of rabbit's cord, at first inactive, then gradually more and more potent, until cords containing the " virus fixe " were given. He found that if this 220 RABIES treatment were commenced within five days of the bite nearly 100 per cent, of cases failed to develop the disease, and even later a large proportion could be saved. Sometimes transient paraplegia supervenes after the eighth day of treatment, but this usually passes off in four to five days. The cause of this is unknown, but it may be that during the process of destruction of the virus by the cells of the body some toxin is set free which has a selective action on the cells of the central nervous system. Morbid anatomy. — The naked-eye appearances of the brain and cord show nothing characteristic of the disease. General congestion of the grey matter of the cord, and medulla, is present, and frequently small haemorrhages are to be seen, most commonly in the floor of the fourth ventricle. The incidence of the disease seems to fall chiefly on the medulla, then on the cord, and least on the cortex and central ganglia. Microscopically, the most obvious change in the infected region is infiltration of the perivascular adventitial spaces with small round cells. There is also considerable glial pro- liferation, and in the most affected areas the ganglion nerve cells are surrounded with a ring of parasitic glia cells. The nerve cells themselves undergo considerable degenerative changes, being swollen and containing hyaline bodies near the nucleus. Negri's bodies may be found in them with appropriate staining methods. Golgi has described alterations in the dendrites, which are diffusely swollen or show small swellings along their stems. This change is diffusely spread throughout the cortex. In the nerves along wjiich the infection has run degenerative changes are often found, such as breaking up of the myelin, swelling of the axons, and small-celled infiltration. The salivary glands show accumulations of leucocytes and vascular congestion. REFERENCES MacCallum, W. G. : Textbook of Pathology, 1919. MuiR, R., AND Ritchie, J.: Manual of Bacteriology, 1919. Nageotte, J., AND RiCHE, A.! Coriiil and Ranvier, Man. d'histol. Path. Paris, 1907, vol. iii. Landry's" Paralysis. Buzzard, E. F.: Allbutt and Rolleston, System of Medicine, vol. vii., 1910, p. 670, INFFXTIVE DISEASES 221 Acute Poliomyelitis. Batten, F. E.: Lumleian Lectures. Brain, 1916, vol. xxxix. Herpes Zoster. Head, Henry: Allbutt and Rolleston, System of Medicine, vol. vii., 1910, p. 470. Trypanosomiasis. MoTT, F. W.: Archives of Neurol. , 1907, vol. iii., p. 581. Muir, R., and Ritchie, J.: Manual of Bacteriology, 1919. Tetanus. CouRTOis-SuFFiT AND GiROUS I Lcs formcs anormaUs du Tetanos. Paris, 1916. Muir, R., and Ritchie, J.: Manual of Bacteriology, 1919- Teale, F. H., and Embleton, D.: Journ of Path, and Bad., 1919, vol. xxiii., P- 50. Chorea. PoYNTON and Paine: Rheumatism. London, 1914. Rabies. Muir, R., and Ritchie, J.: Manual of Bacteriology, 1919- NoGUCHi, H.: Journ. of Exper. Med., 1913, xviii., p. 314. Lethargic Encephalitis. Buzzard, E. F., and Greenfield, J. G. : Brain, iqiq, vol. xlii.. p. 305. McIntosh, J., AND TuRNBULL, H. M. : British Journ. of Exper. Pathology. London, 1920, vol. i., p. 89. Marinesco AND McIntosh, J. : Report to the Local Government Board. No. 121, 1918. Netter, A.: Bull, de I' Acad, de Med., 1918. T. Ixxix., p. 337. CHAPTER VII EFFECTS OF POISONS I. Neuritis. The term neuritis is applied both to inflammations of the nerves and to certain degenerative processes which affect the nerve fibres, without producing any inflammatory reaction in the tissues surrounding them. Any of the constituent parts of a nerve fibre may be primarily and chiefly affected. Thus, the connective tissue of the peri- or endo-neurium may undergo the greatest change, the vessels may be little or much congested, or the nerve fibres themselves may degenerate from a toxic cause with little or no reaction in the supporting structures. For the most part, however, all these structures are affected at the same time to a greater or less degree, but as certain elements tend to be specially damaged by certain causes, two main groups of neuritis are distinguished, the interstitial and the parenchymatous. Interstitial neuritis is always asym- metrical. It is usually localised, but may spread along a nerve. Parenchymatous neuritis, on the other hand, is usually symmetrical. It tends to affect chiefly the peripheral portions of the nerves, and may affect many nerves throughout the body, both cranial and spinal, when it is termed " multiple peripheral neuritis." Aetiology. — In general it may be said that interstitial neuritis is due to lymph-borne or local causes, parenchymatous neuritis to blood-borne or general causes. In both cases the general resistance of the patient plays a great part. A rheumatic or gouty diathesis predisposes to local nerve inflammations, and tuberculosis, diabetes, and blood diseases lower the resistance to general toxins and lead to multiple neuritis. At the same time, the attack tends to be more severe and the recovery more prolonged in debilitated than in strong, healthy subjects. 222 NEURITIS 223 Interstitial neuritis may be more directly caused by exposure to cold, by prolonged or intermittent pressure on a nerve, or by the proximity of some infective focus, such as an abscess, a superficial sore, a diseased joint or an unhealthy mucous membrane. In the first two cases the process is an aseptic one and is usually localised and quickly recovered from. In infective cases, on the other hand, the inflammation may spread up the nerve, affecting other nerves of the plexus, and may reach the spinal cord, causing myelitis {ex neuritide) . The recovery depends on the immunity of the patient and the removal of the cause. It may be extremely prolonged, so that in some cases surgical intervention has been needed to end what appeared to be interminable suffering. Parenchymatous neuritis is due to chemical or microbic poisons, of which the commonest are lead, alcohol and the diphtheria toxin. Of the others we must note : (i) Inorganic : arsenic, mercury, phosphorus and silver. (2) Organic : ether, bisulphide of carbon, dinitro-benzol, aniline, carbon monoxide. (3) Toxins of various bacteria, especially those of influenza, enteric, pneumonia, erysipelas, gonorrhoea, septicaemia of all kinds and malaria. Syphilis also is said to give rise to peripheral parenchymatous neuritis, or at least to aid in its causation. In addition, we must confess that many cases of peripheral neuritis occur which, in the absence of any known cause, we are forced to attribute to some unknown toxin. Recent work has thrown doubt on the role of alcohol as a direct cause of parenchymatous neuritis, and certain observers prefer to consider that it plays a similar part to tuberculosis and diabetes in being merely a predisposing cause or in so altering tissue metabolism that the nerves become an easy prey to other toxins. Morbid anatomy — (i) Interstitial form. — On removal from the body the affected nerve is swollen, often irregularly, along a portion of its course. In the early stages of the disease it is soft, hyperaemic, and may contain lymph between its bundles. In the later stages it becomes firmer from overgrowth of fibrous tissue. On microscopic examination the peri- and endo-neural 224 EFFECTS OF POISONS sheaths show a greater or less degree of infiltration with round cells, and proliferation of the connective tissue and endothelial cells with the production of phagocytes. The vessels are con- gested and their walls filled with cells of various kinds. There may be some exudation of lymph which is in process of organisation. As we pass toward a later stage connective-tissue overgrowth assumes the chief role, and the cellular infiltration becomes less marked. The nerve fibres themselves may be pressed on by lymph or overgrown fibrous tissue and undergo changes similar to those in parenchymatous neuritis. But although there is degeneration of the medullary sheaths of the nerves, the axons tend to persist, and thus function is not greatly affected and is quickly restored. Complete recovery in interstitial neuritis may be rapid when the inflammatory products are completely removed, or it may be more prolonged when organisation of the exudate and connective- tissue hyperplasia has occurred. In such cases the thickening of the nerve may be permanent, and fibrous tissue overgrowth may affect its whole course and even its ganglion. Pathologically it is difficult if not impossible to distinguish between the slighter affections of nerves which are grouped under the term " neuralgia " and severer cases where the nerves are more markedly affected. It has been shown in cases of death from cancer of the tongue, bed-sores, abscesses, etc., that the nerves in relation to these foci are infiltrated with in- flammatory products among which organisms have been found. It therefore seems likely that certain cases of neuralgia and referred pain in connection with some such inflammatory focus are due to true inflammation of the nerve sheaths. Thus, in trigeminal neuralgia a fibrous overgrowth in and around the Gasserian ganglion is the usual pathological picture. (2) Parenchymatous neuritis. — While the various poisons which produce this condition vary to some extent in their mode of attack on the nerves, certain pathological results are common to all. I. The nerves tend to be more affected at their periphery than at their emergence from the spinal canal. In fact, nerves which are severely affected may show little change at the level of the plexus from which they spring. The degree to which this is true varies with the causal toxin. Alcohol affects the NEURITIS 225 terminal fibres of the nerves, especially their fine intermuscular branches. Diphtheria and lead affect the nerve fibres more evenly. But all forms of parenchymatous neuritis assume this peripheral distribution, and are therefore often termed " multiple peripheral neuritis." 2. The most obvious changes are found in the medullary sheath. The axis cylinders also suffer, sometimes very severely, and the cells of the sheath of Schwann undergo great change. But while preparations stained for the medullary sheath may show a great degree of degeneration, investigation by Biel- schowsky's stain for axis cylinders may show comparatively little destruction of these elements. On examination of an affected nerve in Marchi prepara- tions the degenerated fibres appear as chains of black dots, representing globules of fat formed by the breaking down of the myelin. At first these are of small size and are massed irregularly within the neurolemmal sheath. Darkly staining globules are also seen inside the cells of the sheath of Schwann, in cells lying within this sheath which are probably derived from it, and in phagocytic mesodermal cells which lie among the nerve fibres. The affected nerve fibres are irregular in outline, and in later stages many are shrunken. Preparations stained for cellular reaction, as by haematoxylin, polychrome methylene blue or thionin blue, show prolifera- tion of the cells of the sheath of Schwann and, to a greater or less extent, of the connective tissue and other meso- dermal elements surrounding the nerve fibres. Many of the cells of the neurolemma assume phagocytic and, it may be, migratory powers, and are found filled with degeneration products lying free within the sheath. Scharlach R. and other fat stains show that these products are largely composed of fully formed fat, while others are fatty acids, and in the later stages cholesterin-containing substances also appear, chiefly in the mesodermal phagocytes. A comparison of these re- actions with those which take place in degenerations of the brain and spinal cord shows that the cells of the sheath of Schwann take a similar part in fat synthesis from degenerated myelin to that taken by their analogues, the neuroglia cells of the central nervous system (p. 22). Bielschowsky's silver impregnation method shows that the 15 226 TOXIC MYELITIS axis cylinders do not undergo any great destruction. In the less affected nerve fibres the axis cylinders do not differ to any great extent from the normal. In more affected fibres they are in a state of granular degeneration. Other nerve fibres contain one or more fine fibrils studded with spindle-shaped or globular varicosities, and either running a straight or wavy course or twisting round one another in spiral form. These probably represent regenerated fibres; and it appears that regeneration of axons often takes place rapidly at an early stage in peripheral neuritis. The investment with medullary sheath which is necessary for restoration of function may, however, be delayed until the poison causing the disease has ceased to act. 2. Toxic Myelitis. This rare form of myelitis differs from the infective type in its less acute onset and course, and still more perhaps in the fact that recovery is more frequently complete. Little is known of its aetiology, except in those cases which have been described as associated with pregnancy and often accompanied by serious disturbances in the functions of the heart and kidneys. The close relationship between child-bearing and this form of myelitis is shown by the improvement in the latter after con- finement, and by its recurrence in subsequent pregnancies. Morbid anatomy. — Gross changes are not very obvious, but some oedema and slight softening of the spinal tissues may be observed. Microscopically, the chief changes consist in swelling of the myelin sheaths and of the axis cylinder processes, especially in the white matter. The change is generally patchy in distribution, and does not affect systems as a whole. In addition, there are toxic changes in the nerve cells and in the neuroglia, although the latter do not show the proliferation which is characteristic of the infective form. The Marchi stain demonstrates fatty changes in the myelin sheaths, and the Nissl stain illustrates the changes in the ganglion cells. Owing to the fact that recovery is the rule rather than the exception, the opportunities for studying this form of myelitis have not been frequent. EFFECTS OF POISONS 227 3. Encephalopathy (Effects of Poisons on the Brain). Various organic and inorganic chemical substances have long been known to have definite actions on the brain, some affecting primarily the higher centres, others the vital mechanism of the medulla oblongata. For this knowledge we are chiefly indebted to the experimental work of pharmacologists, and as a result our knowledge of the direct action of the poisons rests on a firmer basis than that of the remoter effects produced when the poison is absorbed in smaller doses over a longer period. These changes result from faulty nutrition of the nerve cells and their processes and may be directly due to the prolonged action of the poison, or may be caused indirectly by changes in the walls of the blood vessels. The commonest poison to exert a noxious effect on the brain is ethyl alcohol. Some of the well-known results of excessive indulgence in this drug are due to its direct action on the nerve cells and fibres when taken in an overdose. Others, not so well understood, seem to be due to prolonged use of alcohol in smaller quantities. And here we must also reckon with the effects of changes in metabolism induced by the drug, which seem to have a considerable bearing on the aetiology of delirium tremens. In the polyneuritic psychosis we seem to be dealing with the chronic direct poisoning of nerve cells and their processes throughout the nervous system. In this condition Ballet has described changes in the pyramidal cells of the cortex consisting of swelling and vacuolation of the cytoplasm along with perinuclear chromatolysis and displacement of the nucleus. Degeneration of the tangential and supraradial systems of fibres in the cortex has also been found. These lesions are not constant, and it is noticeable that the symptoms are much more pronounced than would be suggested by the pathological changes which have been described. On the other hand, in the accompanying peripheral neuritis extensive pathological changes are found even when the symptoms are minimal. It is probable that in both cases the underlying pathological basis is the same, viz. a chronic intoxication resulting in failure of nutrition of nervous elements. In chronic alcoholic insanity the pathological changes are very various, and many of them appear to be due to degenera- 228 LEAD ENCEPHALOPATHY tion of the blood vessels of the brain, which is the only constant feature. They consist of diffuse sclerosis of the cerebral cortex with widening of the sulci and shrinkage of the brain substance. " Pachymeningitis interna haemorrhagica " is often associated with this disease. In the more acute cases of lead poisoning convulsions fre- quently occur, and they may be associated with optic neuritis. In fatal cases of this kind the brain is found to be pale and oede- matous, with an excess of fluid in the meshes of the arachnoid. These changes appear to be due to spasm of the cerebral arteries and diminution of the blood supply to the brain. In a case of chronic lead poisoning Mott found increase of connective tissue in the pia-arachnoid and the walls of the small vessels, but no evidence of small cell infiltration. Occasional haemorrhages into the perivascular sheath were seen, and adherent to the outermost layer of the vessels were large fibre-forming glial cells. The glia was everywhere undergoing proliferation with formation of new fibres, especially in the superficial and deep layers of the cortex, where the glial increase was out of pro- portion to the wasting of the neural elements. Under the pia also the glia was very dense. The Betz cells of the motor cortex were degenerated, showing perinuclear chromatolysis and some neuronophagy, as well as pigmentary changes, but there was no gross atrophy or degeneration of the fibres of the cortex or of the pyramidal tract, except a slight diffuse sclerosis of the latter in the lumbar region. Some cases of arsenical neuritis develop a similar condition to the polyneuritic psychosis of chronic alcoholism. Optic atrophy has been observed to follow the administration of large doses of arsenical compounds, such as atoxyl. It may also result from poisoning by methyl-alcohol and quinine. Carbon monoxide has both direct and indirect actions on the brain. Its direct action on the nerve cells may be explained by its power of combining with the haemoglobin of the blood and thus depriving the brain of its supply of oxygen. In addition to this action on the nerve cells, small perivascular haemorrhages have been found in the white matter of the brain, and these may account for certain nervous symptoms, such as the partial paralysis remaining after the immediate effects of the poison have passed off. EFFECTS OF POISONS 229 4. Ergotism. This disease, though rare in England, has occurred in the form of severe epidemics on the Continent, especially in France. But since, in the seventeenth century, it was proved to be due to the common fungus of rye, " ergot of rye," the disease has seldom appeared in epidemic form in civilised communities, although it is endemic in parts of Russia. Its two forms, the convulsive and the gangrenous, seem usually to occur in separate epidemics. Localities where one is prevalent seem to be immune from the other. Aetiology. — Although the cause of the disease has been known for more than two centuries, certain points in connection with the aetiology have not been entirely cleared up. Thus, it is known that ergot grown in certain localities and under certain conditions is more toxic than that grown elsewhere, and that ergot grown medicinally in this country is compara- tively inert, but the exact conditions of growth favourable to the production of poisonous active principles are not known. These may lie in the nature of the soil or in conditions of moisture, heat, etc., in the climate. It is known that foggy fen country is most liable to epidemics of ergotism, especially when there has been a hot summer. Conditions of poverty and destitution in the community and individual susceptibility play a great part in the incidence of the disease. Pharmacological research has shown that ergot contains three active principles which have not been isolated in crystallisable form, and which may be combinations of one or more active principles with more inert substances. These active principles have been called (i) ergotinic acid, (2) cornutine, and (3) spha- celinic acid. The first does not appear to have any direct relationship with the disease. Cornutine has an action on the medullary centres and may produce clonic convulsions. To its action the convulsive type of ergotism is probably due. Sphacelinic acid has a constrictor action on plain muscle fibre, especially on that of the vessel walls. This takes effect through the sympathetic nerves, as perfusion of isolated vessels with extracts of ergot has no effect in diminishing the calibre of the vessels. Ergotism is thus due to the direct action on the nervous 230 EFFECTS OF POISONS system of one or other of these two poisons, cornutine and sphaceUnic acid. Whether the symptoms are those of the convulsive type or the gangrenous type is probably due to the relative amounts of the two poisons present in the grain causing the epidemic. Individual susceptibility may also play a part, but that it is a small one is indicated by the way in which epidemics are limited to one form. Pathology. — A very constant feature of the disease is degenera- tive change in the cord. This is similar to that seen in pellagra, except that, being less chronic, there is less sclerosis, and more evidence of recent destruction of nerve fibres. The long columns of the cord suffer most, especially the dorsal columns. The resemblance to the degeneration of tabes may^ be very close in sections stained for myelin, but a more minute examination with other tissue stains shows that this resemblance is only superficial, and that the lesion is not really systematised. Some degeneration of the sensory nerves has also been described. The arteries show thickening of their middle coat and hyaline changes of the intima. The smaller arterioles are thrombosed. Congestive changes in the lungs are very common. 5. Pellagra. Definition. — Pellagra is a disease which occurs sporadic- ally or in endemic form. It has long existed in the north of Italy, and recently has spread in the Southern States of North America. It is characterised by cutaneous eruptions, gastro- intestinal disturbances, and a tendency to degeneration of the nervous system. Aetiology. — The causation of this disease is still obscure. For long it has been known to attack communities where maize is the staple article of diet, and where general poverty exists. The common occurrence of the disease in spring, and its similarity to ergotism as regards pathological changes, led to the view that it was due to the growth of some mould or fungus on maize. Many cases have, however, occurred in which maize could with certainty be ruled out of court as a causative factor. More recent work has indicated that pellagra is a deficiency disease due to a diet in which the proteids are small in quantity and of poor quality. It has been shown that proteids vary PELLAGRA 231 greatly in their value as foodstuffs, and that those derived from the seeds of plants are far inferior to animal proteids in their power of maintaining nitrogenous equilibrium. This is attri- buted, among other things, to their poverty in tryptophan and lysine. Goldberger, in 1914, was able to reproduce pellagra in prisoners by feeding them on diets similar to those consumed in pellagrous districts. These diets have been shown to be composed largely of the seeds of plants, such as maize, beans, and rice, the proteids of which have a low biological value. Morbid anatomy. — Whatever may be the aetiological agent, there is no doubt that the lesions found in the nervous system are of a toxic character. They have been chiefly studied in the spinal cord, but the nerves and the brain are also affected. Macroscopically no marked abnormality is seen beyond some slight changes in the dorsal columns. Histologically the cord shows a pseudo-systematised lesion in the dorsal columns. In the cervical region the column of GoU is always affected, but the column of Burdach may be intact or show irregular bands of degeneration. In the thoracic and lumbar regions the appearance of systematisation is not so marked. The dorsal root zone may or may not be involved, and the cornu-commissural zone may be spared. In fact the process is essentially a diffuse one, affecting not only the dorsal, but, to a less extent, the lateral columns. In the latter it is not limited to the pyramidal tracts, but is spread diffusely, often implicating the spino-cerebellar tracts. Under a higher power of the microscope the picture is extremely like that seen in ^' subacute combined degeneration," although the changes are more confined to the dorsal columns and are of a more chronic type. Degeneration of myelin is evidenced by the numerous granular corpuscles lying in the degenerated areas and filling the adventitial lymphatics of the veins. Marchi's stain shows that these are filled with fat. Glial proliferation is well established, both amoeboid glia cells and spider cells being seen, and there is a greater formation of new glial fibres than in subacute combined degeneration. There is a complete absence of any lymphocyte or plasma cell infiltration. The changes in the nerve cells are those common to all 232 EFFECTS OF POISONS subacute toxic diseases: perinuclear chromatolysis, with eccentricity of the nucleus and a more globular outline of the cell body. The cells of Clarke's column are most affected, but the ventral horn cells, the intermedio-lateral tracts and the spinal ganglion cells may also suffer. Similar changes are found in the Purkinje cells of the cerebellum, in the cranial nerve nuclei and in the pyramidal cells of the cortex. The peripheral nerves have been studied particularly by Mario Zalla and Kinnier Wilson. Changes in them, although usually slight in degree, are practically constant. Slight fragmentation of the myelin and the appearance of scattered droplets staining by Marchi's method are common. The axis cylinders may show no change, or only slight irregularities of contour; occasionally the axis cylinders may split up into fibrils. There is often some oedema of the nerve bundles and thickening of the interstitial tissue. Mario Zalla and Wilson have called attention to the great increase of the ** tt granules of Reich " in the nerves of pellagrins. These are small, irregularly-shaped granules, staining metachromatically and appearing pinkish when coloured with thionin blue. They are chiefly seen in the cells of the sheath of Schwann, but may be found in the adventitia of blood vessels. Their exact nature is not yet determined, but they appear to be a form of degenera- tive product of nervous matter. 6. Lathyrism. Aetiology. — This is a paraplegia of fairly rapid onset and tending towards cure, which is, however, never complete. It was mentioned by Hippocrates, and in more recent times has occurred in France, Italy and Algiers, but during the present century it has been confined to certain parts of India. It occurs in those who subsist almost entirely on a diet of pulse [Lathyrus sativus or L. cicera). It is probable that the grain is not itself toxic, but either from the presence of some parasitic fungus (cf . ergotism) , or from lack of some constituent necessary to life (cf. beri-beri), its too exclusive use leads to the disease. In India it is to be noted that lathyrism only occurs during famine years, when, in addition to the shortage of other foods, there is general poverty, and that its incidence is confined to the rainy season; also that it attacks men much more LATHYRISM 233 Fig. 68. The cord in a case of lathyrism stained by the Weigert-Pal method. 234 EFFECTS OF POISONS frequently than women, a proportion put by some observers as high as ten men to one woman affected. Pathology. — We have only had the opportunity of examining one case of this disease. In this case the lesions were very similar to those seen in ergotism, i.e. a pseudo-systematised degeneration of the long ascending and descending tracts of the cord. This was particularly marked in the pyramidal tracts, both lateral and ventral, in the direct cerebellar tracts, and in the dorsal columns. In the lumbar and thoracic regions the margins of the cord showed the same loose, honeycombed structure as is seen in the more rapid cases of " subacute com- bined degeneration," with oedema at the point of entrance of the dorsal roots. The nerve roots showed a definite increase of connective tissue, and in the peripheral nerves a similar increase was seen, chiefly in the epi- and peri-neurium. These changes are all of the same character as those found in ergotism, and it is probable that the two diseases are closely connected. 7. Beri-beri. This disease is characterised by polyneuritis or anasarca or both, and occurs either endemically, chiefly in the islands of the Indian and Pacific Oceans, or in epidemics. It is not confined to the human race, but may affect several species of mammal and is easily produced in domestic birds, such as pigeons and poultry. Aetiology. — Although the use of a too exclusive diet of rice has long been considered to be the cause of beri-beri, it is only within the last decade that the exact dietary conditions necessary to produce the disease have been worked out. It has been shown that an " accessory factor " is necessary in diets if beri-beri is to be avoided, and that the addition of this substance to the diet of patients suffering from the disease brings about a rapid cure. This factor has been called " anti- neuritic vitamine " or " water-soluble B." It has not been isolated in a pure condition. This vitamine is present in varying amounts in most natural foodstuffs, but may be removed or destroyed by the artificial measures adopted for their preparation. The " polishing " of rice and the milling of wheat to a white flour completely remove the vitamine whiQh BERI-BERI 235 is present in the germ and aleurone layer. The vitamine is fairly stable at ordinary temperatures, and can be preserved in the dry state for long periods. It withstands boiling or steaming at 100° C, but rapidly disappears when foodstuffs are autoclaved at 120° C. It is therefore absent from most forms of tinned food. The chief sources of this vitamine, as deter- mined by experimental work on pigeons, are rice husks, the germ of wheat and yeast. It is also present in fairly large amounts in the yolk of eggs, in ox-liver, and in certain cereals, such as linseed and lentils. Small amounts are present in milk, fresh meat, potatoes and other vegetables. Although normally the vitamine is present in milk, nursing women who are suffering from beri-beri even in a mild form may transmit the disease to their infants in a much more acute form. These infantile cases, which are always of the dropsical type, are rapidly cured by a change of milk, and are due to the absence of the anti- neuritic vitamine from their mother's milk. In experimental work on poultry the disease usually commences about fourteen days after the adoption of a diet from which the anti- neuritic vitamine is absent. In human epidemics the time of onset has been shown to be between eighty and ninety days. It is to be noted that although practically any diet from which the " water-soluble B " vitamine is absent will produce the disease in fowls within the periods stated, yet if they are deprived of everything except water, they survive for longer than this period without showing any signs of polyneuritis. This and the similarity of the neuritis to that produced by diphtheria toxin and various poisons suggest that the action of the vitamine is not a simple nourishment of nerve cells, but that it neutralises or inhibits the action of some poison which is produced in the metabolism of carbohydrates. The onset of the disease is so rapid as not to be easily explained on any simple theory of deficiency alone. The disease occurs in two clinical forms, the " wet " charac- terised by anasarca, and the " dry " characterised by paralysis and wasting of the limbs. These two types may occur together, and in the " wet " form there is usually a certain amount of paralysis. On the other hand, the " dry " form usually shows complete lack of dropsical symptoms. 236 BERI-BERI Morbid anatomy. — The two forms differ somewhat in their morbid anatomy, the same essential lesions being present but in different proportions. Anasarca and hydropericardium are present in the wet form, along with congestion of the spleen and a nutmeg condition of the liver. In all fatal cases of the disease there is some enlargement of the right side of the heart, with degeneration of the heart muscle. This may take the form of fatty infiltration, or there may be fragmentation of the muscle fibres with loss of their striation ; or again there may be areas from which the muscle fibres have completely disap- peared, their place being taken by connective and granulation tissues. The essential nervous lesion of the disease appears to be a multiple neuritis, which has been found in every case examined to this end. The peripheral nerves of the limbs may be healthy, when there is degeneration of such nerves as the phrenic, vagus, and sympathetic cardiac plexus, or of branches of the solar plexus. In the dry form there is always some polyneuritis of the limb plexuses. The neuritis is of the pure degenerative type, without any alteration in the connective-tissue com- ponents of the nerves. It consists, in the earlier stages, in slight degeneration of the myelin sheaths, with the appearance of black granules by Marchi's stain. Later the whole medullary sheath of certain nerve fibres may become broken up into fatty globules, without any obvious change in the axis cylinders. At a still more advanced stage there may be irregular swelling and fragmentation of the axis cylinders, but this is not usual. These changes affect only a small proportion of the nerve fibres, but those which show degeneration are affected throughout their course. Vedder has shown that the earliest changes in the nerve affect the structure of the neurokeratin network. In feeding fowls on a diet of polished rice he found changes in this structure as early as the seventh day, before any symptoms of paralysis had appeared. Along with the changes in the peripheral nerves there are always changes in the ganglion cells of the brain-stem and cord. These have been observed especially in the dorsal root ganglia and in the motor cells of the ventral horns and motor cranial nerves. They consist of chromatolysis affecting first the Nissl granules at the periphery of the cell along with vacuolation of EFFECTS OF POISONS 237 the cytoplasm and collections of pigment in the cell body. The cells which are most affected may be swollen and globular and contain a displaced nucleus. Degeneration of the dorsal columns, especially the tract of Goll, has been frequently observed. This appears to be secondary to the changes in the cells of the dorsal root ganglia, and is not so extensive as the degeneration in the corresponding peripheral nerves. Some observers, using Marchi's method, have found a few scattered degenerated fibres in all the tracts of the cord. The changes in the muscles are of the neuritic type and similar to those seen in other forms of motor-nerve palsy. Relationship of anatomical changes to clinical phenomena.— It has been frequently observed that the first steps in the cure of the paralysis may be extremely rapid, a few days of altered diet sufficing to get the patient on to his legs again. After that it may be weeks or months before he regains the full use of his limbs. The reason for this lies in the different periods required by the various elements affected to return to normal functional activity ; and as histological changes are found in a much larger proportion of nerve cells than of myelin sheaths, it is justifiable to assume that the rapid initial recovery is due to resumption of functional activity by the nerve cells, whereas the slower progress towards complete cure is associated with the remyeli- nation of peripheral nerve fibres. In the present state of our knowledge it is impossible to say whether the " wet " form of the disease is due primarily to degeneration of the cardiac plexus of nerves, to primary degeneration of the heart muscle, or to alterations of tissue metabolism favouring oedema. The evidence seems, on the whole, to be against the nervous theory, as the dropsy comes and goes without any change in the paralysis of the limbs. REFERENCES Neuritis. Bury, J. S.: Allbutt and Rolleston, System of Medicine, vol. vii., 1910, p. 415. Encephalopathy . MoTT, F. W. : (Lead). Arch, of Neurol., 1909, vol. iv.. p. 117. (Carbon Monoxide). Proc. Roy. Soc. of Med., 191 7, vol. x.; Path., pp. 73-90; Arch, of Neurol., 1907, vol. iii., p. 246. Oliver, T. : Allbutt and Rolleston, System of Medicine, vol. ii., part i, p. 988. 238 EFFECTS OF POISONS Ergotism. Allbutt, T. C: Allbutt and Rolleston, System of Medicine, vol. ii., part i, p. 884. Nageotte, J., AND RiCHE, A. I Comil and Ranvier, Manuel d'Hist. Path., vol. iii. Pellagra, Wilson, S. A. K.: Proc. of Roy. Soc. of Med., 1914, vol. vii., pp. 31-40. Beri-beri. Vedder, E. B. : Beri-beri. London, 1913. See also Report to Med. Research Council on the Present State of Knowledge concerning Accessory Food Factors {Vitamines). Special Report Series, No. 38, 1919. CHAPTER VIII TUMOURS OF THE BRAIN AND SPINAL CORD A. — Tumours of the Brain. It is customary to classify under tumours of the brain all forms of intracranial new growth ; in other words, all forms of tumour which tend, by increasing the cranial contents, to affect the functions of the brain. They are not all, strictly speaking, tumours of the brain substance. Some of them are metastatic, others arise from the vascular connective tissues, others from the meninges, and others from the cranial bones. Parasitic and other forms of cysts are also included among brain tumours on account of the similarity of the symptoms to which they give rise. Tumours of the brain have no special aetiology. For the most part their origin is unknown, except in the case of the parasitic and metastatic forms. No direct relationship to trauma has been proved, although, as in other parts of the body, there are plenty of records in which, as a by no means surprising coin- cidence, the onset of symptoms has been preceded by the history of some more or less trivial injury. In this connection it is perhaps worth remembering that the more severe the injury the less likely does it appear that a tumour should subsequently make its appearance. Certain forms of cysts, simple or haemor- rhagic, are undoubtedly of traumatic origin, and may give rise to all the symptoms of a new growth. The age incidence of cerebral tumours do3s not altogether resemble that of tumours in other parts. This is accounted for in part by the fact that tuberculomata and syphilomata, growths of early and mid-life, are often included in statistics, and in part by the fact that gliomata, one of the commonest forms of cerebral tumours, are more prevalent before than after forty-five years of age. It must be remembered also that carcinomata, so frequently the cause of death in elderly people, 239 240 TUMOURS OF THE BRAIN are comparatively rare in the brain, and, when they do occur, are almost invariably secondary to tumours in other organs. Tumours of the brain, therefore, are comparatively rare in infancy and in old age, but common enough in youth and middle life. It has been pointed out that subtentorial tumours are more common in early life, and supratentorial tumours more common in adults, but this is probably due to the inclusion of the tuberculomata of the cerebellum and pons which are not infrequent in children. Tumours of the brain are more common in men than in women. Glioma. This form of neoplasm rivals syphilitic gumma and tubercu- loma as the commonest form of intracranial tumour. Gliomata are diffuse infiltrating tumours arising from glial tissue. They may spread among the nervous elements of the grey or white matter of the brain, and attain considerable size without much destruction of nerve cells or fibres. While they are most common in the white matter of the centrum ovale, they are by no means unusual in the brain-stem, where they may so resemble the normal tissue as to have given rise to the old-fashioned name of " hypertrophy of the pons." They may be of any size and shape, and their limits are often extremely difficult to define. Owing to this and their infiltrating character, gliomata can very rarely be removed by surgical means. Their blood vessels are thin- walled and in parts very scarce, and the tumours are therefore liable both to softening and to haemorrhage into their substance. Both of these accidents may cause destruction of surrounding nerve cells and fibres, and thus it is common in the clinical history of such a case for focal symptoms to appear suddenly some weeks or months after there has been evidence of cerebral compression. These changes are characterised macro- scopically by patches of a reddish-brown or yellowish hue, interspersed among the greyish-red and more opaque white regions distinctive of the tumour itself. As a result of de- generative changes cysts, sometimes of large size, may appear in the tumour. The growth arises usually in the white substance of the brain, and is not directly connected with the blood vessels of the meninges, but it may reach the surface and form a prominent mass. GLIOMA 241 Fig. 69. i Glioma in parieto-occipital region seen on the mesial surface of the left hemisphere. Fig. 70. Glioma of corpus c^llosum with numerous haemorrhages. 16 242 GLIOMA Microscopic appearances. — As pathological neuroglia cells are of very various forms, so it is natural that tumours arising from glial tissue should be characterised by the variety of their histological appearances. Not only is this true of different Fig. 71. Cystic glioma of left frontal lobe. tumours, but different parts of the same tumour often present pictures which at first sight have no resemblance to one another. Thus, one part may be a mass of small rounded cells with com- paratively little intercellular network, and may closely resemble Fig. 72. Glioma of a rather fibrous type Fig. 73. Diffuse glioma of pons {hypertrophy of pons) 244 GLIOMA PONTIS Fig. 74. Diffuse glioma of brain-stem {sagittal section) . Fig. 75. Glioma of pons invading fourth ventricle. TUMOURS OF THE BRAIN 245 round-celled or mixed-celled sarcomata ; in another the tumour may consist of a felted mass of spider cells, while in yet another part cellular elements may be scanty, some small and rounded, others giant cells resembling multinucleated amoeboid glia cells. Cells with polymorphous or ring-shaped nuclei are not uncommon. Round cells with little tendency to the formation of processes are common in all forms of gliomata, especially in the more rapidly growing portions, but everywhere the variety Fig. 76. Glioma pontis. The nerve cells in the infiltrated area are undergoing degenerative changes, but one is practically normal. of cells met with is a most striking feature. Some tumours show a great excess of the large forms of cells ; others tend to approximate to the appearances of sarcoma. The latter form has often been called gliosarcoma, but as this term suggests a tumour which springs partly from mesoblastic and partly from epiblastic (nervous) elements, it is apt to lead to confusion. Careful examination of other parts of the tumour will usually resolve all doubts as to the category in which the tumour should be placed. 246 NEUROBLASTOMA The blood vessels are usually scanty. Their walls are thin and, except for the definite lining of endothelial cells, which is always present, supported almost entirely by glial fibres. This fact, and the tendency of all young glial tissue to de- generative changes, make haemorrhage and softening common occurrences in gliomata. Ependymal Glioma (Neuro-epithelioma Gliomatosum). This is a form of tumour which is most commonly met with in or near the ventricles, and also as a central intramedullary tumour of the cord. It usually forms a circumscribed mass of more or less rounded character, often with haemorrhages into its substance. It arises from cells destined to form ependymal cells, and presents the histological appearances of an irregular collection of tubules, between which is found a varying amount of gliomatous tissue similar to that found in other gliomata. Various names have been given to these growths, which are by no means common; they have been called by Rosenthal " neuro-epithelioma gliomatosum." Similar tubules of cells may often be found in various parts of true gliomata. The " rosette " formations of retinal gliomata are probably of a similar nature. Neuroblastoma (Ganglio-neuroma). These tumours may occur in the brain as fairly well defined, rounded masses, often showing some haemorrhage or softening. While they are more easily enucleated from the brain tissue than the gliomata, they do not come away clean, but separate through a zone of softening or gliosis which has formed around them. Histologically they are characterised by the presence of nerve cells and fibres. Some of the cells are large polygonal cells with clear vesicular nuclei resembling large ganglion cells; others are small rounded cells. By silver staining methods it is possible to make out numerous nerve fibres, which can be traced to the larger cells. The blood vessels are usually fairly well formed, with a definite endothelial lining and one or more layers of fibrous tissue. TUMOURS OF THE BRAIN 247 Neuro-fibroma. The incidence of neuro-fibromata within the skull is chiefly on the acoustic nerve. Tumours of this nature arising from the eighth nerve are commonly found in the cerebello-pontine angle, and give rise to very characteristic clinical symptoms. Usually they are single, but they may be bilateral, in which case they may be associated either with generalised neuro- fibromatosis, or with numerous neuro-fibromata of the cranial Fig. 77. Ganglip-neuroma. High-power view of ganglion cell area. Note amitotic division of nerve cells. (Bielschowsky's stain.) and spinal nerve roots. In a few such cases large psammomata have also been found on the cortex of the brain. Acoustic nerve tumours are irregularly shaped, rounded masses, rarely larger than a hen's egg, covered over with a fibrous layer derived from the arachnoid. On section they are firm and fibrous, and may show smaller or larger cysts or areas of rarefaction. Histologically they are composed of elongated cells, which take a brownish terra-cotta colour with van Gieson's stain, quite unlike that of connective tissue. These 248 NEURO-FIBROMA cells occur in bundles and whorls, and are occasionally seen lying side by side, nucleus by nucleus, and cytoplasm by cytoplasm, giving a banded " palisade " appearance to the field. In between aggregations of such cells are areas of a looser reticular appearance suggesting glial or myxomatous tissue. These appearances account for the terms " myxo- FiG. 78. N euro-fibroma of acoustic nerve. fibroma '' and " fibro-glioma " having been used to describe these tumours. The blood vessels are formed of a definite endothelial lining surrounded with one or more layers of fibrous tissue which may have become hyaline. Several authors have considered these tumours to be derived from the cells of the sheath of Schwann, and the term " neuri- noma " has been used with this meaning. TUMOURS OF THE BRAIN 249 Carcinoma. Pure carcinomata of the brain are always secondary to tumours in other organs, especially to scirrhus of the breast. Primary intracranial carcinomata have been described, but, with the exception of tumours of the pituitary gland, these are found on investigation to belong either to the form of " epithelioma gliomatosum " described above (p. 246), or to the endotheliomata. Carcinoma-like tumours arising from the choroid plexus are occasionally found in relation to the ventricles. Fig. 79. Acoustic nerve tumour, showing " palisade " arrangement of nuclei. Angeioma. This form of growth is found within the skull either as a well-defined mass with a more or less definite capsule or more usually as a diffuse meningeal naevus, which is occasionally associated with a similar structure involving the scalp. Angeiomata are also described in connection with the choroid plexus. They are not malignant tumours, but may give rise to severe symptoms. 250 TUMOURS OF THE BRAIN Sarcoma. These tumours may be primary, arising from the cranial bones or the meninges, or secondary, when they may be found in the substance of the brain. They are very variable in size and in rate of growth. On the whole, their consistence is firmer than that of the gliomata, and sometimes they are sufficiently well defined to be easily removable by the knife of the surgeon, especially when the growth lies on the surface of the brain and Fig. 8o. Carcinoma in the cerebral peduncles, secondary to scirrhus cancer of the breast. has not penetrated the soft meninges. They are always more clearly defined from the surrounding brain tissue than are the gliomata. Diffuse sarcomatosis is a rare condition characterised by a general invasion of the meninges of the brain, and sometimes those of the spinal cord as well, by sarcomatous cells. It may be primary or secondary to a mass of growth elsewhere. In general characters the sarcomata within the skull resemble those found in other parts. The majority of these tumours TUMOURS OF THE BRAIN 251 are spindle-celled; others may be round-celled or mixed-celled. Sarcomata arising from the dura mater occasionally erode the cranial bones and protrude on the surface of the skull. Perivascular Sarcoma (Perithelioma). These tumours may grow in the white matter of the brain without any obvious meningeal attachment. They are slow- growing tumours, usually single and showing little tendency Fig. 81. Perivascular sarcoma. High-power view of small vessel. to form metastases. In shape they are irregularly rounded, and fairly well defined from the surrounding brain tissue, from which, however, they do not peel away cleanly. On section they are rough and " woolly," with a tendency to the forma- tion of cysts or small areas of softening. Histologically they are composed of masses of small rounded or stellate cells surrounding a central, thick-walled blood vessel. In some of the slower growing forms the vessel wall shows very 252 SARCOMA great hyaline thickening, which may exceed the width of the lumen. In the more rapidly growing forms there appears to be proliferation of the smaller vessels to form several channels which run together in a " leash " formation. Their walls show considerable fibrous tissue and endothelial proliferation, so that the lumen is small as compared with the thickness of the tissues forming its wall. At a definite distance away from the vessel the tumour cells tend to degenerate, and thus in the older parts of the growth there are columns of cells with a central blood vessel separated from one another by necrotic cell debris. At the edge of the tumour these cell masses are seen to invade the neighbouring brain substance as finger-like processes. It has been suggested that these tumours arise from an out- ward proliferation of the cells lining the adventitial lymph spaces, and the name " perithelioma " has been used to convey this meaning, but as this theory of histogenesis has by no means met with universal acceptance it is better to retain the older name. Cylindroma (Myxo-endothelioma). i This curious form of tumour occasionally occurs in the brain, usually attached to the meninges, but sometimes deeply imbedded in the brain substance. It is a slow- growing tumour, which occurs in children more often than in adults. In form it is usually an irregularly rounded, bossed tumour which may be single or made up of two or more masses, easily separated from one lanother. On section the appearance is that of a sponge, thin strands of tissue separating numerous rounded areas of all sizes, which are filled with a grey gelatinous substance. Histologically they present a very definite picture of trabeculae of darkly staining epithelioid or fusiform cells surrounding areas of myxomatous tissue. The trabeculae are usually covered with one or more layers of cubical cells which have a considerable resemblance to cubical epithelium, but towards the centre of the trabeculae the cells assume a more and more fusiform type. In histological appearances these tumours thus resemble to a certain extent the endotheliomata, to which it is probable that they are allied. They are also TUMOURS OF THE BRAIN 253 closely allied to the " mixed tumours " of the parotid and other salivary glands. Endothelioma. This type of tumour occurs not uncommonly within the cranium. They are slow-growing benign tumours, usually arising from the arachnoid or the dura mater. They form rounded, well-defined masses often flattened as if by com- pression between the skull and the brain. Sometimes they seem to grow inwards from the pia-arachnoid as globular w t*.'i ^. Fig. 82. Cylindroma. tumours attached by a pedicle to the meninges. They are usually single, but may be multiple. They are readily removed by surgical intervention, as they usually push the pia mater in front of them. Sometimes they grow outwards, infiltrating the cranial bones and producing a " bossing "on the surface of the skull. The bone in such cases may be over an inch in thickness, porous and soft. It is common to find slight bony thickening or roughness over the tumour, and this may be visible by the 254 ENDOTHELIOMA Fig. 83. Etidothelioma growing from falx cerebri. tifiiltliC«l<( €) Fig. 99. Disseminated sclerosis. Sections of cerebellar cortex and of cord (one longitudinal and several transverse at various levels in the same cord). 288 DISSEMINATED SCLEROSIS where they often seem to spread out from the walls of the ventricles. All parts of the cord appear to be equally affected. Most observers agree that the cerebellar lobes are comparatively rarely attacked. The optic nerve suffers more frequently than any other cranial nerve. Patches have also been discovered occasionally in the ventral and dorsal spinal nerve roots. A good general idea of the distribution, size and shape of the islets of sclerosis can best be obtained by staining a number of sections transversely and longitudinally from different levels of the central nervous system by means of the Weigert-Pal method. The examination of such a series of sections brings into prominence the notable absence of deformity resulting from the presence of the sclerosis. It also serves to emphasise what is one of the most striking features of this disease — namely, the comparatively small amount of secondary degeneration which results from the morbid process. In our experience, secondary degeneration in the long tracts of the spinal cord is not quite so un- common as some observers suggest, but the discrepancy is probably explained by the fact that secondary degeneration only occurs when the long fibres of a particular tract have been exposed at various levels to numerous or perhaps unusually severe attacks. This explanation appears to be confirmed by clinical experience in a way which will be referred to below. If the patch be old, no trace of myelin will be contained within its area. If, on the other hand, it is of more recent origin, the Marchi method may reveal more or less numerous droplets of altered myelin, either in the medullary sheaths or in the sub- stance of compound granular corpuscles. Such fatty masses, stained a deep brown or black tint, may sometimes be seen collected in the immediate neighbourhood of the blood vessels. The axis cylinders may be shown in many cases, by Bielschowsky's impregnation method, to remain intact, although deprived of their medullary sheaths. They pass unaltered through the islet of sclerosis. In other instances the axis cylinders do not present an altogether healthy appearance. They may be swollen, varicose, fusiform, or generally thinned and atrophied in appearance. The changes in the neuroglia vary considerably in different patches, and depend probably DISSEMINATED SCLEROSIS 289 to a large extent upon the age of the particular plaque examined. In some, probably the older plaques, there is a thick network of neuroglial fibres and a sparsity of glial cells. In other patches the glial cells are more plentiful, and the ground substance shows a more finely granular structure and a more open network of fibres. The blood vessels seen in the diseased area may present a perfectly normal appearance. On the other hand, there may be indications of subacute inflammatory changes. For instance, the perivascular spaces may be dilated and contain lymphocytic and leucocytic cells of various sizes. In other patches the vessel walls may be notably thickened and show evidence of hyaline degeneration. When a patch reaches the surface the adjacent pia mater may be more adherent to the medullary tissue than is normally the case, and may show a certain amount of unnatural opacity due to proliferation of the connective- tissue elements. It is surprising how little the ganglion cells included in a sclerotic area may suffer, and in many cases it is impossible to say that any changes are present. Occasionally chromatolytic alterations can be detected, and still more rarely there is an obvious diminution in the number of the cells. Too much importance must not be attached to the presence of pigmentation, which is always of doubtful pathological significance. Relationship of anatomi al to clinial phenomena. — The clinical course of disseminated sclerosis is characterised by the occurrence of exacerbations and remissions. The sudden loss of power in a limb or equally sudden loss of sight in one eye may be the first symptom, and the recovery may be so complete in the course of a few weeks that an hysterical origin for the trouble is suspected. Other attacks follow at varying intervals, with the result that, although partial recovery always ensues, each leaves the patient somewhat more disabled. This clinical fact may be readily understood if we regard each attack as the effect of an inflammatory patch in the central nervous system which, while temporarily impairing the con- ducting power of the nerve fibres, does not destroy the axis cylinders. The latter are denuded of their myelin sheaths, but with subsidence of the inflammation regain their function 19 290 DISSEMINATED SCLEROSIS either completely or to a very considerable degree. In the course of time, however, the fibres of a particular tract are subjected to many pathological insults at different levels, and so tend to be permanently injured. In this way symptoms referable to disease of certain tracts become constant and the spastic paraplegia, which is so common a feature of the disease, is easily accounted for. Similarly the ataxy of limbs, which is another frequent symptom, may be brought into line with patches of sclerosis affecting the dorsal columns and cere- bellar tracts at various levels. The cranial nerve symptoms, particularly the optic atrophy, have the same origin. Oc- casionally optic neuritis, which represents an affection of the optic disc, may be observed. On the other hand, the absence of lower motor neuron disturbances — in other words, the absence of muscular atrophy — is due to the integrity of the ventral horn cells even when they are included in a diseased area. 4. Syringomyelia. Syringomyelia is a chronic and usually progressive disease of the spinal cord, characterised anatomically by the presence of one or more pathological cavities. It is more commonly found in men than in women, and the average age at which the symptoms first show themselves is somewhere between twenty and thirty. With the spinal disease, which is frequently regarded as congenital in origin, other congenital anomalies are often associated. For instance, the patient may be unduly small or infantile in his proportions, presenting a large head with diminutive trunk and limbs; or he may have unusual deformities of the skull, of the bones, of the limbs or of the vertebral column, such as a spina bifida. Cases of combined acromegaly and syringomyelia have been recorded sufficiently often to make the association interesting. Pathogenesis. — The number of theories which have been put forward to explain the origin of the morbid process underlying syringomyelia suggests that various cases of spinal cavita- tion, although roughly resembling one another, may have different sources. It is impossible to elaborate a genetic theory applicable to every case, and brief reference must be made to certain developmental, anatomical and pathological facts SYRINGOMYELIA 291 which may throw some light on particular examples of the disease. (a) Developmental. — The medullary canal of early foetal life is only represented clearly in the child by the central canal of the cord, which is a cylindrical tube lined by ependymal cells of epiblastic origin running throughout the length of that organ. This central canal, however, represents only the ventral limb of the medullary canal, the dorsal and lateral limbs disappearing as the dorsal columns of either side gradually coalesce in the region of the dorsal median fissure. Under normal conditions there are no cells resembling the ependymal cells of the central canal, either in the lateral parts of the grey commissure, in the bases of the dorsal horn, or in the walls of the dorsal median fissure. It is not unreason- able, however, to suppose that in some cases there may be in these three contiguous parts embryonic remnants capable of renewed activity, and so of produfcing neuroglia which is the most primitive tissue of epiblastic origin in the central nervous system. Since the neuroglial hyperplasia or gliomatosis of syringomyelia nearly always originates either in the grey commissure, the bases of the dorsal horn, or in the ventral third of the dorsal columns, its source may justifiably be connected with the developmental process just referred to. The occasional presence of groups of ependymal cells away from the central canal in otherwise normal cords may be regarded as a congenital accident which, under certain stimulating conditions, might give rise to a peri-ependymal hyperplasia of the glial tissues just as such hyperplasia certainly originates in some instances from the walls of the central canal itself. [h) Anatomical. — A consideration of the spinal vascular supply brings out one important point. That part of the spinal cord which includes the grey commissure, the ventral third of the dorsal columns, and the bases of the dorsal horns, is the centre of the cord from a vascular point of view. It receives blood from various radiating vessels, of which the most important are those entering along the dorsal roots and along the dorsal septum, as well as from some of the terminal branches of the ventral spinal arteries. It contains, therefore, vessels of small calibre, some of which have a transverse and 292 SYRINGOMYELIA some a longitudinal course. This fact is important in two ways. In the first place, any general disease of the spinal arteries which impairs the elasticity of their walls will produce its greatest effect on the circulation within the central zone. In the second place, any morbid process leading to pressure upon, or strangulation of, the pial vessels on the dorsal surface of the cord will have a profound effect upon the blood supply of the same area. In the same way the central zone of the spinal cord is supplied by lymphatics entering with the dorsal roots. For this reason, some exceptional cases of syringomyelia have been regarded as infective in origin, the infection being carried in along these lymph channels, and perhaps being further spread by the fluid of the central canal. The closure of the central canal at one level of the cord has been suggested as a sufficient reason to explain dilatation of that canal at other levels. Such an explanation is open to question when it is remembered that in normal adult cords the canal may be obliterated in some segments, and yet patent without dilatation above and below. (c) Pathological. — The partiality of intramedullary haemor- rhages and abscesses to make a track from segment to segment along the tissues at the base of the dorsal horn in the central grey matter is so striking that this area has been termed the zone of least resistance. The fact that gliomatosis spreads along similar lines has led to the belief that some cases of syringomyelia have their remote origin in spinal haemorrhage occurring at birth. It is possible, therefore, although not proved, that an old pathological process may, after a con- siderable lapse of time, be the starting-point for gliomatosis. These developmental, anatomical and pathological data taken into consideration with clinical observations suggest that various influences are at work in the production of spinal cavities. Thus there are cases in which a congenital defect appears to carry great weight; others in which a morbid condition of the vascular supply, produced perhaps by a chronic meningitis on the dorsal surface of the cord, seems too important to overlook; and others, again, in which some chronic lymphatic infection or a history of some traumatic haematomyelia require to be taken into consideration. The fact that gliomatosis is always associated in greater or less SYRINGOMYELIA 293 degree with the formation of cavities is best explained by the tendency of ghomatous tissues after reaching a certain bulk to undergo central degeneration, which tendency is doubtless promoted by the deficiency in its blood supply. With regard to the ordinary cases of chronic progressive syringomyelia, which form the majority of those with typical clinical features, little hesitation can be felt in placing fore- most some congenital anomalies of the central embryonic tissue, the latent activity of which is prone to display itself during the early years of adult life in the form of a slow-growing hyper- plasia, having some of the characters of a benign neoplasm and strong tendencies towards the formation of cavities. Morbid anatomy. — The spinal cord appears altered in shape, often very irregularly. In most instances the cervico-thoracic region is enlarged and flattened, the lateral dimension being increased out of proportion to the antero-posterior. Some- times the deformity may extend as high as the medulla or even the pons, and in other cases may involve a large part of the thoracic region. The lumbo-sacral region is occasionally the site of the greatest change. The meninges in the majority of cases present no abnormality ; in others there may be a definite chronic meningitis, probably of syphilitic origin. The ventral spinal roots at the level of the greatest deformity may be atrophied and translucent in appearance. On palpation the swollen parts are soft and usually flaccid. The narrow parts may be normal in consistence or suggest to the palpating finger the presence of a hard core lying within an outer ring of more normal tissue. A series of transverse slices demonstrates striking changes which vary in appearance at different levels. In the region of the greatest swelling, usually the cervico-thoracic region, the cross-section appears at first sight to have traversed an area of necrotic softening, but closer investigation shows that near the centre of the cord is a cavity from which clear fluid may be oozing, and that this cavity is surrounded by a mass of gelatinous material either pale and translucent or yellowish-brown, according to the amount of altered blood pigment it contains. The actual wall of the cavity sometimes stands out as an opaque yellow membrane which has been thrown into folds by the escape of fluid and the collapse of the surrounding parts. Fig. ioo. Syringomyelia, a, Note the excessive gliomatosis and the invasion of the grey matter from the dorsal columns, b, The cavity is distinct from the central canal, c, The cavity has a very well-defined lining membrane. Note the descending degeneration in the lateral columns. (Compare with Fig. 12.) SYRINGOMYELIA 295 Further sections at different levels will display the length of the cavity and sometimes the presence of more than one cavity. Rarely the tube is prolonged into the fourth ventricle, although in one case it extended as far as the right internal capsule. When two cavities are present side by side it is the rule to find on further investigation that one is an offshoot of the other. The syringomyelic cavity generally lies behind the ventral commissure either in the grey matter near the central canal or more laterally in the base of one or other dorsal horn. Diverticula in various directions are common enough and Fig. ioi. Syringobulbia. produce varieties in the shape of the cavity at different levels. The general result is to produce asymmetry rather than symmetry in the relation of the two halves of the cord to one another. In the medulla, as in the spinal cord, the morbid process has a favourite site. Frequently the only evidence of disease in this region is the presence of one or more fissures originating in the floor of the fourth ventricle a little to one side of the mid-line, and extending forward and outward in such a way as to cut off the restiform body from the central parts of the medulla. 296 SYRINGOMYELIA Such a fissure may destroy the descending root of the fifth nerve, the sohtary fasciculus, and some of the nuclei belonging to the vago-glossal pharyngeal nerves. The disposition of the surrounding gliomatous tissue follows roughly the various diverticula, forming prolongations of its own and extending to levels beyond the limits of the cavity. The gliomatous tissue varies greatly in thickness. It may present the characters of a thin lining membrane to a large cavity, or it may be so voluminous as to overshadow the narrow slit-like tube it contains. The gelatinous-looking material is composed of fibres and cells in varying proportion. There are generally more cells than in normal neuroglial tissue, and less cells than in a true glioma. The glial tissue may actually form the wall of the cavity or may be separated from its lumen by a narrow border of ependymal cells. The peripheral parts of the gliomatous mass merge into the more normal neuroglia and nerve tissues. Often when there is no central cavity there may be areas of softening containing finely granular or homogeneous material. The syringomyelic cavity may sometimes represent the central canal, but more often is found distinct from the latter. In some instances the cavity is fused with the central canal over several segments and separated from it in regions beyond. The blood vessels of the cord are often perfectly healthy, but in other cases those which radiate from the surface into the central gliomatous tissue show hyaline and other degenerative changes. The lumen may be diminished or obliterated, es- pecially when it reaches the glial mass, and in the neighbourhood of the cavity the remnant of a vessel may be represented by an undulating ribbon of hyaline connective tissue, which has been unfortunately called a papillary membrane. Old or recent haemorrhages are not infrequent, and the pigment may give a brownish tinge to the tissues. The effects of gliomatosis and cavitation are due in part to pressure and in part to direct invasion. The latter method is the one by which the dorsal columns are principally affected, the new-formed tissue eating its way, as it were, between the bundles of nerve fibres and causing their slow destruction. Pressure leads to the production of oedema around the glial mass, and this is often responsible for the SYRINGOMYELIA 297 rarefaction of the grey matter of the ventral horns. The nerve cells become isolated and their processes disappear, with the result that a species of cavity may be formed to which the name " perigHomatous " as opposed to " endogliomatous " has been given. Secondary degenerations in the pyramidal tracts, in the dorsal columns and in the ascending ventro-lateral tracts, are of common occurrence, and are mainly produced by pressure on those regions. Secondary degenerative changes occur in other tissues, such as the nerves, muscles, skin, joints, and bones, but do not present features which are peculiar to this disease. Some reference must be made to the not infrequent con- currence of a true neoplasm with the syringomyelic process. At any level of the cord, commonly in the cervical region, may be found a true tumour involving more or less the entire transverse area, which has evidently originated from a part of the gliomatous mass. Such tumours may be very cellular gliomata, vascular angiogliomata, or more rarely sarcomata. The relation of the anatomical to the clinical phenomena. — The fact that the cervico-thoracic region of the cord is generally the site of most change explains the frequency of a particular type of syringomyelia, which is characterised by an atrophic palsy beginning in the hands and forearms and a spastic paralysis of the trunk and legs. The muscular atrophy is, of course, the result of the secondary effects on the ventral grey matter described above, and the spastic paraplegia is due to the interference, chiefly by pressure, with the pyramidal tracts. The characteristic dissociative anaesthesia of syringomyelia can be explained by the incidence of the disease on the central parts of the cord. There is reason to believe that the fibres conducting thermal and painful impulses cross from one side to the other in the grey commissure, whereas those carrying tactile impulses are represented both in the lateral and the dorsal columns, and are thus less profoundly influenced by a central lesion at any particular level. The trophic changes in bones, joints, and skin in this disease have not yet received a complete pathological explanation, and can hardly be brought into relationship with the anatomical changes in the spinal cord at the present time. 298 PARALYSIS AGITANS 5. Paralysis Agitans (Parkinson's Disease). This is a slowly progressive disease of unknown causation, usually beginning in middle life or later, but in a minority of cases in the second or third decade. It is characterised by the three cardinal symptoms of tremor, rigidity and weakness. Usually all the limbs, as well as the face, neck and trunk, are affected, but at first the disability is hemiplegic or monoplegic in distribution, and may remain for many years more pronounced on one side. Morbid anatomy. — For a long time the site of the lesion in this disease was unknown, but a number of cases examined during the past few years have presented changes in the corpus striatum. The earliest worker to observe changes in this region was Jelgersma, who, in 1908, reported a case with atrophy of certain tracts leading from the corpus striatum to the mid-brain. This observation has been largely confirmed by subsequent writers, some of whom have found areas of rarefaction or softening in the corpus striatum. In 1917 Ramsay Hunt, in the examination of a case of the juvenile form of the disease which started in the second decade, found that the essential lesion was a primary atrophy of the large cells of the " pallidal system."* These were greatly reduced in numbers, and the remaining cells showed various stages of chromatolytic degeneration often associated with neuronophagy. Along with this there was a great increase in the number of glial cells in the affected areas. Except for a moderate thickening of the walls of the blood vessels in certain areas, no vascular changes were present to account * According to Hunt the cells of the pallidal system are large multipolar cells found chiefly in the globus pallidus, but also scattered among the smaller cells of the'putamen, caudate nucleus and basal nucleus of Meynert. In histological appearances they resemble the large motor cells of the ventral horns and the Betz cells of the precentral convolution. They represent an alternative motor system to the pyramidal system, than which they are phylogenetically older, being present in the lower forms of fishes. They are connected with the optic thalamus through the medium of the smaller cells of the corpus striatum, but apparently have no direct connection with the cortex or the internal capsule. Their efferent fibres run in the ansa system — the ansa radiations, ansa lenticularis and ansa peduncularis. Kinnier Wilson has divided these fibres into two groups : (i) coarser fibres which run transversely to end in the ventral and lateral surface of the thalamus and in the nucleus ruber; (2) smaller fibres which terminate in the corpus Luysii, the nucleus ruber and the substantia nigra. PARALYSIS AGITANS 299 for the cell degeneration, which he considered to be a primary atrophy. All other parts of the brain and brain-stem were normal. The tracts leading from the corpus striatum to the mid-brain showed some thinning of the fibres composing them, but there was no such definite tract degeneration as is seen, for example, in hemiplegia. This case suggests that the symptoms of paralysis agitans may be due to a primary progressive atrophy of the cells of the pallidal system. More recently Hunt has reported the results of examining two cases of the ordinary presenile form of the disease, which also showed degeneration of the pallidal system of cells. In some cases the degeneration may be a form of primary atrophy, associated with some toxic absorption or error of metabolism. In others, it is a senile atrophy due to abiotrophy or to vascular disease. The clinical picture of paralysis agitans may be produced by gross lesions, such as encephalitis, haemorrhage, softening or tumour, affecting the globus pallidus of the lenticular nucleus. Relation of anatomical with clinical phenomena. — The pallidal system of cells seems to have some connection with the faculty of synergic movement, which is preserved in lesions of the pyramidal system, whereas its loss constitutes one of the chief disabilities of paralysis agitans. The festinant gait, and the phenomena of propulsion and retropulsion, are connected with this loss. The peculiar tremor of the disease has long been associated with mid-brain lesions. 6. Myasthenia Gravis. Aetiology. — The essential cause of this disease is still a matter of speculation, and the various theories which have been pro- pounded are not based on substantial foundations. Age has but little influence, although the majority of cases occur in the third and fourth decades of life. Both sexes are attacked in about equal proportion. There are no familial or hereditary tendencies. The onset of symptoms may follow infective fevers, exertion, chills, and emotions, but not more frequently than may be explained by the laws of coincidence. Although some victims of the disease have presented abnormalities, such 300 MYASTHENIA GRAVIS as bifid uvula, Polydactyly and webbing of the toes, there is not sufficient evidence to justify emphasis being laid on the presence of congenital defects. Pathogenesis. — Numerous hypotheses have been put forward to throw light on the obscurity in which the pathogenesis of myasthenia gravis is shrouded. Before any morbid anatomy had been recognised, it was regarded as a neurosis of congenital origin. With the discovery of enlarged thymus glands in certain cases, there was a natural inclination to assume that poisons produced by them and capable of modifying nervous and muscular function were thrown into the circulation. This view has been upset by the fact that many cases of the disease do not present any abnormalities in connection with the thymus gland. Other observers have found congestion and enlargement of the parathyroids, and consider myasthenia to be due to hyperactivity of these organs. The influence of the parathyroids on the functional activity of muscles is shown by experimental tetany which, on this theory, is the converse of myasthenia. One author maintains that the disease is characterised by hyperoxygenation, which is shown especially in connection with voluntary movements, and which results in the production of chemical alterations in the musculature owing to some inter- ference with the formation of antibodies of fatigue. It has been suggested, on the other hand, that the clinical manifesta- tion of muscular fatigue may possibly be explained by as- suming a diminished functional activity on the part of the sarcoplasmic, as compared to the fibrillar, elements of the muscles, with the result that the fibrillar constituents, acting at a disadvantage, become readily and rapidly exhausted when excited by the will or by the faradic current. No pathogenetic theory can afford to neglect the oft-con- firmed observation that women who suffer from myasthenia may lose all their symptoms during pregnancy, only to relapse into their former condition after confinement. Although this is not an invariable rule, it is sufficiently common to indicate that the foetal tissues may temporarily provide some substance which either counteracts a circulating poison or supplies a deficiency in the body of the mother. We may not be wrong in supposing that some disturbance of MYASTHENIA GRAVIS 301 glandular function is responsible for a modification of muscular activity, and for such sensory and mental symptoms as are occasionally met with in myasthenic patients. Morbid anatomy. — The absence of obvious and constant changes in the nervous system in cases of myasthenia gravis is now generally acknowledged, and the only characteristic feature is the presence of cellular deposits in many and various organs of the body, and particularly in the skeletal muscles. Reference must also be made to certain abnormalities of the thymus gland which only occur, however, in a certain percentage of cases. In the central nervous system developmental defects have been noted in rare instances, and a very few observers have recorded slight chromolytic changes in ganglion cells. As a rule, examination of the nervous system reveals a perfectly normal condition. Small deposits of mononuclear cells {lymphorrhages) have been observed in the spinal root ganglia, the medulla and elsewhere, but their occurrence is relatively infrequent when compared to that of similar deposits in the muscles. Recent capillary haemorrhages, when present, are the result of the respiratory embarrassment which so commonly brings about the fatal termination of the disease. Lymphorrhages are more commonly described in the muscles than in other organs, possibly because they have been more thoroughly investigated. The constituent cells are found in ill-defined clumps between the muscle fibres and generally in the vicinity of a capillary vessel. There may be a serous as well as a cellular exudate. The size of a lymphorrhage varies con- siderably ; it may be very minute or large enough to be detected in a stained section by the naked eye. Lymphorrhages have been demonstrated in many muscles, especially the ocular muscles, and also in the myocardium, but it may be necessary to examine very many sections in order to prove their existence. Speaking generally, the neighbouring muscle fibres are healthy, but occasionally they have undergone degenerative changes, and have been even invaded by the lymphocytes. In addition to the presence of lymphorrhages the muscle fibres are wont to display early changes in the form of plasmatic swelling, proliferation of nuclei, hyaline and granular degeneration. In other cases there has been distinct muscular atrophy, 302 MYASTHENIA GRAVIS #* *» Aly isths/iia gfavis. a, Drawing of transverse section of an ocular muscle showing a " lymphorrhage." b, Photograph of tranverse section of a skeletal muscle. MYASTHENIA GRAVIS 303 resembling that seen in progressive muscular atrophy of the spinal type. The thymus gland is often the seat of morbid changes; sometimes it is only represented by the remnants of lymphatic tissue characteristic of the organ in adult life. The abnormali- ties are not specific, and may be divided into three classes: (i) simple hypertrophy, (2) hypertrophy with degenerative and proliferative changes, and (3) new growth. In the first class may be included those cases in which the gland, not having undergone the ordinary regressive changes, is as large or larger than that of infants. It may be histologically normal or lacking in eosinophil cells. Large glands with multilocular cysts are instances of the second class. New growths are represented by lympho-sarcomata and other rarer forms of neoplasm. The thyroid gland has been found to be the site of lymphor- rhages, interstitial fibrosis, colloid degeneration of the fibrous stroma, and proliferation of the epithelium with formation of new vesicles. The pituitary body has, in one case, presented a large adenoma. The liver may be the seat of numerous lymphorrhages, especially in the neighbourhood of the biliary ducts. Serous as well as cellular exudation is often found, and a moderate degree of fatty change in the hepatic tissue has been observed. The adrenals, the kidneys, the lungs, and the pancreas have all been known to contain lymphorrhages, but other abnormalities are rare. The bone marrow is free from notable changes, and the blood and cerebro-spinal fluid have been repeatedly searched in vain for abnormal features. The relation of clinical to anatomical phenomenon. — We are too ignorant at present of the exact nature of this disease to draw any important deductions. Lymphorrhages are certainly not the cause of the alteration in muscular function, but the slight degenerative changes in the fibres may confirm the im- pression, gained from clinical experience, that the character- istic exhaustion is of muscular rather than of nervous origin. This is also supported by the absence of constant changes in the nervous tissues. 304 DISEASES OF OBSCURE ORIGIN REFERENCES Motor Neuron Disease. Beevor, C. E., Batten, F. E., and Holmes, G. : Allbutt and Rolleston, System of Medicine, vol. vii., 1910, p. 699. Subacute Combined Degeneration. Collier. James: Allbutt and Rolleston, System of Medicine, vol. vii., 1910, p. 786. Disseminated Sclerosis. Da Fano, C: Journ. New. and Ment. Dis., 1920, vol. li., p. 428. Dawson, J. W. : Transactions Roy. Sac. Edinburgh, 1916, vol. i., part 3 (No. 18), PP- 517-740- Russell, J. S. R.: Allbutt and Rolleston, System of Medicine, vol. vii., 1910, p. 809. Syringomyelia. Nageotte, J., AND Riche, A.: Cornil and Ranvier, Manuel d'Hist. Path., vol. iii., 1907. Starr, M.A.: Allbutt and Rolleston, System of Medicine, vol. vii., 1910, p. 852. Paralysis Agitans, Hunt, Ramsay: Brain, 191 7, vol. xl., p. 58. Wilson, S, A. K.: Brain, 1914, vol. xxxvi., p. 427, Myasthenia Gravis. Buzzard, E. F.: Allbutt and Rolleston, System of Medicine, vol. vii., 1910. p. 50. Claude, H., and Porak, R.: L'enciphale, 1920. p. 425. APPENDIX I STAINING METHODS Introductory. As a general rule celloidin is preferable to paraffin for imbedding and cutting tissues of the nervous system, as it causes less shrinkage of the tissues and preserves the glial structure better. Celloidin sections are also easier to cut and to handle in certain special methods, such as those of Marchi and Weigert-Pal. On the other hand, for Nissl's method, especially when used on serial sections, paraffin sections are more convenient. Frozen sections are necessary for some of the routine methods, and may be used for all, their only disadvantage being their fragility. It is not proposed to give here an account of all the methods wljich have been and may be applied to the nervous system, and many, especially those dependent on special primary fixation, have been omitted. The majority of the methods described in this section are those which are used as a routine in the examination of nervous tissue. It is not always necessary nor desirable to use all these methods in any given case, but for the full examination of a case of rapidly progressive disease it is necessary to stain for nerve cells by Nissl's method or one of its modifications ; for lipoids, by ScharlachR. or Sudan III; for neuro-fibrils, by theCajal or Bielschowsky method ; for myelin sheaths, both by Weigert-Pal and Marchi methods ; and for neuroglial cells and fibres, by tissue stains, such as haematoxylin and van Gieson. Fixation. Although certain special methods (notably that of Cajal for neurofibrils, in which alcohol-ammonia fixation is a sine qua non, and Nissl's method for cells, for which alcohol is the best fixative) demand the fixation of small pieces of tissue directly in a special fluid, as a general rule the brain and cord are fixed in toto in lo per cent. formaHn sahne, which is changed on the second or third day. We insist on the addition of i per cent, sodium chloride to the formalin solution, as it not only helps to retain the original size and shape of the tissues, but also allows the formalin to penetrate better. It also allows the first sHces across the brain to be made at the end 305 • 20 3o6 STAINING METHODS of five to seven days without any resulting deformity of the brain, which is of great advantage for early diagnosis. When this is done the brain fixes more rapidly and completely, and will usually be found to be completely fixed at the end of a fortnight. For the fixation of a whole brain, four litres of formalin saline are necessary, and the pot used should have a lid which fits sufficiently well to prevent the formalin evaporating out of the fluid, and handles at either side, to which a string can be tied which passes under the basilar artery and suspends the brain in the fluid. At the end -of a week it is possible to select pieces for examination by special methods. It is often convenient to run a horsehair through these, so that they are retained in their proper order in their passage through the various fluids. Those for the VVeigert-Pal and Marchi methods are then put directly into Muller's fluid, or Weigert's primary mordant, while those for freezing and for Nissl and tissue staining are either fixed for two days further in formalin or preferably in Zenker's fluid for twelve to twenty-four hours. In either case they are washed for twenty-four hours in running water before anything further is done. Imbedding in Celloidin. After washing in water, pieces up to 5 mm. in thickness are passed through increasing strengths of alcohol up to absolute alcohol, in which they remain for twenty-four to forty-eight hours. They are then transferred for a similar time to equal parts of alcohol and ether, and from this passed into a thin solution of celloidin in alcohol and ether (which should be so thin as to give only a slight indication of viscosity when it is shaken about in the bottle). After remaining in this for at least a week, they are changed into thicker celloidin (about 6 per cent.). In this they remain any time from four days to a fortnight. They are then placed in order in a shallow flat- bottomed glass vessel, such as an evaporating basin, the side from which the first sections are to be taken being uppermost. Celloidin as thick and viscid as possible is then poured on to cover them . The dish is covered completely for the first day in order to allow of the escape of any bubbles of air, and then the cover is raised shghtly the second dav, and almost completely thereafter until the celloidin is so firm that the pressure of the finger makes no impression on it. The celloidin is then removed entire from the dish, by cutting round the edges, and trimmed into small pieces containing the blocks of tissue. These are then mounted on hard wood blocks by means of 6 per cent, celloidin, the side of the tissue which was next the glass resting on the TISSUE STAINS 307 wood. Cheap and excellent blocks are made from elm or oak, which has been soaked in methylated spirit to rid it of resin, and cut to the required shape. Smoother woods, such as birch or sycamore, afford less grip for the celloidin, but can be used if shallow saw cuts are made across the surface on which the blocks rest. After the tissue is mounted on the blocks they are named and numbered with Chinese ink, and when the celloidin is dry are stored in spirit in large-mouthed bottles for at least twenty-four hours before cutting. Wooden blocks with tissue mounted on them can be kept in 50 to 60 per cent, alcoholfor an indefinite time, and it is always preferable to keep them in this way until they are finished with. A. — General Tissue Stains. Haematoxylin and van Gieson. Overstain sections, either frozen, paraffin, or celloidin, in any good haematoxylin mixture. Mayer's haemalum, Delafield's alum haematoxylin and Weigert's iron haematoxylin are to be recom- mended. Decolourise with acid alcohol, not so completely as for the eosin counterstain. Wash thoroughly in tap water. Stain for eight to twelve seconds in van Gieson 's mixture, which should be prepared from the stock solutions within a few days of use. Van Gieson 's stain: Sol. A. — Acid fuchsin . . . . . . . . i gramme. 0*6 per cent. sol. of picric acid in distilled water . . . . . . 100 c.c. Sol. B. — Sat. sol. (o'6 per cent.) picric acid in distilled water. For use take i part of A and 9 parts of B. Wash rapidly in distilled water or in tap water, to which a few drops of acetic acid have been added to render it acid. Transfer for a few seconds to 95 per cent, alcohol. Paraffin and frozen sections are then transferred to absolute alcohol and xylol, celloidin sections to carbol-xylol (phenol crystals i, xylol 3), and when clear to xylol. It has been recom- mended that the xylol should be acidified either by saturation with salicylic acid or with acetic acid (2 drops for each 100 c.c.) : this is not necessary if the xylol is neutral. Mount in Canada balsam which has been saturated with salicylic acid. This is absolutely necessary if the sections are to retain their pink staining, and its omission has been responsible in the past for the unpopularity of this stain. By this method fibrous tissue is stained red, elastic tissue and red blood corpuscles yello.w, and muscular and nervous tissue brownish or terra-cotta. 3o8 STAINING METHODS B. — Nissl Granules. The best fixative for Nissl granules is 95 per cent, alcohol for three to five days, changed every day. Only thin pieces of brain should be fixed in this way, but the cord can, if desired, be fixed entire. In most cases, however, this is not desirable, as fixation with alcohol prevents any study of degeneration in the myelin sheaths. Formalin or any of the corrosive sublimate fixatives (e.g. Zenker's or Dominici's fluids) may also be used, in which case the pieces must be washed for at least twenty-four hours in running water after fixation. Sections may be made by the freezing, paraffin or celloidin methods. Nissl attached pieces of alcohol-fixed tissue to blocks of wood by means of gum arabic, and cut them as if they had been celloidin blocks. Celloidin sections are most easily and satisfactorily stained after removal of the celloidin by placing them in alcohol and ether for twelve to twenty- four hours. They are transferred through absolute and 80 per cent, alcohols to distilled water before being stained. Paraffin sections are treated by the usual method for removing paraffin and placed in distilled water. Stain with either polychrome methylene blue, i per cent, toluidin blue, or 2 per cent, thionin blue, for twenty to thirty minutes in the paraffin oven from 50° to 55° C, or stain in Nissl's methylene blue-soap solution, which has been allowed to ripen for at least three months. Methylene blue . . . . . . • • 375 grammes. Venetian soap shavings .. .. .. 175 grammes. Distilled water . . . . . . . . 1,000 c.c. Shake well. The sections are stained in this solution by heating the stain rapidly over a fiailie until bubbles rise. Wash rapidly in water. Differentiate first in 90 per cent, alcohol. If the sections are deep blue in colour after the first few minutes in this, they may be differentiated more rapidly by either of the following solutions : 1 . Anilin oil (colourless) Alcohol, 95 per cent, or absolute . . 2. Gothard's solution: Creosote (pure beechwood) . . Cajuput oil Xylol Absolute alcohol 10 c.c, 90 c.c, 50 c.c, 40 c.c, 50 c.c. 50 c.c NISSL GRANULES 309 In either case the differentiation should only be continued until the nerve cells stand out plainly from the surrounding tissue. Continue the differentiation with absolute alcohol until the Nissl granules stand out clearly from the rest of the nerve cell, which, with the exception of the nucleolus, should be almost or completely colourless. This process is best watched under the microscope after transferring the sections ternporarily to xylol. The differentia- tion need not be carried far enough to decolourise the nuclei of the glial or connective-tissue cells, or of the smallest nerve cells. Wash for about fifteen minutes in xylol. Place on slides, blot and mount with a coverslip by means of pure cedar- wood oil. The slides are then put into the incubator at 37° C. for twenty-four hours, after which the cedar-wood oil will have set firmly. It is possible to use Canada balsam if this is neutralised by satura- tion with lithium carbonate. It offers no advantages over cedar- wood oil, in which the sections should remain for years without losing their colour. If it is necessary to keep the celloidin in the sections during the whole of the process, the following variations in the technique must be made. The sections should be thin, not more than 8 {jL. After staining they are transferred to 95 per cent, alcohol; then, if necessary, to anilin oil — alcohol made up with 95 per cent, alcohol and returned for final differentiation to fresh 95 per cent, alcohol. They are then cleared with cajuput or origanum oil, washed in xylol, and mounted as already described. Only the purest celloidin can be used for this purpose, and even this will be found to retain the stain to some extent. Rosin's Method for Nissl Bodies. Rosin's method will be found useful, both for micro-photography and where it is wished to counterstain preparations in which the myelin sheaths have been stained black. He uses saturated watery solution of neutral red in place of basic blue solution. Differentia- tion is done rapidly with alcohol alone, and sections are mounted as before. C— Neuro-Fibrils. Ramon-y-CajaVs Method. Fix small pieces of tissue for twenty-four to forty-eight hours in : 96 per cent, alcohol . . . . . . . . 100 c.c. Ammonia -880 . . . . . . . . . . 0-25 c.c. 310 STAINING METHODS Wash rapidly in distilled water, and transfer to 3 per cent, silver nitrate solution for five to six days in the incubator at 37° C. Wash rapidly in distilled water. Reduce in the following mixture for twenty-four hours in the dark at room temperature (two stock solutions are prepared and kept separately) : Sol. A. — Sod. sulphite Formol Distilled water Sol. B. — Pyrogallic acid Distilled water 5 grammes. 50 c.c. 200 c.c. 20 grammes. 800 c.c. For use take 25 c.c. of A and 80 c.c. of B. Wash for a few minutes in distilled water and imbed in paraffin or celloidin. Cut thin sections. After the sections have been passed into distilled water they are placed in a combined toning and fixing bath, where they remain for a few minutes until they become of a grey- violet colour. Toning and fixing bath: Ammon. sulpho-cyanide . . . . . . 3 grammes. Sod. thiosulphate . . . . . . . . 3 grammes. Distilled water to 100 c.c. Add just before use a few drops of i per cent, gold chloride. Wash, dehydrate, clear, and mount in Canada balsam. Bielschow sky's Method {Modified by Da Fano). Fix small pieces of nervous tissue for at least eight days in 20 per cent, formol (which should not be acid). Cut at 15 to 20 // on the freezing microtome- After washing in distilled water for one or more hours the sections are transferred to a mixture of equal parts of 20 per cent, formol, and of either methyl alcohol or pyridin for twenty-four hours. Wash for six to twenty-four hours in several changes of distilled water. Put into 2 per cent, silver nitrate solution for twenty-four hours in the incubator at 37° C. Wash for a few seconds only in distilled water, and transfer to Bielschowsky's solution made up as follows : To 5 c.c. of 20 per cent, silver nitrate solution, in a clean 50 c.c. cylindrical measure, add 2 drops of 40 per cent. NaOH. A heavy brown precipitate is formed. (It is sometimes preferable to wash this with distilled water once or twice, allowing it to settle and NEURO-FIBRILS S" pipetting off the top fluid before proceeding. In any case, care must be taken that the powdery white substance which tends to gather round the neck of the bottle of caustic soda is washed off with distilled water before it is used.) Then dissolve up the precipitate of silver hydroxide with strong ammonia, adding only just enough ammonia to dissolve the precipitate with thorough stirring. Dilute to 40 c.c. This solution should be perfectly colourless. In this the sections remain for thirty minutes. Wash rapidly, for a few seconds only, in two changes of distilled water. This washing is absolutely necessary, but reduces the intensity of the staining considerably if at all prolonged. It also takes the silver stain out of the fibrous tissue, and must be prolonged to half to one minute when staining the peripheral nerves, in order to differentiate the neuro-fibrils from the surrounding tissue. The resulting decolourisation may be compensated for by leaving the sections in 2 per cent, silver nitrate for a longer time. Transfer to 20 per cent, formol made up with tap water, and leave in this for two to twenty-four hours, changing after the first few minutes. Wash thoroughly in distilled water, and " tone " in a very weak solution of gold chloride (10 drops of i per cent, gold chloride solution to 25 c.c. of distilled water). " Fix " in sodium thiosulphate (5 per cent, watery solution) for about five minutes. Wash in distilled water for at least twenty minutes, dehydrate, and mount in Canada balsam. All glassware used for this method must be perfectly clean and washed out with alcohol, flamed, and finally washed with distilled water before use. The distilled water must be fresh and neutral. Da Fano recommends that it should be doubly distilled over potassium permanganate. The sections should only be handled with glass rods drawn out fine and bent at the tip in the Bunsen burner. If these precautions are observed, it is possible to get uniformly good results with the above method with most forms of nervous tissue. Bielschow sky's Method for Blocks of Tissiie. This method cannot be relied on to show the fine intracellular neuro-fibrils, and it is not possible in using it to differentiate the neuro-fibrils from the neighbouring fibrous tissue. It is, however, of value in examining tissue the consistence of which does not allow of frozen sections, e.g. softenings of the brain or cord, and soft 312 STAINING METHODS tumours, when the condition of the axis cyHnders and dendrites is all that is in question. Fix small pieces of tissue, as thin as possible, in 20 per cent, formol for at least eight days. Wash for a few hours in running water. Place in pure pyridin for twenty-four hours. Wash for twenty-four hours in running water, and then for a few hours in several changes of distilled water. Put into 3 per cent, silver nitrate solution in the dark for four to five days (or for two to three days in the incubator at 37° C). Wash rapidly in distilled water. Place in Bielschowsky's solution, made up as already described, for four to five hours. Wash rapidly in two changes of distilled water, and transfer to 20 per cent, formol for twelve hours. Wash in distilled water, dehydrate, clear, and imbed in paraffin in the usual way. Cut thin sections. Tone, fix, clear, and mount as described for preparations by Cajal's method. D. — Myelin. Weigeri's Method. Fix in formol saline one week. Put pieces up to i cm. in thickness into Muller's fluid for six to eight weeks. The mordanting process may be hastened by putting the bottle containing the pieces in Muller's fluid in the incubator 3-t 37° C. It is necessary to change the fluid as soon as it shows the slightest turbidity. Alternatively the mordanting process may be greatly shortened by using Weigert's primary mordant, in which the pieces should remain for seven to fifteen days according to their thickness. (Weigert's primary mordant: Potassium bichromate 5 grammes. Fluorchrome .. .. .. ..2-5 grammes. Water 100 c.c. Boil the bichromate and add the fluorchrome while boihng, cool, and filter.) Wash in running water. Pass through increasing strengths of alcohol into alcohol and ether. Imbed in celloidin. Cut at 15 to 20 jLi. MYELIN SHEATHS 313 Prepare some clean glass plates, e.g. J-plate or J-plate negatives from which the gelatin has been removed, and coat these thinly with Obregia's solution, which is allowed to dry. Obregia's solution : Syrupus simplex Dextrin syrup Alcohol, 95 per cent. 30 CO. 20 c.c. 20 c.c. The sections are received directly after being cut in methylated spirit, and are transferred from this on to the plates in order. They may be arranged exactly as they will be mounted up, or in serial order. For routine work it is often of advantage to mount up one section from each block in order from the medulla to the sacral end of the cord. After they are carefully arranged, they are blotted firmly, and a very thin film of thin celloidin is poured on to them. This is allowed to dry partially in the air for about five minutes, and the plate is then put in methylated spirit to allow the celloidin film to harden. It is then transferred to a dish of warm water, which dissolves the gummy solution and allows the film of celloidin containing the sections imbedded in it to float off. This film is then treated as a large celloidin section and is stained, differentiated, and mounted up, either whole or after being cut into convenient sizes. (From this stage the films may be stained either by the Weigert or the Kultschitsky-Pal method [v. inf.]). Place in Weigert 's secondary mordant (" gliabeize," p. 317) in the incubator at 37° C. for twenty-four hours. Wash in water. Stain in Weigert 's iron haematoxylin, made by mixing equal volumes of the following solutions A and B. Stain for twenty-four hours. Sol. A. — Haematoxylin .. .. .. i gramme. Absolute alcohol . . . . . . 100 c.c. (Ripen in the sun for two to four weeks.) Sol. B. — Liq. ferri perchlor. . . . . . . 4 c.c. Pure hydrochloric acid . . . . i c.c. Distilled water to 100 c.c. Wash in several changes of water. Differentiate in: Borax . . . . . . . . . . 2 grammes. Pot. ferricyanide .. .. .. .. 2-5 grammes. Distilled water . . . . . . . . 100 c.c. 314 STAINING METHODS Renew the differentiator several times, continuing the differentia- tion until the myehnated fibres stand out dark blue or black against a yellowish background. Wash for twenty-four hours. Dehydrate in increasing strengths of alcohol up to 95 per cent. Clear in carbol-xylol. Wash in xylol. Mount in Canada balsam. KuUschitsky-Pal Method. Mordant, imbed, cut and plate as for Weigert's method. Transfer the films for half an hour to a half-saturated solution of copper acetate in distilled water. Stain in Kultschitsky's haematoxylin in the incubator at 37° C. for twenty-four to forty-eight hours. (Kultschitsky's haematoxylin : Ripened 10 per cent. alcohoHc solution of haema- toxylin . . . . . . . . . . . . 10 c.c. Glacial acetic acid .. .. .. .. .. 2 c.c. Distilled water 90 c.c.) Transfer directly to Muller's fluid for five to fifteen minutes. Differentiate by Pal's method: Transfer the films for quarter to half a minute to 0-25 per cent, potassium permanganate. Wash in tap water. Transfer for a similar time to Pal's solution. Wash again. Transfer again to permanganate, and repeat until the myehnated fibres stand out blue-black against a colourless background. Pal's solution: Oxalic acid . . . . . . . . . . 0-5 grammes. Sod. sulphite 0-5 grammes. Distilled water . . . . . . . . 100 c.c. Wash for several hours in tap water, adding at first a few drops of lithium carbonate solution to give a blue colour to the sections. Dehydrate and mount as in Weigert's method. Lithium carmine or Van Gieson's stain may be used as a counter- stain. Weigert's Method applied to Frozen Sections. This method gives very good and constant results, and has the advantage of greatly reducing the time needed for diagnosis. It also has the advantage that it allows of similar sections being examined by Nissl's, Bielschowsky's, and W^eigert's methods, as well as by Marchi's or other stains for myelin degeneration. Its disadvantages are those common to all frozen sections, e.g. loss of meninges and nerve roots, and difficulty in cutting serial sections. MYELIN DEGENERATION 315 After formalin fixation, cut on the freezing microtome at 20 to 30/^- Put sections into MuUer's fluid for five to seven days in the incubator at 37° C, or into Weigert's primary mordant for two to three days in the incubator. Plate with celloidin on gummed plates, as described for celloidin sections. This step may be taken before mordanting in the chrome solutions if alcohol not stronger than 75 per cent, is used. Stain, differentiate, and mount as for celloidin sections. Marchi*s Method for Degeneration. After fixation in formol-saline, mordant thin pieces of brain and spinal cord, not more than 3 mm. in thickness, in Muller's fluid, for two to three weeks according to the size and thickness of the pieces. The fluid should be changed as soon as it shows any signs of turbidity. Without washing, pass into Marchi's fluid in a well-stoppered bottle, taking care that the pieces do not rest directly on the bottom or on one another. This can be avoided by stringing the pieces on a horsehair. The pieces remain in this fluid for two to four weeks, according to the temperature of the laboratory and the size of the pieces. A few c.c. of i per cent, osmic acid are added once or twice every week, or as soon as the fluid ceases to smell strongly of osmic acid. Marchi's fluid: Muller's fluid . . . . . . . . . . 2 parts. I per cent, osmic acid . . . . . . . . i part. Add just before use i drop of acetic acid for every 20 c.c. of Marchi's fluid. Wash for twelve to twenty-four hours in running water. Dehydrate rapidly and imbed in celloidin or paraflin. Cut at 15 to 30 ju. Transfer to spirit and thence to 95 per cent, alcohol. Clear in carbol-xylol ; mount in chloroform balsam, preferably without a coverslip, which may be easily done by blotting the sections on a slide, covering with a drop of cTiloroform balsam and putting them in the incubator at 37° C. overnight. Another drop of balsam may be added next day if it is thought necessary. Busch's Modification of Marchi's Method. Fix and mordant small pieces up to 5 mm. thick, as in Marchi's method. Wash for twelve to twenty-four hours in running water. 3i6 STAINING METHODS Put in Busch's fluid in a well-stoppered bottle with the same precautions as in Marchi's method. Leave for eight to fourteen days, preferably in the incubator, at 37° C. Busch's fluid: I per cent, osmic acid . . . • • • . . i part. 1-5 per cent, sodium iodate . . . . . . 2 parts. Wash, imbed, cut, and mount as in Marchi's method. In this method the ground substance is less stained than in the original Marchi method, and the osmic acid penetrates better. It is, therefore, more suitable for fairly wide sections, as in studying the tract degenerations in the human brain and brain-stem, and for photographic reproduction. On the other hand, it is perhaps not quite so accurate as the Marchi method, and sometimes black dots are found in sections stained in this way which are not due to fat. These will usually be avoided if the pieces are washed thoroughly between Muller's and Busch's fluids. Marchi's method or its modifications can easily be appHed to frozen sections, which are most conveniently cut after fixation in formahn. In this case the individual steps are the same, but the time required for each can be considerably reduced. E.— Fat. Osmic Acid. This is used in the Marchi method, where the primary mordanting with Muller's fluid prevents the osmic acid from staining the normal myelin. When it is desired to stain all the lipoid substances in nervous tissue, e.g. in peripheral nerves, the tissues may be fixed in formol, and after washing in water cut on the freezing microtome. The sections are then stained for twenty-four hours in i per cent, osmic acid, counterstained in alum carmine or neutral red, if desired, washed for several hours in 80 per cent, alcohol, and mounted in Kaiser's glycerin gelatin. Or they may be mounted in benzene balsam, after being passed rapidly through absolute alcohol and benzene. In this case it is better not to use a coverslip, but to let the balsam dry on the sections. Sudan III and Scharlach R. These stains are particularly valuable for staining sections of the central nervous system. They stain myelin and intracellular lipochromes, but not so deeply as degenerated myelin and other SCHARLACH R. 317 neutral fats. They are, however, of little value for studying the paths of degenerated fibres. After formalin fixation, cut sections as thin as possible on the freezing microtome. Pass into 70 per cent, alcohol for a few minutes. Stain for two or three minutes in a covered vessel in a solution of Scharlach R. made by saturating a mixture of equal parts of 70 per cent, alcohol and of acetone with the dye, filtering or decanting the clear fluid immediately before use. Or stain in a saturated solution of Sudan III in 70 to 80 per cent, alcohol for fifteen to twenty minutes. Wash for a^few seconds in 70 per cent, alcohol, and transfer to water. Counterstain for a minute or two in alum-haematoxylin, dif- ferentiating rapidly in acid alcohol (made up with 70 per cent, alcohol) if necessary. Wash thoroughly in water. A few drops of ammonia or of a saturated solution of lithimn carbonate in the water brings up the blue colour of the haematoxylin more rapidly. Mount in glycerin gelatin. Kaiser's glycerin gelatin : Finest French gelatin ., 40 grammes. Water 210 c.c. Glycerin . . . . . . . . . . 250 c.c. Carbolic acid crystals 5 grammes. Soak the gelatin in the water for two hours. Add the glycerin and carbolic acid and warm for two to fifteen minutes, stirring all the time until the mixture is smooth. Filter through filter paper in the paraffin oven at 50° to 55° C. F. — Neuroglia. Weigert's Method. Small pieces of nervous tissue taken at the autopsy within a few hours of death are placed directly in Weigert's "gliabeize." "Gliabeize": Dissolve 2-5 grammes of fluorchrome (CrFg) in 100 c.c. of distilled water. Add, while boiling, 5 c.c. of 36 per cent, acetic acid and 5 grammes of neutral copper acetate. Cool and add 10 c.c. of formol. The pieces of tissue remain in this for eight days or more. Wash rapidly and imbed in celloidin. Place sections in 0-3 per cent, solution of potassium permanganate for ten minutes. 3i8 STAINING METHODS Wash thoroughly. Place for two to four hours in the following mixture, made by adding 90 c.c. of A to 10 c.c. of B.- Sol. A. — Chromogen 5 grammes. Formic acid, sp. gr. 1-2 . . . . 5 c.c. Distilled water . . . . . . 100 c.c. Sol. B. — Sod. hyposulphite (thiosulphate) 10 grammes. Distilled water . . . . . . 100 c.c. Wash rapidly in water. Place in 5 per cent, solution of chromogen (carefully filtered) for ten to twelve hours. Wash in water. Place sections on a clean slide, dry with filter paper, and stain for half to one minute in the following stain: 70 to 80 per cent, alcohol saturated with methyl violet by heating . . . . . . . . 100 c.c. 5 per cent, aqueous solution of oxalic acid . . 5 c.c. Remove the excess of stain, blot with filter paper, put on a few drops of concentrated Gram's iodine (iodine i, potassium iodide 2, water 100) for about thirty seconds. Dry again with filter paper. Decolourise with a mixture of equal parts of xylol and pure anilin oil. Wash thoroughly in xylol. Mount in Canada balsam. This method sometimes gives remarkably fine results, but is not to be relied on to give good results in all cases. The length of time between the death of the patient and the autopsy seems to play a more than usually important part in the success or otherwise of the method, and sometimes the tissues do not stain well for no ascer- tainable reason. The method is not apphcable to the tissues of the lower animals, for which Cajal's silver method, after fixation in formol-ammonium bromide, is the best (p. 320). It is allowable to fix the tissues first in 10 per cent, formol for not more than twenty-four hours. Lhermitte's Method. Fix for at least fourteen days in 10 per cent, formol. If the brain is fixed in toto in formalin, sections should be made through it on the day after the autopsy to allow of more rapid fixation. It is best at this stage to take out the pieces on which the method is NEUROGLIA STAINS 319 to be performed and fix them in fresh formol, which should be renewed several times. Cut frozen sections, not too thin. The sections after washing are refixed and mordanted in the following solution: Chromic acid, i per cent. . . . . . . • . 50 c.c. Osmic acid, i per cent. . . . . . . . . 12 c.c. Acetic acid, 2 per cent. . . . . . . . . 8 c.c. Distilled water 30 c.c. The sections remain in this for twenty-four to forty-eight hours. Wash rapidly in water and spread the sections, one by one, on glass shdes, which are covered with pieces of cigarette paper. This remains fixed to the slide in water, and the sections remain on it more firmly than on the slide itself. Further, it allows the stain to penetrate better. Stain in 1-5 per cent, solution of Victoria blue in distilled water, heating the slide carefully several times as in Ziehl-Neelsen staining. Pour off the excess of stain and add a few drops of concentrated Gram's solution (iodine i, potassium iodide 2, water 200). Allow to act for one minute. Pour this off and put on directly equal parts of pure anilin oil and xylol. After this has cleared the heavier stain from the section the cigarette paper is eliminated by lifting it off the slide and turning it upside down on to a fresh clean slide, the section thus coming to rest on the slide. Blot carefully and peel the cigarette paper off, leaving the section attached to the slide. Then continue the differentiation with anilin oil-xylol, controlling under the microscope. Wash thoroughly with xylol. Mount in Canada balsam. In this method the sections should be handled with glass rods, especially in passing them into and out of the osmic and chromic acid fixative. Mallory's Method. Fix in Zenker's fluid either primarily or after formalin fixation for twenty-four hours. Wash for twenty-four hours in running water and imbed in celloidin or paraffin. Cut sections, and treat them with iodine solution to remove the mercury left in the tissues. Wash thoroughly in 95 per cent, alcohol to remove the iodine. Wash in water. 320 STAINING METHODS Treat with J per cent, potassium permanganate for five to twenty minutes. Wash in water. Treat with 5 per cent. oxaHc acid five to twenty minutes. Wash thoroughly in several changes of water. Stain in phosphotungstic acid haematoxylin for twelve to twent}^- four hours. Haematoxylin .. .. .. .. .. o-i c.c. Water . . . . . . . . . . . . 80 c.c. 10 per cent, solution phosphotungstic acid . . 20 c.c. Hydrogen peroxide (U.S. Ph.). . .. .. 0-2 c.c. Transfer directly to 95 per cent, alcohol, and dehydrate rapidly with absolute alcohol. Clear in xylol and mount in Canada balsam. This method is apphcable to all nervous tissues fixed in Zenker's fluid, which it is thus possible to stain for nerve cells and by ordinary tissue stains, as well as for neuroglia fibres. To stain the neuroglia cell bodies various methods have been devised. Alzheimer has adopted special methods to give differential staining of the inclusions in these cells, but they offer no advantages over more simple methods for demonstrating the outlines of the cell body and its larger processes. The best of these is Heidenhain's iron alum haematoxylin method. Ramon-y-Cajal' s Silver Method. Fix small pieces of brain or cord in — Formol . . . . ' . . . . . . 70 c.c. Ammon. bromide .. .. .. .. 10 grammes. Water . . . . . . . . . . 430 c.c. The pieces remain in this for four to five weeks. Cut at 15 to 20 /^ on the freezing microtome. Put the sections in formol-ammonium bromide solution in the paraffin oven at 50° to 55° C. for ten to fifteen minutes. Wash rapidly in plenty of water to clear them of formol. Transfer to an ammoniacal silver carbonate solution at 50° to 55° C. until they take on a deep brown colour. This solution is made by adding to 10 c.c. 10 per cent, silver nitrate, 30 c.c. of a 5 per cent, solution of sodium carbonate, dissolving the precipitate in strong ammonia and adding distilled water to 150 c.c. Wash rapidly in distilled water, and transfer to 20 per cent, formol for one minute. Wash again in water, tone in weak gold solution, fix in hyposulphite, wash, dehydrate, clear, and mount as in Bielschowsky's method for frozen sections. APPENDIX I 321 REFERENCES Walker Hall and Herxheimer: Methods of Morbid Histology and Clinical Pathology. Edinburgh, 1905. Mallory and Wright: Pathological Technique. Seventh edition. Phila- delphia, 1918. RoussY AND Lhermitte: Les techniques anatomo-pathologiques du syst^me nerveux. Paris, 1914. Spielmeyer, W.: Technik der Mikroscopischen Untersuchung des N erven- systems. 2 Aiif. Berlin, 1914. 21 APPENDIX II. METHODS OF EXAMINATION OF THE CEREBRO- SPINAL FLUID Cell examination. — This is best done by means of a special counting chamber, the most convenient being that of Fuchs-Rosenthal. The fluid is stained either with a i per cent, solution of toluidin blue, or, if it is desired to dissolve up the red blood corpuscles, with a saturated solution of methyl violet in lo per cent, acetic acid. This stain is sucked up to the mark i of a Thoma leucocyte pipette, and cerebro-spinal fluid sucked up after it to the mark ii. A better method when a number of fluids are to be examined is to put 0-45 c.c. of cerebro-spinal fluid in a small clean test-tube, and to add 0*05 c.c. of the stain. In either case the stain is allowed to act for ten to fifteen minutes before the counting chamber is fiUed. The depth of the Fuchs-Rosenthal chamber is y% mm., and the diameter of the squared area 4 mm. The fluid over this area is, therefore, -V- c.mm. As the fluid is diluted 9 in 10, the total number of cells counted within the outer lines of the square re- presents the number in V XyV = "Vir c.mm. of cerebro-spinal fluid. For practical purposes dividing the total number of cells counted by 3 gives the number in i c.mm. of the fluid. It is also possible to use a Thoma counting chamber by counting fields after the method of Strong. The field is adjusted by altering the tube length until its diameter corresponds exactly with 10 of the small squares of the ruling. Each field will then be almost exactly \ sq. mm. in area. As the cell is ^^ mm. deep, by counting 50 fields we get the number of cells in i c.mm. of diluted fluid. To get a more exact picture of the individual cells Quincke's original technique may be employed: 5 to 10 c.c. of fluid are centri- fugalised; the fluid is poured out, and the cells adhering to the bottom of the tube are scraped off with a capillary glass pipette up which the fluid remaining in the tube runs. This is blown on to a clean albuminised slide, dried, fixed in alcohol, and stained by Giemsa's or Pappenheim's stains, or with toluidin blue. 322 EXAMINATION OF CEREBRO-SPINAL FLUID 323 Alzheimer's method may be used, but is rather cumbersome. He mixes the cerebro-spinal fluid with four times its volimie of absolute alcohol. The resulting precipitate, which contains the albumen and cells, is centrifugalised ; the supernatant fluid is poured off, absolute alcohol added, and the tube centrifugalised again. The resulting deposit, which is now a firm mass, is imbedded in celloidin, cut, and stained by any methods applicable to celloidin sections. Methods of albumen examination. —Where large quantities of fluid are available, it is possible to use Esbach's, Sicard's, or Aufrecht's sedimentation tubes. These, however, are not exact for small percentages of albumen, such as are normally found. Much better is Mestrezat's method of comparing the precipitate given by boiling with trichloracetic acid, with a standard scale of tubes containing known quantities of albumen. This is made by measuring exactly the amount of albumen, either in a highly albuminous cerebro-spinal fluid or in a dilution of blood serum. A number of narrow test- tubes of equal calibre are then taken, and into each is put 2 c.c. of albuminous fluid in strengths varying from o-i per cent, to o-oi per cent. : o -3 c.c. of 30 per cent, trichloracetic acid is added to each tube, and these are boiled and then closed off in the blowpipe and sterilised at 56° C. For examination 2 c.c. of the cerebro-spinal fluid to be tested are put into a test-tube of similar calibre, and trichloracetic acid added as before. The tube is boiled and left for half an hour to allow the precipitate to become flocculent, and is then compared with the " standard scale." This can be done with great exactness by noting which of Jaeger's test types can be read through the fluid in the tubes. Globulin estimation — (i) Nonne-Apelt reaction (phase !).• — Add to I c.c. of cerebro-spinal fluid i c.c. of saturated ammonium sulphate solution. Shake the tube. The appearance of a definite faint opalescence, within three minutes after mixture of the fluids, is considered a weak positive reaction. A milky turbidity constitutes a strong positive reaction. (2) Noguchi {butyric acid) reaction. — To 0-2 c.c. of cerebro-spinal fluid in a test-tube is added 0-5 c.c. of 10 per cent, butyric acid in normal saline. The mixture is boiled, o -i c.c. of normal caustic soda added, and the mixture boiled again for a few seconds. The appearance of a granular or flocculent deposit which commences to settle into a peUicle at the bottom of the tube within three hours constitutes a positive reaction. Glucose estimation. — For rough practical work the reduction of Fehling's solution may be used. Normally i c.c. of cerebro-spinal 324 EXAMINATION OF CEREBRO-SPINAL FLUID fluid reduces almost completely 0-25 c.c. of mixed Fehling's solution, giving a heavy red precipitate with a faint blue colour in the super- natant fluid. For quantitative estimation the only satisfactory methods are those devised for blood analysis. Of these probably the best is that of McLean (v. Cole's Practical Physiological Chemistry) . Chlorides. — These are estimated by titration with standard silver nitrate, using potassium chromate as an indicator. The silver nitrate solution used contains 5-814 grammes AgNOg to the litre; 2 c.c. of cerebro-spinal fluid are put into about 10 c.c. of distilled water, a few drops of potassium chromate solution added, and a pinch of pure calcium carbonate to insure alkalinity. Silver nitrate is run in out of a burette until the lemon yellow colour changes to orange. Each c.c. of silver solution used will then indicate i part per thousand of chlorides in the cerebro-spinal fluid. Urea. — This may be estimated by any of the hypobromite methods in common use. If possible, 5 c.c. of cerebro-spinal fluid should be used. Or Kennaway's soy-bean method may be used {v. Brit. Journ. of Exper. Path., vol. i. (1920), p. 135). Organic acids. — These may be roughly estimated by Kopetsky's method. A standardised Uffelmann's reagent is prepared by. taking — ■ 5 per cent, ferric chloride . . . . . . i part. I per cent, carbolic acid . . . . . . 5 parts. Mix. Take 6 drops. Cerebro-spinal fluid is added drop by drop to this until the purple colour becomes yellow. If this appears on i to 3 drops it indicates great excess of organic acids; if on 3 to 6 drops it indicates moderate excess of organic acids; if on 6 to 10 drops it indicates slight excess of organic acids. Reaction of Boveri. — One c.c. of cerebro-spinal fluid to be tested is placed in a small test-tube. Allow i c.c. of o-i per cent, solution of potassium permanganate to run down the side of the tube on to the top of the fluid, taking care that the fluids do not mix. A brownish ring forms between the two fluids in certain pathological conditions of the cerebro-spinal fluid. On mixing the fluids the colour of the permanganate may be changed slightly or completely to brown. A positive reaction is said to indicate inflammatory disease in the cord. Lange's colloidal gold reaction. — All the water used to make up solutions for this test must be freshly distilled twice in good glass, using cork instead of rubber for the connections. The glass-ware LANGE'S GOLD-SOL REACTION 325 used should be absolutely clean. It should be washed with 50 per cent, hydrochloric acid, washed out first with tap and then with freshly distilled water, and sterilised in the hot-air oven for half an hour. A litre of fresh double-distilled water is heated in a Jena glass flask to 60° C. Then add i c.c. of 10 per cent, gold chloride and 10 c.c. of 2 per cent, potassium carbonate solution. Shake well, and heat rapidly to 90° to 95^ C, not higher. Then take the flask away from the flame, and add drop by drop with constant shaking 10 c.c. of I per cent, formalin solution. Stop as soon as the solution mGR^M OF LANGE'S COLLOIDAL GOLD REACTION. (After Miller, Brush, Hammers and Felton.) - - - - = MENINGEAL FIELD = PARETIC FIELD ++^.+^.+ = LUETIC FIELD DILUTIONS 5 1 <55 1 1 <5a CM i 5 COLOURLESS 4 PALE OR GREYBLUE ■ rr- — i J BLUE ^+4 ^\ / / \ \ 2 LILAC OR PURPLE *. % *5 / / f \ \ \ 1 RED BLUE * X X / / / t k \ ', \ • > N RED X + 4--t- • + ■». + + +v lv>" ;v ~ — Fig. 103. begins to turn red. Usually the red colour develops rapidly once it has commenced. If not, a few drops more of formaHn may be added. When cool, place in several clean, well-stoppered flasks or bottles, and store in a dark place. The gold solution thus prepared should be bright cherry red in colour, with no fluorescence when viewed by reflected light, and perfectly clear to transmitted light. If there is any tendency to purple or " old rose " shades, or more than the slightest degree of fluorescence, it should be discarded. The final test of the gold solution is that it gives a curve of the " paretic type," with a known sample of fluid from a case of general paralysis. 326 LANGE'S GOLD-SOL REACTION For each test, ten clean test-tubes are set out in a rack, and another one or two tubes are set out for a control. Into the first tube of each test 0-9 c.c. of 0-4 per cent, saline is put, and 0-5 c.c. into each of the other tubes of the test and the controls. With a clean dry pipette o-i c.c. of the cerebro-spinal fluid to be tested is put into the first tube of the test. Mix and take up 0'5 c.c, which is put into the second tube. Mix and take up 0-5 c.c. and transfer it to the third tube, and so on, until the tenth tube of the test is reached. From this take out 0-5 c.c. of fluid and throw it away. Then to every tube including the controls add 2-5 c.c. of the colloidal gold solution. Shake and leave at the temperature of the laboratory for twelve to twenty-four hours. Then read each tube from the i in 10 dilution in order up to the i in 5,120 dilution, giving each degree of alteration in colour a number from o (no change) up to 5 (complete loss of colour in the tube except for a bluish precipitate) . The curve of progressive general paral37sis will thus read some- thing like 5.5.5.5.5.4.3.2.1.0 (fig. 103). INDEX Numbers in heavy type refer to illustrations. Abiotrophy, i Abscess of the brain, 24, 85, 163, 179, 181, 183 — of the spinal cord, 184 — • after wounds of the brain 85 Acetone in cerebro-spinal fluid, 37 Achromatosis, 6 Acidosis, Effects on vasomotor centre, 81 Acoustic nerve tumours, 247, 243, 249 Acromegaly, 256 — and syringomyelia, 290 Actinomycosis, 163 Acute leptomeningitis, 171 Cerebro-spinal fluid in, 39 Acute myelitis, 184 Cerebro-spinal fluid below level ot 45 Acute poliomyelitis, 193, 195 Addison's disease causing disease in spinal cord, 24 Adventitial lymph spaces, 27, 34 Agenesis, i Albumen examination in cerebro- spinal fluid, 323 Albumoses, Bence- Jones', 266 — in cerebro-spinal fluid, 44 Alcohol in cerebro-spinal fluid, 37 — in relation to tabes dorsalis, 147 Alcoholic insanity, 227 — neuritis, 223 Alzheimer's method of cell examina- ' tion in cerebro-spinal fluid, 323 Amaurotic family idiocy, 58, 60 i Amnesia in concussion of the brain, 82 I Amyotonia congenita, 78 j Amyotrophic lateral sclerosis, 271, 1 273, 275 Anaemia in subacute combined de- generation, 277, 282 I Anencephaly, 50 Aneurysm of cerebral arteries, 120, 259 Angeioma, 249 Aniline poisoning causing neuritis, 223 Anterior horns. See Ventral horns — roots. See Ventral roots Anthrax meningitis, 1 75 Aortitis, Syphilitic, in tabes dorsalis, 153 Arachnoid cysts, 259 ■ Cerebro-spinal fluid below, 44 Argyll- Robertson pupil, 152 — in hypertrophic interstitial neuritis, 72 Arsenical neuritis, 223 Arteritis syphilitica, 138 Ascending neuritis after injury, 97 Asphalter's drop-foot, 102 Atrophic sclerosis, 52 Atrophy of brain. Localised, 52 Auditory nerve. Injuries of, 94 Tumours of. 247, 248, 249 Aufrecht's albumino meter, 323 Axis cylinders in cord. Changes due to concussion, 89 Bandelette of Pierret in tabes dor- salis, 150 Basilar artery. Atheroma of, 107 Bell's paralysis, 18, 100 Bence- Jones' albumoses, 266 Beri-beri, 234 Betz cells in chronic lead poisoning, 228 after lesions of the pyramidal tract, 10 in motor neuron disease, 274 Bielschowsky's method for neuro- fibrils 4, 49, 225, 305, 310 Birth injuries, 58 Bisulphide of carbon neuritis, 223 Blood stream, Infection by, 24 Blood vessels in central nervous sys- tem. Structure of, 27 Bosses on skull in endothelioma, 253 Boveri reaction, 324 327 328 PATHOLOGY OF THE NERVOUS SYSTEM Brachial plexus, Injuries to, loo Brain, Abscess of, 24, 85, 163, 179, 181, 183 — Injuries to, 81 Bronchiectasis causing abscess of the brain, 24, 179 Brown- Sequard paralysis in syphilitic myelitis, 145 Bulbar nuclei in bulbar palsy, 273 — palsy, 271 Busch's modij&cation of March i method, 315 Caisson disease, 93 Cajal's method for neuro-fibrils, 305, 309 for neuroglia, 320 Cancer of spine, 264 • and herpes zoster, 213 — and subacute combined degenera- tion, 282 Carbol- xylol, 307 Carbon-monoxide poisoning causing neuritis, 223 affecting the brain, 228 Carcinoma of brain, 249, 260 • — of vertebrae, 264 Carotid artery. Changes following ligature of, 13 Cauda equina, Neuro-fibromatosis of, 268 Cavities in cord due to pachy- meningitis, 137 Cell changes after lesions of axis cylinders, 6 Celloidin method for imbedding nervous tissues, 306 Cells in cerebro-spinal fluid, 38 Examination of, 322 Central canal of cord, 18 — Development of, 291 Cerebello-pontine angle. Tumours in, 247 Cerebral abscess, 24, 85, 163, 179 due to streptothrix, 163 — asymmetry, 52, 54 — embohsm, 105, 106, 112 and chorea, 215 — haemorrhage, 117, 119 — syphilis, 133, 135 Cerebro-spinal fluid in, 41 Cerebro-spinal fluid, 30 in acute poliomyelitis, 202 Amount of, 33 Cells in, 38 Changes in appearance, 37 — — Circulation of, 31 Examination of, 322 — — Function of, 35 in lethargic encephalitis, 203 Mode of absorption, 33 Cerebro-spinal fluid. Mode of forma- tion. 30 Normal composition of, 35 — — Pathological alterations in, 36 in Pott's disease, 165 Pressure of, ^^ in trypanosomiasis, 217 Cervical form of tabes dorsalis, 146 — ribs, loi Charcot joints, 153 Charcot-Marie Tooth paralysis. 70 Chlorides in cerebro-spinal fluid, 43 — ■ — — Examination of, 324 Chlorosis, Sinus thrombosis in, 124 Cholesteatoma, 256 Chorea, 214 — Progressive familial (Hunting- ton's), 68 Choroid plexus, 30 — ■ — Failure of, 42 Tumours of, 249 Chromatolysis, 6, 12 — Time relations of, 6, 7, 9 Chromophobe struma of pituitary, 256 Circulatory disturbances of brain, 103 Effect of, on nerve cell, 1 3 Circumflex nerve injuries, 10 1 Cirrhosis of liver in progressive lenticular degeneration, 68 Cisterns at base of brain. Function of. 82 Clarke's column, 10, 199, 210, 232, 282 Clawhand in peroneal atrophy, 70 Clubfoot in peroneal atrophy, 70 Colloid formation by pituitary gland, 256, 259 Colloidal gold reaction in cerebro- spinal fluid, 47, 324, 325 Comma tract in tabes dorsalis, 149 Compound granular corpuscles, 28, 89, no, 143, 183 Compression of spinal cord, 269 Concussion of brain, 82 — of the spinal cord, 88 Contrecoup effect in concussion of the brain, 8^ Convolutions, Foetal type of, 52, 53 — Parchment-like, 52 Corpora amylacea, 23, 24 in subacute combined degenera- tion, 282 Corps granuleux. See Compound granular corpuscles Corpus callosum. Degeneration in motor neuron disease. 274 — striatum. Lesions of, in paralysis agitans, 298 Cranial nerves. Injuries to, 100 Lesions of, in disseminated sclerosis, 288 in syphilitic meningitis, 136 INDEX 329 Cranial nerve palsy in cerebral tumour, 262 Cranio-cleido-dysostosis, 52 Cranium, Tuberculosis of, 164 Crutch paralysis, loi Cyclencephaly, 51 Cylindrical cavities in the spinal cord, 88 Cylindroma, 252, 263 Cysticerci, 259 Cysts of brain, 258 — Dermoid, 256, 268 — Parasitic, 259 Da Fano's modification of Biels- chowsky's staining method, 310 Dejerine-Sottas' disease, 71 Dementia in Huntington's chorea, 69 Dendrites, Swellings of, in amaurotic family idiocy, 59 Dermoid cysts as cerebral tumours, 256, 268 Developmental disease, 50 Development of canal of spinal cord, 291 Diabetes, Cerebro-spinal fluid in, 37 Diffuse infarction of brain, 86 Dinitro-benzol neuritis, 223 Diphtheritic neuritis, 2, 223 Diplococci in Landry's paralysis, 209 Diplococcus rheumaticus, 214 Dislocation of vertebrae, 91 Disseminated sclerosis, 283, 286, 287 — ; — Cerebro-spinal fluid in, 39 Dissociative anaesthesia in haemato- myelia, 127 Disuse of nerve cells, 2 Dorsal root degeneration in cerebral tumour, 263 in herpes zoster, 213 in tabes dorsalis, 147 Dorsal root ganglia in beri-beri, 236 in herpes zoster, 213 in leprosy, 162 in tabes dorsalis, 152 Double hemiplegia. Chronic progres- sive, 105 Echinococcus cysts, 259 Electrical changes in muscle in family periodic paralysis, 73 — — in myasthenia gravis, 300 in myotonia, 74 Embolism of cerebral arteries, 105, 106, 112 Encephalitis lethargica, 202, 204-206, 208 — Suppurative, 179, 181 — Tubercular, 169 — in wounds of the brain, S$ Encephalopathy, 227 End-bulbs on injured nerves, 15, 17, 97 Endocarditis and chorea, 214 Endothelioma of the brain, 253, 254, 266 j — of the spinal meninges, 267 I Ependyma, Origin of, 18 ; Ependymal glioma, 246 I Epilepsy (Jacksonian) after injuries ^ to the brain, ?>t, ! Epileptic convulsions from cerebral ' anaemia, 104 Erb-Duchenne paralysis, 100 Erb's syphilitic paraplegia, 145 — type of myopathy, 77 Ergotism, 229 Exencephaly, 50 Exogenous fibres of spinal cord in i tabes dorsalis, 149 j External popliteal nerve injuries, loi I Extradural haemorrhage, 123 ; — spinal tumours, 266 I Facial herpes. Causation of, 26 ; — paralysis in tetanus, 192 False porencephaly, 121 : Falx cerebri. Function of, 82 i Endothelioma of, 264 . Famihal cerebral degeneration with macular changes, 61 — hypertrophic neuritis, 71 — periodic paralysis, 73 Fat granule cells. See Compound granular corpuscles I — Stains for, 316 ' Fibrin in cerebro-spinal fluid, 37, 40, 41, 44 ! Fibro-glioma, 248 ■ Fixation of nervous tissues, 305 ' Fluorchrome as mordant, 312, 317 Foetal type of convolutions, 52 Foramina of Magendie and Luschka, 31 Foreign bodies in the brain, 86, 180 Formaldehyde in the cerebro-spinal fluid, 37 Formalin-saline fixation, 306 " Four reactions " in the cerebro- spinal fluid, 46 Fracture of the skull, 84, 100 — of the vertebrae, 91, 92 Fxiedreich's ataxia, 63, 66 Froin, Syndrome of, 38, 43, 45 Fuchs- Rosenthal counting chamber, 322 Ganglion cells. See Nerve cells Ganglio -neuroma, 246, 247 Gas gangrene of the brain, 85 Gasserian ganglion in herpes zoster, 213 330 PATHOLOGY OF THE NERVOUS SYSTEM Gasserian ganglion in leprosy, i6i General paralysis of the insane, ^5, i54> 166, 157 Cerebro-spinal fluid in, 39, 46, 47 Gennari, Fibres of, 59 Gitterzell. See Compound granular corpuscles Glia. See Neuroglia " Gliabeize," 317 Glial cells. Fibre-forming, 56 Glioma, 240, 241-245 Gliomatosis and syringomyelia, 292 Gliosis in Friedreich's ataxia, 64 — in progressive lenticular degenera- tion, 67 Globulin in cerebro-spinal fluid, 41, 323 Glucose in cerebro-spinal fluid, 36, 42, 323 Glycerin gelatin, 317 , Gold-Sol reaction of Lange, 47, 324, 325 Granular cells. See Compound granular corpuscles Gummata, 133, 240 — of spinal cord, 267 Gummatous arteritis, 138, 139, 140 — meningitis, 134, 135 Gutter fracture of skull, 84 Haematomyelia, 125 Haematorrhachis, 127 Haemorrhage, Extradural, 123 — into areas of softening, 83 — into brain, 117, 119 — into dorsal root ganglia, 213 — into ventricles, 118 — Subarachnoid, 82, 122 — Subdural, 122 Hemisection of spinal cord, 91 Herpes zoster, 212 — • — Cerebro-spinal fluid in, 39 His, Perivascular space of, 34 ■ in lethargic encephalitis, 207 Huntington's chorea, 68 -Hyaline degeneration of muscle in myopathy, 77 Hydatid cysts of the brain, 259 of the spine, 266 Hydrocele of the fourth ventricle, 58 Hydrocephalus, Causation of, 47 — from cerebral tumour, 260, 261 — after concussion of the brain, 83 — External, 52, 57 — Internal, 57 — in acute meningitis, 1 72 — in syphilitic meningitis, 136 — in tubercular meningitis, 168 Hydromyelus, 57, 58 Hydrophobia, 217 Hyperactivity of cells, 2 Hyperpyrexia, Effect on nerve cells, 13 Hypertrophic interstitial neuritis of children, 71 — ■ tuberous sclerosis, 56 Hypopituitarism in cerebral tumour, [ 262 Icterus neonatorum. Relation to pro- j gressive lenticular degeneration, 68 ! Idiocy, Amaurotic family, 58 ! Infarction of brain, 86, no in lethargic encephalitis, 203 Infection by blood stream, 24 I Inflammatory reaction in central nervous system, 27 I Injuries to the brain, 81 I — to nerves, 94. 95, 98, 98, 99 — to the spinal cord, 87 — to spine and herpes zoster, 213 Intermedio-lateral columns, 11 Interstitial neuritis, 71, 224 Intoxication of nerve cells, 2 Intramedullary haemorrhage, 125 i — tumours of spinal cord, 267 i Intraventricular haemorrhage, 38 I Iodides in cerebro-spinal fluid, 31 Ischaemia, 2 I Ischaemic softening of brain, 105 I Jacksonian fits after concussion of j the brain, 83 I Jaundice in progressive lenticular ' degeneration, 67 Kaiser's glycerin gelatin, 317 Klumpke type of paralysis, 100 Kopetsky's method of organic acid estimation, 324 Kornchenzell. Se^ Compound granu- lar corpuscles Kultschitsky-Pal method for myelin, 314 Kypho-scoliosis in hypertrophic fami- lial neuritis, 72 Lacunar softening of the brain, 114 Landouzy-Dejerine type of myo- pathy, 75 Landry's paralysis, 209, 211 Lange's gold reaction, 47, 324, 325 Lathyrism, 25, 232, 233 Lead neuritis, 223 — encephalopathy, 228 Lenticular degeneration. Progressive, familial, 66 Lenticulo-striate arteries in cerebral haemorrhage, 117 Leprosy, 160, 162 INDEX 331 Leptomeningitis, Acute, 171 — Syphilitic, 136, 143 — Tubercular, 167 Lethargic encephaUtis, 202, 204-206, 208 Cerebro-spinal fluid in, 39 Lhermitte's neuroglia method, 318 Ligature of vessels, Effect on nerve cells, 13 Lipochrome granules in nerve cells, 6, 14 in amaurotic family idiocy, 61 Lipomata of spinal canal, 266 Liver cirrhosis in progressive len- ticular degeneration, 68 Locomotor ataxia, 145 Loculation syndrome in cerebro- spinal fluid, 38, 43, 45 Longitudinal sinus thrombosis, 124, 177 Luschka, Foramina of, 31 Lymphangitis of syphilis, 131 Lymphatics, Infection of nervous system through, 26 — Perivascular, 27, 34 Lymphatic leukaemia and subacute combined degeneration, 282 Lymphorrhages in myasthenia gravis, 301, 302 Magendie, foramen of, 31 Mallory's neurogha method, 319 Marchi method, 14, 16, 375 Mastoiditis causing abscess of brain, 180 " Median triangle " in tabes dorsalis 149 Melanin pigment, 14 Meningeal haemorrhage, 121 Meningism; 176 Meningitis . See Leptomeningitis and Pachymeningitis Meningitis from wounds of the brain, 35 of nose, 86 — Gummatous, 135 — Post-basic, 173 — and syringomyelia, 293 Meningocele, 50 Meningococcal meningitis, 1 73 Meningo-myelitis, Syphilitic, 141 — Tubercular, 171 Mestrezat's method of albumen esti- mation, 323 Microcephaly, 52, 54 Microgyria, 52 Middle root zone in tabes dorsalis, 150 Monakow's bundle in motor neuron disease, 274 Motor neuron disease, 271 I Multiple sclerosis. 5ee Disseminated sclerosis Muscles in amyotrophic lateral sclerosis, 275, 276 i — in myasthenia gravis, 302 : — in myopathy, y^, 77 Myasthenia gravis, 299 I Myatonia congenita. 5ee Amyotonia » congenita Myelin, Degeneration of, 16 — destruction, 29 — Regeneration of, in peripheral ! nerves, 17 ' — • Stains for, 312 Myelitis, Acute, 184 — ex neuritide, 223 — Infective, 184 — ■ from lymphatic infection, 26 — ■ Septic, in wounds of the cord, 91 — Suppurative, 184 i — Syphilitic, 141 I — Toxic, 226 i — Transverse, 186 Myeloma of spine, 265 Myopathy, 75 Myotonia, 74 — atrophica, 75 Myxo-endothelioma, 252 ; Myxo-fibroma, 248 I Negri bodies, 218 I Negro lethargy, 216 j Nephritis, Cerebro-spinal fluid in, 37 Nerve cell, 4, 5, 7, 8 Changes after amputation, 9 — — Changes due to circulatory dis- orders, 13 Changes in exhaustion, 12 Changes in hyperpyrexia, 13 Changes in intoxication, 12 Coagulative necrosis of, 13 Condition during activity, 12 Condition during rest, 12 Parapyknomorphic state, 12 Perinuclear chromatolysis, 12 Pigmentary changes, 14 Pyknomorphic state, 12 — fibre, 14, 15, 16 — grafts, 1 7 Nerve sheaths, in relation to cerebro- spinal fluid, 31 Nerves, Injuries of. 94. 95, 96, 98-99 — Regeneration after injury, 97 Neuralgia, 224 Neural lymphatics in tetanus, 189, 191 Neurinoma, 248 Neuritis, 222 — Ascending after injury, 97 — in beri-beri, 236 — in leprosy, 160 332 PATHOLOGY OF THE NERVOUS SYSTEM Neuritis, Progressive hypertrophic interstitial form of, 71 Neuroblastoma, 246 Neuro-epithelioma gliomatosum, 246 Neuro-fibrils, 4, ii — Outgrowth of, in regeneration of nerves, 15 — Reaction of intracellular, to lesions of axis cylinders, 1 1 — Stains for, 309 Neuro- fibroma. Intracranial, 247, 248-249 Neuro-fibromata of spinal roots, 266, 268 Neuro-fibromatosis of cauda equina, 268 Neuroglia, Function of, 19 — Origin of, 18 — Overgrowth of, 2, 20, 22 — Powers of defence of, 28 — Reaction of, 23 — Stains for, 317 Neuroglial cells, Amoeboid forms, 20 " Fibre-forming," 21 Forms of, 1 8 , Miniature, 20 Neurokeratin network in beri-beri, 236 Neurolemma sheath. See Sheath of Schwann Neuronophagy, 12, 199, 205, 213, j 220, 298 1 Neuron theory, 2 Nissl granules, 4, 5, 6 Stains for, 308 Nitrates in cerebro-spinal fluid, 31 Noguchi reaction in cerebro-spinal fluid, 323 Nonne-Apelt reaction in cerebro- spinal fluid, 323 Obersteiner, Pericellular space of, 34 Obregia's solution, 313 Obstetrical injuries of spinal cord, 126 Odontoid process. Fracture of, 90 Oedema of the brain after wounds of the head, 85 in lead encephalopathy, 228 — of the spinal cord after injury, 89 Olfactory nerves, in relation to cerebro-spinal fluid, 31 Injuries to, 94, 100 Optic atrophy in amaurotic family idiocy, 61 in disseminated sclerosis, 290 in general paralysis, 156 in tabes dorsalis, 152 — chiasma, Pressure of tumours on, 262 • — nerves. Injuries to, 100 — neuritis in cerebral tumour, 262 Optic neuritis in disseminated sclero- sis, 290 Organic acids in cerebro-spinal fluid, 43 Osteo-porosis in cerebral tumour, 262 Otitis media, Pyogenic, 177 Tubercular, I64 Oval field of Flechsig in tabes dorsalis, 149 Pachymeningitis cervicalis hyper- trophica, 136 — haemorrhagica interna, 123 -in alcoholic insanity, 228 — Pyogenic, 177 — Tubercular, 165 " Palisade " appearance in neuro- fibroma, 248, 249 Pallidal system of cells, 298 Papilloedema, 263 Paralysis agitans, 298 Parasitic cysts, 259 : Parathyroids in myasthenia gravis, 300 Parenchymatous neuritis, 223 Parkinson's disease, 298 Paths of infection in the central nervous system, 24 Pellagra, 230 — Cord changes in, 25 Periodic paralysis, Familial, 73 Peripheral chromatolysis, 13 — nerves, Rapid regain of function of, 18 — neuritis, 224 Perithelioma, 251 Perivascular infiltration, 27, 131, 169, 175, 183, 187, 196, 204, 207, 217, 220, 269, 289 — lymphangitis in syphilis, 131, 142, — lymphatic space, 27, 34 — sarcoma, 261 Pernicious anaemia, causing disease of spinal cord, 24 Peroneal atrophy, 70, 71 Pes cavus in Friedreich's ataxia, 66 Pes equinovarus in hypertrophic interstitial neuritis, 72 Pes equinus in peroneal atrophy, 70 Phenol-xylol, 307 Pineal tumours, 258 Pituitary body in myasthenia gravis, — symptoms in cerebral tumour, 262 — tumours, 256, 267, 269 Plasma cells, 28, 39, 133, 135. 15^. 168, 175, 183, 187, 198, 207 Poliomyelitis, Acute, 193 Cerebro-spinal fluid in, 37, 39, 41 INDEX 333 Polyneuritis, 224 — of beri-beii. 236 — Cerebro-spinal fluid in, 41 — gallinarum, 235 Pons, Hypertrophy of , 240, 243. 245 Pontine thrombosis, 107, 108, il6 Porencephaly, False, 56, 121 — True, 51 Post-basic meningitis, 173, 174 Cerebro-spinal fluid in, 37 Posterior horns. See Dorsal horns — roots. See Dorsal roots — longitudinal bundle in motor neuron disease, 274 Pott's disease, 90, 164, 166 Cerebro-spinal fluid in, 44 and herpes zoster, 213 Prenatal diseases, 52 Pressure cone, 48, 261, 262 — on nerves, 100 Progressive dementia from cerebral anaemia, 104 — lenticular degeneration, 66 — muscular atrophy, 271 — spinal muscular atrophy of in- fants, 62 Psammoma of cerebral meninges, 247, 256 — of spinafl meninges, 267 Pseudo-hypertrophic form of myo- pathy, 75, 77 Pseudo-pofencephaly, 56, 121 Pyaemia causing abscess of brain, 179 Pyknomorphic state of nerve cell, 7 Pyogenic pachymeningitis, 177 Quincke's method of examining cells in cerebro-spinal fluid, 322 Rabies, 217 Radiculitis of dorsal roots in tabes dorsalis, 147 Ramon-y-Cajal. See Cajal Red nucleus. Changes in, after lesions of Monakow's bundle, 10 Regeneration of nerves after injury, 97 Reich's ir granules in pellagra, 232 Reparation of nerve cells, 7 Retinal changes in amaurotic family idiocy, 59, 61 Rosettes in gliomata, 246 Rosin's method for Nissl granules, 309 Sarcoma of brain, 250 — spinal meninges, 266, 267 — of vertebrae, 265 Scharlach R. stain for fat, 14, 316 Schwann, Sheath of, in injuries to nerves, 15, 16 _ — , — in neuritis, 72, 225 Schwann, Sheath of, in neuro-fibro- mata, 248 Sclerosis, Hypertrophic tuberous, 56 Scoliosis in Friedreich's ataxia, 66 Septicaemia, causing cerebral ab- I scess, 24 I Septomarginal tract in tabes dorsalis, i 149 I Serous meningitis, 176 ' Sheath of Schwann. See Schwann, Sheath of Sicard's albumen tube, 323 Sinus thrombosis, Primary, 124 Secondary, 125, 177 in wounds of the brain, 86 i Sixth nerve palsy in cerebral tumour, I 262 j Skull in general paralysis, 155 I — in endotheliomata of brain, 253 I Sleeping sickness, 216 Softenings of brain, 108, 109, 112, 113, 115 Spider cells, 20, 21, 23, 158, 245 Spina bifida, 51 Spinal compression. Pathology of, 269 Spinal cord. Concussion of, 88 Injuries of, 87 Suspension of the, 87 Wounds of, 88, 90 — ganglia. See Dorsal root ganglia — tumour, 264 — tumour, Cerebro-spinal fluid in, 44 Spirochaeta pallida, 130 in general paralysis, 154 Spirochaetes in disseminated sclerosis, 284 Staining methods, 305 Streptothrix infection, 163 of wounds of the brain, 86 Subacute combined degeneration of spinal cord, 24, 277, 279, 281 • and pellagra, 231 Subarachnoid haemorrhage, 38, 82, 122 — space, 32 Subdural haemorrhage, 91, 122 — space, -^T) Sudan III stain for fat, 316 Superior longitudinal sinus. Wounds of, 86 Rupture of, 122 Thrombosis of, 177 Sympathetic ganglia in leprosy, 161 — nerve cells, Changes in, after lesion of sympathetic chain, 1 1 Syndrome of Froin, 38, 43 Syphilis, 130 — and neuritis, 223 SyphiUtic aortitis in tabes dorsalis, ^5Z 334 PATHOLOGY OF THE NERVOUS SYSTEM Syphilitic leptomeningitis, 131, 136 — meningitis, Cerebro-spinal fluid in, and syringomyelia, 293 — myelitis, 141 — pachymeningitis, 134, 136 Syringobulbia, 295 Syringomyelia, 2, 290, 294 — due to pachymeningitis, 137 Tabes dorsalis, 145, 151, 157 Cerebro-spinal fluid in, 39, 46 Tay- Sachs disease, 58, 60 Tentorium cerebelli. Action in in- juries to brain, 82, 83 Tetanus, 26, 188 Thomsen's disease, 74 Thrombosis of cerebral arteries, 105, 107, 108, 109, 112 ■ — sinuses, 124, 177 vessels in lethargic en- cephalitis, 203, 208 Thymus gland in myasthenia gravis, 300, 303 Thyroid gland in myasthenia gravis, 303 and pituitary tumour, 256 Torcular HerophiH, Pressure in, ^^ Tract degeneration from cerebral softening, 114 Transverse myelitis, 186 Trauma of nervous system, 81 — and syringomyelia, 292 — and tabes dorsalis, 146 Trigeminal neuralgia, 26, 224 Trjrpanosomiasis, 216 Tubercular arteritis, 169 — leptomeningitis, 167 Cerebro-spinal fluid in, 37 — meningo-myelitis, 171 • — pachymeningitis, 165 Tuberculomata of brain, 169, 170, 240 — of spinal cord, 170, 267 Tuberculosis, 163 Tuberous sclerosis. Hypertrophic, 56 Tumours of brain, 239 General pathology of, 260 — of the pineal body, 258 — of spine and spinal cord, 264 — and syringomyelia, 297 — of vertebrae, 90, 264 Typhoid myelitis, 185 Ulcerative endocarditis causing cere- bral abscess, 24 Ulnar nerve injuries, loi Uraemia, Cerebro-spinal fluid in, 41, 43 Urea in cerebro-spinal fluid, 43, 324 Vagus neuritis in beri-beri, 236 van Gieson's stain, 307 Vasomotor mechanism of medulla, 81 Venous sinuses. Escape of cerebro- spinal fluid into, 33 Ventricular haemorrhage, 118 Vertebrae, Tuberculosis of, 164 Vertebral tumours, 264 Virchow-Robin space. See Peri- vascular lymphatic space Vitamines and beri-beri, 234 Wallerian degeneration, 2, 16, 97, 213 Wassermann reaction in cerebro- spinal fluid, 46, 130 - — ■ — in general paralysis, 154 in tabes dorsalis, 145 Weigert-Pal staining method, 312 Weigert's myelin method, 312 — neuroglia method, 19, 317 Weigert's primary mordant, 306, 312 Werdnig- Hoffmann paralysis, 62 Wounds of brain, 84, 180 Xanthochromia fluid, 38, 43 of cerebro-spinal PRINTED IN GREAT BRITAIN BY BIILINO AND SONS, LTD., GUILDFORD AND ESHER THIS BOOK IS DUE ON THE LAST DATE STAMPED BELOW AN INITIAL FINE OF 25 CENTS WILL BE ASSESSED FOR FAILURE TO RETURN THIS BOOK ON THE DATE DUE. THE PENALTY WILL 'increase to SO CENTS ON THE FOURTH DAY AND TO $1.00 ON THE SEVENTH DAY OVERDUE. BIOLOGY LIBRARY m< ■ ' rr APR l^^l^'^S .<" , 1 APR 2 1938 .4^^ LD21-5m-7,'33 1 ^0 LIBRARY * ' ( i-> .'..> UNIVERSITY OF CALIFORNIA LIBRARY