THE LJUBLJANA ABORTION STUDY 1971—1973 Editor: L. Andoléek Ljubljana, 1974 Published by the National Institute of Health Center for Population Research, Md., U. S. A. Printed by Tiskarna Slovenija, Ljubljana, Yugoslavia ERRATA Inside the title page should be: . Published bythe National Institutes of Health Center for Population Research, Bethesda.'Md.. U. S. A. Page 2, heading for chapter 13 should be: summary and recommendations Page 9, line 3 should be: whereas subsequent chapters describe Page 23, table 31, column 7, line 1‘ should be: 0.7 column 7, line 5 should be: 97.9 Page 28, item 10.1., para 1., line 15 should be: We take less than Hegar 8 to be the normal width'of'the cervical canal. Page 29, line 6 should be: may be too high. Page 33, item 11.2., line 18 should be: in this time indicate a regular Page 36, table- 60, column ‘1, line 7 missing: Proportion Page 39, item 13..,1 line 13 should be. group 20 to 39 years (85.7 °/o) item 13...21 should be. In 157 cases (3.3 %) item 13.2.2 shoulld be: We found 3 perforations of the uterus (0.60/00) and 33 injuries of the cervix (0.7 %), Page 40, item 13.24.. line; should be. We registered 104 cases (2.1 %) Page 40, item 1331 line should be: forty-eight women Page 53, References 3. 3hown, B. On a search for rhesus antibodies in very young fetuses . . , .\ Comparison of the Medical Effects of Induced Abortion by Two Methods, Curettage and Vacuum Aspiration Final Report Principal Investigator: F. Novak Coinvestigator: L. Andoisek This study was supported by the Yugoslavian — American Cooperative Medical Research Program, National Institutes of Health Agreement Number 02-039—1. CONTENTS 25- Page Preface 3 Study participants 4 Chapter 1: Study objectives 5 Chapter 2: Clinical procedures 7 Chapter 3: Data collection and processing 8 Chapter 4: Patient population (D. Bole) 9 4.1. Selected social characteristics 9 4.2. Selected fertility characteristics 9 4.3. Selected characteristics of reproductive behaviour 10 4.3.1. Time of pregnancy termination 10 4.3.2. Sexual activity after abortion 10 4.3.3. Contraception before and after abortion 10 Chapter 5: Operation and operator (L. Andolsek and M. Oven) 12 Chapter 6: Operative events (L. Andolsek) 15 6.1. Anesthesia complications 15 6.2. Uterine injury 15 Chapter 7: Blood loss (L. Andolsek and M. Oven) 17 7.1. Bleeding at and immediately after operation 17 7.1.1. Estimation of blood loss on the basis of the treatment 17 7.1.2. Blood loss at and after surgery 18 7.1.3. Blood loss by operative procedure 18 7.1.4. Blood loss by gestational age 20 7.1.5. Blood loss by duration of operation and operator‘s training 20 7.1.6. Hemoglobin and hematocrit level before and after operation 20 7.1.7. Blood loss and recurettage 21 7.2. Postabortal bleeding 21 7.2.1. Bleeding after discharge 21 7.2.2. Treatment of bleeding 21 Chapter 8: Infection (L. Andolsek) 22 8.1. Events during first and second day of hospitalization 22 8.2. Influence of abnormal vaginal flora and bacteria on incidence of infection 23 8.3. Symptoms of infection reported after discharge 23 Chapter 9: Rhesus factor immunization (M. Hocevar and L. Glonar) 9.1. Goal and study strategy 25 9.2. Immunization of Rh negative women before abortion 25 9.3. Immunization of Rh negative women after abortion 26 9.3.1. Immunization by prior delivery and abortion 26 9.3.2. Immunization by gestational age 26 9.3.3. Immunization by operative procedure 26 9.4. Alteration in titer of anti-Rh agglutinins in already immunized women 26 9.5. Rh immunization two years after abortion 27 Page Chapter 10: Cervico isthmic incompetence (B.KraU and V.Lavflé) 28 10.1. Test applied for diagnosis of CH 28 102.lncidence of CH afler operafive procedure 28 10.3. Factors affecting CII after operative procedure 29 1031. C“ by gefiafionalage 29 10.3.2. Oil by dilatation 29 10.3.3. Oil by shape of cervical canal 29 10.3.4. Oil by operative procedure 30 10.3.5. Oil by skill of operators training 30 10A“ Synechiae afler operafive procedure 30 Chapun 11:0hangesin menshualpanern(L.Ando%ek) 32 11A. Fhstinensnual peflod afler aborfion 32 112.8m4nonfim posbabonwn mensUum paflerns 33 Chapter 12: Ectopic pregnancy (M. Hren, T.Tomazewc and D.Se@eh 34 12.1. Goal and study strategy 34 122.AnaWficalflndhgs 34 12.2.1. Risk of EP by age 34 12.2.2. Risk of EP by previous intrauterine pregnancies 35 12.2.3. Risk of EP by previous deliveries 35 12.2.4. Risk of EP by previous total aborfions 36 12.2.5. Risk of EP by previous induced aborfions 36 12.2.6. Risk of EP by previous gynecologic treatment 37 12.2.7. Risk of EP by previous abdominal operafions 37 12.3. Factors other than induced abortion govenfing EP 38 Chamer13:Sumarw andiecomnwndauon(L.Andomem 39 131. Study and study setfing 39 13.2. Complications during hospitalization 39 132 L Anefihema compHcafions 39 13.2.2. Injury of uterus 39 13.2.3. Blood loss 39 13.2.4. Infection 40 13.3. Complications after discharge from the hospnal 40 13.3.1. Infection and fever without other wmpmms 40 1332.Pomabonm bmemng 40 1333.Pmn m abdomen 40 13.3.4. Pain in lower back 40 13.3.5. Rh Immunization 40 13.3.6. Cervico-isthmic incompetence 41 13.3.7. Menstrual pattern 41 13.3.8. Ectopic pregnancy 41 13A“ Reconunendafions 41 Tabbs 42 Figures 45 References 53 Appendmes 54 Appendix 1: Admission form 54 Appendix 2: Operation form 54 Appendix 3: Vacuum aspiration type VA-3 55 Appendix 4: Hospitalization form 55 Appendix 5: Six-weeks questionnaire 56 Appendm 6:8M-monflm quesfionnane 56 Appendix 7: Ectopic pregnancy form 57 Appendm 8:PubHcafionstesuMng honttMs fiudy 57 One of the significant social changes occuring in recent years has been the pressure of public opinion to permit couples to determine freely the number of children they wish to bear. Such control over child- bearing requires a variety of methods, including abortion. Because of recent achievements in medi- cine, particularly the development of antibiotics and new surgical techniques, abortions performed in a hospital during the first ten weeks of pregnancy have become relatively safe. The death rate of women undergoing legal abortion is significantly lower than the death rate of women at delivery. Despite these advantages, health workers persevere in their search for even safer methods for pregnan- cy termination. One such method, vacuum aspira- tion, appeared in the medical literature of East Euro- pean countries after 1960. This technique was intro— duced at the University Teaching Hospital of Obstetrics and Gynecology in Ljubljana in 1964 and within a few years all the doctors had become fami- liar with the new technique and have used it since for routine cases. Although most authors emphasize the advantages of vacuum aspiration over curettage, we have not found any report of a study in which both methods are compared simultaneously in a controlled clinical trial. This monograph describes such a study. It also estimates the frequency and intensity of complica- tions after both types of abortion. A primary activity of «the Family Planning Institute of Ljubljana is research on abortion and other methods of fertility regulation. In an earlier study (1) we de- termined the complications which occur with first trimester interruptions performed by the routine suction procedure used at the University Teaching Hospital. The work described here is a continuation of this investigation. PREFACE The study was organized and coordinated by the Family Planning Institute and carried out by three Institutions: the Obstetrical and Gynecological Clinic of the University Teaching Hospital, the Family Plan- ning Institute, and the Blood Transfusion Institute of Slovenia. The report includes patients studied in the years 1971 and 1972. The research plan was prepa— red by the chief investigators in cooperation with Dr. P. Corfmian and Dr. D. Seigel of the National In- stitute of Child Health and Human Development and the study was financed through the Yugoslavian— American cooperative medical research program. Our results are presented chronologically, from the detection of pregnancy through the procedure itself, events occuring during hospitalization, and during the first few months after the procedure. Late se- quelae are referred to in the monograph but will be described in greater detail elsewhere. Some chapters are revisions of articles which have already been published. The chief investigators of the various portions of the study are identified as authors of the separate chapters. The authors also appear on the roster of all cooperating participants. The report concludes with a list of publications pre- pared from the study. We ta‘kethis.,opportunity to thank all of our collea- gues in Ljubljana and Bethesda, physicians, (nurses, laboratory personnel, social workers and office workers, for their enthusiastic cooperation. We also thank Dr. M. Cavié, Assist. Prof. M. Gu’bina, Prof. S. Rainer and Prof. D. Vilfan for their consultation on specific topics, Dr. P. Corfman for ‘his assistance in the preparation of this monograph, and Dr. R. Ber- nard for his valuable advice. Finally we hope that the results and experiences gai- ned from our study and reported in this monograph will prove of value to workers all over the world who deal with problems relating to abortion. Ljubljana, February 1974 F. Novak STU DY PARTICIPANTS Obstetrical and Gynecological Department of the University Teaching Hospital, Ljubljana Meta Hren, M. 0., Senior Gynecologist Boio Kralj, M. D., Senior Gynecologist Vito Lavrié, M. D., Professor of Obstetrics and Gynecology Franc No'vak, M. D., Professor of Obstetrics and Gynecology, Former Director of the Obstetrical and Gynecological Clinic Tomai Tomaievié, M. D., Assist. Prof. of Obstetrics and Gynecology Family Planning Institute, Ljubljana Lidija Andolsek, M. D., Assoc. Prof. of Obstetrics and Gynecology, Director of the Family Planning Institute Daéa Bole, Graduate Sociologist Majda Oven, Graduate Economist and Statistician Transfusion Institute, Ljubljana Miran Hoéevar, M. D., Specialist in Transfusion Ljerka Glonar, M. 0., Specialist in Transfusion National Institute of Child Health and Human Development, Bethesda, Maryland Philip Corfman, M. D., Director, Center for Population Research Daniel Seigel, S. D., Chief, Biostatistics Branch Chapter 1: Complications following induced abortion, both le— gal and illegal, have always been the subject of se- rious discussion. Nevertheless, many gynecologists now are of the opinion that complications of legal abortion are few and that they are even fewer when the vacuum aspiration (VA) technique is used. Other gynecologists have the opposite opinion, na- mely, that the classic curettage method (D & C) is much safer. Indeed, the use of VA has been discou- raged by government decree or medical opinion in certain countries. The basic aim of our study is to compare the fre- quency and intensity of immediate and early com- plications following these two techniques. Many authors are of the opinion that a D & C indu- ces heavier bleeding during and after the operation, more infection, and more Rh sensitization. They also feel that more dilation is required than for VA. But there is also the unproved hypothesis that VA might induce ectopic pregnancies in conceptions occuring subsequent to the procedure. As noted in the Preface, the University Obstetrical and Gynecological Clinic in Ljubljana adopted VA in preference to D & C several years ago (1954). This was done because it was felt that VA has the follow- ing advantages: 1. the operation is of shorter duration, 2. the uterus is evacuated more completely and efficiently, 3. there seems to be less uterine injury, particularly perforation, and 4. there appear to be fewer major complications. t is clear, therefore, that our group has a bias in favor of VA. Nevertheless, we felt that there was a real need to establish with the best possible study design which technique is in fact better. Even though we chose VA for our hospital, we conceded that the choice was based on clinical judgement rather than an analysis of scientific data. Also we knew that oth- er gynecologists of considerable reputation had decided that VA was not significantly better than D & C. We feel that the organization of our Clinic makes us uniquely capable of undertaking this work since STUDY OBJECTIVES we are able to conduct a double blind study in which the procedure is randomly assigned. Both techni- ques were used in the same time period by random allocation according to the last digit in the patient’s birth year: odd number—VA; even number— D & C. Furthermore, and of great importance, the medical staff which cared for the women after surgery was not aware of which technique was used. Also, neither the operators nor the medical care staff par- ticipated in data processing and analysis. The factors in which we were primarily interested include the following: . anesthesia complications . uterine injuries blood loss . infection Rh sensitization . cervico-isthmic incompetence . menstrual patterns ectopic pregnancies in subsequent pregnancies . future fertility. copnxiovgntspam—t Observations on these various effects are described in detail in subsequent chapters. A few of the essen- tial issues are elaborated here. The study population was 4733 women, admitted to the Obstetric&Gyneco- logical Clinic from the period January 1971 through December 1972 for abortion for social reasons. During this period, 67 abortions were done for medical reasons and they are omitted from the report. In all study women we monitored the events of anesthesia, uterine injury, blood loss, and infection sub-, sequent to the procedures. Blood loss was monitor- ed more closely in a smaller group of women in whom hemoglobin and hematocrit determinations were made before and 48 hours after the procedure. A ten percent sample of the women (all those with a case number ending in 0) were asked to return a questionnaire (Appendix 6) six months after the pro- cedure. The questionnaire included such topics as the menstrual cycle, blood loss, and abdominal and lower back pains. As noted in a subsequent chapter, the return of this questionnaire (53%) was some- what Iess than we had expected based on previous use of questionnaires with our patient population. 5 Later sequelae were studied in yet smaller groups of women selected as described in the appropriate chapters. One such sequela liS Rh sensitization, well known to be induced by abortion. Nevertheless, it has not been established in which period of preg- nancy this sensitization is most apt to occur and whether or not there is a relationship with the tech- nique employed. We did anticipate however, that VA would induce less sensitization because we feel that it is less traumatic and therefore less apt to permit fetal antigens to enter the maternal blood stream. Another late sequelae of considerable interest is the possibility of cervicmisthmic insufficiency. Our stu- dy group for this problem were all nulliparous wo- men who agreed to be part of the study. Another se- rious late sequelae is ectopic pregnancy. We are particularly concerned with this because it has been hypothesized that this problem is somehow linked to abortion. The last problem of concern is the effect of the pro- cedure on future fertility. Does abortion increase the risk of subsequent sterility? For this study we chose women who declared at the time of abortion that they wished to have a child in the future. Since we intend to send questionnaires to such women two times in two years, this \portion of the study is still under way and the data are not at hand for this monograph. L Chapter 2: All women in Yugoslavia who wish to terminate a pregnancy must appiy to a committee selected by local health authorities and composed of a gyneco- logist, an internist, and a social worker. This com- mittee must approve of the abortion before it may be done. If the applicant is denied permission, she may apply to a second committee nominated by the Health Department of the Republic and whose deci- sion is final. Committee members, particulary the social worker, consult and advise the applicant concerning abor- tion and family planning. In this way, every woman is informed about family planning and about various contraceptives that are available. Some women will ask for immediate protection after the abortion through the insertion of an lUD or the injection of depogestagens. Other women come for consultation four to six weeks after the procedure. Women who chose lUDs or depogestagens were excluded from this study. Women were invited to participate in the study by the social worker of the abortion review committee. The women were again offered the opportunity to join when admitted to the hospital. The study inclu- ded all women who agreed to be hospitalized for 2—3 days, were willing to cooperate by returning a questionnaire four to six weeks after the procedure (Appendix 5) and, in a ten percent sample, return another questionnaire after the sixth month (Appen- dix 6). Women not willing to spend two to three days in the hospital and those who for various reasons were not willing to participate or were too far advan- ced were excluded. Blood samples for determination of blood groups, including Rh factor, were taken when the subjects first met the committee. Women were admitted to the hospital on the mor- ning of the day of the operation. At that time, a phy- sician again ascertained the indications for abortion and confirmed the existence of the abortion com- mittee‘s permission. The procedure was done the same day for healthy women with no contraindi- cations. After admission formalities, the social worker filled in the admission form (Appendix 1). At the same time she handed each patient a questionnaire (Appen- CLINICAL PROCEDURES dix 5) with instructions concerning its use. Immedia- tely before the operation, the social worker deter- mined whether a VA or D & C would be done accor- ding to the patient’s birth year, as previously de- scribed. The operator examined the patient again, determi- ning the position, shape, and size of the uterus. The anesthesiologist reviewed the history with the ope- rator and if necessary, examined the patient. In this report the period of gestation is always stated in completed weeks. The pelvic exam was done to confirm this estimation. In 28 women (5.9%) the duration of gestation was based only on the physical examination because either the women did not re- member her last menstrual period (LMP) or had had no period since her last confinement. An operation was performed on all patients diagno- sed as pregnant. During the 2 year study period 50 of these women were shown by histological exami- nation not to have been pregnant. These cases are excluded from the study. Most women received the same premedlcation (atro- spin 0.5 mg and Hyoscin — N — butylbromid 20 mg, i. v.). Only 0.4% of the patients received pre- medication of some other type, a proportion that we feel too small to affect our results. Most abortions (98.7 %) were performed under general anesthesia using sodium thiopental, usually in a range of 1.2 to 1.3 grams. Only 1.3 % of the procedures were per- formed under analgesia. We recognize that the local technique has received favorable attention in recent times but our preference is for general anesthesia which we have used for over twenty years. We have all the necessary facilities, adequate means to monitor the recovery period, and a suffi- cient number of qualified anesthesiologists.* After emptying the bladder, introducing the vaginal speculum and cleansing the vagina with an antisep- tic solution, we hold the anterior lip of the cervix with a single-tooth tenaculum and apply traction to straighten the uterus. We then introduce a uterine sound to determine the depth and position of the uterus and dilate the cervical canal with Hegar’s di- ' At the printing of this report a second controlled study compa- res local versus general anesthesia and outpatient versus Inpa- tlent care. Sponsor: WHO, Geneva. 7 lator-s. The extent of dilation is one Hegar size less than the week of pregnancy, with a maximum of He- gar 13. Shortly after dilation, we give 10 units intra- venously of synthetic pitocin dissolved in 10 cc of 0.9 % physiological saline. The form used to descri- be the operation can be seen in Appendix 2 . The D & C is performed with standard sharp curet- tes. VA is carried out with a suction apparatus made in Ljubljana and fitted with metal cannulae (Appen- dix 3). After VA some physicians check the uterus with a sharp ourette to ascertain that the uterine cavity is empty. This ”curette-check" was not consi- dered a second procedure. Occasionally, some pla- cental residue rs found. For this reason, we are obliged to divide our VA g-‘roup of patients into three subgroups: Subgroup 1. Vacuum aspiration only. Subgroup 2. VA checked with curette, no placental tissue found. Subgroup 3. VA with curettage, placental tissue found. in the third subgroup a second procedure was requ- ired to complete the abortion. Therefore the cases in this group are declared a ”failure" of VA. All aspirated material is examined macroscopically and when macroscopiC‘exami-nation is insufficient for diagnosis the aspirated material is subjected to histological examination to verify the pregnancy. The patients awake from anesthesia in the operating room or their hospital beds. Although we are aware that post abortal complications after 48 hours are rare, we keep most of our patients (95 %) for two days or longer after the procedure. During this pe- riod we take temperatures twice a day and register all symptoms such as pain, tenderness and bleeding (see Appendix 4). We also record all medication the women take. After discharge the women receive 7 to 10 days of sick leave. Chapter 3: Most of the data obtained in this study was derived by use of a randomized clinical trial. The only excep- tion is the sub-study on ectopic pregnancy which is case-controlled. Data were obtained from the patients through direct interview and through the use of questionnaires which are in the appendices. Direct interviews took place before the operation, and the social worker filled in the questionnaire. At the same time she explained the study purposes and asked the patient if she were willing to cooperate. The woman was given an addressed envelope with the 6-week que- stionnaire and a careful explanation of how to answer the questions and when to return the form. Every 10th woman also was given the 6-month questionnaire, a menstrual calendar, and instruc- tions for the appropriate handling of both. Data were also obtained from hospital records and from special questionnaires completed by physi- cians and nurses. Hospitalization data were obtained from the patients’ hospital charts prepared espe- 8 DATA COLLECTION AND PROCESSING cially for this purpose. Observations were noted by nursing staff and physicians during the post-ope- rative hospitalization period. Data were collected in special ways for the different sub-studies. Admittance to an individual study was determined by the social worker assigned to the patient before the procedure. These sub-studies also required special questionnaires which are de- scribed in the appropriate chapters. The questionnaires were prepared to permit direct coding based on a coding list prepared at the be- ginning of the study. This information was keypun— ched and transferred to a magnetic tape. Tabulation was done in part with an IBM 1130 system and with a CDC 6600 system available at the Ljubljana Com- puter Center. Data for all sub-studies except Rh sensitization and sterility were also computerized. Tables were prepared for the main study and sub- studies by the statistician and the programmer in collaboration with the physicians responsible for in- dividual studies. Chapter 4: The purpose of this chapter is to describe some characteristics of the entire patient population whereas subsequent chapter describe a variety of experimental and control groups formed to answer specific questions. It must be pointed out that our patients were self-selected volunteers. Also certain women (67) were excluded from the study for vari— ous medical reasons. Consequently our study group cannot be taken as representative of the complete population of women undergoing abortion in our region. 4.1. SELECTED SOCIAL CHARACTERISTICS Table 1 provides some of the most general social characteristics of our patients. The over-repre- sentation of urban residents in the abortion group is compatible with findings from all regions where abortion is practiced. The majority of the women selected were of lower or middle educational levels, a finding which to some extent reflects the edu- cational composition of our general female popu- lation. About three-fourths (76.5 %) of the patients were income-earners, most of whom being employ- ed, others being owners of small private businesses and farmers. Only 12.7 % were housewives not work- ing outside the home and 10.2 % were students. The patient characteristics are illustrated in Figure 1. Table 1: Selected social characteristics Characteristics No. of patients % of the total Residence Urban 3266 69.0 Not urban 1467 31.0 TOTAL 4733 100.0 Education 0—4 school years 385 8.1 5—8 1891 39.9 9—12 1911 40.4 13—16 539 11.4 17—20 or more 7 0.2 TOTAL 4733 100.0 Occupation Income-earner 3623 76.5 Housewife 600 12.7 Student 483 10.2 Other (invalids, pensioners, etc.) 27 0.6 TOTAL 4733 100.0 PATIENT POPULATION D. Bole 4.2. SELECTED FERTILITY CHARACTERISTICS As indicated in Table 2 candidates for surgery were mostly in their 20’s or 30’s, i. e., in their best repro— ductive period. The fact that 8.4 % of the patients were 19 or less is a matter of some concern. A large portion (77.7 %) of the patients had had at least one live birth before the operation, whereas 33.6 % had undergone at least one prior induced abortion. it is interesting to note that 35.7% of the women said they intended to have a child in the future. Table 2: Selected fertility characteristlcs Characteristics No. of patients % of the total Age 14 years or less 2 0.1 15—19 393 8.3 20—24 1337 28.3 25—29 992 21.0 30—34 895 18.9 35—39 715 15.1 40—44 367 7.6 45 or more 32 0.7 _ TOTAL 4733 100.0 Previous pregnancies a) Live births 1 1302 27.5 2 1756 37.1 3 428 9.0 4 138 2.9 5 39 0.8 6 15 0.3 7 2 0.0 l 8 1 0.0 0.1 9 and more 2 0.0 l Subtotal 3683 77,7 b) Pregnancies not terminating in live birth 126 2.8 c) First pregnancy 924 19.5 Total 4733 100.0 Previous induced abortions No 3144 66.4 Yes 1589 33.6 Total 4733 100.0 Woman wants additional children No 2684 56.7 Yes 1690 35.7 Undecided 359 7.6 Total 4733 100.0 Although our experimental methods prevent us from undertaking extensive sociological interpretations, we may draw some conclusions. For instance, there is high participation in abortion practiced by high risk women: primigravidae, women with repeated in- duced abortions, teenagers, and women with an expressed wish for additional children. This certain- ly accentuates the need to find abortion proce- dures with minimal health risks and to establish suc- cessful ways of motivating women for contraception. 4.3. SELECTED CHARACTERISTICS OF REPRODUCTIVE BEHAVIOUR 4.3.1. Time of pregnancy termination Usually, highly educated women first visit a physician in the fifth or sixth week of amenorrhoea, while other women come later. The current legisla— tion and the administrative requirements, such as a patient’s interview with the abortion committee, do not facilitate the implementation of early abortion on a mass scale. As shown in Table 3, almost all of our patients (89.1 %) were aborted between the 7‘h and 11‘h week of pregnancy and most of these in the 8th (25.1 %) and in the 9th week (24.8 °/0). Table 3: Pregnancy termination, by gestational age % distribution Duration of pregnancy No. of patients of total 4 weeks 6 ’ 0.1' 5 13 0.3 6 82 1.7 . 7 569 12.0 8 1188 25.1 . 9 1173 24.8 10 786 16.6 11 503 10.6 12 231 4.9 13 102 2.2 14 or more 52 1.1 NN 28 0.6 TOTAL 4733 100.0 4.3.2. Sexual activity after abortion Table 4 examines the interval between the abortion and the time subsequent sexual intercourse was said to occur. Table 4: Interval between abortion and first Intercourse Interval between . . No. of 0 Corrected fntigggflrggd my patients /° Of total “/0 of total 1 week 64 1.4 1.4 2 165 3.5 3.5 3 410 8.8 8.8 4 640 13.7 13.8 5 787 16.9 17.5 6 544 11.7 14.4 7 200 4.3 8.0 8 84 1.8 5.0 9 and more 90 1.9 27.5 Not yet 1679 36.0 -— TOTAL 4663* 100.0 100.0 ' Out of 4733 questionnaires, 66 were not returned and 4 contained no answer to this question. 10 Out of 4663 women in the study, 1679 said they had not resumed intercourse at the time they returned their questionnaire. More information on these wo- men is provided in Table 5. Table 5: Patients stating no postoperative intercourse, by time of questionnaire return Time of returning the questionnaire No. of patients % of total 1 week 0 0.0 2 0.1 3 1 0.1 4 37 2.2 5 145 8.7 6 663 39.6 7 368 22.0 S 159 9.5 9 weeks and more 298 17.8 1674* TOTAL 100.0 ' For 5 out of 1674 questionnaires the date of return is not known. Using the life table method the information in Table 5 was used to correct the distribution percentages and these corrections appear in the last column of Table 4. According to this evidence, the frequency of first sexual intercounse in the study group increased somewhat in the fourth through sixth week after abortion, but a considerable number of women (27.5%) did not report intercourse even after nine or more weeks following the procedure. These ans- wers are to be taken with certain reserve. We have reliable information that our women in most cases resume sexual relations alreadyvin the third week following «delivery which may be a far greater phy- sical stress than abortion. 4.3.3. Contraception before and after abortion Of considerable interest was the contraceptive prac- tice of the women in our study*. This topic is dis- played in Table 6 which shows that before abortion only 9 % of our patients used the two most effec- tive means, (orals and lUDs); 10 % used methods of less reliability, (the diaphragm, condom and che- micals); 39 % used poor methods, (coitus interrup— tus and rhythm); and 42% reported no contra- ception. After abortion there was considerable increase in the use of oral contraceptives and lUDs, (from 9 % to 27%) and there was some shift in the use of coitus interruptus and rhythm. Most remarkable is the fact that the percentage of non-users remained essentially unchanged. ' These calculations were done only tor the women who answered the questions on contraceptlve practice. Table 6: Contraceptive practice before and after abortion Contraceptive method Orals IUD Diaph. Condom Chem. C. I. Rhythm None BEFORE ABORTION No. of women 242 157 137 176 1412 308 1852 % of women 5.5 3.6 3.1 4.0 32.0 7.0 41.9 CHANGE No. of women +752 +31 —20 —52 —75 ~1145 +710 —134 0/0 of women +16.8 +0.6 —0.4. —1.2 —2.4 —-26.0 +15.9 —3.3 AFTER ABORTION No. of women 994 188 85 71 267 1018 1718 % of women 22.3 4.2 2.5 1.9 , » 1.67 6.0 22.9 38.6 We examined this topic further by constructing a cross tabulation.» of contraceptive methods used before and after the abortion procedure, as shown In Table 7. Table 7: Contraceptives used before and after abortion \ Before N 0 ~ None Orals IUD Diaph. Condom Chem. C. l. Rhythm Subtotal NK Total After \ quest. None 903 350 63 25 26 19 76 385 1847 3 2 1852 Orals 65 93 9 2 2 2 19 44 236 0 6 242 IUD 48 29 11 5 1 3 13 45 155 0 2 157 Diaph. 27 20 9 47 O 3 13 10 129 0 0 129 Condom 39 34 1 2 2 4 41 14 137 0 0 137 Chemicals 39 49 14 3 5 26 7 32 175 1 0 176 C. I. 468 327 65 18 28 9 81 415 1411 1 0 1412 Rhythm 101 78 13 4 20 5 16 71 308 0 0 308 Subtotal 1690 980 185 106 84 71 266 1016 4398 5 10 4413 NK 0 0 0 0 0 0 O 0 0 0 0 0 No quest. 28 14 3 3 1 0 1 2 52 0 268 320 TOTAL 1718 994 188 109 85 71 267 1018 4450 5 278 4733 In 73.9 % of cases there was a change in method, but this change was not necessarily to better met- hods. Unfortunately, the greatest stability was found in non-users who in 48.8 041 continued to ignore con- traception. The next most stable group were those who used oral contraceptives, with 38.4 % remai- ning faithful to this method. We observe from this table that there is a great variety of responses and much instability in contra— ceptive practice. This may be characteristic of the time periodinuolved, since the questionnaires were returned only a few weeks after the procedure (82 % before 9 weeks had passed and most in the second to fourth week). We feel that this period is too short to allow contraceptive practice to stabi- lize and a longer period for follow-up is required to permit meaningful conclusions. 11 Chapter 5: As described, the women were selected for the type of operative technique on the basis of the last digit of the birth year. The D &C was done by the usual technique and the VA technique was uniform for ail operative procedures with differences only in the last part of the operation. Some of the operators used only VA. They did not deem it necessary to check the uterus with a curette in order to make certain that the uterus had been evacuated completely. In this group 23.8 % such operations were done by junior and 28.6 % by se- nior operators. Approximately two thirds of VA procedures were routinely checked at the end of operation to make certain that the uterine cavity was empty. This was done with the smallest size curette in 67.8 % of ope- rations by junior and in 56.5% of operations by senior operators. Such curette—check after VA is a well established practice at the Gynecological Cli- nic. Since we introduced the suction technique in 1964, the majority of gynecologists use this method. In some cases the operators found placental tissue with the check-curette. It is interesting that a pro- portionally larger number of this class of operations were performed by senior operators (senior 14.9 %, junior 8.4 °/o). Also, incomplete evacuation was more frequent after the twelfth week. Furthermore, we know that additional curettage increases the number and intensity of complications, especially bleeding. The distribution of the procedures is shown in Table 8. Table 8: Operative procedures, by pregnancy duration and the operator’s training VA+ Total VA VA + Total D & C curette . VA only check curettage NO. 0/0 c'/o 0/o 0/o 0/0 Total 4733 51.7 48.3 27.0 60.2 12.8 Week of pregn. 7 and less 920 51.6 48.4 28.5 63.6 7.9 8 1617 52.1 47.9 34.1 56.9 9.0 9 589 51.1 48.9 28.8 63.2 8.0 10 926 50.8 49.2 20.8 63.4 15.8 11 243 52.7 47.3 20.0 67.0 13.0 12 397 53.7 46.3 13.6 54.4 32.0 13 and more 41 43.9 56.1 0.0 21.7 78.3 Operator Senior 3161 51.7 48.3 28.6 56.5 14.9 Junior 1572 51.8 48.2 23.8 67.8 8.4 THE OPERATION AND THE OPERATOR L. Andolse‘k and M. Oven It is evident that proportional relations exist bet- ween D & C and VA subgroups, with regard to the advancement of pregnancy and the operator’s skill. Table 9 shows 28 operators who performed over 50 abortions, with regard to the kind of VA operation they chose. The table shows that five operators per- formed 58.0 % to 81.6 % VA only operations. Eight operators performed up to only 9 % VA only proce— dures, and checked all others. Therefore the deci- sion whether to check the uterus lies individually with every operator and does not seem to depend upon the circumstances of the particular operation. Table 9: Operators with more than fifty operations, by operative procedure VA+ Number of D 8 C Total VA cu rette VA+ operations % ‘5: oglly choeck CU'E/Ztage O 301 53.2 46.8 81.6 16.3 2.1 436 52.7 47.3 11.7 76.5 11.7 290 47.9 52.1 69.5 21.9 8.6 335 54.3 45.7 26.8 69.9 3.3 99 48.5 51.5 25.5 66.7 7.8 121 51.2 43.8 3.4 71.2 25.4 150 53.3 46.7 70.0 22.9 7.1 124 49.2 50.8 20.6 36.5 42.9 165 47.3 52.7 21.8 42.5 35.6 241 51.9 48.1 0.9 63.8 35.4 146 52.7 47.3 27.5 72.5 0.0 75 45.3 54.7 2.4 95.1 2.4 71 50.7 49.3 22.9 68.6 8.6 101 54.5 45.5 15.2 69.6 15.2 125 57.6 42.4 9.4 77.4 13.2 59 57.6 42.4 4.0 76.0 20.0 110 50.0 50.0 9.1 83.6 7.3 51 45.1 54.9 14.3 25.0 60.7 228 51.3 48.7 18.0 76.6 5.4 171 41.5 58.5 58.0 40.0 2.0 215 49.3 50.7 11.0 81.7 7.3 158 50.6 49.4 14.1 74.4 11.5 183 64.5 35.5 73.9 21.5 4.6 73 57.5 42.5 0.0 90.3 9.7 102 55.9 44.1 4.4 77.8 17.8 93 48.4 51.6 0.0 79.2 20.8 70 55.7 44.3 6.5 80.7 12.9 61 44.3 55.7 5.9 85.3 8.8 12 It is clear therefore that there were in fact four kinds of operative procedures and outcome in this study as illustrated in Figure 4. 1. D & C 51.7 % 2. Total ‘VA 48.3 % (100 %) — VA only 13.0 % (27.0 0/o) — VA plus curette check without placental tissue 29.1 % (60.2 %) — VA plus curettage, with placental tissue 6.2 % (12.8 %) The operator‘s training and experience were also considered for their possible influence on the con- sequences of the operation. As shown in Table 10, most of the surgical procedures were carried out by senior operators (gynecologists who have passed the specialty examination) and the rest by junior operators. Table 10: Operators, by training Operator No. of operations % of total Senior 3161 66.8 Junior 1572 33.2 4733 100.0 Total The operators performed up to 14 abortions each day, usually not more than eight. The D & C lasted on the average 7.2 minutes, the average duration of VA was 5.8 minutes. There were six VA operations (0.3%) lasting 19 minutes or longer and in the D 8 C group there were 19 such cases (0.8 °/o). D & C by senior gynecologists lasted on average 6.4 minutes but non—specialists took 8.6 minutes. The difference is significant. VA by senior gyneco- logist took on average 5.5 minutes whereas the ju- nior operators required 6.5 minutes. This differential pattern is also significant and is illustrated in Fi- gure 5. Whereas all four perforations (3 confirmed and 1 suspected) were done by senior operators, injuries to the cervix were distributed between juniors and seniors as seen in Table 11. Table 11: Uterine injury, by the operator’s training No. of operations No. of injuries % Senior 3161 26 0.82 Mior 1572 7 0.45 Total 4733 33 0.70 It is evident that operations by senior operators were associated with 26 injuries (0.82 % of all ope- rations) and procedures by junior operators with 7 injuries (0.45%). This difference is marginally significant. From data obtained for each operator a conclusion can be drawn that operators who per- formed less than 100 abortions during the two year period had more injuries (0.92%) than did those who performed between 200 and 300 procedures (0.41 °/o). Operators who performed more than 300 operations had a higher percentage of injuries (0.71 %) suggesting they may have been overschedu— led. One operator performed more than 450 ope- rations and was associated with 5 cervical injuries (1.01 °/o). This V—shape pattern of uterine injuries is illustrated lin Figure 6.* The order of the operation executed by the same operator on the same day is also an attribute of in— terest. The intent is to measure possible effects of fatigue on the outcome and, thus, help in establi- shing an optimal number for each day. Table 12 demonstrates a marked drop in the number of ope- rations after the eighth. We did not find an asso- ciation between the order of operation and the ap- pearance of complications until after the eighth procedure. Table 12: Operators, by operation order attained at same day Order of operation No. of operations °/o of total 1 907 19.2 2 777 16.4 3 678 14.3 4 601 12.7 5 519 11.0 6 442 9.4 7 365 7.7 8 265 5.6 9 125 2.6 10 38 0.8 11 or more 16 0.3 4733 Total 100.0 The quality of operation was assessed by 1) the percentage of immediate complications (lacerations, perforations, and bleeding which required additional uterotonic treatment), 2) incomplete evacuation of the uterus (placental residua established during ho- spitalization or at readmission to the hospital) and 3) the failure of VA as a first procedure when curet- tage was necessary. Table 13 and Figures 7 and 8 give the relevant information. Table 13: Complications, by operation order Lacerations Placental VA cases and Bleeding tissue requiring perforations found curettage Order of operation 16 ":6 E < a 8 8 ‘6 .._ > ‘6 o a C 0 c I: o o I 0 ° TU l— 2 o\ 0 2 o\ 0 2 o\ 0 2 03 E 1 907 3 0.3 47 5.2 16 1.8 70 15 4 2 777 8 1.0 38 4.9 8 1.0 42 11 1 3 678 5 0.7 29 4.3 12 1.8 39 12 2 4 601 8 1.3 34 5.7 13 2.2 37 12.5 5 519 5 1.0 27 5.2 7 1.3 28 11.5 6 442 3 0.7 19 4.3 5 1.1 25 12.1 7 365 1 0.3 20 5.5 9 2.5 22 12.6 8 265 0 0.0 14 5.3 2 0.8 20 16.4 9 125 3 2.4 6 4.8 2 1.6 8 12.7 10 38 1 2.6 3 7.9 1 2.6 0 0.0 11 or more 16 0 0.0 0 0.0 0 0.0 1 16.7 Total 4733 37 1 6 292 12.8 0.8 237 5.0 75 it will be noted in Figure 7 that the three types of complications increase after the 9th procedure. Figure 8 shows the percentage of VA cases requiring additional curettage, by the order of operation. It was expected that the order of operation is related to failure of the first procedure and the subsequent need for additional curettage. But as evident in ’ Because of the significance the data in this section are presen- ted in two decimals. 13 Table 13 and in Figure 8 this speculation was not confirmed. We are also interested in correlating the operators’ estimations of gestational age with those obtained from the date of UMP. Table 14 shows that the operators tend to diagnose by even numbers of weeks. Table 14: Pregnancy duration, by last menstrual period and by the surgeon’s estimate in weeks According Surgeon’s estimate to LMP 4 5 6 7 8 9 10 11 121314 Total 4 0 O 1 2 2 1 0 0 0 0 0 6 5 0 0 2 2 3 1 1 2 2 0 O 13 6 0 5 21 9 25 11 6 3 1 0 1 82 7 8 18 126 96 218 52 35 4 12 0 0 569 8 11 19159124 534157132 22 30 0 01188 9 2 8 83 101 475 187 222 32 59 0 4 1173 10 0 5 33 39 211 116 259 63 55 1 4 786 11 1 2 14 7 81 45 189 73 81 3 7 503 12 0 0 4 6 43 10 48 36 80 1 3 231 13 O 0 0 2 7 3 17 6 57 5 5 102 14 and more 0 0 2 2 12 1 11 1 16 5 2 52 Total 22 57 445 390 1611 584 920 242 393 15 26 4705 This observation is also illustrated in Figure 9. The operators generally estimate pregnancies to be less advanced than calculated from the LMP. The dif- ference between the mean duration of pregnancies is significant at P < 0.01; only 46.3 % of estimations by operators cite pregnancies over eight weeks as compared with 60.5 % of estimations by LMP. 14 Chapter 6: The following factors may influence the frequency of abortion complications: the woman's health sta- tus, parity, gestational age, anesthesia, the opera- tive technique, the duration of hospitalization, the duration of sick leave, the technical skill of the ope— rators. 6.1. AN ESTH ESIA COMPLICATIONS Table 15 shows the major complications of anes- thesia occuring in healthy women having abortions for social indications. We calculate that anesthesia complications do not differ significantly with regard to the type of operation technique. Table 15: Complications of anesthesia, by operative procedure Episodes D & C All VA Grand total NO. 0/0 NO. °/o NO. 0/0 Laryngospasm 43 1.8 46 2.0 89 1.9 Anaphylaxis 5 0.2 5 0.2 10 0.2 Shock 1 0.0 0 0.0 1 0.0 Vomiting 20 0.8 8 0.4 28 0.6 Convulsions 15 0.6 7 0.3 22 0.5 Combination* 6 0.3 1 0.0 7 0.1 Subtotal 90 3.7 67 2.9 157 3.3 None 2358 96.3 2218 97.1 4576 96.7 Grand total 2448 100.0 2285 100.0 4733 100.0 VA + VA + . VA on] curette Episodes Y check curettage No. % No. % No. % Laryngospasm 7 1.1 32 2.3 7 2.4 Anaphylaxis 0 0.0 4 0.3 1 0.3 Shock 0 0.0 0 0.0 0 0.0 Vomiting 1 0.2 5 0.4 2 0.7 Convulsions 2 0.3 3 0.2 2 0.7 Combination 1 . 0.2 0 0.0 0 0.0 Subtotal 11 1.8 44 3.2 12 4.1 None 606 98.2 1332 96.8 280 95.9 Total VA 617 100.0 1376 100.0 292 100.0 The duration of the operation also does not signi- ficantly influence the number and intensity of an- esthesia complication-s as shown in Table 16. It is evident that almost 86% of the operations lasted * This group includes 5 women who experienced laryngospasm and vomiting, and 2 women with convulsion and vomiting. OPERATIVE EVENTS L. Andolsek from 1 to 10 minutes and that in this period most anesthesia complications appear. Something less than 25 % of the anesthesia complications occured during operations lasting over 10 minutes. Table 16: Complications, by length of operation 1_10 11 and Time Episodes Total minutes mriiiiitZS kngevtn NO. 0/0 0/o 0/o Laryngospasm 89 76.4 22.5 1.1 Anaphylaxis 10 80.0 20.0 0.0 Shock 1 100.0 0.0 0.0 Vomiting 28 71.4 25.0 3.6 Convulsions 22 90.9 9.1 0.0 Combination 7 42.9 57.1 0.0 None 4576 86.3 6.5 7.2 Grand total 4733 86.0 7.0 7.0 6.2. UTERINE INJURY Some surgical complications of induced abortion are not immediately detectable and their incidence is probably underreported. Indeed, reported perfo- ration is rare, especially using VA. In 4733 induced abortions we diagnosed and con- firmed only 3 perforations (0,6 °/oo). One additional case was reported as suspicious. All perforations occured in the D & C group during dilatation, sug- gesting that the instruments producing the perfo- rations were Hegar dilators. In all cases the ope- rators were senior gynecologists. All women were multigravidas, three were pregnant 8 weeks and one was pregnant 10 weeks. Three of these women underwent laparotomy and suture of the perforation; the suspicious case was managed by observation alone. The women remained in the hospital for 5—10 days and their postoperative courses were normal. Table 17 displays all patients who suffered injury of the cervix at the time of dilation. We include all lesions of the cervix — cervical tears and lacera- tions. Tears may be caused by stabilization instru- ments (the single tooth tenacula) and lacerations by dilators. There is no marked difference in cer- vical lesions regarding the technique used. 15 Table 17: Injuries of the cervix, by operative procedure Treatment of D & C All VA Grand total injury No. % No. % No. °/o Suture 2 0.1 6 0.3 8 0.2 No suture 15 0.6 10 0.4 25 0.5 Subtotal 17 0.7 16 0.7 33 0.7 No injury 2431 99.3 2269 99.3 4700 99.3 Grand total 2448 100.0 2285 100.0 4733 100.0 Treatment VA only VA :hgzLette VA + curettage of ”W” No. % No. % No. % Suture 2 0.3 3 0.2 1 0.3 No suture 1 0.2 8 0.6 1 0.3 Subtotal 3 0.5 11 0.8 2 0.7 No injury 614 99.5 1365 99.2 290 99.3 Total VA 617 100.0 1376 100.0 292 100.0 As can be noted from the Table, cervical suture was deemed necessary in 24 % of cervical injuries. Other injuries were minor and did not demand special therapy. There was no difference in these events among the different types of operation. 16 Chapter 7: We divided bleeding events with abortion into two groups: 1. bleeding at operation, and 2. ‘postabortal bleeding. In the first group are included all bleedings registered in the course of the operation and du— ring later hospitalization. It should be noted that bleeding at and immediately after operation was estimated ‘by the medical staff (doctors and nurses), whereas postab‘ortal bleeding was interpreted from reports by patients. 7.1. BLEEDING AT AND IMMEDIATELY AFTER OPERATION We did not measure the immediate blood loss at abortion in the entire group of 4733 women, rather blood loss at operation was the subject of a special sub-study. 7.1.1. Estimation of blood loss on the basis of the treatment For most of the women in the study we estimated blood loss on the basis of need for uterotonic the- rapy and transfusions, as shown in Table 18. Routi- nely all women received 10 units of Syntocinon intravenously and in cases of unusual bleeding more was given. By the operators’ judgement, heavy bleeding occured in 237 women (5.0 °/o). Table 18: Treatment of bleeding at operation, by operative procedure Treatment D & C All VA Grand total NO. °/o NO. °/o NO. °/o Roufine Uterotonics 2300 94.0 2196 96.1 4496 95.0 Additional uterotonics 147 6.0 89 3.9 236 5.0 Transfusion 1 0.0 0 0.0 1 0.0 Subtotal 148 6.0 89 3.9 237 5.0 Total 2448 100.0 2285 100.0 4733 100.0 VA + curette VA + Treatment Nc\>/A onLy check curettage ' 0 NO. °/o NO. 0/0 Routine Uterotonics 607 98.4 1330 96.7 259 88.7 Additional uterotonics 10 1.6 46 3.3 33 11.3 Transfusion 0 0.0 0 0.0 0 0.0 Subtotal 10 1.6 46 3.3 33 11.3 Total VA 617 100.0 1376 100.0 292 100.0 BLOOD LOSS L. Andolsek and M. Oven Table 18 shows that there were more women who needed extra treatment in the D&C group than in the VA group (M2 = 11.05). The Highest percent of such cases appeared in the VA + curettage group (11.3 We). One woman received a blood transfusion. Since this patient was histologically verified to have had an endocervical pregnancy we cite it in detail. Case report Patient V. E., age 43 years, gravidity 10, para 8. Admitted on August 12, 1971 for abortion on social indications. LMP: June 15, 1971. On admission the operator found a 10 weeks preg- nancy. By the principle of rand-om allocation a D & C was performed under general anesthesia (Pen- tho‘tal). August 12, 1971: Operation; Length of uterus 12cm. Cervical canal dilated to Hegar 12.5. With curette and forceps an 80m fetus and pla- centa were removed. Bleeding did not stop despite administration of Syntocinon and Methergin, as well as 10 l. U. Syntocinon ’i.v. Tamponage of uterus and vagina done. Loss of blood 500 grams. Therapy: Haemacell plus blood transfusion. At 6.50 pm. the same day the packing was removed from the vagina and uterus, followed by 100 cc of fresh blood which coagulated immediately. The ute- rus continued to bleed, the BP dropped to 80/60, and the uterus became atonic. Accordingly we de- cided to reexamine the patient at 7.45 pm. Total blood loss estimated to be about 1000 ml. Repeat examination + curettage performed under anesthesia. Uterus length 13cm. The right isthmic portion of the uterine wall was slightly recessed. No perforation was found. We removed a walnutsized portion of placental tissue and coagulated blood. We placed two stitches on the right lateral side of the cervix and one stitch on the lower side, and packed the uterus, stopping the bleeding. We also packed the vagina. A Foley catheter was inserted. 17 August 13, 1971: Under general anesthesia the packs were removed. Since bleeding from the cervical canal continued, we decided to remove the uterus. At operation under endotracheal anesthesia we found about 250 cc of dark uncoagulated blood in the abdominal cavity. We performed a total extirpation of uterus with both adnexae without difficulty. The uterus was moderately enlarged. No lacerations were vi- sible. Total blood loss during operation was 400 cc. The operation was well taken by the patient. Pathology report: Uterus and both adnexae very enlarged, measuring 14 X 8 X 6.5 cm. The mucous membrane was erythroplastic. In the cervical canal was found a stitch. The cervix was dilated and fill- ed with coagulated blood. The cervical wall and isthmus were very thin and the canal was rough. Both ovaries were of normal size, the oviducts were dilated. No perforations were noted. Pathological diagnosis confirmed by histological examination: Giravid. cervicalis, residua post ab. Pla- centa increta. After the operation the patient received analgesics and antibiotics. In total she received 1400 c: A Rh D positive blood. Her post-operative course was uneventful and she was discharged on August 31, 1971. 7.1.2. Blood loss at and after surgery In a subgroup of 579 women pregnant up to 14 weeks we measured blood loss during surgery and for 48 hours after operation. This study was per- formed from April 14, 1971 to July 13, 1971. Final analyses were done in 532 women ,for whom all measurements of hemoglobin and hematocrit were performed. The interruptions were done with the same criteria and conditions as for the main group of 4733 women. Table 19: A subgroup of 532 patients, by age, prior pregnancies and their outcome, and by maior operative procedure Age in D & C VA Total years No. % No. % No. % 14 or less 0 0.0 0 0.0 0 0.0 15—19 4 1.5 16 6.1 20 3.8 20—24 72 26.6 60 23.0 132 24.8 25—29 48 17.7 71 27.2 119 22.4 30—34 83 30.6 49 18.8 132 24.8 35—39 28 10.3 45 17.2 73 13.7 40—44 33 12.2 14 5.4 47 8.8 45 and more 3 1.1 6 2.3 9 1.7 All ages 271 100.0 281 100.0 532 100.0 Prior pregnancies None 46 17.0 49 18.8 95 17.9 One or more births 222 81.9 205 78.5 427 80.3 One or more abortions 108 39.9 110 42.2 218 41.0 As can be seen in Table 19, the majority of women were in the age group 20—39 years (85.7 ”M; apprO- ximately 18 % were primigravida. The mean duration 18 of pregnancy was 8.7 weeks (range 4—14 weeks) and the mean number of prior pregnancies was 2.6 (confinements and abortions). Table 20: Pregnancy duration, by major operative procedure Pregnancy D & C VA Grand total age: k': No. 0/0 No. % No. 0/0 4 0 0.0 1 0.4 1 0.2 5 5 1.8 2 0.8 7 1.3 6 26 9.6 24 9.2 50 9.4 7 21 7.8 21 8.0 42 7.9 8 96 35.4 95 36.4 191 35.9 9 46 17.0 41 15.7 87 16.3 10 40 14.8 49 18.8 89 16.7 11 6 2.2 10 3.8 16 3.0 12 29 10.7 15 5.7 44 8.3 13 0 0.0 1 0.4 1 0.2 14 2 0.7 2 0.8 4 0.8 All preg. ages 271 100.0 261 100.0 532 100.0 The pregnancy age is tabulated in Table 20 by the operator’s estimation. The entire group of 532 women was divided into 4 subgroups: 271 (50.9 %) of D & C, 93 (17.5 °/o) of VA only, 144 (27.1 %) of VA + curette-check, and 24 (4.5 %) of VA —l— curettage cases. Prior to the ope- ration, blood samples were taken for hemoglobin and hematocrit. These tests were repeated 2 days after operation. We determined hemoglobin with the cyanmethemoglobin colorimetric method, and hema- tocrit by the micromethod. All material obtained from the uterus by D 8 C was collected through a special funnel into a container. Material obtained by VA was also collected in a special container. No liquids were added. After operation every woman was supplied with a package of sanitary napkins which she placed in a special bag after use. The following morning the nurse weighed all soiled napkins. We also used this method to calculate the quanttty‘a-of blood loss on the second day of hospitalization. 7.1.3. Blood loss by operative procedure in the following Tables we indicate the amount of blood loss in the course of operation, hospitaliza- tion, and the total, all according to the technique of interruption. Table 21: Blood loss at operation, by operative procedure Blood loss D & C Total VA Grand total in mi No. °/o N0. % No. % 00— 30 5 1.8 26 10.0 31 5.8 31— 60 60 22.1 64 24.5 124 23.3 61— 90 67 24.7 51 19.5 118 22.2 91—120 36 13.3 46 17.6 82 15.4 121—150 49 18.1 28 10.7 77 14.5 151—180 14 5.2 20 7.7 34 6.4 181—210 8 3.0 11 4.2 19 3.6 211—240 12 4.4 4 1.5 16 3.0 241—270 10 3.7 4 1.5 14 2.6 271—300 1 0.4 2 0.8 3 0.6 301 —330 4 1.5 1 0.4 5 0.9 331—360 2 0.7 0 0.0 2 0.4 361 and more 3 1.1 4 1.5 7 1.3 All patients 271 100.0 261 100.0 532 100.0 Blood loss VA + curette VA + VA + curette VA + in ml VA only check curettage {340:3 loss VA only check curettage NO. °/o NO. °/o N0. °/o NO. °/o NO. l’/o NO. °/o 00— 15 3 3.2 1 0.7 1 4.2 00— 30 7 7.5 19 13.2 0 0.0 16— 30 15 16.1 24 16.7 1 4.2 31— 60 23 24.7 36 25.0 5 20.8 31— 45 28 30-1 36 25-0 7 29-2 61_ 90 18 193 30 208 3 125 46— 60 23 24.7 44 30.6 6 25.0 61— 75 12 12.9 20 13.9 2 8.3 91—120 21 22.6 22 15.3 3 12.5 76— 90 5 5‘4 13 9'0 3 125 121—150 12 12.9 13 9.0 3 12.5 91_105 3 3,2 4 2.8 1 42 151——180 8 86 1o 53 2 83 106—-120 3 32 1 OJ 0 00 181—210 2 2.2 7 4.9 2 8.3 1:25:13?) (1) 3(1) (1) 3'3 ‘1’ 2'3 211—240 1 1-1 2 1-4 1 4-2 151 and more 0 0.0 0 0.0 2 8.3 241—270 0 0.0 2 1.4 2 8.3 . Al t t 3 100.0 144 1 .0 24 100.0 271—300 1 1.1 0 0.0 1 4.2 I pa 'e" s 9 00 301—330 0 0-0 0 0-0 1 4-2 Table 22 shows blood loss in the 48 hours after sur- 331—350 0 0-0 0 0‘0 0 ‘10 gery. Most patients lost up to 75 ml (D & C: 82.7 %; 361 and more 0 0-0 3 2-1 1 4-2 total VA: 85.4 °/o). Losses of more than 150 ml occu- red in only 0.8 %. The average losses for individual Total 93 100.0 144 100.0 24 100.0 groups are as follows: From Table 21 Wis—evident that blood loss with D&C and total VA ranged up to 150 ml (80.1% and 82.4 °/o), whereas in VA + curettage cases more than 40% lost more than 150 ml. Heavier bleeding than 300 ml is uncommon in D & C (3.3 %) and even less in total VA (1.9%), but 8.4% in VA + curet- tage. Blood loss during operation regardless of the method used was up to 150ml in more than 80 % and we had 14 patients (2.6 0/o) with blood loss of more than 300 ml. The average blood loss during operation in the in- dividual groups is as follows: VA +7 VA + D & C Total VA VA only curette . check curettage 116.35 ml 103.90 ml 95.38 ml 98.11 mi 171.67 ml The average blood loss at operation for VA only, was 18 ”/0 below that for D & C. Noteworthly is also the statistical difference in blood loss for D & C and total VA, at the level of P < 0.05. We also wish to emphasize the higher level in the VA + curettage group, significant at the level P < 0.01. Table 22: Blood loss during 48 hours hospitalization, by operative procedure Blood loss D & C Total VA Grand total in ml No. % No. % No. % 00— 15 9 3.3 5 1.9 14 2.6 16— 30 27 10.0 40 15.3 67 12.6 31— 45 54 19.9 71 27.2 125 23.5 46— 60 84 31.0 73 28.0 157 29.5 61— 75 50 18.5 34 13.0 84 15.8 76— 90 23 8.5 21 8.1 44 8.3 91—105 13 4.8 8 3.1 21 4.0 106—120 4 1.5 4 1.5 8 1.5 121—135 3 1.1 2 0.8 5 0.9 136—150 2 0.7 1 0.4 3 0.6 151 and more 2 0.7 2 0.8 4 0.8 All patients 271 100.0 261 100.0 532 100.0 VA-l- VA + D & C Total VA VA only curette check curettage 56.92 mi 52.86 mi 51.43 ml 51.65 mi 65.67 mi The average blood loss during the first 48 hours after operation for VA only was 9.7 % below that for D & C. The VA + curettage group showed significantly more blood loss. Table 23: Total blood loss, by operative Procedure Blood loss D & C Total VA Grand total in ml No. % No. % No. % 00— 30 0 0.0 0 0.0 0 0.0 31— 60 5 1.8 9 3.5 14 2.6 61— 90 23 8.5 42 16.1 65 12.2 91—120 43 15.9 52 19.9 95 17.9 121—150 62 22.9 49 18.8 111 20.9 151—180 41 15.1 40 15.3 81 15.2 181—210 39 14.4 24 9.2 63 11.8 211—240 14 5.2 16 6.1 30 5.6 241—270 11 4.1 8 3.1 19 3.6 271—300 9 3.3 7 2.7 16 3.0 301—330 10 3.7 5 1.9 15 2.8 331—360 5 1.8 2 0.8 7 1.3 361 and more 9 3.3 7 2.7 16 3.0 All patients 271 100.0 261 100.0 532 100.0 VA + curette VA + 120311 ICSS VA only check curettage " 'NO. lyo NO. 0/0 N0. 0/0 00— 30 0 0.0 0 0.0 0 0.0 31— 60 3 3.2 6 4.2 0 0.0 61— 90 12 12.9 29 20.1 1 4.2 91—120 17 18.3 32 22.2 3 12.5 121—150 23 24.7 20 13.9 6 25.0 151—180 18 19.3 19 13.2 3 12.5 181—210 9 9.7 14 9.7 1 4.2 211—240 6 6.5 10 6.9 0 0.0 241—270 2 2.2 4 2.8 2 8.3 271—300 1 1.1 3 2.1 3 12.5 301—330 2 2.2 2 1.4 1 4.2 331—360 0 0.0 2 1.4 0 0.0 361 and more 0 0.0 3 2.1 4 15.7 All patients 93 100.0 144 100.0 24 100.0 Total blood loss in the hospital (Table 23) was as much as 210 ml in most patients. Heavier losses than 360 ml occured only in a small number of cases 19 (D & C: 3.3 %; total VA: 2.7 °/o). Again VA + curet— tage showed more loss; we found 16.7 % cases with losses over 360 ml. The average total loss for individual groups is as follows: med within 10 minutes (D & C: 84.8%; total VA: 91.5 °/o). The longest-lasting operations were VA + curettage. In the D & C group, blood loss increased according to the operation time very markedly (regression coefficient: b = 13.99) whereas in total VA this increase is not so explicit (regression coef- VA + VA + ficient: b = 6.01). D & C Total VA VA only curette curettage check Table 25: Total average blood loss, by duration 173.26 ml 165.17 ml 146.81 ml 149.76 ml 237.33 ml of operation and by operative procedure Duration — The average total blood loss in the hospital for VA of oper. D & C Tmal VA Grand total only was 15.3% below that for 080. in min. Mean NO- Mean N°- Mean NO- The difference between D & C and total VA is signi- :15— 2 £38 2: 11212 4% 13;? 7g flcant at P <.0.01. The difference of the VA + curet- 5__ 6 143.9 88 1563 99 1505 187 tage groups is also Significant. 7— 8 172.4 57 143.8 59 157.9 116 9—1 0 179.6 55 179.7 32 179.7 87 7.1.4. BIOOd loss I) estation I a 11—12 247.9 19 123.1 14 194.9 33 ’ y 9 a ge 13—14 259.2 13 270.0 1 259.9 14 Table 24: Total average blood loss, by gestational age 15 and more 307-1 9 306-5 7 306.9 16 and Operahve Pf°cedwe All patients 173.3 271 156.8 261 169.7 532 Week of D & C Total VA Grand total PYGQ- Mean No. Mean No. Mean No. Duration of oper. VA only VA—lf—Icuiette VAtt+ 4 0.0 0 120.0 1 120.0 1 in min. C 9° cure age 5 99.8 5 57.5 2 87.7 7 Mean No. Mean No. Mean No. 6 130.8 26 113.8 24 122.7 50 1_ 2 155.0 1 120.0 1 0.0 0 7 124.1 21 112.7 21 118.4 42 3— 4 145.2 34 143.2 13 0.0 0 8 164.7 96 137.5 95 151.2 191 n 5— 6 151.5 39 159.2 56 160.7 4 9 164.8 46 151.6 41 158.6 07 - 7— 8 148.3 6 141.8 47 154.3 6 10 171.2 40 190.9 49 182.0 89 9—10 120.5 6 175.5 19 241.9 7 11 206.8 6 219.1 10 214.5 16 11—12 128.2 6 88.5 6 211.5 2 12 282.3 29 213.7 15 258.9 44 13—14 270.0 1 0.0 O 0.0 0 13 0.0 0 260.0 1 260.0 1 15 d 0 0 0 67 5 2 402 2 5 14 388.5 2 647.0 2 517.7 4 a” ”me - - - All patients 173.3 271 156.8 261 169.7 532 A” ”“9“ 146-8 93 149-8 144 237-3 2‘? 7 W k f VA l VA+curet-te VA+ We studied differences in blood losses with different mg: O my check curettage techniques with regard to the training of the ope- _" Mean NO- Mean N°- Mean NO- rator. Noteworthy differences occured in the D & C 4 0.0 o 0.0 ' 0 120,0 1 group where special-lists had less blood loss (107.8 ml) 5 0.0 0 57.5 2 0.0 0 in comparison with non—specialists (133.8 mi). 6 126.3 6 110.7 16 101.5 2 7 108.2 4 113.4 16 120.0 1 . . 8 145.4 45 126-9 45 162-8 5 7.1.6. Hemoglobin and. hematocrit level before 9 142.6 17 152.6 22 216.0 2 and after Operation 10 155.8 16 196.1 2" 244.5 8 11 207.7 3 244.0 9, 0.0 0 We used two laboratory methods to measure blood 12 190.0 2 211.6 11 248.5 2 loss by comparing hemoglobin and hematocrit val- 13 0-0 0 0-0 0 250-0 1 ues prior to and after operation. The average values 14 0'0 0 0'9- 0 6470 2 for hemoglobin in the four different operation types A” Patients 145-8 93 149-8 144 237-3 24 before and after the procedure are as follows: The association ‘between total blood loss and age of pregnancy is shown in the Table 24. Regression coefficients calculated for D & C and total VA show differences attributable to technique (22.09 ml for D & C and 15.59 ml for total VA). The difference is significant on the level of P<0.01. Furthermore it appears that blood loss increases with the age of pregnancy in all groups. We also noted the influence of the number of pre— vious pregnancies on blood loss. In 17.9 % of primi- glravidae the average blood loss was 180.2 mi where- as the average blood loss in multigravidae was 161.9 mi. 7.1.5. Blood loss by duration of operation and operator’s training Table 25 shows than the duration of the operation affected blood loss. Most operations were perfor— 20 VA+ VA + D & C VA only curette check curettage Before: 12.65 12.80 12.62 12.27 After: 12.46 12.72 12.48 11.91 The differences between the average values of he- moglobin before and after operation are significant in the D & C group and the VA + curettage group. VA only showed the smallest difference. The average values for hematocrit in the four different operation types before and after procedure are as follows: VA+ VA+ D & C VA only (2:11:38? curettage Before: 39.84 39.83 39.76 38.56 After: 39.13 39.79 39.17 38.10 A significant differences between values of hemato- crit before and after operation can be established only in the D & C group again, VA only showed the smallest difference. 7.1.7. Blood loss and recurettage In our group of 4733 women we registered 14 cases of hemorrhage which did not stop after the proce- dure. In all cases recurettage was done and diffe- rent quantities of placental tissue found. Eight cases were registered in the D & C group and the remai- ning 6 cases were as follows: VA only: 1 case; VA + curette-check: 4 cases; and VA + curettage: 1 case. In addition we found one primigravida with a duplex uterus. At the first procedure, the non pregnant side was evacuated. Curettage of the preg- nant side was done three days later. 7.2. POSTABORTAL BLEEDING 7.2.1. Bleeding after discharge Discharge bleeding is usually not a serious problem with first trimester abortions. However the incidence varies according to the way it is reported. We asked about the appearance of any — even the slightest bleeding, occurring from the time of discharge until the first regular men- struation. Table 26 shows the response. Table 26: Bleeding after discharge in days, by operative There is a question of how to interpret bleeding during the first three weeks. It could be simply post- abortal flow, or a sign of some pathological process such as endometritis or retained placenta, or ano- vulatory bleeding. Actually we noted that 9.2% of the women reported having had their first menstru- ation after abortion within 3 weeks after discharge and this percentage does not differ according to technique. 7.2.2. Treatment of bleeding Table 28: Bleeding after discharge requiring ambulatory treatment, by operative procedure D & C Total VA Grand total Treatment NO. 0/o NO. 0/0 NO. 0/0 No 2172 90.0 2037 90.4 4209 90.2 Yes 242 10.0 216 9.6 458 9.8 All patients 2414 100.0 2253 100.0 4667 100.0 VA + cu rette VA + Treatment VA only check curettage No. % No. % No. % No 559 91.6 1221 90.0 257 89.5 Yes 51 8.4 135 10.0 30 10.5 All patients 610 100.0 1356 100.0 287 100.0 In Table 28 we consider bleeding for which women had to seek a physician’s help or were hospitalized. The differences among techniques are not signifi- cant. The percentage for all types varies between 8.4 and 10.5 °/o. Also, from the records we ascer- tained that the bleeding was light in character. pr°°°dure We also studied the hospital records and found that Da 3 D & C Total VA Grand total 79 Women (1.7 %) were hospitalized. These women y No. % No. % No. % belonged to the following groups: No 639 26.5 665 29.5 1304 27.9 VA + 1—9 1122 46.5 955 42.4 2077 44.5 VA 10 and more 653 27.0 633 28.1 1286 27.6 D & C Total VA VA only ccuhrett: “Henge ec All patients 2414 100.0 2253 100.0 4667 100.0 45 (19%) 34 (15%) 11 (1.8%) 14 (1.0%) 9 (3.1%) It can be said that thereis no difference between Days VA only VAgggflette chQttige the D & C group and the total VA group (if = 0.67), No. % No. % No. % nor betv;t2ae_n1t27e D & C and the VA + curettage No 199 32.6 390 28.8 76 26.5 groups ( 7 ' )' 1—9 246 40.3 587 43.3 122 42.5 . . . Table 29: Bleeding due to retained placental tissue and 10 and more 165 27'1 . 379 28'0 89 31'0 bleeding without placental tissue, by operative All patients 610 100.0 1356 100.0 287 100.0 procedure We have correlated the quantity of all blood lost at Bleeding N2. 8‘ (02/0 (135%: fig“ tg/Eal abortion With the occurrence of bleeding after dis- Retained tissue 31 1'3 20 0.9 51 1.1 charge for all women who returned the first ques- N0 tissue 14 0.6 14 0'6 28 05 ttionnaire (526). We recorded postdischarge blee- Total 2414 2253 4667 dung regardless of duration and 1intens1ty. Such bleeding did not appear to be related to bleeding . . . . . VA during operation and hospitalization (%2 = 3.17). Bleeding VA only ctggfiette chQttit—ge NO. o/o NO. 0/0 N0. 0/0 Table 27: Bleeding after discharge, by quantity of blood R . . etained tissue 5 0.8 8 0.6 7 2.4 '°ss at ab°'“°" No tissue 6 1.0 6 0.4 2 0.7 Bleeding T | 151_ 301 m] ota 610 1356 287 agiiharge -150 ml —300 ml and more Total Table 29 shows a higher percent of retained tissue figs 133 133, 9; $33 after D & c (1.3 %) than after VA (0.9 %), but the difference is not significant (*2 = 1.01). The VA + Total 285 206 35 526 curettage group showed the highest level (2.4 04:). 21 Chapter 8: Infection was studied in three groups: 1. in the entire group of 4733 women we establi- shed the incidence of infection during hospitali- zation by monitoring temperature, abdominal plains, tendernes, and the need for drug treat- ment. 2. In a subgroup of 532 women, we took smears for vaginal flora and bacteriology before opera- tion in order to establish if the presence of abnor- mal bacteria influenced the incidence of infec- tion. 3. We analyzed the records of all rehospitalized wo- men for infection. 8.1. EVENTS DURING FIRST AND SECOND DAY OF HOSPITALIZATION In the 4733 women, we classified temperature data int-o afebrile, subfe‘brile and febrile cases. We also specified the day of hospitalization when the women had abdominal pains and tenderness, and the admi- nistration of antibiotics and uterotonics. From Table 30, it is eVirdent that more febrile cases, more abdo~ minal pains, and more tenderness were registered for all patients on the second day than the first day. These symptoms appeared on the first and on the second day more frequently in the combined group. The D&C and VA only groups are essentially si- milar. Table 30: Events during the first and second day of hospitalization for 4733 women. by operative procedure Event D & C Total VA Grand total N0. 0/0 N0. 0/0 NO. 0/0 Subfebrile 674 27.5 608 26.6 1282 27.1 Febrile 43 1.8 36 1.6 79 1.7 Abd. pain 89 3.6 94 4.1 183 3.9 Tenderness 94 3.8 112 4.9 206 4.4 Antibiotics 100 4.1 86 3.8 186 3.9 Uterotonics 113 4.6 89 3.9 202 4.3 Ant. + Uterot. 27 1.1 36 1.6 63 1.3 No. 1St day 3 0.1 7 0.3 10 0.2 22 INFECTION L. Andolsek VA+curette VA+ Event VA only check curettage NO. 0/0 N0. °/o NO. °/o Subfebrile 141 22.9 388 28.2 79 27.1 Febrile 10 1.6 20 1.5 6 2.1 Abd. pain 18 2.9 61 4.4 15 5.1 Tenderness 20 3.2 76 5.5 16 5.5 Antibiotics 19 3.1 49 3.6 18 6.2 Uterotonics 21 3.4 57 4.1 11 3.8 Ant. + Uterot. 5 0.8 24 1.7 7 2.4 No.1st day 2 0.3 4 0.3 1 0.3 Second day Event D & C Total VA Grand total NO. 13/0 NO. ”/0 NO. °/o Subfebrile 439 17.9 392 17.2 831 17.6 Febrile 61 2.5 52 2.3 113 2.4 Abd. pain 114 4.7 101 4.4 215 4.5 Tenderness 142 5.8 125 5.5 267 5.6 Antibiotics 163 6.7 152 6.7 315 6.7 Uterotonics 67 2.7 40 1.8 107 2.3 Ant. + Uterot. 50 2.0 58 2.5 108 2.3 No.2nd day 117 4.8 107 4.7 224 4.7 VA+curette VA+ Event VA only check curettage NO. c'/o N0. 0/0 NO. 0/0 Subfebrile 100 16.2 237 17.2 55 18.8 Febrile 9 1.5 28 2.0 15 5.1 Abd. pain 20 3.2 65 4.7 16 5.5 Tenderness 32 5.2 73 5.3 20 6.9 Antibiotics 37 6.0 84 6.1 31 10.6 Uterotonics 17 2.8 22 1.6 1 0.3 Ant. + Uterot. 13 2.1 34 2.5 11 3.8 No.2nd day 23 3.7 73 5.3 11 3.8 We cannot consider all cases presented in Table 30 as infections. In surveying the individual records we established that unequivocal infection is compara- tively rare. Pelvic infection is defined for the ana- lysis as the presence of one or more of the follow- ing symptoms: temperature 37.600 or higher for at least two days, pains in the abdomen, heavy blee- ding, and tenderness in the adnexal regions. These events are shown in Table 31. We registered 35 such cases (0.7 °/o). It is evident that pelvic infections and fever are most frequent in the VA and curettage group. Table 31: Complications during hospitalization, by operative procedure Event D & C Total VA Grand total NO. °/o NO. 0/0 N0. 0/0 Pelvic infection 17 0.7 18 0.8 35 0.8 Fever only 19 0.8 20 0.9 39 0.8 Heavy bleeding & infection 4 0.1 7 0.3 11 0.2 Other 9 0.4 10 0.4 19 0.4 No events 2399 98.0 2230 97.6 4629 97.8 Total 2448 100.0 2285 100.0 4733 100.0 VA+curette VA+ Event VA only check curettage N0. °/o NO. °/o NO. °/o Pelvic infection 4 0.7 9 0.6 5 1.7 Fever only 4 0.7 12 0.9 4 1.4 Heavy bleeding & infection 2 0.3 2 0.1 3 1.0 Other 2 0.3 5 0.4 3 1.0 No events 605 98.0 1348 98.0 277 94.9 Total 617 100.0 1376 100.0 292 100.0 Notes on Table 31: — The pelvic infection group (35 cases: 0.7 %) includes endometritis and salpingitis. We did not observe para- metritis, peritonitis or septicemia. —- Fever only (39 cases: 0.8 %) includes women who de- monstrated a rise of temperature to 37.60 C or more for two or more days. Some of the cases probably belong to the pelvic infection group, although we did not always find symptoms to confirm this diagnosis. Some of the women complained of headaches, cough, and pains in the lower back. All these symptoms disap- peared with or without specific therapy within a few days and without an established diagnosis. — Heavy bleeding (11 cases: 0.2 %) includes women with retained placental tissue plus fever (10 cases). In one case, a cervical pregnancy (reported elsewhere) was associated with fever and was hospitalized. — The "other“ category includes 19 cases (0.4 %) of mis- cellaneous complications not included in other cate- gories. In 10 cases the fever was a symptom of other events: (two cases of phlebitis of the upper extremity after i.v. anesthesia, four cases of acute pharyngitis, two of influenza, and one each of herpes simplex and tonsillitis). The remaining 9 cases involved infections of the urinary tract. 8.2. INFLUENCE OF ABNORMAL VAGINAL FLORA AND BACTERIA ON INCIDENCE OF INFECTION In the subgroup of 532 women, two cervical smears were taken immediately before abortion, for direct microscopic study (Giemsa stain) and bacteriolo- gical culture. The microscopical findings were classified after Jellinek into six groups: |. Normal flora with Doderlein bacilli ll. Normal flora with Doderlein bacilli and a few leucocytes lll. Purulent bacteria and a large number of leuco- cytes IV. Presence of diplococcus (gonorrhea) V. Presence of trichomonas vaginalis Vl. Presence of monilia albicans The culture findings were divided into the following eleven groups: 1. Sterile 2. Mixed saprophytic flora 3. Streptococci 4. Staph. aureus 5. Lactobacilli 6. Fungi of Candida origin 7. Staph. aureus — streptococci 8. Staph. albus — streptococci 9. E. coli 10. Proteus mirabilis, klebsiella, enterobacteriaceae 11. Mixed (Gram neg. bacilli — staphilococci) Table 32: Hospitalization events, by vaginal flora as classified by Jellinek Classification I II III V/VI ll/VI III/V III/VI TotaI Pelvic infection 1 1 1 0 0 0 0 3 Event Fever only 1 1 2 0 0 O 4 Other 1 1 1 0 0 0 3 No events 6 522 Total (No.) 188 184 127 10 13 6 532 35.3 34.6 23.9 0 1.9 2.4 13100.0 0 O 185 181 123 4 10 13 4 .7 Total [o/o] As seen in Table 32, 372 women (69.9 %) had nor— mal vaginal flora (classes I and II) before abortion; 30.1 % had pathological vaginal flora. Of these, 127 (23.9 %) had flora Ill, and the rest had mixed infec- tions. There was no evident influence of vaginal flora on the appearance of infection and fever. None of these patients received antibiotics before ope- ration. From the Table it is evident that 4 women (40 %) of those having any kind of febrile symptoms had vaginal flora lll but only one showed symptoms of infection. Table 33: Events during hospitalization, by microbiological status Classification Event 2 3 4 5 6 7 8 9 10 11 Total Pelvic in- fection 2 1 3 Fever only 1 2 1 4 No events 44 271 20 8 125 8 2 6 30 5 6 525 Total 45 275 20 8 126 8 2 6 30 5 7 532 From Table 33 it is evident that half of the women (51 %) had mixed saprophytic flora on the cervix. in all 3 women developed pelvic infection, 4 wo- men developed fever, and there were 3 women with infections of the urinary tract. All infections were minor and responded to antibiotic therapy. None of these women returned after discharge. 8.3. SYMPTOMS OF INFECTION REPORTED AFTER DISCHARGE All women in the study were asked to indicate fever and other symptoms on the 6 week questionnaire. This enabled us to consider the possibility of infe- ction. We realize of course that such data are sub- jective. 23 Table 34: Reported fever, by ordinal day of onset after discharge and by operative procedure Fever onset in days D & C Total VA Grand total after discharge No. % No. % No. % 1— 3 355 14.7 327 14.5 682 14.6 4— 6 134 5.6 129 5.7 263 5.6 7— 9 43 1.8 40 1.8 83 1.8 10—1 2 24 1.0 26 1.2 50 1.1 13—15 16 0.7 16 0.7 32 0.7 16 and more 16 0.7 8 0.4 24 0.5 of D&C cases and in 0.6% of all VA cases. It appears that subsequent hospitalization because of pelvic infection is more frequent with D & C than with VA cases (9%2 = 4.17). Aside from bleeding and fever without other symp- toms, the women reported abdominal pain and pain in the lower back. Almost half reported abdominal pain, especially the first six days after abortion as seen in Table 36. Such pains caused 7.2 % of wo— men to seek medical treatment in the out-patient department. No significant differences were esta- Total no . , fever 1826 75.6 1707 75.8 3533 75.7 “'3th according '00 m‘et'h‘Od- Table 36: Outpatient and hospital treatment of abdominal Fever onset VA+curette VA pain, by operative procedure + in days VA only check curettage after Treatment D & C Total VA Grand total discharge No. % No. % No. % No. % No. % No. % 1— 3 81 13.3 194 14.3 52 18.1 Abd. pains 4— 6 30 4.9 81 6.0 18 6.3 reported 1095 45.4 996 44.2 2091 44.8 7— 9 10 1.6 27 2.0 3 1.1 Sought 10—12 8 1.3 15 1.1 3 1.1 treatment 174 7.2 162 7.2 336 7.2 13—15 5 0.8 9 0.7 2 0.7 Treated 2 0.1 5 0.2 7 0.3 16 d o e 1 0.2 5 0.4 2 0.7 a" m r Total 2414 2253 4667 Total no *— fever 475 77.9 1025 75.5 207 72.1 VA + curette VA + VA only Out of 24.3 % of women reporting fever, only 17.4 % Treatment No. % NgheCIE/o figtfitagf sought aid in the out-patient department. We noted Abd . no differences by the operative technique. In the -PaInS r time between discharge until the first menstrual pe- geopuogrt‘ed 250 41-0 501 44-3 145 00.5 riod, we noted complications in 48 women (1.0 °/o) treatment 34 55 105 7_7 23 8.0 of all women that returned the questionnaire. Treated 1 0.2 4 0.3 0 0.0 Total 610 1356 287 Table 35: Infection and fever requiring rehospitalization, by operative procedure D & C Total VA Grand total Event NO. 1’/o No. °/o NO. °/o Pelvic infection 30 1.2 14 0.6 44 0.9 Fever only 2 0.1 2 0.1 4 0.1 No event 2382 98.7 2237 99.3 4619 99.0 Total 2414 100.0 2253 100.0 4667 100.0 VA + curette VA + Event VA only check curettage NO. 0/0 NO. l1/1) N0. 0/0 Pelvic infection 1 0.2 11 0.8 2 0.7 Fever only 0 0.0 2 0.2 0 0.0 No event 609 99.8 1343 99.0 285 99.3 Total 610 100.0 1356 100.0 287 100.0 In evaluating the results‘ “shown in Table 35, we established that pelvic infections appeared in 1.2 ”/0 24 Pain in the lower back is difficult to interpret be- cause of its frequency and non-specificity. If we compare the appearance of such pain before and after operation, we ascertain that about 33 % of wo- men reported ~pain before surgery. Of those women who were asked about pain after the operation, a little over 40 % reported pain in the lower back. The pain appeared in the following proportions, according to the method: VA + curette VA + D 8‘ C VA only check curettage 40.3 % 41.1 % 41.1 % 42.6 °/o The results of this portion of the study confirm our experience that one often cannot make conclusions about the frequency and intensity of complications on the basis of patient's reports. Chapter 9: Women are immunized when they carry an Rh posi- tive infant because the erythrocytes of the fetus can pass to the mother during pregnancy. Rh positive infants of subsequent pregnancies will demonstrate signs of morbus "'haémoliticus neonatorum (MHN). Surgical abortion can also induce immunization since passage of fetal erythrocytes through the pla- centa is facilitated by the operation. Rh D agglutinogens in the fetus develop very early. Bergstrom (2) demonstrated it in a fetus of only 38 days and Chown (3) in two fetuses of only 32 mm and 59 mm length. The fact that erythrocytes pass from the fetus to the mother was proved by iden- tifying fetal hemoglobin in erythrocytes in the ma- ternal circulation. It had once been thOught that the appearance of fetal erythrocytes in the maternal circulation is not always followed by anti Rh immu— nization but, on the contrary, immunization is found even without the presence of fetal erythrocytes (4). 9.1. GOAL AND STUDY STRATEGY As noted in Chapter 2, all patients in our study admitted for abortion underwent blood testing, in- cluding the Rh factor. The purposes of substudy were to ascertain if there are correlations between anti-Rh immunization and the procedure, whether the method is a factor, and if there are any corre- lations to the month of pregnancy, to parity, and to the woman’s prior immunization status. We did not ascertain the Rh status of the women’s sexual partners but in our patient population 82% of the men are Rh D positive, either ‘in homozygosis or heterozygosis. Therefore, we calculate that 60 % of the Rh negative women in our population deliver Rh positive infants. Blood was drawn from all Rh negative women be- fore the abortion and again approximately three months post-operatively. The serum was tested 24 hours after being drawn with erythrocytes of the genotype CDe/cde (Rrr), cDE/cde (Rzl'), CDe/cDE (RIRZ), and cde/cde (rr). We used one stage and two stage techniques with papain (5,6) and the in- direct Coombs test. The papain preparations were read separately by two technicians as a confirming procedure. In positive cases, analysis on cell panels and titration procedures were also carried out. RHESUS FACTOR IMMUNIZATION M. Hoéevar and L. Glonar 9.2. IMMUNIZATION OF RH NEGATIVE WOMEN BEFORE ABORTION In this study 789 Rh D negative women were exa- mined during the two years study period. Ta-ble 37 shows the number of these women who demon- strated anti-Rh agglutinins before the operative pro- cedure. Forty-one of these women were found to be immunized. We believe that the history of previous abortions in our patient population is not always reliable since many women are reluctant to admit to a previous abortion. Another uncertain factor is the fact that we do not know the Rh genotype of children previously born. However, in spite of these limitations, the available statistical evidence is strong enough to suggest stimulation of immunization by both prior delivery and induced abortion. Table 37: Immunization of Rh negative women before abortion, by prior delivery and abortion C U, U, 5 Anti Rh agglu- "8’8 93 .9 g a E tinins before 0 .t‘ o .3 L . C “'1 ,_ a: ,_ 'z n 0 abortion ; 3 ,__ <9 2% .95 Egg—#3898 E g 5; g E ‘5 absent present \0 E 8. '5. 0 0 163 161 2 1.2 0 1 or more 24 23 1 4.2 1 0 137 134 3 2.2 1 1 or more 68 63 5 7.3 2 0 132 123 9 6.8 2 1 or more 166 154 12 7.2 3 or more 0 36 33 3 8.3 3 or more 1 or more 63 57 6 9.5 TOTAL 789 748 41 5.2 One hundred and sixty three of our patients were pregnant for the first time. Two of them showed anti Rh agglutinins and gave no history of prior blood transfusions. One of these patients was aged 15 and the other over 20. We can only suppose that in these cases erythrocytes passed to the mother before the interruption of pregnancy sometime be- tween the second and third month and resulted in the formation of antibodies. 25 The second group is 24 women, none of whom had previously delivered but all of whom had had one or more prior abortions. One of these women was shown to be sensitized before the procedure and later we found she had had four spontaneous abor- tions. In the other groups the percentage of sensitization increased progressively so that the more the women had had deliveries or abortions, the more likely they were to be sensitized. In the women with one prior delivery, immunization was detected in only 2.2 %, When such women had also had abortions, the per- centage increased to 7.3 %. In the women with two deliveries only, immunization occured in 6.8%, but when combined with abortions this figure became 7.2%. There is an even bigger difference in the group with three or more deliveries: those with de— liveries only were sensitized in 8.3 % whereas those with abortions in addition demonstrated a sensiti— zation rate of 9.5%. We conclude from this that abortion clearly stimulates immunization. 9.3. IMMUNIZATION OF RH NEGATIVE WOMEN AFTER ABORTION 9.3.1. Immunization by prior delivery and abortion Table 38 is an analysis of the 748 women who show- ed no anti-Rh agglutinins before the abortion. In 161 of those with no prior deliveries, immunization was found three months after abortion in 7 cases (4.3%). The risk of immunization to this group is tempered by the fact that only 82 % of men are Rh positive in our population and as noted above, we calculate that 60 % of Rh negative women will have an Rh D positive infant. Therefore the risk of immu— nization from abortion with an Rh positive fetus is approximately 6 °/o, less than the risk of immuniza- tion after the first delivery, which has been calcu— lated to range between 8.2 % and 17 04: (4,7). The same percentage of immunization was found in the second group, where 23 women declared that they have had abortions before. Table 38: Development of anti-Rh agglutinins three months after abortion in women with no agglutinins prior to the operative procedure, by prior delivery and abortion C m .,, c Anti Rh agglu- “£3 5‘5 g 23 2 tinins 3 months g-E . g .E D 0 after abortion ._ :: 0= 2 o E a —— o s 11:. 8 E a 5 “5 absent present o\° § 0 0 161 154 7 4.3 0 1 or more 23 22 1 4.3 1 0 134 130 4 3.0 1 1 or more 63 61 2 3.2 2 0 123 115 8 6.5 2 1 or more 154 148 6 3.9 3 or more 0 33 28 5 15.1 3 or more 1 or more 57 56 1 1.7 TOTAL 748 714 34 4.5 In the 197 women with one previous delivery, the percentage immunized after abortion is lower than 26 in the first group: only 3.0 °/o—3.2 %. Women who had two deliveries developed agglutinins in 3.9 % or 6.5 % of the cases and in the groups with three or more deliveries, 1.7 % or 15.1 % developed anti Rh agglutinins. We speculate that some of our patients were immunized at previous pregnancies but that the agglutinins disappeared. Before the abortion, none of the 748 women demonstrated antibodies but as the Table shows, the women with the greater number of prior pregnancies were more likely to demonstrate anti-Rh agglutinins after the abortion, thus providing a case for the hypothesis that prior pregnancies make women more likely to develop anti-Rh agglutinins during the pregnancy under study. 9.3.2. Immunization by gestational age Table 39 displays the 748 cases with no prior agglu- tinins according to the month in which the preg- nancy was interrupted. As can be seen, abortion was carried out in the first month in only 13 cases and no immunization was observed; the sample taken alone is too small to draw any conclusion. Wo- men who had abortions in the second and third month showed an immunization rate of 4.5% or 4.8%. These data could suggest, that interruption of pregnancy in the second or third month does not differ essentially in iRh immunization. Table 39: Induction of anti-Rh agglutinins in women with no antibodies prior to the operative procedure, by gestational age at pregnancy termination . Anti Rh agglutinins Gestational three months 0 age at Number of after abortion /0 of women pregnancy women immunized termination absent present 1st month (4 week] 13 13 0 0.0 2nd month (5, 6,7,8 462 441 21 4.5 week] 3rd month (9,10,11,12 273 260 13 4.8 week) TOTAL 748 714 34 4.5 9.3.3. Immunization by operative procedure Table 40 relates the induction of immunization in 714 patients with no prior agglutinins according to the operative procedure selected. Immunization was detected most frequently after abortion by VA (7.7 %), less after D&C (4.2 %), and least (2.5 0/o) in cases where suction was followed by currettage. Table 40: Development of anti-Rh antibodies three months after abortion, by operative procedure Anti Rh agglutinins three Operative .‘ "/0 of women procedure atggégs after 33:23? immunized D & C 364 16 4.2 VA only 155 13 7.7 VA + D & C 195 5 2.5 TOTAL (748 W) 714 34 4.5 9.4. ALTERATION IN TITER OF ANTI-RH AGGLUTININS IN ALREADY IMMUNIZED WOMEN Table 41 analyzes the 41 women who demonstrated anti Rh titer before abortion according to the trend of the titer following the procedure. It can be noted that in almost half (41.5 %) there was no change in titer, whereas, in 24.4 % and 34.1 0/o respectively, there was either an increase or decrease. Table 41: Alteration in titer of anti-Rh agglutinins after abortion in already-immunized women Anti Rh agglutinins three Number of months after abortion women % 0f total Titer increased 10 24.4 Titer unchanged 17 41.5 Titer decreased 14 34.1 TOTAL 41 100.0 9.5. RH IMMUNIZATION TWO YEARS AFTER ABORTION We were able to test the antibodies of some of our patients 2 years after the study. These cases are shown in Table 42 and 43 according to whether or not the study abortion was found to be im- munizing. Table 42 shows that none of the 372 who were not immunized originally were immunized 2 years later, regardless of their obstetrical history. Table 43 shows that the antibodies disappeared in 13 out of 19 patients, again regardless of the obstet‘ rical history. The above data do not provide information concern— ing the occurrence of MHN at the following deli— very, since the time of observation (two years) was too short, and there was only one woman who had delivered an Rh-positive child (Table 43) and had Rh agglutinins after abortion. it may be stated only that in most cases it is not possible to detect immunization after two years, because the anti Rh- agglutinins disappear. Table 42: Anti-Rh agglutinins in women two years after a non-immunizing abortion, by pregnancy status Number Anti-Rh agglutinins of wo- at time of test :/°t°|f o a men absent present No pregnancies 302 302 0 81.2 Pregnant at time of test 16 16 0 4.3 Delivered 25 *25 0 6.7 [12 Rh- neg. chil- dren] Aborted 29 29 0 7.8 100.0 TOTAL 372 372 0 “ Inocculated with Immunoglobulin Anti-D after delivery. Table 43: Anti-Rh agglutinins in women two years after an immunizing abortion, by pregnancy status Number . . . of wo- Anti-Rh agglutinins at men time of test immuni- present absent zed No pregnancies 16 3' 13 Pregnant at time of test 1 0 1 Delivered 2 0 2 (1 child Rh neg.) Aborted 0 0 0 TOTAL 19 3 16 ‘ Women with 2 or 3 deliveries in their history. We conclude from this study that the appearance of Rh antibodies after abortion and alterations of the titer in those previously immunized indicates that induced abortion stimulates immunization. This observation confirms that of other investigators and strengthens the recommendation that treatment of Rh negative women with immunolglobulin is de« sirable. However, this treatment may fail in those cases where women have been sensitized already. We also conclude that the likelihood of immuniza- tion is somewhat lower in women who have had previous deliveries but did not develop anti-Rh agglutinins, and that the appearance of agglutinins is not influenced by the duration of pregnancy be- yond the first month nor statistically significant by the operative procedure. 27 Chapter 10: The causes of incompetence of the internal 03 of the uterine cervix (CII) are considered to, be congenital, traumatic and functional. The congenital form is found in infantile and malformed uteri, particulary those of the arcuate and bicornuate types. In these cases, irregularity of the canal is shown by hystero— graphy where one sees uterine deformities and the so-cal’led tunnel cervix. Less frequently the uterine cavity is in the form of a sack. The abnormality may also be detected by use of the Palmer test where the internal 03 will pass Hegar dilators of size 8 or more. The second and most common cause is traumatic. In this type, the hysterogram shows a typical picture of »direct insufficiency<< and again the Palmer test is positive. The third form of incompetence is functional and is seen only at the time of pregnancy and simply means that the internal os of the cervix looses its integrity once uterine contractions have begun. The purpose of this substudy was to see how often CII occurs after surgical abortion and what factors may be involved. Our patients were seen during the third month after the abortion and examined in the second half of the menstrual cycle when the inter- nal os is considered to be most constricted. Our patients were those who had never delivered at or near term, and said that they had had no more than one abortion. in all we asked 924 patients to parti- cipate and 275 accepted our invitation (29.8 °/o). 10.1. TESTS APPLIED FOR DIAGNOSIS OF CII To test for the existence of CH we used four tests in all 275 patients (except for the Palmer test in which we tested 272*): the Palmer test, Bergman- -S‘ve>nnerunrd test, and cervical shape and cervical measure by hysterography. 1. The Palmer test is used to determine the width of the cervical canal by means of insertion of Hegar dilators. The largest dilator which can be inserted without resistance is considered to be a measure of the width of the cervical canal in milimeters. In our clinic we use the »inverted« ' The Palmer test was made on all 275 women but we failed to include the data for 3 women. 28 CERVICO ISTHMIC INCOMPETENCE B. Kraij and V. Lavrit': Palmer test which means that we begin the test with a dilator bigger than the anticipated width of the cervical canal and test with increasingly smaller dilators. in this way we can detect which dilator slips into the cervical canal without resi— stance. This reverse method is used to prevent the unwanted dilation of the cervical canal which can happen with the regular Palmer test. We take Hegar 8 to be the normal width of the cervical canal. 2. We also use the traction test of Bergman-Svenne- rund. This technique involves insertion of a Foley catheter size No.16 into the uterine cavity, filling the balloon with 1 mi of physiological saline, and attempting to extract the catheter from the ute- rus using an instrument which measures the force necessary for such extraction. According to Berg- man-Svennerund standards, CII exists if less than 600 grams are needed to extract the catheter. 3. We used hysterography to ascertain cervical shape and measure. We use high viscosity Sal- pix which has been warmed before insertion and which contains {no bubbles. No cannula is used in order not to interfere with the clarity of the pho- tograph. We use only two to three milliliter of solution, sufficient to fill the cervix and the ute- rine cavity but not the oviducts. 10.2. INCIDENCE OF C" AFTER THE OPERATIVE PROCEDURE As shown in Table 44, we found that there are great differences among these four techniques. According to the Palmer test, only five out of the 272 patients had Oil (1.8 °/o), but the Bergman-S‘venneru'nd test was positive in 160 women (58.2 0/0). Table 44: Incidence of Oil by type of test Women with Oil by Tests the criteria described NO. 0/0 1. Palmer test (272 patients) 5 1.8 2. Bergman-Svennerund test 160 58.2 3. Cervical shape 24 8.7 4. Cervical measure 57 20.7 Performing the Bergman-S‘vennerund test may create several opportunities for error. For instance, one must be careful that there is no air in the balloon. Very important is also the technique used to extract the catheter. The limit of 6009 needed for extrac- tion may be to high. We think that the Palmer test and the assessment of the shape of the cervix are the most valuable test for CH. These tests show an excellent correlation. The data of our other study on women with healthy cervices and on women with proved Cll speak for these two tests as :being the best (8). Despite these differences in the tests, we have no doubt that abortion may induce Cll. A female popu- lation with no abortion shows Cll in only three out of 1.000 women examined (9). 10.3. FACTORS AFFECTING Cll AFTER THE OPERATIVE PROCEDURE The purpose of the rest of our substudy was to examine which factors influence the appearance of CH after abortion. Our goal was to identify those factors with the hope that we might be able to elimi- nate or minimize them during the abortion proce- dure. Our wonk was organized according to the following questions: 10.3.1. C" by gestational age The occurrence of CH after abortion done not later than 8 weeks was compared to those done between 9 and 12 weeks. As can be noted in Table 45, the differences between the two time peniods were not significant and we conclude that the time of abortion has no demonstrable effect on the subsequent oc- currence of Oil. Table 45: Incidence of C", by duration of pregnancy and type of test Palmer test Bergman-Svenne- rund test Cervical shape Cervical measure 5 5 5 c E _ _ E _ _ E _ _ GE) Duration of 0-0 5 a °-o (3 5 0'0 a 5 0'5 = = O 0 Pregnancy 5‘5 s :3 :3 s s 53 s :E E .“=_’ s s o (E5 E E o E E E ° E E E E .E, E E tr 5 it s i ‘ii a ti s 1 ‘li a it s , 1t 3 1i s Up to 8 weeks 79 1 13 79 52 65.8 79 5 7.5 L 79 1‘8“— 22.8 9 to 12 weeks 181 4 2.2 183 99 44.8 183 17 9.3 183 38 20.8 All durations 260 5 1.9 262 151 57.6 262 23 8.8 262 56 21.4 Note: Palmer test: 12 examined women were pregnant over 12 weeks and are not considered in analysis. Bergman-Sven- nerund test, cervical shape and cervical measure: the number of examined than in the group with Palmer test. The Palmer test was made on all 275 women, but we failed to include the data for 3 women (data of 2 women pregnant up to 12 weeks were lost, the third woman with lost data was pregnant beyond 12 weeks). 10.3.2. Cll by dilatation In examining this question, we divided the 275 wo- men into three groups: those requiring 9mm dila- tion or less, those requiring 9.5 to 10.5 mm dilation, and those requiring 11 mm or more. As shown in Table 46, there was no difference in subsequent Cll found among these three groups. women in these groups is by 2 higher 10.3.3. Oil by shape of cervical canal In considering this question, we decided to calculate only whether there is any difference in subse— quent outcome between cervices which are »short« or »long«.* We postulated that long cervices may be more difficult to dilate and therefore, subject to greater trauma and subsequent Cll. As will be noted in Table 47, whether the cervix was short or long there was no relationship to the subsequent occur- rence of CH. Table 46: Incidence of Oil, by degree of cervical dilatation Palmer test Bergman-Svenne- Cervical shape Cervical measure rund test : C C l I: C 15° 5 = E 'c 5 = “2’ 'c 5 = E 'o 5 = 'U Di'atafion 2.3 .s. .2 3.2 5 5:3 3.2 a E $.13 .c. E “5 g E E “5 E E E “5 g E E ”a g E "E ft 5‘, it s 1t :5 it s° ‘H~ 5 it 3 1i 3 it s Up to 9 mm 26 — — 26 17 65.4 26 — — 6 4 15.4 9.5 to 10.5 mm 62 2 3.2 62 32 51.6 62 4 6.5 62 14 22.6 11 mm and more 184 3 1.6 187 111 59.6 187 20 10.7 187 39 20.9 All gestations 272 5 1.8 275 160 58.2 275 24 8.7 275 57 20.7 ' This is a clinical term or expression- We compare the length of the cervix with the examiner's distal phalanx. If the cervix is shorter than the phalanx we describe the cervix as ushortx, if longer, it is rlongx. 29 Table 47: Incidence of Oil, by length of cervix Palmer test Bergman-Svennerund Cervical shape Cervical measure test * 1: l 1: i c l 1: ‘ E : : ) g : : g : : OEJ : : o '0 O O o "C O O Q 'C! O O o "U O O [nggtflaxn 3 fig 5 a 3 E 5 s E E a 55 3 .“s’ a 5 “5 g E E “5 g 'E E ”5 g 'E E “5 g E E 11 a 11 as 11 ti 1i s ‘ ‘ll 5 ‘H~ s 11 a: 11 6’ Short 139 4 2.9 142 86 60.5 142 13 9.2 142 34 20.4 Long 133 1 0.8 133 74 55.6 133 11 8.3 133 28 21.1 All lengths 272 5 1.8 275 160 58.2 275 24 8.7 275 62 20.7 10.3.4. CII by operative procedure In attempting to answer the subsequent occurrence of CH we divided our patients into four groups ac- cording to the procedure done. As shows Table 48, Table 48: Incidence of Oil, by operative procedure the type of procedure carried out appeared to have no effect on the subsequent development of CH. This result was anticipated since all four procedures begin in the same way, with dilation up to a maxi- mum of Hegar 13. Bergman-Svennerund Palmer test test Cervical shape Cervical measure 1 c r: 1: \ 1.: , a: __ _ (D _ _ i Q) __ _ CD _ __ E8 5 5 E3 5 a l EB a o , SE a a Operative 3 5 s a E .s as a 3 .s a .—C ‘ 3 .r_—- as a procedure '15 g 'E E “5 g E E ‘ '45 g E E 1 “o- g E E 11 a 11 s ( 1t 3 1+ as l 11 a 11 s l 11 a 11 s D & C 141 2 1 4 143 79 55.2 143 14 9.8 143 34 23.8 VA only 32 — 32 22 68.8 32 3 9.4 32 8 25.0 VA + curette check 78 3 3.8 78 43 54.8 78 2 2.6 78 9 11.5 VA + curettage 21 — — 22 16 28.8 22 5 22.7 22 6 27.3 Op. procedure 272 5 1.8 275 160 58.2 275 24 8.7 275 57 20.7 10.3.5. Cll by skill of operators training As a measure of the skill of the operators we di- vided the surgeons into two groups, those who were candidates for the specialist examination and those Table 49: Incidence of CH, by operators’ training who had passed it. The result of this study, shown in Table 49, shows again that there was no diffe- rence between these groups in subsequent outcome of CH. Bergman-Svennerund Palmer test test Cervical shape Cervical measure C E C C (D _ _ CD _ _ a.) _ _ (D _ _ Eg a a 63 a o } SE a 6 Eu 5 o Operators 3 .s 5 5 3 .5 s 5 3 .s e a 3 E 5 a o E E E o g E E l o E E E “5 g E E ‘H~ o ‘H~ E" ti o 1+ 6° 1 ti o ti E" ‘H~ 8 ti s" Junior 94 2 2.1 95 51 53.7 95 9 9.5 95 20 21.1 Senior 173 3 1 7 180 109 60.6 180 1 8.3 180 37 20.6 All operators 272 5 1.8 275 160 58.2 275 24 8.7 275 57 20.7 10.4. SYNECHIAE AFTER THE OPERATIVE PROCEDURE A number of the hysterograms showed unexpected synechiae in the cervical canal (27 patients), in the uterine cavity (17 patients), or in both locations (15 patients). It is interesting to note that most of 30 these women appeared to have had no difficulties and that the synechiae were discovered only by chance during examination for CH. We studied the synechiae in relation to the same 5 factors as CH and, as shown in Table 50, none of these factors seem to be related to the accidental findings of synechiae in the uterine cavity. Table 50: Incidence of synechiae, by five factors . . Syn. in the Syn. in the Syn. 1n the . TOTAL . cervux and Inter - - cerVIx cor us pretation 1. Duration of pregnancy % o P o corpus as to difference NO. /0 NO. /0 NO. 0/0 Up to 8 weeks 79 30.2 9 11.4 3 3.8 2 2.5 Not significant 9 to not later than 12 weeks 183 69.8 18 9.8 14 7.7 13 7.1 2. Dilatation Up to 9 mm 26 9.45 3 11.45 1 3.85 — — 9.5 to 10.5 62 22.55 6 9.68 5 8.06 5 8.06 Not significant 11 mm and more 187 68.00 18 9.63 14 7.49 11 5.88 3. Shape of the cervix Long 142 51.64 16 12.03 8 6.02 6 4.51 Short 133 48.36 11 7.75 12 8.45 10 7.04 Not significant 4. Operative procedure D & C only 143 52.00 15 10.49 8 5.54 6 4.20 VA only 32 11.64 2 6.25 3 9.38 2 6.25 VA and curretage without tissue 78 28.36 7 8.79 7 8.97 5 6.41 Not significant VA and curretage with tissue 22 8.00 3 13.64 2 9.09 3 13.64 5. Operators ‘ Junior 95 34.55 6 6.32 10 10.53 8 8.42 Senior 180 65.45 21 11.67 10 5.66 8 4.44 Not significant Total 275 100.0 27 9.82 20 7.27 16 5.82 In sum we prefer the Palmer test and assessment of the cervical shape. None of the five variables examined (duration of pregnancy, cervical dilation, lengths of cervix, operative procedure and training of operators) affected the incidence of Oil. in addi- tion accidentally discovered synaechia-e in 22.9% of 275 volunteer nuliparous women were not found to be related to any of the five factors examined. 31 Chapter 11 : There is no real evidence for menstrual disorders after induced abortion. There are a few studies, but these are primarily case reports without control groups (10). In our study, we collected data on the first period and all subsequent menstrual periods for six months after the abortion. We know from expe- rience with this and other studies that the return of the 4 week questionnaire is usually very good, (98.6 We), but that comparatively few women return the 6 month questionnaire (32 0/o). Because of this difference in return rate we separate our description of the first period from that of subsequent periods. 11.1. FIRST MENSTRUAL PERIOD AFTER ABORTION The day of onset of the first period after abortion was obtained for 4664 women. Because of the high return rate we feel these data represent our entire patient group. We do not count bleeding occuring during the first three weeks as a menstrual period but as post- abortal bleeding. Excluding these cases (428), the not yet occuring, and the not reported cases from the total, we find that in 93.9 % of women, menstrual bleeding appeared from the 4th to the 6th week. (Table 51). Table 51: First menstrual period after abortion, by operative procedure - Weeks D & C Total VA Grand total after abortion No. % No. % No. % 1 9 0.4 14 0.6 23 0.5 2 38 1.6 32 1.4 70 1.5 3 184 7.6 151 6.7 335 7.2 4 899 37.2 839 37.2 1738 37.2 5 633 26.2 609 27.0 1242 26.6 6 244 10.1 225 10.0 469 10.1 7 61 2.5 50 2.2 111 2.4 8 31 1.3 26 1.2 57 1.2 9 and more 28 1.2 29 1.3 57 1.2 Not yet occuring 286 11.9 276 12.3 562 12.0 Not reported 1 0.0 2 0.1 3 0.1 32 CHANGES IN MENSTRUAL PATTERN L. Andolsek VA+curette VA + Weeks VA only after abortion check curettage N0. l’/o NO. 0/0 N0. °/o 1 6 1.0 7 0.5 1 0.4 2 5 0.8 23 1.7 4 1.4 3 36 5.9 97 7.1 18 6.3 4 218 35.7 515 38.0 106 36.9 5 151 24.8 391 28.8 68 23.3 6 56 9.2 135 10.0 34 11.9 7 13 2.1 32 2.4 5 1.7 8 7 1.2 15 1.1 4 1.4 9 and more 6 1.0 15 1.1 8 2.8 Not yet occuring 110 18.0 126 9.3 40 13.9 Not reported 2 0.3 0 0.0 0 0.0 At the time the questionnaire was returned, 12.0 % of women had not yet had a menstrual period. The life table method was used to correct the percent distribution of cases, and to include these cases in the distribution. These distributions are shown in Table 52; no essential differences appear. Table 52: Cases with appearance of menstrual period after abortion, by ordinal week after operative procedure: basic distribution and corrected distribution by life table method Weeks after Basic % Corrected abortion distr. °/o distr. 1 0.5 0.5 2 1.5 1.5 3 7.2 7.2 4 37.2 37.5 5 26.6 27.7 6 10.1 11.4 7 2.4 4.3 8 1.2 3.0 9 and more 1.2 6.9 Most gynecologists believe that the first period after abortion is usually heavier and longer than normal. Some treat it as postabortal bleeding. Wo- men from our study assessed the intensity of the first menstrual period as shown in Table 53. it is evident that 41 % of women estimate the first menstrual period after abortion ‘to be heavier but it is also evident [that approximately every seventh woman claimed that the period was lighter. The dif- ferences among the treatment groups are minimal. Table 53: Intensity of first menstrual period after abortion relative to intensity before the operative procedure . D & C Total VA Grand total IntenSIty No. We NO. °/o NO. l3/0 More 914 42.9 786 39.8 1700 41.4 Same 911 42.8 899 45.5 1810 44.1 Less 302 14.2 289 14.6 591 14.4 Not stated 1 0.1 3 0.1 4 0.1 Grand total 2128 100.0 1977 100.0 4105 100.0 . VA + curette Va + lntensrty VA only check curettage N0. c'/o No. “/0 No. ”/0 More 191 38.2 490 39.8 105 42.5 Same 242 48.4 545 44.3 112 45.3 Less 65 13.0 194 15.8 30 12.2 Not stated 2 0.4 1 0.1 0 0.0 500 100.0 1230 100.0 247 100.0 Table 54 shows that/the abortion method had little if any effect on the duration of flow. Table 54: Duration of first menstrual period after abortion relative to duration before the operative VA + curette VA + 23:2; VA only check curettage No. % No. % No. % Worse 95 19.0 245 19.9 52 21.0 Same 326 65.2 792 64.4 158 64.0 Less 77 15.4 192 15.6 37 15.0 Not stated 2 0.4 1 0.1 0 0.0 All cases 500 100.0 1230 100.0 247 100.0 11.2. SIX-MONTHS POST-ABORTION MENSTRUAL PATTERNS From our experience and others we know that the rate of questionnaire returns worsens with the post- operative interval. We decided to request data about menstrual period patterns during the first six months after abortion in a 10% sample of women. We asked every woman whose study number ended with zero to return a questionnaire after six calendar months. We also gave her a diary in which she in- dicated all menstrual flow. We studied menstrual regularity after abortion, its nature, abdominal pain and other possible complications occuring in the time period. As shown in Table 56, we found 68.5 % of cases with regular menstrual periods during the first six months and 31.5 % with irregular periods. Table 56: Regularity of the six-month postoperative menstrual patterns, by operative procedure procedure . D 8. C Total VA Grand total Duration No. % No. % No. % Longer 792 37.2 654 33.1 1446 35.2 Same 1072 50.3 1074 54.3 2146 52.3 Shorter 263 12.4 246 12.4 509 12.4 Not stated 1 0.1 3 0.2 4 0.1 Grand total 2128 100.0 1977 100.0 4105 100.0 VA + curette VA + Duration VA only check curettage No. % No. % No. % Longer 156 37.2 408 33.2 90 36.4 Same 280 50.4 661 53.7 133 53.9 Shorter 62 12.4 160 13.0 24 9.7 Not stated 2 0.1 1 0.1 0 0.0 500 100.0 247 100.0 100.0 1230 Table 55 shows that the procedure also seems to have had no effect on dysmenorrhea, even though the answers were subjective and we did not have a control group. Table 55: Dysmenorrhea with the first menstrual period after abortion relative to dysmenorrhea before the operative procedure Dysme- D & C Total VA Grand total norrhea No. % No. ”/0 No. °/o Worse 456 21.4 392 19.8 848 20.7 Same 1325 62.2 1276 64.5 2601 63.3 Less 346 16.3 306 15.5 652 15.9 Not stated 1 0.1 3 0.2 4 0.1 All cases 2128 100.0 1977 100.0 4105 100.0 Regularity D & C Total VA Grand total No. % No. °/o No. % Regular 82 64.6 92 72.4 174 68.5 Irregular 45 35.4 35 27.6 80 31.5 VA + curette VA + Regularity VA only check curettage 'No. » % No. % No. °/o Regular 29 87.9 48 62.3 15 88.2 Irregular 4 12.1 29 37.7 2 11.8 Another assessment of regularity was gained from the number of periods during the six months. We assume that 5 or 6 periods in this time indicates a regular pattern. in this group, 67 % of the patients indicated 5 or 6 periods; over 4% did not note bleeding and 28 % noted irregular bleeding. The nature of menstrual bleeding as estimated by the patients may be classified as follows: Duration of bleeding SAME LONGER SHORTER 185 — (72.8 0A.) 36 — (14.2 %) 33 — (13.0 %) Dysmenorrhea SAME MORE LESS 186 — (73.2 %) 32 — (12.6 %) 36 —— (14.2 0/0) In sum these data show that interruption of preg- nancy has no apparent influence on duration of subsequent menstrual bleeding or on dysmenorrhea. 33 Chapter 12: 12.1. GOAL AND STUDY STRATEGY The primary purpose of this sub-study was to deter- mine whether abortions lead t-o an increased risk of subsequent ectopic pregnancy (EP), a hypothesis that has been suggested by some authors (11). Two hundred women operated on for EP at the Gynecological clinic from January 1971 through July 1973 were included in the study. The control group consisted of 600 women obtained by selecting the three women giving birth immediately following each EP operation. Since we required a representative sample of all intrauterine pregnancies, information was also obtained from women who sought induced abortions. Information on 4733 such women was obtained from the main part of this study. As will be seen, a pooled control was then derived, com- bining the statistics from the women with term deli- veries with those of the women obtaining abortions, in the ratio of 600 to 240. This ratio reflects the relative frequency of term deliveries and abortions during the time that the study was in progress. Most statistical comparisons in this paper were made on the women having E‘P and on the pooled control data. Since EP is a relatively rare complication of preg- nancy, we elected to use a case—control design to obtain etiologic information (12). This design consists of taking samples of cases of a disease, and of controls who do not have the disease, and com- paring them with respect to factors which may be related to the disease being investigated. In this study, the disease is EP, the controls are our ,,poo- led controls”, and the factor being investigated is a history of induced abortions, although other fac— tors were also examined. As will be seen, the sum- mary statistic which we will commonly use is called the ”relative-risk”. This statistic is typically used in case-control studies, and is defined as the risk of disease (EP) for persons who have some characte— ristic (eg. a positive history for induced abortions) divided by the risk of disease in persons who do not have the characteristic. All women included in the study were interviewed to obtain information on previous pregnancies, espe- cially about abortions, previous treatment of gyne- 34 ECTOPIC PREGNANCY M. Hren, T. Tomaievic and D. Seigel cologic illnesses, such as infection and sterility, and about previous abdominal operations. For wo- men whose Iast pregnancy terminated in an abor— tion, the method of the abortion was verified by reviewing the clinic records. The clinical diagnosis of EP was confirmed by his- tologic examination whenever possible. In 89 % of cases the diagnosis was histologically confirmed. In 11 % this was not possible because of the very con- servative operative technique, though the clinical status was clear. 12.2. ANALYTICAL FINDINGS 12.2.1. Risk of EP by age Table 57 presents the distribution of the ages of the women with ectopic pregnancies, and of the women [in the pooled control. The group having the great— est number of ectopic pregnancies were those aged 30—34. Control women hag their greatest num- ber in the age group less than 25. The relative risks for each age group of having EP as compared to an intrauterine one are provided in the bottom row. Women aged 25—29 are 3 times more likely than younger women to have EP. Women over 30 have a six time greater risk than women less than 25 years. Younger women are clearly at less risk. Table 57: The risk of EP, by age Study series A g e < 25 25—29 30—34 35 + Total Term deliveries 290 177 82 51 600 Abortions 1732 992 895 1114 4733 Abortions (Tot. adj.)* 87.8 50.3 45.4 56.5 240.0 Pooled control“ 377.8 227.3 127.4 107.5 840.0 Proportion 0.45 0.27 0.15 0.13 Ectopic pregs. 30 56 61 53 200 Proportion 0.15 0.28 0.30 0.26 Relative risk of EP 1 3.1 6.1 6.2 5.1 " These figures were obtained by multiplying abortions by a con- stant 240/4733, so that total is 240. " Pooled control equal to abortions (Tot. adj.) p‘lus term deliveries. 12.2.2. Risk of EP by previous intrauterine pregnancies Table 58 permits a comparison of' the number of previous intrauterine pregnancies in women in the EP group as compared with the women in the poo- led control. In each of four age groups the number of women with no previous intrauterine pregnan- cies, with one or two, and with three or more is presented. Women in the EP group at every age category are more likely to present a history of no previous intrauterine pregnancy than are control wo- men. This is reflected in the relative risks within each age group, which tend to be equal to or less than one. The average relative risks across all age cate- gories for histories of 1—2 pregnancies and 3 + pregnancies are 0.6 and 0.5. The interpretation of these data is that women with a previous history of intrauterine pregnancy have about one-half the risk of having their next pregnancy be ectopic, as com— pared with women with no previous pregnancy. Table 58: The risk of EP, by age and previous intrauterine pregnancies < 25 25—29 30—34 35 + Study series Prev. pregnancies Prev. pregnancies - 2 Prev. pregnancies Prev. pregnancies 0 1—2 3 + 0 1—2 3 + 0 1—2 3 + 0 1—2 3 + Term deliveries 199 90 1 48 115 14 14 44 24 9 23 19 Abortions 820 801 111 71 580 341 17 377 501 14 352 748 Abortions (Tot. adj.) 41.6 40.6 5.6 3.6 29.4 17.3 0.9 19.1 25.4 0.7 17.9 37.9 Pooled control 240.6 130.6 6.6 51.6 144.4 31.3 14.9 63.1 49.4 9.7 40.9 56.9 Proportion 0.64 0.35 0.02 0.23 0.64 0.13 0.12 0.50 0.39 0.09 0.38 0.53 Ectopic pregs. 22 8 0 12 37 7 18 25 18 8 22 23 Proportion 0.73 0.27 0.00 0.21 0.66 0.13 0.30 0.41 0.30 0.15 0.42 0.43 Relative risk of EP 0.67 0.00 1.10 0.38 0.33 0.96 0.65 0.49 1. Average rel. risk for women with 1—2 prev. pregnancies = 0.6 2. Average rel. risk for women with 3 + prev. pregnancies = 0.5 Table 59 presents an analysis of the same data from another point of view. For each age category the average number of previous intrauterine preg- nancies is presented for women with EP and the control series. At every age the EP had lower average numbers of pregnancies. Further, if aver- ages are computed with women with no previous pregnancies excluded, the average number of pre- vious pregnancies still tends to be higher in the pooled control. Tables 58 and 59 are consistent; women with EP tend to have had fewer intrauterine pregnancies than women with no EP. Table 59: Average number of previous intrauterine pregnancies, by age Study series Age < 25 25—29 30—34 35 + Term Average 0.38 1.11 1.88 2.16 Average (multigrav) 1.21 1.52 2.26 2.62 Abortions Average 0.82 2.17 2.88 3.44 Average (multigrav) 1.56 2.21 2.94 3.48 Pooled control Average 0.51 1.41 2.17 2.53 Average (multigrav) 1.31 1.72 2.45 2.87 Ectopic Average 0.30 1.30 1.74 2.26 Average (multigrav) 1.13 1.66 2.47 2.67 12.2.3. Risk of EP by previous deliveries Tables 60 and 61 present data on previous deli- veries for EP and controls. Average relative risks (Table 60) across all age categories are equal to 1.1 for women with 1 previous delivery, 0.4 for women with two or more previous deliveries. The interpre- tation of this is that women with 1 previous delivery have the same risk of their next pregnancy termi- nating as EP as do women with no previous deli- very. Further, women with two or more previous deliveries have only half the risk that their next pregnancy will be EP, as compared to women with no previous deliveries. The analysis of means pre- sents a similar picture (Table 61). In every age group the average number of preceding deliveries is grea— ter for the women in the pooled control series than for women in..the ectopic.series. Similary, the ave— rage number of preceding deliveries is higher for the pooled control series when the women with no previous deliveries are excluded. In both of these tables, the same pattern emerges: at all ages, wo— men with EP have histories of fewer deliveries than are found in histories of women with intrauterine pregnancies. Note 1: Pooled control average is weighted average of term and abortion averages. Note 2: Average (multigrav) rows give average No. of previous in- trauterine pregnancies among women with at least one p're- vious pregnancy. 35 Table 60: The risk of EP, by age and previous deliveries < 25 25—29 30—34 35 + Study series Prev. deliveries Prev. deliveries Prev. deliveries Prev. deliveries 0 1 ’2 + 0 1 2 + 0 1 2 + 0 1 2 + Term deliveries 230 52 8 67 94 16 16 43 23 12 14 25 Abortions 886 602 244 105 345 544 28 166 701 25 179 910 Abortions (Tot. adj.) 44.9 30.5 12.4 5.3 17.5 27.4 1.4 8.4 35.5 1.3 9.1 46.1 Pooled control 274.9 82.9 20.4 72.3 111.5 43.4 17.4 51.4 58.5 13.3 23.1 71.1 Proportion 0.73 0.22 0.05 0.32 0.49 0.19 0.14 0.40 0.46 0.12 0.21 0.66 Ectopic pregs. 25 5 0 17 28 11 22 20 19 11 33 9 0.83 0.17 0.00 0.30 0.50 0.20 0.36 0.33 0.31 0.21 0.62 0.17 Relative risk of EP 1 0.67 0 1 1.07 1.08 1 1.08 0.26 1 1.73 0.15 1. Average rel. risk for women with 1 previous delivery = 1.1 2. Average rel. risk for women with 2 + previous deliveries : 0.4 Table 61: Average number of previous deliveries, by age Study series Age < 25 25—29 30—34 35 + Term Average 0.21 0.75 1.24 1.43 Average (multigrav) 1.13 1.20 1.55 1.87 Abortions Average 0.64 1.56 2.02 2.31 Average (multigrav) 1.31 1.75 2.09 2.37 Pooled control Average 0.33 0.98 1.46 1.68 Average (multigrav) 1.18 1.36 1.70 2.01 Ectopic Average 0.17 0.93 1.02 1.40 Average (multigrav) 1.00 1.33 1.59 1.76 Note 1: Pooled control average is weighted average of term and abortion averages. Note 2: Average (multigrav) rows give average No. of previous in— trauterine pregnancies among women with at least one pre- vious pregnancy. 12.2.4. Risk of EP by previous total abortions Table 62 presents the history of all kinds of abor- tions for each age category for‘both groups. Average relative risks for women with 1 previous abortion, and for those with 2 or more previous abortions are 1.1 and 0.6 respectively. The implication to be drawn is that the risk of EP as opposed to intrauterine pregnancy in women with a history positive for abortion is the same or slightly less than it is for women whose histories are negative. Table 62: The risk of EP, by age and total previous abortions < 25 25—29 30—34 35 + Study series Total abortions Total abortions Total abortions Total abortions 0 1 25+ 0 1 2 + 0 1 2 + 0 1 2 + Term deliveries 253 33 4 127 41 9 54 13 15 27 14 10 Abortions 1458 238 36 556 311 125 389 312 194 391 381 342 Abortions (Tot. adj.) 13.9 12.1 1.8 28.2 15.8 6.3 19.7 15.8 9.8 519.8 19.3 17.3 Pooled control 326.9 45.1 5.8 155.2 56.8 15.3 73.7 28.8 24.8 46.8 33.3 27.3 Proportion 0.86 0.12 0.02 0.68 0.25 0.07 0.58 0.23 0.19 0.44 0.31 0.25 Ectopic pregs. 26 4 0 39 14 3 33 18 10 27 15 11 Proportion 0.87 0.13 0.00 0.70 0.25 0.06 0.54 0.30 0.16 0.51 0.28 0.21 Relative risk of EP 1.1 0.0 1.0 0.8 1.4 0.9 0.8 0.7 1. Average rel. risk for women with 1 previous abortion = 1.1 2. Average rel. risk for women with 2 + previous abortions = 0.6 12.2.5. Risk of EP by previous induced abortions Table 63 presents the history of induced abortion, for each age category, for both groups. Average relative risks for women with 1 previous induced abortion, and for those with two or more previous 36 induced abortions are 1.0 and 0.8 respectively. Since these relative risks are close to 1, the implication to be drawn ‘is that the risk of ectopic pregnancy as opposed to intrauterine pregnancy is the same in women with a history positive for induced abortion as it is for women whose histories are negative. Table 63: The risk of EP, by age and previous induced abortions < 25 25—29 30—34 35 + Study series induced abortions induced abortions induced abortions induced abortions 0 1 2 + 0 1 2 + 0 1 2 + 0 1 2 + Term deliveries 269 19 2 152 25 0 66 9 7 39 6 6 Abortions 1505 206 21 635 270 87 489 260 146 515 350 249 Abortions (Tot. adj.) 76.3 10.4 1.1 32.2 13.7 4.4 24.8 13.2 7.4 26.1 17.7 12.6 Pooled control 345.3 29.4 3.1 184.2 38.7 4.4 90.8 22.2 14.4 65.1 23.7 18.6 Proportion 0.91 0.08 0.01 0.81 0.17 0.02 0.71 0.17 0.11 0.61 0.22 0.17 Ectopic pregnancies 27 3 0 45 9 2 47 8 6 33 15 5 Proportion 0.90 0.10 0.00 0.80 0.16 0.04 0.77 0.13 0.10 0.62 0.28 0.09 Relative risk of EP 10 1.3 0.0 1.0 0.9 1.9 1.0 0.7 0.8 1.0 1.3 0.5 1. Average rel. risks for women with 1 previous induced abortion 21.1 2. Average rel. risks for women with 2 + previous induced abortions = 0.8 The method used to perform the last abortion in the EP and term delivery groups is presented in Table 64. This information is not available for women in the abortion series. For every age category, the pro— portion of abortions for which curettage was used was not very different for the two groups: in no age category was there a statistically significant dif- ference. Neither of these two tables leads one to believe that previous abortions -or the methods by which such abortions are performed have any re- lation to the risk of ectopic pregnancy. Table 64: Proportion of abortions in EPs and term deliveries, by age and operative procedure of last pregnancy termination < 25 25—29 30—34 35 + Study series Method Method Method Method None D&C VA Unkn None D&C VA Unkn None D&C VA Unkn None D&C VA Unkn Ectopic pregs. 26 2 1 1 39 10 4 3 33 23 4 1 27 22 3 1 Proportion 0.87 0.07 0.03 0.03 0.70 0.18 0.07 0.05 0.54 0.38 0.07 0.02 0.51 0.42 0.06 0.02 Term deliveries 254 22 10 4 127 37 6 7 54 21 3 4 27 19 4 1 Proportion 0.88 0.08 0.03 0.01 0.72 0.21 0.03 0.04 0.66 0.26 0.04 0.05 0.53 0.37 0.08 0.02 Note: 1. Method of last abortion not available for abortion series. 2. Some women had not only induced abortions but also spontaneous abortion as last abortion. Therefore number of women without abortions in this Table does not correspond to values in previous Table. 12.2.6. Risk of EP by previous gynecologic treatment Table 65 presents the number of women in the EP and term delivery series who reported having had significant gynecologic treatment. Such treatment was primarily for sterility or pelvic inflammatory di- sease (PID). Such treatment was clearly «more com- mon in women with EP. The relative risks of EP summarize the importance of the examined factors. On the average, women with sterility problems or PlD are 7 to 9 times more likely to have their next pregnancy turn out to be ectopic. Table 65: The risk of EP, by age and previous gynecological treatment <25 25—29 30—34 35 + Study series Gyn. treatment Gyn. treatment Gyn. treatment Gyn. treatment None Ster. PID Any None Ster. PID Any None Ster. PlD Any None Ster. PlD Any Ectopics 23 2 4 7 37 9 11 19 41 9 10 20 35 11 6 18 Proportion 0.77 0.06 0.13 0.23 0.66 0.16 0.20 0.34 0.67 0.15 0.16 0.33 0.66 0.21 0.11 0.34 Term deliveries 281 3 5 9 167 3 7 10 77 3 2 5 45 2 4 6 Proportion 0.97 0.01 0.02 0.03 0.94 0.02 0.04 0.06 0.94 0.04 0.02 0.06 0.88 0.04 0.08 0.12 Relative risk of EP 1.0 8.1 9.8 9.5 1.0 13.5 7.1 8.6 1.0 5.6 9.4 7.5 1.0 7.1 1.9 3.9 Average relative risk for sterility = 9 Average relative risk for PID = 7 Average rel. risk for any gyn.treatment = 7 Note: Sterility and PID are not mutually exclusive in this Table. 37 12.2.7. Risk of EP by previous abdominal operations Table 66 presents the numbers of women with EP and the pooled control series who have had abdo- minal operations. The abdominal operations are di- vided into operations for previous EP, and ,,other" operations, which include appendectomies, tubal surgery, and caesarean sections. Women with EP have had a previous EP more fre— quently in every age group than control women. Estimates of relative risk tend to be unreliable be— cause of the small numbers. Nevertheless, the fact that they range from 10 to several hundred indicates that this is an important factor in the history of women with EP. Any woman with a history of EP is therefore at considerably greater risk than other women to experience another ectopic pregnancy. Other abdominal operations on the other hand, do not appear to be more common in the histories of women with EP than in histories of women in the term delivery series. On the average, the relative risk is very close to 1. Table 66: The risk of EP, by age and previous abdominal operations <25 25—29 30—34 35+ . Operations Operations Operations Operations Study series No No No No 0 Ect. Oth. ect. 0 Ect. Oth. ect. 0 Ect. Oth. ect. 0 Ect. Oth. ect. Term deliveries 272 0 18 290 161 14 175 74 1 7 81 42 O 9 51 Abortions 1 1731 990 4 891 11 1103 Abortions (Tot. adj.) 0.05 87.8 0.1 50.2 0.2 45.2 0.6 55.9 Pooled control 0.05 377.8 2.1 225.2 1.2 126.2 0.6 106.9 Proportion 0.0001 0.9999 0.092 0.991 0.009 0.991 0.006 0.994 Ectopic pregs. 27 2 1 28 48 5 3 51 45 10 6 51 33 12 8 41 Proportion 0.07 0.93 0.09 0.91 0.16 084 0.23 0.77 Relative risk of EP* 540 0.6 11 0.7 21 1.4 52.1 1.1 " Note: Relative risk for previous ectopic pregnancy based on ectopic pregnancy series and pooled control. Relative risk for other abdominal operations based on ectopic pregnancy series and term deliveries only. 12.3. FACTORS OTHER THAN INDUCED ABORTION GOVERNING EP We believe that our data do not support the hypo- thesis that induced abortlon leads to .an increased risk of subsequent ectopic pregnancy. In this study, women with several induced abortions are in fact at slightly reduced risk of subsequent ectopic pregnan- cies. This is not surprising, since having had several induced abortions indicates a normally functioning reproductive system. The suggestion that a history of several induced abortions permits prediction of a slightly decreased risk of ectopic pregnancy is even more dramatically illustrated in patients with preceding intrauterine pregnancies and preceding deliveries. Women who have had several previous deliveries have only one half the risk that their next pregnancy will be ecto- pic, as compared with women who have never had a delivery. This is consistent with the point of view that ectopic pregnancies occur in women who have difficulty having a normal pregnancy. 38 Age is clearly a factor in the risk of ectopic preg- nancy, older women being subject to as much as six times the risk of younger women. Any study which attempts to examine the etiologic factors of ectopic pregnancies would have to take age into consideration in the design and the analysis. An interesting story emerges from the examination of previous abdominal operations and gynecologic treatment. Women with a history of surgery for ecto- pic pregnancy are likely candidates for future sur- gery. This is, as a matter of fact, perhaps the best marker for women of high risk. Other abdominal surgery, on the other hand, appears to be entirely unrelated to subsequent risk of ectopic pregnancy. Women seeking gynecologic treatment, for ste- rility and pelvic inflammatory disease, are candi- dates for future ectopic pregnancy to a greater extent than women who do not require such treat- ment. The reasons are probably different. Sterility may be caused by the very same problems that in turn cause ectopic pregnancies, but vpelvic inflam- matory disease may produce tissue changes that cause later ectopic pregnancies. Chapter 13: 13.1. STUDY AND STUDY SETTING A comparison of the medical effects of induced abortion by two principal methods, curettage (D & C) and vacuum aspiration (VA), was carried out in Ljubljana during the years 1971 and 1972. Immediate complications such as anesthesia events, uterine injuries, bleeding, and infection were followed in the entire study group of 4733 women for whom abortion was performed for social reasons. Sixty- seven for whom the abortion was performed for medical indications during «this time were excluded from the study. We examined correlations of the procedure with Rh sensitization, cervico-isthmic in— competence, ectopic pregnancy, menstrual patterns, subsequent ectopic pregnancies, in subgroups se- lected from the main study group. All subjects were volunteers to whom the objectives of the study were explained. The majority were in the age group 20 to 39 years (83.3 °/o). Most of them came from urban areas (69.0 0/0). At least 80.5% had had one previous pregnancy, 33.6 % had had at least one induced abortion. Fifty-eight percent used contraceptives before abortion. Of these 39.0 % used coitus interruptus or rhythm, and 9.1% used lUDs and orals. After abortion 61.4 % used contra- ceptives: 28.9 0/o coitus interruptus and rhythm and 26.5 % lUDs and orals. Most important was the fact that abortion did not seem to stimulate the subse- quent use of modern contraception. Women wanting an abortion must apply to a special commission for permission. A majority applied bet- ween 6 and 10 weeks of pregnancy. The procedure was performed on the same morning the women were admitted to the hospital. In 98.7 % the abor- tion was done under general anesthesia. The ope— rative technique was selected with a random allo- cation principle based on the last digit of the pa- tient’s birth year: odd number, VA — even number, D&C. For VA we used a locally-made apparatus with metal cannulae. After VA, 73.0 % of the opera- tions were checked with a small curette to make certain the uterine cavity was empty. If the operator discovered placental tissue, he did a complete curet- tage. Accordingly, we divided the VA group into 3 subgroups: 1. VA only (27.0 °/o) SUMMARY AND RECOMMENDATION L. Andoléek 2. VA with curette-check and no placental tissue (60.2 0/o) 3. VA with curettage and placental tissue obtained (12.8 °/o) This group has to be considered as a failure of the first procedure. The operations were performed by senior (66.8%) and junior (33.2 %) gynecologists who usually per- formed no more than 8 abortions daily. A slight in- crease in complications occurred after the ninth ope- ration. The D & C lasted an average of 7.2 minutes; the average duration of VA cases was 5.8 minutes. There were 6 VA operations (0.3 0/o) lasting longer than 19 minutes; in the D & C group there were 19 such cases (0.8 °/o). 4499 (95.0%) of the women were hospitalized for two days or longer. After discharge they received 7—10 days of sick leave. 13.2. COMPLICATIONS DURING HOSPITALIZATION 13.2.1. Anesthesia complications In 157 cases (3.4 °/o), laryngospasm, anaphylaxis, collapse, vomiting and convulsion were noted. These complications did not differ significantly with regard to the type of operative technique. 13.2.2. Injury of uterus We found 3 perforations of the uterus (0.82 %) and 33 injuries of the cervix (1.27 0/o), mostly lacerations. All perforations occured in the D & C group during dilation with Hegar dilators in multigravid women, pregnant 8 to 10 weeks; all perforations were done by senior operators. 13.2.3. Blood loss Blood loss was measured quantitatively in a sub— group of 532 women. Loss in the entire group was estimated by the operator and by the need for addi- tional uterotonics. They were needed in 6.0% of the D & C group and in 3.9% of the VA group, especially in those cases where VA was concluded with curettage (11.3 °/o). In the entire group only one 39 woman bled so much to need a transfusion. This proved to be caused by an endocervical pregnancy which required hysterectomy. In the 532 in whom blood loss was measured, we found that blood loss was heavier after 10 weeks of pregnancy regardless of the method. If VA & curett- age was used, blood loss was greater even in ear- lier abortions. Blood loss during the operation rang- ed in most cases up to 150ml and depended on the method: average loss in D & C: 116.35 ml; in VA: 103.90 ml. When VA plus curettage was used, bleed- ing in 40% of cases was over 150ml; in 8.4% bleeding exceeded 300ml and average blood loss was 171.67 ml. Blood loss was heavier with longer operations and when D & C was used, but the loss did not increase proportionally with time with VA. While still in hospital we registered among 4733 women 14 cases (0.3 %) of bleeding which did not stop after routine treatment. In all cases curettage was done and different quantities of placental tissue found. Eight cases were registered in the D & C group and the remaining 6 cases divided into VA only (1 case) VA+curette check (4 cases) and VA + curettage (1 case). 13.2.4. Infection We registered 104 cases (2.2 0/o) of infection during hospitalization. Among them were 35 cases (0.7 %) of pelvic infections, 39 (0.8%) of fever-only, 11 (0.2 0/o) of hemorrhage and infection, and 19 (0.4 %) of other infections. Of the later, almost half (9) had urinary infections. in another subgroup we ascertained the influence of vaginal and cervical flora on the appearance of infection and found that abnormal flora (classes iii, N, V and VI according to Jellinek) have no influence on the appearance of infections. The presence of pathogenic bacteria on the cervix before the pro- cedure does not seem to be related to the occur- rence of infection during hospitalization or up to the first menstrual period. 13.3. COMPLICATIONS AFTER DISCHARGE FROM THE HOSPITAL Events from discharge until the first period were recorded by the patients who noted temperature, bleeding, pains in the lower back and abdomen, and the character of the period. The questionnaires from which we evaluated these events were return- ed by 4667 (98.6 % of the total). 13.3.1. Infection and fever without other symptoms Fever was noted by 1134 (24.3 °/o). Of these 17.4 % sought help in the outpatient department. No dif- ferences were found according to method. vForty-eigth women (1.0 %) were rehospitalized: 32 (1.3 %) from the D & C group, 16 (0.7 %) from the VA group. We concluded that D & C is more likely to lead to sub- sequent hospitalization from fever than VA (14-2 = 4.17 d. f. = 1). 13.3.2. Postabortal bleeding Various amounts of postabortal bleeding were re- ported by 3363 women (72.1 °/o). Medical assistance 40 was sought by 458 (13.6 °/o), or 9.8 % from the total study group. The differences among techniques are not significant. Hospitalization for bleeding occurred in 79 (1.7 %) and these patients received conservative treatment in 19 cases (24.1 °/o); recu'rettage was required in 60 (75.9%). There is no difference between D & C group and total VA group in hospitalization because of bleeding; a higher percentage of retained tissue is shown after D & C (1.3 %) than after VA (0.9 %) but the difference is not significant. The VA + addi- tional curettage group showed the highest level (2.4 °/o). 13.3.3. Pain in abdomen Less than 45% (2091 women) recorded abdominal pains after discharge. We assume that these pains were not of a serious character since only 7.2 % sought medical assistance. Only 7 (0.3 %) were re— hospitalized. The surgical technique seemed to have no significance. 13.3.4. Pain in lower back Exactly 1272 women (40.8 %) who returned questio- nnaire complained of pains in the lower back. Dif— ferences regarding method were insignificant. We are doubtful if such pains were the consequence of abortion since 33.2% complained of them before the operation. 13.3.5. Rh immunization During the study we took blood from 5942 women who applied for abortion. Blood groups ABO and Rh were determined. We found 789 Rh D negative women. Before abortion we took a history concer- ning previous abortions, deliveries and transfusions. We tend to withhold judgment on the veracity of previous abortion histories. All sera were tested for the presence of anti Rh agglutinins. Forty-one women (5.2%) had anti-Rh agglutinins in the serum before the abortion, pro— bably from previous pregnancies. Two women had anti-Rh agglutinins even though they had no history of previous abortions, deliveries and transfusions. These women were excluded from the study. Three months after abortion we checked for anti—Rh agglu- tinins. We also tested serum 1 to 2 years after the abortion. The abortion immunized women against Rh agglu- tinins in 4.3 % of cases but such agglutinins, found 3 months after abortion, disappeared from the serum in most cases within one to two years. In wo- men immunized before abortion we found a change of titer three months after abortion. immunization does not seem to depend on whether the abortion is performed in the first Weeks of pregnancy or later up to 12 weeks. It does seem that the method of abortion has some influence in our sample but differences are not sta- tistically significant. Rh sensitization was detected more often in the VA group (7.7 %) than after D & C (4.2 °/o). In the VA group there was less immunization (2.5 %) detected, when suction was followed by additional curettage. 13.3.6. Cervico-isthmic incompetence In 275 nulliparous women we measured the influence of abortion on the appearance of subsequent cer- vico4isthmic incompetence (OH). We examined the women three months after abortion in the second half of the menstrual cycle. To determine CII we used 1) the Palmer test, 2) the Bergman-Svenne- rund test, and 3) cervical shape and 4) cervical measurements by x-ray. We found great differences among the tests. With the Palmer test we found 5 cases (1.8 %); with the Bergman-Svennerund test, 160 cases (58.2%); by x-ray shape of the cer- vix, 24 cases (8.2 We). We think that the Palmer test and assesment of the shape of the cervix are the most valuable tests. We did not find significant differences in CH among the surgical technique used or were such factors as pregnancy length, amount of cervical dilation or the skill of the operators found to be important. Incidental to our x-rays of the cervix, we discovered asymptomatic synechiae of the cervix, the isthmus, and the uterine cavity. In total we found synechiae In the cervix in 27 cases (9.8 0/o), in the corpus in 20 cases (7.3 °/o), and in both in 16 cases (5.8 0/o). 13.3.7. Menstrual pattern We collected data on the first menstrual period after abortion and on menstrual patterns during the first six months. On the basis of 3674 returned question- naires the first menstrual period occurs most often (93.9%) between 4—6 weeks after abortion. 1700 women (41.4%) estimated the first menstrual pe- riod to be heavier than before. The duration of the period was the same in 2146 (52.3%). Dysmenor- rhea was greater in 848 women (20.7 0M. We con— clude that abortion does not significantly influence menstrual patterns, and there are no significant differences according to method. In 254 women we recorded menstrual patterns du— ring the first six months. 174 (68.5 %) reported re- gular periods and 80 (31.5%) reported irregular periods. 185 (72.8 %) reported the same duration of bleeding as before and 186 (73.2%) indicated the same amount of dysmenorrhea. It seems that abor- tion has no effect on the regularity and quality of subsequent menstrual periods. 13.3.8. Ectopic pregnancy We tried to determine if the techniques influence the appearance of ectopic pregnancies. For this We analyzed data from 200 women with ectopic preg- nancy. These data were compared with data ob- tained from a control group of 600 parturients deli- vered at the same time in the same hospital and 240 women who sought induced abortion at the same time. Our data do not support the hypothesis that in— duced abortion leads to an increased risk of sub- sequent ectopic pregnancy. In our experience wo— men with several induced abortions are at slightly reduced risk of subsequent ectopic pregnancies. The risk factors for subsequent ectopic pregnan- cies are: age (older women having six times the risk of younger women), history of therapy with or with— out surgery for infertility; and pelvic inflammatory disease. The significant finding is that women who have had several previous deliveries have only one- half the risk that their next pregnancy will be ecto- pic, as compared with women who have never had a delivery. There are no significant differences ac- cording to the operative method. 13.4. RECOMMENDATIONS 1. Vacuum aspiration (VA) is better than curettage for abortion because it is quicker, has a lower perforation rate, induces less blood loss at ope- ration, and involves less subsequent infections which require rehospitalization. 2. Our study also showed that VA should be per- formed within the first ten weeks of pregnancy. In more advanced pregnancies, additional curet- tage as a second procedure is often necessary. 3. This study shows that complication rates should not be calculated on the basis of patients’ reports since such reports are too subjective. indeed we feel that patients over-report such events as bleeding and pain. Objective evaluation must be obtained from medical records. 4. Treatment of Rh negative women after abortion with Rh D anti immunog‘lobulin is recommended. 5. In our study a relatively high percentage of pa- tients showed cervico—isthmic incompetence and uterine synechiae. The possibility of these com- plications should be explained, particularly to the primiparous patient. 41 TABLES Page 4. PATIENT POPULATION Table 1: Selected social characteristics 9 Table 2: Selected fertility characteristics 9 Table 3: Pregnancy termination, by gestational age 10 Table 4: Interval between abortion and first intercourse 10 Table 5: Patients stating no postoperative intercourse, by time of questionnaire return 10 Table 6: Contraceptive practice before and after abortion 11 Table 7: Contraceptives used before and after abortion 11 5. OPERATION AND OPERATOR Table 8: Operative procedures, by pregnancy duration and the operator’s training 12 Table 9: Operators with more than fifty operations, by operative procedure 12 Table 10: Operators, by training 13 Table 11: Uterine injury, by the operator’s training 13 Table 12: Operators, by operation order attained at same day 13 Table 13: Complications, by operation order 13 Table 14: Pregnancy duration, by last menstrual period and by the surgeon’s estimate in weeks 14 6. OPERATIVE EVENTS Table 15: Complications of anesthesia, by operative procedure 15 Table 16: Complications, by length of operation 15 Table 17: Injuries of the cervix, by operative procedure 16 7. BLOOD LOSS Table 18: Treatment of bleeding at operation, by operative procedure 17 Table 19: A subgroup of 532 patients, by age, prior pregnancies and their outcome, and by major operative procedure 18 Table 20: Pregnancy duration, by major operative procedure 18 Table 21: Blood loss at operation, by operative procedure 18 Table 22: Blood loss during 48 hours hospitalization, by operative procedure 19 Table 23: Total blood loss, by operative procedure 19 Table 24: Total average blood loss, by gestational age and operative procedure 20 42 10. 11. Table 25: Table 26: Table 27: Table 28: Table 29: Table 30: Table 31: Table 32: Table 33: Table 34: Table 35: Table 36: . RHESUS Table 37: Table 38: Table 39: Table 40: Table 41: Table 42: Page Total average blood loss, by duration of operation and by operative procedure Bleeding after discharge in days, by operative procedure Bleeding after discharge, by quantity of blood loss at abortion Bleeding after discharge requiring ambulatory treatment, by operative procedure Bleeding due to retained placental tissue and bleeding without placental tissue, by operative procedure . INFECTION Events during first and second day of hospitalization for the 4733 women, by operative procedure Complications during hospitalization, by operative procedure Hospitalization events, by vaginal flora as classified by Jellinek Events during hospitalization, by microbiological status Reported fever, by ordinal day of onset after discharge and by operative procedure Infection and fever requiring rehospitalization, by operative procedure Outpatient and hospital treatment of abdominal pain, by operative procedure FACTOR IMMUNIZATION Immunization of Rh negative women before abortion, by prior delivery and abortion Development of anti-Rh agglutinins three months after abortion in women with no agglutinins prior to the operative procedure, by prior delivery and abortion Induction of anti-Rh agglutinins in women with no antibodies prior to the operative procedure, by gestational age at pregnancy termination Development of anti-Rh antibodies three months after abortion, by operative procedure Alteration in titer of anti-Rh agglutinins after abortion in already-immunized women Anti-Rh agglutinins in women two years after a non-immunizing abortion, by pregnancy status Table43: Anti-Rh agglutinins in women two years after an CERVICO Table 44: Table 45: Table 46: Table 47: Table 48: Table 49: Table 50: immunizing abortion, by pregnancy status ISTHMIC INCOMPETENCE Incidence of Oil, by type of test Incidence of Oil, by duration of pregnancy and type of test Incidence of Oil, by degree of cervical dilatation Incidence of Oil, by length of cervix Incidence of Oil, by operative procedure Incidence of Oil, by operators’ training Incidence of synechiae, by five factors CHANGES IN MENSTRUAL PATTERN Table 51: 43 First menstrual period after abortion, by operative procedure 20, 21 21 21 21 22 23 23 23 24 24 24 25 26 26 27 27 28 29 29 30 30 30 31 32 12. Table Table Table Table Table 52: 53: 54: 55: 56: ECTOPIC Table Table Table Table Table Table Table Table Table Table 57: 58: 59: 60: 61: 62: 64: 65: 66: Page Cases with appearance of menstrual period after abortion, by ordinal week after operative procedure: basic distribution and corrected distribution by life table method 32 Intensity of first menstrual period after abortion relative to intensity before the operative procedure 33 Duration of first menstrual period after abortion relative to duration before the operative procedure 33 Dysmenorrhea with the first menstrual period after abortion relative to dysmenorrhea before the operative procedure 33 Regularity of the six-month postoperative menstrual patterns, by operative procedure 33 PREGNANCY The risk of EP, by age 34 The risk of EP, by age and previous intrauterine pregnancies 35 Average number of previous intrauterine pregnancies, by age 35 The risk of EP, by age and previous deliveries 36 Average number of previous deliveries, by age 36 The risk of EP, by age and total previous abortions 36 : The risk of EP, by age and previous induced abortions 37 Proportion of abortions in EPs and term deliveries, by age and operative procedure of last pregnancy termination 37 The risk of EP, by age and previous gynecologic treatment 37 The risk of EP, by age and previous abdominal operations 38 44 FIGURES Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure Figure mmAwN—t \l 10: 11: 45 : Selected social characteristics : Selected fertility characteristics : Contraceptive practice before and after abortion : Operative procedures used in this study : Operation time, by technique and experience : Uterine injuries, by operator’s training. The V-shaped pattern : Complications, by order of operation : Vacuum cases requiring curettage, by order of operation : Pregnancy duration, by last menstrual period and by the surgeon‘s estimate in weeks Immunization of Rhesus — negative women prior to the present abortion, by deliveries and abortions Shapes of cervix. Standards for cervical measurement test 2% 47 47 48 49 50 50 51 52 52 Other I7 + at Not 13—16 |S—itu ent-f Urban ousewr e 9-12 Income Urban Earner 5—8 0-4 EDUCATION OCCUPATION RESIDENCE in School Years Figure 1: Selected social characteristics 1‘ + Yes Live Births Undecided None No Stillbirths First Pregnancies AG E PREVIOUS PREVIOUS WOMAN WANTS PREGNANCIES ABORTIONS ADDITIONAL CHILDREN Figure 2: Selected fertility characteristics 46 BEFORE AFTER The Present Abortion The Present Abortion 5.5 22.3 6.9 Diophr - Jelly CONDOM RHYTHM 32.0 41.9l NONE jsss METHOD I I I l I I I I 100 75 50 25 0 0 25 50 75 100 Percent Women Percent Women Figure 3: Contraceptive practice before and after abortion D & C VACUUM _ . .V _. . 483 100.0 13.0 29.1 6.2 220 ' 60.2 72.8 VACUUM VACUUM VACUUM DILATATION and CURETTAGE and and ALONE CURETTE CHECK CURETTAGE Figure 4: Operative procedures used in this study 47 Junior - Doctors Senior 10 - 8.6 § 8 — D .E 2 .E 6.4 <1) 6 _ E i: C .9 ‘5 a 4 - o. O C o o E 2 _ D & C VACUUM Technique and Experience Figure 5: Operation time, by technique and experience 48 Uterine Injuries Perforation Lacerotion,ect. 12 - N : 980 1 656 974 636 486 10.2 10.3 1° ' I 8.5 E .9 8 _ E I Q) C. O 6.3 8 6 — O B O. 4.1 f.’ 4 -— 0 DC 2 1—- 100 200 300 ‘ 100 199 299 399 400+ Abortion Experience (Frequency) of the Operators during the two-year study period Figure 6: Uterine injuries, by operator’s training. The V-shaped pattern. 49 Percent of Complications Bleeding Placental Tissue Found Laceration and/or Perforation ." . ‘- 1 2 3 4 5 6 7 8 9 10 Order of Operation Figure 7: Complications, by order of operation Percent 'Vacuum plus Curettage 15 10 ./'/.\./’/ Order 0F Operation Figure 8: Vacuum cases requiring curettage, by order of operation 50 o ..... ' ' 34.16 0 Operator 5 Estimate :'. 0—0 Since Last Menstrual Period 30— ,: : 25.25 ‘- 21.93 0*. 20 — '. c '-. G.) I E ‘. o . 3 “E a) U 83 Q. 10 - 0 K“; 4.92 ‘. \ 2.17 ‘1..'\ 1.11 o I ' l I L 1 ‘8'9-2- -------- 3°55 8 9 10 11 12 13 14+ Gestational Age in Weeks Figure 9: Pregnancy duration, by last menstrual period and by the surgeon’s estimate in weeks 51 0—0 One or more previous Abortions o ------- o No previous Abortions N: — l 725 5/68 lZ/léé 6/63 N: 2/i63 3/137 9/132 3/36 10 t— 8 _ E E 1*:— G — E U) m E m B 4 '- 0 D— 2 .— I I I I Previous Deliveries Figure 10: Immunization of Rhesus-negative women prior to the present abortion, by deliveries and abortions W 1 2 3 4 NORMAL DIRECT 1:5 BAG-SHAPED 2) TUNNEL-SHAPED 3) UTERUS INSUFFICIENCY UTERUS UTERUS l) The cervix is in all its length of the some width and is continuing directly into the corpus. 2) The upptr part of the cervix is included in the corpus. 3) An infantile uterus. Figure 11: Shapes of cervix. Standards for cervical measurement test 52 REFERENCES 10. 11. 12. . Andolsek, L. et al: immediate and early sequelae of induced abortion. Zdrav. vestn., 40; 93, 1971. O . Bergstrom, H., Nilsson, L. A., Nilsson, L. A., Ryttinger, L.: Demonstration of Rh antigens in a 38-day old fe- tus. Amer. J. Obstet. Gynec., 99, 130, 1967. . Chown, B.: On a search for rhesus antibodies in every young fetuses. Arch. Dis. Childh. 30, 232, 1955. .Woodrow, J. C.: Rh immunisation and its prevention (Munksgaard-Copenhagen, 1970). .Stratton, F., Renton, P. H.: Practical blood grouping (Blackwell Oxford, 1958). . Boorman, K. E., Dodd, B. E.: An introduction to blood group serology. (Churchill, London 1970). . Pollack, W., Gorman, J. G., Freda, V.: Prevention of Rh hemolytic disease. Progress in hematology — Vol. Vl. (Grune & Stratton, New York and London, 1969). . Kralj, B., Lavric, V.: The diagnosis of incompetence of the internal es of the cervix. Will be read at the Congress of Danube countries in Belgrade in 1975. . Picot, H., Thompson, H. G., Christopher, I. M.: A consi- deration of the incompetent cervix. Amer. J. Obstet. Gynec., 77, 706—718, 1959. Edstrom, K.: Early complications and late sequelae of induced abortion. A Review of the Literature. Working Paper for Consultation on Abortion Care. WHO, Ge- neva, 17—21 April 1972. Sawazaki, 0., Tanaka, S.: The relationship between artificial abortion and 'extrauterine pregnancy. in: Harmful Effects of Induced Abortion, Family Planning Federation of Japan, 49, 1966. Mac Mahon, B., Pugh, T. F., lpsen, |.: Epidemiologic Methods, Little, Brown and Company, Boston, Toronto, 1960. 53 Appendix 1: ADMISSION FORM Date: APPENDICES Surname and name: Maiden name: Date of birth: Address: Occupation Office: No. of school years: Tel. No. Fertility history No. of conft'vnements No. of live births No. of still births No. of all ab. Spont. ab. Induced ab. Ect. preg. Last (pregnancy terminated: Date: How terminated: first ipreg. confinement still birth induced ab. spont. ab. ectopic preg. Previous menstrual data and present pregnancy Menstrual cycle: regular irregular ; Duration of flow days Intensity: severe medium light . Menstrual pain: no yes Lower abdominal pain: no yes only at menses Lower back pain: no yes only at menses Date of last menstr. uperiod: Duration of flow days Contraception used at the time of this conception: no yes IUD orals diaphragm spermaoides: condom C. I. rhythm Health center where pregnancy was diagnosed Indication for abortion: social medical other If medical, diagnosis: Planning another birth: no yes not know Remarks: Appendix 2: OPERATION FORM Date: Study number: 1. Duration of pregnancy in: weeks (operator's estimate) 2. Uterus position: anteverted retroverted elongated Adnexis: normal enlarged Portio: short long conic cylindric laceration: no yes : left right both sides Cauterization of portio: no yes 54 3. Cervical canal: dilation to Hegar 'resistance starts at Hegar resistance stops at Hegar resistance does not stop no resistance 4. Anesthesia: — Premedication: no yes — Analgesia: no yes — Anesthesia: no local regional general — Anesthesia complications: no laryngospasm anaphylaxis collapse vomiting convulsion combination — Other medical conditions — no connection with pregnancy: — Other medication: 5. Termination complications: — Perforation: no suspect yes — Laceration: no yes — without suture yes with suture — Shock (with B. P. below 80 systolic) no yes — Therapy of bleeding: oxytocin comb. of uterotonics transfusion — Death: no yes 6. Termination procedure: D§C VA only VA+D§C VA+D§C (placental tissue) 7. Operation order number (for specific operator) 8. Is the surgeon also her gynaecologist: no yes 9. Duration of operation: minutes. Name of anesthesist: Name of surgeon: Appendix 3: VACUUM ASPIRATOR TYPE VA—3 manufactured by lnstitut za Elektroniko in Vakuumsko Tehniko Ljubljana/Yugoslavia The vacuum aspirator VA-3 is designed for hospital and ambulatory medical treatments, for various operative interventions; for aspiration of fluids and secretions, and for pregnancy interruptions. The de- vice operates almost noiselessly. Beside this advantage, aspiration can be very easily and quick regu- lated. In general use, the apparatus is equipped with an ordinary probe for aspiration and for gynecolo- gical interventions it has a special probe. Three different probe sizes are available. The apparatus is fitted with a rotary pump, which produces a vacuum up to 76 cm Hg. By means of a special valve, pressure may be continuously adjusted. The aspirator is equipped with a pressure gauge for the range from 0—76 cm Hg to control the pressure and with two two lite-r containers. The operator can control the pump motor by a foot-switch. TECHNICAL DATA — pumiping speed 331/min (rotary pump with a single-phase motor) — continuously adjustable pressure from O to 76 cm Hg — vacuum gauge measuring range 0 to 76 cm Hg — volume of container for aspirated material — 2 liters — apparatus movable on wheels —— power supply 220 V, 50 Hz, 250 W — weight approx. 32 kg Appendix 4: HOSPITALIZATlON FORM Study number: Protocol number: 1St Day: temperature: afebrile subfebrile febrile ‘ lower abdominal pain: no yes tenderness: no yes therapy: none antibiotics uterotonics 55 2nd Day: temperature: afebrile subfebrile febrile lower obdominal pain: no yes tenderness: no yes therapy: none antibiotics uterotonics Duration of hospitalization: days Hospitalization prolonged beyond 2 days: no yes Complications in hospital: no yes : injury inflammation bleeding Discharge date: Remarks: Doctor’s signature: * Afebrile: 37° C and less Subfebrile: 37.1° C — 375° C Febrile: 316° C and more Appendix 5: SIX-WEEKS QUESTIONNAIRE Study No.: 1. Days sick leave after discharge from hospital: 2. Bleeding after discharge? no yes days Did you see a physician because of bleeding? no yes Where Have you been readmitted to the hospital because of bleeding? no yes Where 3. Fever following discharge from hospital? no yes days Did you see a :physician because of fever? no yes Where Have you been readmitted to the hospital because of fever? no yes Where 4. Lower abdominal pain? no yes days Did you see a physician because of lower abdominal pain? no yes Where Have you been readmitted to the hospital because of lower abdominal pain? no yes Where 5. Back pain? no yes 6. First menstrual period after operation: not yet onset date: Duration in days: 7. First menstrual period compared with before operation: a) amount of bleeding: more same less b) duration of bleeding: longer same shorter c) menstrual pain: more same less 8. First intercourse following operation? no yes approximately date 9. Have you had the gynecological examination after operation? no yes When Where 10. What type of contraception are you using now? none IUD orals diaphragm chem. rhythm condom C. l. Appendix 6: SIX~MONTHS QUESTIONNAIRE Study No.: Please, answer the foil-owing questions and return the questionnaire and the menstrual calendar to the institute: 1. Menstrual periods following abortion: regular irregular 2. Menstrual flow compared with before abortion: duration of bleeding: same longer shorter menstrual pain: same less more 56 a) Abdominal pains following abortion: no yes b) Back pains following abortion: no yes Contraception used now: none IUD orals diphrag-m spermatocides rhythm condom coitus interruptus Please, do not hesitate to mention any other side effects you experienced in the period of six months. Social worker Appendix 7: ECTOPIC PREGNANCY FORM Study number Operation protocol No. Date: Delivery protocol No. 1. Surname and name: Maiden name: Home address: Telephone number: 2. Year of birth: 3. Onset of last regular menstruation: 4. Number of live births: 5. Number of stillbirths: 6. Total number of abortions: 7. Number of spontaneous abortions: 8. Number of induced abortions: 9. Number of prewi-ous ectopic pregnancies: 10. Date last pregnancy terminated: (month and year) 11. How terminated: first pregnancy: livebirth stillbirth induced aib. spont. ab. ectopic preg. 12. Last induced abortion performed by: no abortion D§C VA combination not known 13. Contraception used during last 12 months: none IUD orals diaphragm spermatocides condom coitus interruptus rhythm 14. Previous abdominal surgery: no yes Surgery of adinexis Caesarean section Myomeotomy Ectopic preg. Other X8 2 Study group P-Z P-S = Controls Appendix 8: PUBLICATIONS RESULTING FROM THIS STUDY 1. 2. Andoléek, L., Pretnar, A., Oven, M.: Immediate and Early Complications After Legal Abortion. Zbor- niik radova Vll. kongresa ginekologa-opstetriéara Jugoslavije, I, Beograd: 175, 1972. Kralj, B., Lavric, V., Sancin, A.: CerviCOuisthmian Insufficiency as a Complication of Artificial Abor- tion. Zbornik radova VII. kongresa ginekologa-opstetrioara Jugoslavije, ll, Beograd: 517, 1972. . Glonar, L., Hocevar, M.:The Effect of Aborti0n on Rh-isosensitization (Preliminary report). lbid II, 399, 1972. . Hren, M., Tomaievié, T.: Extrauterine Pregnancy as a Late Complication of Artificial Aibortion. lbid ll, Beograd: 507, 1972. Hccevar, M., Glona-r, L.: Isoimimuniization during the First Pregnancy. 1St International Congress on Immunology in Obstetrics and Gynaecology. Padova, June 7, 8, 9, 1973. It will be published by Excenpta Medica. Hren, M., Tomaievié, T.: The Importance of the Aborti0n in the Etiology of Ectopic Pregnancy (Pre- liminary Report). Zdrav. vestn. 43, 5—6 1974. 57 GENERAL UBflAflY-ILC.BEIKELEY BODDHfilL71