key: cord-0006408-cbelsymu authors: Gross, Peter A. title: Current Recommendations for the Prevention and Treatment of Influenza in the Older Population date: 2012-11-18 journal: Drugs Aging DOI: 10.2165/00002512-199101060-00003 sha: 31f4b44422268b514ecada6a2ad1c4ef16abc98f doc_id: 6408 cord_uid: cbelsymu Influenza is a major cause of morbidity and mortality in the elderly. Influenza vaccine is recommended for people aged 65 years and older and those in long term care. Currently only 30% of high risk persons are vaccinated. Vaccination generally stimulates an adequate immune response, is well tolerated and is to be encouraged. Prophylactic amantadine 100 mg/day should be given for 2 weeks with influenza vaccine in the aged population when they have not been previously immunised. Broad application of these preventive measures would have a significant impact on reducing influenza prevalence in the elderly and other high risk groups. Influenza is a major cause of morbidity and mortality in the elderly. Influenza vaccine is recommended for people aged 65 years and older and those in long term care. Currently only 30% of high risk persons are vaccinated. Vaccination generally stimulates an adequate immune response, is well tolerated and is to be encouraged. Prophylactic amantadine 100 mgjday should be given for 2 weeks with influenza vaccine in the aged population when they have not been previously immunised. Broad application of these preventive measures would have a significant impact on reducing influenza prevalence in the elderly and other high risk groups. If we were more aware of the full impact of influenza on our lives, we would be more concerned about preventing it. In the United States alone more than 10 000 deaths are attributed to influenza virus in an average epidemic year (Recommendations of the Immunization Practices Advisory Committee 1990). In some epidemics, the death toll has risen to more than 40 ODD. The majority of deaths -80 to 90% -occur in persons 65 years of age and older . The estimated economic loss from influenza in the United States is estimated to be more than Influenza is a major cause of morbidity and mortality in the elderly. Influenza vaccine is recommended for people aged 65 years and older and those in long term care. Currently only 30% of high risk persons are vaccinated. Vaccination generally stimulates an adequate immune response, is well tolerated and is to be encouraged. Prophylactic amantadine 100 mg/day should be given for 2 weeks with influenza vaccine in the aged population when they have not been previously immunised. Broad application of these preventive measures would have a significant impact on reducing influenza prevalence in the elderly and other high risk groups. If we were more aware of the full impact of influenza on our lives, we would be more concerned about preventing it. In the United States alone more than 10 000 deaths are attributed to influenza virus in an average epidemic year (Recommendations of the Immunization Practices Advisory Committee 1990). In some epidemics, the death toll has risen to more than 40 ODD. The majority of deaths -80 to 90% -occur in persons 65 years of age and older . The estimated economic loss from influenza in the United States is estimated to be more than $USI billion each year (Schoenbaum 1987) . In reality, the cost is probably 3 to 5 times greater. While much has been written on the cost-benefit ratio for influenza vaccine, in persons 65 years of age and above the benefit clearly outweighs the cost. Despite the favourable impression gained from economic models, however, individual patients and their physicians do not as a rule perceive the benefit. As a result, only a small percentage of the elderly population are vaccinated each year. What is influenza? Is it easy to diagnose? How effective is influenza vaccine? Are the side effects a frequent concern? What efforts have been and can be made to improve vaccine acceptance? These are some of the important questions we attempt to answer in this article. I Influenza virus was first isolated from humans in the 1930s (Ruben 1987) . The clinical syndrome of a typical case of influenza begins with fever, cough and myalgias that last for a few days to a week . The clinical course may be complicated by laryngotracheobronchitis and pneumonia. Secondary bacterial pneumonia may occur simultaneously or a few days later. Hospitalisation is more likely to occur in individuals who are elderly or have pre-existing cardiac or pulmonary disease. Children aged under 12 years taking aspirin during the early phases of infection are more likely to get Reye's syndrome. Late complications of infection such as fatigue, chronic cough, and small-airway disease may last for weeks. Making a diagnosis of influenza virus infection is difficult because the acute respiratory syndromes may be confused with other viral respiratory agents such as parainfluenza viruses, adenoviruses, respiratory syncytial viruses as well as the more common rhinoviruses and coronaviruses. Laboratory I Educational materials on influenza are available from the Centers for Disease Control [Technical Information Services, Center for Prevention Services, Mailstop E06, CDC, Atlanta, GA 30333, USA; phone (404) 639-18191 as well as from state and local health departments in the USA. Drugs & Aging 1 (6) 1991 tests can distinguish among these agents, and rapid, sensitive tests are now available to diagnose influenza virus (Reichelderfer et al. 1987) . A specific diagnosis of influenza can be made in a few hours to a few days, rather than in a week or more as was the case several years ago. While influenza virus infection may be difficult to distinguish from other respiratory viruses on clinical grounds, influenza virus is unique among the respiratory viruses for the high mortality associated with it. Since the sixteenth century, people have recognised the increased mortality resulting from an influenza epidemic (Monto 1987) . Excess mortality from influenza has been documented since 1889. On occasion, influenza outbreaks reach pandemic proportions, as occurred in 1918. Influenza epidemics occur almost every winter. Quantifying the mortality from influenza begins with describing the known seasonal variation in mortality from all causes, as shown in figure 1. To reduce the chance of error a dashed line equal to 1.645 standard deviations (the 'epidemic threshold') is drawn above the known season incidence line. When the mortality rate exceeds the standard deviation, it is considered significant. This so-called excess mortality is invariably associated with an outbreak of influenza. Excess deaths, therefore, are considered to be due to epidemics of influenza virus. Quantifying morbidity from influenza is more difficult. While excess mortality occurs in the elderly and those with certain chronic diseases, morbidity is generally more common among healthy children and young adults. The severe morbidity associated with influenza, such as hospitalisation, is most common among the elderly and chronically ill. A number of studies have documented the toll influenza exacts among the infirm and the elderly. The most recent large scale study on influenza-related mortality is by Barker and Mullooly (1980) . In 2 influenza A (H3N2) epidemics, they report that 11 to 13 excess deaths occurred per 100 000 persons. When they examined those aged 432 $US 1 billion each year (Schoenbaum 1987) . In reality, the cost is probably 3 to 5 times greater. While much has been written on the cost-benefit ratio for influenza vaccine, in persons 65 years of age and above the benefit clearly outweighs the cost. Despite the favourable impression gained from economic models, however, individual patients and their physicians do not as a rule perceive the benefit. As a result, only a small percentage of the elderly population are vaccinated each year. What is influenza? Is it easy to diagnose? How effective is influenza vaccine? Are the side effects a frequent concern? What efforts have been and can be made to improve vaccine acceptance? These are some of the important questions we attempt to answer in this article.! Influenza virus was first isolated from humans in the 1930s (Ruben 1987) . The clinical syndrome of a typical case of influenza begins with fever, cough and myalgias that last for a few days to a week . The clinical course may be complicated by laryngotracheobronchitis and pneumonia. Secondary bacterial pneumonia may occur simultaneously or a few days later. Hospitalisation is more likely to occur in individuals who are elderly or have pre-existing cardiac or pulmonary disease. Children aged under 12 years taking aspirin during the early phases of infection are more likely to get Reye's syndrome. Late complications of infection such as fatigue, chronic cough, and small-airway disease may last for weeks. Making a diagnosis of influenza virus infection is difficult because the acute respiratory syndromes may be confused with other viral respiratory agents such as parainfluenza viruses, adenoviruses, respiratory syncytial viruses as well as the more common rhino viruses and coronaviruses. Laboratory Drugs & Aging 1 (6) 1991 tests can distinguish among these agents, and rapid, sensitive tests are now available to diagnose influenza virus (Reichelderfer et al. 1987) . A specific diagnosis of influenza can be made in a few hours to a few days, rather than in a week or more as was the case several years ago. While influenza virus infection may be difficult to distinguish from other respiratory viruses on clinical grounds, influenza virus is unique among the respiratory viruses for the high mortality associated with it. Since the sixteenth century, people have recognised the increased mortality resulting from an influenza epidemic (Monto 1987) . Excess mortality from influenza has been documented since 1889. On occasion, influenza outbreaks reach pandemic proportions, as occurred in 1918. Influenza epidemics occur almost every winter. Quantifying the mortality from influenza begins with describing the known seasonal variation in mortality from all causes, as shown in figure 1. To reduce the chance of error a dashed line equal to 1.645 standard deviations (the 'epidemic threshold') is drawn above the known season incidence line. When the mortality rate exceeds the standard deviation, it is considered significant. This so-called excess mortality is invariably associated with an outbreak of influenza. Excess deaths, therefore, are considered to be due to epidemics of influenza virus. Quantifying morbidity from influenza is more difficult. While excess mortality occurs in the elderly and those with certain chronic diseases, morbidity is generally more common among healthy children and young adults. The severe morbidity associated with influenza, such as hospitalisation, is most common among the elderly and chronically ill. A number of studies have documented the toll influenza exacts among the infirm and the elderly. The most recent large scale study on influenza-related mortality is by Barker and Mullooly (1980) . In 2 influenza A (H3N2) epidemics, they report that 11 to 13 excess deaths occurred per 100 000 persons. When they examined those aged 65 years and older, the incidence of excess deaths increased to between 68 and 104 per 100 000. These estimates should in reality be even higher because pneumonia or influenza is often omitted from the diagnoses listed on death certificates . showed that the risk ofhospitalisation for acute respiratory disease (ARD) during influenza epidemics was 19.7 per 10 000 persons with high risk conditions for which influenza vaccine is indicated, compared to 9.3 per 10 000 for persons without these conditions. Chronic pulmonary disorders followed by chronic cardiac conditions were the most common highrisk conditions. Renal failure, diabetes and long term care residence are other risk factors. For persons over 65 years of age with a concomitant chronic pulmonary condition the hospitalisation rate for ARD soared to 87.5 per 10 000. The peak of hospitalisations for ARD typically followed by one week the peak of influenza virus isolations (Perrotta et al. 1985) . The disease severity varies according to the influenza strain and the age ofthe patient. The H3N2 subtype of influenza A causes the most severe ilness, while the HI N 1 subtype of influenza A is associated with the mildest illness (Monto et al. 1985) . Type B influenza causes an illness intermediate in severity between the 2 A subtypes. A third type of influenza, type C, is rarely responsible for epidemic disease. But why is influenza a problem year after year? Becoming infected with each of the 3 strains of influenza does not confer long-lasting immunity; in contrast, becoming infected naturally with, for example, measles does confer long-lasting immunity. Why the difference? There is only I strain of measles and its nonsegmented genome preserves its Prevention and Treatment of Influenza 65 years and older, the incidence of excess deaths increased to between 68 and 104 per 100 000. These estimates should in reality be even higher because pneumonia or influenza is often omitted from the diagnoses listed on death certificates . showed that the risk of hospitalisation for acute respiratory disease (ARD) during influenza epidemics was 19.7 per 10000 persons with high risk conditions for which influenza vaccine is indicated, compared to 9.3 per 10 000 for persons without these conditions. Chronic pulmonary disorders followed by chronic cardiac conditions were the most common highrisk conditions. Renal failure, diabetes and long term care residence are other risk factors. For persons over 65 years of age with a concomitant chronic pulmonary condition the hospitalisation rate for ARD soared to 87.5 per 10 000. The peak of hospitalisations for ARD typically followed by one week the peak of influenza virus isolations (Perrotta et al. 1985) . The disease severity varies according to the influenza strain and the age of the patient. The H3N2 subtype of influenza A causes the most severe ilness, while the HI N 1 subtype of influenza A is associated with the mildest illness (Monto et al. 1985) . Type B influenza causes an illness intermediate in severity between the 2 A subtypes. A third type of influenza, type C, is rarely responsible for epidemic disease. But why is influenza a problem year after year? Becoming infected with each of the 3 strains of influenza does not confer long-lasting immunity; in contrast, becoming infected naturally with, for example, measles does confer long-lasting immunity. Why the difference? There is only I strain of measles and its nonsegmented genome preserves its singularity. Influenza virus, on the other hand, has a segmented genome. The multiple RNA segments -there are 8 in all -predispose to the genetic instability of the influenza virus (table I, fig. 2 ). The nucleotide sequences in the RNA segments are inherently unstable, and when 2 different strains of the same subtype infect the same host the RNA segments can reassort, giving rise to progeny that are different from the 2 parents. The naming of influenza viruses is straightforward. For example, influenza A/Shanghai/16/89 (H3N2) is so named because this strain is a type A influenza virus first isolated in Shanghai in 1989. It was the sixteenth isolate of the subtype H3N2 detected in that year. Haemagglutinin (H) and neuraminidase (N) proteins exist in types A, B, and C influenza virus. But only haemagglutinin and neuraminidase proteins of influenza A are sufficiently diverse to warrant subtype designations. For the type A influenza viruses, 3 haemagglutinins (H\, H2 and H3) and 2 neuraminidases (N \ and N2) have been described. Currently, only 2 subtypes of influenza A are circulating and causing disease in humans, and they are the H3N2 and the HIN\ subtypes. The genetic changes that occur have been referred to by the terms antigenic drift and antigenic shift. Antigenic drift is a minor change that occurs (H3N2) is an example of this . Minor changes also occur when antigens are glycosylated and camouflage the previously exposed antigenic site. Antigenic shift is a major change that occurs by gene reassortment resulting in a new subtype. The change from the influenza A (H2N2) to the influenza A (H3N2) subtype that occurred in 1968 is such an example. Gene reassortment probably occurs when 2 different subtypes coinfect the same host. Serum antibody to the surface haemagglutinin and neuraminidase proteins is important in protection against infection and the development of disease. Humoral antibody to the haemagglutinin component is the most critical, but the importance of cell-mediated immunity is less clear . The humoral immune response is highly specific for certain epitopes on the surface of the virus. The cellular response, in contrast, is less specific and more cross-reactive. The B cells which produce humoral antibody recognise antigenic determinants, or epitopes, adjacent in space but not in sequence, while T cells recognise epitopes adjacent in sequence but not necessarily in space. The 434 singularity. Influenza virus, on the other hand, has a segmented genome. The multiple RNA segments -there are 8 in all -predispose to the genetic instability of the influenza virus (table I, fig. 2 ). The nucleotide sequences in the RNA segments are inherently unstable, and when 2 different strains of the same subtype infect the same host the RNA segments can reassort, giving rise to progeny that are different from the 2 parents. The naming of influenza viruses is straightforward. For example, influenza A/Shanghai/16/89 (H3N2) is so named because this strain is a type A influenza virus first isolated in Shanghai in 1989. It was the sixteenth isolate of the subtype H3N2 detected in that year. Haemagglutinin (H) and neuraminidase (N) proteins exist in types A, B, and C influenza virus. But only haemagglutinin and neuraminidase proteins of influenza A are sufficiently diverse to warrant subtype designations. For the type A influenza viruses, 3 haemagglutinins (HI, H2 and H3) and 2 neuraminidases (N I and N2) have been described. Currently, only 2 subtypes of influenza A are circulating and causing disease in humans, and they are the H3N2 and the HIN I subtypes. The genetic changes that occur have been referred to by the terms antigenic drift and antigenic shift. Antigenic drift is a minor change that occurs . Minor changes also occur when antigens are glycosylated and camouflage the previously exposed antigenic site. Antigenic shift is a major change that occurs by gene reassortment resulting in a new subtype. The change from the influenza A (H2N2) to the influenza A (H3N2) subtype that occurred in 1968 is such an example. Gene reassortment probably occurs when 2 different subtypes coinfect the same host. Serum antibody to the surface haemagglutinin and neuraminidase proteins is important in protection against infection and the development of disease. Humoral antibody to the haemagglutinin component is the most critical, but the importance of cell-mediated immunity is less clear . The humoral immune response is highly specific for certain epitopes on the surface of the virus. The cellular response, in contrast, is less specific and more cross-reactive. The B cells which produce humoral antibody recognise antigenic determinants, or epitopes, adjacent in space but not in sequence, while T cells recognise epitopes adjacent in sequence but not necessarily in space. The implication here is that B cell epitopes are intact external molecules such as haemagglutinins, while T cell epitopes appear to be linear sequences exposed by enzymatic breakdown of the original molecule. A killed (inactivated) vaccine against influenza virus was first developed in the 1940s. Since then the manufacturing process has resulted in a more pure vaccine. The virus is grown in embryonated eggs. Most of the egg protein is now eliminated and the side effects have been reduced markedly. Killed vaccines are referred to as either whole virus or split virus: whole virus vaccines are inactivated by formalin; split or subunit vaccines use whole virus which is treated with a detergent to free up the haemagglutinin and neuraminidase subunits from the viral surface, then formalin is added. In children who have not previously been vaccinated, split virus vaccines cause fewer side effects such as fever and myalgias than whole virus vaccines. Live attenuated vaccines have been in the development stages for over 10 years and are currently being tested; they are not commercially available. A genetically stable method of production was the initial problem but this has since been solved. However, it has been difficult to show a clear advantage of live over killed vaccines. The currently licensed influenza vaccine contains 3 strains: influenza A/Shanghai/I 6/89 (H3N2); influenza A/Taiwan/I/86 (HINl); and influenza B/ Yamagata/I 6/88. There are 15ILg of haemagglutinin for each strain in the vaccine. It should be given intramuscularly in the deltoid in adults and in the anterolateral thigh in infants and young children. Administration in the buttock is not recommended because it may result in a poorer immune response due to inadequate absorption from the subcutaneous fat. The strains to be included in the vaccine change from year to year according to World Health Organization (WHO) recommendations. 435 The primary target groups are as follows (Recommendations of the Immunization Practices Advisory Committee 1990). I. Persons aged 65 years or older. 2. Residents of nursing homes and other chronic-care facilities housing persons of any age with chronic medical conditions. of the pulmonary or cardiovascular systems, including children with asthma. 4. Adults and children who have required regular medical follow-up or hospitalisation during the preceding year because of chronic metabolic diseases (including diabetes mellitus), renal dysfunction, haemoglobinopathies, or immunosuppression (including immunosuppression caused by medications). I. Physicians, nurses and other personnel in both hospital and outpatient-care settings who have contact with high-risk persons in all age groups, including infants. 2. Employees of nursing homes and chronic-care facilities who have contact with patients or residents. 3. Providers of home care to high risk persons (e.g. visiting nurses, volunteer workers). Other groups merit consideration (Recommendations of the Immunization Practices Advisory Committee 1990): (a) in the general population, anyone who wishes to reduce the chance of acquiring influenza infection; (b) anyone who provides essential community services and persons living or working in an institutional setting; (c) persons infected with HIV should be vaccinated, although they may not respond as well if their disease is far advanced; and (d) foreign travellers should be immunised when travelling to the trop-Prevention and Treatment of Influenza implication here is that B cell epitopes are intact external molecules such as haemagglutinins, while T cell epitopes appear to be linear sequences exposed by enzymatic breakdown of the original molecule. A killed (inactivated) vaccine against influenza virus was first developed in the 1940s. Since then the manufacturing process has resulted in a more pure vaccine. The virus is grown in embryonated eggs. Most of the egg protein is now eliminated and the side effects have been reduced markedly. Killed vaccines are referred to as either whole virus or split virus: whole virus vaccines are inactivated by formalin; split or subunit vaccines use whole virus which is treated with a detergent to free up the haemagglutinin and neuraminidase subunits from the viral surface, then formalin is added. In children who have not previously been vaccinated, split virus vaccines cause fewer side effects such as fever and myalgias than whole virus vaccines. Live attenuated vaccines have been in the development stages for over 10 years and are currently being tested; they are not commercially available. A genetically stable method of production was the initial problem but this has since been solved. However, it has been difficult to show a clear advantage of live over killed vaccines. The currently licensed influenza vaccine contains 3 strains: influenza A/Shanghai/I 6/89 (H3N2); influenza A/Taiwan/I/86 (HIN J); and influenza B/ Yamagata/I 6/88. There are 15J.Lg of haemagglutinin for each strain in the vaccine. It should be given intramuscularly in the deltoid in adults and in the anterolateral thigh in infants and young children. Administration in the buttock is not recommended because it may result in a poorer immune response due to inadequate absorption from the subcutaneous fat. The strains to be included in the vaccine change from year to year according to World Health Organization (WHO) recommendations. 435 The primary target groups are as follows (Recommendations of the Immunization Practices Advisory Committee 1990). I. Persons aged 65 years or older. 2. Residents of nursing homes and other chronic-care facilities housing persons of any age with chronic medical conditions. 3. Adults and children with chronic disorders of the pulmonary or cardiovascular systems, including children with asthma. 4. Adults and children who have required regular medical follow-up or hospitalisation during the preceding year because of chronic metabolic diseases (including diabetes mellitus), renal dysfunction, haemoglobinopathies, or immunosuppression (including immunosuppression caused by medications). I. Physicians, nurses and other personnel in both hospital and outpatient-care settings who have contact with high-risk persons in all age groups, including infants. 2. Employees of nursing homes and chronic-care facilities who have contact with patients or residents. 3. Providers of home care to high risk persons (e.g. visiting nurses, volunteer workers). Other groups merit consideration (Recommendations of the Immunization Practices Advisory Committee 1990): (a) in the general population, anyone who wishes to reduce the chance of acquiring influenza infection; (b) anyone who provides essential community services and persons living or working in an institutional setting; (c) persons infected with HIV should be vaccinated, although they may not respond as well if their disease is far advanced; and (d) foreign travellers should be immunised when travelling to the trop-ics at any time of the year or from the Southern to the Northern Hemisphere between August and March and the Northern to the Southern Hemisphere between April and September. The vaccine is contraindicated in persons with anaphylaxis to eggs and in those with an acute febrile illness. A protocol is available for giving vaccine to allergic patients at high risk of influenza complications (Recommendations of the Immunization Practices Advisory Committee 1990). Other vaccines can be given along with influenza vaccine, such as pneumococcal, measlesmumps-rubella, Haemophilus inJluenzae type b, and oral polio vaccines. Pertussis vaccine, however, should not be given within 3 days of influenza virus vaccine. For optimal effect, the vaccine should be given in November in the Northern Hemisphere or May in the Southern Hemisphere, before the winter influenza season begins. Effective antibody levels develop within 2 weeks in most previously immunised persons. The levels decline a few months later, when they may be suboptimal. Studies investigating vaccine efficacy are difficult to conduct and to evaluate. They are hard to conduct because randomisation to control and vaccinated groups raises ethical questions in high risk patients. Evaluation is difficult because other respiratory infectious agents also cause colds, pneumonia, occasionally death, and often result in hospitalisation -the criteria used to determine efficacy. Vaccine immunogenicity and protective efficacy in the elderly have had mixed reviews. Immunogenicity, the vaccine's ability to stimulate protective levels of antibody, appears to be adequate in many studies. Protective levels of serum antibody are considered to be equal to or greater than the reciprocal of serum dilutions at 1 : 32 or 1 : 40, depending on the dilution method. Protective levels are usually achieved in 70 to 90% of vaccine recipients. reviewed 17 papers on this subject published between 1968 and 1988. A total Drugs & Aging 1 (6) 1991 of 30 comparisons were available in the 17 papers. In 10, the immune response was superior in young subjects; in 4, the elderly responded better, and in 16, no significant difference was noted between the young and old subjects; in this instance, the lack of differences may have been due to a type 2 error because the numbers studied were low. pointed out 3 methodological limitations. The studies included, firstly, subjects with illnesses or taking drugs, which affect the immune system; secondly, patients immunised previously with influenza vaccine; and finally, those with protective antibody levels before immunisation. Inclusion of patients with any of these 3 factors would tend to underestimate the ability of elderly subjects to respond to immunisation. The association between increasing age and a poor immune response to influenza vaccine has thus not yet been convincingly made. Perhaps a small subgroup of the elderly population respond poorly while most respond normally. This was originally suggested in a study by Phair et al. (1978) examining the T cell response of infirm elderly to influenza antigens. More recently, in a study by , bedridden, infirm elderly patients appeared to respond less well to influenza vaccine than did healthy, ambulatory elderly patients. Alternative approaches to the standard immunisation have been tried. Two doses 1 month apart did not improve the immune response , neither did 2 and 3 times the standard dose . From currently available information we can conclude that the recommended single standard dose of killed influenza vaccine will stimulate an adequate immune response in most elderly persons. Finally, and most importantly, in a number of recent studies influenza vaccine has been shown to reduce mortality from influenza virus infection Gross et al. 1988a ). Recent studies on the adverse effects of influenza vaccine show that the vaccine is typically well 436 ics at any time of the year or from the Southern to the Northern Hemisphere between August and March and the Northern to the Southern Hemisphere between April and September. The vaccine is contraindicated in persons with anaphylaxis to eggs and in those with an acute febrile illness. A protocol is available for giving vaccine to allergic patients at high risk of influenza complications (Recommendations of the Immunization Practices Advisory Committee 1990). Other vaccines can be given along with influenza vaccine, such as pneumococcal, measlesmumps-rubella, Haemophilus injluenzae type b, and oral polio vaccines. Pertussis vaccine, however, should not be given within 3 days of influenza virus vaccine. For optimal effect, the vaccine should be given in November in the Northern Hemisphere or May in the Southern Hemisphere, before the winter influenza season begins. Effective antibody levels develop within 2 weeks in most previously immunised persons. The levels decline a few months later, when they may be suboptimal. Studies investigating vaccine efficacy are difficult to conduct and to evaluate. They are hard to conduct because randomisation to control and vaccinated groups raises ethical questions in high risk patients. Evaluation is difficult because other respiratory infectious agents also cause colds, pneumonia, occasionally death, and often result in hospitalisation -the criteria used to determine efficacy. Vaccine immunogenicity and protective efficacy in the elderly have had mixed reviews. Immunogenicity, the vaccine's ability to stimulate protective levels of antibody, appears to be adequate in many studies. Protective levels of serum antibody are considered to be equal to or greater than the reciprocal of serum dilutions at 1 : 32 or 1 : 40, depending on the dilution method. Protective levels are usually achieved in 70 to 90% of vaccine recipients. reviewed 17 papers on this subject published between 1968 and 1988. A total Drugs & Aging 1 (6) 1991 of 30 comparisons were available in the 17 papers. In 10, the immune response was superior in young subjects; in 4, the elderly responded better, and in 16, no significant difference was noted between the young and old subjects; in this instance, the lack of differences may have been due to a type 2 error because the numbers studied were low. pointed out 3 methodological limitations. The studies included, firstly, subjects with illnesses or taking drugs, which affect the immune system; secondly, patients immunised previously with influenza vaccine; and finally, those with protective antibody levels before immunisation. Inclusion of patients with any of these 3 factors would tend to underestimate the ability of elderly subjects to respond to immunisation. The association between increasing age and a poor immune response to influenza vaccine has thus not yet been convincingly made. Perhaps a small subgroup of the elderly population respond poorly while most respond normally. This was originally suggested in a study by Phair et al. (1978) examining the T cell response of infirm elderly to influenza antigens. More recently, in a study by , bedridden, infirm elderly patients appeared to respond less well to influenza vaccine than did healthy, ambulatory elderly patients. Alternative approaches to the standard immunisation have been tried. Two doses 1 month apart did not improve the immune response , neither did 2 and 3 times the standard dose . From currently available information we can conclude that the recommended single standard dose of killed influenza vaccine will stimulate an adequate immune response in most elderly persons. Finally, and most importantly, in a number of recent studies influenza vaccine has been shown to reduce mortality from influenza virus infection Gross et al. 1988a) . Recent studies on the adverse effects of influenza vaccine show that the vaccine is typically well tolerated. For example, Margolis et a1. (1990a,b) reported that in elderly chronically ill patients fever and significant disability could not be attributed to the vaccine compared to a control group. A flu-like illness, however, was attributable to vaccine in 5.5%. They concluded that the overall frequency of adverse effects was low. Another area of concern is the effect of vaccine on drug metabolism. The mechanism postulated is vaccine suppression of oxidative hepatic metabolism. Interferon production induced by vaccination is supposed to block the oxidative cytochrome P450 system, resulting in accumulation of certain drugs. Metabolism of theophylline, chlordiazepoxide, lorazepam, warfarin, phenytoin and phenobarbital (phenobarbitone) has been studied, with variable results. For theophylline, in particular, most studies seem to indicate that the effect of vaccine is minor (Gomolin et a1. 1985; Grabowski et a1. 1985; Meredith et a1. 1985) . Less than 30% of high risk persons are vaccinated annually. Consequently many studies have been undertaken to determine the best strategies for immunising appropriate groups. The current strategies for implementing the influenza vaccine recommendations include immunising persons in the following settings (Recommendations of the Immunization Practices Advisory Committee 1990). 1. In outpatient clinics and physicians' offices label medical records of high risk persons; use mail or telephone reminders; begin vaccinating in September (Northern Hemisphere) [Buchner et a1. 1987] and March (Southern Hemisphere). 2. In facilities providing episodic or acute care (e.g. emergency rooms, walk-in clinics) healthcare providers be aware of high risk groups; provide written information in appropriate languages. 3. In nursing homes and other residential long term care facilities vaccinate all residents with the prior aggreement of attending physicians; do not require individual physician orders; obtain permission for vaccination on admission to facility. 437 4. In acute-care hospitals encourage vaccination before discharge between September and April (Northern Hemisphere), and March to August (Southern Hemisphere) of all persons aged 65 years or over, or with high risk conditions at any age . 5. In outpatient facilities providing continuing care to high risk patients (e.g. haemodialysis centres, hospital speciality care clinics, outpatient rehabilitation programmes); facilities providing services to people 65 years and older (e.g. retirement communities, recreation centres); clinics and other centres providing healthcare for travellers; healthcare workers, visiting nurses and others providing home care to high risk persons provide educational materials; encourage vaccination; emphasise to staff in intensive care units, medical/surgical units, employees in nursing homes and long term care facilities; make vaccine available for all work shifts, including night and weekend. Two antiviral agents are effective in the prevention and treatment of influenza type A infections -amantadine and rimantadine. Only amantadine is currently available commercially in the United States. It has no effect against influenza type B (Douglas 1990). The primary use of amantadine is in preventing influenza A infection. Preventive efficacy varies between 50 and over 90%. When influenza A is in the community, an unvaccinated elderly person should receive vaccine plus 2 weeks of amantadine at a dose of 100 mg/day (Nicholson & Wiselka 1991) . Amantadine does not interfere with the immune response to vaccine, and after 2 weeks the vaccine-induced immune response should be sufficient to prevent influenza. Amantadine treatment should not be stopped if the high risk person cannot be vaccinated or is immunodeficient. It should be continued for as long as influenza is epidemic -usually about 5 to 6 weeks. Amantadine has also been used for the treatment of influenza A. When the drug is started within 48 hours of the onset of illness it produces Prevention and Treatment of Influenza tolerated. For example, Margolis et al. (1990a,b) reported that in elderly chronically ill patients fever and significant disability could not be attributed to the vaccine compared to a control group. A flu-like illness, however, was attributable to vaccine in 5.5%. They concluded that the overall frequency of adverse effects was low. Another area of concern is the effect of vaccine on drug metabolism. The mechanism postulated is vaccine suppression of oxidative hepatic metabolism. Interferon production induced by vaccination is supposed to block the oxidative cytochrome P450 system, resulting in accumulation of certain drugs. Metabolism of theophylline, chlordiazepoxide, lorazepam, warfarin, phenytoin and phenobarbital (phenobarbitone) has been studied, with variable results. For theophylline, in particular, most studies seem to indicate that the effect of vaccine is minor . Less than 30% of high risk persons are vaccinated annually. Consequently many studies have been undertaken to determine the best strategies for immunising appropriate groups. The current strategies for implementing the influenza vaccine recommendations include immunising persons in the following settings (Recommendations of the Immunization Practices Advisory Committee 1990). 1. In outpatient clinics and physicians' offices label medical records of high risk persons; use mail or telephone reminders; begin vaccinating in September (Northern Hemisphere) and March (Southern Hemisphere). 2. In facilities providing episodic or acute care (e.g. emergency rooms, walk-in clinics) healthcare providers be aware of high risk groups; provide written information in appropriate languages. 3. In nursing homes and other residential long term care facilities vaccinate all residents with the prior aggreement of attending physicians; do not require individual physician orders; obtain permission for vaccination on admission to facility. 437 4. In acute-care hospitals encourage vaccination before discharge between September and April (Northern Hemisphere), and March to August (Southern Hemisphere) of all persons aged 65 years or over, or with high risk conditions at any age . 5. In outpatient facilities providing continuing care to high risk patients (e.g. haemodialysis centres, hospital speciality care clinics, outpatient rehabilitation programmes); facilities providing services to people 65 years and older (e.g. retirement communities, recreation centres); clinics and other centres providing healthcare for travellers; healthcare workers, visiting nurses and others providing home care to high risk persons provide educational materials; encourage vaccination; emphasise to staff in intensive care units, medical/surgical units, employees in nursing homes and long term care facilities; make vaccine available for all work shifts, including night and weekend. Two antiviral agents are effective in the prevention and treatment of influenza type A infections -amantadine and rimantadine. Only amantadine is currently available commercially in the United States. It has no effect against influenza type B (Douglas 1990). The primary use of amantadine is in preventing influenza A infection. Preventive efficacy varies between 50 and over 90%. When influenza A is in the community, an unvaccinated elderly person should receive vaccine plus 2 weeks of amantadine at a dose of 100 mg/day (Nicholson & Wiselka 1991) . Amantadine does not interfere with the immune response to vaccine, and after 2 weeks the vaccine-induced immune response should be sufficient to prevent influenza. Amantadine treatment should not be stopped if the high risk person cannot be vaccinated or is immunodeficient. It should be continued for as long as influenza is epidemic -usually about 5 to 6 weeks. Amantadine has also been used for the treatment of influenza A. When the drug is started within 48 hours of the onset of illness it produces a therapeutic effect. The duration of illness decreases by 50%, virus shedding decreases more rapidly, and peripheral airway resistance is reduced faster. Serum antibody still develops despite the use of amantadine. Amantadine treatment for the acute illness should be continued for 5 days. Studies on the therapeutic efficacy of amantadine have been done in healthy adults; there are no data on its efficacy in high risk persons (Recommendations of the Immunization Practices Advisory Committee 1990). Resistance of influenza A virus to amantadine has been reported . The adverse effects of amantadine at a dose of 200 mg/day involve the central nervous system (nervousness, anxiety, insomnia, difficulty concentrating and lightheadedness) and the gastrointestinal tract (anorexia and nausea). They usually decrease or stop after the first week of use or when amantadine is discontinued. Using a smaller dose, 100 mg/day, reduces the adverse effects, apparently without diminishing efficacy. The manufacturer's recommendations should be read carefully to adjust the dose for age, renal function, weight, and interactions with other drugs and diseases (see table II) . Rimantadine is a structural analogue of amantadine. It has 2 advantages over amantadine: it has fewer central nervous system side effects at doses of 200 mg/day, and, because most of the drug is Drugs & Aging 1 (6) 1991 metabolised, elimination from the body is not dependent on renal excretion. The use of influenza vaccine in the elderly is no longer an oPtion"to be exercised only occasionally. Administration of influenza vaccine and the use of amantadine when indicated should become a standard part of the care given our elderly population (Nicholson 1990 ). a therapeutic effect. The duration of illness decreases by 50%, virus shedding decreases more rapidly, and peripheral airway resistance is reduced faster. Serum antibody still develops despite the use of amantadine. Amantadine treatment for the acute illness should be continued for 5 days. Studies on the therapeutic efficacy of amantadine have been done in healthy adults; there are no data on its efficacy in high risk persons (Recommendations of the Immunization Practices Advisory Committee 1990). Resistance of influenza A virus to amantadine has been reported . The adverse effects of amantadine at a dose of 200 mg/day involve the central nervous system (nervousness, anxiety, insomnia, difficulty concentrating and lightheadedness) and the gastrointestinal tract (anorexia and nausea). They usually decrease or stop after the first week of use or when amantadine is discontinued. Using a smaller dose, 100 mg/day, reduces the adverse effects, apparently without diminishing efficacy. The manufacturer's recommendations should be read carefully to adjust the dose for age, renal function, weight, and interactions with other drugs and diseases (see table II) . Rimantadine is a structural analogue of amantadine. It has 2 advantages over amantadine: it has fewer central nervous system side effects at doses of 200 mg/day, and, because most of the drug is Drugs & Aging 1 (6) 1991 metabolised, elimination from the body is not dependent on renal excretion. The use of influenza vaccine in the elderly is no longer an option to be exercised only occasionally. Administration ~f influenza vaccine and the use of amantadine when indicated should become a standard part of the care given our elderly population (Nicholson 1990) . Experiences in the use and efficacy of inactivated influenza vaccine in nursing homes Impact of epidemic type A influenza in a defined adult population Underestimation of the role of pneumonia and influenza in causing excess mortality Effectiveness of inactivated influenza vaccine among non-institutionalized elderly persons Resistance of influenza A virus to amantadine and rimantadine: results of one decade of surveillance Antibody induced by influenza vaccines in the elderly: a review of the literature Influenza vaccination in community elderly: a controlled trial of postcard reminders Clinical manifestations and consequences of influenza Antimicrobial agents: antiviral agents Immunizations for health care workers and patients in hospitals Survey of underlying conditions of persons hospitalized with acute respiratory disease during influenza epidemics in Houston Lack of effect of influenza vaccine on theophylline levels and warfarin anticoagulation in the elderly The effect of split virus influenza vaccination on theophylline pharmacokinetics Association of in fluenza immunization with reduction in mortality in an elderly References Experiences in the use and efficacy of inactivated influenza vaccine in nursing homes Impact of epidemic type A influenza in a defined adult population Underestimation of the role of pneumonia and influenza in causing excess mortality Effectiveness of inactivated influenza vaccine among non-institutionalized elderly persons Resistance of influenza A virus to amantadine and rimantadine: results of one decade of surveillance Antibody induced by influenza vaccines in the elderly: a review of the literature Influenza vaccination in community elderly: a controlled trial of postcard reminders Clinical manifestations and consequences of influenza Antimicrobial agents: antiviral agents Immunizations for health care workers and patients in hospitals Survey of underlying conditions of persons hospitalized with acute respiratory disease during influenza epidemics in Houston Lack of effect of influenza vaccine on theophylline levels and warfarin anticoagulation in the elderly The effect of split virus influenza vaccination on theophylline pharmacokinetics Association of in fluenza immunization with reduction in mortality in an elderly population: a prospective study Immunization of elderly people with high doses of influenza vaccine Relation of chronic disease and immune response to influenza vaccine in the elderly Immunization of elderly people with two doses of influenza vaccine Effect of influenza vaccine on serum anticonvulsant concentrations Epidemiologic implications of changes in the influenza virus genome T-cell recognition of influenza viral antigens Impact of influenza epidemic on mortality in the United States from Frequency of adverse reactions to influenza vaccine in the elderly Frequency of adverse reactions after influenza vaccine Effects of influenza virus vaccine on hepatic drug metabolism Infleunza: quantifying morbidity and mortality Tecumsah study of illness. XIII: influenza infection and disease Influenza vaccination and the elderly Amantadine for influenza A Acute respiratory disease hospitalizations as'll.Ijleasure of impact of epidemic influenza Failure to respond to influenza vaccine in the aged: correlation with B-cell number and function Prevention and control of influenza. And addendum clarification Economical laboratory support system for influenza virus surveillance Prevention and control of influenza: role of vaccine Economic impact of influenza: the individual's perspective Prevention and Treatment of Influenza population: a prospective study Immunization of elderly people with high doses of influenza vaccine Relation of chronic disease and immune response to influenza vaccine in the elderly Immunization of elderly people with two doses of influenza vaccine Effect of influenza vaccine on serum anticonvulsant concentrations Epidemiologic implications of changes in the influenza virus genome T-cell recognition of influenza viral antigens Impact of influenza epidemic on mortality in the United States from Frequency of adverse reactions to influenza vaccine in the elderly Frequency of adverse reactions after influenza vaccine Effects of influenza virus vaccine on hepatic drug metabolism