INFLUENZA VIRUSES: CLINICAL 15 SPECTRUM AND MANAGEMENT Karl Nicholson

15.1 Introduction 15.1 INTRODUCTION 15.2 History Influenza is an acute, febrile respiratory illness 15.3 Pandemic influenza of global importance caused by influenza A or 15.3.1 Influenza pandemics - 1847 to B virus. Influenza in humans occurs in two 1977 epidemiological forms - pandemic influenza 15.4 Interpandemic influenza which results from the emergence of a new 15.4.1 Herald waves 15.4.2 Importance of children in the influenza A virus to which the population pos• spread of influenza sesses little or no immunity, so it spreads with 15.4.3 Influenza in residential homes a high attack rate in all parts of the world; and 15.5 Transmission and pathogenesis interpandemic influenza A or B, occurring as 15.5.1 Transmission sporadic infections, a localized outbreak, or 15.5.2 Infection and replication epidemic, the latter representing an outbreak 15.6 Clinical features in a given community which usually occurs 15.6.1 Adults abruptly, peaks within 2-3 weeks, lasts 5-6 15.6.2 Infants and children weeks, and is associated with a significant drift 15.7 The complications of influenza of the surface haemagglutinin and neu• 15.7.1 Infection in the elderly and raminidase antigens. Epidemics occur virtu• excess deaths ally every year almost exclusively in the 15.7.2 Influenza during pregnancy 'winter' months in the northern hemisphere and the effects on the fetus and (October to April), and May to September in offspring the southern hemisphere. Explosive epi• 15.7.3 Respiratory tract complications demics of influenza can exert an enormous toll 15.7.4 Complications in diabetics in terms of morbidity, mortality, and economic 15.7.5 Myocarditis and pericarditis and social costs. 15.7.6 Neurological complications Influenza has no pathognomonic features 15.7.7 Toxic shock syndrome so a precise picture of its impact was impossi• 15.7.8 Myositis, myoglobinuria and ble until the first isolation of influenza A virus renal failure in 1933 [170], and influenza B virus in 1940. 15.8 Antivirals Further advances came with the recognition of 15.8.1 Adamantanes influenza virus replication in hens' eggs in 15.9 References 1936 [20], discovery of the haemagglutinating Viral and Other Infections of the Human Respiratory Tract. properties of influenza virus in 1941, and the Edited by S. Myint and D. Taylor-Robinson. Published in 1996 by Chapman & Hall.ISBN 978-94-011-7932-4 subsequent development of serological 276 Influenza viruses: clinical spectrum and management methods based on haemagglutination-inhibi• 'influence' of the planets at times of respira• tion. In the absence of these tools, a combina• tory epidemics [178]. 'Influenza' was used in tion of the explosive nature of influenza, its England during the outbreak of 1743 [50], but tendency for seasonality, high attack rates, earlier references to the malady included and respiratory and systemic features allow an 'newe acquayantance', the gentle correction, overview of the disease since ancient times. epidemic catarrh, or catarrhal fever [32, 50].

15.2 HISTORY 15.3 PANDEMIC INFLUENZA Langmuir and colleagues [111] proposed that 15.3.1 INFLUENZA PANDEMICS - 1847 TO 1977 the plague of Athens in the years 430-427 BC The influenza pandemic of 1847-1848 was caused by influenza associated with toxic shock syndrome and named the association Review of mortality statistics for 'influenza' in 'the Thucydides syndrome'. The epidemic in London since 1840 reveals a 39-fold rise in 412 BC described by Hippocrates was proba• deaths during the fourth quarter of 1847 as bly influenza [158]. Clear historical accounts of compared with the average for the same quar• influenza in Italy, Germany and England can ter during the 7-year period 1840-1846 [151]. be traced back to the epidemic of 1173 [89]. The increase was 11-fold for the first quarter of Hirsch [89] and Thompson [181] provide 1848 as compared with the average for the accounts of epidemics from the twelfth and same period during the eight preceding years. sixteenth centuries respectively, and Hirsch Such abrupt increases in mortality, with out• [89] referred to 'pandemics' occurring over a breaks occurring in both eastern and western great part of the globe and noted their occur• hemispheres, suggest the occurrence of a pan• rence in the years 1510, 1557, 1580, 1593, demic associated with antigenic shift. London 1732-1733, 1767, 1781-1782, 1802-1803, mortality statistics and the 1847-1848 pan• demic enabled William Farr to provide the first 1830-1833, 1836-1837, 1847-1848, 1850-1851, estimate of influenza-associated mortality. 1855, 1857-1858 and 1874-1875. Interestingly, Farr noted that 'the epidemic was most fatal to several of these pandemics, such as the ones in adults and the aged' ... 'the mortality in child• 1781-1782, 1801 (1802-1803), 1830-1833 and hood was raised 83%, in manhood 104%, in 1847-1848, seem to have spread to Europe old age 247%' ... 'from the age 4 to 25, however, across Russia from the Far East and Hirsch the mortality was comparatively not much commented upon the 'regular progress of the increased' . disease from east to west'. Modern influenza surveillance has shown that the virus affects much of the globe in most years; 'pandemics' The influenza pandemic of 1889-1890 with high attack rates are associated with anti• The number of influenza deaths in London genic 'shift' of the virus rather than antigenic gradually decreased from 1739 during the 'drift', and such pandemics occur infrequently winter of 1847-1848 to less than 10 during each with two of the most recent (' Asian' and 'Hong winter from 1884 to 1889 [151]. Some 558 Kong') originating in the Far East. The short 'influenza' deaths were recorded during the intervals between several 'pandemics' in first quarter of 1890, and outbreaks of respira• Hirsch's review question whether they were tory disease with high attack rates were noted associated with antigenic shift. throughout the United Kingdom [151]. The The origin of the term 'influenza' is uncer• first cases were described in May 1889 in tain, although the chronicles of a Florentine North America and in Bokhara, and the family used it in reference to the possible Russian doctors at that time called the disease Pandemkinfiuenza 277 'Chinese' influenza. The outbreak reached 1890, beyond that during the three previous Western Europe via Central Asia and years, amounted to at least 5500 cases, but only European Russia and reached London by 213 of all deaths were attributed by the certify• November 1889. High attack rates throughout ing medical attendants to influenza [151]. Europe were associated with excess deaths as Table 15.1 shows the relative increase in compared with those in previous years. deaths from chronic conditions including dia• Seroprevalence studies carried out during betes, 'wasting', alcoholism, heart disease, the emergence of 'Asian' influenza in tuberculosis, and chronic bronchitis during 1957-1958 revealed the presence of pre-exist• the epidemic as compared with the average ing antibody in people aged ~ 71 years of age during the three previous years. These and [139]. This suggested that the viruses responsi• more recent observations identify 'risk factors' ble for both the 1889-1890 and 1957-1958 pan• and provide the basis for current vaccine rec• demics contained an H2 haem agglutinin and ommendations. A further interesting observa• led to the proposal that influenza virus might tion, which has been noted in more recent recycle. However, during the Asian influenza pandemics, was the increase in influenza mor• pandemic (1957-1958) there was no evident tality during the second and third waves in reduction in pneumonia and influenza death 1891 and 1892 as compared with the first. This rates with increasing age, implying that the may reflect an increasing awareness of elderly had little or no protection following influenza among attendant physicians at the prior exposure to related H2 viruses [166]. time or an increase in virulence of the virus. The 1889-1890 pandemic is the first pan• demic for which there are good public health Possible H3 pandemic during the late records. Data collected during this period con• nineteenth century firmed and extended those obtained during The number of 'influenza' deaths in England 1847-1848, notably that during periods of and Wales fell after 1892 to about 100 per mil• influenza activity, certified influenza deaths lion of the population in 1896 but exceeded represent only a small proportion of the 400 per million in 1900 [178]. Some historians excess. In Paris, the excess mortality between suspect the emergence of an H3 strain at the 15th December 1889 and the end of January turn of the century based on the finding by

Table 15.1 Increase in deaths during the 1889-90 epidemic in Paris. (Data abstracted from [151])

Cause of death Ratio, 1889-1990 (Dec 1S-Jan 31) to average number in the same period during 3 previous years *

Pneumonia 3.19 Paralysis 2.50 Pleurisy 1.99 'Wasting' 1.94 ,Alcoholism' 1.86 Heart disease 1.78 Acute bronchitis 1.67 Tuberculosis 1.58 Chronic bronchitis 1.44

• Period 15th December to 31st January 278 Influenza viruses: clinical spectrum and management independent investigators in Asia, Europe and garglings from patients with influenza in 1933 North America that a large number of elderly [170]. The subsequent recognition that persons, principally those over 75 years of age influenza virus would replicate in hens' eggs, in 1968, had high levels of antibody to the and the development of serologic techniques A;Hong Kong (H3) strain of influenza before provided the means to monitor the spread and the pandemic in 1968 [35,55,118,122]. Lower impact of influenza. The 'Asian' influenza mortality rates during the A;Hong Kong pan• pandemic was apparently first encountered in demic in 1968-1969 in the over 75s as com• an outbreak in the Kweichow province of pared with those aged 65-74 years provides China in February, 1957 [178] and the additional evidence for infection with an H3 causative virus - N Asian/57 (H2N2) - was first virus at the turn of the century [96]. identified in cases in Hong Kong in April 1957. The outbreak spread rapidly to South-East The influenza pandemic of 1918 Asia and Australasia. It travelled westwards to the Middle East and Europe and eastwards The first 18 years of the twentieth century saw through Japan to the USA. Influenza reached influenza death rates of ~ 300 per million of the Holland, apparently by sea, between the end population in England and Wales, but in 1918 of May and the beginning of July [138]. influenza mortality increased more than ten• Sporadic cases occurred in Britain during the fold to 3129 per million [178]. The pandemic in summer months, and outbreaks developed in 1918 was due to the appearance of a new virus many places in the latter half of August, but which seroepidemiological studies have identi• October was the peak month and the first fied as related to swine influenza (HI). The wave had virtually disappeared by the end of 1918-1919 pandemic is unequalled in recent his• November [52,148,194]. tory, causing an estimated 20 million deaths The attack rate caused by the N Asian virus worldwide and 675 000 in the USA alone [33]. was unusually high and varied from about The reason for the exceptionally high toll is 20% to almost 100% in different communities unclear, but unlike other outbreaks during the worldwide, with the higher rates (>50%) nineteenth and twentieth centuries high mor• occurring among school-age children, the tality was seen in otherwise healthy young institutionalized, and those living in crowded adults. During the 1889-1890 pandemic and conditions [45,52,69,138,194]. The attack rate in those since 1918-1919, age-specific mortality England and Wales averaged 31 % for all age curves showed a U-shaped pattern with higher groups, but the highest overall attack rate, death rates in the very young and the elderly, 49%, occurred in the group aged 5 to 14 years but in 1918, the mortality curve was W-shaped, [194]. Similarly, during the first wave in the with high fatality in young adults and at the USA, the overall incidence of infection extremes of age. This may not be unique in the history of influenza. During the pandemic of assessed serologically was 55%, with rates of 1782 several observers noted ' ... I think that the 78% among 10 to 14-year-olds and 24% among middle age felt it most' ... 'I mean from 16 to 45 all adults [103]. The incidence of clinical or so'; 'children and old people escaped entirely influenza in Kansas and Louisiana was or were affected in a slighter manner' [50]. approximately 60% in high-school children, whereas the attack rate was 10-20% in adults above the age of 30 years [23]. Children were The influenza pandemic of 1957 paramount in the spread of infection during The research effort generated by the 1918 pan• the 1957 pandemic. In a Kent general practice, demic culminated in the successful transmis• families without school-children suffered sion of human influenza to ferrets using about half the incidence of those with school- Pandemkinfluenza 279 children (18% versus 33%), and attack rates in During the influenza A virus epidemic of 1953 the elderly were three times greater in house• and the B virus epidemic of 1955 Fry noted holds with school-children (31 %) when com• chest complications in 10%, whereas in 1957 pared with childless households [194]. only 3% of patients had comparable complica• From 21st August until the end of the year tions [52]. Woodall, Rowson and McDonald the number of new claims for national insur• [194] similarly observed an incidence of 3% for ance sickness benefit in Britain was about 2.5 pneumonia and 2% for otitis media among 187 million more than average for the same peri• patients with serologically confirmed influenza ods during the previous 5 years - that is, an in a general practice. increase equivalent to about 12.5% of the Although elderly individuals had antibod• insured population [194]. Data collated by the ies to previously circulating H2 viruses in 1957, Ministry of Health in London indicated that there was no evident reduction in pneumonia 7.5-9 million persons suffered some incapacity and influenza deaths with increasing age from influenza and that at least 5.5 million vis• [166]. Indeed in Fry's general practice the inci• ited a doctor. To illustrate the effect of the epi• dence of bronchitis and pneumonia actually demic in general practice, Fry [52] in his increased from 2-3% in those aged < 50 years, practice compared the weekly volume of work to 5% in those 50-59 years, 9% in those aged for common respiratory infections during 60-69 years, and 15% in those over 70 [52]. September and October 1957 with the corre• Elsewhere in Britain the incidence of pul• sponding weeks over the previous 5 years. At monary complications was 5% of 542 patients the peak of the epidemic in 1957 there were 3.6 in a general practice, [42] and 5% of 391 RAP times as many attendances as previously. recruits [59]. Asthma exacerbations occurred in Sickness statistics for a large factory in south• 38% of 124 people with atopy and influenza in east London showed that the proportion of a general practice in Barrow in Furness [12]. employees who were absent also rose approxi• There is remarkably little published work mately four-fold to 23% during the influenza on influenza as met in hospitals in England epidemic compared with a level of about 6% and Wales during autumn 1957. Most admis• in the recent past [194]. The demand for hospi• sions with influenza were for the complication tal accommodation appears to have been of pneumonia, but some people with uncom• heavy. In Liverpool during September and plicated influenza were admitted because of October 1957 the number of discharges from difficulty of treatment at home [l00]. During general medical departments increased by the 6-week period commencing 1st October 27% compared with the average for the same 2 1957, 262 (48%) of 541 patients admitted to months during the years 1953 to 1956 inclusive Dundee hospitals had influenzal pneumonia [128]. During the period 23rd September to 5th based on clinical and radiological evidence, November, the London Emergency Bed and an additional 135 (25%) had abnormal Service dealt with 2.8 times as many patients chest signs [l00]. On admission Streptococcus suffering from acute respiratory disease as pneumoniae and Haemophilus influenzae were compared with the same period during the isolated from 13% and 8% of cases, respec• three previous years [128]. tively, and staphylococci from only 2.5%. The 1957 Asian pandemic was generally However, after 4 and 7 days of hospitalization regarded - on the basis of complications and staphylococci were isolated from 32% and 36% deaths - as rather mild. A much higher rate of implying nosocomial acquisition. The overall complications occurred during the 1950-1951 death rate was 6%, but was 11% in those with epidemic when 20% of those seen in a general pneumonia and 1 % in the remainder. Among practice developed chest complications [52]. those aged <20 years the case-fatality rate was 280 Influenza viruses: clinical spectrum and management

0.5%, rising to 2.2% in those aged 20-49 years, 477 fatal cases from over 100 hospitals [159]. and 14% in those over 50 [100]. Of 30 patients The disease progressed more rapidly in the dying from pneumonia, eight (27%) died young than the old. From the onset of pneu• within 24 hours of admission, 11 (37%) within monia deterioration was usually rapid and 86 48 hours, and 18 (60%) within 96 hours. All of (18%) of the 477 patients died before admis• these patients were gravely ill on admission sion to hospital. An outstanding feature was and the part played by the nosocomially the speed with which patients died after acquired staphylococci was difficult to assess. admission - two-thirds died within 48 hours of Elsewhere the presence of staphylococci arrival, implying that in most instances the rendered the prognosis grave. Oswald, staphylococci were acquired in the commu• Shooter and Curwen [148] compared fatality nity. In this series virtually no deaths occurred in relation to age and sex in a series of 155 in children or young adults in the absence of S. staphylococcal and 145 non-staphylococcal aureus, whereas in the very young and elderly, influenzal pneumonias. The mortality was in whom 'milder' infections might well be 28% for staphylococcal pneumonia, 12% for fatal, staphylococci were found less often. It is non-staphylococcal pneumonia, and the clear that H. influenzae, despite its prominence greater fatality rate of staphylococcal pneumo• in the past did not play an important part in nia was observed at all ages. Among survivors, 1957, even though technical difficulties may staphylococcal pneumonia was the more have led to some underestimate of frequency. severe with a median stay in hospital of 26 The number of deaths arising from the days compared with 16 days for the others. As influenza pandemic is not known with cer• in Dundee [100], there was evidence of noso• tainty. In Britain, the average number of comial acquisition of staphylococci and of deaths for the years 1950-1956 in England and multiple antibiotic resistance; 93% of staphylo• Wales for the September and December quar• coccal isolates were penicillin-resistant, 74% ters was 228 760, whereas during this period in were resistant to tetracycline, 67% were resis• 1957 it was 262191. The difference between tant to sulphonamides, and 62% were resistant these figures (33431) has been taken as a to streptomycin. However, most were sensi• rough estimate of the toll in deaths [128]. In tive to chloramphenicol and erythromycin. the USA the excess deaths from influenza Overall, 49% of staphylococci isolated within 4 were estimated as 35 400 during October to days of admission - which were presumed to December 1957, and 34100 during January to have been acquired outside hospital - were March 1958 [95]. penicillin-resistant [148]. Reports from different parts of the world The influenza pandemic of 1968 showed that the case fatality varied greatly from place to place. In Malaya it was about 1 in An increase in influenza was noted in Hong 40000 and in Pakistan around 1 in 46000, Kong during mid-July 1968. The virus isolate whereas in the Philippines it was 1 in 220 reacted poorly with antiserum to 'Asian' [163]. In Abadan, Iran it was estimated at 1 in strains and was sent to WHO reference centres 9000 [163], in Kuwait 1 per 1000 [69] and in in London and Atlanta where it was con• Aden 1 in 350 [45]. Estimates of the case-fatal• firmed that the strain differed substantially ity rate in Britain ranged from 13 to 35 deaths from previous isolates. Although it was not the per 10 000 [128]. The most comprehensive usual season for such outbreaks, the virus review of fatal cases in England and Wales spread rapidly and involved about 500 000 was undertaken by the Public Health persons [189]. The peak incidence in Hong Laboratory Service, which received records for Kong occurred in late July, only a fortnight Pandemkinfluenza 281 after the first isolation of the virus, and the 1957 [112]. Pneumonia was a complication in whole outbreak was over in 6 weeks. 20 (16%) of cases. Excess mortality from all The new virus appeared in Singapore and causes above that expected from seasonal vari• the Philippines in August and by the end of ation was estimated at 56 300 in the USA dur• the month there had been outbreaks in Taipei, ing the 12-month period 1st October to 30th Taipan, Kuala Lumpur, Malaysia and Vietnam September, 1968--1969, but at only 9500 for [164]. By September it had spread to northern 1969-1970, and there was a deficit of deaths Australia, and also to Madras, India, after during 1970-1971. arrival of a ship from Singapore with 16 cases Hong Kong virus reached Europe in of influenza on board. A week later influenza autumn 1968, but no widespread outbreaks was reported in Bombay. The virus was intro• were reported until the winter and the disease duced into Japan in August and September was generally mild. Poland was probably the but, as in the USA and Britain, there was little hardest hit country with an estimated four mil• general spread until the onset of cooler lion cases of influenza-like illnesses [24]. The weather in the autumn. virus was isolated in London in August from a The Hong Kong virus was introduced into solitary family infection, and in September the USA in September and October by people there was a small school outbreak, but then the returning from the Far East and sporadic cases infection apparently halted [177]. In January were subsequently reported throughout the 1969 an increase in national insurance claims country. Outbreaks were first noted in began which plateaued in March and April California; there was a gradual spread east• [162]. Returns from the Royal College of wards throughout November and December General Practitioners in England indicated and by the end of the year all states were that, unlike the situation in Hong Kong, the involved. A retrospective questionnaire sur• epidemic was prolonged, lasting from late vey of 6994 students from a high school in December to early April; the incidence did not Kansas City, Missouri, and their families, for reach high levels in anyone week and there histories of influenza-like illness indicated an was no great pressure on hospital accommoda• overall attack rate of 39% [36]. In contrast to tion. Deaths from pneumonia and other causes the Asian pandemic in 1957 and studies were little elevated and only in March 1969 did reported by many other authors, the age-spe• the mortality curve suggest the existence of an cific attack rates of all age groups were similar influenza epidemic [177]. and adults were as likely as school-children to Hope-Simpson reported his experience of have been the first family case. This epidemio• the first wave of Hong Kong influenza as it logical observation has implications for the use presented in a Gloucestershire general prac• of influenza vaccine as it has been suggested tice with a list size of 3620 persons [91]. The that vaccination of school-children should pre• outbreak lasted 13 weeks, from 15th January to vent dissemination of influenza virus in the 15th April 1969, and is estimated to have community. Overall, some 27% of the popula• attacked only 4% of the practice population. tion in the USA were ill during the epidemic Within households, the secondary attack rate and schools had 50% absenteeism [168]. A was 17%. As in the USA, liability to infection review of the clinical and pathological features was almost independent of age except that in 127 adults - 93 outpatients, 23 who were persons over 65 years were less frequently more seriously ill and were admitted, and 11 attacked. In household outbreaks neither patients who acquired influenza nosocomially school-children nor children under school age in the Mayo Clinic - suggested that the mani• were more commonly the first case than their festations were similar to Asian influenza in elders, and the majority of contacts evidently 282 Influenza viruses: clinical spectrum and management escaped disease. The difference in attack rates HINI and H3N2 subtypes are circulating between different countries was remarkable; throughout the world. indeed, during the conference on Hong Kong influenza in October 1969, it was suggested 15.4 INTERPANDEMIC INFLUENZA that two virus strains were circulating simulta• The interpandemic period is represented by rel• neously - the more virulent one in the USA, atively minor, but immunologically significant and a less virulent one in Europe. It is now rec• changes in the haemagglutinin or neu• ognized that mutations, including substitu• raminidase of influenza virus types A and B. tions, deletions and insertions, are extremely Epidemics occur with cycles every few years, important in producing variation in the but a recent feature has been the co-circulation behaviour of influenza viruses. It is therefore of one or more strains of H3N2 and HINI sub• conceivable that geographic variation in types of influenza A virus together with strains had occurred and that the virus in influenza B virus. All three types may be Britain became more virulent or pneumotropic detected nationally in anyone year but it is overtime. usual that one type predominates and annual Although the first wave of Asian influenza outbreaks of influenza are not uncommon. Not only in large countries such as the USA and was mild, infection reappeared simultaneously Australia, but also in smaller ones such as France as a major epidemic in Britain, Spain, Portugal, where there is extensive reporting by medical Switzerland, Norway, Italy, Yugoslavia and practitioners, it is not uncommon to observe France in November, 1969. In contrast to the major outbreaks in some regions while others previous year's experience mortality was experience little or no activity - at least for a heavy, the deaths reported during the first 5 week or two. The spread of influenza through a weeks of the epidemic amounted to over 8700 community typically produces a bell-shaped and the total for the epidemic exceeded that of curve of the incidence of visits to medical practi• all individual outbreaks since 1951 [177]. The tioners for influenza and influenza-like illness. It age-specific attack rate was concentrated upon usually lasts 6-8 weeks and depending upon those aged 40--60 years and chest complica• the epidemic's size there are concomitant tions occurred at a rate of 25% [53]. increases in claims for sickness benefit, hospital admissions, school absenteeism and death regis• The HINl'pandemic' in 1977 trations. Interpandemic influenza is responsible for considerable morbidity and mortality which Significant pandemics have arisen with major exceeds that associated with the introduction of changes in both the haemagglutinin and neu• the pandemic strain. raminidase, but an exception to this general• Influenza infections are often identified ization occurred when NUSSR/77 (HINl) did during years when no excess mortality is not cause a pandemic - this may have been detected. Silent transmission of virus for a because much of the world's population in long period in a small community was demon• 1977 were alive before 1957 during the previ• strated in 1935, more than 2 months after the ous HINI era. Most people over the age of 23 last ship left a Greenland village [99]. years possessed antibodies to its surface Influenza occasionally persists beyond its nor• haemagglutinin antigen and the new virus mal season, and in the absence of recognizable was confined almost entirely to children and community outbreaks of disease [71,82,109,133, teenagers. Previously, when new influenza A 180]. Failure to culture influenza virus from subtypes emerged the preceding antigenic over 20 000 specimens collected during the subtype disappeared rapidly throughout the summer from persons with febrile respiratory world, but not on this occasion. Currently both disease in Houston over 11 years of commu- Interpandemic influenza 283 nity surveillance suggests that persistent infec• of antigenic shift and drift - the following tion in a community seldom occurs [31]. observations serve as examples. Influenza outbreaks occur nearly all months of Buchan and Reid [18] describing the impact the year somewhere in the world, and the of influenza A virus on the Scottish island of introduction of influenza virus into one coun• Lewis in 1969-1970, observed that the presence try from another is generally accepted as the of school-children in a household almost dou• origin of epidemics. bled the chance of infection in the household. Intensive surveillance in Seattle, Washington, 15.4.1 HERALD WAVES of families with school-age children during In certain years, viruses similar to those 1975-1979 encompassed type B virus infections appearing in the next year have appeared at in 1975-1976, type A (H3N2) during 1975-1976 the end of outbreaks. These so-called 'herald and 1977-1978, and type HINI in 1978-1979 [48]. Age-related infection rates were highest in waves' have been observed in some years in community studies in several geographic loca• children aged 5 to 9 years for H3N2 viruses and tions within the USA [47,49]. For example, in in teenagers for HINI and type B viruses. During the 1975-1976 season the overall infec• Houston during the influenza ANictoria epi• demic of 1976, 34 infections with influenza tion rate was 18.4% for influenza type A (H3N2) and 16.6% for type B virus. The rates B;Hong Kong virus were detected and this fell during 1976-1977 to 5.8% for influenza virus became epidemic the following year [62]. type A (H3N2) and 1.6% for influenza B virus, Similarly a herald wave of 48 Nfexas• but during 1977-1978 they increased to 23.7% ANictoria (H3N2) virus infections occurred for influenza type A (H3N2), 4.8% for influenza during the latter half of the influenza B;Hong type A (HINl), and 3.2% for influenza B virus. Kong epidemic of 1977 [63]. These H3N2 Finally during 1978-1979 the rate for influenza viruses were epidemic during the next season. type A (H3N2) fell to 0.4%, but for influenza In Japan, during the 5-year period 1985-1990, type A (HINl) it rose to 30.6%, and was 2.4% the influenza virus type or subtype isolated in for influenza B virus. Thus, during three of the spring was shown by HA gene sequenc~s four periods of virus activity influenza affected to be in most cases genetically close to the epl• almost one-third of the population, the great• demic strain of the next influenza season [142]. est burden occurring in young children. These same investigators compared the HA Houston families with and without infec• gene sequences of influenza B virus isolates tion were also studied during a period of during herald waves and later epidemic sea• ANictoria/75 (H3N2) virus activity in 1976 sons in the same districts in Japan in 1987 and [179]. The overall frequency of infection in 1989 [143]. Analysis revealed that the herald family members was 27.7%. Three determi• viruses in one wave (1987) were genetically nants characterized those families who were close to the winter isolates and were consid• infected with the virus: family size, density, ered to be the parental viruses for the follow• and the presence of school-age children or ing influenza season, while in the other wave children in day-care facilities. Indeed, as in the (1989) winter isolates were genetically and study of Buchan and Reid in Lewis [18] the antigenically different from the herald viruses. incidence of infection and illness attributed to influenza among persons in families with 15.4.2 IMPORTANCE OF CHILDREN IN THE school-children (38.5% ) was more than double SPREAD OF INFLUENZA that in families without children attending Pre-school and school-age children are gener• school or day-care facilities (16.9%). However, ally major vectors of infection during periods during the Hong Kong pandemic it was noted 284 Influenza viruses: clinical spectrum and management the age-specific attack rates of all age groups residents have been reported and many die or were similar and adults were as likely as chil• require hospitalization for severe compli• dren to have been the first family case [36,91]. cations [4]. During the winter of 1982-1983 in Influenza virus certainly has the potential Genesee County, Michigan, Patriarca et al. for rapid spread within schools. During the [153] compared the characteristics of seven pandemic in 1957, high attack rates were the nursing homes with outbreaks caused by rule in the closed communities of residential NBangkok/1/79 (H3N2)-like viruses with schools in England and Wales, frequently those of six homes in which there were spo• reaching 90% and often affecting the whole radic infections only. The homes were similar school within a fortnight [128]. Blunting of the in many respects including the physical char• epidemic curve and delay in the upswing until acteristics of the facilities, visiting and staffing classes re-start have been observed when patterns, infection control practices and demo• influenza first appeared in the community graphy of the residents. However, homes with before the Christmas holidays [60]. During epi• outbreaks were almost twice the size of those demics, a consistent finding has been a shift in without and a lower proportion of residents the age-distribution of persons with virus-posi• were vaccinated. When assessed by multivari• tive cultures [61]. During the early phase of ate analysis, the number of susceptible individ• epidemics a disproportionate number of cases uals was the most important factor in have been older school-age children in the 10 predicting the occurrence of an outbreak. to 19-year age range, but as epidemics progress Arden et al. [5] conducted a similar study in 64 an increasing proportion of infections were nursing homes in Michigan. Preliminary analy• detected in adults and pre-school children. sis revealed a mean vaccination rate of 71 % in A similar shift in age distribution seen with 22 outbreak homes and 79% in 39 homes that hospital admissions and influenza deaths dur• did not experience an outbreak (non-significant ing the first wave of Asian influenza further difference). Overall, 25% of homes with vacci• supports the concept that influenza virus first nation rates of 80% or more experienced an spreads mostly in children and then among outbreak of pyrexial influenza-like illness, younger sibs and adults [100,156]. A review of implying that target vaccination rates of 80% or 541 patients admitted to hospitals in Dundee, greater can not guarantee herd immunity. revealed that during the first week patients under 20 years of age predominated (52 of 88, 15.5 TRANSMISSION AND PATHOGENESIS 59%); this contrasted with subsequent weeks when the number of patients over 50 years of 15.5.1 TRANSMISSION age increased, reaching a maximum of 55% (21 Influenza is spread via virus-laden respiratory of 51) in the fifth week [100]. In Holland dur• secretions from an infected to a susceptible ing the early part of the epidemic 146 of 301 person. It is generally accepted that influenza deaths (49%) occurred in under-29-year-olds; viruses are transmitted by droplets several during the mid-phase 213 of 767 deaths (28%) microns in diameter that are expelled during occurred in this age group; and during the late coughing and sneezing, rather than fine• epidemic period 25 of 137 deaths (18%) were droplet nuclei. Influenza virus can evidently among those aged up to 29 years [156]. survive drying at room temperature for some days and has been demonstrated in dust for as 15.4.3 INFLUENZA IN RESIDENTIAL HOMES long as 2 weeks [43]. The human infectious Influenza A and B viruses are important causes dose 50 (HIDso) of influenza A virus is 127-320 of infection within nursing homes and chronic tissue culture infectious dose 50s (TCIDso) for care facilities. Outbreaks affecting over 60% of virus administered by nosedrops [29,30] but is Clinical features 285 significantly less, 0.6-3.0 TCID5(), for virus dren virus shedding at high titres is generally delivered by aerosol [2]. Pathological evidence more prolonged and virus can be recovered suggests initial or early involvement of pul• up to 6 days before and 21 days after the onset monary alveolar cells, a site only accessible to of symptoms [51,72,74]; in adults the quantity droplets up to 5 f.Lm in diameter. of virus shedding is related to the severity of illness and temperature elevation [140]. 15.5.2 INFECTION AND REPLICATION Because of the generalized symptoms present in uncomplicated influenza, viraemic spread The binding of influenza virus to erythrocytes from the respiratory tract has been suspected and to host cells is mediated by interaction of but attempts to demonstrate viraemia have the viral haemagglutinin with cell surface been inconclusive - a few investigators have oligosaccharides containing sialic acid. Human demonstrated viraemia, even before the onset H3 isolates have been shown to bind almost of symptoms [107,141,174] but other attempts exclusively to sialyloligosaccharide structures have been unsuccessful [108,131,136]. terminated by the SA a-2, 6-Gal sequence and The incubation period of influenza ranges are sensitive to inhibition of binding and infec• from 1 to 7 days but is commonly 2 to 3 days. tion by a glycoprotein present in non-immune This short period, coupled with the relatively horse serum. Histological studies of fatal cases, high titres in nasopharyngeal secretions, the nasal exudate cells, and tracheal biopsies indi• fairly lengthy periods of virus shedding (espe• cate that virus replication may occur through• cially in children), and the relatively small out the entire respiratory tract [87,112,114,116], amounts necessary to initiate infection, explain the principal site of infection occurring in cili• the explosive nature of influenza outbreaks. ated columnar epithelial cells [86,187]. It has been suggested that infection begins in the tra• 15.6 CLINICAL FEATURES cheobronchial epithelium and then spreads. Indeed, lesions in the tracheobronchial 15.6.1 ADULTS mucosa have been identified in bronchoscopic Infection with influenza viruses can result in a biopsies from young adults with uncompli• wide spectrum of clinical responses ranging cated Asian influenza [187] which correspond from asymptomatic infection to fulminant pri• with, but are less severe than, those found in mary viral pneumonia (Table 15.2). The onset the trachea and bronchi of fatal cases [86]. of illness is typically abrupt after an incubation In vitro studies suggest that the cycle of period of several days which may depend in replication takes about 4-6 hours [165]. part on the severity of exposure and the Thereafter virus is released for several hours immune status of the individual. Review of before cell death and progeny virions initiate the symptoms and signs of illnesses due to infection in adjacent cells, so that within a HON1, H1N1, H2N2, and H3N2 viruses short period many cells in the respiratory tract reveals no important differences [19,52,90,112, are either infected, releasing virus, or dying. 164,176]. However, there have been differ• The pattern of virus replication in relation to ences in the proportion and character of the clinical symptoms and immune responses has complications that have been noted. been studied by several investigators. Virus Influenza B virus can also cause the same can be detected shortly before the onset of ill• spectrum of disease as that seen following ness, usually within 24 hours, rises to a peak of influenza A virus infection, and the frequency 103-107 TelOs/ml of nasopharyngeal wash, of severe influenza B virus infections requiring remains elevated for 24-72 hours, and falls to hospital admission may be several-fold less low levels by the fifth day [140]. In young chil- than that for influenza A [123] or may be similar 286 Influenza viruses: clinical spectrum and management

Table 15.2 Clinical features of influenza illness in serologically confirmed family outbreaks. (Data abstracted from [194]).

Asian influenza (% of 186 cases with feature)

Onset Gradual 54 Sudden 46 Symptoms Cough 87 Headache 70 Sneezing 67 ~asalsymptoms 65 Sore throat 57 Sweating 48 Shivers 46 Aches and pains 42 Malaise 15 Prostration 14 Drowsy 11 Delirium 10 ~ose bleeds 9 Faint and giddy 7 Hoarseness 5 Vomiting 18 Abdominal pain 7 Diarrhoea 3 Complications Pneumonia 3 Otitis media 2 Most frequent initial symptom Headache, sore throat Most troublesome symptoms Headache, cough Days fever (median) 3 Days in bed (median) 3

[154]. In Tecumseh, Michigan, longitudinal Systemic features are often evident initially studies of influenza in the community showed and include shivering, feverishness, headache, that type B influenza was associated with ill• myalgia affecting the back and limbs, malaise nesses intermediate in severity compared with and anorexia. Headache and sore throat are those associated with influenza A viruses and the most frequent initial symptoms; headache, rhinoviruses [133]. This pattern was unrelated myalgia and cough are usually the most trou• to age. During the period 1976--1981 influenza blesome. The early features are almost invari• A H3N2 viruses produced the most severe ill• ably accompanied by a non-productive cough, nesses, HINI viruses the mildest, and type B sneezing, nasal discharge or obstruction and infections were intermediate [134]. less frequently by productive cough and sub- Clinical features 287 sternal soreness. Photophobia and other ocu• iting, abdominal pain and diarrhoea lar symptoms including lacrimation, burning [60,84,85,157). Infection in neonates may and pain on moving the eyes and nausea manifest itself as a pyrexial condition of occur in up to 20% of cases, hoarseness and uncertain aetiology [126); it may present as abdominal pain are reported by less than 10%, croup or bronchiolitis, and lower respiratory and diarrhoea is a feature in less than 5%. tract complications seem more frequent than Fever is the most prominent sign of infec• in adults [16,109,195). Finally there is a much tion and although it is as high as 41°C in about higher incidence of drowsiness and delirium one-quarter of cases it is more commonly than in adults. 38-40°C. The pyrexia peaks at the height of Infants beyond the age when passively systemic features and is typically of 3 days' acquired antibody would provide protection duration, but may last for 1-5 days. Clinical represent an especially susceptible group. examination often reveals a toxic appearance The risk of hospital admission for infants early in the course of the illness, the skin is from low-income families in Harris County, often hot and moist, the face appears flushed, Texas, was estimated to be 4 per 1000 during the eyes may have a glistening, injected, some• the ANictoria influenza outbreak, and the times weepy appearance, the mucous mem• mortality rate for hospitalized infants was branes of the nose and pharynx are 7.5% [64). Review of three later epidemics hyperaemic and devoid of an exudate, and (HINl, H3N2 and influenza B viruses) in breathing may appear laboured with a clear Harris County identified hospitalization rates nasal discharge or blocked nose. Small, tender, of at least 5 per 1000, regardless of the type of cervical lymph nodes are palpable in up to influenza virus [154). Both influenza A and B one-quarter of cases and crackles and wheezes viruses represent a significant cause of seri• may be heard in a similar proportion. This ous lower respiratory tract infection in chil• constellation of symptoms and signs typically dren. In one prospective study of 121 young persists for 3-4 days, but cough, lassitude, and seronegative children during an H3N2 virus malaise may persist for 1-2 weeks after the outbreak in 1974, 60 children were infected fever settles. These features relate to a typical and five developed clinical and X-ray evi• case, but during an outbreak cases may be dence of pneumonia [195). Kim and col• either subclinical or present with other acute leagues associated influenza virus infection respiratory syndromes such as the common with more than 5% of admissions to a chil• cold, pharyngitis, tracheobronchitis, or a sys• dren's hospital for acute respiratory tract ill• temic illness without respiratory features. ness [109) and showed that 14.3% of 860 patients with croup were infected with influenza virus. The mean period of hospital• 15.6.2 INFANTS AND CHILDREN ization for either influenza A or B was about 8 Although the manifestations of influenza A days, but serious infections with influenza A virus infection in infants and children are virus were five times more common than similar to those in adults there are some dif• with influenza B virus. ferences. Maximal temperatures tend to be In a review of five studies of manifesta• higher among children, and below the age of tions of influenza infections in hospitalized 5 years febrile convulsions are prominent, children it was reported that about one-half occurring in 35-40% of hospitalized patients had complications affecting systems other [16,60,64,104,157,195,196). There is also a rela• than the respiratory tract [16,21,106,150,157). tive prominence of coryza, otitis media and Abdominal pain mimicking acute appendici• gastrointestinal manifestations, notably vom- tis was seen in patients infected with 288 Influenza viruses: clinical spectrum and management influenza B virus, and mortality rates of 1-4% In the 3 months subsequent to the were reported in four of the five studies. Other 1989-1990 epidemic in Great Britain there was reports refer to the occurrence of acute myosi• a deficit of almost 11 000 deaths [6], indicating tis in association with both influenza A and B that almost one-third of the deaths would viruses [9,38,127]. have occurred in the absence of influenza. A small but consistent deficit of observed deaths 15.7 THE COMPLICATIONS OF INFLUENZA was also noted by Eickhoff, Sherman and 15.7.1 INFECTION IN THE ELDERLY AND Serfling [44], in relation to mortality in the USA after the 1957-1958 pandemic, principally EXCESS DEATHS from June to August 1958. This reinforces a Analyses of mortality data have shown consis• generally held opinion that influenza often tently that the elderly, especially those with kills those individuals who would die of their certain chronic medical conditions, are at primary disease alone, in the absence of epi• greatest risk from death during outbreaks of demic influenza, within 6 to 12 months. influenza. In temperate climates the incidence Using a computer model, Serfling et al. [166] of deaths from cerebrovascular, cardiovascular estimated excess pneumonia-influenza deaths and respiratory diseases increases during win• during three major influenza epidemics in the ter, and this seasonal variation is influenced by USA in late 1957, early 1958, 1960 and 1963. low temperature and respiratory viruses such Mortality among infants aged less than 12 as influenza and respiratory syncytial (RS) months ranged from 3.7-11.6 per 100 000 (mean viruses. Additional deaths above the normal 7.3/100 000), but fell to 0.8-3.1 per 100 000 (mean winter increase are recorded regularly in asso• 1.8/100 000) among those aged 1 to 4 years, and ciation with influenza epidemics and about to 0.1-1.4 per 100 000 (mean 0.5/100 000) among 90% of these excess deaths are among people 5 to 14-year-olds. Mortality then increased aged 65 years and over [6,173,182,183]. steadily, to 0.2-2.5 per 100 000 (mean 0.95/ Over 10 000 excess deaths were docu• 100 000) among 15 to 19-year-olds, to 0.9-2.7 per mented in each of seven epidemics in the USA 100 000 (mean 1.6/100 000) among 20 to 44-year• from 1977 to 1988, and more than 40 000 excess olds, to 5-8.9 per 100 000 (mean 7/100 000) among 45 to 64-year-olds, and to 16.7-33.1 per deaths occurred during each of two of these 100 000 (mean 25/100 000) among 65 to 74-year• epidemics [132]. About 120 000 excess deaths olds. Mortality was maximal in the over 75s, at were attributable to influenza in England and 51.3-101.6 per 100 000 (mean 73.1/100 000). Wales during the 10 winters after influenza Comparable rates for deaths certified as due NHong Kong (H3N2) virus first arrived [182]. to influenza were seen in England and Wales None was recognized in the UK during seven during the 15 years between 1974 and 1988. consecutive winters from 1978-1979 to The lowest mortality (0.04 per 100 000) was 1984-1985 [22], but there were almost 30 000 again noted in 5 to 14-year-olds; mortality excess deaths in Great Britain during the 56 increased in successive la-year age bands by days between 17th November, 1989 and 11th three-fold, four-fold, six-fold, 11-fold, 32.5- January, 1990 associated with the influenza fold, lOa-fold and 765-fold to a mean of 30.6 epidemic at that time [6]. Generally about half per 100 000 (range 3.4-190.2) in those aged 75 the excess deaths during influenza epidemics years and over. Examination of 1989-1990 are attributed to influenza, bronchitis and mortality data for England and Wales [34] pneumonia, and many of the remainder to showed that certified influenza deaths repre• cerebrovascular and cardiovascular disease - sent fewer than 10% of the excess mortality implying that influenza is responsible for associated with influenza, so it is possible that many hidden deaths. mortality from influenza in the over 75s could The complications of influenza 289 be nearer 2% during the most severe epidemics. in Leicestershire during the 1989-1990 epidemic This high estimate is supported by mortality [145]. Influenza was reported as the cause of data for non-vaccinated, elderly, people in nurs• death for 18 men and 29 women aged 67 to 97 ing homes. In Cardiff, for example, mortality years (mean age 84 years); 41(87%) were aged rates attributable to influenza were 4-8% during ~ 75 years. There were no certified influenza a 3-year period [98]; rates of 3.7% and 0.2% were deaths in younger subjects with chronic medical seen in Paris during NVictoria/75 (H3N2) and conditions, and the estimated mortality for the BlHong Kongl73 virus outbreaks [167]; a rate of fit elderly was approximately 7 per 100 000 4.5% was seen in seven nursing homes in (Table 15.3), i.e. similar to the rate observed by Michigan when NBangkok/1/79 (H3N2) virus Barker and Mullooly [8]. Among non-residential was circulating [152]; and the mortality was 17.7 subjects the rate for influenza-certified deaths % during an influenza NArizona/80 (H3N2) was 11.6 and 23.1 per 100 000 for those with virus outbreak in a New York home for the lung and heart disease, respectively. The major elderly [67]. impact of influenza was seen in the residential Barker and Mullooly [8] studied pneumonia care facilities where the rates were 343, 499 and and influenza deaths among a population of 230 2703 per 100 000, respectively, for people with 000 in relation to age and presence of chronic one, two and three or more medical conditions. medical conditions during two epidemics On the basis of these mortality data it appears (1968-1969 and 1972-1973). The death rate was 9 that pneumonia and influenza deaths in per 100 000 for the over 65s without high-risk Oregon, and influenza deaths in Leicestershire, conditions and increased to 217 per 100 000 for are more numerous in the fit elderly than in the presence of one high-risk condition and 306 younger subjects with a single 'high-risk' med• per 100 000 for two or more. The death rate ical condition. among the fit elderly was comparable with that of 10 per 100 000 among 45 to 64-year-olds with one high-risk condition, and higher than that in 15.7.2 INFLUENZA DURING PREGNANCY AND 15 to 44-year-olds with one high-risk condition THE EFFECTS ON THE FETUS AND (no deaths among 6260 subjects). Nguyen-Van• OFFSPRING Tam and Nicholson examined certified Women during the second and third influenza deaths among a population of 892 000 trimesters of pregnancy are at increased risk of

Table 15.3 Influenza-associated death rates (certified as influenzal) during the 1989-1990 influenza A epidemic in Leicestershire. (Data abstracted from [145].)

No. of Estimated no. Deaths per Status influenza deaths in the popUlation 100000

Residential patients With 1 medical condition 8 2334 343 With 2 medical conditions 5 1002 499 With ~ 3 medical conditions 6 222 2703 With lung disease 5 1590 314 With heart disease 12 1668 719 Non-residential patients Without medical condition 3 45399 6.6 With lung disease 6 51534 11.6 With heart disease 15 65031 23.1 290 Influenza viruses: clinical spectrum and management hospital admission, severe pulmonary compli• stenosis in addition. In the series reported by cations of influenza and death. The Public Petersdorf et al., one of 10 pregnant women Health Laboratory Service Report on the with influenza went into premature labour, influenza epidemic in England and Wales dur• delivered a dead fetus, and died 15 hours later ing 1957-1958 notes that 12 of 103 fatal cases in [155]. Neither she nor the other nine pregnant females aged 15 to 44 years of age were preg• women had any other risk factor. Ramphal, nant, which is about double the expected pro• Donnelly and Small [160] describe a fatal portion for this age group [159]. In Holland, A(fexas/77 virus infection in a woman with during the first wave of Asian influenza, 1230 ventricular septal defect and term pregnancy. influenzal deaths were reported among a pop• Two of 15 women with Asian influenzal pneu• ulation of 11 million, and 11 deaths occurred monia reported by Louria and colleagues were during pregnancy [156]; in the group aged pregnant [116]; one had rheumatic heart dis• 20-39 years the mortality of pregnant women ease with mitral stenosis but both recovered. was twice that of non-pregnant women. During the same epidemic Martin et al. Similarly, 10 of 24 females of childbearing age reported 10 fatal cases among women aged 15 among an American series of 91 patients with to 44 years, of whom four were pregnant [119]. Asian influenzal pneumonia were pregnant, The number who were pregnant was several all but one in the last trimester or at term, and times higher than expected, and only one of this was three to four times the expected pro• the four pregnant women had a chronic med• portion [155]. During the Asian influenza epi• ical condition, myasthenia gravis. Giles and demic in New Orleans, Burch, Walsh and Shuttleworth [58] reported the death of a 17- Mogabgab [19] admitted 34 women of whom year-old pregnant girl among 14 women who 13 were pregnant - three in the second died from influenza in Stoke on Trent; post• trimester and 10 in the third. Here again, the mortem examination revealed cardiac hyper• proportion of admissions who were pregnant trophy but she was not known previously to was several times higher than expected. The have had cardiac disease. From this small sam• evident increase in mortality associated with ple it would appear that about 75% of the pregnancy has also been recognized during an deaths during pregnancy occur in women interpandemic period. Ashley, Smith and without a chronic medical risk factor. Dunell [6] compared cause of death among a 1 Influenza complicating heart disease in in 15 random sample of deaths that occurred pregnancy appears especially serious. Govan between 25th November 1989 and 8th January and Macdonald identified nine such patients, 1990 with deaths occurring during the same of whom four died [65]. They presented with period in 1985-1986. Analysis revealed a four• an acute febrile illness and within 12 hours of fold increase in deaths associated with preg• onset developed fulminating bronchopneumo• nancy during the 1989-1990 epidemic, and the nia. Ketosis was a prominent finding. Five data suggest that the epidemic in the UK patients went into labour at this stage, of whom accounted for about 90 excess maternal deaths three died; two of the babies were stillborn and during pregnancy. three died within 48 hours. Post-mortem exam• Most of the above reports do not make it inations revealed extensive haemorrhagic bron• clear whether severe influenza during preg• chopneumonic consolidation. nancy was associated with a chronic medical Few attempts have been made to demon• condition. In the series reported by Oswald, strate transplacental passage of influenza virus Shooter and Curwen [148], seven of 379 to the fetus. Martin et al. [119] and Ramphal, patients with influenzal pneumonia were Donnelly and Small [160] failed to identify the pregnant and two died, one having mitral presence of influenza antigen in fetuses from The complications of influenza 291 four mothers with fatal influenza. Yawn et al. influenza epidemics during the period, [198], however, recovered the virus from the notably in 1951,1953,1959 and 1961; the excep• extra pulmonary tissues of a mother who died tion was the Asian flu epidemic of the autumn in the third trimester and from the amnion and of 1957. Ashley, Smith and Dunell [6] showed myocardium of the fetus, although there were that perinatal deaths increased 1.6-fold during no abnormalities in the tissues from which the the 1989-1990 epidemic, namely an increase of virus was isolated. Influenza NBangkok 255 from approximately 465 to approximately (H3N2) virus was isolated from the amniotic 720, as compared with a similar period in fluid of a mother with 'amnionitis' and 1985-1986. Hardy et al. [76] reported that the influenza infection at 36 weeks of gestation. incidence of stillbirths was higher in 332 symp• The infant, who was born at 39 weeks, had evi• tomatic pregnant women with serologically dence of infection but remained well [125]. confirmed influenza than in 206 asympto• There have been several reports of a slight matic women with serologically confirmed increase in congenital abnormalities following influenza or in 73 uninfected women [72]. influenza virus infections during pregnancy [25,66,70] but there is no consistent association 15.7.3 RESPIRATORY TRACT COMPLICATIONS between specific defects and influenza, and the virus has not been conclusively implicated. Pneumonia Hardy et al. [76] noted major congenital abnor• The H3N2 influenza virus epidemic in malities in 5.3% of 80 women whose influenza 1989-1990 provided Connolly, Salmon and infections occurred during the first trimester Williams with the opportunity of reviewing compared with 2.1% of 183 women infected 342 episodes reported as influenza in the gen• during the second trimester and 1.1 % infected eral practice surveillance of infectious diseases during the third. Wilson and Stein [191], during in Wales [26]. Pneumonia was the second most the 1957 outbreak, noted no increase in congen• common respiratory complication, occurring ital abnormalities among infants conceived at a rate of 2.9%. Comparable figures for pneu• during the 3-month period when influenza was monia complicating Asian influenza in the pri• epidemic and whose mothers had serologic evi• mary care setting are 4.8% of 391 [59] 1.8% of dence of infection with influenza virus. Other 170 [90],3% of 187 [194], 1.9% of 930 [52], 1.5% studies in the USA and the UK have been of 1990 [69], 3.5% of 750 [54], 4.9% of 161 [12], reported which indicate a possible relationship and 5% of 542 [42]. Overall, the incidence of between maternal influenza and childhood pneumonia complicating influenza appears to leukaemia [7] and schizophrenia. be approximately 3%. In the hospital setting Review of influenza deaths during the pneumonia was seen in 11.1% of 494 people period 1921-1932 revealed an increase in mor• with severe or complicated Asian influenza tality from premature births in association [163], 13% of 77 children admitted with Hong with influenza [175]. Wynne-Griffith et al. Kong influenza [16], 19.7% of 76 subjects with [197] showed that early neonatal mortality in Asian influenza [19], and 48% of 541 cases with the second quarter of 1970 after a major Asian influenza [100]. influenza epidemic was slightly but signifi• The course and outcome of influenzal cantly higher (from 9.88 per 1000 live births in pneumonia is influenced by the presence of 1969 to 10.77 in 1970) than in the correspond• pre-existing disease, secondary staphylococcal ing quarter of the previous year. Analysis of infection and age. The series of patients infant mortality over the previous 25 years reported by Jamieson et al. [100] exemplifies indicated that similar increases occurred in the importance of pneumonia, age and pre• relation to four of the other five major existing disease. Of 541 hospitalized cases of 292 Influenza viruses: clinical spectrum and management

Asian influenza, a worse prognosis was noted Jamieson et al. [100]. The chest pain is of two in those with pneumonia, and of the 262 types - substernal, which presumably reflects patients with pneumonia the course was par• the tracheitis commonly seen at post mortem, ticularly severe among those with pre-existing and pleuritic. During the Asian influenza out• disease, mostly chronic respiratory or cardiac break the onset of illness in fatal cases was typi• disease [100]. Case-fatality rates increased cally abrupt and within 24 hours of onset of from 1 % (3 of 279) among admissions with symptoms pneumonic or other serious symp• uncomplicated influenza to 11% (30 of 262) toms were observed in one-third of patients among those with pneumonia; from 0.5% (1 of [159]. Deterioration was much more rapid in the 200) in those aged <20 years, through 2.2% (3 young than in the elderly with 56% of patients of 135) in those aged 20-49 years, to 14% in under 5 years of age gravely ill within the day of those 50 years and over; and from 3% in pneu• onset compared with 16% of those > 64 years monia patients aged <50 years to 22% in those [159]. In fatal cases deterioration was particu• aged ?! 50 years. The overall mortality among larly rapid; in the Public Health Laboratory those with pre-existing disease was 51 %. Giles Service series of 477 deaths, no fewer than 86 and Shuttleworth [58] found that 35 of their 46 occurred before admission to hospital and two• fatal cases had pre-existing disease. thirds died within 48 hours of admission [159]. Children with congenital malformations and chronic diseases are also at increased risk Primary viral pneumonia of severe lower respiratory tract disease. Among 77 children hospitalized with Hong Louria et al. [116] and Martin et al. [119] pro• Kong influenza, lower respiratory tract infec• vided the first detailed accounts of primary tion occurred in 13 (56%) of 23 with pre-exist• influenza virus pneumonia that occurred dur• ing conditions, compared with 10 (19%) of 54 ing the Asian influenza pandemic in without [16]. Oswald, Shooter and Curwen 1957-1958. The incidence of primary influenzal [148] compared staphylococcal and non• pneumonia is unclear, but Louria et al. saw six staphylococcal influenzal pneumonias; in the cases among 33 with pulmonary complications staphylococcal group, 71 (46%) of the patients of influenza that were sufficiently severe to be were severely ill, compared with 34 (23%) of admitted. All six had underlying cardiac dis• those with non-staphylococcal pneumonia, ease. Since the overall incidence of pneumonia many of whom were elderly and had associ• during influenza is about 3%, viral pneumonia ated chronic respiratory or cardiac disease. uncomplicated by secondary bacterial infec• The overall mortality rate of 28% for tion evidently occurs in at least 1 in 200 infec• staphylooccal pneumonia was higher than tions. Hers, Masurel and Mulder [88] isolated that of non-staphylococcal pneumonia (12%). influenza virus from the lungs of 72% of 148 Moreover, the mortality in the staphylococcal virologically confirmed fatal cases. Overall, group was similar at all ages, whereas that in 20% were identified as having influenza virus the non-staphylococcal group was concen• pneumonia without secondary bacterial infec• trated in those aged over 55 years. tion. These observations indicate that primary Influenza complicated by pneumonia gener• viral pneumonia, without secondary bacterial ally presents with influenzal symptoms and infection, is not uncommon and can be life• although there are no clear distinguishing fea• threatening, and that the majority of bacterial tures, cough and chest pain were observed pneumonias complicating influenza occur in more frequently than in uncomplicated lungs infected with the virus. influenza - 54% versus 79%, and 31% versus After a brief period of typical influenzal 51 %, respectively, in the series reported by symptoms each of the patients in Louria's The complications of influenza 293 series [116] became severely dyspnoeic (mak• enced an improvement in influenzal symp• ing speech difficult) and cyanosed. None had toms before pulmonary involvement, and pul• chest pain, four produced bloody sputum and monary symptoms blended with the initial two had mucoid sputum. Auscultation illness, but occurred several days after its onset revealed fine inspiratory crackles, inspiratory in the remainder. Chest X-radiography and expiratory wheezes, and no signs of con• revealed lobar or lobular involvement; and solidation. The total leucocyte counts were ele• only one of the 15 patients died. Angeloni and vated with a polymorphonuclear leucocyte Scott [3] reported nine deaths among 41 adult predominance despite the absence of concur• patients who developed pneumonia from 1-14 rent bacterial infection. Chest radiography days after onset of influenza. Post-influenzal mostly revealed a diffuse perihilar infiltrate. pneumonia caused a higher mortality in Profound hypoxia was resistant to oxygen patients with long-standing chest disease (six administered by face-mask and oxygen tent deaths in 13 cases) than in those who were pre• but none of the patients was ventilated. viously fit (three deaths in 28 cases). Influenza virus was recovered from the lungs In the cases with secondary bacterial infec• tions reported by Louria et al. [116] S. aureus of each of five patients who died. Histological was recovered from four of five fatal cases and studies revealed tracheitis, bronchitis, bronchi• from eight of 19 who survived (NS). Although olitis with haemorrhagic areas, and loss of cili• fatal influenzal-staphylococcal pneumonia ated epithelial cells, especially in the lower had been recognized previously, the Asian lobes. The alveolar spaces contained neu• influenza epidemic highlighted its importance trophils, mononuclear cells, fibrin and fluid, in fatal cases. It was commonly encountered in and intra-alveolar haemorrhage was common pathological material- in 62% of 467 deaths in in the lower lobes. the UK [159], 71 % of 62 deaths in London hos• pitals [148], 59% of 148 fatal cases in Holland Combined bacterial and viral pneumonia [88], 45% of 11 post mortems reported by Louria et al. [116] report nine patients in whom Martin et al. [120],33% of 46 fatal cases in Stoke concomitant viral and bacterial pneumonia on Trent [58], and in eight of nine cases of ful• was considered likely, and 15 in whom the bac• minating influenzal pneumonia [161]. Two terial infection was a 'late' secondary complica• large studies of hospitalized patients in tion. All nine patients with concomitant viral Dundee [100] and London [148] illustrate that and bacterial pneumonia were severely ill on many of these staphylococcal infections were admission; eight were cyanosed, all had puru• acquired nosocomially. This fact was again lent or bloody sputum with signs of consolida• emphasized when the bacterial flora of the tion, crackles or wheezes. Six of the nine had lung or sputum was related to the interval underlying chronic disease and one was 6 between admission to hospital and death months pregnant. Chest radiography revealed [159]. The proportions of patients over 65 lobar infiltrates and/or bilateral nodular perihi• years who yielded staphylococci were 13% lar infiltrates; sputum examination revealed among those who died on the day or day after large numbers of polymorphonuclear leuco• admission, 20% after 2 or 3 days, 45% after 5 to cytes, even in patients with leukopaenia, and 7 days, and 80% after 8 days or more. large numbers of bacteria. Influenza virus was isolated from the lungs of three of the four Lung abscess patients who died. Seven of the 15 patients in Comparison of chest radiographs and whom bacterial pneumonia was considered a necropsy specimens indicates that radiology 'late' event had chronic underlying disease does not reflect the true incidence of abscess and one was 8 months pregnant. Ten experi- formation as a complication of influenza [148]. 294 Influenza viruses: clinical spectrum and management

Of 300 influenza-associated pneumonias Bendkowski [12], a general practitioner, com• reported by Oswald, Shooter and Curwen pared the incidence of respiratory complica• [148] radiological cavitation was observed in tions in three groups of patients, 124 with 14% of secondary staphylococcal pneumonias, allergies (of whom 55 had asthma), 59 who and in 2% of the others. Lung abscess was were non-allergic but came from allergic fami• diagnosed radiologically in eight (29%) of 28 lies, and 161 who were non-allergic. He noted previously fit adults who developed symp• that asthma complicated 38% of influenzal toms of pneumonia up to a fortnight after an infections in the allergic group and clinical influenzal illness [3] and frank abscess forma• improvement was slow. Influenza virus infec• tion was found at post mortem in four of 41 tions have consistently precipitated attacks of patients by Giles and Shuttleworth [58]. wheezing in both asthmatic children and adults. For example, in a study by Minor et al. Croup [129] which involved 41 children aged 3-17 years and eight adults with a history of 'infec• Influenza viruses are among the aetiological tious' asthma, 55% of all respiratory infections agents associated with acute laryngotracheo• precipitated asthma. Five patients had bronchitis (croup), and croup associated with influenza A virus infections and four of these influenza A virus appears to be more severe, were associated with asthma. In adult asthmat• resulting in more frequent hospitalizations and ics, almost 50% of upper respiratory tract viral tracheostomies [97] but less frequent than that infections are followed by asthma and a similar associated with parainfluenza virus or RS virus percentage of asthmatic episodes are associ• infections. Kim et al. [109] showed that whereas ated with a virus, chiefly rhinoviruses and influenza viruses (particularly influenza A) coronaviruses [147]. Severe epidemics of ranked second overall after the parainfluenza influenza result in a small but significant excess viruses as a cause of croup, influenza A virus mortality attributed to asthma. Housworth and infection was detected in proportionately more Langmuir [95] studied asthma excess mortality croup patients than the most important croup• during seven periods of influenza activity producing virus, parainfluenza type I, when between 1957 and 1966. Asthma deaths the peak months of influenza A virus and increased by 19-46% during influenza A virus parainfluenza virus activity were compared. outbreaks in 1957, 1958, 1960 and 1963, but H3N2 viruses appear to have a greater were either insignificant or barely significant predilection to produce croup than H2N2 during the milder epidemics with some viruses. During the years 1957-1976, influenza influenza B virus activity. During the worst A virus infection was detected in approxi• period excess mortality from asthma increased mately 8% of hospitalized children with croup by an estimated 3.8 per million population. during the H2N2 virus era and 24% of such patients during the H3N2 era [109]. Bronchitis Acute bronchitis is the most common lower Asthma respiratory tract complication of influenza. In normal individuals with uncomplicated Bendkowski [12] studied 161 subjects without influenza, pulmonary function tests have a history of allergy and noted acute bronchitis revealed frequent airway hyperreactivity, during Asian influenza in 30%. Among 350 peripheral airway dysfunction, and abnormali• admissions with Asian influenza in Kuwait, ties in gas exchange that can be prolonged for Guthrie, Forsyth and Montgomery [69] diag• some weeks after clinical recovery [73,93,102]. nosed bronchitis in 22%. Among 76 children The complications of influenza 295 admitted to hospital during the Hong Kong ing the five most severe epidemics of 1921 to epidemic, Brocklebank et al. [16] found bron• 1932 [175] and by 5-12% during four of five chitis in 12%. During the H3N2 virus outbreak USA epidemics during 1960-1966 [95]. in 1989-1990, Connolly, Salmon and Williams Endocrine deaths (mostly diabetic) increased [26] found evidence of acute bronchitis in 19% by about 1350 (i.e. by 30%) in England, Wales and noted the risk of bronchitis complicating and Scotland during the 1989-1990 epidemic, influenza to be greater in patients with pre• as compared with 1985-1986, but the actual existing diseases, regarded as an indication for rise was eight-fold lower than excess circula• vaccination, and also in the elderly. tory deaths and approximately 17.5-fold lower Studies of chronic bronchitics have shown than excess respiratory deaths [6]. Older-onset that one-quarter to two-thirds of exacerbations are associated with viruses, and data collated diabetics are 1.7 times more likely to die from from several studies have revealed serological pneumonia and influenza as compared with evidence of influenza A and B virus infection the general population, and 1 in 33 die from in approximately 4% of exacerbations. In these conditions overall [137]. A similar but Stock's review of influenza epidemics, mortal• non-significant trend for increased pneumo• ity from both acute and chronic bronchitis was nia and influenza deaths has also been noted to have increased during the years 1921 observed among younger, insulin-dependent to 1933 [175]. During the worst periods deaths diabetics [137]. from acute bronchitis increased by 93% and Hospitalization rates for 'diabetes' were from chronic bronchitis by 52%. marginally increased (by 4%) during the 1989-1990 epidemic in the USA, but mortality 15.7.4 COMPLICATIONS IN DIABETICS among 'diabetes' admissions increased by two-thirds to 6% [123]. During 1976 and 1978, As long ago as 1889-1890 it was noted that hospital admissions in Holland for influenza there were twice as many' diabetes' deaths in were six times more common among diabetics Paris during the pandemic as compared with than controls with duodenal ulcer, and admis• the average number during the three previous sions for ketoacidosis increased by 50%, as years [151]. In Holland in 1957 the first wave of compared with years with low influenza activ• the 'Asian' influenza pandemic saw an ity, from 78 in 1976 to 152 in 1978 [15]. There increase of 61 'diabetes' deaths among a 14.9 were four times as many pneumonia deaths million population, representing a rise of 25% among diabetics in 1976 and 1978 (57 versus compared with the same period in previous 15) compared with years with little influenza years [156]. The actual increase was 4.6-fold activity, and twice the number of ketoacidosis lower than excess respiratory deaths and 15.5- deaths (63 versus 29). Bouter et al. [15] esti• fold lower than excess cardio- and cerebrovas• cular-related deaths. Almost 90% of 'diabetes' mated that during the 1978 epidemic 1 in 1300 deaths occurred in the over 70s, and 99% were patients with diabetes mellitus were hospital• over 50 years of age. The same pandemic in ized for pneumonia, 1 in 260 with insulin the USA saw' diabetes' deaths increase by an dependence were hospitalized for ketoacido• estimated 3.62 per million population (by sis, and 1 in 1000 with insulin dependence 9.2%) in late 1957, and by 3.25 per million died in hospital. The combination of pneumo• (7.9%) in early 1958 [95]. nia and diabetes appears especially serious; in The interpandemic years also see increased the series of influenzal pneumonias reported death rates among diabetics. Thus, 'diabetes' by Oswald, Shooter and Curwen [148] six of deaths increased by approximately 10% dur- nine diabetics died. 296 Influenza viruses: clinical spectrum and management

15.7.5 MYOCARDITIS AND PERICARDITIS that the pyrexia, hypoxia and pH abnormali• Myocarditis and pericarditis occasionally com• ties that accompany influenza are responsible plicate fatal influenza, and electrocardio• for a toxic encephalopathy in many cases. graphic abnormalities have been reported in The Asian epidemic appears to have stimu• some uncomplicated cases of influenza A virus lated an interest in the neurological complica• infection [19,57,105]. Louria et al. [116] studied tions of influenza, or conceivably H2N2 virus the course of influenza in 15 patients with was particularly neurotropic. Dubowitz [41] underlying heart disease and found no elec• observed two children who became comatose trocardiographic evidence of myocarditis. and recovered rapidly. Homer [92] described However, autopsies on eight patients revealed four children and a young woman who devel• one with a widespread, acute, necrotizing oped neurological disorders several weeks after myocarditis, and the striking polymorphonu• an influenzal illness. Three were described as clear leucocyte infiltration of the myocardium having encephalopathies; one recovered com• appeared to be of recent origin. Martin et al. pletely except for a mononeuritis, one recov• [119] described one case among 32 who died ered but was left with severe sequelae, and one from influenza. The diagnosis was not sus• died and showed extensive but patchy pected antemortem and no signs of myocardi• demyelinization of the central nervous system. tis other than tachycardia were observed. A fourth patient developed acute cerebellar Histological studies revealed marked muscle ataxia which improved completely within 12 fibre necrosis with a cellular infiltrate of days, and the final patient developed a dissemi• mononuclear cells, lymphocytes and occa• nated encephalomyelitis with prominent fea• sional plasma cells. Focal interstitial myocardi• tures of a transverse myelitis. Four cases were tis of low degree and questionable significance reported by McConkey and Daws [124]; all four was observed in nine of the remaining 31 patients developed neurological features 4 or 5 cases. Attempts at virus isolation from cardiac days after onset of the respiratory illness. Three tissue of patients with myocarditis associated became drowsy or semi-comatose and one with fatal influenzal pneumonia have gener• developed a vertical gaze palsy. Electro• ally been unsuccessful. Virus has been recov• encephalograms (EEGs) in two of three patients ered from pericardial fluid and both influenza were grossly abnormal; all recovered and the A and B virus infections have been associated EEGs soon reverted to normal. with myopericarditis [1]. Post-mortem studies Flewett and Hoult [46] in an analysis of of cardiac tissue of 46 patients dying from neurological complications in 18 cases seen Asian influenza revealed myocardial oedema during the 1957-1958 outbreak, categorized as the only constant finding [58]. patients into those with 'encephalopathy' seen at the height of influenza, post-influenzal encephalitis which developed 3 days to 2 15.7.6 NEUROLOGICAL COMPLICATIONS weeks after onset of influenza, and Neurological complications of influenza have Guillain-Barre, syndrome. Fatal encephalopa• long been described but appear to be rare [41]. thy and encephalitis appears to be rare. In the Apart from Reye's syndrome, a range of neu• UK during the four seasons 1955-1956 to rological complications have been reported in 1958-1959 there were 19622 certified association with influenza, including irritabil• influenza deaths with 52 fatal cases of ity, confusion, convulsions, psychoses, neurtis, , encephalitis' - a rate of 2.6 fatal cases of Guillain-Barre, syndrome, coma, trans• , encephalitis' per 1000 fatal cases of influenza. verse myelitis and encephalomyelitis. Their Virus has occasionally been recovered from pathogenesis remains unclear but it is likely the cerebrospinal fluid (CSF) or brain of The complications of influenza 297 patients with 'encephalopathy' [148]. Hoult periods revealed a 37.5% increase in the vari• and Flewett [94] described the histological fea• able fore-period simple reaction time, a 13% tures in the brain of five fatal cases of reduction in response time to repeated num• encephalitis. No gross abnormalities were bers detection task, and a significant reduction found except in one case in which there was in a categorical search task [172]. acute haemorrhagic leucoencephalitis. Other investigators have also commented on the Reye's syndrome haemorrhagic leucoencephalitis and noted Reye's syndrome, a multisystem disorder vessel necrosis, perivascular oedema, inflam• which is characterized by encephalopathy and matory infiltrates, haemorrhage and demyeli• fatty liver degeneration, typically follows an nation. Cerebral abscess formation and acute viral infection in young children, most bacterial meningitis are also recognized [148]. commonly influenza B and varicella, but also According to Osler, 'almost every form of dis• gastrOintestinal infections. Clustering of cases ease of the central nervous system may follow occurs during the winter months and out• influenza', and others have also commented breaks of Reye's syndrome have been associ• on a variety of mental disturbances associated ated with epidemics of both influenza A and B with the 1918-1919 pandemic. and salicylate use. During the 1974 influenza Acute psychosis developing 2 to 10 days outbreak, Reye's syndrome occurred with an after the onset of Asian influenza possibly rep• estimated rate of 31 to 58 cases per 100 000 resents another manifestation of encephalitis; influenza B virus infections in children [27,28]. electroencephalograms in three patients with Otherwise, the evaluated incidence of Reye's psychoses were all diffusely abnormal and syndrome has been approximately 0.5 per improved slowly over a period of weeks [13]. 100 000 population aged less than 18 years in Lloyd-Still [115] noted the onset of acute psy• the USA. An increasing body of data has choses, with auditory and visual hallucina• revealed a strong association between the use tions, in 19 patients during or after an attack of of salicylates and Reye's syndrome and, possi• presumed influenza. All recovered quickly. bly because of reduced salicylate usage, recent trends in the USA and UK indicate a decreased Subtle changes in brain function incidence of cases. In the UK, a recent active surveillance of Reye's syndrome revealed an Influenza B virus was one of several viruses annual incidence of 1 case per million popula• studied at the MRC Common Cold Unit in tion aged <16 years [144]. The pathogenesis of Salisbury for its ability to affect simple mea• the condition remains obscure. Isolation of sures of performance [171]. Volunteers influenza virus from the liver, muscle biopsy infected with influenza B viruses had signifi• specimens, CSF and muscle from a survivor of cant impairments of reaction times as mea• Reye's syndrome has led to the proposal that sured by the ability to press a button as soon as viral dissemination may playa role. possible after seeing a black spot appear on a television screen. A visual search task involv• ing five possible target letters was also signifi• 15.7.7 TOXIC SHOCK SYNDROME cantly impaired in the group infected with Toxic shock syndrome, an illness character• influenza B virus. Various tests were applied ized by fever, hypotension and erythroderma but only the attention tasks were impaired. followed by desquamation, is found in associa• Similar tasks were assessed in students tion with TSST-1- or enterotoxin-producing s. infected naturally with influenza B virus. aureus infection. MacDonald et al. [117] identi• Comparison of baseline and symptomatic fied nine cases of severe hypotension or death 298 Influenza viruses: clinical spectrum and management compatible with toxic shock syndrome as a the children often refused to walk, or did so complication of influenza or influenza-like ill• with a bizarre gait. Muscle enzymes were ele• ness during an epidemic of influenza in vated in two-thirds of the cases, but attempts Minnesota. Four of the patients had a proven to isolate influenza virus from blood were influenza B virus infection, and S. aureus was unsuccessful. In adults, the myositis tends to cultured from two individuals; one isolate pro• be more diffuse. Muscle cell necrosis with and duced TSST-1 and both produced enterotoxin. without cellular infiltrates can occur, but the The authors concluded that toxic shock syn• pathogenesis of the condition remains drome may be a rare complication of obscure. The condition is usually benign and influenza, and that it may occur more fre• of short duration, but rhabdomyolysis with quently than Reye's syndrome. The mortality myoglobinaemia, myoglobinuria and acute rate from toxic shock syndrome complicating renal failure have been described in severe influenza is 60%. It has been proposed that a adult cases occurring in association with plague in Athens in 430 Be was due to the influenza A virus infection [130,135,169,199]. complications of a toxin-producing S. aureus During outbreaks influenza is probably the during an influenza epidemic. Interestingly, leading cause of acute myoglobinuric renal Jaimovich et al. [101] recently reported four failure. Histological examination reveals nor• cases of influenza B virus infection associated mal glomeruli with focal tubular necrosis and with a non-bacterial septic shock-like illness. pigmented casts in some tubules. Altogether 14 cases of toxic shock syndrome following influenza or influenza-like illness Other renal disease have been described, but influenza A virus infection has been documented serologically Glomerulonephritis and Goodpasture's syn• in only two cases [184]. drome have been documented in several patients with influenza infections [188,192]. A lesion in the kidney with tubular changes was 15.7.8 MYOSITIS, MYOGLOBINURIA AND reported by Beswick and Finlayson [14]. RENAL FAILURE Judging by the absence of clinically important Myalgia affecting the legs and back is a well• renal complications during influenza and the recognized feature of influenza and occurs dearth of pathological abnormalities in renal early during the course of the illness. In con• tissue, it seems that influenza rarely affects the trast, myositis (and myoglobinuria with or kidneys. without renal failure) usually occurs in the recovery phase from a typical influenzal illness. 15.8 ANTIVIRALS Middleton, Alexander and Szymanski [127] reported 26 children who developed bilateral 15.8.1 ADAMANTANES lower-limb myositis, characteristically after a Amantadine and other adamantanes such as period of rest and at a time when the respira• rimantadine were first shown to inhibit tory symptoms and signs began to subside. influenza A virus in the 1960s. Amantadine Influenza B virus infection was proved in 20 and rimantadine inhibit H1N1, H2N2 and cases and type A infection in one. In a further H3N2 strains of influenza type A virus [68,80] series of paediatric cases, similar evidence of including 'new' epidemic strains, so it is to be influenza B virus infection was found in 11 of expected that future variants, including pan• 17 cases [39]. In Middleton's series, leg pains demic strains, will be similarly inhibited. and muscle tenderness lasted 1-5 days and no Neither agent inhibits influenza B virus strains true muscle weakness was apparent, though and higher concentrations than can be Antivirals 299 achieved safely in humans are required to but not rabbits, receiving 15 times the usual inhibit parainfluenza and RS viruses. Drug• human dose [149], treatment with amantadine resistant strains of influenza A virus, which is not justified in women of childbearing exhibit complete cross-resistance between potential, except perhaps in life-threatening amantadine and other adamantanes [10,77] influenzal pneumonia. Amantadine is con• can be readily produced in the laboratory and traindicated in epilepsy and gastric ulceration, are being recovered increasingly from humans and should be used cautiously in patients with [11,81,83,110,121]. The molecular basis of resis• cardiovascular or renal disorders or those with tance involves a single amino acid change in cerebral atherosclerosis, i.e. individuals who one of four amino acid residues in the M2 are at increased risk from influenza. membrane protein [10,77,78]. Rapid selection Before amantadine is prescribed, laboratory and apparent transmission of drug-resistant and epidemiological evidence of an outbreak influenza A viruses have resulted in treatment of influenza A in the community should exist. failures when index cases and family contacts Taken prophylactically amantadine is about as were given rimantadine [11,81] and the emer• effective as influenza virus vaccine in prevent• gence of drug resistance was implicated in ing influenza A virus infection. Controlled tri• occasional treatment failures when amanta• als suggest that prolonged administration of dine prophylaxis was initiated during a nurs• amantadine, 200 mg daily, during community ing home outbreak of influenza [121]. Illnesses outbreaks of influenza is about 50% effective caused by resistant strains are probably no in the prevention of infection but 70-100% more severe than those caused by wild strains effective in the prevention or amelioration of of the virus, and although the emergence of illness [40,185]. This distinction may be a desir• virus drug resistance is of paramount concern, able feature of prophylaxis since subclinical no reduction of drug efficacy was observed in infection could confer immunity against re• the Soviet Union over a 20-year period when infection. If amantadine or rimantadine is 142227 patients were treated with rimanta• used for both treatment and post-exposure dine [110]. prophylaxis in homes, however, rapid selec• Currently, only amantadine is licensed for tion and apparent transmission of drug-resis• use in the UK. Amantadine is almost com• tant influenza A viruses can occur, and the pletely absorbed after oral administration and drug may provide little or no protection is excreted unchanged in the urine. Because of [11,56,81,121] reduced renal clearance in the elderly and in Because of minor CNS side effects pro• patients with impaired renal function, care duced by amantadine, prophylaxis is most should be taken to ensure that the drug does appropriate for persons in high-risk groups for not accumulate to toxic levels [146]. Minor whom vaccination is indicated [37,190] neurological symptoms including insomnia, Indeed, WHO recommends amantadine or light-headedness, difficulty in concentration, rimantadine prophylaxis for elderly and high• nervousness, dizziness and headaches risk people in institutional settings to augment develop in 5-20% of individuals receiving 200 protection afforded by vaccination [190]. If mg daily [17]. Other adverse effects include vaccination has been overlooked high-risk anorexia, nausea, vomiting, dry mouth, consti• individuals should still be vaccinated after an pation and urinary retention. They arise outbreak appears locally, but the development mostly during the first few days of medication of an immune response usually takes several and disappear quickly when amantadine is weeks and this vulnerable period (and discontinued. Because embryotoxic and ter• beyond) can be covered by chemoprophylaxis. atogenic effects have been described in rats, When vaccine is unavailable, or the influenza 300 Influenza viruses: clinical spectrum and management

A strain causing an epidemic differs markedly useful clinically and deserve wider distribu• from the vaccine strain, amantadine should be tion as an adjunct to (not a substitution for) given for the entire duration of the outbreak, a vaccination, but doctors must continue to period of about 4-8 weeks. Chemoprophylaxis monitor efficacy and the emergence of resis• should be considered for all unvaccinated tant strains in formal clinical trials. household members and medical and para• medical workers in frequent contact with 15.9 REFERENCES high-risk persons in the home, hospital or 1. Adams, C.W. (1959) Postviral myopericarditis institutional setting. It is also advocated to associated with the influenza virus; report of control established outbreaks in facilities that eight cases. Am. J. Cardiol., 4, 51Hi7. care for high-risk persons, regardless of vacci• 2. Alford, RM., Kasel, J.A., Gerone, P.J. et al. nation status, but in this setting the rapid (1966) Human influenza resulting from aerosol emergence of resistance may be a problem. inhalation. Proc. Soc. Exp. BioI. Med., 122, 800-4. Treating established influenza A virus 3. Angeloni, J.M. and Scott, G.w. (1958) Lung infections with amantadine, when started abscess and pneumonia complicating influenza. Lancet, i, 1255--6. within the first 48 hours of symptoms, reduces 4. Arden, N.H., Patriarca, P.A. and Kendal, A.P. the duration of fever and other effects by 1-2 (1986) Experiences in the use and efficacy of days and accelerates the resolution of the inactivated influenza vaccine in nursing peripheral airways abnormalities that usually homes, in Options for the Control of Influenza, accompany influenza [113,146,186,193]. The (eds A.P. Kendal and P.A. Patriarca), A.R Liss, reduction in symptoms far outweighs the New York, pp. 155--68. 5. Arden, N., Bidol, 5., Ohmit, S. and Monto, A drug's toxic effects [186,193]. Early treatment, (1993) Effect of nursing home size and i.e. before laboratory confirmation of the diag• influenza vaccination rates on the risk of insti• nosis is generally available, seems essential. tutional outbreaks during an influenza (H3N2) Treatment for several days is usually effective epidemic, in Options for the Control of Influenza and short courses may lessen the selection of II, (eds C. Hannoun, A.P. Kendal, H.D. Klenk, resistant strains of virus [75]. During a known A. McMichael, K.G. Nicholson and A Oya), influenza outbreak most people with acute Excerpta Medica, Amsterdam, pp. 161-5. 6. Ashley, J., Smith, T. and Dunell, K. (1991) onset of nasal symptoms, feverishness, shiver• Deaths in Great Britain associated with the ing, cough, headache, myalgia or anorexia, influenza epidemic of 1989/90. Population without vomiting or diarrhoea, will have Trends, 65, 16-20. influenza [79,186] and can be considered for 7. Austin, D.F., Karp,S., Dworsky, Rand treatment, particularly those in high-risk Henderson, B.E. (1975) Excess leukemia in groups, in whom complications can be cohorts of children born following influenza expected. epidemics. Am. J. Epidemiol., 101, 77-83. 8. Barker, W.H. and Mullooly, J.P. (1982) Pne• The recommended prophylactic and thera• umonia and influenza deaths during epi• peutic dose of amantadine is 200 mg daily, demics. Arch. Intern. Med., 142, 8~9. reduced to 100 mg in those 10-15 years of age 9. Barton, L.L. and Chalhub, E.G. (1975) Myositis or over 65; the suggested dose in children associated with influenza A infection (letter). J. aged 1-9 years is 2-4 mWkg. The likelihood Pediatr., 87,1003-4. that drug resistance will increase with exten• 10. Belshe, RB., Smith, M.H., Hall, C.B., Betts, R sive use of amantadine, together with minor and Hay, AJ. (1988) Genetic basis of resistance to rimantadine emerging during treatment of CNS adverse effects and the logistic difficulties influenza virus infection. J. Virol., 62,1508-12. in organizing timely prophylaxis and treat• 11. Belshe, RB., Burk, B., Newman, F., Cerruti, ment, underscore the importance of immu• RL. and Sim, 1.5. (1989) Resistance of influenza nization. On balance the adamantanes are still A virus to amantadine and rimantadine: results References 301

of one decade of surveillance. f. Infect. Dis., 159, 25. Coffey, V.P. and Jessup, W.J.E. (1959) Maternal 430-5. influenza and congenital deformities. Lancet, ii, 12. Bendkowski. B. (1958) Asian influenza (1957) in 935-8. allergic patients. Br. Med. ]., 2,1314-15. 26. Connolly, A.M., Salmon, RL. and Williams, 13. Bental, E. (1958) Acute psychoses due to D.H. (1993) What are the complications of encephalitis following Asian influenza. Lancet, influenza and can they be prevented? ii,I8-20. Experience from the 1989 epidemic of H3N2 14. Beswick, 1.P. and Finlayson, R. (1959) A renal influenza A in general practice. Br. Med. ]., 306, lesion in association with influenza. f. c/in. 1452-4. Pathol., 12, 280-5. 27. Corey, L., Rubin, R.J., Thompson, T.R. et al. 15. Bouter, A.M., Diepersloot, R.J.A., van (1976) A nationwide outbreak of Reye's syn• Romunder, L.K.J. et al. (1991) Effect of epidemic drome: its epidemiologic relationship to influenza on ketoacidosis, pneumonia and influenza B. Am. f. Med., 61, 204-8. death in diabetes mellitus: a hospital register 28. Corey, L., Rubin, RJ., Thompson, T.R et al. survey of 1976--1979 in the Netherlands. Diab. (1977) Influenza B- associated Reye's syn• Res. c/in. Prac., 12, 61-8. drome: incidence in Michigan and potential for 16. Brocklebank, J.T., Court, S.D.M., McQuillin, J. prevention. f. Infect. Dis., 135, 398-407. and Gardner, p.s. (1972) Influenza A infections 29. Couch, R.B., Douglas, RG., Fedson, D.S. et al. in children. Lancet, ii, 497-500. (1971) Correlated studies of a recombinant 17. Bryson, Y., Monahan, c., Pollack, M. and influenza-virus vaccine. III. Protection against Shields, W.D. (1980) A prospective double• experimental influenza in man. J. Infect. Dis., blind study of side-effects associated with the 124, 473-80. administration of amantadine for influenza A 30. Couch, R.B., Kasel, J.A., Gerin, J.L. et al. (1974) prophylaxis. f. Infect. Dis., 141, 543-7. Induction of partial immunity to influenza by a 18. Buchan, K.A. and Reid, D. (1972) neuraminidase-specific vaccine. f. Infect. Dis., Epidemiological aspects of influenza in Lewis. 129,411-20. Health Bulletin, 30, 183-6. 31. Couch, RB., Kasel, J.A., Glezen, W.P. et al. 19. Burch, G.E., Walsh, N. and Mogabgab, W.J. (1986) Influenza: its control in persons and (1959) Asian influenza - clinical picture. Arch populations. J. Infect. Dis., 153, 431-40. Intern. Med., 103, 696--707. 32. Creighton, C. (1965) A History of Epidemics in 20. Burnet, F.M. (1936) Influenza virus on the Britain, Vols 1 and 2, 2nd edn, Frank Cass, developing egg. 1. Changes associated with the London, and Barnes & Noble, New York. development of an egg-passage strain of virus. 33. Crosby, A.W. (1989) America's Forgotten Br. f. Exp. Pathol., 17, 282-93. Pandemic, Cambridge University Press, 21. Caul, ED., Waller, D.K. and Clarke, S.K.R. Cambridge. (1976) A comparison of influenza and respira• 34. Curwen, M., Dunnell, K. and Ashley, J. (1990) tory syncytial virus infections among infants Hidden influenza deaths. Br. Med. ]., 300, 896. admitted to hospital with acute respiratory 35. Davenport, F.M., Minuse, E., Hennessy, A.V. infections. f. Hyg. (Camb), 77,383-92. and Francis, T. (1969) Interpretations of 22. Chakraverty, P., Cunningham, P., Shen, G.Z. influenza antibody patterns in man. Bull. and Pereira, M.s. (1986) Influenza in the WHO, 40, 453-60. United Kingdom 1982-85. f. Hyg. (Cam b), 97, 36. Davis, L.E., Caldwell, G.G., Lynch, RE., Bailey, 347-58. R.E. and Chin, T.D.Y. (1974) Hong Kong 23. Chin, T.D.Y, Foley, J.F., Doto, LL., Gravelle, influenza: the epidemiologic features of a high C.R and Weston, J. (1960) Morbidity and mor• school family study analysed and compared tality characteristics of Asian strain influenza. with a similar study during the 1957 Asian Pub!. Health Rep., 75, 148-58. influenza epidemic. Am. f. Epidemiol., 92, 240-7. 24. Cockburn, W.c., Delon, P.J. and Ferreira, W. 37. Department of Health, Welsh Office, Scottish (1969) Origin and progress of the 1968-69 Home and Health Department. (1992) Hong Kong influenza epidemic. Bull. WHO,41, Immunization against infectious diseases, 345-8. Influenza, London HMSO, pp. 95-9. 302 Influenza viruses: clinical spectrum and management

38. DiBona, F.J. and Morens, D.M. (1977) 54. Fry, J. and Hume, E.M. (1957) Contrasts with Rhabdomyolysis associated with influenza. previous epidemics. Br. Med. J., ii, 1057. Am. J. Pediatr., 91, 943-5. 55. Fukumi, H. (1969) Interpretation of influenza 39. Dietzman, D.E., Schaller, J.G., Ray, CG. et al. antibody patterns in man. Bull. WHO 41, (1976) Acute myositis associated with influenza 469-73. B infection. Pediatrics, 57, 255-8. 56. Galbraith, A.W., Oxford, J.S., Schild, G.C and 40. Dolin, R, Reichman, R.C, Madore, H.P., Watson, G.!. (1969) Study of amantadine Maynard, R, Linton, P.M. and Webber-Jones, J. hydrochloride used prophylactically during (1982) A controlled trial of amantadine and the Hong Kong influenza epidemic in the fam• rimantadine in the prophylaxis of influenza A ilyenvironment. Bull. WHO, 41, 677-82. infection. N. Engl. ]. Med., 307, 580-4. 57. Gibson, T.C, Arnold, J., Craige, E. and Curnen, 41. Dubowitz, V. (1958) Influenzal encephalitis. E.C. (1959) Electrocardiographic studies in Lancet, i, 140-1. Asian influenza. Am. Heart J., 57, 661-8. 42. Edmundson, P. and Hodgkin, K. (1957) 58. Giles, C. and Shuttleworth, E.M. (1957) Post• Protean symptomatology. Br. Med. /., 2, 1058. mortem findings in 46 influenza deaths. Lancet, 43. Edward, D.G. (1941) Resistance of influenza ii,1224-5. virus to drying and its demonstration in dust. 59. Gilroy, J. (1957) Asian influenza. Br. Med. ]. 2, Lancet, ii, 664-6. 997-8. 44. Eickhoff, T.C, Sherman, I.L. and Serfling, RE. 60. Glezen, W.P. (1980) Consideration of the risk of (1961) Observations on excess mortality associ• influenza in children and indications for pro• ated with epidemic influenza. JAMA, 176, phylaxis. Rev. Infect. Dis., 2, 408-20. 776-82. 61. Glezen, W.P. (1982) Serious morbidity and 45. Fawdry, A.L. (1957) Influenza epidemic in mortality associated with influenza epidemics. Aden. Lancet, ii, 335--6. Epidemiol. Rev., 4, 25-44. 46. Flewett, T.H. and Hoult, J.G. (1958) Influenza 62. Glezen, W.P. and Couch, RB. (1978) encephalopathy and postinfluenza encephali• Interpandemic influenza in the Houston area, tis. Lancet, ii, 11-15. 1974-76. N. Engl. J. Med., 298, 587-92. 47. Fox, J.P., Cooney, M.K., Hall, CE. et al. (1982) 63. Glezen, W.P., Couch, RB., Taber, L.H. et al. Influenza virus infections in Seattle families, (1980) Epidemiologic observations of influenza 1975-79: I. Study design, methods and the B virus infections in Houston, Texas, 1976-77. occurence of infections by time and age. Am. J. Epidemiol., 116, 212-27. Am. J. Epidemiol., 111, 13-22. 48. Fox, J.P., Cooney, M.K., Hall, CE. et al. (1982) 64. Glezen, W.P., Paredes, A. and Taber, L.H. Influenza virus infections in Seattle families, (1980) Influenza in children: relationship to 1975-79: II. Pattern of infection in invaded other respiratory agents. JAMA, 243, 1345-9. households and relation to age and prior anti• 65. Govan, A.D.T. and Macdonald, H.RF. (1957) body to occurrence of infection and related ill• Influenza complicating heart disease in preg• ness. Am. J. Epidemiol., 116, 228-42. nancy. Lancet, ii, 891. 49. Foy, H.M., Hall, CE., Cooney, M.K. et al. (1983) 66. Griffiths, p.o., Ronalds, c.J. and Heath, R.B; Influenza surveillance in the Pacific Northwest, (1980) A prospective study of influenza infec• 1976-80. Int. J. Epidemiol., 12, 353--6. tions during pregnancy. J. Epidemiol. Commun. 50. Francis, T. (1953) Influenza: the newe Health, 34, 124-8. acquayantance. Ann. Intern. Med., 39, 203-21. 67. Gross, P.A., Quinnan, G.V., Rodstein, M. et al. 51. Frank, A.L., Taber, L.H., Wells, CR., Wells, (1988) Association of influenza immunization J.M., Glezen, W.P. and Paredes, A. (1981) with reduction in mortality in an elderly popu• Patterns of shedding of myxoviruses and lation. Arch. Intern. Med., 148, 562-5. paramyxoviruses in children. J. Infect. Dis., 144, 68. Grunert, R.R and Hoffman, CE. (1977) 433-41. Sensitivity of influenza NNew Jersey/8/76 52. Fry, J. (1958) Clinical and epidemiological fea• (Hsw1N1) virus to amantadine HCl. J. Infect. tures in a general practice. Br. Med. J., i, 259~1. Dis., 136, 297-300. 53. Fry, J. (1970) A report on the influenza epi• 69. Guthrie, J., Forsyth, D.M. and Montgomery, H. demic (NHong Kong) 1969-70. Update, 2, 369. (1957) Asiatic influenza in the middle east: an References 303

outbreak in a small community. Lancet, ii, rimantadine-resistant influenza A virus in fam• 590-2. ilies. N. Engl. J. Med., 321, 1696-702. 70. Hakosalo, J. and Saxen, L. (1971) Influenza epi• 82. Hayslett, J., McCarroll, J., Brady, E., Deuschle, demics and congenital defects. Lancet, ii, K, McDermott, W. and Kilbourne, E.D. (1962) 1346-7. Endemic influenza. I. SerolOgic evidence of 71. Hall, C.E., Cooney, M.K and Fox, J.P. (1973) continuing and subclinical infection in dis• The Seattle virus watch. IV Comparative epi• parate populations in the post-pandemic demiological observations of infections with period. Am. Rev. Respir. Dis., 85, 1-8. influenza A and B viruses., 1965-1969, in fami• 83. Heider, H., Adamczyk, B., Presber, H.W., lies with young children. Am. J. Epidemiol., 98, Schroeder, c., Feldblum, R, and Indulen, M.K. 365-80. (1981) Occurrence of amantadine and rimanta• 72. Hall, C.B. and Douglas, RG. (1975) Nosocomial dine resistant influenza A virus strains during influenza infection as a cause of intercurrent the 1980 epidemic. Acta Virol., 25,395-400. fevers in children. Pediatrics, 55, 673-7. 84. Heikkinen, T., Waris, M. and Ruuskanen, O. 73. Hall, W.J., Douglas, RG., Hyde, RW., Roth, (1993) Efficacy of influenza vaccination in F.K., Cross, AS. and Speers, D.M. (1976) reducing otitis media in children, in Options for Pulmonary mechanics after uncomplicated the Control of Influenza II, (eds C. Hannoun, AP. influenza A infection. Am. Rev. Respir. Dis., 113, Kendal, H.D. Klenk, A McMichael, K.G. 141-7. Nicholson and A. Oya), Excerpta Medica, 74. Hall, C.B., Douglas, R.G., Geiman, J.M. and Amsterdam, pp. 25-37. Meagher, M.P. (1979) Viral shedding patterns 85. Henderson, F.W., Collier, A.M., Sanyal, M.A. et of children with influenza B infection. ]. Infect. al. (1982) A longitudinal study of respiratory in Dis., 140, 610-13. viruses and bacteria the etiology of acute oti• tis media with effusion. N. Engl. J. Med., 306, 75. Hall, C.B., Dolin, R, Gala, c.L. et al. (1987) Children with influenza A infection: treatment 1377-83. 86. Hers, J.F.P. (1966) Disturbances of the ciliated with rimantadine. Pediatrics, 80, 275-82. epithelium due to influenza virus. Am. Rev. 76. Hardy, J.M.B., Azarowicz, EN., Mannini, A., Respir. Dis., 93, 162-71. Medearis, DN. and Cooke, RE. (1961) The 87. Hers, J.F., Gosling, W.RO., Masurel, N. and effect of Asian influenza on the outcome of Mulder, J. (1957) Death from Asiatic influenza 1957-1958. pregnancy, Baltimore Am. J. Publ. in the Netherlands. Lancet, ii, 1164-5. Health, 51, 1182--8. 88. Hers, J.F.P., Masurel, N. and Mulder, J. (1958) 77. Hay, AJ., Wolstenholme, A.J., Skehel, J.J. and Bacteriology and histopathology of the respira• Smith, M.H. (1985) The molecular basis of the tory tract and lungs of fatal asian influenza. specific anti-influenza action of amantadine. Lancet, ii, 1141-3. EMBO ].,4,3021-4. 89. Hirsch, A. (1883) Influenza, in Handbook of geo• 78. Hay, AJ., Zambon, M.C., Wolstenholme, A.J., graphical and historical pathology, Vol I, Acute Skehel, J.J. and Smith, M.H. (1986) Molecular infective diseases, New Sydenham Society, basis of resistance of influenza A viruses to London, pp. 7-54. amantadine. J. Antimicrob. Chemother., 18 90. Holland, W.W. (1957) A clinical study of (suppl. B), 19-29. influenza in the Royal Air Force. Lancet, ii, 79. Hayden, F.G., Hall, W.J. and Douglas, RG. 840-1. (1980) Therapeutic effects of aerosolized aman• 91. Hope-Simpson, RE. (1970) First outbreak of tadine in naturally acquired infection due to Hong Kong influenza in a general practice influenza A virus. J. Infect., 141, 535-42. population in Great Britain. A field and labora• 80. Hayden, F.G., Cote, KM. and Douglas, R.G. tory study. Br. Med. J., 3, 74-7. (1980) Plaque inhibition assay for drug suscep• 92. Homer, F.A. (1958) Neurologic disorders after tibility testing of influenza viruses. Antimicrob. Asian influenza. N. Engl. J. Med., 258, 983-5. Agents Chemother., 17, 865-70. 93. Homer, G.J. and Gray, F.D. (1973) Effect of 81. Hayden, F.G., Belshe, RB., Clover, RD., Hay, uncomplicated, presumtive influenza on the A.J., Oakes, M.G. and Soo, W. (1989) diffUSing capacity of the lung. Am. Rev. Respir. Emergence and apparent transmission of Dis., 108, 866-9. 304 Influenza viruses: clinical spectrum and management

94. Hoult, J.G. and Flewett, T.H. (1958) Influenzal Influenza A (Asian strain) viruses from encephalopathy and postinfluenzal encephali• influenza patients with pulmonary complica• tis. Histological and other observations. Br. tions. Details of virus isolation and characteri• Med. ]., i, 1847-50. zation of isolates, with quantitative comparison 95. Housworth, J. and Langmuir, A.D. (1974) of isolation methods.]' Clin. Invest., 38, 266-74. Excess mortality from epidemic influenza, 109. Kim, H.W., Brandt, C.W., Arrobio, J.O., 1957-1966. Am.]. Epidemiol., 100, 40-8. Murphy, B., Chanock, R.M. and Parrott, R.H. 96. Housworth, Wl and Spoon, M.M. (1971) The (1979) Influenza A and B infections in infants age distribution of excess mortality during A2 and young children during the years Hong Kong influenza epidemics compared 1957-1976. Am.]. Epidemiol., 109, 464-79. with earlier A2 outbreaks. Am. ]. Epidemiol., 94, 110. Kubar, 0.1., Brjantseva, E.A., Nikitina, L.E. and 348-50 Zlydnikov, D.M. (1989) The importance of 97. Howard, J.B., McCracken, G.H. and Luby, J.P. drug-resistance in the treatment of influenza (1972) Influenza A2 virus as a cause of croup with rimantadine. Antiviral Res., 11, 313-16. requiring tracheotomy.]. Pediatr., 81,1148-50. 111. Langmuir, A.D., Worthen, T.D., Slomon, J., 98. Howells, C.H.L., Vesselinova-Jenkins, C.K., Ray, C.K. and Patersen, E. (1985) The Evans, A.D. and James, J. (1975) Influenza vac• Thucydides syndrome: a new hypothesis for cination and mortality from bronchopneumo• the cause of the plague of Athens. N. Eng/. ]. nia in the elderly. Lancet, i, 381-3. Med., 313, 1027-30. 99. H0ygaard, A. (1939) Acute epidemic diseases 112. Lindsay, M.I., Herrmann, E.C., Morrow, G.W. among Eskimos in Angmagssalik. Lancet, i, and Brown, A.L. (1970) Hong Kong influenza: 245-6. clinical, microbiologic, and pathologic features 100. Jamieson, W.M., Kerr, M., Green, D.M. et al. in 127 cases. lAMA, 214, 1825-32. (1958) Some aspects of the recent epidemic of 113. Little, J.W., Hall, W.H., Douglas, R.G., influenza in Dundee. Br. Med.]., 1, 908-13. Mudholkar, G.S., Speers, D.M. and Patel, K. 101. Jaimovich, D.G., Kumar, A., Shabino, c.L. and (1978) Airway hyperreactivity and peripheral Formoli, R. (1992) Influenza B virus infection airway dysfunction in influenza A infection. associated with non-bacterial septic shock like Am. Rev. Respir. Dis., 118, 298-303. illness.]. Infect., 25, 311-15. 114. Liu, C. (1956) Rapid diagnosis of human 102. Johanson, W.G., Pierce, A.K. and Sanford, J.P. influenza infection from nasal smears by (1969) Pulmonary function in uncomplicated means of fluorescein-labelled antibody. Proc. influenza. Am. Rev. Respir. Dis., 100,141-6. Soc. Exp. Bioi. Med., 92, 883-7. 103. Jordan, W.S. (1961) The mechanism of spread 115. Lloyd-Still, R.M. (1958) Psychosis following of Asian influenza. Am. Rev. Respir. Dis., 83, Asian influenza. Lancet, ii, 20-1. 29-35 116. Louria, D.B., Blumenfeld, H.L., Ellis, J.T., 104. Jordan, W.S., Denny, F.W., Badger, G.F. et al. Kilbourne, E.D. and Rogers, D.E. (1959) Studies (1958) A study of illness in a group of on influenza in the pandemic of 1957-1958. II. Cleveland families. XVII. The occurrence of Pulmonary complications of influenza. l. Clin. Asian influenza. Am. ]. Hyg., 68, 190-212. Invest., 38, 213-65. 105. Karjalainen, J., Nieminen, M.S. and Heikkila, J. 117. MacDonald, K.L., Oserholm, M.T., Hedberg, (1980) Influenza Al myocarditis in conscripts. C.W. et al. (1987) Toxic shock syndrome. A Acta Med. Scand., 207, 27-30. newly recognised complication of influenza 106. Kerr, A.A., Downham, M.A.P.s., McQuillin, J. and influenza-like illness. lAMA, 257, 1053-S. and Gardner, p.s. (1972) Gastric 'flu influenza 118. Marine, W.M. and Workman, W.M. (1969) B causing abdominal symptoms in children. Hong Kong influenza immunologic recapitula• Lancet, i, 291-5. tion. Am. l. Epidemiol., 90, 406-15. 107. Khakpour, M., Saidi, A. and Naficy, K. (1969) 119. Martin, C.M., Kunin, C.M., Gottlieb, L.S., Proved viraemia in Asian influenza (Hong Barnes, M.W., Liu, C. and Finland, M. (1959) Kong variant) during incubation period. Br. Asian influenza A in Boston, 1957-1958. Arch. Med. ]., 4, 208-9. Intern. Med., 103, 515-31. 108. Kilbourne, E.D. (1959) Studies on influenza in 120. Martin, C.M., Kunin, C.M., Gottlieb, L.S. and the pandemic of 1957-1958. III Isolation of Finland, M. (1959) Asian influenza A in Boston, References 305

1957-1958: Severe staphylococcal pneumonia 131. Minuse, E., Willis, P.W .. Davenport, F.M. and complicating influenza. Arch. Intern. Med., 103, Francis, T. (1962) An attempt to demonstrate 532-42. viremia in cases of Asian influenza. J. Lab. Clin. 121. Mast, E.E., Davis, J.P., Harmon, M.W., Arden, Med., 59,1016-19. N.H., Circo, Rand Tyszka, G.E. (1989) 132. Center for Disease Control. Editorial. (1992) Emergence and possible transmission of aman• Prevention and control of influenza. tadine-resistant viruses during nursing home Recommendations of the immunization prac• outbreaks of influenza A(H3N2), in Program tices advisory committee. MMWR, 41, No. and Abstracts of the 29th Interscience Conference on RR-9,1-17. Antimicrobial Agents and Chemotherapy, 133. Monto, A.S. and Kioumehr, F. (1975) The September 18, 1989, Houston, Washington, Tecumseh study of respiratory illness. IX DC. American Society for Microbiology, pp. Occurrence of influenza in the community, 111 (abstract). 1966-1971. Am. J. Epidemiol., 102, 553-63. 122. Masurel, N. (1969) Relation between Hong 134. Monto, A.S., Koopman, J.S. and Longini, I.M. Kong virus and former A2 isolates and the (1985) The Tecumseh study of illness. XIII NEqui 2 virus in human sera collected before influenza infection and disease, 1976-1981. Am. 1957. Lancet, i, 907-10. I. Epidemiol., 121, 811-22. 123. McBean, A.M., Babish, J.D., Warren, J.L. and 135. Morgensen, J.L. (1974) Myoglobinuria and Melson, E.A. (1993) The effect of influenza epi• renal failure associated with influenza. Ann. demics on the hospitalisation of persons 65 Intern. Med., 80, 362-3. years and older, in Options for the Control of 136. Morris, J.A., Kasel, J.A., Saglam, M., Knight, V. Influenza II, (eds C. Hannoun, A.P. Kendal, and Loda, F.A. (1966) Immunity to influenza as H.D. Klenk, A. McMichael, KG. Nicholson and related to antibody levels. N. Engl. I. Med., 274, A. Oya), Excerpta Medica, Amsterdam, pp. 527-35. 25-37. 137. Moss, S.E., Klein, R and Klein, B.E.K. ( 1991) 124. McConkey, B. and Daws, RA. (1958) Cause-specific mortality in a population-based Neurological disorders associated with Asian study of diabetes. Am. I. Publ. Health, influenza. Lancet, ii, 15-17. 81,1158--62. 125. McGregor, J.A., Burns, J.c., Levin, M.J. et al. 138. Mulder, J. (1957) Asiatic influenza in the (1984) Transplacental passage of influenza Netherlands. Lancet, ii, 334. NBangkokl(H3N2) mimicking amniotic fluid 139. Mulder, J. and Masurel, N. (1958) Pre-epidemic infection syndrome. Am. I. Obstet. Gynecol., 149, antibody against 1957 strain of Asiatic 856-9. influenza in serum of older people living in the 126. Meibalane, R., Sedmak, G.V., Sasisharan, P., Netherlands. Lancet, i, 810-14. Garg, P. and Grausz, J.P. (1977) Outbreak of 140. Murphy, B.R, Baron, S., Chalhub, E.G., influenza in a neonatal intensive care unit. I. Uhlendorf, c.P. and Chanock, RM. (1973) Infect. Dis., 91, 974-6. Temperature-sensitive mutants of influenza 127. Middleton, P.J., Alexander, RM. and virus. IV. Induction of interferon in the Szymanski, M.T. (1970) Severe myositis during nasopharynx by wild-type and a temperature• recovery from influenza. Lancet, ii, 533-5. sensitive recombinant virus. I. Infect. Dis., 128, 128. Ministry of Health. (1960) The influenza epidemic 488-93. in England and Wales, 1957-1958. HMSO, 141. Naficy, K (1963) Human influenza infection London. with proved viraemia. N. Engl. I. Med., 269, 129. Minor, T.E., Dick, E.c., Baker, J.W., Ouellette, 964-6. J.J., Cohen, M. and Reed, c.E. (1976) 142. Nakajima, S., Nakamura, K., Nishikawa, F. and Rhinovirus and influenza type A infections as Nakajima, K. (1991) Genetic relationship precipitants of asthma. Am. Rev. Respir. Dis., between the HA genes of type A influenza 113,149-53. viruses isolated in off-seasons and later epi• 130. Minow, RA., Gorbach, S., Johnson, B.L. et al. demic seasons. Epidemiol. Infect., 106, 383-95. (1974) Myoglobinuria associated with 143. Nakajima, S., Nishikawa, F., Nakamura, K and influenza A infection. Ann. Intern. Med., 80, Nakajima, K (1992) Comparison of the HA 359-361. genes of type B influenza viruses in herald 306 Influenza viruses: clinical spectrum and management

waves and later epidemic seasons. Epidemiol. symptoms in young children. c/in. Pathol., 15, Infect., 109, 559-68. 361-7. 144. Newton, L. and Hall, S.M. (1993) Reye's syn• 158. Pryor, H.B. (1964) Influenza: that extraordinary drome in the British Isles: report for 1990/91 malady. Notes on its history and epidemiol• and the first decade of surveillance. Comm. Dis. ogy. c/in. Pediatr., 3,19-24. Rep. Rev., 3, Rl1-16. 159. Public Health Laboratory Service. (1958) 145. Nguyen-Van-Tam, J.5. and Nicholson, KG. Deaths from Asian influenza, 1957. Br. Med. ]., (1992) Influenza deaths in Leicestershire dur• i,915-19. ing the 1989-90 epidemic: implications for pre• 160. Ramphal, R., Donnelly, W.H. and Small, P.A. vention. Epidemiol. Infect., 108, 537-45. (1980) Influenza pneumonia in pregnancy: fail• 146. Nicholson, KG. (1984) Antiviral agents in clini• ure to demonstrate transplacental transmission cal practice: properties of antiviral agents. of influenza virus. Am. ]. Obstet. Gynecol., 138, Lancet, ii, 562-4. 347-8. 147. Nicholson, KG., Kent, J. and Ireland, D.C. 161. Roberts, G.B.5. (1957) Fulminating influenza. (1993) Respiratory viruses and exacerbations of Lancet, ii, 944-5. asthma in adults. Br. Med. ]., 307, 982-6. 162. Roden, A.T. (1969) National experience with 148. Oswald, N.C., Shooter, R.A. and Curwen, M.P. Hong Kong influenza in the United Kingdom, (1958) Pneumonia complicating Asian 1968-69. Bull. WHO, 41, 375-80. influenza. Br. Med.]., ii, 1305-11. 163. Rowland, H.A.K (1958) The influenza epi• 149. Oxford, J,S. and Galbraith, A. (1980) Antiviral demic in Abadan. Br. Med. ]., i, 422-- 5. activity of amantadine: a review of laboratory 164. Saenz, AC., Assad, FA and Cockburn, W.c. and clinical data. P1uirmacol. Ther., 11, 181-262. (1969) Outbreak of A2/Hong Kongl68 influenza 150. Paisley, J.W., Bruhn, F.W., Lauer, B.A. and at an international medical conference. Lancet, McIntosh, K (1978) Type A2 influenza viral i,91-3. infections in children. Am. ]. Dis. Child., 132, 165. Scholtissek, c., Rott, R. and Klenk, H.-D. (1975) 34-6. Two different mechanisms of the multiplica• 151. Parsons, H.F. (1891) Report on the influenza epi• tion of enveloped viruses of glucosamine. demic of 1889-90. Local Government Board, Virology, 63, 191-4. HMSO, London. 166. Serfling, R.E., Sherman, !.L. and Housworth, 152. Patriarca, P.A., Weber, J.A., Parker, R.A. et al. W.J. (1967) Excess pneumonia-influenza mor• (1985) Eficacy of influenza vaccine in nursing tality by age and sex in three major influenza homes. Reduction in illness and complications A2 epidemics, United States, 1957-58, 1960 and during an influenza A (H3N2) epidemic. 1963. Am.]. Epidemiol., 86, 433-41. JAMA, 253, 1136-9. 167. Serie, c., Barme, M., Hannoun, c., Thibon, M., 153. Patriarca, P.A., Weber, JA, Parker, R.A. et al. Beck, H. and Aquino, J.P. (1977) Effects of vac• (1986) Risk factors for outbreaks of influenza in cination on an influenza epidemic in a geriatric nursing homes. Am.]. Epidemiol., 124, 114-19. hospital. Dev. BioI. Stand., 39, 317-21. 154. Perrotta, D.M., Decker, M. and Glezen, W.P. 168. Sharrar, R.G. (1969) National influenza experi• (1985) Acute respiratory disease hospitaliza• ence in the USA, 1968-69. Bull. WHO, 41, 361-6. tions as a measure of impact of epidemic 169. Simon, N.M., Rovner, R.N. and Berlin, B.S. influenza. Am. ]. Epidemiol., 122, 468-76. (1970) Acute myoglobinuria associated with 155. Petersdorf, R.G., Fusco, J.J., Harter, D.H. and type A2 (Hong Kong) influenza. JAMA, 212, Albrink, W.S. (1959) Pulmonary infections com• 1704-5. plicating Asian influenza. Arch. Intern. Med., 170. Smith, W., Andrewes, C.H. and Laidlaw, P.P. 103,262--72. (1933) A virus obtained from influenza 156. Polak, M.F. (1959) Influenzasterfte in de herfst patients. Lancet, ii, 66-8. van 1957. Ned. Tijdschr. Geneeskd., 103, 171. Smith, AP., Tyrrell, DAJ., Al-Nakib, W., 1098-109. Barrow, G.!., Higgins, P.G., Leekam, S. and 157. Price, D.A., Postlethwaite, R.J. and Longson, M. Tricket, S. (1989) Effects and after effects of the (1976) Influenza virus A2 infections presenting common cold and influenza on human perfor• with febrile convulsions and gastrointestinal mance. Neuropsychobiology, 21, 90-3. References 307

172. Smith, A.P.,Thomas, M., Brockman, P., Kent, J. Successful treatment of naturally occurring and Nicholson, KG. (1993) Effect of influenza influenza NUSSR/77 HlN1. lAMA, 245, B infection on human performance. Br. Med. ]., 1128-31. 306,760-1. 187. Walsh, J.J., Dietlein, L.F., Low, F.N., Burch, G.E. 173. Sprenger, M.J.W., Beyer, W.E.P., Kempen, B.M. and Mogabgab, W.J. (1961) Bronchotracheal and Mulder, P.G.H. (1993) Risk factors for response in human influenza. Arch. Intern. influenza mortality? in Options for the Control of Med., 108, 376--88. Influenza II, (eds C Hannoun, A.P. Kendal, 188. Whitaker, A.N., Bunce, I. and Graeme, E.R. H.D. Klenk, A. McMichael, KG. Nicholson and (1974) Disseminated intravascular coagulation A. Oya), Excerpta Medica, Amsterdam, pp. and acute renal failure in influenza A2 infec• 15-23. tion. Med.]. Aust., 2, 196--201. 174. Stanley, E.D. and Jackson, G.G. (1966) Viraemia 189. WHO. (1968) Influenza: The Hong Kong virus. in Asian influenza. Trans. Assoc. Am. Physicians, WHO Chron., 22, 528-9. 1,376--87. 190. WHO. (1985) Current status of amantadine 175. Stocks, P. (1935) The effect of influenza epi• and rimantadine as anti-influenza-A agents: demics on the certified causes of death. Lancet, memorandum from a WHO meeting. Bull. ii, 386--95. WHO, 63, 51-6. 176. Stuart-Harris, CH. (1961) Twenty years of 191. Wilson, CB. and Stein, A.M. (1969) Teratogenic influenza epidemics. Am. Rev. Respir. Dis., 83, effects of Asian influenza. An extended study. 5~1. lAMA, 210, 336--7. 177. Stuart-Harris, CH. (1970) Pandemic influenza: 192. Wilson, CB. and Smith, R.C (1972) an unresolved problem in prevention. ]. Infect. Goodpasture's syndrome associated with Dis., 122,108-15. influenza A2 virus infection. Ann. Intern. Med., 178. Stuart-Harris, CH. and Schild, G.C (eds) 76,91-4. (1976) Influenza, the Viruses and Disease, Edward 193. Wingfield, W.L., Pollack, D. and Grunert, R.R. Arnold, London. (1969) Therapeutic efficacy of amantadine HCl 179. Taber, L.H., Paredes, A., Glezen, W.P. and and rimantadine HCl in naturally occurring A2 Couch, R.B. (1981) Infection with influenza respiratory illness in man. N. Engl. ]. Med., 281, ANictoria virus in Houston families, 1976. 1 579-84. Hyg. (Camb), 86,303-13. 194. Woodall, J., Rowson, KE.K and McDonald, 180. Thacker, S.B. (1986) The persistence of J.C (1958) Age and Asian influenza. Br. Med. l., influenza in human populations. Epidemiol. 4, 1316--18. Rev., 8, 129-42. 195. Wright, P.F., Ross, KB., Thompson, J. and 181. Thompson, T. Annals of Influenza in Great Karzon, D.T. (1977) Influenza A infections in Britain, Sydenham Society, London 1852. young children. Primary natural infection and 182. Tillett, H.E., Smith, J.w.G. and Clifford, R.E. protective efficacy of live-vaccine induced or (1980) Excess morbidity and mortality associ• naturally-acquired immunity. N. Engl. l. Med., ated with influenza in England and Wales. 296, 829-34. Lancet, i, 793-5. 196. Wright, P.F., Thompson, J., McKee, KT., 183. Tillett, H.E., Smith, J.W.G. and Gooch, CD. Vaughn, W.K., Sell, S.H.W. and Karzon, D.T. (1983) Excess deaths attributable to influenza in (1981) Patterns of illness in the highly febrile England and Wales: age at death and certified young child; epidemiologic, clinical, and labo• cause. Int. ]. Epidemiol., 12, 344-52. ratory correlates. Pediatrics, 67, 694-700. 184. Tolan, R.W. (1993) Toxic shock syndrome com• 197. Wynne-Griffith, G., Adelstein, A.M., Lambert, plicating influenza in a child: case report and P.M. and Weatherall, J.A.C (1972) Influenza review. Clin. Infect. Dis., 17, 43-5. and infant mortality. Br. Med.]., 3, 553-6. 185. Tominack, R.L. and Hayden, F.G. (1987) 198. Yawn, D.H., Pyeatte, J.C, Joseph, J.M., Eichler, Rimantadine hydrochloride and amantadine S.L. and Bunuel, R.G. (1971) Transplacental hydrochloride use in influenza A infections. transfer of influenza virus. lAMA, 216, 1022-3. Infect. Dis. Clin. North Am., 1, 459-78. 199. Zamkoff, K and Rosen, N. (1979) Influenza 186. Van Voris, L.P., Betts, R.F., Hayden, F.G., and myoglobinuria in brothers. Neurology, 29, Christmas, W.A. and Douglas, R.G. (1981) 340-5.