Influenza is a potentially severe acute respiratory illness caused by various strains of the influenza virus. The different strains all produce characteristic symptoms, and because major outbreaks are associated with increased mortality, occurrences can be identified in history. Outbreaks consistent with influenza can be traced back at least to the court of Elizabeth I. Some have speculated that the Plague of Athens described by Thucydides was influenza complicated by bacterial superinfection. The influenza syndrome, commonly known as the flu, with its fever, cough, rapid onset and body aches, is not only typical enough to be recognized in the past, but it also allows physicians to recognize it, especially when it is known that the virus is circulating. Unfortunately, death is the other consistent phenomena associated with influenza. Mortality statistics are the principal way the intensity of an influenza outbreak is quantified, and are so characteristic that viral identification of etiology is not required.
THE VIRUS AND ITS ANTIGENS
The influenza viruses contain RNA (ribonucleic acid) and are somewhat unusual in that they have a segmented genome, which means that there are eight distinct segments to the single-stranded RNA. Influenza types A and B are the only strains with epidemic potential; type C viruses are difficult to work with in the laboratory and are one of the multiple agents able to cause the common cold. While the viruses are classified into type A and B on the basis of their internal components, it is the surface antigens that are important in eliciting antibodies that will protect against future infection. These surface antigens and their changes make influenza challenging to control. Two types of changes are recognized.
One change occurs in both type A and B viruses and is a result of point mutations in the segments of the genome coding for two specific surface antigens (the neuraminidase [N] and the hemagglutinin [H] segments). These mutations are the reason that both type A and B viruses change regularly from year to year, though type A changes somewhat more rapidly than type B. Such changes are referred to as "antigen drift." Another change is more dramatic, only occurring with type A viruses, and is an example of "antigen shift." It takes place when one or two gene segments are replaced in a circulating virus. The same two antigens, or proteins, are involved in both types of change. The various influenza A viruses are categorized into subtypes by the differences in those two antigens, such as A (H1N1) or A (H3N2).
The most widely accepted theory explaining this antigen shift is that the segments come from animal influenza viruses. Type B influenza is confined to humans, while type A exists in numerous species of birds and domestic animals. There are fifteen types of hemagglutinin in the influenza virus of birds, but only three in human viruses, which gives an ample opportunity for the segment coding for the hemagglutinin to move from avian viruses to human. This has apparently happened in the past, and is likely to occur in the future, either directly or through pigs. In 1997, in Hong Kong, an avian virus infected humans directly, but did not become adapted to humans by exchange of gene segments. If it had, a pandemic undoubtedly would have resulted.
PANDEMICS
While some trace influenza pandemics back to ancient Greece, the first documented occurrence was in 1889 (see Table 1). In that and subsequent years, outbreaks of influenza were reported in many areas of the world, and in the United States, deaths reported in the state of Massachusetts for the first time demonstrated the U- or J-shaped mortality curve—an elevated mortality in young children, low mortality until age forty-five, followed by gradually increasing mortality with a relatively sharp inflection upward at age sixty-five (see Figure 1). By testing blood specimens of persons who lived through this period, researchers have been able to hypothesize about the strain of virus that caused this pandemic. In 1899, there was an apparent antigen shift, but this was determined serologically, not on the basis of an observed pandemic.
It is now certain that a virus resembling one isolated from pigs in the 1930s caused the devastating 1918 pandemic. No influenza viruses were isolated until the 1930s, so that any identification of viruses responsible for events occurring before that time has traditionally been done by testing the blood of people living through the period of an outbreak. Confirmation of this approach has recently taken place using modern molecular technique involving tissue of individuals who died during the 1918 pandemic. The virus is now termed A(H1N1). The estimated death toll from this pandemic has been revised upwards from 20 million to 40 million, since large segments of the world— mainly the current developing countries—were originally omitted from the counts. The lethality of this pandemic was related in large part to the death of an unexpectedly large number of healthy
young adults. This resulted in a W-shaped agespecific mortality curve (see Figure 1). It is hoped that genetic research with lung tissue, either stored or recovered from bodies, will enable epidemiologists to predict the potential behavior of future pandemic strains of influenza when they are identified. However, this has not as yet been possible, so it is only by observing the epidemiology of infection that the age-specific pattern of illness can be determined.
The first influenza viruses were isolated from humans in the early 1930s. However, the next pandemic did not occur until 1957, when the A(H2N2) virus appeared in South China (see Figure 1). The pandemic that resulted was the most severe since 1918, but again exhibited the more typical U-shaped mortality curve, concentrated in very young children and older individuals. A little more than ten years later, in 1968, the hemagglutinin changed and the resulting pandemic was similar to 1957 in age distribution, but more moderate in overall impact.
Two more episodes have occurred since 1968 that had the potential to be full pandemics. In 1977, the A(H1N1) virus returned, with outbreaks occurring first in China and then in the former Soviet Union. Since the virus had circulated twenty or more years before, when worldwide outbreaks occurred, these epidemics were confined to younger individuals. This virus has continued to circulate, along with the A(H3N2) and B viruses. Finally, in 1997, A(H5N1) moved from chickens to humans in Hong Kong. There were eighteen confirmed cases, with six deaths that were not restricted to older individuals. Fortunately, this avian virus did not
Figure 1
become fully adapted to humans. No human-to-human transmission was observed, but this episode showed how a catastrophic pandemic might have occurred had such adaptation taken place.
PREVENTION AND CONTROL OF INFLUENZA
A vaccine for the prevention of influenza was developed during World War II in order to maintain military readiness. This was done in recognition of the high morbidity that could result among troops exposed to the virus. A similar inactivated vaccine is still in use, improved in both potency and lack of side effects. It is known to be 70 to 90 percent efficacious in healthy young adults as long as the vaccine viruses resemble those circulating. This necessitates updating the viruses in the vaccine each year. For this and other reasons, the vaccine must be given annually. Since vaccination programs must be sustained, the goal in most countries has been to reduce influenza mortality by vaccinating older individuals and those with chronic underlying diseases. An exception to this has been Japan, where, for a time, school-age children were vaccinated in an effort to control influenza morbidity. It has been repeatedly demonstrated that the inactivated vaccine is effective in preventing hospitalization and death in older individuals and, as such, is also cost effective. The inactivated vaccine is cost effective in healthy adults only when the attack rates are above 12 percent. A live attenuated influenza vaccine has been used in the former Soviet Union for many years, and another is in development in the United States. Because of its delivery—intranasally rather than by injection—it may prove to be particularly useful in children and younger adults.
Antiviral drugs have been available both for treatment and prophylaxis. Two of these are active only against type A viruses. A new group of drugs, acting as neuraminidase inhibitors, is active against both type A and B viruses. These drugs have been shown to have a prevention efficacy similar to vaccines. They start protecting more quickly than the vaccine, but have to be taken daily to continue protection. Therefore, vaccination will continue to be the usual means of prophylaxis. The neuraminidase inhibitors also significantly shorten the duration of illness, reducing severity and preventing complications. Influenza can be debilitating, even in the absence of complications, so that the drugs will be used for treatment during defined influenza outbreaks. They are likely also to be useful prophylactically, especially for outbreak control in nursing homes.
ARNOLD MONTO
BIBLIOGRAPHY
Belshe, R. B.; Mendelman, P. M.; Treanor, J. et al. (1998). "The Efficacy of Live Attenuated, Cold-adapted, Trivalent, Intranasal Influenza Virus Vaccine in Children." New England Journal of Medicine 149:1113–1117.
Dolin, R.; Reichman, R. C.; Madore, H. P.; Maynard, R.; Linton, P. N.; and Webber-Jones, J. (1982). "A Controlled Trial of Amantadine and Rimantadine in the Prophylaxis of Influenza A Infection." New England Journal of Medicine 307:580–584.
Douglas, R. B., Jr. (1990). "Drug Therapy: Prophylaxis and Treatment of Influenza." New England Journal of Medicine 322:443–450.
Eickhoff, T. C.; Sherman, I. L.; and Serfling, R. E. (1961). "Observations on Excess Mortality Associated with Epidemic Influenza." Journal of the American Medical Association 176:776–782.
Gubareva, L. B.; Kaiser, L.; and Hayde, F. G. (2000). "Influenza Virus Neuraminidase Inhibitors." New England Journal of Medicine 355:827–835.
Lui, K. J., and Kendal, A. P. (1987). "Impact of Influenza Epidemics on Mortality in the United States from October 1972 to May 1985." American Journal of Public Health 77:712–716.
Monto, A. S., and Arden, N. H. (1992). "Implications of Viral Resistance to Amantadine in Control of Influenza A." Clinical Infectious Diseases 15:362–367.
Monto, A. S., and Kioumehr, F. (1975). "The Tecumseh Study of Respiratory Illness. IX. Occurrence of Influenza in the Community, 1966–1971." American Journal of Epidemiology 102:553–563.
Patriarca, P. A.; Weber, J. A.; Parker, R. A. et al. (1985). "Efficacy of Influenza Vaccine in Nursing Homes: Reduction in Illness and Complications During an Influenza A (H3N2) Epidemic." Journal of the American Medical Association 253:1136–1139.
Sullivan, F. M.; Monto, A. S.; and Longini, I. M., Jr. (1993). "Estimates of the U.S. Health Impact of Influenza." American Journal of Public Health 83: 1712–1716.
Wilde, J. A.; McMillan, J. A.; Serwint, J.; Butta, J; O'Riordan, M. A.; and Steinhoff, M. C. (1999). "Effectiveness of Influenza Vaccine in Health Care Professionals: A Randomized Trial." Journal of the American Medical Association 281:908–913.
