Lisa Danzig MD
Keiji Fukuda MD, MPH
Essentials of Diagnosis
Influenza is a highly contagious, acute, febrile respiratory illness caused by influenza A and B viruses. The hallmark of these viruses is their ability to undergo rapid ongoing antigenic change and to cause annual or near-annual epidemics of febrile respiratory disease affecting all age groups. In addition, the unpredictable emergence of new influenza A subtypes can lead to explosive global pandemics of disease. Although most cases are self-limited, influenza is a major source of mortality among those at increased risk for influenza-related complications. Annual vaccination of persons at high risk for serious complications is the most effective approach for reducing influenza-related morbidity and mortality.
Influenza viruses, which include influenza A virus, influenza B virus, influenza C virus, and thogotovirus, belong to the Orthomyxoviridiae family. Influenza A and B viruses cause epidemic disease in humans (influenza C viruses usually cause a mild, coldlike illness) and are designated by type, site, and year of isolation. Influenza A (but not B or C) viruses are further designated by subtype. For example, influenza A/Texas/36/91/(H1N1) refers to an influenza type A virus of the H1N1 subtype that was isolated in Texas in 1991 and has a strain designation of 36 to distinguish it from other 1991 Texas isolates of this subtype.
Influenza A(H1N1), influenza A(H3N2), and influenza B viruses are in worldwide circulation. Any one or any combination of these three viruses may be the cause of a season's epidemic activity. During each influenza season, an estimated 10–20% of the U.S. population may develop influenza, but attack rates of 40–50% within institutions are not unusual. In communities, influenza cases often appear first among school-age children, and attack rates in this group are usually the highest (and lowest among the elderly).
By contrast, rates of serious complications are highest among the elderly, the very young, and those with underlying chronic cardiopulmonary conditions that place them at high risk for complications. Influenza-related mortality is usually tabulated in terms of “excess” deaths. This concept was developed by William Farr and refers to the idea that influenza-related mortality can be quantified during influenza epidemics by determining the increase in deaths beyond the expected number of deaths if there was no epidemic. In the United States, the average annual toll of excess deaths from influenza-related complications is ~20,000; > 40,000 deaths may occur during severe years. Although pandemics are often associated with dramatic increases in mortality, the cumulative death toll from seasonal epidemic influenza since the 1968 pandemic exceeds the total for the 1918 pandemic (see below).
Pandemics of influenza occur when a new subtype of influenza A virus emerges and leads to increased morbidity and mortality in all age groups. This is a dramatic event that occurs relatively infrequently and unpredictably. The most devastating pandemic of the twentieth century began in 1918 and was associated with the emergence of the influenza A(H1N1) virus. Between the spring of 1918 and the spring of 1919, three waves of “Spanish flu” swept around the world leading to > 550,000 deaths in the United States and > 20 million deaths worldwide. In contrast to the usual mortality pattern, influenza-related deaths in this pandemic occurred for unknown reasons predominantly among people 20–40 years of age.
Two subsequent pandemics, which began in 1957 (the “Asian flu”) and in 1968 (“Hong Kong flu”), were associated with the emergence of the influenza A (H2N2)virus and the influenza A (H3N2) virus, respectively. In contrast to these events, the reappearance of influenza A (H1N1) virus in 1977 was not considered a true pandemic because illness largely was confined to people younger than 20 years of age. Based on these events, evidence of earlier pandemics, and the continuing evolution of influenza A viruses, the occurrence of another pandemic is likely.
The ability of influenza viruses to cause recurrent epidemics and pandemics can be attributed in large part to their propensity for antigenic change. This occurs through two distinct processes known as antigenic “drift” and “shift.” In antigenic drift, which both influenza A and B viruses exhibit, point mutations in the viral RNA result in immunologically significant alterations to HA and NA. Drift occurs more rapidly in influenza A subtypes than in influenza B viruses. Eventually, one of the newer influenza strains becomes the predominant strain because neutralizing antibody levels to older strains rise in the general population and exert a selective evolutionary pressure favoring the emergence of a new strain. Typically, a virus strain may predominate for a few years before a significantly different variant emerges to replace it.
Antigenic shift refers to the appearance of a virus bearing either a novel HA or a novel combination of HA and NA. Shift is exhibited only by influenza A viruses and may occur in one of two ways. Genes from influenza A viruses normally in circulation among pigs or birds may reassort (ie, mix) with genes from influenza A viruses circulating among humans, resulting in the appearance of a new influenza A virus subtype containing genes from both viruses. Pigs are thought to serve as a “mixing vessel” for genetic reassortment among influenza viruses because they are susceptible to avian, swine, and human influenza A viruses. Avian species are thought to serve as the ultimate reservoir of influenza A viruses. Alternatively, new influenza A virus subtypes may appear among humans as a result of direct transmission of animal viruses to humans. An example of this type of antigenic shift was seen recently in Hong Kong. During May through December 1997, a total of 18 people in Hong Kong were hospitalized because of disease related to influenza A (H5N1) infections. Previous to this outbreak, H5N1 viruses had been known to cause disease only in birds. The potential for this influenza A virus to cause a pandemic remains uncertain.
Antigenic shift results in a radical alteration in antigenicity. The emergence of a novel influenza A virus can lead to a pandemic but only if the new virus is sufficiently transmissible among humans to maintain epidemic activity and is capable of causing disease.
The viral envelope also contains two matrix proteins (M1 and M2), which form the basis for classifying influenza viruses into types A or B. M1 protein is thought to add rigidity to the lipid bilayer, whereas M2 protein functions as a pH-activated ion channel. The M2 protein is therapeutically important because it appears to contain the site of action for amantadine and rimantadine, two antiviral agents with specific activity against influenza A (but not influenza B) viruses.
Infection leads to the induction of both B- and T-cell responses. Neutralizing serum and mucosal antibodies to HA are the primary mediators of protection against infection and clinical illness. Serum antibodies to HA can persist for decades. Antibodies to NA are inefficient in neutralizing influenza viruses but help restrict the release of virus from infected cells, reduce the intensity of infection, and enhance recovery. The role of T-cell-mediated immunity is less well understood.
The spectrum of influenza infection ranges from subclinical cases to fulminating viral pneumonia.
In the elderly, fever may be absent and the presenting signs may be anorexia, lassitude, confusion, and rhinitis. In children, fevers are often higher and can lead to febrile seizures. Gastrointestinal manifestations, such as vomiting, abdominal pain, and diarrhea, and other complications such as myositis, croup (tracheobronchitis), and otitis media also occur more frequently in children. Unexplained fever may be the primary manifestation in neonates.
Virus can be isolated from nasal washing and nasopharyngeal swab specimens obtained within 3–4 days of illness. Virus is grown either in embryonated hens' eggs or in primary tissue culture systems, such as Madin-Darby canine kidney cells. Viral culture offers specific information and the ability to further characterize the isolate, but the sensitivity of this technique is highly dependent on the timing of when the specimen is obtained. Results usually are not available for at least 3 days.
Rapid diagnostic techniques to identify viral antigens in clinical specimens include immunofluorescence, enzyme immunoassay, and time-resolved fluoroimmunoassay. Several rapid test kits are now commercially available that either can 1) detect influenza A but not influenza B virus; 2) detect influenza A or B virus but not distinguish between them; or 3) detect influenza A or B virus and distinguish between them. In general, these tests are more specific than sensitive but head-to-head comparison data are not available. These tests can yield results within 30 min. Reverse transcriptase polymerase chain reaction assays have also been used to detect influenza virus RNA in clinical specimens.
Serologic techniques for measuring antibody against influenza include hemagglutination inhibition, neutralization, enzyme immunoassay, and complement fixation. In general, serology is a sensitive technique for establishing influenza infections. However, serologic tests usually require acute and convalescent serum samples to demonstrate a significant increase in antibody level. The measurement of influenza antibody levels in a single-serum sample is rarely helpful. Ideally, acute and convalescent blood samples should be collected, respectively, within 2–3 days of illness and at 3 weeks after the start of illness. The most commonly used serologic test to document influenza virus infection is hemagglutination inhibition because it (and neutralization) is more sensitive than complement fixation and allows subtype and strain-specific antibody to be measured.
BOX 29-1 Influenza
Other respiratory viruses, including respiratory syncytial virus, adenovirus, parainfluenza virus, and rhinovirus, as well as other organisms, such as Mycoplasma pneumoniae, can produce illness similar to influenza. However, outbreaks of febrile respiratory illness cases during the winter through spring months are characteristic of influenza. Information on local influenza activity is usually available from the local health department.
The most common serious complications of influenza include exacerbation of underlying chronic pulmonary and cardiopulmonary diseases, such as worsening of chronic obstructive pulmonary disease, asthma, and congestive heart failure, as well the development of pneumonia. Secondary bacterial pneumonias occur much more frequently than primary viral pneumonia and usually are associated with Streptococcus pneumoniae, Staphylococcus aureus, and Haemophilus influenzae. With complicating bacterial pneumonia, the patient typically reports a period of improvement followed by the appearance of signs and symptoms suggestive of pneumonia, such as pleuritic chest pain, productive cough, and fever. On chest radiography, lobar consolidation can be seen and sputum smears show polymorphonuclear leukocytes with bacteria.
Primary viral pneumonia is an infrequent but often fatal complication in which influenza progresses within the first 24–48 h of illness and leads to increasing dyspnea, tachypnea, and cyanosis. On presentation, fever and cough are usually present and the patient appears uncomfortable and dyspneic. On chest auscultation, diffuse fine rales with wheezes or coarse breath sounds may be evident. Sputum may be scanty but can be blood streaked. Chest roentgenograms usually show bilateral interstitial infiltrates or a picture consistent with acute respiratory distress syndrome. The Gram stain of the sputum often shows few polymorphonuclear cells or bacteria. Virologic cultures of the respiratory secretions often yield virus. The value of antiviral medications in this setting is unknown. Cases of viral pneumonia described in the 1918 and 1957 pandemics were associated with underlying cardiac valvular disease (frequently mitral stenosis from rheumatic heart disease) and pregnancy.
In addition to these pneumonias, mixed viral and bacterial pneumonias with features of both etiologies have been described. Other respiratory tract complications include bacterial sinusitis, croup, and otitis media.
Reye's syndrome has been described primarily in children < 18 years of age. In almost all cases, Reye's syndrome appears to be a complication resulting from the use of salicylates, most commonly aspirin, to treat certain viral illnesses. Reye's syndrome usually presents several days after an unremarkable viral illness. The syndrome usually presents with nausea and vomiting followed by central nervous system (CNS) changes such as lethargy, delirium, seizures, or coma. Reye's syndrome has been primarily seen in children treated with aspirin for influenza B virus and varicella-zoster virus infections but also for infections by influenza A virus. Since the 1980s, the incidence of Reye's syndrome has decreased dramatically in the United States after warnings were issued regarding the link between the treatment of children with aspirin and this syndrome.
Myocarditis and pericarditis were reported in association with influenza during the 1918–1919 pandemic but have been documented infrequently since then. Minor electrocardiogram changes in the setting of influenza can often be seen in patients with underlying heart disease. Myositis with rhabdomyolysis and myoglobinuria has been reported but is uncommon. Toxic shock syndrome associated with secondary staphylococcal infection after acute influenza also has been reported. In addition, a number of CNS complications including encephalopathy, encephalitis, transverse myelitis, and Guillain-Barré syndrome have all been reported, but their association with influenza remains unclear.
Uncomplicated cases of influenza are usually treated symptomatically. Salicylates should be avoided in children < 18 years of age because of the risk of Reye's syndrome. There are two classes of licensed antiviral agents, the adamantines and the neuraminadase inhibitors, with specific activity against influenza viruses. The adamantines, amantadine hydrochloride and rimantadine hydrochloride, have specific activity against influenza A virus but not B virus. The neuraminadase inhibitor drugs, zanamivir and oseltamivir, have activity against both influenza A and B viruses. Amantadine is licensed for use in children and adults, whereas rimantadine is licensed for use in adults. Both drugs can be given for treatment of or prophylaxis against influenza A virus but are ineffective against influenza B virus. These chemically related drugs are equally effective and appear to inhibit viral replication by blocking the ion channel function of the viral M2 protein. Viral resistance to both compounds is associated with changes in the M2 protein. Although the rapid emergence of resistant viruses has been demonstrated both in vitro and in vivo, the risk of transmission of these resistant viruses remains unclear. Resistant viruses have been most often isolated from individuals receiving treatment and less often from contacts.
When administered prophylactically, both agents are ~70–90% effective in preventing illness caused by influenza A viruses. Subclinical infections still can occur while taking these drugs. Although chemoprophylaxis can be used alone in persons for whom vaccination is contraindicated, it is preferable to administer these agents as an adjunct to vaccination in high-risk groups. Since these drugs do not interfere with antibody response to vaccination, they can be used to provide prophylaxis to persons who were vaccinated but who have not yet had adequate time (usually 2 weeks) to mount a vaccine antibody response. To control an institutional influenza outbreak, antiviral agents are most effective when administered to all residents.
When administered within 48 h of illness onset, these agents have also been shown to reduce the severity and duration of influenza in young and healthy adults and children. Similar controlled studies have not been conducted among persons at high risk for complications.
The major pharmacological differences between these agents are their pharmacokinetic and side-effect profiles. More than 90% of amantadine is excreted unchanged by the kidneys, whereas ~75% of rimantadine is metabolized by the liver (however, unmetabolized rimantadine and its metabolites are renally excreted).
Both medications can lead to CNS and gastrointestinal side effects. In one study, the incidence of CNS side effects in young and healthy adults using 200 mg/d was higher among those taking amantadine (14%) than rimantadine (6%) or placebo (4%). CNS side effects include nervousness, anxiety, difficulty concentrating, and light-headedness. More serious side effects (eg, marked behavioral changes, delirium, hallucinations, agitation, and seizures) have been associated with high plasma drug concentrations, particularly among persons with renal insufficiency, seizure disorders, or certain psychiatric disorders, or among elderly persons who were taking amantadine at doses of 200 mg/d. Approximately 3% of those taking either drug develop gastrointestinal side effects, such as nausea and anorexia. Side effects cease soon after stopping the drug, and lower doses appear to be associated with a lower incidence of side effects.
The usual therapeutic and prophylactic dosage of amantadine and rimantadine in adults < 65 years is 100 mg orally twice a day (Box 29-2). Doses should be reduced in children younger than 10 years, adults 65 years and older, especially those in nursing homes, and persons with renal insufficiency or severe hepatic dysfunction (for rimantadine).
The neuraminadase inhibitors were approved in 1999 for treatment of uncomplicated influenza. Zanamivir is orally inhaled and was approved for treatment of persons > 7 years. Oseltamivir is orally administered and was approved for treatment of persons > 1 year. Similar to the adamantines, both neuraminadase inhibitors can be effective in reducing illness by approximately one day when used within 2 days of illness. Although only oseltamivir has been approved for chemoprophylaxis, recent community studies suggest that both zanamivir and oseltamivir are approximately 80% effective in reducing febrile influenza illness when administered as chemopropylaxis. Oseltamivir was approved in 2000 for chemoprophylaxis of persons > 13 years.
In placebo controlled studies of persons with uncomplicated influenza, persons receiving zanamivir or placebo reported similar rates of adverse events including diarrhea, nausea, sinusitis, nasal signs and symptoms, bronchitis, cough, headache, dizziness, and ear, nose throat infections. In patients with asthma or chronic obstructive pulmonary disease, more patients taking zanamivir than placebo had a > 20% decrease in forced expiratory volume in 1 second (FEV1) or peak expiratory flow. In addition, persons with underlying asthma or chronic obstructive pulmonary disease have been reported to experience respiratory deterioration following use of zanamivir. Caution should be exercised when prescribing zanamivir to patients with asthma or chronic obstructive pulmonary disease. Such patients should have a fast acting bronchodilator available when inhaling zanamivir. Oseltamivir has been associated with higher levels of nausea or vomiting (approximately 9-10%) than in persons taking placebo.
Resistance to the neuraminadase inhibitors can be induced in influenza A and B viruses in-vitro but there is little information to indicate the clinical significance of these findings. Available information suggests that resistance to these compounds develops less frequently than with the adamantines.
The recommended dosage of zanamivir is two inhalations (a total of 10 mg) twice daily about 12 hours apart for five days. The manufacturer does not recommend changes in dosage based on age or renal function. Inhaled zanamivir has a half-life of about 2.5–5.1 hours and is excreted unchanged in urine. The recommended dosage of oseltamivir for treatment in persons > 13 years or for younger children who weigh > 40 is 75 mg twice daily. In children > 13 years and who weigh > 40 kg the recommended dosages vary by weight: 30 mg twice daily for children > 15 kg; 45 mg twice daily for children > 15 kg to 23 kg; and 60 mg twice daily for children > 23 to 40 kg. Oseltamivir is approved for chemoprophylaxis in children > 13 years and the dosage is 75 mg once a day. There is no recommended change in dosage for elderly persons. However, in patients with renal dysfunction and a creatinine clearance between 10 and > 30, the recommended treatment dosage is 75 mg once a day and the recommended chemoprophylaxis dosage is 75 mg every other day. No recommendations are available for patients undergoing renal dialysis.
Ribavirin has been reported to have efficacy against influenza A and B infections when administered as an aerosol for treatment.
In most cases, illness from influenza resolves within a week, but cough and malaise may persist for several days to a few weeks longer. In a minority of patients, fatigue may persist for months.
Prevention & Control
Annual administration of influenza vaccine is the most effective approach for preventing illness caused by influenza. In the United States, the currently licensed vaccine is an inactivated vaccine (either killed whole virus or subunit preparations) that contains three contemporary circulating strains of influenza A (H1N1), influenza A (H3N2), and influenza B virus. Because influenza viruses exhibit ongoing antigenic
changes, one or two of the vaccine viruses are updated almost every year.
BOX 29-2 Treatment and Prophylaxis for Influenza.*
The recommended timing for influenza vaccination is from September through mid-November. However, influenza activity frequently peaks after December and unvaccinated persons who are at high risk for complications should continue to be offered vaccine after November.
Live attenuated influenza virus (LAIV) vaccines have been under development since the 1960s. LAIVs have several potential advantages over inactivated influenza vaccine, including greater induction of mucosal IgA and intranasal administration as a spray or nose drops. Currently, studies on trivalent formulations of LAIV vaccines are under way.
The effectiveness of inactivated influenza vaccine depends in large part on the degree of match between circulating viral strains and vaccine strains as well as the age and health status of the recipient. In controlled trials among children and young adults, influenza vaccines are ~70–90% effective in reducing influenza when there is a good match between vaccine and circulating viruses. A meta-analysis of influenza vaccine studies among the elderly found the effectiveness of vaccines to be 56% in preventing illness, 50% in reducing hospitalization, and 68% for preventing death.
In elderly nursing home residents, vaccine effectiveness is ~30% for preventing illness, but ~47–95% for reducing hospitalization, pneumonia, and death.
Each year, comprehensive recommendations on the use of influenza vaccines are published by the Centers for Disease Control and Prevention in an April or May issue of Morbidity and Mortality Weekly Report (Table 29-1, adapted from Centers for Disease Control and Prevention, 2000). In general, influenza vaccination is recommended for the groups listed in Table 29-1.
Information on vaccinating people with human immunodeficiency virus (HIV) infection against influenza is limited. The main issues in question are whether persons with HIV are at elevated risk of influenza or serious complications from influenza, whether immunization poses a risk of accelerating HIV replication, and whether immunization is protective. Recent studies suggest that persons with HIV are at high-risk for developing complications from influenza. Because vaccine may result in protective antibody levels, it is felt that influenza vaccination will benefit many HIV-infected patients.
Table 29-1. Target population for influenza vaccination.*
Arruda E, Hayden FG: Update on therapy of influenza and rhinovirus infections. In Mill J et al: Antiviral Chemotherapy, Elsevier, 1996;1/75.
Betts RF: Influenza Virus. In Mandell: Principles and Practices of Infectious Diseases, 1995;1546.
Breese-Hall C, Dolin R, Gala CL, et al: Children with Influenza A infection: treatment with rimantidine. Pediatrics 1987;80:275.
Centers for Disease Control and Prevention: Prevention and control of influenza: recommendations of the Advisory Committee on Immunization Practices (ACIP). Morbid Mortal Weekly Rep 2000;49(RR-3).
Centers for Disease Control and Prevention: Update: influenza activity—United States and worldwide, 1996–1997 season, and composition of the 1997–98 influenza vaccine. Morbid Mortal Weekly Rep 1997; 46:15.
Cox NJ, Kawaoka Y: Orthomyxoviruses: influenza. In Collier L et al: Topley & Wilson's Microbiology & Microbial Infections, 9th ed. Virology 1998;1:385.
Douglas RG. Prophylaxis and treatment of influenza. N Engl J Med 1990;322:443.
Govaert TM, Thijs CT, Masurel N, Sprenger MJ, Dinant GJ, Knottnerus JA: The efficacy of influenza vaccination in elderly individuals: a randomized double-blind placebo-controlled trial. J Am Med Assoc 1994;272:1661.
Gross PA, Hermogenes AW, Sacks HS, Lau J, Levandowski RA. The efficacy of influenza vaccine in elderly persons: a meta-analysis and review of the literature. Ann Intern Med 1995;123:518.
Hayden FG, Belshe RB, Clover RD, Hay AJ, Oakes MG, Soo W: Emergence and apparent transmission of rimantidine-resistant influenza A virus in families. N Engl J Med 1989;321:1696.
Hayden FG, Treanor JJ, Betts RF, Lobo M, Esinhart JD, Hussey EK: Safety and efficacy of the neuraminadase inhibitor GG167 in experimental human influenza. J Am Med Assoc 1996;275:295.
Hayden FG et al: Efficacy and safety of the neuraminadase inhibitor Zanamivir in the treatment of influenzavirus infections. New Engl J Med 1997;337:874.
Kilbourne ED: Influenza. Plenum, 1987.
Kilbourne ED, Arden NH: Inactivated influenza vaccines. In Plotkin, M, Vaccines, 3rd ed. Saunders (In press).
Lasky T et al: The Guillain-Barré syndrome and the 1992–1993 and 1993–1994 influenza vaccines. N Engl J Med 1998;339:1797.
Murphy BR, Webster RG: Orthomyxoviruses. In Fields BN et al: Virology, 2nd ed. Raven, 1990;1091.
Neuzil KM et al: Influenza-associated morbidity and mortality in young and middle-aged women. JAMA 1999;281:901.
Nichol KL, Margolis KL, Wuorenma J, Von Sternberg T: The efficacy and cost effectiveness of vaccination against influenza among elderly persons living in the community. N Engl J Med 1994;331:778.
Subbarao K et al: Characterization of an avian influenza A (H5N1) virus isolated from a child with fatal respiratory illness. Science 1998;279:393.
Waner JL, Todd SJ, Shalaby H, Murphy P, Wall LV. Comparison of Directogen Flu-A with viral isolation and direct immunofluorescence for the rapid detection and identification of influenza A virus. J Clin Microbiol 1991;29:479.
Yuen KY et al: Lancet 1998;351:467.