MJA
MJA

Pandemic influenza: clinical issues

Mark Boyd, Kate Clezy, Richard Lindley and Rod Pearce
Med J Aust 2006; 185 (10): S44. || doi: 10.5694/j.1326-5377.2006.tb00706.x
Published online: 20 November 2006
Uncomplicated influenza

Uncomplicated influenza is typically characterised by acute onset of symptoms of an upper respiratory tract infection (eg, dry cough, sore throat) accompanied by a constitutional illness (eg, various combinations of fever, headache, chills, myalgias, anorexia and malaise). The typical incubation period for influenza is 1–4 days, with an average of 2 days.1 In the context of a declared influenza pandemic, the presence of cough and fever may be diagnosed as influenza infection with a reasonable degree of certainty.2

History
Examination and investigation

Abnormal physical findings are sparse in patients with uncomplicated influenza. One hospital series found that only half the patients with proven influenza satisfied the criteria for influenza-like illness (temperature > 37.8°C, cough or sore throat), so a high index of suspicion is required to recognise influenza.11 Examination of the chest is usually unremarkable. The respiratory rate, estimated haemoglobin oxygen saturation (assessed by pulse oximetry) and chest x-ray are usually normal.

The full blood examination in uncomplicated influenza may be normal or consistent with the presence of a viral infection — that is, the blood haemoglobin concentration will likely be normal; the platelet count and total white cell count may be normal, decreased, or raised; and a blood film will probably suggest infection, with a neutrophil “left shift”, “band forms”, and “toxic” changes. In general, the serum urea, creatinine and electrolytes will be normal, or exhibit mild, clinically insignificant abnormalities. Liver function tests may demonstrate a mild hepatitis (raised serum alanine aminotransferase or asparagine aminotransferase concentrations) or non-specific cholestatic changes (eg, raised serum gamma glutamyl transferase and alkaline phosphatase concentrations). Study of C-reactive protein (CRP) in assessment of respiratory viral infections suggests that uncomplicated infection with influenza A or B viruses tends to produce higher CRP levels when compared with other upper respiratory viruses.12,13 However, it is neither sufficiently sensitive nor specific to support its use as a marker of influenza infection.

Immunopathogenesis

The major target of influenza infection is the ciliated epithelial cells in the mucous layer of the respiratory tract, leading to their necrosis, with oedema and infiltration by lymphocytes, plasma cells, histiocytes and neutrophils. In uncomplicated infection, repair starts about 3–5 days after illness, corresponding with the time of defervescence. However, restoration of ciliated cell function and normal mucous production may be delayed for 2 or more weeks after the onset of the illness. In fatal cases of influenza pneumonia, there have been varying degrees of interstitial cellular infiltrate, alveolar oedema and hyalin membrane deposition described. The virus may also infect neutrophils and lymphocytes, resulting in a reduced response to chemotactic stimuli and cellular function in general. This, together with necrosis and desquamation of the ciliated epithelial cells and abnormal mucus secretion, favours the development of secondary bacterial infection, including bronchitis and pneumonia, as well as other complications such as middle ear infections and sinusitis.14,15

The severity of clinical disease during an influenza pandemic is determined by intrinsic properties of the virus and the immunological status of the affected individual. For instance, the “cleavability” of the haemagglutinin (HA) glycoprotein has an association with viral pathogenicity. Anti-HA antibodies are the primary neutralising antibodies, and participate in complement-mediated lysis of infected cells, aggregation of virions, and cell cytotoxicity. Anti-neuraminidase reduces the number of infectious particles released from infected cells, and may reduce disease severity. The replication of influenza in a new host activates an inflammatory cytokine cascade, which leads to the febrile response and symptoms. Lavage specimens of nasal secretions typically contain interleukin 6 (IL-6), tumour necrosis factor (TNF), interferon γ, interleukin 10, monocyte chemotactic protein 1, and macrophage inflammatory proteins. While these cytokines may be associated with decreases in viral titre, very high levels of cytokines (eg, IL-6 and TNF) have been found in patients who manifest complicated disease.14,16,17

Complicated influenza

There are a number of individuals who are at increased risk for complicated influenza infection. They are:

Avian influenza A/H5N1 in birds and humans

In 1997, an epizootic avian influenza A/H5N1 virus of high pathogenicity began to cross the species barrier from birds to humans. This first epidemic occurred in China and Hong Kong, and 18 human infections were described (six deaths), which ceased after a mass cull of the entire Hong Kong chicken population. In mid 2003, the H5N1 virus began to circulate widely in poultry in the South-East Asian region as a result of the commercial flow of poultry stocks between neighbouring countries. Its adaptation to migratory birds in 2005 has allowed widespread dissemination of the virus, which is now present in at least 50 countries in Asia, the Indian subcontinent, Africa and Europe.

Since December 2003, 10 of those 50 countries have reported a total of 256 laboratory-confirmed human cases of H5N1 influenza, of which two were asymptomatic cases detected on contact screening.26 The case fatality rate (of symptomatic cases) was 59%. Although this mortality rate is high, it must be recognised that the extent of subclinical infection or mild illness is not certain — it cannot be assumed that these confirmed cases are representative of all human H5N1 infections. However, recent epidemiological surveys have detected only very low rates of asymptomatic seropositive cases of H5N1 virus among health care contacts of patients with documented H5N1 infection, suggesting a substantial symptomatic infection rate.27

Most cases of confirmed human H5N1 influenza to date have been in previously healthy young children or adults, probably reflecting the age-related behaviours that increase risk of exposure to infected birds (ie, poultry workers in the affected countries often tend to be young women). Median duration from symptom onset to hospitalisation was 4–5 days. The median time from symptom onset to death was 9 days. Case fatalities have been highest in those aged 10–39 years, lowest among those older than 50 years, and intermediate among children younger than 10 years. This age profile differs from the typical age-related case fatality for seasonal influenza, in which the highest mortality is seen among people at the extremes of age. The age distribution is similar to that described for the 1918 “Spanish flu” epidemic, in which fatality rates were higher among young adults.28

The clinical presentation of H5N1 infection has been with fever (typically > 38°C) and an influenza-like illness, with lower respiratory tract symptoms more frequent than upper respiratory tract symptoms. Most patients (> 88%) have had pulmonary infiltrates at time of diagnosis.29 Limited microbiological data gathered to date suggest that this pneumonic process is a primary viral pneumonia. Gastrointestinal manifestations have been a relatively prominent aspect of the presentation; watery diarrhoea has preceded respiratory manifestations by up to 1 week.29 The fatality rate of hospitalised patients since 2003 is 78%.26

Common laboratory findings have included leukopenia (particularly lymphopenia), mild to moderate thrombocytopenia, and mild to moderately elevated serum aminotransferase levels. In Thailand, an increased risk of death was associated with decreased leukocyte, platelet, and, particularly, lymphocyte counts at admission.30

H5N1 infection may be associated with a higher frequency of viral detection and higher viral DNA levels in pharyngeal than in nasal samples.17 Commercial rapid antigen tests have been less sensitive than reverse transcription polymerase chain reaction assays in detecting H5N1 influenza.30

Most hospitalised patients with H5N1 influenza have required ventilatory support within 48 hours of admission, and intensive management for multiorgan failure.30,31 Empirical therapy has generally consisted of broad-spectrum antibiotics and antiviral agents, alone or with corticosteroids; the emergency situation has not allowed a rigorous assessment of their effectiveness. Initiation of antibiotics and antivirals relatively late in the disease course has not resulted in any apparent reduction in mortality, although early initiation of antivirals does appear to be of some benefit.27,30,31 After treatment with oseltamivir, it has not been possible to culture the virus from patients who survived, and reductions in pharyngeal viral load have been described within 72 hours of oseltamivir initiation. However, clinical deterioration and eventual death have occurred despite these observations. Observations like this, as well as the apparent risk of serious disease in the otherwise healthy, have suggested a possible role of the innate immune response in the pathogenesis of H5N1 influenza. Elevations of various cytokines, including IL-6, TNF-α, interferon γ, soluble interleukin 2 receptor, interferon-inducible protein 10, monocyte chemoattractant protein 1, and monokine induced by interferon γ, have been described. In one study, the average levels of plasma interferon α in people who died of H5N1 influenza were about three times that found in healthy controls. Such responses may account, at least in part, for the sepsis syndrome, acute lung injury and multiorgan failure seen in many patients who died.17,27

Online responses are no longer available. Please refer to our instructions for authors page for more information.