Unmasking influenza virus infection in patients attended to in the emergency department.

BACKGROUND

Infection by the influenza virus may pass undetected in many adult patients attended to in the emergency department because its diagnosis usually relies on clinical manifestations, which can be distorted by symptoms of a preexisting disease, superposed complications or nontypical manifestations of influenza virus infection (confusing symptoms).

PATIENTS AND METHODS

We performed this observational, prospective study with an antigen detection test by indirect immunofluorescence assay (IFA) to estimate the presence of influenza virus infection in such patients. No confirmatory test was performed to validate a positive or negative IFA result. Then we compared those who were antigen positive to those who were negative and also analyzed those who were positive classified by age, comorbidity and clinical presentation. We also evaluated the use of medical and hospital resources and vaccination status. Posterior pharynx swab specimens from 136 consecutive adult patients, 74 women and 62 men with a mean age of 68.7 +/- 17.9 (range: 18-97) years attended to in the emergency department of a university hospital in Barcelona during the 1999-2000 influenza epidemic were examined. Tested patients presented either a classical influenza syndrome, a deterioration of a previous condition or any abrupt onset of symptoms without an obvious cause.

RESULTS

Influenza A virus antigen was detected in 99 (72.8%) of the 136 patients included in the study. Confusing symptoms were present in 86 patients with laboratory-confirmed influenza and 40 of them lacked influenza syndrome. Prostration, aching and fever out of proportion to catarrhal symptoms (disproportionate prostration) and cough were independent predictors for this diagnosis (OR = 5.14; 95% CI: 1.98-13.35, p = 0.001 and OR = 4.40, 95% CI, 1.65-11.75, p = 0.03, respectively). Among the 99 patients who tested positive, 72 were >or= 65 years of age. This older positive group compared to the 27 also positive < 65 (non-old) had a tendency to show symptoms mediated by cytokines less frequently: malaise was present in 76.4% of the older positive patients vs 92.6% in the non-old positive ones, p = 0.07. The equivalent percentages for muscle ache were: 56.9% vs 77.8%, p = 0.06; for dysthermia: 54.2% vs 70.4%, p = 0.08; for headache: 35.2% vs 66.7%, p = 0.005, and for disproportionate prostration: 47.2% vs 66.7%, p = 0.08. Cough was more frequent in the older positive group: 94.4% vs 77.8%, p = 0.02. Older positive patients were also hospitalized and received antibiotics more frequently than the non-old positive ones: 65.3% vs 40.7%, p = 0.03 and 81.9% vs 63.0%, p = 0.046, respectively. Hospitalization was independently correlated with the presence of complications (OR = 4.5, 95% IC 1.27-15.95, p = 0.02). Patients with the highest comorbidity, evaluated with the Charlson scale, were more inadequately vaccinated than those with moderate or low comorbidity.

CONCLUSION

Influenza virus infection has a great and underestimated impact in the emergency department during influenza epidemics. High frequency of confusing symptoms, which overcome classical influenza syndrome in adult people with comorbidity, may explain this effect. Disproportionate prostration and cough are symptoms that independently predict its diagnosis in the global adult population, whereas in the elderly, fever and cough should arouse this suspicion whether or not they present classic symptoms. In our setting, individuals with high comorbidity are inadequately vaccinated.