Objective: To estimate the incidence of influenza infection with a rapid diagnostic test in patients hospitalized in an intensive care unit (ICU) during the epidemic periods of two consecutive winters (2005-2006).
Methods: This study tested nasal aspirate of all patients admitted to the ICU for acute respiratory failure during the influenza epidemic period with the QuickVue influenza rapid diagnostic test.
Results: The study included 39 patients: 22 men and 17 women. Their mean age was 69.7+/-13 years, their mean SAPS II score 54+/-21, their mean length of stay 14.7+/-20 days, and 43.5% had been vaccinated against influenza. The mortality rate was 33.3% (n=13). Four patients (10.2%) had a positive QuickVue test and were considered to have influenza associated with their primary diagnosis: pneumococcal pneumonia, haemophilus pneumonia, acute bronchitis, and acute bronchitis with acute cardiogenic edema. These patients were older than those with a negative test (80+/-7 versus 68+/-13 years, p=0.02). Their SAPS II, length of stay and mortality rate did not differ statistically from those with negative results. Chronic heart failure (but not COPD or non-vaccination) was more frequent in these patients (p=0.01).
Discussion: Several published studies report a satisfactory specificity and sensitivity for QuickVue. With this test, we estimated the incidence of influenza in patients hospitalized in our ICU for acute respiratory failure during flu epidemics at around 10%, close to that reported in previous studies.