Can bacteriological upper airway samples obtained at intensive care unit admission guide empiric antibiotherapy for ventilator-associated pneumonia?

BACKGROUND: : Ventilator-associated pneumonia is associated with an increase in morbidity and mortality. The delay before adequate antibiotherapy is known to influence patients' outcome. We hypothesized that the results of upper airways samples performed at immediately intensive care unit admission could help the clinician to choose the adequate empiric antibiotherapy for a ventilator-associated pneumonia occurring during the first 5 days of intensive care unit admission. OBJECTIVES: : To compare the bacterial content of the upper airways samples to that of the pulmonary plugged specimen performed when ventilator-associated pneumonia was suspected. DESIGN: : Prospective observational study. SETTING: : Twenty beds in a surgical intensive care unit of a teaching hospital. PATIENTS: : All patients between 1996 and 2001, who were ventilated for more than 48 hours and presented a suspicion of ventilator-associated pneumonia, occurring during the first 5 days. INTERVENTIONS: : As compared to the results of pulmonary plugged specimen, upper airways samples performance was tested by determining sensitivity, specificity, and positive likelihood ratios for each microorganism. MEASUREMENTS AND MAIN RESULTS: : Five hundred eighty-eight patients ventilated for more than 48 hours were suspected to suffer from a VAP and benefited from a pulmonary plugged specimen: 136 (48%) patients had a positive pulmonary plugged specimen and received antibiotics. Of these 136, 125 (92%) had had a positive upper airway samples at intensive care unit admission. For all microorganisms, upper airway sample specificity exceeded 85%. For all bacteria except Streptococcus species, the likelihood ratios exceeded 6, threshold considered as significant to rule in the diagnosis. CONCLUSIONS: : In this study, we found high specificities and likelihood ratios for upper airways samples to predict the microorganisms involved in a ventilator-associated pneumonia. These results suggest that upper airways samples might provide an adjunctive assistance in selecting therapy for ventilator-associated pneumonia.