Incidence, etiology, and outcome of nosocomial pneumonia in ICU patients requiring percutaneous tracheotomy for mechanical ventilation.

OBJECTIVE To determine the epidemiology of pneumonia in patients with tracheotomy receiving short-term mechanical ventilation. DESIGN Observational prospective study. SETTING A 14-bed medical-surgical ICU. SUBJECTS Ninety-nine critically ill acute patients requiring percutaneous dilatational tracheotomy for mechanical ventilation. INTERVENTIONS Tracheal aspirate obtained 48 h before tracheotomy. MEASUREMENTS AND MAIN RESULTS Eighteen patients (18.1%) acquired pneumonia (median of 7 days after tracheotomy). Pseudomonas aeruginosa was the most frequently identified pathogen, found in eight of the episodes (four not documented by prior tracheal colonization), followed by other Gram-negative bacilli. The development of ventilator-associated pneumonia (VAP) was not anticipated by any clinical variable. A positive tracheal aspirate (TA) culture result obtained before tracheotomy was associated with a risk of acquiring pneumonia of 19.7%, whereas sterile TA cultures were associated with a risk of 14.3% (p > 0.20). VAP prolonged ICU stay or the ventilation period for a median of 19 days and 15 days, respectively. Overall mortality was 34.3%, but the presence of VAP did not increase the mortality rate. CONCLUSIONS Percutaneous tracheotomy in patients receiving short-term mechanical ventilation predisposes to pneumonia. Pneumonia was associated with prolonged ventilation and ICU stay, but was not associated with increased mortality. Pseudomonas is a common pathogen after tracheotomy, and this observation should be considered in selecting an antibiotic regimen, because TA obtained prior to the tracheotomy often failed to identify this pathogen.