Seventy-eight (24%) episodes of nosocomial pneumonia (NP) were detected in 322 consecutive mechanically ventilated patients admitted to a 1,000-bed teaching hospital from April 1987 through May 1988 to assess the incidence, risk, and prognosis factors of NP acquired during mechanical ventilation (MV). The risk and prognosis factors for developing NP during MV were studied using both univariate and multivariate statistical techniques. Multivariate analysis selected the following variables significantly associated with a higher risk for developing ventilator-associated pneumonia: more than one intubation during MV (p= 0.000012), a prior episode of aspiration of gastric content (p= 0.00018), a MV period longer than 3 days (p= 0.015), the presence of chronic obstructive pulmonary disease (COPD)(p= 0.048), and the use of positive end-expiratory pressure (PEEP) during MV (p= 0.092). The presence of an ultimately or rapidly fatal underlying disease (p= 0.0018), worsening of acute respiratory failure caused by pneumonia (p= 0.0096), the presence of septic shock (p= 0.016), an inappropriate antibiotic treatment (p= 0.02), and the type of intensive care unit (ICU) hospitalization (noncardiac surgery and nonsurgical ICU compared with postcardiac surgery ICU)(p= 0.08) were those factors selected by a stepwise logistic regression analysis as independently worsening the prognosis. The overall fatality rate was 23%(73 of 322). The mortality of patients with NP was higher (33%; 26 of 78; p< 0.01) when compared with fatality rates of patients without NP (19%; 47 of 244). An etiologic agent was isolated in 36 cases (46%). Polymicrobial associations of microorganisms and Acinetobacter calcoaceticus were the most frequent etiologies of NP. Because some of the factors identified as influencing the risk and the prognosis could be modified by medical intervention, a percentage of NP during MV may be prevented and the prognosis improved.