Objective: Ventilator-associated pneumonia (VAP) is a serious complication of critical illness, conferring increased morbidity and mortality. Many interventions have been studied to reduce the risk of VAP. We systematically reviewed the influence of airway management on VAP in critically ill patients.

Data sources: Studies were identified through searching MEDLINE and EMBASE from 1980 through July 1997 and by searching SCISEARCH, the Cochrane Library, bibliographies of primary and review articles, personal files, and contact with authors of the randomized trials.

Study selection: We selected randomized trials evaluating ventilator circuit and secretion management strategies on the rate of VAP.

Data extraction: Two investigators independently abstracted key data on design features, the population, intervention, and outcome of the studies.

Data synthesis: The frequency of ventilator circuit changes and the type of endotracheal suction system do not appear to influence VAP rates (3 trials, none with significant difference; range of relative risks [RRs], 0.84-0.91). However, lower VAP rates may be associated with avoidance of heated humidifiers and use of heat and moisture exchangers (5 trials, only 1 showing a significant difference; range of RRs, 0.34-0.86), use of oral vs nasal intubation (1 trial; RR, 0.52; 95% confidence interval, 0.24-1.13), subglottic secretion drainage vs standard endotracheal tubes (2 trials, 1 showing a significant difference; range of RRs, 0.46-0.57), and kinetic vs conventional beds (5 trials, only 1 showing a significant difference; range of RRs, 0.35-0.78).

Conclusions: Some ventilator circuit and secretion management strategies may influence VAP rates in critically ill patients. Whether these strategies are adopted in practice depends on several factors such as the magnitude and precision of estimates of benefit and harm, as well as access, availability, and costs.