Sixteen studies (3,630 patients, range five to 954) were included in the review. Nine trials were double blind and seven had incomplete or no blinding. Most of the double blind and single blind trials also met other quality criteria. Trials with no blinding tended to be generally poorly reported in terms of quality criteria.
Pulmonary outcomes (16 RCTs): Incidence of pulmonary outcomes was statistically significantly reduced in patients receiving chlorhexidine (RR 0.73, 95% CI 0.58 to 0.92; 16 RCTs; Ι²=43%). Subgroup analyses indicated that reductions remained statistically significantly significant in cardiac patients (RR 0.56, 95% CI 0.41 to 0.77; three RCTs; Ι²=0%) but when stratified by study design the results were significant only in double blind studies. There were no statistically significant differences in incidence of ventilator-associated pneumonia in non-cardiac trials regardless of study design. Results for other analyses were not statistically significant (fully reported in the review).
Mortality (12 RCTs): There were no statistically significant differences in rates of mortality between treatment groups for the overall estimate or any additional analyses.
Other outcomes: There were no statistically significant differences between treatment groups for duration of mechanical ventilation (six RCTs), intensive care unit hospital stay (six RCTs) or hospital length of stay (three RCTs). Results on antibiotic use for individual trials were reported in the review.
There was no evidence of significant publication bias for pulmonary and mortality outcomes.