Effect of open and closed endotracheal suctioning on cross-transmission with Gram-negative bacteria: a prospective crossover study

Abstract

Objective: Cross-transmission of Gram-negative bacteria increases the likelihood of acquisition of infections and emergence of antibiotic resistance in intensive care units. Respiratory tracts of mechanically ventilated patients are frequently colonized with Gram-negative bacteria and endotracheal suctioning may facilitate cross-transmission. It is unknown whether closed suction systems, as compared with open suction systems, prevent cross-transmission. The objective was to determine whether closed suction systems, as compared with open suction systems, reduce the incidence of cross-transmission of Gram-negative bacteria in intensive care units.

Design: We performed a prospective crossover study in which both systems were tested unitwide in four intensive care units.

Setting: Two intensive care units from a university hospital and two from a teaching hospital participated in the trial between January 2007 and February 2008.

Patients: All patients admitted to the intensive care unit for >24 hrs were included.

Intervention: Closed suction systems and open suction systems were used for all patients requiring mechanical ventilation during 6-month clusters with the order of systems randomized per intensive care unit.

Measurements and main results: Acquisition and cross-transmission rates of selected Gram-negative bacteria were determined through extensive microbiological surveillance and genotyping. Among 1,110 patients (585 with closed suction systems and 525 with open suction systems), acquisition for selected Gram-negative bacteria was 35.5 and 32.5 per 1,000 patient-days at risk during closed suction period and open suction period, respectively (adjusted hazard ratio, 1.14; 95% confidence interval, 0.9-1.4). During closed suction period, adjusted hazard ratios for acquisition were 0.66 (95% confidence interval, 0.45-0.97) for Pseudomonas aeruginosa and 2.03 (95% confidence interval, 1.15-3.57) for Acinetobacter species; acquisition rates of other pathogens did not differ significantly. Adjusted hazard ratios for cross-transmission during closed suction period 0.9 (0.4-1.9) for P. aeruginosa, 6.7 (1.5-30.1) for Acinetobacter, and 0.3 (0.03-2.7) for Enterobacter species. Overall cross-transmission rates were 5.9 (closed suction systems) and 4.7 (open suction systems) per 1,000 patient-days at risk.

Conclusion: Closed suction systems failed to reduce cross-transmission and acquisition rates of the most relevant Gram-negative bacteria in intensive care unit patients.