Multiple Reservoirs Contribute to Intraoperative Bacterial Transmission

BACKGROUND: Intraoperative stopcock contamination is a frequent event associated with increased patient mortality. In the current study we examined the relative contributions of anesthesia provider hands, the patient, and the patient environment to stopcock contamination. Our secondary aims were to identify risk factors for stopcock contamination and to examine the prior association of stopcock contamination with 30-day postoperative infection and mortality. Additional microbiological analyses were completed to determine the prevalence of bacterial pathogens within intraoperative bacterial reservoirs. Pulsed-field gel electrophoresis was used to assess the contribution of reservoir bacterial pathogens to 30-day postoperative infections. METHODS: In a multicenter study, stopcock transmission events were observed in 274 operating rooms, with the first and second cases of the day in each operating room studied in series to identify within- and between-case transmission events. Reservoir bacterial cultures were obtained and compared with stopcock set isolates to determine the origin of stopcock contamination. Between-case transmission was defined by the isolation of 1 or more bacterial isolates from the stopcock set of a subsequent case (case 2) that were identical to reservoir isolates from the preceding case (case 1). Within-case transmission was defined by the isolation of 1 or more bacterial isolates from a stopcock set that were identical to bacterial reservoirs from the same case. Bacterial pathogens within these reservoirs were identified, and their potential contribution to postoperative infections was evaluated. All patients were followed for 30 days postoperatively for the development of infection and all-cause mortality. RESULTS: Stopcock contamination was detected in 23% (126 out of 548) of cases with 14 between-case and 30 within-case transmission events confirmed. All 3 reservoirs contributed to between-case (64% environment, 14% patient, and 21% provider) and within-case (47% environment, 23% patient, and 30% provider) stopcock transmission. The environment was a more likely source of stopcock contamination than provider hands (relative risk [RR] 1.91, confidence interval [CI] 1.09 to 3.35, P = 0.029) or patients (RR 2.56, CI 1.34 to 4.89, P = 0.002). Hospital site (odds ratio [OR] 5.09, CI 2.02 to 12.86, P = 0.001) and case 2 (OR 6.82, CI 4.03 to 11.5, P < 0.001) were significant predictors of stopcock contamination. Stopcock contamination was associated with increased mortality (OR 58.5, CI 2.32 to 1477, P = 0.014). Intraoperative bacterial contamination of patients and provider hands was linked to 30-day postoperative infections. CONCLUSIONS: Bacterial contamination of patients, provider hands, and the environment contributes to stopcock transmission events, but the surrounding patient environment is the most likely source. Stopcock contamination is associated with increased patient mortality. Patient and provider bacterial reservoirs contribute to 30-day postoperative infections. Multimodal programs designed to target each of these reservoirs in parallel should be studied intensely as a comprehensive approach to reducing intraoperative bacterial transmission.