Does an intra-operative safety checklist reduce postoperative morbidity and mortality in high risk surgical patients?

Abstract

BACKGROUND. Structured communication using a checklist between health care providers may improve performance: a simple checklist used during surgery was associated with a reduction in death rates and complications. The aim of this study was to evaluate if a similar checklist reduces postoperative mortality and morbidity in high risk surgical patients. METHODS. Design: pre-implementation (period I) and post-implementation (period II) cohort study; duration: 2x3 months. Inclusion:16 years, ASA>2. Exclusion: low risk surgery, obstetrical, gynaecologic surgery, vital interventions. Primary outcomes were unplanned returns to operating room (OR), unplanned admissions to intensive care unit (ICU) or death within 30 days as well as a composite endpoint including all three events. Changes in outcome variables between the two periods were evaluated by calculating risk differences and their 95% confidence intervals (CI) for categorical variables as well as mean differences and their 95% CIs for continuous variables. Furthermore, we used uni- and multivariate logistic regression to obtain odds ratios (OR). RESULTS. Comparing period I to period II, 609 and 553 patients were included, respectively. Mean age was 70.1 15.5 vs. 69.1 15.9 years. The ASA physical status 3/4/5 distribution was 85.2%/14.0%/0.8 vs. 84.1%/15.4%/0.5%. In period I 289 patients (47%) were operated in emergency vs. 293 (53%) in period II. Postoperative admission to ICU was planned for 154 patients (25.3%) vs. 156 patients (28.2%). Unplanned return to OR was observed in 79 patients (13.0%) vs. 61 (11.0%); unplanned return to OR related to surgical site infection was found in 29 patients (4.8%) vs. 17 patients (3.1%). Unplanned return to ICU was observed in 32 (5.3%) vs. 31 patients (5.8%). The number of in-hospital death was 26 (4.3%) vs. 42 (7.6%), risk difference 3.3% (95%CI 0.6;6) and unadjusted OR 1.8 (95% CI 1.1;3.0). After adjustment for age, sex, ASA score and type of surgery the OR was 1.7 (95% CI 1.0;2.9). The number of composite events was 122 (20.0%) in period I versus 107 (19.3%) in period II. CONCLUSION. Implementation of an intra-operative checklist seems not to improve postoperative outcome of high risk surgical patients immediately after introduction. The increase in 30-day mortality in period II may be related to unmeasured confounding factors. Of note, period II was during the summer holidays, period with fewer seniors.