K. L. Haines1, C. D. Goldman1, J. Franko1 1Mercy Medical Center,Surgery,Des Moines, IA, USA
Introduction: Recent decades have shown a substantial reduction in mortality and morbidity of Whipple procedures. Perioperative outcomes are thought to be volume-and surgeon-related. Therefore, using the NSQIP dataset, we examined effects of resident/trainee participation in Whipple procedures on mortality and morbidity.
Methods: The NSQIP database was queried for patients undergoing Whipple from 2008-2012. Data was categorized by resident participation (operation with resident versus attending only). Proportion of complications in each group was compared using chi-square test. Logistic regression model was built to analyze impact of multiple covariates on perioperative mortality within 30-days and first postoperative day. Data are presented as proportions and mean±SD.
Results: 12,104 Whipples were reported to NSQIP between 2008-2012. After excluding cases with missing data on resident participation, 7,605 were available for analysis. These cases were divided into two groups: attending alone (group A, n=1105) versus attending with resident (group B, n=6500). Patient comorbidities and demographics were the same between the groups. Operative time was increased with resident involvement (364±139 vs. 380±132 minutes, p<0.001). Length of stay was significantly decreased when residents were involved (13.7±15 vs. 12.7±12 days, p<0.0001). When comparing attending alone versus with resident, there was no significant difference in patient outcomes including incidence of renal failure (p<0.0001), septic shock (p<0.0001), reintubation (p<0.0001), death on operative day (p<0.0001), 30-day mortality (p<0.0001), and readmission rates (p<0.0001). Resident participation was not a significant 30-day mortality predictor in unadjusted (OR=0.73; CI 0.51-1.04, p=0.087) or risk-adjusted models (adjusted OR 0.79; CI 0.55-1.14, p=0.22).
Conclusion: Resident involvement did not influence mortality and morbidity rates of Whipple procedure in this study. Resident participation is associated with longer operative time, but also significantly shortened hospital stay. Reduction in length of stay for this complex operation may reflect that, contrary to oft-reported criticisms in the literature, present-day residents are, in fact, taking ‘ownership’ of the patients under their care.