The Impact of Surgeon Volume on Outcomes After Pancreaticoduodenectomy: a Meta-analysis

Despite significant improvement in operative mortality rates following pancreaticoduodenectomy (PD), morbidity remains high. Outcomes following PD may be improved in high-volume centers and with high-volume surgeons. We sought to evaluate the association between surgeon experience and postoperative outcomes after PD.

An online database search of MEDLINE and EMBASE was performed; key bibliographies were reviewed. Studies comparing operative outcomes of high-volume surgeon (HVS) and low-volume surgeon (LVS) performing PD were included. Odds ratios with the corresponding 95% confidence intervals (CI) by random fixed effects models of pooled data were calculated. Definition of HVS varied among the studies, ranging from 6 to >20 PD/year. The primary endpoint was 30-day mortality, and secondary outcomes were complication rates, length of stay (LOS), hospital costs, and readmission rates. Study quality was assessed using STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) criteria.

Search strategy yielded 360 publications. Eleven studies met the inclusion criteria comprising 36,449 patients. Among these patients, 12,512 (34.3%) PDs were performed by HVS and 23,937 (65.7%) by LVS. Meta-analysis of included studies showed that HVS had significantly lower mortality rates than LVS (2.4 vs. 6.7%, OR 2.88; 95% CI 2.51-3.27, p < 0.001). They also had significantly lower overall complication rates (36.3 vs. 50.3%, OR 1.71; 95% CI 1.62-1.81, p < 0.001), hospital costs (range $10,818-141,322 vs. $12,114-198,678, OR 0.13; 95% CI 0.07-0.19, p < 0.001), and LOS (range 11-35 vs. 14-38 days, OR 2.86; 95% CI 2.03-3.68, p < 0.001).

HVS performing PD have significantly better outcomes than LVS in terms of decreased mortality, morbidity, LOS, and hospital costs. Efforts toward increased regionalization of care should be discussed. Consensus regarding definition of HVS needs to be undertaken.