Open Pancreaticoduodenectomy Case Volume Predicts Outcome of Laparoscopic Approach: A Population-based Analysis.

OBJECTIVE: To determine if laparoscopic pancreaticoduodenectomy (LPD) is safe and offers benefits over open pancreaticoduodenectomy (OPD) at institutions with lower pancreaticoduodenectomy (PD) volume.

BACKGROUND: Although a hospital-based case volume-outcome relationship for morbidity, mortality, and oncologic quality has been reported for OPD, comparative trends for LPD have yet to be investigated.

METHODS: A total of 4739 patients with complete data were identified in National Cancer Data Base between 2010 and 2011; 4309 patients had OPD and 430 patients had LPD. Institutions were categorized into quartiles based on PD case volume. For the entire cohort and within each quartile, LPD and OPD were compared for 30-day and 90-day mortality, length of hospital stay, 30-day unplanned readmission rate, and margin status. Binary logistic regression, linear regression, and propensity score matching was performed.

RESULTS: Hospitals with low PD case volume (≤25 PDs per year; 91% of all hospitals in the US and 25% of cases) had the highest 30- and 90-day mortality, highest margin positivity rates, and lowest lymph node counts. These trends were more pronounced in the LPD group. Only in the highest-volume hospitals was LPD associated with shorter hospital stay and lower readmission compared with OPD.

CONCLUSIONS: These findings confirm that risks of postoperative mortality and suboptimal oncologic surgical quality following PD are higher in low-volume hospitals. Furthermore, these risks are more profound with LPD compared with OPD. These data suggest that the putative benefits of LPD are unlikely to be observed in institutions performing ≤25 PDs per year.