Comparison of laparoscopic to open pancreaticoduodenectomy in elderly patients with pancreatic adenocarcinoma

The purpose of the study is to compare perioperative and survival outcomes in elderly patients undergoing laparoscopic pancreaticoduodenectomy (LPD) to those undergoing open pancreaticoduodenectomy (OPD). Patients aged 75 years with pancreatic adenocarcinoma undergoing LPD or OPD were identified from the NCDB (2010-2013). Baseline characteristics and perioperative outcomes were compared using a chi (2) and Student's t test. The Kaplan-Meier method was used to generate survival curves, and differences were tested using a log-rank test. A multivariate cox proportional hazard model was applied to estimate the hazard ratio (HR) of LPD on overall survival (OS). We identified 1768 patients aged 75 years who underwent LPD (n = 248, 14.0%) or OPD (n = 1520, 86.0%). The majority of patients in the LPD group had their surgery at facilities performing less than 5 LPDs per year (n = 165, 66.5%). 90-day mortality was significantly lower in the LPD compared to the OPD (7.2 vs. 12.2%, p = 0.049). The laparoscopic conversion rate was 30% (n = 74) and was associated with higher readmission rates (13.5 vs. 8.1%), 30-day mortality (8.0 vs. 3.8%), and 90-day mortality (10.4 vs. 6.0%), but these did not reach statistical significance. Median OS was significantly longer in the LPD group (19.8 vs. 15.6 months, p = 0.022). After adjusting for patient and tumor-related characteristics, there was a trend towards improved survival in the LPD group (HR 0.85, 95% CI 0.69-1.03). The vast majority of the NCDB participating facilities perform less than 5 LPD cases per year, which was associated with an increased risk of perioperative mortality. Overall 90-day mortality was significantly lower in the LPD group and there was a trend towards improved OS in the LPD group compared to the OPD group after adjusting for patient and tumor-related characteristics. Studies with increased sample size and longer follow-up are needed before definitive conclusions can be made.