Association of pylorus preservation with outcomes of pancreaticoduodenectomy across the United States.

Background: Pylorus-preserving pancreaticoduodenectomy (PPPD) was developed to improve postoperative gastrointestinal function while maintaining oncologic adequacy. However, conflicting evidence and concerns persist regarding increased delayed gastric emptying, warranting a national-level investigation. Methods: This retrospective cohort study analyzed 40,063 adult pancreaticoduodenectomy procedures from the ACS NSQIP database (2014-2023). Patients were stratified into pylorus-preserving and non-pylorus-preserving groups. Multivariable regression models evaluated independent associations between pylorus-preserving status and clinical outcomes including infectious complications, blood transfusions, delayed gastric emptying, and length of stay. Results: Of 40,063 pancreaticoduodenectomy procedures, 13,882 (34.6 %) were pylorus-preserving. PPPD was associated with lower rates of infectious complications (25.9 vs 27.5 %, P = 0.01) and blood transfusions (13.6 vs 16.4 %, P < 0.001), but higher delayed gastric emptying rates (17.1 vs 16.1 %, P = 0.02) and shorter length of stay (7 vs 8 days, P = 0.01). Readmission rates were similar (17.4 vs 17.9 %, P = 0.16). Following risk adjustment, PPPD remained associated with reduced infectious complications (AOR 0.94, 95 % CI 0.89-0.99) and blood transfusions (AOR 0.88, 95 % CI 0.82-0.95), but increased delayed gastric emptying (AOR 1.12, 95 % CI 1.04-1.21). Conclusion: Despite declining utilization over the study period, PPPD offers significant advantages in reducing infectious complications and blood transfusion requirements while shortening operative times and hospital stay. However, the trade-off of increased delayed gastric emptying requires careful consideration in surgical decision-making. These findings support individualized approach selection based on patient complexity and surgeon expertise to optimize perioperative outcomes.