International Validation of Reduced Major Morbidity After Minimally Invasive Distal Pancreatectomy Compared With Open Pancreatectomy

Ann Surg. 2019 Oct 17. doi: 10.1097/SLA.0000000000003659. Online ahead of print.

Abstract

Objective: To quantify the nationwide impact of minimally invasive distal pancreatectomy (MIDP) on major morbidity as compared with open distal pancreatectomy (ODP).

Background: A recent randomized controlled trial (RCT) demonstrated significant reduction in time to functional recovery after MIDP compared with ODP, but was not powered to assess potential risk reductions in major morbidity.

Methods: International cohort study using the American College of Surgeons' National Quality Improvement Program (ACS-NSQIP) (88 centers; 2014-2016) to evaluate the association between surgical approach (MIDP vs ODP) and 30-day composite major morbidity (CMM; death or severe complications) with external model validation using Dutch Pancreatic Cancer Group data (17 centers; 2005-2016). Multivariable logistic regression assessed the impact of nationwide MIDP rates between 0% and 100% on postoperative CMM at conversion rates between 0% and 25%, using estimated marginal effects. A sensitivity analysis tested the impact at various scenarios and patient populations.

Results: Of 2921 ACS-NSQIP patients, 1562 (53%) underwent MIDP with 18% conversion, and 1359 (47%) underwent ODP. MIDP was independently associated with reduced CMM [odds ratio (OR) 0.50, 95% confidence interval (CI) 0.42-0.60, P < 0.001], confirmed by external model validation (n = 637, P < 0.003). The association between rising MIDP implementation rates and falling postoperative morbidity was linear between 0% (all ODP) and 100% (all MIDP). The absolute risk reduction for CMM was 11% (95% CI 7.3%-15%) at observed conversion rates and improved to 14% (95% CI 11%-18%) as conversion approached 0%. Similar effects were seen across subgroups.

Conclusion: This international study predicted a nationwide 11% risk reduction for CMM after MIDP versus ODP, which is likely to improve as conversion rates decrease. These findings confirm secondary outcomes of the recent LEOPARD RCT.