Multivisceral and extended resections during pancreatoduodenectomy increase morbidity and mortality

Background. Improvements in outcomes after pancreatoduodenectomy (PD) have permitted more complex resections. Complete extirpation at PD may require multivisceral resection (MVR-PD); however, descriptions of morbidity of 1VIVR-PD are limited to small, single-institution series. Methods. The National Surgical Quality Improvement Project database (2005-2011) was used to compare 30-day postoperative morbidity of PD with 1VIVR-PD. Concurrent resection of colon, small bowel, stomach, kidney, or adrenal gland defined MVR-PD. Results. Of 9,927 PDs, 1VIVR-PD was performed in 273 patients (3%). MVR included colon (58%), small bowel (:30%), and gastric (12%) resections. Preoperative comorbidities were similar between groups. Pancreatic, duodenal, or periampullary cancer was present in 75% of patients. Mortality (8.8% vs 2.9%) and major morbidity (56.8% vs 30.8%) were much greater for MVR-PD versus PD alone (P < .001). MVR-PD patients also experienced greater rates of wound, pulmonary, cardiac, thromboembolic, renal, and septic complications. On multivariable regression, MVR was an independent predictor of death (odds ratio [OM, 3.4; P < .001), overall morbidity (OR, 3.01; P < .001), major morbidity (OR, 3.21; P < .001), and minor morbidity (OR, 1.65; P = .03). Among patients undergoing PD-FMVI?, colectomy was an independent predictor of increased overall morbidity (OR, 1.96; P = .03) and major morbidity (OR, 1.90; P = .02). Conclusion. Margin-negative resection may require MVRs at the time of PD. MVR at is associated with 3-fold mortality and substantial morbidity after adjusting for comorbidities. Colectomy independently predicted major morbidity. At PD, the morbidity of MVR should be approached with caution when attempting margin-negative resection.