Multivisceral pancreaticoduodenectomy with vascular resection: are aggressive surgical approaches justified for pancreatic head cancer?

The aim of this study was to evaluate the oncological justification for multivisceral pancreaticoduodenectomy with resection of the venous confluence (superior mesenteric and/or portal vein) in patients with pancreatic ductal adenocarcinoma (PDAC) of the pancreatic head. In this retrospective observational cohort study, we enrolled 107 consecutive patients with PDAC of the pancreatic head who underwent pancreaticoduodenectomy with SMV/PV resection at two specialized oncological centers between January 2011 and April 2024. We estimated overall survival (OS) and recurrence-free survival (RFS) by the Kaplan–Meier method, assessed prognostic factors by Cox proportional-hazards regression, and built multivariate models by forced entry (enter) of the variables that reached p < 0.10 in univariate analysis, with the number of predictors limited by the events-per-variable rule. Fifteen patients underwent multivisceral surgery with pancreaticoduodenectomy (MVS + PD; concurrent resection of one or more adjacent organs in addition to venous resection) and 92 underwent standard pancreaticoduodenectomy (PD) with venous resection alone. Severe postoperative complications (Clavien–Dindo grade ≥3a) occurred in 26.2% of patients; the 30-day and in-hospital mortality rate reached 7.5%. In the MVS + PD group, there was no postoperative mortality (0%), and immediate results were comparable to standard PD with venous resection in terms of severe complications (20.0% vs. 27.2%; p = 0.755), postoperative pancreatic fistula (POPF; 13.3% vs. 7.6%; p = 0.500), and delayed gastric emptying (DGE; 20.0% vs. 42.4%; p = 0.153). In addition, postoperative hospital stay was shorter in the MVS + PD group (median 17 vs. 20 days; p = 0.037), with no significant differences in operative duration or intraoperative blood loss. The median overall survival (OS) was significantly lower in the MVS + PD group than in the standard PD group (13 versus 31 months; p = 0.02), and the median recurrence-free survival (RFS) was 12 versus 16.5 months ( p = 0.049). According to multivariate analysis, independent adverse prognostic factors of OS were MVS + PD (HR 2.53; 95% CI 1.25–5.13; p = 0.010) and POPF (HR 2.72; 95% CI 1.24–5.97; p = 0.013), whereas adjuvant chemotherapy independently improved OS (HR 0.50; 95% CI 0.28–0.88; p = 0.017). For RFS, independent unfavorable factors were MVS + PD (HR 3.40; 95% CI 1.19–9.75; p = 0.023) and elevated preoperative CA 19 − 9 levels (HR 1.001; 95% CI 1.000–1.001; p = 0.018). MVS + PD with concomitant venous resection for PDAC of the pancreatic head is technically feasible at specialized centers but confers significantly worse long-term oncological outcomes than standard PD with venous resection alone. These findings do not support its routine adoption; clinicians should rigorously individualize patient selection following multidisciplinary tumor-board review, with adjuvant systemic therapy considered mandatory in eligible patients.