Radiographic Splenic Artery Involvement Is a Poor Prognostic Factor in Upfront Surgery for Patients with Resectable Pancreatic Body and Tail Cancer.

PURPOSE: The prognostic impact of radiographic splenic vessel involvement in pancreatic cancer remains unclear. We evaluate its oncological significance in resectable pancreatic body/tail cancer. PATIENTS AND METHODS: We retrospectively review 102 cases of resectable pancreatic cancer and 51 of borderline resectable pancreatic cancer (BRPC) who underwent pancreatectomy for pancreatic body/tail cancer. Resectable pancreatic body/tail cancer was classified into one of three categories based on radiographic splenic vessel involvement. RESULTS: Among 102 cases of resectable pancreatic cancer, 37 (36.3%), 35 (34.3%), and 30 cases (29.4%) were classified as no splenic vessel involvement (Rnone), splenic vein involvement (RV), and splenic artery involvement (RA), respectively. Disease-free survival (DFS) among patients with Rnone, RV, RA, and BRPC was 58.5, 18.4, 10.8, and 9.2 months, respectively. Patients with RV and RA had significantly poorer DFS than patients with Rnone (P = 0.010, P < 0.001, respectively). Median survival among Rnone, RV, RA, and BRPC was 80.6, 23.4, 15.1, and 21.3 months, respectively. Patients with RV and RA had significantly poorer survival than patients with Rnone (P = 0.001, P < 0.001, respectively) and had short survival similar to that of those with BRPC. Multivariate Cox proportional hazard analysis detected preoperative CA19-9 ≥ 37 IU/L, radiologic splenic vein involvement, radiologic splenic artery involvement, intraoperative bleeding ≥ 500 ml, transfusion, positive washing cytology, and noncompletion of adjuvant therapy as independent prognostic factors. CONCLUSIONS: Radiographic splenic artery involvement is a poor prognostic factor in resectable pancreatic body/tail cancer and may have a role in stratification of treatment strategy.