Adverse prognostic impact of splenectomy on survival in gastric carcinoma patients: Regression and propensity score matching analysis of 1074 patients.

Abstract

BACKGROUNDS:Patients with proximal gastric carcinoma undergo total gastrectomy with concomitant splenectomy to ensure the complete removal of splenic hilar lymph nodes. However, the impact of splenectomy on survival remains uncertain. This study aimed to investigate the impact of splenectomy on survival among patients with gastric carcinoma. METHODS:Of 1074 patients who underwent total gastrectomy for proximal gastric carcinoma between 2006 and 2014, 229 patients underwent concomitant splenectomy or pancreaticosplenectomy during surgery. We investigated the prognostic impact of splenectomy using a regression and propensity score matched model. RESULTS:The splenectomy and non-splenectomy groups differed in many baseline characteristics, including tumor stage, and had respective crude 5-year survival rates of 55% and 81% (p <0.001). In a multivariate analysis adjusted for TNM stage and other prognostic factors, splenectomy was an independent poor prognostic factor for overall survival (hazard ratio [HR] = 1.67, 95% confidence interval [CI] = 1.11-2.51) and disease-free survival (HR = 1.61, 95% CI = 1.24-2.10). A survival evaluation stratified by TNM stage showed that splenectomy adversely affected survival among patients with stage III, but not stage I, II, and IV disease. In the propensity score-matched sample, splenectomy group also showed significantly worse overall survival (5-year, 65% vs. 79%, p = 0.010) and disease-free survival (5-year, 55% vs. 72%, p = 0.025) and was an independent poor prognostic factor in a multivariate analysis adjusting TNM stage and other prognostic factors. CONCLUSIONS:Splenectomy adversely affects survival, particularly among patients with stage III gastric carcinoma, and should be avoided unless there is direct invasion to the splenic hilum.