Does Splenic Hilar Lymph Node Harvesting Impact Outcomes After Total Gastrectomy for Cancer? A Multi-Institutional Retrospective Study With Propensity Score-Matching.

Abstract

Benefit of splenic hilar lymph nodes (group n°10 or LN10 according to the Japanese gastric cancer association classification) harvesting during total gastrectomy for cancer is still debated, especially because of both uncertain positive impact on oncologic outcomes and reported increased intra- and postoperative morbidity. Our aim was to investigate the real value of LN10 harvesting in total gastrectomy. Data were analyzed from the ADENOKGAST multi-institutional retrospective database, conducted between 2007 and 2017. A propensity score analysis was performed using the inverse probability of treatment weighting method. Preoperative factors for recurrence were analyzed in univariate and multivariate models to create a risk score (proportional to Odds Ratios), allowing two groups of patients to be distinguished: low and high risk for recurrence. Survivals were studied using Kaplan-Meier methods and Cox models. 514 patients were analyzed. Patients who underwent LN10 harvesting (LN10+) (n=84) or not (LN10-) (n=430) were compared. 30-day severe postoperative morbidity (p=0.7) and 90-day postoperative mortality (p=0.3) were similar. Mean length of hospital stay was slightly increase in LN10+ patients (19.7 vs 20.6 days; p=0.03). Recurrence rates (36% vs 40%; p=0.7) and recurrence location (OR=1.2; p=0.85) were similar, as well as both overall survival (OS) (HR=0.84; IC95=[0.44;1.58]; p=0.6) and recurrence-free survival (RFS) (HR=1.1; IC95=[0.64;1.91]; p=0.7). Among patients with high risk for recurrence, there was no benefit neither on OS nor RFS of LN10 harvesting. LN10 harvesting did not improve neither OS nor RFS following total gastrectomy for gastric cancer, even in patients with high risk for recurrence.