RESUMO
INTRODUCTION: Positive pathologic margins following gastric cancer (GC) resection carries a poor prognosis. We evaluated intraoperative frozen section (IFS) analysis of resection margins (RMs) as a quality indicator in GC surgery. METHODS: Patients referred to a provincial cancer agency with surgically resected non-metastatic GC between 2004 and 2012 were included. Associations between IFS analysis, other baseline characteristics, RMs, and overall survival (OS) were assessed using logistic regression, Kaplan-Meier analyses, and Cox proportional hazards modeling. RESULTS: Among 377 patients, median age was 67 years, 68% were male, and 16% had +RMs. Thirty-four percent of patients underwent IFS analysis, which protected against +RMs (odds ratio [OR]: 0.34, 95% confidence interval [CI]: 0.16-0.73, p = 0.006) and improved OS (hazards ratio [HR]: 0.72, 95% CI: 0.54-0.98, p = 0.037). OS following re-resection of IFS positive patients was similar to IFS negative patients (69 vs. 54 months, p = 0.317). Stage III disease (OR: 12.8, 95% CI: 3.00-55.0, p = 0.001) and gastroesophageal junction tumors (OR: 2.25, 95% CI: 1.05-4.78, p = 0.036) predicted +RMs. Stage III disease led to worse OS (HR: 2.89, 95% CI: 1.92-4.34, p < 0.001) while intestinal histology improved OS (HR: 0.67, 95% CI: 0.50-0.90, p = 0.007). CONCLUSIONS: IFS analysis reduce +RMs and improve OS and should be incorporated in curative intent GC surgery for patients with locally advanced GC.