Introduction: This study addressed whether the positive node-ratio (N-ratio) for patients who underwent curative-intent treatment was an independent prognostic factor of overall survival (OS) for gastric adenocarcinoma (GA).
Methods: Consecutive patients who underwent gastrectomy for GA with at least 15 harvested nodes were evaluated for 5-year OS. The best threshold was determined using the area under an receiver operating characteristic (ROC) curve. Univariate and multivariate models were assessed looking for independent prognostic factors for OS.
Results: From 1994 to 2015, 398 consecutive patients were evaluated. The N-ratio ≥11% had an accuracy of 0.764, the sensitivity of 71.1%, the specificity of 81.7%, positive predictive value (PPV) of 91.7%, and odds ratio (OR) of 11. After multivariate analysis for OS, age ≥70 years (HR 1.44), need for total gastrectomy (HR 1.45), need for extended resection (HR 1.7), and N-ratio ≥11% (HR 3.7) were unfavorable prognostic factors. D2 lymphadenectomy (HR 0.53) was a protective factor. The median OS according to N-ratio was 14 months for N-ratio >11 vs 58 months for N-ratio <11%.
Conclusion: The N-ratio ≥11% was an independent negative prognostic factor for patients who underwent treatment for GA with curative intent. The N-ratio ≥11% presented high specificity, high PPV and high OR for risk of death for 5 years after surgery.
Keywords: gastric cancer; lymphadenectomy; oncologic outcomes; surgery; survival.
© 2019 Wiley Periodicals, Inc.