Gastric cancer: predictors of recurrence when lymph-node dissection is inadequate

BACKGROUND The TNM classification (sixth edition) requires at least 15 lymph nodes to be examined to allow an accurate staging. However, in our environment, only 20% of patients have the recommended minimum of 15 nodes removed. PURPOSE To evaluate clinicopathological predictors of recurrence in patients with gastric cancer undergoing radical resection with an inadequate number of lymph nodes examined. METHODS 101 patients were included in this retrospective cohort. We evaluated age, gender, tumoral location, Borrmann type, Lauren histotype, type of gastrectomy, grade, invasion depth of tumor, lymph node involvement, ratio between metastatic and total number of excised lymph nodes keeping 20% as the cutoff value (LNR) and adjuvant treatment. The association between these variables and recurrence was investigated by using univariate methods and multivariate logistic regression analysis. RESULTS Median (range) age was 63 years (44-85). 63% males, 37% females. Median follow-up time for the whole patients population was 36 months (10-104). Median number of lymph nodes retrieved was 6 (0-14). Recurrence: 50 of 101 cases (49,6%); 41 hematogeneus dissemination, 9 locoregional recurrences. The following factors were found to be correlated with the recurrence risk: tumoral location, invasion depth of tumor, lymph node involvement and LNR. A multivariate analysis revealed that depth of invasion [odds ratio (OR) 2.80, 95% confidence interval (CI) 1.03-7.58, P = 0.04] and LNR (OR 2.34, 95% CI 1.05-5.21, P = 0.03) were independent risk factors for recurrences of gastric cancer. Median time to recurrence: 16 months (2-50). 82% of recurrences occurred within the first two years after surgical treatment. The estimated cumulative risk of recurrence at five years: 61% in the whole patients population, with serosal invasion and LNR > and < 20% was 82% and 44%, without serosal invasion 73% and 39% respectively. CONCLUSION Invasion depth of tumor and LNR were independent predictors of recurrence in gastric cancer after potentially curative resection with an inadequate number of lymph nodes examined.