[The value of metastatic lymph nodes ratio in predicting the prognosis of patients with T3 gastric carcinoma]

To evaluate the value of the metastatic lymph node ratio (MLR) in predicting the postoperatively survival time of patients with T(3) gastric carcinoma.Eighty-nine patients with T(3) gastric carcinoma who underwent curative gastrectomy were analyzed retrospectively. The correlations between MLR, positive nodes and the total lymph nodes (15 or more) in histologic examination were analyzed using Spearman's correlation analysis. The influence of MLR and positive nodes on survival time of patients was identified with univariate Kaplan-Meier survival analysis and multivariate Cox proportional hazard model analysis. The predicting accuracy of MLR to death of patients within 2 years postoperatively was determined by receiver working characteristic curve and was compared with that of positive nodes.(1) The MLR did not correlate with the total lymph nodes in histologic examination (Spearman's correlation coefficient was -0.0022, P0.05), whereas positive lymph nodes did (correlation coefficient was 0.2504, P0.05). (2) Kaplan-Meier survival analysis identified that the MLR influenced significantly the survival time postoperatively (Log-rank chi(2) = 35.7470, P0.01). Cox proportional hazard model showed the high MLR was an independent poor prognostic factor (chi(2) = 7.9708, P0.01). (3) There was not difference between the area under the receiver working curve of MLR and positive nodes to predict the death of patients within 2 years postoperatively (P0.05).The MLR in T(3) gastric carcinoma is not correlated with the number of total lymph nodes examined on the condition that 15 or more lymph nodes were assessed in pathology. The predicting accuracy of MLR to death of patients with T(3) gastric carcinoma within 2 years postoperatively is same as, but not better than that of positive nodes if the extent of lymphadenectomy is optimal.