The number of axillary lymph nodes involved and retrieved are important prognostic factors in breast cancer. The purpose of our study was to investigate whether the lymph node ratio (LNR) is a better prognostic factor in predicting disease-free survival (DFS) for breast cancer patients as compared with pN staging. The analysis was based on 804 breast cancer patients who had underwent axillary lymph node dissection between 1999 and 2008 in Sun Yat-Sen University Cancer Center. Optimal cutoff points of LNR were calculated using X-tile software and validated by bootstrapping. Patients were then divided into three groups (low-, intermediate-, and high-risk) according to the cutoff points. Predicting risk factors for relapse were performed according to Cox proportional hazards analysis. DFS was estimated using the Kaplan-Meier method and compared by the log-rank test. The 5-year DFS rate decreased significantly with increasing LNRs and pN. Univariate analysis found that the pT, pN, LNR, molecule type, HER2, pTNM stage and radiotherapy well classified patients with significantly different prognosis. By multivariate analysis, only LNR classification was retained as an independent prognostic factor. Furthermore, there was a significant prognostic difference among different LNR categories for pN2 category, but no apparent prognostic difference was seen between different pN categories in any LNR category. Therefore, LNR rather than pN staging is preferable in predicting DFS in node positive breast cancer patients, and routine clinical decision-making should take the LNR into consideration.