BACKROUND: It was reported that the number of positive lymph nodes (PLN) and negative lymph nodes (NLN) resected in lymphadenectomy were both independent prognostic factors in esophageal squamous cell carcinoma (ESCC). However, it is unclear which kind of combination of them would have a stronger impact on the prognosis of the ESCC patients. METHODS: A total of 229 patients with ESCC were retrospectively investigated. The Cox Proportional Hazards Model was used to investigate the relationship between the PLN or NLN and the overall survival of patients. Survival analysis was performed using the Kaplan-Meier method, and the survival difference between sub-groups were explored with log-rank tests. The cut-off number of the ratio of NLN count to PLN count resulted in an improved outcome for patients was investigated by a series of log-rank tests. RESULTSs: The numbers of PLN and NLN were both independent prognostic factors, supported by the results of Cox regression model analysis (hazard ratio [HR], 1.12, P<0.001 and HR, 0.97, P=0.024, respectively).The survival differences among sub-groups from N0 to N3 stage were statistically significant (P<0.001). Analysis using a Cox regression model revealed that the number of NLN inside the thoracic cavity had a stronger impact on the prognosis compared to the number of NLN outside the thoracic cavity (HR,0.959, P=0.042 vs.HR, 0.973, P=0.179). As a combination analysis consisting of the number of PLN and NLN, the log-rank test result demonstrated that patients with the ratio of NLN count inside thoracic cavity to the PLN count less than 8 could achieve a better survival.CONCLUSIONS: The ratio of NLN count inside thoracic cavity to the PLN count was a independent prognostic factor.