Objective：To explore the incidence of pulmonary infection between minimally invasive surgery and traditional thoracotomy for the esophagus cancer，and to provide theoretical basis for reducing the infection rate after esophagus surgery. Methods：Totally 272 patients admitted to our hospital from February of 2012 to April 2014，who were going to receive esophagus surgery，were randomly divided into two groups：traditional thoracotomy group（134 patients） and minimally invasive group（138 patients）. Pulmonary infection of the patients in hospital and at 48 h after discharge was observed. Results：Totally 47 patients were infected；the infection rates were 16.42%（22/134） in traditional thoracotomy group，and 18.11%（25/138） in minimally invasive group，without significant differences between two groups（P<0.05）. Most of bacteria were acinetobacter baumannii and pseudomonas aeruginosa. The infection rate within elder patients（≥60 years old） was much higher than that of younger patients（<60 years old）（P<0.05）.There were statistic differences in infection rate of between operation time longer than 5 h group and operation time shorter than 5 h group（P<0.05）. For the patients in two groups，incidence of three operative incisions was higher than that of two operative incisions（P<0.05）. The mortality of the pulmonary infection was not difference between traditional thoracotomy group and minimally invasive surgery group（P>0.05）. Conclusion：Pulmonary infection is not associated with the minimally invasive surgery and traditional thoracotomy in esophagus cancer. In order to reduce the pulmonary infection，we should improve the management during perioperative period and deal with the potential factors which can induce the infection.