Objective: To determine the prognostic predictive value of preoperative Glasgow prognostic score (GPS) in rectal cancer patients after surgery. Methods: The clinical data of 219 patients with rectal cancer undergoing surgical treatment over the past six and a half years were reviewed. The related factors that affect prognosis were analyzed, and the patients were divided into score-2 group, score-1 group and score-0 group according to their preoperatvie GPS, to analyze the relations of GPS with the clinical features of the patients, and to compare the postoperative survivals among the 3 groups. Results: Univarate analysis showed that the preoperative levels of carcinoembryonic antigen (CEA), C reactive protein (CRP), and albumin, TNM classification and GPS were associated with the overall survival of the patients (all P<0.05); GPS of the patients was relevant to their preoperative levels of CEA, CA19-9, CA724, CRP, and albumin, Dukes’ stage and TNM classification (all P<0.05). Univariate Logistic regression analysis suggested that the preoperative levels of CEA, CA19-9, CA724, CRP, and albumin, differentiation degree of the tumor, Dukes’ stage and TNM classification and GPS were associated with the increased risk of postoperative death (all P<0.05), and further multivariate Logistic regression analysis identified that preoperative albumin level, differentiation degree, TNM classification and GPS were independent risk factors for postoperative death (all P<0.05). The 5-year survival rate of score-2, score-1 and score-0 group was 13.8%, 59.9% and 88.4% respectively, and the difference among them had statistical significance (P<0.001). Conclusion: Preoperative GPS can be used as a prognostic predictor for postoperative survival of rectal cancer patients.