Currently, the optimal therapy of primary liver cancer (PLC) remains to be hepatic resection. For better management of the patients, we evaluated the prognostic predicting value of red blood cell (RBC) count, a routine laboratory parameter, on the long-term survival of patients who underwent surgical treatment.Clinical and laboratory data of 758 patients, who underwent surgical hepatic resection, were retrospectively studied by χ2 tests and logistic regression. All patients were enrolled at Henan Cancer Hospital, Zhengzhou, China, from February 2009 to July 2013, and none of them received any other treatments before surgery. Kaplan–Meier survival analysis and Cox proportional hazard models were used to examine the influence of RBC counts on patients’ survival.The Cox univariate and multivariate analyses showed that preoperative RBC count was an independent risk factor of poor prognosis after surgical treatment. The Kaplan–Meier curves showed that the overall survival (OS) of patients without reduced preoperative RBC counts was significantly better than those patients with reduced preoperative RBC counts (P < 0.001). Concordantly, compared with the patients with either reduced preoperative and/or postoperative RBC counts, patients without reduced RBC counts preferred to be low Child–Pugh grades (P = 0.0065), which implies a better hepatic function. In addition, low RBC count was found to be significantly associated with patients of female (P = 0.003), younger age (P = < 0.001), and with higher AST/ALT ratio (P = 0.005).This study revealed that patients with preoperative RBC counts lower than normal had worse OS rates than those without reduced preoperative RBC counts, perhaps due to the significant correlation of reduced preoperative RBC count to patients’ worse Child–Pugh grade that reflect the loss of liver functions.