Objectives: To investigate the value of dynamic monitoring of serum procalcitonin (PCT) in anti-infective therapy of patients with acute stroke. Methods: This is a case control retrospective study of acute stroke patients conducted from July 2016 to October 2018, in the Department of Neurology, Affiliated Hospital of Hebei University, who who reached within twenty four hours. They, were selected as the study subjects who were divided into infection group and non-infection group according to the inclusion and exclusion criteria. The serum PCT and CRP levels were compared between the two groups at 24 hours, 48 hours and 72 hours. In order to judge the changes of PCT level and the infection of stroke patients, different kinds of antibiotics were used for corresponding treatment. Retrospective analysis of the cases that did not monitor PCT anti infective treatment before July 2016 were compared with the cases that monitored PCT to guide anti infective treatment after July 2016, and compared the efficacy of antibiotics. Results: The serum PCT level of patients in the infection group was significantly higher than that of patients in the noninfection group (P<0.001). For the patients whose PCT<0.5 ng/ml within 72 hour, antiinfective therapy was not administered. However, for those patients whose PCT<0.5 ng/ml and CRP rose significantly, WBC, body temperature and chest CT were closely monitored. For the patients whose PCT increased slightly (0.5 ng/mlPCT>2 ng/ml), mezlocillin/ sulbactam or ceftriaxone/ tazobactam was administered. For patients whose PCT increased significantly (PCT>5 ng/ml), carbapenem antibiotic or a combination of two antibiotics was administered. Conclusion: Dynamic detection of serum PCT concentration can make accurate judgment on the severity of bacterial infection in patients with acute stroke and guide the rational application of antibiotics.