Objective To investigate the risk factors for postoperative intracranial infection after suboccipital retrosigmoid surgery， and to construct a risk prediction model.Methods The clinical data of 258 patients undergoing surgeries via the suboccipital retrosigmoid approach in the Affiliated Hospital of Xuzhou Medical University from December 2018 to December 2020 were collected and randomly divided into modeling group （180 patients） and validation group （78 patients） in a 7：3 ratio， with a random seed of 20210528. Univariate and multivariate logistic analyses were used to screening the risk factors for postoperative intracranial infection after surgeries via the approach， and the risk factors were assigned according to the partial regression coefficient （β value） to construct the infection risk prediction model. The internal validation of the model was performed using the data of the modeling group， and the risk scores of patients were evaluated. The external validation was performed using the data of the validation group. The area under the receiver operating characteristic curve （AUC） and Hosmer-Lemeshow （H-L） test were used to evaluate the discrimination and calibration of the model.Results The multivariate analysis showed that 2 points of postoperative modified Glasgow prognostic score， no hydrogen peroxide irrigation before dural incision， the surgery method of endoscopy combined with microscopy， venous sinus rupture， and an operation time of ≥3.5 h were risk factors for postoperative intracranial infection after surgeries via the approach， with the corresponding scores in the model of 6， 6， 6， 5 and 4 points. A score of 20-27 points indicated high-risk patients. The AUC was 0.896 in the modeling group （95% confidence interval ［CI］： 0.840-0.952， P <0.001） and 0.896 in the validation group （95% CI： 0.782-0.999， P <0.001）. The results of H-L tests in the two groups were P >0.05， indicating that the model had good discrimination and calibration.Conclusion The risk prediction model for postoperative intracranial infection after suboccipital retrosigmoid surgery constructed in this study has a good prediction performance and can be used to screen for high-risk groups of postoperative intracranial infection after surgeries via the approach.