Objective: To establish a predictive model for spinal canal labor analgesia-related intrapartum fever and validate its predictive efficacy. Methods: A total of 2 276 parturients who received labor analgesia from January to December 2021 were selected as the training set, aged ≥ 18 years, BMI 18.5-40.0 kg/m2, ASA physical status Ⅰ or Ⅱ. The patients were divided into fever group and non-fever group according to the occurrence of intrapartum fever (body temperature ≥ 38.0 ℃). The independent risk factors of intrapartum fever were screened by multivariate logistic regression, and the predictive model was established. A total of 568 parturients who received labor analgesia in the same hospital from January to March 2022 were selected as the verification. The inclusion criteria were the same as the training set, and the model was externally verified by R language. Results: There were 197 parturients (8.7%) in the training set and 46 parturients (8.1%) in the validation set experienced intrapartum fever. The multivariate logistic regression analysis showed that primiparity, a high neutrophil count, anemia, and a heavier estimated fetal weight were risk factors of intrapartum fever, while a large body surface area and large cervical dilatation degree before labor analgesia were protective factors against intrapartum fever. According to the predictors, the predictive model for spinal canal labor analgesia-related maternal fever was established. The area under the receiver operating characteristic (ROC) curve (AUC) was 0.698 (95% CI 0.660-0.732), the sensitivity and specificity was 83.2% and 47.9%, respectively. Using R language for the external validation, the AUC of the predictive model was 0.703 (95% CI 0.634-0.772), the sensibility and specificity was 65.2% and 71.3%, respectively. The results showed that the prediction model for spinal canal labor analgesia-related intrapartum fever is effective. Conclusion: Primiparity, a high neutrophil count, anemia, and a heavier estimated fetal weight were risk factors of intrapartum fever, while a large body surface area and large cervical dilatation degree before labor analgesia were protective factors. The predictive model established based on these indicators can effectively predict the occurrence of intrapartum fever before labor analgesia. |