Establishment of predictive models for acute complicated appendicitis during pregnancy—A retrospective case–control study
Ping Li, Zhuo Zhang, Shanshan Weng, Hu Nie
Abstract
Abstract Objective To develop a scoring system based on clinical and imaging features to distinguish complicated appendicitis (CA) from uncomplicated appendicitis (UCA) during pregnancy. Method This was a retrospective case–control study. Patients diagnosed with acute appendicitis during pregnancy were included, and they were divided into a CA group and a UCA group based on the intraoperative findings and the biopsy results. Multivariate logistic regression and machine learning were employed to establish a predictive model. Results A total of 342 patients were included in this study. Among them, 141 (41.23%) patients were diagnosed with CA. The predictive model contained six indices, including symptom duration time more than 24 h, fever, heart rate at least 98 beats/minute, monocyte count at least 0.72 × 10 9 /L, lymphocyte count at least 1 × 10 9 /L and direct bilirubin at least 4.75 μmol/L. The total score was 31 points, and a score of more than 15.5 points predicted the development of CA during pregnancy with area under the curve (AUC) of 0.80 (95% confidence interval 0.75–0.84) and specificity of 0.84. A decision flow chart for distinguishing CA from UCA during pregnancy was developed by Decision Tree with an AUC of 0.78. Conclusion The models combining clinical findings and laboratory tests, developed by two methods, can distinguish CA from UCA in pregnancy in a convenient and visualized way. Trial Registration The research has been registered in Chinese Clinical Trial Registry on January 7, 2022 with registration ID ChiCTR2200055339.