Recurrence prediction model of perianal abscess after incision and drainage was established based on Logistic regression
10.3760/cma.j.cn115807-20240514-00167
- VernacularTitle:基于Logistic回归建立肛周脓肿切开引流术后复发预测模型
- Author:
Nan LI
1
;
Guodan JIANG
1
;
Huifeng LIU
1
Author Information
1. 首都医科大学附属北京潞河医院普外科，北京 101100
- Publication Type:Journal Article
- Keywords:
Perianal abscess;
Incision and drainage;
Diabetes mellitus;
Recurrence
- From:
Chinese Journal of Endocrine Surgery
2024;18(6):887-891
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To establish a recurrence prediction model for perianal abscess after incision and drainage surgery based on Logistic regression.Methods:A total of 180 patients with perianal abscess admitted to our hospital from Jan. 2018 to Dec. 2023 were selected as the study objects, all of whom were treated with perianal abscess incision and drainage. According to the follow-up results, the subjects were divided into relapse group and non-recurrence group, and the clinical and follow-up data of the two groups were collected. Multivariate Logistic regression analysis was used to screen the risk factors of postoperative recurrence in patients with perianal abscess. A risk prediction nomogram model was constructed based on the selected risk factors, and the constructed model was evaluated.Results:Of the 180 patients, 35 relapsed during follow-up, with a recurrence rate of 19.44% (35/180). The proportion of patients with diabetes mellitus, abscess site (deep), BMI≥25 kg/m 2, history of chronic diarrhea, and surgical method (simple incision and drainage) in relapse group was higher than that in non-recurrence group ( P<0.05). Combined diabetes mellitus, abscess site (deep), BMI≥25 kg/m 2, history of chronic diarrhea, and surgical method (simple incision and drainage) were the main influencing factors for the recurrence of perianal abscess after incision and drainage. The area under ROC curve (AUC), 95%CI, sensitivity and specificity were 0.873, 0.649-0.996, 96.50% and 79.30% ( P<0.001), respectively, in the prediction of recurrence after incision and drainage of perianal abscess by the nomogram risk model. Calibration curve results showed that both predicted and actual predicted values were near the ideal curve, and the Hosmer-Lemeshow goodness of fit curve test χ 2=3.154, P=0.578. Conclusions:Combined diabetes mellitus, abscess site (deep), BMI≥25 kg/m 2, history of chronic diarrhea, and surgical method (simple incision and drainage) were the main influencing factors for the recurrence of perianal abscess after incision and drainage. The risk prediction model based on the above indexes has good differentiation, calibration and clinical practicability.
