探讨急性肛周脓肿术后并发肛瘘的发生情况和危险因素，建立并验证可预测肛瘘早期风险的列线图模型。

选取2023年1月至2025年1月西安交通大学附属西安市中心医院收治的144例急性肛周脓肿患者作为研究对象，其中男性104例，女性40例；年龄21~79岁[（50.77±6.23）岁]。将所有患者按术后3个月是否临床诊断肛瘘分为肛瘘组44例和无肛瘘组100例。收集所有患者临床资料，包括性别、年龄、吸烟史、饮酒史、高血压、便秘、糖尿病、体质指数（BMI）、脓肿大小、脓肿位置、脓肿深度、术前发热史、术前抗生素使用情况、脓肿病史及致病菌类型等，将单因素分析中P<0.05的变量纳入多因素Logistic回归模型，筛选出独立危险因素；再基于多因素Logistic回归分析筛选出的独立危险因素构建列线图预测模型，并完成模型验证。

两组患者年龄、BMI、吸烟史、饮酒史、高血压、脓肿大小、便秘、术前发热史、脓肿位置及术前抗生素使用比较差异无统计学意义（P>0.05）；而两组患者性别、糖尿病、脓肿深度、致病菌及脓肿病史比较差异有统计学意义（P<0.05）。Logistic回归分析结果显示，男性、糖尿病、深部脓肿、致病菌为肠道来源菌群、有脓肿病史均是急性肛周脓肿术后发生肛瘘的独立危险因素（P<0.05）。列线图模型预测急性肛周脓肿术后发生肛瘘风险C-index为0.844，区分度良好，校正曲线趋近于标准曲线，预测值同实测值基本一致。受试者工作特征（ROC）曲线分析结果显示，曲线下面积为0.945（95%CI：0.915~0.979），表明该列线图模型有良好的预测区分度。

男性、糖尿病、深部脓肿、致病菌肠道来源及有脓肿病史均是急性肛周脓肿术后发生肛瘘的独立危险因素，相关列线图模型可为临床早期识别高风险患者、优化个体化治疗决策提供参考，但其预测效能仍有待多中心、大样本的外部验证。

To investigate the incidence and risk factors of anal fistula after surgery for acute perianal abscess, and to establish and validate a nomogram model for predicting the risk of anal fistula in the early postoperative period.

A total of 144 patients with acute perianal abscess admitted to Xi'an Central Hospital Affiliated to Xi'an Jiaotong University from January 2023 to January 2025 were enrolled as the research subjects, including 104 males and 40 females, aged 21-79 years (mean age 50.77±6.23 years). All patients were categorized into an anal fistula group (n=44) and a non-anal fistula group (n=100) according to whether anal fistula was diagnosed within 3 months after surgery. Clinical data of all patients were collected, including gender, age, body mass index (BMI), smoking history, drinking history, hypertension, constipation, diabetes mellitus, hypertension, abscess size, location and depth, preoperative fever history, preoperative antibiotic use, history of abscess and origin of pathogenic bacteria. Variables with P<0.05 in univariate analysis were included in the multivariable Logistic regression model to identify independent risk factors. A nomogram prediction model was constructed based on statistically significant indicators, and model validation was completed.

The incidence of anal fistula was 30.6% (44/144). There were no statistically significant differences in age, BMI, smoking history, drinking history, hypertension, abscess size, constipation, preoperative fever history, abscess location and preoperative antibiotic use between the anal fistula group and the non-anal fistula group (P>0.05). However, statistically significant differences were found in gender, diabetes mellitus, depth of abscess, pathogenic bacteria origin and history of abscess between the two groups (P<0.05). Logistic regression analysis showed that male gender, diabetes mellitus, depth of abscess, intestinal origin of pathogenic bacteria, and history of abscess were independent risk factors for anal fistula after surgery for acute perianal abscess (P<0.05). The concordance index (C-index) of the nomogram model for predicting the risk of anal fistula after acute perianal abscess surgery was 0.844, indicating good discrimination. The calibration curve was close to the ideal reference line, and the predicted values were generally consistent with the observed values. Receiver operating characteristic (ROC) curve analysis showed that the area under the curve (AUC) was 0.945 (95%CI: 0.915-0.979), suggesting that the nomogram model had favorable predictive discrimination.

Male gender, diabetes mellitus, depth of abscess, intestinal origin of pathogenic bacteria, and history of abscess are independent risk factors for anal fistula after acute perianal abscess surgery. The nomogram model may provide a useful reference for the early clinical identification of high-risk patients and for optimizing treatment decisions, although external validation is still needed.