2025 年第 9 期 第 20 卷

重症急性胰腺炎患者合并腹内高压的影响因素分析及列线图预测模型构建

Analysis of the influencing factors of severe acute pancreatitis complicated with intra-abdominal hypertension and construction of the nomogram prediction model

作者：戢德宇1钱嘉俊2宋志1杨公利3

英文作者：Ji Deyu1 Qian Jiajun2 Song Zhi1 Yang Gongli3

单位：1深圳大学总医院重症医学科，深圳510180；2广东省广州市第一人民医院南沙医院重症监护病房，广州511466；3深圳大学总医院消化内科，深圳510180

英文单位：1Department of Critical Care Medicine Shenzhen University General Hospital Shenzhen 510180 China; 2Intensive Care Unit Nansha Hospital Guangzhou First People′s Hospital Guangdong Province Guangzhou 511466 China; 3Department of Gastroenterology Shenzhen University General Hospital Shenzhen 510180 China

关键词：重症急性胰腺炎；腹内高压；列线图模型；预测效能

英文关键词：Severeacutepancreatitis;Intra-abdominalhypertension;Nomogrammodel;Powerofprediction

• 摘要：

• 目的 分析重症急性胰腺炎（SAP）患者合并腹内高压的影响因素，据此构建列线图模型，并评估列线图模型的预测效能。方法 回顾性分析2022年2月至2024年6月深圳大学总医院重症医学科和广东省广州市第一人民医院南沙医院重症监护病房收治的160例SAP患者的临床资料。根据住院期间是否合并腹内高压，将合并腹内高压的SAP患者纳入发生组（82例），未合并腹内高压的SAP患者纳入未发生组（78例）。比较2组患者入院24 h内临床资料，采用Logistic回归分析法筛选SAP患者合并腹内高压的危险因素。构建列线图模型，采用受试者工作特征曲线、Hosmer-Lemeshow检验、决策分析曲线评估列线图模型区分度、校准度及临床收益。结果 发生组和未发生组体重指数、急性生理学与慢性健康状况评分系统Ⅱ（APACHE Ⅱ）评分、24 h液体负荷量、呼气末正压、降钙素原、C反应蛋白水平比较，差异均有统计学意义（均P＜0.05）。Logistic回归分析结果显示，体重指数（比值比=1.122）、APACHE Ⅱ评分（比值比=1.342）、24 h液体负荷量（比值比=1.229）、呼气末正压（比值比=1.224）、降钙素原（比值比=1.164）、C反应蛋白（比值比=1.182）均是影响SAP患者合并腹内高压的危险因素（均P＜0.05）。基于体重指数、APACHE Ⅱ评分、24 h液体负荷量、呼气末正压、降钙素原、C反应蛋白构建SAP患者合并腹内高压风险列线图模型。Hosmer-Lemeshow检验结果显示，列线图模型预测值与实际值比较，差异无统计学意义（χ2=2.015，P=0.377）；受试者工作特征曲线分析结果显示，该列线图模型预测SAP患者合并腹内高压的敏感度、特异度、曲线下面积分别为89.9%、74.8%、0.885；决策曲线分析结果显示，列线图模型预测价值高，具有临床有效性。结论 基于体重指数、APACHE Ⅱ评分、24 h液体负荷量、呼气末正压、降钙素原、C反应蛋白构建的SAP患者合并腹内高压风险列线图模型具有良好的预测效能。

• Objective To analyze the influencing factors of intra-abdominal hypertension (IAH) in patients with severe acute pancreatitis (SAP), and to construct a nomogram model based on these factors, and to evaluate the predictive efficacy of the nomogram model. Methods The clinical data of 160 SAP patients admitted to the Department of Critical Care Medicine, Shenzhen University General Hospital and Intensive Care Unit, Nansha Hospital, Guangzhou First People′s Hospital, Guangdong Province from February 2022 to June 2024 were retrospectively analyzed. According to the presence or absence of IAH during hospitalization, SAP patients with IAH were included in the occurrence group (82 cases), and SAP patients without IAH were included in the non-occurrence group (78 cases). The clinical data of the two groups within 24 hours after admission were compared. Logistic regression analysis was used to screen the risk factors of SAP patients with IAH. A nomogram model was constructed, and the receiver operating characteristic curve, Hosmer-Lemeshow test, and decision analysis curve were used to evaluate the discrimination, calibration, and clinical benefits of the nomogram model. Results There were significant differences in body mass index (BMI), acute physiology and chronic health evaluation Ⅱ (APACHE Ⅱ) score, 24 h fluid load, positive end-expiratory pressure, procalcitonin and C-reactive protein levels between the two groups (all P<0.05). Logistic regression analysis showed that, BMI (odds ratio=1.122), APACHE Ⅱ score (odds ratio=1.342), 24 h fluid load (odds ratio=1.229), positive end-expiratory pressure (odds ratio=1.224), procalcitonin (odds ratio=1.164) and C-reactive protein (odds ratio=1.182) were all risk factors for IAH in SAP patients (all P<0.05). A risk nomogram model for SAP patients with IAH was constructed based on BMI, APACHE Ⅱ score, 24 h fluid load, positive end-expiratory pressure, procalcitonin and C-reactive protein. Hosmer-Lemeshow test results  showed that there was no significant difference between the predicted value of the nomogram model and the actual value (χ2=2.015, P=0.377). The results  of receiver operating characteristic curve analysis showed that the sensitivity, specificity, and area under the curve of the nomogram model for predicting SAP patients with IAH were 89.9%, 74.8%, and 0.885, respectively. The results  of decision curve analysis showed that the nomogram model had high predictive value and clinical validity. Conclusions The nomogram model based on BMI, APACHE Ⅱ score, 24 h fluid load, positive end-expiratory pressure, procalcitonin and C-reactive protein has a good predictive efficacy for SAP patients complicated with IAH.