中度重症急性胰腺炎和重症急性胰腺炎患者急性肾损伤预测模型的构建与验证研究

Development and validation of a predictive model for acute kidney injury in patients with moderately severe acute pancreatitis and severe acute pancreatitis

目的构建并验证中度重症急性胰腺炎(MSAP)和重症急性胰腺炎(SAP)患者并发急性肾损伤(AKI)风险的预测模型。方法将2019年1月至2022年8月海南医学院第一附属医院收治的确诊为MSAP和SAP的患者纳入本次回顾性研究,按照2∶1的比例将患者随机分为训练组和验证组。使用最小绝对收缩和选择算子(LASSO)法和机器学习筛选发生AKI的相关预测因子,用多因素Logistic回归分析筛选发生AKI的危险因素,并构建列线图预测模型,用校准曲线对模型进行一致性评价,用受试者工作特征(ROC)曲线评估模型的预测性能,用决策曲线分析(DCA)评估模型的临床价值。结果共纳入565例MSAP和SAP患者,其中377例纳入训练组,188例纳入验证组,训练组中96例患者(25.46%)和验证组中44例患者(23.40%)并发AKI。各因素Logistic回归分析显示C反应蛋白、腹内压和胱抑素C是MSAP和SAP并发AKI的危险因素(P<0.05)。校准曲线显示模型的预测值与实际值一致性良好;受试者工作特征曲线显示模型预测性能较高,DCA显示模型的临床价值较高。结论列线图模型在预测MSAP和SAP发生AKI有较好的效果,该模型可帮助临床医师对患者进行分层,以便早期进行预防和治疗,进而有效改善患者预后。

Objective To developed and validated a nomogram prediction model to predict the risk of AKI in patients with moderately severe acute pancreatitis(MSAP)and severe acute pancreatitis(SAP).Methods Patients with a diagnosis of MSAP and SAP admitted to the First Affiliated Hospital of Hainan Medical University from January 2019 to August 2022 were retrospectively included in the study and were randomly divided into the training group and validation group according to a ratio of 2∶1.The prediction factors were screened using least absolute shrinkage and selection operator(LASSO)method and machine learning.Multivariate logistic regression analysis was used to screen the risk factors of AKI,and the prediction model were develope.The consistency of the model was assessed using calibration plots,and the predictive power was assessed by receiver operating characteristic(ROC)curve.Clinical value was assessed using decision-curve analysis(DCA).Results A total of 565 patients were enrolled in this research and 377 patients were allocated in the training group and 188 in the validation group.AKI occurred in 96 patients(25.46%)in the training group and 44(23.40%)in the validation group,respectively.Nultiple Logistic regression analysis showed that CRP,IAP and CysC were risk factors.The calibration curve verified the good agreement of the predicted and actual results.ROC curve verified the good predictive ability of nomogram.DCA verified the good clinical value of nomogram.Conclusion The prediction model has good performance for predicting AKI in MSAP and SAP patients.Application of this model can help clinicians stratify patients for primary prevention and early therapeutic intervention to improve prognosis.