摘要
目的 本研究旨在筛选导致老年急性胰腺炎(AP)患者发展为重症急性胰腺炎的关键风险因素,并建立一个预测模型,精确评估老年AP患者的严重程度.方法 回顾性分析2018年6月至2023年6月安徽医科大学第二附属医院收治的374名老年急性胰腺炎患者(年龄＞65岁),按照严重程度将患者分为重症急性胰腺炎组和非重症急性胰腺炎组,比较两组患者入院24 h内收集的临床指标的差异,利用多因素Logistic回归识别独立危险因素并构建Nomogram临床预测模型.结果 重症急性胰腺炎组患者BISAP评分、合并高血压的比例显著高于非重症急性胰腺炎组患者.在实验室指标上,重症急性胰腺炎组白细胞、中性粒细胞百分比、血红蛋白、血细胞比容、血浆凝血酶原时间、纤维蛋白原、D-二聚体、尿素氮、肌酐、血钾显著高于非重症急性胰腺炎组,抗凝血酶-Ⅲ活性、白蛋白显著低于非重症急性胰腺炎组.多因素Logistic回归分析结果显示,合并高血压(OR=1.88,95％CI:1.02～3.46,P=0.042)、BISAP 评分高(OR=3.62,95％CI:2.28～5.75,P＜0.001)、血细胞比容(OR=1.08,95％CI:1.02～1.14,P=0.006)、白蛋白低(OR=0.93,95％CI:0.88～0.98,P=0.008)是老年急性胰腺炎患者发展为重症急性胰腺炎的独立危险因素.预测模型的一致性指数为0.865(95％CI:0.825～0.906),模型的预测校准曲线与参考曲线接近,这提示该模型具有较好的区分能力和校准能力.临床决策曲线分析表明该模型具有良好的临床实用性.结论 本研究强调高血压、高BISAP评分、高血细胞比容和低白蛋白是老年AP患者发展为SAP的显著危险因素.所建立的预测模型具备高度的预测价值和临床应用潜力,为医生在早期阶段制定治疗策略提供了重要依据.
Abstract
Objective This study aims to identify key risk factors leading to the development of severe acute pancreatitis(AP)in elderly patients and to establish a prediction model for accurately assessing the severity of AP in the elderly.Methods A retrospective analysis was conducted on 374 elderly acute pancreatitis patients(aged＞65 years)treated at the Second Affiliated Hos-pital of Anhui Medical University from June 2018 to June 2023.Patients were divided into severe acute pancreatitis and non-severe a-cute pancreatitis groups based on the severity of their condition.Differences in clinical indicators collected within 24 hours of hospital admission between the two groups were compared.Multivariate logistic regression was utilized to identify independent risk factors and to construct a Nomogram clinical prediction model.Results The severe acute pancreatitis group showed significantly higher BISAP scores and a higher proportion of concomitant hypertension compared to the non-severe acute pancreatitis group.In terms of laboratory indica-tors,white blood cell count,neutrophil percentage,hemoglobin,hematocrit,plasma prothrombin time,fibrinogen,D-dimer,blood u-rea nitrogen,creatinine,and potassium were significantly higher in the severe group,while antithrombin-Ⅲ activity and albumin were significantly lower.Multivariate logistic regression analysis revealed that concomitant hypertension(OR=1.88,95％CI:1.02～3.46,P=0.042),high BISAP scores(OR=3.62,95％CI:2.28～5.75,P＜0.001),hematocrit(OR=1.08,95％CI:1.02～1.14,P=0.006),and low albumin(OR=0.93,95％CI:0.88～0.98,P=0.008)were independent risk factors for the development of se-vere acute pancreatitis in elderly patients.The predictive model had a concordance index of 0.865(95％CI:0.825～0.906),with the predictive calibration curve closely matching the reference curve,indicating good discriminative ability and calibration of the model.Decision curve analysis showed that the model has good clinical utility.Conclusion This study emphasizes that high blood pressure,high BISAP scores,high hematocrit,and low albumin are significant risk factors for the development of severe acute pancreatitis in eld-erly patients.The established prediction model has high predictive value and clinical application potential,providing an important basis for physicians to formulate treatment strategies at an early stage.
维普
万方数据