重症胰腺炎后并发糖尿病的相关危险因素及预测模型

Risk Factors and Predictive Models of Diabetes Mellitus After Severe Acute Pancreatitis

目的探讨重症急性胰腺炎(severe acute pancreatitis,SAP)后并发糖尿病(DM)的相关危险因素;构建相应预测模型,对其结果进行评估。方法选择2015年2月起至2018年2月于西南医科大学附属医院消化内科收治的347例SAP患者的临床资料及实验室检查进行回顾性分析方法研究,将SAP后并发DM的68例作为试验组,SAP未并发DM的279例作为对照组,先找出两组患者中具有统计学意义的指标,进一步采用多因素Logistic回归分析方法对其进行分析;构建相应预测模型,并制作观察者的特征曲线(即ROC曲线)对其预测效能进行评估。结果两组的身体质量指数(BMI)、CT严重指数评分(CTSI)、急性和慢性健康状况评分Ⅱ(APACHEⅡ)、Ranson评分、吸烟差异具有统计学意义(P<0.05)。两组实验室结果相比,凝血酶原时间(PT)、降钙素原(PCT)、C反应蛋白(CRP)、白蛋白(ALB)、丙氨酸氨基转移酶(ALT)、天门冬氨酸氨基转移酶(AST)、甘油三酯(TG)、尿素氮(BUN)、血清肌酐(Scr)、胰淀粉酶、脂肪酶、血糖、血钾、血钙、氧分压(PO2)、肌钙蛋白I(TnI)差异有统计学意义(P<0.05)。通过多因素Logistic回归分析方法进行分析后得出,高血糖、高BMI指数、高CTSI评分、高APACHEⅡ评分、低血钙是SAP后并发DM的独立危险因素(P<0.05)。据上述危险因素建立预测模型:LogitP=0.11×血糖+0.214×BMI+0.276×CTSI评分-1.997×血钙+0.167×APACHEⅡ评分-2.587。并绘制ROC曲线,ROC曲线下面积为0.955,约登指数为0.78时模型敏感度为89%,特异度为89%,表明该模型能较好地预测SAP后并发DM的风险。结论SAP后并发DM的独立危险因素为高血糖、高BMI指数、高CTSI评分、高APACHEⅡ评分、低血钙,且预测模型具有较好的预测风险能力。

Objective To investigate the risk factors associated with diabetes mellitus(DM)after severe acute pancreatitis(SAP),establish a predictive model and evaluate its value.Methods The clinical data and laboratory examination results of 347 SAP patients admitted to the Affiliated Hospital of Southwest Medical University from February of 2015 to February of 2018 were analyzed retrospectively.68 patients with DM after SAP were divided into the observation group while the other 279 patients without DM after SAP were divided into the control group.The statistically significant indicators of the two groups were found out and then analyzed by the multivariate logistic regression analysis.The prediction model was established and the receiver operating characteristic(ROC)curve was used to assess its prediction efficacy.Results There were significant differences between the two groups in the body mass index(BMI),CT severity index(CTSI)score,acute physiology and chronic health evaluationⅡ(APACHEⅡ)and Ranson score(P<0.05).As to the laboratory examination results,there were significant differences between the two groups in prothorombin time(PT),procalcitonin(PCT),C-reactive protein(CRP),albumin(ALB),alanine aminotransferase(ALT),aspartate aminotransferase(AST),triglyceride(TG),urea nitrogen(BUN),serum creatinine(Scr),pancreatic amylase,lipase,blood glucose,serum potassium,blood calcium,partial pressure of oxygen(PO2)and troponin I(TnI)(P<0.05).The results of the multivariate logistic regression analysis showed that hyperglycemia,high BMI index,high CTSI score,high APACHEⅡscore and hypocalcemia were the independent risk factors of DM after SAP(P<0.05).The predictive model was established based on the above risk factors LogitP=0.11×blood glucose+0.214×BMI+0.276×CTSI score-1.997×blood calcium+0.167×APACHEⅡscore-2.587.The ROC curve was drawn and the area under the ROC curve was 0.955.The model sensitivity was 89%and the specificity was 89%when the Youden index was 0.78,which indicated that the model could better predict the risk of DM after SAP.Conclusion The independent risk factors for DM after SAP are hyperglycemia,high BMI index,high CTSI score,high APACHEⅡscore and hypocalcemia.In addition,the predictive model has better ability of predicting risks.