血液肿瘤化疗患者继发糖尿病的危险因素研究及预测模型构建  

Risk factors of secondary diabetes in patients with hematologic tumor received chemotherapy and construction of prediction model

作　　者：周倩 余红霞 巨姣春 贺章亚 陈方尧[3] 罗小琴 Zhou Qian;Yu Hongxia;Ju Jiaochun;He Zhangya;Chen Fangyao;Luo Xiaoqin(Department of Nutrition and Food Safety,School of Public Health,Xi'an Jiaotong University,Xi'an 710061,China;Department of Hematology,Tangdu Hospital,the Second Affiliated Hospital of the Air Force Medical University,Xi'an 710038,China;Department of Epidemiology and Health Statistics,School of Public Health,Xi'an Jiaotong University,Xi'an 710061,China)

机构地区：[1]西安交通大学公共卫生学院营养与食品安全系,710061 [2]空军军医大学第二附属医院唐都医院血液内科,西安710038 [3]西安交通大学公共卫生学院流行病与卫生统计学系,710061

出　　处：《国际内分泌代谢杂志》2021年第5期439-444,共6页International Journal of Endocrinology and Metabolism

摘　　要：目的探讨血液肿瘤化疗患者继发糖尿病的危险因素并建立预测模型。方法回顾性分析唐都医院2017年1月至2019年6月血液肿瘤化疗患者769例,根据是否继发糖尿病分为病例组(n=289)和对照组(n=480)。使用logistic回归分析法建立继发糖尿病风险预测模型,Hosmer-Lemeshow拟合优度检验评价模型校准度,使用曲线下面积(AUC)评价模型区分度,后续收集2019年7—12月183例患者资料进行模型外部验证。结果建立了由糖尿病家族史、高血压史等7项指标组成的继发糖尿病风险预测模型。模型AUC为0.917,灵敏度为89.3%,特异度为83.1%。结论对于有糖尿病家族史、高血压史、体重指数(BMI)≥24 kg/m^(2)、高龄、口服激素且时间长用量大的血液肿瘤化疗患者应警惕继发性糖尿病的发生,本预测模型具有较好的精确度和区分度。Objective To explore the risk factors of secondary diabetes in patients with hematologic tumor and treated by chemotherapy and establish a prediction model.Methods A retrospective analysis was performed on 769 patients with hematologic tumor in Tangdu Hospital from January 2017 to June 2019.The patients were divided into case group(n=289)and control group(n=480)according to whether they had secondary diabetes.Logistic regression analysis was used to establish the risk prediction model for secondary diabetes.Hosmer-Lemeshow goodness of fit test was used to evaluate the model calibration degree,and area under receiver(AUC)of Charateristic curve(ROC curve)was used to evaluate the differentiation degree.The data of 183 patients from July to December 2019 were collected for external validation of the model.Results A risk prediction model of secondary diabetes was established consisted of 7 indicators including family history of diabetes,hypertension,etc.AUC of this model was 0.917 with high sensitivity(89.3%)and specificity(83.1%).Conclusion Patients with family history of diabetes,history of hypertension,body mass index(BMI)≥24 kg/m^(2),advanced age and oral hormone treatment for a long time with large dosage should be vigilant the occurrence of secondary diabetes in patients with nematologic tumor and treated by chemotherapy.The prediction model has good accuracy and discrimination.

关 键 词：血液肿瘤 继发性糖尿病 危险因素 预测模型 

分 类 号：R587.1[医药卫生—内分泌] R733[医药卫生—内科学]