2型糖尿病患者药物性低血糖的临床调查及影响因素

Clinical investigation and influencing factors of drug-induced hypoglycemia in patients with diabetes mellitus type 2

目的调查2型糖尿病(T2DM)患者药物性低血糖发生率,并采用Logisitc模型分析药物性低血糖的影响因素。方法选取2021年1-6月医院收治的168例T2DM患者,调查药物性低血糖发生率。根据低血糖发生情况分为非低血糖组和低血糖组,采用Logistic回归方程分析影响因素,采用受试者工作特征曲线(ROC)分析影响因素对药物性低血糖的预测价值。结果在收治的168例T2DM患者中,药物性低血糖发生率为26.79%(45例),其中发生1次者占71.11%,发生2次者占13.33%,发生3次及以上者占15.56%;45例药物性低血糖患者中症状性低血糖占44.44%,无症状性低血糖占28.89%,相对低血糖占17.78%,可疑症状性低血糖占8.89%。多因素Logistic回归方程分析显示,糖尿病病程、体质量指数(BMI)、糖化血红蛋白(Hb A1c)、丙氨酸氨基转移酶(ALT)和治疗方案均是药物性低血糖的影响因素(P<0.05)。通过Logistic二元拟合回归构建各指标联合预测药物性低血糖的ROC模型示,BMI、Hb A1c和ALT联合预测药物性低血糖的AUC最大。结论受多种因素影响,T2DM患者存在药物性低血糖风险,临床工作中应加强低血糖时段的血糖监测,及时发现低血糖,做到早预防、早发现、早治疗。

AIM To investigate the incidence of drug-induced hypoglycemia in patients with diabetes mellitus type 2(T2DM),and use the Logisitc model to analyze the influencing factors of drug-induced hypoglycemia.METHODS A total of 168 T2DM patients admitted to our hospital from January to June 2021 were selected to investigate the incidence of drug-induced hypoglycemia.According to the occurrence of hypoglycemia,they were divided into nonhypoglycemia group and hypoglycemia group.Logistic regression equation was used to analyze the influencing factors,and receiver operating characteristic curve(ROC)was used to analyze the predictive value of influencing factors on druginduced hypoglycemia.RESULTS In 168 T2DM patients,the incidence of drug-induced hypoglycemia was 26.79%,of which 71.11%occurred once,13.33%occurred twice,and 15.56%occurred three times or more.Among 45 drug induced hypoglycemia patients,symptomatic hypoglycemia accounted for 44.44%,asymptomatic hypoglycemia accounted for 28.89%,relative hypoglycemia accounted for 17.78%,and suspected symptomatic hypoglycemia accounted for 8.89%.Multivariate Logistic regression analysis showed that diabetes duration,BMI,HbA1c,ALT,and treatment regimen were all influencing factors of drug-induced hypoglycemia(P<0.05).The ROC model for the combined prediction of druginduced hypoglycemia by each index was constructed through Logistic binary fitting regression.The results showed that the AUC of BMI,HbA1c,and ALT combined to predict drug-induced hypoglycemia was the largest.CONCLUSION Affected by many factors,patients with T2DM are at risk of drug-induced hypoglycemia.In clinical work,blood glucose monitoring during the period of hypoglycemia should be strengthened to detect hypoglycemia in time,so as to achieve early prevention,early detection and early treatment..