糖皮质激素诱发类固醇糖尿病38例临床分析

Clinical analysis of 38 cases of steroid-induced diabetes mellitus due to glucocorticoid treatment

目的 探讨糖皮质激素诱发类固醇糖尿病的危险因素、临床特点。方法 对2013—2016年在杭州市第三人民医院皮肤科798例系统使用糖皮质激素（简称激素）治疗的患者进行回顾性分析，采用logistic回归模型分析类固醇糖尿病发生的影响因素，重复测量方差分析比较类固醇糖尿病患者三餐后血糖水平，t检验比较类固醇糖尿病与2型糖尿病患者空腹血糖及糖化血红蛋白（HbA1c）的差异。结果 798例患者中，类固醇糖尿病患者38例，年龄（66.86 ± 13.30）岁，显著高于非类固醇糖尿病患者［760例，（39.95 ± 17.01）岁］，两组差异有统计学意义（t = 8.86，P 〈 0.01），但两组间性别分布差异无统计学意义（χ2 = 1.61，P = 0.20）。类固醇糖尿病患者伴脂肪肝、高脂血症、高血压、肝功能异常以及有糖尿病家族史的比例显著高于非类固醇糖尿病组（χ2 = 12.25、19.25、32.69、21.47、16.70，均P 〈 0.01）。Logistic回归分析显示，年龄、脂肪肝、高脂血症、高血压、肝功能异常、激素用量、激素使用时间、免疫抑制剂的使用及糖尿病家族史是类固醇糖尿病发生的危险因素（均P 〈 0.05）。0.50 ~ 0.74、0.75 ~ 0.99、1.00 ~ 1.25 mg·kg？1·d？1激素用量类固醇糖尿病患者组之间空腹血糖及三餐后末梢血糖差异均无统计学意义（P 〉 0.05）。类固醇糖尿病患者早餐、午餐及晚餐后末梢血糖分别为（11.50 ± 2.90）、（16.02 ± 5.81）、（16.81 ± 4.52） mmol/L。类固醇糖尿病组与2型糖尿病组间空腹血糖及HbA1c差异均有统计学意义（t = 3.74、9.92，均P 〈 0.01）。结论 年龄、激素用量、激素使用时间及脂肪肝、高脂血症、高血压、肝功能异常等相关基础疾病、免疫抑制剂的使用和糖尿病家族史是发生类固醇糖尿病的危险因素，类固醇糖尿病患者血糖主要于午餐后及晚餐后升高，空腹血糖及HbA1c升高不明显是类固醇糖尿病区别于2型糖尿病的特征。

Objective To investigate risk factors for and clinical features of steroid-induced diabetes mellitus due to glucocorticoid treatment. Methods Clinical data were collected from 798 patients who received systemic glucocorticoid treatment in Department of Dermatology of Hangzhou Third People′s Hospital from 2013 to 2016, and analyzed retrospectively. Logistic regression analysis was performed to analyze the factors influencing the occurrence of steroid-induced diabetes mellitus （SDM）, repeated-measures analysis of variance to compare peripheral blood glucose levels of patients with SDM after breakfast, lunch and dinner, and t test to compare the levels of fasting blood glucose and glycosylated hemoglobin （HbA1c） between patients with SDM and those with type 2 diabetes mellitus. Results Of the 798 patients, 38 developed SDM due to glucocorticoid treatment. The average age was significantly older in the patients with SDM （[66.86 ± 13.30] years, n = 38） than in those without SDM （[39.95 ± 17.01] years, n = 760; t = 8.86, P 〈 0.01）, but there was no significant difference in the gender ratio between the patients with and thhose without SDM （χ2 = 1.61, P = 0.20）. The prevalence of fatty liver, hyperlipidemia, hypertension, abnormal liver function and family history of diabetes mellitus was significantly higher in the patients with SDM than in those without SDM （χ2 = 12.25, 19.25, 32.69, 21.47, 16.70 respectively, all P 〈 0.01）. Logistic regression analysis showed that age, fatty liver, hyperlipidemia, hypertension, abnormal liver function, dosage of glucocorticoids, duration of glucocorticoid therapy, use of immunosuppressive agents and family history of diabetes mellitus were risk factors for SDM （all P 〈 0.05）. There were no significant differences in fasting blood glucose levels or postprandial peripheral blood glucose levels among the SDM patients receiving glucocorticoid therapy at different dosages of 0.50 - 0.74, 0.75 - 0.99, 1.00 - 1.25 mg·kg-1·d-1 （P 〉 0.05）. The peripheral blood glucose levels after breakfast, lunch and dinner were （11.50 ± 2.90）, （16.02 ± 5.81） and （16.81 ± 4.52） mmol/L respectively in the patients with SDM. The levels of fasting blood glucose and glycosylated HbA1c were both significantly lower in the patients with SDM than in those with type 2 diabetes mellitus（t = 3.74, 9.92 respectively, both P 〈 0.001）. Conclusions The risk factors for SDM are age, dosage of glucocorticoids, duration of glucocorticoid therapy, fatty liver, hyperlipidemia, hypertension, abnormal liver function, use of immunosuppressive agents and family history of diabetes mellitus. The patients with SDM showed obviously elevated blood glucose levels mostly after lunch and dinner, but slightly increased levels of fasting blood glucose and glycosylated HbA1c, which can be used to distinguish between SDM and type 2 diabetes mellitus.