Karges Beate, Rosenbauer Joachim, Kapellen Thomas, Wagner Verena M, Schober Edith, Karges Wolfram, Holl Reinhard W
Division of Endocrinology and Diabetes, RWTH Aachen University, Aachen, Germany.
Institute for Biometrics and Epidemiology, German Diabetes Center, Leibniz Center at University of Düsseldorf, Düsseldorf, Germany.
PLoS Med. 2014 Oct 7;11(10):e1001742. doi: 10.1371/journal.pmed.1001742. eCollection 2014 Oct.
Severe hypoglycemia is a major complication of insulin treatment in patients with type 1 diabetes, limiting full realization of glycemic control. It has been shown in the past that low levels of hemoglobin A1c (HbA1c), a marker of average plasma glucose, predict a high risk of severe hypoglycemia, but it is uncertain whether this association still exists. Based on advances in diabetes technology and pharmacotherapy, we hypothesized that the inverse association between severe hypoglycemia and HbA1c has decreased in recent years.
We analyzed data of 37,539 patients with type 1 diabetes (mean age ± standard deviation 14.4 ± 3.8 y, range 1-20 y) from the DPV (Diabetes Patienten Verlaufsdokumentation) Initiative diabetes cohort prospectively documented between January 1, 1995, and December 31, 2012. The DPV cohort covers an estimated proportion of >80% of all pediatric diabetes patients in Germany and Austria. Associations of severe hypoglycemia, hypoglycemic coma, and HbA1c levels were assessed by multivariable regression analysis. From 1995 to 2012, the relative risk (RR) for severe hypoglycemia and coma per 1% HbA1c decrease declined from 1.28 (95% CI 1.19-1.37) to 1.05 (1.00-1.09) and from 1.39 (1.23-1.56) to 1.01 (0.93-1.10), respectively, corresponding to a risk reduction of 1.2% (95% CI 0.6-1.7, p<0.001) and 1.9% (0.8-2.9, p<0.001) each year, respectively. Risk reduction of severe hypoglycemia and coma was strongest in patients with HbA1c levels of 6.0%-6.9% (RR 0.96 and 0.90 each year) and 7.0%-7.9% (RR 0.96 and 0.89 each year). From 1995 to 2012, glucose monitoring frequency and the use of insulin analogs and insulin pumps increased (p<0.001). Our study was not designed to investigate the effects of different treatment modalities on hypoglycemia risk. Limitations are that associations between diabetes education and physical activity and severe hypoglycemia were not addressed in this study.
The previously strong association of low HbA1c with severe hypoglycemia and coma in young individuals with type 1 diabetes has substantially decreased in the last decade, allowing achievement of near-normal glycemic control in these patients. Please see later in the article for the Editors' Summary.
严重低血糖是1型糖尿病患者胰岛素治疗的主要并发症，限制了血糖控制目标的完全实现。过去曾有研究表明，糖化血红蛋白（HbA1c）水平较低（平均血糖的一个指标）预示着严重低血糖的高风险，但这种关联是否仍然存在尚不确定。基于糖尿病技术和药物治疗的进展，我们推测近年来严重低血糖与HbA1c之间的负相关关系有所减弱。
我们分析了来自糖尿病患者病程记录（DPV）倡议糖尿病队列中37539例1型糖尿病患者（平均年龄±标准差为14.4±3.8岁，范围1 - 20岁）的数据，这些数据前瞻性记录于1995年1月1日至2012年12月31日期间。DPV队列涵盖了德国和奥地利所有儿科糖尿病患者中估计超过80%的比例。通过多变量回归分析评估严重低血糖、低血糖昏迷与HbA1c水平之间的关联。从1995年到2012年，每降低1%的HbA1c，严重低血糖和昏迷的相对风险（RR）分别从1.28（95%CI 1.19 - 1.37）降至1.05（1.00 - 1.09），以及从1.39（1.23 - 1.56）降至1.01（0.93 - 1.10），分别相当于每年风险降低1.2%（95%CI 0.6 - 1.7，p<0.001）和1.9%（0.8 - 2.9，p<0.001）。HbA1c水平在6.0% - 6.9%（每年RR分别为0.96和0.90）以及7.0% - 7.9%（每年RR分别为0.96和0.89）的患者中，严重低血糖和昏迷的风险降低最为显著。从1995年到2012年，血糖监测频率以及胰岛素类似物和胰岛素泵的使用有所增加（p<0.001）。我们的研究并非旨在调查不同治疗方式对低血糖风险的影响。局限性在于本研究未涉及糖尿病教育和体育活动与严重低血糖之间的关联。
在过去十年中，1型糖尿病年轻患者中先前HbA1c水平低与严重低血糖和昏迷之间的强关联已大幅减弱，使得这些患者能够实现接近正常的血糖控制。编者总结请见本文后续内容。
