Determinants of Severe Hypoglycemia Complicating Type 2 Diabetes: The Fremantle Diabetes Study
Determinants of Severe Hypoglycemia Complicating Type 2 Diabetes: The Fremantle Diabetes Study
复制标题
DOI：
10.1210/jc.2009-2828
复制
发表时间：
2010-05-01
影响因子：
5.8
通讯作者：
Davis, Wendy A.
中科院分区：
文献类型：
--
作者：
Davis, Timothy M. E.;Brown, Simon G. A.;Davis, Wendy A.
背景：已发表的关于 2 型糖尿病并发严重低血糖特征的数据有限。目的：本研究的目的是确定社区居住的 2 型糖尿病患者严重低血糖的发生率和预测因素。设计：这是一项纵向观察性队列研究。背景：这是一项基于社区的研究。患者：1998 年对 616 名患者（平均年龄 67.0 岁，52.3% 男性，中位糖尿病病程 7.7 年）进行了评估，并随访至 2006 年 6 月底。 主要结果指标：严重低血糖定义为需要救护车、急诊室服务和/或住院治疗。 Cox 比例风险模型用于确定首次发作的预测因子，泊松、负二项、零膨胀泊松和零膨胀负二项回归模型确定了频率的预测因子。 结果：52 例 (8.4%) 在 3953 个患者年期间经历了 66 次发作（每 100 个患者年发生率 1.7 次）。那些经历严重低血糖的人有一到四次发作。首次发作时间的显着独立预测因素包括胰岛素治疗持续时间、估计肾小球滤过率每 1.73 m(2) 低于 60 ml/min、周围神经病变、小学以上教育程度以及既往严重低血糖情况。零膨胀负二项式提供了严重低血糖频率的最佳模型。较低的空腹血糖和较高的糖化血红蛋白与频率显着相关，而重复严重低血糖风险极低的患者不太可能使用胰岛素或接受短期胰岛素治疗，也不太可能患有肾功能损害或周围神经病变，也不太可能接受小学以上的教育。结论：胰岛素治疗持续时间被证实是严重低血糖的独立危险因素。与受教育程度的新颖关联表明知识驱动的强化血糖自我管理。频率和糖化血红蛋白之间的正相关可以识别血糖控制不稳定的患者。 （临床内分泌代谢杂志 95：2240-2247，2010）
Context: There are limited published data characterizing severe hypoglycemia complicating type 2 diabetes.Objective: The objective of the study was to determine the incidence and predictors of severe hypoglycemia in community-dwelling type 2 patients.Design: This was a longitudinal observational cohort study.Setting: This was a community-based study. Patients: There were 616 patients (mean age 67.0 yr, 52.3% males, median diabetes duration 7.7 yr) assessed in 1998 and followed up to the end of June 2006.Main Outcome Measures: Severe hypoglycemia defined as that requiring ambulance attendance, emergency department services, and/or hospitalization. Cox proportional hazards modeling was used to determine predictors of first episode, and Poisson, negative binomial, zero-inflated Poisson, and zero-inflated negative binomial regression models identified predictors of frequency.Results: Fifty-two (8.4%) experienced 66 episodes during 3953 patient-years (incidence 1.7 per 100 patient-years). Those experiencing severe hypoglycemia had one to four episodes. Significant independent predictors of time to first episode were duration of insulin treatment, estimated glomerular filtration rate less than 60 ml/min per 1.73 m(2), peripheral neuropathy, education beyond primary level, and past severe hypoglycemia. The zero-inflated negative binomial provided the best model of severe hypoglycemia frequency. Lower fasting serum glucose and higher glycosylated hemoglobin were significantly associated with frequency, whereas patients at minimal risk of repeated severe hypoglycemia were unlikely to use insulin or to have short-duration insulin treatment, to have renal impairment or peripheral neuropathy, or to be educated beyond primary level.Conclusions: Duration of insulin treatment was confirmed as an independent risk factor for severe hypoglycemia. The novel association with educational attainment suggests knowledge-driven intensive glycemic self-management. The positive relationship between frequency and glycosylated hemoglobin may identify patients with unstable glycemic control. (J Clin Endocrinol Metab 95: 2240-2247, 2010)