Department of Family Medicine and Primary Care, University of Hong Kong, Hong Kong.
Department of Pharmacology and Pharmacy, the University of Hong Kong, Hong Kong.
Diabetes Obes Metab. 2020 Dec;22(12):2325-2334. doi: 10.1111/dom.14157. Epub 2020 Sep 17.
To investigate the assocation between glycated haemoglobin (HbA1c) level and cardiovascular disease (CVD) risk among patients with type 2 diabetes.
This retrospective cohort study conducted in Hong Kong selected patients aged 45 to 84 years with type 2 diabetes mellitus and without CVD from primary care clinics during the period 2008 to 2010. Usual HbA1c measurement was calculated using a mixed-effects model to minimize regression dilution bias. The association between usual HbA1c and CVD risk was assessed by Cox regression, with adjustment for baseline covariates. Subgroup analyses by patient characteristics were also conducted.
After a median follow-up period of 8.4 years (1.4 million person-years), 174 028 patients with 34 074 CVD events were observed. A curvilinear association was found between usual HbA1c and total CVD, stroke, heart failure and CVD mortality risk. No significant difference was found among patients with usual HbA1c <53 mmol/mol (7%). A positive linear association was found between usual HbA1c and the risks of outcomes when the usual HbA1c was 53 mmol/mol (7%) or above. The adjusted hazard ratios (HRs) for CVD risk per 1% increment in usual HbA1c >7% was 21% (HR 1.21, 95% confidence interval [CI] 1.18-1.23) (HR for CVD per 1mmol/mol increment in usual HbA1c > 53 mmol/mol was 1.7% (HR 1.017, CI 1.015-1.019)). A similar pattern was identified in a patient subgroup analysis, but the effects of usual HbA1c in younger patients were more prominent than in older patients.
Usual HbA1c increments for levels >53 mmol/mol (7.0%) were associated with elevated CVD risk, but no difference was found in the population with usual HbA1c <53 mmol/mol (7.0%), irrespective of patient characteristics. For CVD prevention, strict adherence to an HbA1c target of <53 mmol/mol (7%) should apply to younger patients.
探讨糖化血红蛋白（HbA1c）水平与 2 型糖尿病患者心血管疾病（CVD）风险之间的关系。
本研究为回顾性队列研究，在香港，选取 2008 年至 2010 年期间年龄在 45 至 84 岁、无 CVD 的 2 型糖尿病患者，在基层医疗机构接受治疗。采用混合效应模型计算常规 HbA1c，以最小化回归稀释偏倚。采用 Cox 回归评估常规 HbA1c 与 CVD 风险之间的关系，并根据基线协变量进行调整。还进行了按患者特征的亚组分析。
中位随访 8.4 年（140 万人年）后，观察到 174028 例患者发生 34074 例 CVD 事件。常规 HbA1c 与总 CVD、中风、心力衰竭和 CVD 死亡率风险之间存在曲线关系。常规 HbA1c<53mmol/mol（7%）的患者之间无显著差异。常规 HbA1c 为 53mmol/mol（7%）或更高时，常规 HbA1c 与结局风险之间呈正线性关系。常规 HbA1c 每增加 1%，CVD 风险的调整后的危害比（HR）为 21%（HR 1.21，95%置信区间 [CI] 1.18-1.23）（常规 HbA1c 每增加 1mmol/mol，CVD 风险的 HR 为 1.7%（HR 1.017，CI 1.015-1.019））。在患者亚组分析中也发现了类似的模式，但在年轻患者中常规 HbA1c 的影响比老年患者更为明显。
水平>53mmol/mol（7.0%）的常规 HbA1c 增加与 CVD 风险升高相关，但在常规 HbA1c<53mmol/mol（7.0%）的人群中无差异，无论患者特征如何。对于 CVD 预防，应严格遵守 HbA1c<53mmol/mol（7.0%）的目标，适用于年轻患者。
