Luo Miyang, Lim Wei Yen, Tan Chuen Seng, Ning Yilin, Chia Kee Seng, van Dam Rob M, Tang Wern Ee, Tan Ngiap Chuan, Chen Richard, Tai E Shyong, Venkataraman Kavita
Saw Swee Hock School of Public Health, National University of Singapore, Tahir Foundation Building, 12 Science Drive 2, 117549, Singapore.
Yong Loo Lin School of Medicine, National University of Singapore, NUHS Tower Block, Level 11, 1E Kent Ridge Road, 119228, Singapore; NUS Graduate School for Integrative Sciences and Engineering, National University of Singapore, Singapore.
Diabetes Res Clin Pract. 2017 Nov;133:69-77. doi: 10.1016/j.diabres.2017.08.013. Epub 2017 Aug 24.
This study examined longitudinal trends in HbA1c in a multi-ethnic Asian cohort of diabetes patients, and the associations of these trends with future risk of acute myocardial infarction (AMI), stroke, end stage renal failure (ESRD) and all-cause mortality.
6079 participants with type 2 diabetes mellitus in Singapore were included. HbA1c measurements for the five years previous to recruitment were used to identify patterns of HbA1c trends. Outcomes were recorded through linkage with the National Disease Registry. The median follow-up for longitudinal trends in HbA1c was 4.1years and for outcomes was between 7.0 and 8.3years. HbA1c patterns were identified using latent class growth modeling, and associations with outcomes were analyzed using Cox proportional hazards models.
Four distinct HbA1c patterns were observed; "low-stable" (72·2%), "moderate-stable" (22·0%), "moderate-increase" (2·9%), and "high-decrease" (2·8%). The risk of comorbidities and death was significantly higher in moderate-increase and high-decrease groups compared to the low-stable group; the hazard ratios for stroke, ESRD, and death for moderate increase group were 3.22 (95%CI 1.27-8.15), 4.76 (95%CI 1.92-11.83), and 1.88 (95%CI 1.15-3.07), respectively, and for high-decrease group were 2.16 (95%CI 1.02-4.57), 3.05 (95%CI 1.54-6.07), and 2.79 (95%CI 1.97-3.95), respectively. Individuals in the moderate-increase group were significantly younger, with longer diabetes duration, and greater proportions of Malays and Indians.
Deteriorating HbA1c pattern and extremely high initial HbA1c are associated with increased risk of long-term comorbidities and death. Therapeutic interventions to alter longitudinal HbA1c trends may be helpful in reducing this risk.
本研究调查了多民族亚洲糖尿病患者队列中糖化血红蛋白（HbA1c）的纵向变化趋势，以及这些趋势与未来急性心肌梗死（AMI）、中风、终末期肾衰竭（ESRD）风险和全因死亡率之间的关联。
纳入新加坡6079名2型糖尿病患者。使用招募前五年的HbA1c测量值来确定HbA1c趋势模式。通过与国家疾病登记处的数据关联记录结局。HbA1c纵向趋势的中位随访时间为4.1年，结局的随访时间在7.0至8.3年之间。使用潜在类别增长模型确定HbA1c模式，并使用Cox比例风险模型分析与结局的关联。
观察到四种不同的HbA1c模式；“低稳定”（72.2%）、“中度稳定”（22.0%）、“中度上升”（2.9%）和“高度下降”（2.8%）。与低稳定组相比，中度上升组和高度下降组的合并症和死亡风险显著更高；中度上升组中风、ESRD和死亡的风险比分别为3.22（95%CI 1.27 - 8.15）、4.76（95%CI 1.92 - 11.83）和1.88（95%CI 1.15 - 3.07），高度下降组分别为2.16（95%CI 1.02 - 4.57）、3.05（95%CI 1.54 - 6.07）和2.79（95%CI 1.97 - 3.95）。中度上升组的个体明显更年轻，糖尿病病程更长，马来人和印度人的比例更高。
HbA1c模式恶化和初始HbA1c极高与长期合并症和死亡风险增加有关。改变HbA1c纵向趋势的治疗干预可能有助于降低这种风险。
