Martinell Mats, Dorkhan Mozhgan, Stålhammar Jan, Storm Petter, Groop Leif, Gustavsson Carin
Department of Public Health and Caring Sciences, Uppsala University, Uppsala, Sweden.
Department of Clinical Sciences in Malmö, Lund University, Uppsala, Sweden.
J Diabetes Complications. 2016 Nov-Dec;30(8):1456-1461. doi: 10.1016/j.jdiacomp.2016.08.009. Epub 2016 Aug 14.
To study prevalence of diabetic retinopathy (DR) at diagnosis (DRAD) and to estimate contributing risk by sociodemographic, cardiovascular and metabolic characteristics present in patients recently diagnosed with type 2 diabetes (T2D) or latent autoimmune diabetes in the adult (LADA).
Patients (n=2174) recently diagnosed T2D (93%) or LADA (7%) were included upon arrival for their baseline DR screening. Fundus photographs of 4902 eyes were graded by a senior ophthalmologist according to the International Diabetic Retinopathy Disease Severity Scale. Official registers held by Statistics Sweden provided sociodemographic variables. The National Patient Register and Swedish Prescribed Drug Register were used to assess cardiovascular risk. Beta cell function (HOMA2%b) and insulin sensitivity (HOMA2%s) were estimated from fasting (f) C-Peptide using the homeostasis model assessment (HOMA) 2 calculator. Odds ratios (OR) for DRAD were estimated using generalized estimating equation models.
The prevalence of DRAD was 12% (7% mild and 5% moderate) and of diabetic macular edema it was 11% (all within vascular arch). The prevalence did not significantly differ between T2D and LADA. Due to sample size, the regression analysis of LADA patients did not yield any significant estimates. In T2D low educational level (≤9years) increased risk for DRAD by 44% (OR 1.44; 95% CI 1.07-1.93) and <50% beta-cell function adjusted for HbA1c and insulin sensitivity at diagnosis increased the risk by 77% (OR 1.77; 95% CI 1.28-2.44). For every unit increase in BMI, risk for DRAD decreased by 3% (OR 0.97; 95% CI 0.95-0.99).
DRAD prevalence in patients recently diagnosed with T2D or is 12%. Low educational level and low beta cell function at diagnosis are risk factors for DRAD. Estimation of beta cell function from (f)C-Peptide and (f)P-Glucose may be a valuable tool in identifying patients at risk for DRAD.
研究糖尿病视网膜病变（DR）在诊断时的患病率（DRAD），并通过近期诊断为2型糖尿病（T2D）或成人隐匿性自身免疫性糖尿病（LADA）患者的社会人口统计学、心血管和代谢特征来评估相关风险因素。
近期诊断为T2D（93%）或LADA（7%）的患者（n = 2174）在进行基线DR筛查时被纳入研究。4902只眼睛的眼底照片由一位资深眼科医生根据国际糖尿病视网膜病变疾病严重程度量表进行分级。瑞典统计局持有的官方登记册提供社会人口统计学变量。国家患者登记册和瑞典处方药登记册用于评估心血管风险。使用稳态模型评估（HOMA）2计算器，根据空腹（f）C肽估计β细胞功能（HOMA2%b）和胰岛素敏感性（HOMA2%s）。使用广义估计方程模型估计DRAD的比值比（OR）。
DRAD的患病率为12%（轻度7%，中度5%），糖尿病性黄斑水肿的患病率为11%（均在血管弓内）。T2D和LADA之间的患病率没有显著差异。由于样本量的原因，LADA患者的回归分析没有得出任何显著的估计值。在T2D患者中，低教育水平（≤9年）使DRAD风险增加44%（OR 1.44；95% CI 1.07 - 1.93），诊断时调整糖化血红蛋白和胰岛素敏感性后β细胞功能<50%使风险增加77%（OR 1.77；95% CI 1.28 - 2.44）。BMI每增加一个单位，DRAD风险降低3%（OR 0.97；95% CI 0.95 - 0.99）。
近期诊断为T2D或LADA的患者中DRAD患病率为12%。诊断时低教育水平和低β细胞功能是DRAD的风险因素。根据（f）C肽和（f）血糖估计β细胞功能可能是识别DRAD风险患者的有价值工具。
