Nagi D K, Knowler W C, Charles M A, Liu Q Z, Hanson R L, McCance D R, Pettitt D J, Bennett P H
Diabetes and Arthritis Epidemiology Section, National Institute of Diabetes and Digestive and Kidney Diseases, Phoenix, Arizona, USA.
Diabetologia. 1995 Feb;38(2):187-92. doi: 10.1007/BF00400093.
Risk factors predicting deterioration to diabetes mellitus were examined in 181 subjects with impaired glucose tolerance. Fifty-seven subjects had impaired glucose tolerance on one occasion followed by normal glucose tolerance at a repeat oral glucose tolerance test, and 124 subjects had impaired glucose tolerance on two successive oral glucose tolerance tests. Subjects were followed for a median period of 5.0 years (range 1.0-17.2). The age- and sex-adjusted cumulative incidence of diabetes at 10 years of follow-up was higher in subjects who had impaired glucose tolerance on both tests (70%) than in those whose glucose tolerance was normal at the repeat test (53%), [rate ratio (RR) = 1.6, 95% confidence intervals (CI) = 1.0-2.5]. Proportional hazards analyses were used to identify baseline risk factors (measured at the repeat oral glucose tolerance test) for subsequent diabetes, and incidence rate ratios were calculated for the 90th percentile compared with the 10th percentile of each continuous variable for the whole group. In all subjects, in separate models, higher body mass index [RR = 2.0, 95% CI = 2.2-9.9], high fasting serum insulin concentrations [RR = 2.4, 95% CI = 1.4-4.2], and low early insulin response [RR = 0.5, 95% CI = 0.3-0.8] 30 min after a glucose load were significant predictors for deterioration to diabetes. In a multivariate analysis which controlled for age and sex, 120-min post-load glucose, fasting insulin and late insulin response predicted diabetes.(ABSTRACT TRUNCATED AT 250 WORDS)
在181例糖耐量受损的受试者中,对预测发展为糖尿病的危险因素进行了研究。57例受试者一次糖耐量受损,重复口服葡萄糖耐量试验时糖耐量正常;124例受试者连续两次口服葡萄糖耐量试验均显示糖耐量受损。对受试者进行了为期5.0年的中位随访(范围1.0 - 17.2年)。随访10年时,两次试验均糖耐量受损的受试者经年龄和性别调整后的糖尿病累积发病率(70%)高于重复试验时糖耐量正常的受试者(53%),[率比(RR)= 1.6,95%置信区间(CI)= 1.0 - 2.5]。采用比例风险分析来确定后续发生糖尿病的基线危险因素(在重复口服葡萄糖耐量试验时测量),并计算整个组中每个连续变量第90百分位数与第10百分位数相比的发病率比。在所有受试者中,在单独的模型中,较高的体重指数[RR = 2.0,95% CI = 2.2 - 9.9]、空腹血清胰岛素浓度高[RR = 2.4,95% CI = 1.4 - 4.2]以及葡萄糖负荷后30分钟时早期胰岛素反应低[RR = 0.5,95% CI = 0.3 - 0.8]是发展为糖尿病的显著预测因素。在控制了年龄和性别的多变量分析中,负荷后120分钟血糖水平、空腹胰岛素和晚期胰岛素反应可预测糖尿病。(摘要截断于250字)