The relationship between macrovascular disease and blood glucose control in long-term follow-up of Type 2 (non-insulin-dependent) diabetes mellitus is difficult to study because of the gradual rise in fasting plasma glucose due to ongoing beta-cell failure. We used time-dependent covariates in Cox's proportional hazards model to allow variables measured annually during a 10-year prospective follow-up to be related to risk of myocardial infarction or cerebrovascular accident. Data for 432 newly diagnosed diabetic patients were available, 112 of whom suffered myocardial infarction (fatal or non-fatal). Analysis of baseline measurements only gave relative hazards (95% CL) of 1.04 (0.99, 1.09) per mmol l(-1) increase in fasting plasma glucose, 1.43 (1.12, 1.83) per decade increase in age and 1.07 (0.98, 1.17) per 10 % increase in percentage of ideal weight. Analysis incorporating ongoing measurements gave corresponding figures of 1.07 (1.02, 1.12) for fasting plasma glucose, 1.64 (1.23, 2.20) for age and 1.06 (0.95, 1.18) for percentage of ideal weight. The risk of myocardial infarction while on insulin treatment 1.09 (0.58, 2.06) or oral agents 1.41 (0.86, 2.31) was not significantly elevated relative to dietary treatment. Baseline smoking status, systolic blood pressure, and previous myocardial infarction were also significant predictors of myocardial infarction. Similar relationships were found for cerebrovascular accident and total mortality. Increasing fasting plasma glucose is a significant independent predictor of macrovascular disease in diabetes.
No takes yet. Share an insight, caveat, or question.
Hadden et al. (1997) studied this question.