We read with interest the recent editorial by Sobel (1), wherein it is asserted that “intensive treatment of type 1 diabetes appears to increase central obesity.” We agree with Sobel that “the insulin resistance underlying type 2 diabetes and frequently manifested in those with type 1 diabetes may be the most powerful determinant of coronary disease” (1), but argue that insulin resistance in type 1 diabetes is unrelated to increased central abdominal adiposity.
Attention should be directed to our recent study (2), which examined the relationships between insulin sensitivity (glucose infusion rate [GIR] measured during euglycemic-hyperinsulinemic clamp), abdominal fat, lipid levels, blood pressure, and androgens in 10 premenopausal women with type 1 diabetes (HbA1c 8.1 ± 1% and diabetes duration 24 ± 10 years) and 10 nondiabetic BMI-matched control subjects. We found that GIR was lower in subjects with type 1 diabetes than in control subjects (49.3 ± 14.8 vs. 73.2 ± 21.6 μmol · min−1 · kg fat-free mass−1 respectively, P = 0.01), indicating greater insulin resistance in the former (2). However, we found no difference between control and type 1 diabetic subjects in central abdominal adiposity (measured directly by dual-energy X-ray absorptiometry) and intra-abdominal fat (measured by four-slice computed tomography) (2). Furthermore, unlike control subjects, we found that GIR was unrelated to abdominal obesity in type 1 diabetes (2). The finding that women with type 1 diabetes do not have greater central abdominal fat than nondiabetic control subjects confirms one of the few reports (3) that directly quantified abdominal adiposity in type 1 diabetes. Although a previous study of men found that waist-to-hip ratios were greater in men with type 1 diabetes than control subjects (4), this study relied on indirect and imperfect anthropometric surrogates (5).
We echo Sobel’s conclusion that targeting insulin resistance may reduce excess coronary risk in type 1 diabetes. Although insulin “sensitizers” traditionally used in the treatment of type 2 diabetes, such as metformin, have been reported to improve insulin action in small studies of type 1 diabetes (6), it is not known whether this improvement translates into lower rates of coronary artery disease and whether newer agents, the thiazolidinediones, have a similar effect. We believe this to be an important question, especially due to the excess rates of coronary disease associated with type 1 diabetes and failure of traditional vascular risk factors to fully explain this excess risk.
