Type 2 diabetes is a heterogeneous disorder characterized by insulin resistance and/or defective insulin secretion (1,2). The mechanisms underlying insulin resistance in type 2 diabetes are not fully understood, but the numerous studies in nondiabetic populations have addressed the importance of upper body fat distribution. A study by Vague (3) was the first to show that upper body fat distribution has particularly adverse metabolic abnormalities. Higher concentrations of plasma glucose, insulin, and triglycerides have been shown to be associated with increasing abdominal body fat in nondiabetic subjects (4,5,6,7). However, in diabetic patients, very little has been reported on the associations between insulin resistance and body fat distribution.
Banerji et al. (8) recently disclosed that visceral but not subcutaneous abdominal fat volume is associated with insulin resistance in black subjects with type 2 diabetes. In contrast, Abate et al. (9) recently demonstrated that subcutaneous but not intraperitoneal nor retroperitoneal fat volume is associated with insulin resistance in non-Hispanic whites with type 2 diabetes. The reason for the discrepant result between the two groups is unknown, but it can be speculated that there may be an ethnic difference in the relationship between insulin resistance and body fat distribution in type 2 diabetes.
Nonobese Japanese type 2 diabetic patients are unique in that they are divided into two variants: those with insulin resistance and those with normal insulin sensitivity (10,11,12). However, to the best of our knowledge, the relationship between insulin resistance and body fat distribution has not been investigated in nonobese Japanese type 2 diabetic populations. Thus, the aim of the present study was to investigate the relationships among insulin resistance and subcutaneous or visceral fat area in nonobese Japanese type 2 diabetic patients.
A total of 75 nonobese Japanese type 2 diabetic patients at Kansai-Denryoku Hospital were recruited for the present study. Type 2 diabetes was diagnosed based on the criteria of the World Health Organization (13). Of the 75 patients, 36 were taking sulfonylureas; the rest were treated with diet alone. None of the patients has received insulin therapy. All subjects ingested ≥150 g of carbohydrate for the 3 days preceding the study. None of the subjects had significant renal, hepatic, or cardiovascular disease. Of the 75 patients, 18 had hypertension, and these patients were treated with ACE inhibitors (n = 10), calcium channel blockers (n = 6), or both (n = 2). They did not consume alcohol or perform heavy exercise for ≥1 week before the study.
Blood was drawn in the morning after a 12-h fast. Plasma glucose was measured with the glucose oxidase method, and serum insulin was measured using a two-site immunoradiometric assay (Insulin Riabead II; Dainabot, Osaka, Japan). Triglycerides, total cholesterol, and HDL cholesterol were also measured. The LDL cholesterol level was calculated using the Friedewald formula (14).
The estimate of insulin resistance by the homeostasis model assessment (HOMA-IR) was calculated with the following formula: fasting serum insulin (μU/ml) × fasting plasma glucose (mmol/l)/22.5 (15). One might argue that the use of sulfonylureas in patients with diabetes might significantly affect the estimate of HOMA-IR, because these drugs are known to decrease fasting plasma glucose without substantially changing fasting plasma insulin (16). However, this seems unlikely, because Bonora et al. (17) and Emoto et al. (18) confirmed that in the validation studies of HOMA, the correlation of insulin sensitivity estimated by HOMA and that measured by the glucose clamp was not substantially different in diet-treated and sulfonylurea-treated type 2 diabetes. Therefore, we estimated HOMA-IR in diet-treated and sulfonylurea-treated diabetic patients.
All subjects underwent computed tomography (TSX-012A, X-Vigor; Toshiba, Osaka, Japan) to measure cross-sectional abdominal subcutaneous and visceral fat areas, as described previously (6,19). Briefly, the subjects were examined in the supine position, and computed tomography scans were performed at the umbilical levels. Adipose tissue areas were determined using commercially available software (Fat Scan; N2 System Corporation, Osaka, Japan) (20). Subcutaneous and visceral borders were defined using a manual cursor.
Data are presented as means ± SEM. Statistical analyses were conducted using the StatView 5 system (Statview, Berkeley, CA). Simple (Spearman’s rank) correlation coefficient and stepwise multiple regression analyses were used to examine the relationships between insulin resistance and subcutaneous or visceral abdominal fat area, BMI, or the measures of variables, including triglycerides. P < 0.05 was considered as significant. In multivariate analysis, F values ≥4 were considered significant.
The subjects studied were all Japanese type 2 diabetic patients (49 men and 26 women) with an age range of 39–83 years (61.8 ± 1.1) and a BMI of 16.2–26.8 kg/m2 (22.2 ± 0.3). They all were nonobese (21). The mean fasting plasma glucose was 144 ± 3 mg/dl, and the HbA1c level was 6.9 ± 0.1%. The mean fasting plasma insulin level was 6.2 ± 0.4 μU/ml. Serum triglycerides, total and HDL cholesterol levels were 108 ± 7 mg/dl, 197 ± 4 mg/dl, and 58 ± 2 mg/dl, respectively. The mean serum LDL cholesterol level was 117 ± 3 mg/dl.
There was a wide variation in insulin resistance values calculated from HOMA-IR in our diabetic patients (range 0.64–9.67, mean 2.21 ± 0.16). Of the 75 patients, 23 (31%) had HOMA-IR values >2.5, indicating that they were insulin-resistant (22). Similarly, there was also a wide variation in subcutaneous and visceral abdominal fat areas. Mean subcutaneous and visceral abdominal fat areas were 127.4 ± 6.3 cm2 (range 27.6–267.4) and 80.4 ± 4.1 cm2 (12.2–179.1), respectively.
Insulin resistance calculated from HOMA-IR was positively correlated to subcutaneous (r = 0.583, P < 0.001) and visceral (r = 0.490, P < 0.001) fat areas in our diabetic patients. Furthermore, insulin resistance was positively correlated to BMI (r = 0.458, P < 0.001) and levels of HbA1c (r = 0.253, P = 0.026) and serum triglycerides (r = 0.419, P < 0.001). In contrast, insulin resistance was negatively correlated to serum HDL cholesterol levels (r = –0.334, P = 0.003). However, there was no relationship between insulin resistance and measures of other variables, including total cholesterol.
Multiple regression analyses showed that insulin resistance was predicted by the area of subcutaneous (F = 6.7) and visceral (F = 5.6) abdominal fat and the level of serum triglycerides (F = 10.0), which explained 39.5% of the variability of insulin resistance in our nonobese Japanese type 2 diabetic patients. However, BMI, HbA1c, and HDL and LDL cholesterol levels were not independently associated with insulin resistance in our patients.
Nonobese Japanese type 2 diabetic patients are divided into two variants: those with insulin resistance and those with normal insulin sensitivity (10,11,12). We recently demonstrated that the patients with insulin resistance had significantly higher triglyceride levels, higher RLP cholesterol levels, and lower HDL cholesterol levels compared with those with normal insulin sensitivity (22,23,24). Thus, dyslipidemia is postulated to be associated with insulin resistance in nonobese Japanese type 2 diabetic patients. This idea is supported by our recent study showing that bezafibrate, a triglyceride-lowering drug, reduces insulin resistance and plasma glucose levels without affecting BMI levels in nonobese Japanese type 2 diabetic patients (25). We recently demonstrated that short-term physical training also decreased serum triglycerides, insulin resistance, and glucose levels without affecting BMI levels in nonobese Japanese type 2 diabetic patients (26).
Abdominal fat volume has been implicated in the pathogenesis of insulin resistance in type 2 diabetic patients. However, reports on the relationship between insulin resistance and abdominal fat volume are inconsistent. Banerji et al. (8) disclosed that visceral but not subcutaneous abdominal fat volume is associated with insulin resistance in black populations with type 2 diabetes. Abate et al. (9) recently demonstrated that subcutaneous but not intraperitoneal or retroperitoneal fat volume is associated with insulin resistance in non-Hispanic whites with type 2 diabetes. In the present study, we first documented that insulin resistance is independently associated with subcutaneous and visceral abdominal fat areas in nonobese Japanese type 2 diabetic patients. Thus, racial differences seem to exist in the relationship between insulin resistance and abdominal fat area in type 2 diabetic patients.
In summary, although our present study was performed with a limited number of patients (n = 75), it can be suggested that both subcutaneous and visceral abdominal fat areas seem to be independently associated with insulin resistance in nonobese Japanese type 2 diabetic patients. A larger population study should be undertaken to clarify the relationship between insulin resistance and regional abdominal fat area in nonobese Japanese type 2 diabetic patients.
We are very grateful to Dr. Hakaru Suzuki, Dr. Katsuhisa Oonishi, Dr. Atsunori Kawakami, Dr. Kazuyuki Fukui, and Dr. Issyuu Kimura from the Department of Radiology, Kansai-Denryoku Hospital, for their help in our research. We also thank Dr. Shoichiro Nagasaka from the Division of Endocrinology and Metabolism, Jichi Medical School; Dr. Kentaro Doi from the Graduate School of Medicine, Kyoto University; and Dr. Kumpei Tokuyama from the Laboratory of Biochemistry of Exercise and Nutrition, Institute of Health and Sports Science, Tsukuba University, for encouraging this study.
Address correspondence and reprint requests to Ataru Taniguchi, M.D., First Department of Internal Medicine, Kansai-Denryoku Hospital, 2-1-7 Fukushima, Fukushima-ku, Osaka-city, Osaka 553-0003 Japan. E-mail: email@example.com.