BACKGROUND: Previous epidemiological studies have shown that Korean type 2 diabetic patients were mainly non-obese compared to their Western counterparts. This retrospective study was performed to find the percentage of obese type 2 diabetic patients, and its temporal changes, using the newly proposed Asian criteria for the diagnosis of obesity. In addition, our results were compared with Caucasian data. METHODS: The subjects of our study were all newly diagnosed type 2 diabetic patients; 157 for 1991, 176 for 1996 and 275 for 2001. The all the study subjects were aged over 30 years. They all had visited the Mary Knoll General Hospital for the first time, and were diagnosed with type 2 diabetes mellitus within 1 year. The maximum BMI (Body Mass Index) was calculated from the patients heaviest life-time body weight, and their current BMI from the values obtained at their first visit to our institution. The delta BMI (deltaBMI) was calculated by subtracting the current BMI from the maximum BMI the HbA1c value at the time of the first visit was also recorded. Obesity was defined as a body mass index greater than 25kg/m2. RESULTS: The mean values of maximum BMI were 25.7+/-4.5, 26.4+/-4.3 and 25.9+/-6.0 kg/m2 for the years of 1991, 1996 and 2001, respectively. The mean values of the current BMI were 24.0+/-3.0, 24.2+/-3.0 and 24.8+/-3.6 kg/m2 for the years of 1991, 1996 and 2001, respectively. None of these values showed statistically significant differences. The percentages of obese type 2 diabetic patients in 1991, 1996 and 2001 were 64.3, 69.0 and 66.9%, according to their maximum BMI, respectively. The percentages of obese type 2 diabetic patients, from their current BMI, were 31.8, 39.8 and 43.6% in 1991, 1996 and 2001, respectively, and these values showed statistically significant increases over time (p=0.016). The mean value of the delta BMI was significantly lower in 2001 compared with 1996, and it was positively correlated with the HbA1c at the time of the first visit (p< 0.01). CONCLUSION: The percentage of obese type 2 diabetic patients at the time of the maximum body weight was 60 to 70%, but the percentage at the time of diagnosis had decreased to 30 to 40%. The percentage of obese type 2 diabetic patients at the time of diagnosis significantly increased over time. The mean BMI value of the Korean type 2 diabetic patients was lower than that of Caucasians, but the percentage of obese type 2 diabetic patients and its temporal trends were similar to those of Caucasians. Our study shows that Korean type 2 diabetic patients are as obese as Caucasians when they meet their own diagnostic criteria for obesity.