Increased Risk of Type 2 Diabetes Despite Same Degree of Adiposity in Different Racial Groups

As indigenous people of the South Pacific have rates of type 2 diabetes and obesity among the highest in the world (1,2), diabetes risk reduction strategies include identification of high-risk individuals. Different cutoffs to define overweight and obesity have been suggested for some indigenous people because they have more lean mass and less adipose tissue than Europeans with comparable BMIs (3). However, cutoffs for BMI should be based on risk of comorbidities associated with a given BMI, not simply with lean body mass. To further assist the process of defining appropriate BMI levels for South Pacific people, we have investigated insulin sensitivity in New Zealand women of Maori and European descent who have similar levels of BMI, fat, and lean mass.

A total of 88 European and 23 Maori women consented to participate in an ethically approved study to measure body composition (using dual-energy absorptiometry) and insulin sensitivity (using the euglycemic insulin clamp). Smoking history, weight, height, blood pressure, BMI, and waist circumference were recorded, and fasting plasma glucose and insulin levels were measured. The methodologies are described elsewhere (4). The significance of differences between the two groups was tested by regression analysis.

The Maori women were on average 7 years (95% CI 3–11) younger, had a higher prevalence of smoking (39 vs. 7%), and had higher fasting glucose levels (difference = 0.7 [95% CI 0.3–1.1]) than the European women. The fasting insulin level was 1.5 (1.1–2.0) times higher in Maori women compared with European women. There was no significant difference between Maori and Europeans for weight, BMI, waist circumference, blood pressure, total fat, truncal fat, or lean mass. There was no evidence of reduced adiposity or increased lean mass for a given BMI in Maori women.

Maori had lower levels of insulin sensitivity than Europeans, despite similar BMI levels and total and truncal fat levels. The difference was 1.8 G · mIU⁻¹ · l⁻¹ (0.3–3.3) (expressed for fat-free mass, where G is glucose infused during the euglycemic in mg · kg⁻¹ · min⁻¹). After adjusting for age, BMI, fasting glucose levels, and smoking, the difference in insulin sensitivity was 2.0 G · mIU⁻¹ · l⁻¹ (0.5–3.6). The relationship between insulin resistance and BMI for Maori and European women is illustrated in Fig. 1.

The early report suggesting that Maori and Pacific people in New Zealand have more lean mass and less adipose tissue than Europeans with comparable BMIs (5) did not describe measures of comorbidity. Furthermore, their European group was significantly lighter than the Maori and Samoan groups; therefore, comparisons of lean mass at higher BMI levels may not be valid. The results of our study suggest that for any given level of BMI or total or truncal fat, Maori are more likely than Europeans to have insulin resistance and are therefore at greater risk of type 2 diabetes and cardiovascular disease. After adjusting for age, smoking, and glucose levels, as well as BMI, the difference in insulin sensitivity remained significant. Therefore, there is no evidence to support the use of higher cutoffs for BMI in Maori and, perhaps, other indigenous people.