Association of obesity with cardiovascular outcomes in patients with type 2 diabetes and cardiovascular disease: Insights from TECOS.
Pagidipati NJ., Zheng Y., Green JB., McGuire DK., Mentz RJ., Shah S., Aschner P., Delibasi T., Rodbard HW., Westerhout CM., Holman RR., Peterson ED., TECOS Study Group None.
BACKGROUND: Obesity is a risk factor for type 2 diabetes (T2D) and cardiovascular disease (CVD). Whether obesity affects outcomes among those with T2D and atherosclerotic CVD (ASCVD) remains uncertain. Our objective was to investigate the relationship between body mass index (BMI) and ASCVD outcomes among TECOS participants with T2D and ASCVD. METHODS: BMI categories were defined as underweight/normal weight (BMI <25 kg/m2), overweight (25-29.9 kg/m2), obese class I (30-34.9 kg/m2), obese class II (35-39.9 kg/m2), and obese class III (≥ 40 kg/m2). Asian-specific BMI categories were applied to Asian participants. Kaplan-Meier survival analysis and Cox proportional hazards models were used to examine associations between baseline BMI and a composite CV outcome (CV death, nonfatal myocardial infarction, nonfatal stroke, or hospitalization for unstable angina). RESULTS: For 14,534 TECOS patients with available BMI, mean age was 65.5 years; 29.3% were female, 32.0% non-White, and 23.1% insulin-treated, with median 3 years' follow-up. At baseline, 11.6% (n = 1686) were underweight/normal weight, 38.1% (n = 5532) overweight, 32.2% (n = 4683) obese class I, 12.4% (n = 1806) obese class II, and 5.7% (n = 827) obese class III. The composite CV outcome occurred in 11.4% (n = 1663) of participants; the outcome risk was lower, compared with under/normal weight, in overweight (HR 0.83, 95% CI 0.71-0.98) and obese class I (HR 0.79, 95% CI 0.67-0.93) individuals. Obesity was not associated with worse glycemic control. CONCLUSIONS: The majority of TECOS participants with ASCVD and T2D were overweight or obese, yet overweight or obese class I individuals had lower CV risk than those who were under/normal weight. These results suggest the presence of an obesity paradox, but this paradox may reflect an epidemiological artifact rather than a true negative association between normal weight and clinical outcomes.