Waist-hip-ratio a better indicator of risk of myocardial infarction than BMI in a Mediterranean Southern European population

Aims: To analyse measures of adiposity and risk of myocardial infarction (MI) in a Mediterranean European Population. Methods: A total of 308 MI cases and 437 controls from the Maltese Acute Myocardial Infarction (MAMI) study had hip and waist circumferences measured per WHO guidelines. BMI and Waist-Hip Ratio (WHR) were analysed using WHO cutoffs. Odds ratios adjusted for common risk factors (adjOR) were calculated. Differences between groups were evaluated using the Kruskal-Wallis test. Results: A similar median BMI between male cases and controls (29.2, 28.7, p =0.05) and a similar prevalence of obesity (41.4%, 38.2%) resulted in an AdjOR of 1.2 (95%CI 0.6-2.4). AdjOR was similar in females despite a higher prevalence of obesity in cases (52.9%, N=37) than in controls (32.8%, N=45). Waist circumference (0.94, 1.00, p<0.001) and WHR (1.00, 0.95, p<0.001) were higher in male cases than controls giving an adjOR of 2.6 (95% CI 1.7-4.1) and 6.4 (95% CI, 3.0-14.0) respectively. 95.6% (N=237) of cases had a high Waist-Hip Ratio compared to 73.5% (n=228) of controls. Females showed similar results. Amongst all controls, 23.5% (N=20) of normal BMI controls and 59% (N=111) of overweight controls had a high WHR. Conversely, 21.3% (N=35) of obese individuals had a normal WHR. A stratified analysis of WHR and BMI shows elevated AdjOR only in groups with high WHR. Conclusion: BMI had no predictive value while WHR had a high predictive value for MI in the Maltese Population. BMI miscategorises high WHR individuals as being normal or overweight underestimating their risk.