Although there are studies of different ethnic groups that have reported the differential risk for CHD and examined the contribution of the metabolic syndrome to CVD/CHD risk within the ethnic groups, to our knowledge the work of our group provides the only available data of the role of the metabolic syndrome in the observed ethnic difference in CHD risk (Table 2.2).
Recently we have been able to examine the association of the metabolic syndrome with CHD mortality in Europeans and South Asians in the UK in a prospective follow-up of the Southall Study. Among 2935 men aged 40-69 years at baseline, at a median 14-year follow-up there were 399 total deaths, with 161 CHD deaths (91 in South Asians, 70 in Europeans) (Forouhi etal., 2005). The metabolic syndrome was significantly associated with CHD mortality in both ethnic groups in analyses adjusted for age, smoking and total cholesterol [ATP-III hazard ratio 1.8 (95 per cent CI 1.0-3.0, P=0.035) in Europeans and 2.6 (1.7-4.1, P<0.001) in South Asians; WHO hazard ratio 1.7 (1.0-2.9, P=0.044) in Europeans and 1.9 (1.3-3.0, P=0.003) in South Asians]. Of significant note, however, is the finding that the metabolic syndrome (either definition) did not account for the excess CHD mortality in South Asians compared with Europeans (hazard ratio 1.8 for CHD mortality in South Asian versus European men in analyses adjusted for age, smoking, cholesterol, socio-economic status and either ATP-III or WHO-defined metabolic syndrome). Even when individual components of the metabolic syndrome as continuous variables were entered into the multivariate Cox regression models, the significantly higher CHD mortality rate in South Asians compared with Europeans persisted. This was the case even when analyses were adjusted for prevalent CHD at baseline, or when those with significant co-morbidity at baseline were excluded from analysis. Limitations of this study included: only men were studied; any potential ethnic biases or misclassifications in assigning CHD death could not be studied; and ethnic differences in healthcare access or treatment need were not measured, although socio-economic status (years of education and occupational social class) as a proxy marker was measured and included in the analyses. Strengths, however, included prospective study design with near-complete follow-up at 99.5 per cent and the ability to examine a comprehensive battery of risk factors for CHD, including individual components and composite definitions of the metabolic syndrome.
The failure of the metabolic syndrome to account for the excess CHD mortality in South Asians was contrary to expectations from the cross-sectional analysis of the same cohort, where prevalent CHD was the outcome (McKeigue etal., 1993). A decade ago McKeigue etal. reported that the age-, smoking- and cholesterol-adjusted odds ratio for the electrocardiographic major Q waves in South Asians compared with Europeans was 2.4 (95 per cent CI 1.5-3.8). Adjusting for glucose intolerance and hyperinsulinaemia reduced this ratio to 1.5 (0.9-2.5), leading them to conclude that insulin resistance underlies the high coronary risk in South Asians (McKeigue etal., 1993). Explanations for this difference in findings include:
1. The deaths ascribed to CHD may have been over- or underestimated by ethnic group and, though unlikely, could potentially be checked by a validation study of cause of death.
2. Current metabolic syndrome criteria may not fully 'capture' the vascular risk associated with insulin resistance in South Asians, and by extrapolation may not do so in other ethnic groups. In this respect it is noteworthy that others have recently suggested that the metabolic syndrome as defined by ATP-III criteria has low sensitivity (20-50 per cent) for identifying insulin-resistant individuals (Cheal etal., 2004; Ciao etal., 2004). We were surprised, however, to note that adjustment for individual criteria of the metabolic syndrome, inclusive of fasting triglyceride, fasting glucose, waist circumference and fasting insulin as continuous variables, also did not attenuate the ethnic difference in CHD mortality.
3. The current cut-off values for defining the metabolic syndrome are based on European populations and may not be appropriate for South Asians, where lower thresholds may apply for some factors such as central obesity.
4. Other 'novel' risk factors linked to insulin resistance, such as adipocytokines and pro-inflammatory factors, might be important in South Asians but were not measured in this study.
Was this article helpful?