Impact of waist circumference versus adiponectin level on subclinical atherosclerosis: a cross-sectional analysis in a sample from the general population.

Abstract

OBJECTIVE: Waist circumference is a clinical marker of obesity and an established risk factor for cardiovascular (CV) disease. Adiponectin, an adipocyte-derived hormone and new biomarker of obesity, was recently proposed as the missing link between obesity and increased cardiovascular risk. We evaluated waist and adiponectin in a middle-aged population-based cohort to compare the impact of both obesity-markers on subclinical atherosclerosis, in relation to other CV risk factors. DESIGN, SETTING & SUBJECTS: Seven noninvasive measurements of atherosclerosis (NIMA), as surrogate markers of (subclinical) atherosclerosis, were determined in 1517 participants of the Nijmegen Biomedical Study, aged 50-70 years, who were drawn from the Dutch community. RESULTS: Both men and women with a high waist (M >104 cm; F >95 cm) showed increased pulse wave velocity (PWV) (M: +9.4%; F: +8.3%) and thicker intima-media thickness (IMT) (M: +7.3%; F: +4.3%) and women also showed increased plaque thickness (+16.6%). After adjustment for other CV risk factors both men and women showed increased IMT (M: +4.8%; F: +2.8%) and men also showed increased PWV (+9.6%). Both men and women with a low adiponectin level (M <2.2 mg L(-1); F <3.5 mg L(-1)) showed a decreased ankle-brachial index after exercise (M: -9.5%; F: -3.9%) and increased IMT (M: +3.7%; F: +3.6%) and women also showed increased PWV (+6.8%), but after adjustment for other CV risk factors low adiponectin level was no longer associated with deteriorated outcomes of NIMA. CONCLUSIONS: Waist circumference showed independent associations with noninvasive measurements of subclinical atherosclerosis, whereas the association of adiponectin level with subclinical atherosclerosis was not independent of other CV risk factors. Prospective studies are needed to elucidate, if the atherogenic effect of a low adiponectin level is mediated by other CV risk factors and not by low adiponectin level intrinsically.