Metabolically healthy obese (MHO) phenotype is associated with lower cardiovascular risk, but there is no standard criteria for MHO diagnostic. The aim of our study was to compare MHO prevalence according to different criteria in sample of Saint-Petersburg inhabitants.Design and method:
As a part of all-Russian epidemiology survey ESSE-RF random sampling of 1600 Saint-Petersburg inhabitants (25–65 years) was involved. Anthropometry (weight, height with body-mass index (BMI) calculation), waist circumference (WC)), blood pressure (BP) measurement and fasting blood-tests: glucose, lipids, CRP, insulin (Abbott Architect 8000 (USA)) and HOMA IR were performed. MHO criteria were used in subjects with obesity (BMI >=30 kg/m2):Design and method:
We’ve assumed to use 3 types of CRP and HOMA IR cut-offs in Wildman criteria: 90th percentile in normal-weight (BMI<25 kg/m2) subjects (variant 1), in whole cohort (variant 2) and used criteria from Oliveros et al. study (variant 3 criteria)Results:
Obesity was revealed in 430 subjects.Results:
Table 1. MHO prevalence according to different criteriaResults:
Kappa index was maximal between variant 2 and Meigs criteria (kappa=0,7, p < 0,001) and minimal between variant 1 and Meigs criteria (kappa=0,4, p < 0,001).Conclusions:
MHO prevalence was decreased up to 3 times including CRP and HOMA IR to MHO criteria. Gender differences in MHO phenotype prevalence were found only with applying of the strictest cut-off for CRP and HOMA IR. Using of cut-off of whole cohort data for additional components of metabolic health (CRP and HOMA IR) was associated with best agreement with Meigs criteria.