Background: Black populations are disproportionately burdened with type 2 diabetes, but have a reduced risk for cardiovascular disease. Elevated C-reactive protein (CRP) levels, a marker of inflammation, have been associated with increased cardio-metabolic risk, including insulin resistance (IR) and elevated serum lipid levels. Although not consistent, several studies worldwide have reported higher circulating high-sensitivity CRP (hsCRP) levels in black compared to white women. However, it is not fully known whether the relationship between hsCRP and cardio-metabolic risk differs between ethnic groups, and whether these relationships may be confounded by differences in socioeconomic status (SES), lifestyle factors and/or central obesity. Aims: (1) To investigate whether hsCRP levels differ between black and white SA women; (2) to examine the ethnic-specific associations between hsCRP, insulin resistance (HOMA-IR) and serum lipids (total cholesterol (TC), triglycerides (TG), high-density lipoproteins (HDL-C) and low-density lipoproteins (LDL-C)); and 3) determine whether any ethnic differences may be explained by differences in SES, lifestyle factors and/or central adiposity between black and white SA women. Methods: The study sample comprised 226 black and 179 white apparently healthy, premenopausal, South African women, aged 18-45 years. A questionnaire was administered that captured measures of SES (level of education, asset index and housing density) and lifestyle factors (smoking status, alcohol use, physical activity and contraceptive use). Weight, height and waist circumference (WC) were measured and fasting blood samples were drawn for the determination of hsCRP, glucose, insulin, and serum lipids. Results: Black women had higher age- and BMI-adjusted hsCRP levels than white women [2.5µg/ml vs. 2.0µg/ml; p=0.047]. Age- and ethnicity-adjusted multivariable analysis showed hsCRP was independently associated with HOMA-IR [β(CI): 0.09 (0.05;0.13); p<0.001], TG [β(CI): 0.07 (0.05;0.10); p<0.001], TC [β(CI): 0.02 (0.01;0.03); p<0.05], HDL-C [β(CI): -0.01 (-0.03;-0.01); p<0.05], and LDL-C [β(CI): 0.03 (0.01;0.05); p<0.05]. Notably, the association between hsCRP and all the lipid markers differed by ethnicity, such that hsCRP was positively associated with TG in the white, but not black women [Pearson r (p-value): 0.42 (p<0.0001) vs. 0.12 (p=0.1284)]. hsCRP was positively associated with TC in white women and inversely associated with TC in black women [Pearson r (p-value): 0.21 (p=0.0109) vs. -0.20 (p=0.0110)]. hsCRP was also associated with LDL-C in white women, but not black women [Pearson r (p-value): 0.17 (p=0.0404) vs. -0.04 (p=0.5822)], and inversely associated with HDL-C in black women, but not in white women [Pearson r (p-value): -0.36 (p<0.0001) vs. -0.06 (p=0.4850)]. The associations between hsCRP and cardio-metabolic risk factors were independent of SES and lifestyle factors for all variables, however, upon further adjustment for central adiposity (WC), only TG remained significantly associated with hsCRP. Discussion: Although hsCRP levels were higher in black compared to white women, and independently associated with various cardio-metabolic risk factors, these relationships were largely mediated by central adiposity, and not SES and lifestyle factors.

The role of socioeconomic status, lifestyle factors and adiposity in the association between hsCRP and metabolic risk