Body mass index, circulating sex hormone binding globulin, and dietary linoleic acid are strong predictors of C-reactive protein levels in healthy premenopausal women

3812

C-reactive protein (CRP) is an acute-phase serum protein synthesized in the liver. Circulating CRP may serve as a risk factor and biomarker for inflammation and cancer. In this study we investigated the potential correlates of CRP in 30 to 40 year old multiethnic women (N=249).
 Healthy premenopausal women, who were not pregnant, not breastfeeding, not using any contraceptive medications, and had regular menstrual cycles, provided two fasting blood samples that were analyzed for CRP, sex hormone binding globulin (SHBG), insulin, progesterone, estradiol, testosterone, and blood chemistries. Body weight, body mass index (BMI), waist and hip circumference, lean body mass and fat body mass were determined. Nutrient intake was estimated from three 24-hr food records. Forward selection, multi-level multivariate regression models were constructed to assess the correlates of CRP.
 Mean CRP levels in our study group were 0.64 ± 0.68 mg/dl (95% CL 0.55-0.72). CRP levels correlated positively (all P<0.0001) with body weight, BMI, waist and hip circumference, lean body mass, fat body mass, and the percentage of body mass as fat (R=0.30-0.70), and with serum insulin (R=0.49), triglycerides (R=0.34), glucose (R=0.25), and alkaline phosphatase (R=0.44). CRP levels correlated negatively with height (R=0.17, P=0.007), serum albumin (R=0.22, P<0.001), HDL (R=0.41, P<0.0001), and SHBG (R=0.40, P<0.0001), and with dietary intake of vegetable protein, pectin, insoluble dietary fiber, alcohol, polyunsaturated fatty acid to saturated fatty acid ratio, linoleic acid, magnesium, and folic acid (all P<0.05).
 The final model of multivariate analyses showed that BMI was a positive independent predictor of CRP (for1 kg/m², β=0.48, P<0.0001), while serum SHBG (for 1 nmol/l, β=0.11, P=0.04) and dietary linoleic acid (for 1 mg/day, β=0.16, P=0.02) were negative independent predictors of CRP. Together, these variables could explain 50% of the variance in CRP levels. Ethnicity did not have an effect on CRP levels in fully adjusted models. Our study, though cross-sectional in nature, suggests that serum SHBG and linoleic acid have a negative influence on CRP levels. Lifestyle modifications aimed at decreasing BMI may help decrease circulating CRP levels and the associated disease risk. (Supported by US Army DADM17-01-1-0417, NIH NCRR GCRC MO1 RR00073, NIH CA95545, NIH CA65628 and USPHS CA95545. Clinical Trial Identifier: NCT00204477)