Urinary lignans and inflammatory markers in the US National Health and Nutrition Examination Survey (NHANES) 1999-2004 and 2005-2008.

Author(s): Eichholzer M,  Richard A,  Nicastro HL,  Platz EA,  Linseisen J,  Rohrmann S

Journal: Cancer Causes Control

Date: 2014 Mar

Major Program(s) or Research Group(s): CPFP, NSRG

PubMed ID: 24463788

PMC ID: PMC4696038

Abstract: PURPOSE: Chronic inflammation has been implicated in the etiology of various chronic diseases. We previously found that certain urinary isoflavones are associated with markers of inflammation. In the present study, we examined the associations of serum C-reactive protein (CRP) and white blood cell (WBC) count with lignans, which are more frequent in the Western diet than isoflavones. METHODS: Our analysis included 2,028 participants of NHANES 2005-2008 and 2,628 participants of NHANES 1999-2004 aged 18 years and older. The exposures of interest were urinary mammalian lignans (enterodiol and enterolactone). Outcome variables were two inflammatory markers (CRP [≤10 mg/L] and WBC [≥3.0 and ≤11.7 (1,000 cells/μL)]). Log-transformed CRP concentration and WBC count by log-transformed creatinine-standardized concentrations of mammalian lignans were used for linear regression. RESULTS: Statistically significant inverse associations of urinary lignan, enterodiol, and enterolactone concentrations with circulating CRP and WBC counts were observed in the multivariate-adjusted models: In NHANES 2005-2008, per one-percent increase in lignan concentrations in the urine, CRP concentrations and WBC counts decreased by 8.1 % (95 % CI -11.5, -4.5) and 1.9 % (95 % CI -2.7; -1.2), respectively. Per one-percent increase in enterodiol and enterolactone, WBC counts decreased by 2.1 % (95 % CI -2.8, -1.3) and 1.3 % (95 % CI -1.9, -0.6), respectively. In NHANES 1999-2004, analogous results were 3.0 % (95 % CI -5.6, -0.3), 1.2 % (95 % CI -2.0; -0.4), 1.0 % (95 % CI -1.8, -0.2), and 0.8 % (95 % CI -1.4, 0.2). CONCLUSIONS: Mammalian lignans were inversely associated with markers of chronic inflammation. Due to the cross-sectional design, our findings require confirmation in prospective studies.