High Total Serum Testosterone is Associated With Increased Risk of Advanced Hepatic Fibrosis and Inflammatory Activity in Male Veterans With Chronic Hepatitis C

Abstract

Background: Increased relative adiposity, measured as BMI or waist circumference, is a recognized risk factor for HCV-related liver disease. Bioelectric impedance analysis (BIA)based measurements of adiposity offer advantages including less inter-observer variability compared with widely used anthropometric calculations. There have been no studies of BIA in association with HCV-related liver disease. Methods: HCV+ male veterans with confirmed HCV viremia and without ascites or decompensated disease were recruited at a single VA medical center. We obtained direct 8-segment multi-frequency BIA measurements of body composition with Biospace Inbody scale (98% correlation with DXA, 99% reproducibility). BIA measurements (<5 minutes in duration) were taken after entering study measured height, gender and age. Fibrosure test was used to assess grade of fibrosis and inflammation. We also obtained other conventional anthropometric measurements (waist, hips). ANOVA was employed to compare mean BIA and anthropometric measurements with advanced compared to mild fibrosis (F3-F4, F4 vs. F0-F3), and with advanced compared to mild inflammatory activity (A2, A2-A3, A3 vs. A0, A1, A1-A2). Logistic regression was employed to evaluate these factors and risk of advanced liver disease after adjusting for ethnicity, age and current alcohol use. Results: We recruited 198 HCV+ male veterans (mean age 56 yrs; 52% AfricanAmerican; 43% non-Hispanic White). BIA-measured body fatness (%BF) and BMI were significantly associated with advanced hepatic fibrosis in univariate (Table 1) and multivariate (Table 2) analyses. Similar associations were observed with advanced inflammation (data not shown). However, waist and hip circumference obtained from conventional tapemeasurements were not significantly associated with fibrosis in univariate or multivariate analyses, though both were marginally significant for inflammation. There were no significant differences in total body water or lean body mass among groups with advanced and mild hepatic disease (Tables 1, 2). This indicates that the observed associations with %BF and BMI are primarily related to differences in relative fatness and not in water retention or muscular atrophy. The correlation between BMI and %BF was significant and moderately strong (Kendal tau-b=64%), with similar correlation observed for both BMI and %BF with waist and hip circumference respectively. Conclusion: Bioimpedance analysis (BIA) measured %BF and BIA calculated BMI both predict the presence of advanced hepatic fibrosis and inflammation in HCV+ male veterans. BIA measurements were easy, quick, and highly reproducible. Given the degree of correlation among these measures, additional research is needed to clarify their joint and individual contributions in risk of advanced liver disease. Table 1. Univariate assessment of bioelectric impedance analysis (BIA)and anthropometricbased measurements in 198 male veterans with chronic HCV.