Association of chronic viral hepatitis B with insulin resistance

AIM: To investigate the relationship between chronic viral hepatitis B (CVHB) and insulin resistance (IR) in Korean adults. METHODS: A total of 7880 adults (3851 men, 4029 women) who underwent a comprehensive medical examination were enrolled in this study. Subjects diagnosed with either diabetes mellitus, or any other disorder that could influence their insulin sensitivity, were rejected. Anthropometry, metabolic risk factors, hepatitis B surface antigen, hepatitis B surface antibody, hepatitis B core antibody, fasting plasma glucose and insulin were measured for all subjects. Homeostasis model assessment (HOMA), quantitative insulin check index (QUICKI), and Mffm index were used for determining insulin sensitivity. Each participant was categorized into a negative, recovery, or CVHB group. To compare variables between groups, a t-test and/or one-way analysis of variance were used. Partial correlation coefficients were computed to present the association between insulin resistance and other variables. Multiple logistic regression analysis was used to assess the independent association between CVHB and IR. RESULTS: The mean age of men and women were 48.9 and 48.6 years, respectively. Subjects in the CVHB group had significantly higher waist circumference [(86.0 ± 7.7 cm vs 87.3 ± 7.8 cm, P = 0.004 in men), (78.3 ± 8.6 cm vs 80.5 ± 8.5 cm, P < 0.001 in women)], cystatin C [(0.96 ± 0.15 mg/dL vs 1.02 ± 0.22 mg/dL, P < 0.001 in men), (0.84 ± 0.15 mg/dL vs 0.90 ± 0.16 mg/dL, P < 0.001 in women)], fasting insulin [(5.47 ± 3.38 μU/mL vs 6.12 ± 4.62 μU/mL, P < 0.001 in men), (4.57 ± 2.82 μU/mL vs 5.06 ± 3.10 μU/mL, P < 0.001 in women)] and HOMA index [(1.24 ± 0.86 vs 1.43 ± 1.24, P < 0.001 in men), (1.02 ± 0.76 vs 1.13 ± 0.87, P = 0.033 in women)] compared to control group. The HOMA index revealed a positive correlation with body mass index (BMI) (r = 0.378, P < 0.001), waist circumference (r =0.356, P < 0.001), percent body fat (r = 0.296, P < 0.001), systolic blood pressure (r = 0.202, P < 0.001), total cholesterol (r = 0.134, P < 0.001), triglycerides (r = 0.292, P < 0.001), cystatin C (r = 0.069, P < 0.001) and uric acid (r = 0.142, P < 0.001). The QUICKI index revealed a negative correlation with BMI (r = -0.254, P < 0.001), waist circumference (r = 0-0.243, P < 0.001), percent body fat (r = -0.217, P < 0.001), systolic blood pressure (r = -0.132, P < 0.001), total cholesterol (r = -0.106, P < 0.001), triglycerides (r = -0.205, P < 0.001), cystatin C (r = -0.044, P < 0.001) and uric acid (r = -0.096, P < 0.001). For subjects identified with IR, the odds ratio of an accompanying diagnosis of chronic hepatitis B was 1.534 (95% CI: 1.158-2.031, HOMA index criteria) or 1.566 (95% CI: 1.124-2.182, QUICKI criteria) after adjustment for age, gender, BMI, and amount of alcohol consumption. CONCLUSION: Our study demonstrates that CVHB is associated with IR. CVHB may need to be monitored for occurrence of IR and diabetes mellitus.