Independent expression of serological markers of thyroid autoimmunity and hepatitis virus C infection in the general population: results of a community-based study in north-western Sardinia

To assess the relationship between serological markers of thyroid autoimmunity and chronic hepatitis C, we surveyed the general population of two villages in the region of Sardinia, Italy, where infection with hepatitis viruses is endemic and the prevalence of autoimmune diseases is elevated. A total of 1310 subjects aged 6-88 years (65% of the total resident population) participated in the survey, and 1233 (94%; 444 males and 789 females) agreed to provide a blood sample. Autoantibodies to thyroid peroxidase (anti-TPO) were measured by radioimmunoassay; antibodies to HCV (anti-HCV) by a third generation enzyme immunoassay and borderline positive results confirmed by recombinant immunoblot assay. For both anti-HCV and anti-TPO the age- and gender-standardized prevalence rates (SPR) were calculated and the significance of the association between the two antibodies tested by Yates corrected chi2 test. The overall SPR for anti-HCV was 50.7x10(-3) (86/1,233), similar between men [49.1x10(-3) (22/444)] and women [52.3x10(-3) (64/789)]. The overall SPR for anti-TPO was 136.9x10(-3) (204/1,233), and that among women [201x10(-3) (174/789)] was almost 3-fold that among men [71.6x10(-3) (30/444)]. A concurrent anti-HCV and anti-TPO positivity was found in a small minority of subjects [8/1,233 (0.65%)], all women aged 57-81 years. The SPR for the two concurrent events was 3.3x10(-3), which was not significantly different (Yates corrected chi2 test = 0.65) from that expected under the assumption of unrelated events. To explore whether HCV infection is a risk factor for anti-TPO positivity, we designed a case-control study with anti-TPO positive subjects as the cases, and anti-TPO negative subjects as the controls. The age- and gender-adjusted odd ratio (OR) was 0.4 (95% CI 0.2,0.7), indicating a negative association. In conclusion, no evidence for epidemiological association of circulating thyroid autoantibodies and antibodies to HCV was found. Our findings do not therefore support a pathogenetic link between HCV infection and thyroid autoimmunity.