The hypothesis of the role of iron overload associated with HFE gene mutations in the pathogenesis of nonalcoholic steatohepatitis (NASH) has been raised in recent years. In the present study, biochemical and histopathological evidence of iron overload and HFE mutations was investigated in NASH patients. Thirty-two NASH patients, 19 females (59%), average 49.2 years, 72% Caucasians, 12% Mulattoes and 12% Asians, were submitted to serum aminotransferase and iron profile determinations. Liver biopsies were analyzed for necroinflammatory activity, architectural damage and iron deposition. In 31 of the patients, C282Y and H63D mutations were tested by PCR-RFLP. Alanine aminotransferase levels were increased in 30 patients, 2.42 ± 1.12 times the upper normal limit on average. Serum iron concentration, transferrin saturation and ferritin averages were 99.4 ± 31.3 g/dl, 33.1 ± 12.7% and 219.8 ± 163.8 µg/dl, respectively, corresponding to normal values in 93.5, 68.7 and 78.1% of the patients. Hepatic siderosis was observed in three patients and was not associated with architectural damage (P = 0.53) or with necroinflammatory activity (P = 0.27). The allelic frequencies (N = 31) found were 1.6 and 14.1% for C282Y and H63D, respectively, which were compatible with those described for the local population. In conclusion, no evidence of an association of hepatic iron overload and HFE mutations with NASH was found. Brazilian NASH patients comprise a heterogeneous group with many associated conditions such as hyperinsulinism, environmental hepatotoxin exposure and drugs, but not hepatic iron overload, and their disease susceptibility could be related to genetic and environmental features other than HFE mutations.

The hemochromatosis gene, HFE, is located on chromosome 6 in close proximity to the HLA-A locus. Most Caucasian patients with hereditary hemochromatosis (HH) are homozygous for HLA-A3 and for the C282Y mutation of the HFE gene, while a minority are compound heterozygotes for C282Y and H63D. The prevalence of these mutations in non-Caucasian patients with HH is lower than expected. The objective of the present study was to evaluate the frequencies of HLA-A antigens and the C282Y and H63D mutations of the HFE gene in Brazilian patients with HH and to compare clinical and laboratory profiles of C282Y-positive and -negative patients with HH. The frequencies of HLA-A and C282Y and H63D mutations were determined by PCR-based methods in 15 male patients (median age 44 (20-72) years) with HH. Eight patients (53%) were homozygous and one (7%) was heterozygous for the C282Y mutation. None had compound heterozygosity for C282Y and H63D mutations. All but three C282Y homozygotes were positive for HLA-A3 and three other patients without C282Y were shown to be either heterozygous (N = 2) or homozygous (N = 1) for HLA-A3. Patients homozygous for the C282Y mutation had higher ferritin levels and lower age at onset, but the difference was not significant. The presence of C282Y homozygosity in roughly half of the Brazilian patients with HH, together with the findings of HLA-A homozygosity in C282Y-negative subjects, suggest that other mutations in the HFE gene or in other genes involved in iron homeostasis might also be linked to HH in Brazil.

Parvovirus B19 has been associated by some investigators with cases of severe hepatitis. The aim of the present study was to determine the presence of active parvovirus B19 infection among 129 Brazilian patients with non-A-E hepatitis. The patients were assayed for antibodies against parvovirus B19, IgM class, by ELISA. In IgM-positive cases, parvovirus B19 DNA was assayed by PCR in serum and liver tissue and parvovirus VP1 antigen in liver tissue was assayed by immunohistochemistry. Antibodies against parvovirus B19, IgM class, were detected in 3 (2.3%) of 129 patients with non-A-E hepatitis. Previous surgery and blood transfusions were reported by these 3 patients. One patient was a 56-year-old female with severe hepatitis, with antimitochondrial antibody seropositivity and submassive necrosis at liver biopsy, who responded to corticosteroid therapy. Strong evidence for active parvovirus B19 infection was found in this patient, with parvovirus B19 DNA being detected by PCR in liver tissue. Furthermore, parvovirus VP1 antigen was also detected in liver tissue by immunohistochemistry. The other two IgM-positive patients were chronic hepatitis cases, but active infection was not proven, since neither viral DNA nor antigen were detected in their liver tissues. This and other reports suggest a possible relation between parvovirus B19 infection and some cases of hepatitis.