We read with interest the article by Lecube et al. (1) regarding the strong association they observed between diabetes and serum ferritin in chronic hepatitis C. We sought to analyze the contribution of biochemical, metabolic, and histological parameters to high ferritin levels detected in hepatitis C.
We investigated a large consecutive series of 177 patients with chronic hepatitis C who underwent a diagnostic liver biopsy. Serum ferritin was tested in a univariate analysis against demographics, biochemical parameters, and histological features. Patients with cirrhosis or with any alcohol intake were excluded. The median age of the patients was 48.4 years (range 19–71) and 97 (54.8%) were men. Using the same cutoff values of Lecube et al., serum ferritin was raised in 92 cases (52.0%). The prevalence of impaired glucose tolerance or diabetes was 9.6% (17 cases) in our series. Overall, 66 patients (37.3%) had mild fibrosis (F0–F1) and 111 patients (62.7%) had moderate to severe fibrosis (F2–F3) according to METAVIR. Hepatic iron deposits were found in 68 patients (38.4%). Hepatic steatosis was detected in 132 patients (74.6%). Serum ferritin correlated by univariate analysis with male sex (P = 0.05), BMI (P = 0.0001), aspartate aminotransferase and alanine aminotransferase levels (P = 0.003 and P = 0.0009, respectively), γ-glutamyl transferase levels (P < 0.00001), hepatic iron (P < 0.00001), and hepatic steatosis (P = 0.01). No correlation between serum ferritin and fasting glucose could be observed. Moreover, no significant difference in serum ferritin was observed in patients with impaired glucose tolerance or diabetes in comparison with other patients.
The following considerations arise from the comparison of our data with those reported by Lecube et al. First, mean serum ferritin in our study was much higher than that observed by Lecube et al. even if we excluded cirrhosis and alcohol consumption. Second, prevalence of diabetes was much higher (21.7%) in the Spanish series compared with our own cohort, probably because Lecube et al.’s study was conducted in a tertiary reference center for both diabetes and hepatitis C. Third, we did not observe any association between raised serum ferritin and diabetes. In chronic hepatitis C there are many conditions that could elevate serum ferritin such as necroinflammation, steatosis, and hepatic iron deposition (2). In our series, we found an association with metabolic factors and with markers of inflammation. On the other hand, there was also a strong association with proper hepatic iron deposition.
We can therefore conclude that the increase of serum ferritin in chronic hepatitis C could be linked to diabetes, as Lecube et al. have clearly suggested, but the pathogenesis is multifactorial. The weight of different determinant factors on the elevation of ferritin depends on their prevalence in the analyzed series of patients.
