The hepatic iron index was originally described as a useful test to discriminate genetic hemochromatosis from alcoholic siderosis. To evaluate the hepatic iron index as a diagnostic criterion, it is essential to evaluate a cohort of hemochromatosis patients in whom the diagnosis has been established with great certainty. The presence of a sibling with identical human leukocyte antigen (HLA) and with iron overload was considered to be the gold standard for the diagnosis. Hepatic iron index was reviewed retrospectively in 55 homozygotes and in 189 patients who did not have hemochromatosis and were referred for hepatic iron analysis. Four of 55 homozygotes (7%) had a hepatic iron index of < or = 1.9. Hepatic iron concentration was increased in all 4 patients, ranging from 36 to 100 micromol/gm dry weight, (normal value <35.5 micromol/gm). Twelve of 189 (6%) patients without hemochromatosis had hepatic iron indexes > 1.9. The positive likelihood ratio for a hepatic iron index of 1.9 was 12.4. Area under the receiver operating characteristic curve was 0.94 (0.9 to 0.99, 95% confidence interval). The hepatic iron index remains a useful tool in the diagnosis of genetic hemochromatosis. However, it should not be an absolute criterion for the diagnosis and should be interpreted in combination with clinical assessment and genetic studies.