The relationship between alcoholism and hereditary hemochromatosis performed before the use of these dihemochromatosis remains controversial. Previous studagnostic tools most likely included patients with alcoies have included patients with alcoholic siderosis holic siderosis rather than hereditary hemochromatorather than hereditary hemochromatosis. In this retrosis. [3][4][5][6][7][8][9] In this study, we have re-examined the spective study, the clinical features, iron status, alcohol prevalence of alcohol use in hereditary hemochromatohistory, liver histology, and long-term survival were resis using rigid criteria for the diagnosis, including the viewed in 105 homozygotes for hemochromatosis using presence of an HLA identical sibling with hemochromarigid diagnostic criteria including an HLA identical sibtosis.
ling with iron overload. Heavy alcohol consumption (ΓΊ80 g ethanol/day) was found in 15% of hemochromato-PATIENTS AND METHODS sis patients. Histological features of alcoholic liver disease (Mallory's hyaline bodies, pericentral fibrosis, poly-Patient records were reviewed from a large database of 170 morphonuclear infiltrate, and fatty infiltration) were hemochromatosis homozygotes that have been documented uncommon in hemochromatosis. Hemochromatosis pasince 1965. Family investigations at this center ranged from tients with heavy alcohol consumption had a higher no further investigation in the case of isolated proband cases prevalence of cirrhosis than hemochromatosis patients to extensive pedigree analysis involving up to 45 family memwithout heavy alcohol consumption. Hepatic iron conbers of a single family over multiple generations. Data was centration and hepatic iron index did not significantly available on 483 family members assessed at this tertiary differ between these two hemochromatosis groups. referral center for possible hemochromatosis. The clinical fea-Long-term survival was significantly reduced in patients tures of this patient cohort has been previously described. [10][11][12] with heavy alcohol consumption (mean follow-up, 9.22 From this database, homozygotes were selected that had an years). This suggests that chronic alcohol consumption HLA identical sibling with hemochromatosis to establish a has an additive hepatotoxic effect despite the paucity of hereditary disorder of iron metabolism (n Γ
105). There were histological features of alcoholic liver disease. (HEPATOL-65 homozygotes excluded from this study because no HLA OGY 1996;23:724-727.) identical sibling had been identified. The diagnosis was confirmed by liver biopsy results in all probands and most discovered siblings (n Γ
89/105). No hemochromatosis patients had
The clinical distinction between patients with heredconcomitant chronic viral hepatitis B or C on histological itary hemochromatosis and alcoholic siderosis continexamination of the liver biopsy specimen. Liver biopsy re- ues to be a diagnostic problem. Clinical tools that have ports were reviewed for the presence of Mallory's hyaline, improved this diagnostic distinction include measurepericentral fibrosis, polymorphonuclear infiltrate, and fatty ment of the hepatic iron concentration, hepatic iron infiltration by a single pathologist in 85 cases. These histolog- index (hepatic iron/age), 1 and family studies using HLA ical features were considered to be typical but not pathog- typing as a surrogate genetic test because of the close nomic of alcoholic liver disease. The hemochromatosis pa- tients had an alcohol history recorded by a physician and linkage of the putative hemochromatosis gene to the research nurse at two additional times as part of a large HLA locus on chromosome 6. 2 database collection project on patients and family members Previous studies on the prevalence of alcoholism in with hemochromatosis. Referral letters from primary care physicians were also consulted at the time of initial evaluation. Patients were excluded if the medical history regarding
From the