Article Text

HFEand alcoholic liver disease
  1. Gartnavel General Hospital,
  2. 1053 Great Western Road,
  3. Glasgow G11 6NT, UK
  4. Clinical Sciences Unit,
  5. Queensland Institute of Medical Research,
  6. 300 Herston Road,
  7. Brisbane, Queensland, Australia
  1. Dr George.
  1. C P DAY
  1. Department of Medicine,
  2. The Medical School,
  3. Framlington Place,
  4. Newcastle upon Tyne NE2 4HH, UK

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Editor,—We read with interest the recent paper by Grove et al (Gut1998;43:262–6) which concludes that the two haemochromatosis mutations (C282Y and H63D) influence neither the liver iron content nor risk of fibrosis in alcoholic liver disease. Two factors in this study may have led to an underestimation of the contribution of the haemochromatosis gene (HFE) to hepatic iron loading.

Most of the patients in the study had established cirrhosis. Cirrhosis, particularly alcoholic cirrhosis, may itself be a potent cause of hepatic iron loading once it has developed.1 This rapid iron loading may result in hepatic iron concentrations usually associated with the homozygous haemochromatosis state and might obscure the effect of any iron loading that might occur due to the heterozygous genotype. Also, in haemochromatosis, excess alcohol consumption, although not affecting the hepatic iron concentration, seems to cause iron to redistribute from hepatocytes to reticuloendothelial cells.2 The same may be true of hepatic iron stores in heterozygous haemochromatosis when excess alcohol is consumed, thus causing an underestimation of the hepatic iron stores if only hepatocellular iron is scored histologically and hepatic iron concentration (HIC) is not measured biochemically. In their study, Grove et al noted the presence of perisinusoidal and portal tract iron but did not estimate the degree of this staining and did not measure HIC.

These two factors may have led to the finding that significant (grade 2 or more) hepatocyte iron staining was not significantly commoner in age/sex matched alcoholics with either of the haemochromatosis gene mutations. However, although not quite reaching statistical significance, in this population twice as may alcoholics with one copy of the C282Y mutation did have grade >2 hepatocyte iron compared with those without this mutation (36%v 18%) and twice as may alcoholics with one copy of the C282Y mutation had any degree of histological iron compared with those without the C282Y mutation (46%v 24%).

In order to eliminate the effect of cirrhosis per se on hepatic iron loading it would be interesting to know whether there was any greater association between the HFE gene mutations and hepatic iron stores in the 59 patients with precirrhotic alcoholic liver disease. Perhaps more importantly it would be interesting to know in these 59 patients whether there was any relation between hepatic iron stores and the degree of hepatic fibrosis.

The fact that the HFE gene mutations were not significantly commoner in patients with fibrotic/cirrhotic alcoholic liver disease than controls does seem to suggest that these mutations do not have a major role in the progression of alcoholic liver disease. However, it is likely that any effect thatHFE might have on the progression of chronic liver disease is mediated through raising iron stores, as we have shown in non-alcoholic steatohepatitis,3 rather than a direct effect of HFE on fibrogenesis. Only a minority of heterozygotes show some phenotypic expression of iron overload,4 other causes of iron overload clearly occur1 ,5 ,6 and many factors may be involved in the progression of alcoholic liver disease. Therefore a much larger fibrotic alcoholic population may be required to investigate definitively whether or not the HFE gene mutations are enriched in this population.

We agree with Grove et al that further research is required to investigate the factors that determine liver iron content in alcoholic liver disease and the prognostic significance of this iron. We do not think, however, that the effect of the haemochromatosis gene mutations on liver iron content and disease progression can yet be completely discounted.



Editor,—We thank Drs George and Powell for their comments on our paper. We agree that confining our iron studies to the assessment of iron staining in hepatocytes of cirrhotic patients may have lead to underestimation of the effect of theHFE genotype on iron content in alcoholic liver disease. However, as they point out, our results are fully consistent with the findings in non-alcoholic C282Y heterozygotes—only a minority of whom show evidence of iron overload—and any small effect of C282Y heterozygosity on liver iron content is likely to be swamped by the other causes of iron overload present in heavy drinkers. On the question of whether or not heavy drinkers with the C282Y mutation have more liver iron than drinkers without the mutation, in contrast to George and Powell, we believe our results ofHFE genotyping do largely exclude a role for this mutation in determining the risk of fibrotic alcoholic liver disease. This is based not only on the similar frequency of the mutation in patients and controls, but also the identical mean age of presentation (and therefore cumulative alcohol intake) in patients with and without the mutation. Moreover, the lack of any increased frequency of the C282Y mutation in patients with alcoholic liver disease compared with controls has been reported by others.1-1

With respect to the relation between iron stores and the degree/risk of fibrosis in heavy drinkers, we have no data on iron staining in the patients with fibrosis. However, we have reported recently that the liver iron content of patients with alcoholic fatty liver does not predict the risk of progression to fibrosis or cirrhosis.1-2If we accept that the C282Y mutation increases liver iron content but not the risk of alcoholic liver disease, taken together, these results would appear largely to exclude a role for excess liver iron in the pathogenesis of alcohol related liver disease.


  1. 1-1.
  2. 1-2.