—  SPECIALTY CONFERENCE  —

Liver Pathology

Case 1 - Hereditary Hemochromatosis

Kenneth Batts
Abbott Northwestern Hospital
Minneapolis, Minnesota


Click on each slide thumbnail image for an enlarged view
Clinical History:
The patient is a 54 year old woman who has Type II diabetes, hypertriglyceridemia, and obesity. She had a mild increase in serum AST level with normal serum alkaline phosphatase and total bilirubin values. A liver biopsy was performed to assess the putative fatty liver disease, the presence of which was confirmed. However, hemosiderin was present in the liver, prompting several issues for the pathologist:
  1. Would you quantitate the liver iron? If so, how would you do it? Is it necessary to perform quantitative iron analysis in this case?
  2. In your daily practice, would you describe the iron and hope the clinician knows what to do to work it up, or would you actively intervene and join the attempt to determine the cause of the hemosiderosis?
  3. What additional clinical testing would you advise?

(See end of syllabus for my answers to these.)


Case 1 - Figure 1 - Fatty liver disease - non-cirrhotic
Case 1 - Figure 2 - Mild zone 1 hemosiderosis (Prussian blue stain)
Case 1 - Figure 3 - Mild zone 1 hemosiderosis - hepatocellular (Prussian blue stain)

Discussion

A Brief Medical History of Hemochromatosis
In 1889 von Recklinghausen identified excess tissue iron obtained at autopsy and termed the condition "hamochromastose".1  This iron was felt to be derived from the blood until 1935 when the British gerontologist Sheldon published his findings from 311 selected patients, concluding that the iron overload was secondary to increased iron absorption .2 

It was not until 1952 that a therapy became available, when Davis and Arrowsmith reported benefits from treating these patients with phlebotomy.3  In 1962 the antemortem diagnosis of hemochromatosis became available with the developing of a grading system for iron in liver samples was developed by Scheuer and colleagues.4  Easier antemortem diagnosis was made available by the development of reliable serum iron, iron binding capacity, and ferritin testing in the 1960's and 70's. While an HLA locus association was identified by Simon et al in 1975,5  it was not until 1996 when Feder and colleagues identified two missense mutations on chromosome 6p that accounted for the vast majority of cases of hemochromatosis.6 

Terms and Current Definition of Hemochromatosis:
Traditionally , "hemochromatosis" had been defined clinically based on the presence of a combination of an otherwise unexplained iron overload, frequent presence of iron overload in relatives, and in later stages end-organ damage (liver, pancreas, etc). Following the discovery of the major gene mutations described below, a "genetic" definition came into being: ". . . hemochromatosis is now defined as a disorder of iron metabolism that is inherited as an autosomal recessive trait due to two mutant HFE alleles" 7  This definition has the advantage of correctly characterizing individuals early in the disease course, it does not account well for the approximately 10% of individuals who have phenotypic expression of the disease but lack the major mutations referred to in this definition (C282Y and H63D).

Genetics of Hereditary Hemochromatosis
Population Genetics There is a current general consensus that the initial hemochromatosis mutation occurred within the last 100-120 generations (@2800) years or so in Northwestern Europe in a Celtic, Viking, or Germanic population.(Fairbanks) The propagation of this gene likely occurred because the adverse complications did not become evident until after the reproductive years and the iron stores were of benefit to a younger adult population living in a region of possibly limited dietary intake of iron. Affected women were better able to tolerate menstrual or child-birth related blood loss and all affected individuals could better cope with blood loss associated with trauma or blood letting (Fairbanks). The worldwide distribution of hemochromatosis reflects the migrations of the originally affected populations – Caucasians in Northwestern Europe, Canada, USA, Australia, New Zealand, and South Africa all have roughly similar gene frequencies of 0.06-1.0 and heterozygous carriage in 10-15% for the major gene mutations.

The HFE Gene Feder et al identified two mutations in an MHC class I gene on chromosome 6 which is now named the "HFE" gene. The "major" mutation was a G to A substitution at nucleotide 845 which lead to a cysteine to tyrosine substitution at the amino acid 282, now commonly referred to a the "C282Y" mutation. The "minor" mutation was a C to G change at nucleotide 187 which resulted in a histidine to aspartic acid substitution at position 63 ("H63D").

Among patients with marked iron overload ("phenotypic hemochromatosis"), approximately 80-85% (range 59-100%) will demonstrate a C282Y/C282Y genotype, 5% (range 4-8%) a C282Y/H63D phenotype, and the remainder are not clearly explained currently.(Fairbanks) Hardy-Weinberg analyses indicate that the C282Y/wild type, H63D/wild type, and H63D/H63D phenotypes are not associated with phenotypic disease and these patients are regarded as "carriers". There continues to be searching for genes to explain the 5-10% of hemochromatosis patients without the aforementioned mutations, however at the time of this writing no strong candidates have emerged.

Histopathology of Hereditary Hemochromatosis
The hallmark of genetic hemochromatosis is the deposition of hemosiderin in hepatocytes and biliary epithelium. Hemosiderin is insoluble and particulate in nature and will appear granular with iron stains.9  Alternatively, iron may be deposited in the form of ferritin which is soluble and characterized by diffuse, non-granular light blue staining of the hepatocyte or macrophage cytoplasm and is very non-specific but not characteristic of genetic hemochromatosis.

In early phases of disease, hemosiderin is deposited in periportal (zone 1) hepatocytes .As progressive iron accumulation takes place, midzonal (zone 2) and centrilobular (zone 3) hepatocytes will progressively accumulate iron as will biliary epithelium. Neither inflammation nor fatty infiltration is a feature of uncomplicated hemochromatosis and their presence suggests alcoholic liver injury or chronic hepatitis C.

As iron accumulates, eventually individual hepatocytes will accumulate lethal levels of iron and undergo "sideronecrosis". Generally sideronecrosis will become evident when liver iron is in the 12-15,000 microgm/gm dry weight liver range although there seems to be considerable variation in this threshhold. The locally released iron is then taken up into macrophages, however hepatocellular iron will continue to dominate. By contrast, iron deposition in hematologic disorders occurs primarily in the reticuloendothelial system (Kupffer cells) and, when present in large quantities, "spills over" into the hepatocytes to a lesser degree. Thus, the finding of iron deposited primarily in Kupffer cells is not in keeping with hemochromatosis and does not require measurement of hepatic iron concentration for diagnostic purposes. Quantification of hepatic iron may, however, be useful in assessing the need for treatment in exogenous iron overload.

With continuing iron accumulation and sideronecrosis, progressive fibrosis and eventually cirrhosis may occur. Iron related fibrosis tends to begin when approximately 15,000 microgm Fe/gm dry weight liver is present although this is just a generalization. Precirrhotic portal and periportal fibrosis takes on a "holly leaf" configuration. When cirrhosis is present, it is bland in nature with fine fibrous tissue septa surrounding regenerative nodules.

Another histologic feature described in the hemochromatotic liver is the iron free focus characterized by a relative reduction of hepatocytic iron and dysplastic changes.10  This lesion is purported to represent an early step in the development of hepatocellular carcinoma. Both the clinician and the pathologist should maintain a high index of suspicion for hepatocellular carcinoma as it may occur in up to one-third of patients with cirrhotic stage hemochromatosis.11 

Differential Diagnosis of Hemosiderosis
It should be recognized that hepatic hemosiderin deposition can occur in a wide variety of disease states other than genetic hemochromatosis. In general, thorough clinical evaluation and careful histologic examination can distinguish these conditions from genetic hemochromatosis although biochemical analysis of liver tissue iron (see below) or genotypic analysis may be necessary in ambiguous cases.

One of the most common non-hemochromatotic cause of hemosiderosis is alcoholic steatohepatitis. Histological features include fatty infiltration and varied degrees of inflammatory reaction. Generally, stainable hepatic iron is only mild in such instances, but at times it is sufficient to cause confusion with regard to the question of genetic hemochromatosis. Measurement of hepatic iron concentration usually allows one to distinguish these conditions. The mechanism whereby mild hemosiderosis occurs in conditions other than hemochromatosis is unknown;12  increased iron absorption has been postulated in alcoholic liver disease,13  but has not been supported by measurement of iron absorption following administration of alcohol.14  The demonstration of increased iron absorption in a small number of chronic alcoholic men may be pertinent, however.15  In some other instances, mild hepatic iron deposition likely reflects the heterozygous state of hemochromatosis, given the fact that approximately 10-15% of the Caucasian North American population is so affected and mild iron overload is manifested in about one-third.

Transfusions and chronic hemolytic disorders commonly lead to hepatic hemosiderin deposition. Transfusional iron tends to accumulate in Kupffer cells and is thus easily distinguished from the hepatocellular iron of genetic hemochromatosis. Iron following hemolysis tends to be deposited in both hepatocytes and Kupffer cells and thus demonstration of hemolysis by laboratory means is very helpful in distinguishing hemolysis-related hemosiderosis from genetic hemochromatosis.

Chronic viral hepatitis is not uncommonly accompanied by hepatic hemosiderin. The hemosiderin deposits are largely in Kupffer cells and often contain diffuse, ferritin-type iron, both of which are in contrast to genetic hemochromatosis.

Accumulation of hemosiderin is fairly common in cirrhosis of any type. Ludwig et al noted stainable hemosiderin in 145/449 (32.4%) of cirrhotic livers and increased chemically-determined iron concentration in 91/449 (20.3%), including 38 cases(8.5%) in which the level was in the hemochromatosis range (hepatic iron index > 1.9).16  In this study, it was felt that hemozygous hemochromatosis only accounted for 5 instances of iron overload. While many of the other cases may have reflected incidental heterozygous genetic hemochromatosis, it seems clear that iron deposition can occur in cirrhosis as an secondary phenomenon although the etiology remains unclear. Patients with biliary cirrhosis seem to be less prone to accumulate iron (7-20%) than do patients with non-biliary cirrhosis(22-67%). In cases of cirrhosis with iron deposition, one is usually able to determine whether homozygous genetic hemochromatosis is present using traditional means of assessment, however in ambiguous cases genetic analysis may play a helpful role.

Quantitative Analysis of Liver Tissue Iron
Introduction In the past 25 years, measurement of iron concentration in liver biopsy specimens has proven to be a valuable diagnostic tool which has also served to broaden our concept of hemochromatosis.17  Tissue iron analysis must be performed in a highly qualified laboratory where rigorous control of processing and analytical procedures are followed to assure accurate results. The actual analysis is performed in a graphite atomic absorption spectrophotometer. The results are reported in μmol/g dry weight or in μg/g dry weight. Normal values in the Mayo Metals Laboratory are 7.2-39.4 μmol (400-2,200 μg) per gram dry weight in men and 1.8-28.6 μmol (100-1,600 μg) per gram dry weight in women. The variation of repeated determinations on the same test sample is approximately 5%.

Interpretation of Hepatic Iron Concentration Despite the somewhat subjective nature of histochemical assessment of deposited iron, it correlates reasonably well with the hepatic iron concentration as measured chemically. This is especially true at higher concentrations, which almost always show grade 3 or 4 stainable iron. Nonetheless, occasional instances of disparity are encountered. Also, the significance of moderate degrees of iron overload (2-3+) may be difficult to interpret, particularly in adults in the 30-60 years of age range. In such cases, measurement of hepatic iron concentration is of particular value (see below regarding histologic hepatic iron index). Although the hepatic iron concentration provides a very useful piece of diagnostic information, in routine clinical practice this determination is not necessary in every biopsy specimen obtained for suspected hemochromatosis. For instance, if the pathologist finds little in the way of stainable iron, measurement of hepatic iron concentration is unnecessary. Conversely, if there is histological evidence of heavy iron deposition, the hepatic iron concentration may be superfluous.

Heavy iron overload (defined arbitrarily as hepatic iron concentration >10,000 μg/g dry weight or iron stores as estimated by quantitative phlebotomy exceeding 10 g) is the hallmark of genetic hemochromatosis. As mentioned above, iron accumulation to this degree is almost never seen in other conditions. With the advent of HLA typing, however, it has become apparent that homozygous hemochromatosis can be associated with only modest degrees of iron overload, especially in young individuals and in premenopausal women. Furthermore, only a portion of those who are genotypically homozygous manifest disease-related mobidity with advancing age.

The concept of the hepatic iron index was introduced by Bassett, et al. in 1986,17  and it has proven useful, especially in the interpretation of mild to moderate degrees of hepatic iron overload. The index is simply calculated by dividing the hepatic iron concentration in μmoles/g by the patient's age and is expressed in micromoles/g/year. Since the hepatic iron concentration is often reported in μg/g, the value must be divided by 55.8 to convert it to μmoles/g. The utility of this determination is based on the observation that hepatic iron concentration rises progressively with age only in homozygous hemochromatosis. Thus, an hepatic iron index of 1.9 or greater (in normals, the index is usually <1.0.) generally indicates homozygous hemochromatosis, whereas values <1.9 are compatible with the heterozygous state, alcoholic liver disease, or other conditions not accompanied by significant iron overload. Borderline values may be difficult to interpret and a number of caveats exist about the use of the hepatic iron index (below). In ambiguous cases, genetic analysis is often helpful.

Although the validity of the hepatic iron index has been confirmed in several studies, there are occasional instances where available evidence and clinical judgment support another interpretation. Firstly, iron quantitation and hepatic iron index determination for diagnostic purposes should be restricted to cases in which the histologic distribution of suggests genetic hemochromatosis, i.e. hepatocellular iron predominates over Kupffer cell iron, and a reasonable clinical explanation for iron overload does not exist. Transfusional hemosiderosis, particularly in younger individuals, can easily lead to hepatic iron indices in excess of 1.9. Secondly, the utility of the hepatic iron index in the pediatric population has not been demonstrated and thus should be used in this group with utmost caution if at all. Thirdly, as discussed above, it should be kept in mind that the index was originally applied to cases of early hemochromatosis and must be interpreted cautiously in patients with advanced chronic liver disease where exceptions appear to occur quite frequently.16  Lastly, sampling variations may occur (see below).

Because measurement of hepatic iron concentration is not always available, Deugnier, et al. ,18  have investigated the utility of a histologic hepatic iron index based on detailed and systematic grading of iron deposits in various locations of the liver lobule. In their hands, this provided satisfactory differentiation of heterozygous and homozygous patients, but other conditions were not evaluated. This method or the computerized assessment of hepatic iron described by Olynyk, et al. ,19  can be performed on a standard histologic slide stained for iron. The findings from both methods correlate reasonably well with the chemical measurement of hepatic iron and support the opinion that determination of hepatic iron concentration may not be essential in every instance. A pathologist, experienced in assessing hepatic iron deposition, should assist the clinician in this decision.

Potential Difficulties in Interpretation of Liver Biopsy and Hepatic Iron Concentration With regard to measurement of hepatic iron concentration, certain potential problems can give rise to spurious results. The distribution of iron, although generally uniform in the non-cirrhotic liver, might be quite irregular when cirrhosis is present. Irregular distribution is not likely to affect the final interpretation if hepatic iron concentration is very high, but it might cause confusion when iron overload is moderate in degree. As mentioned above, a sample that consists largely of scar tissue might lead to underestimation of the degree of iron deposition.

The finding of modest hepatic siderosis in a patient with histological evidence of established cirrhosis should be interpreted with caution as mild iron overload is unlikely to be the cause of cirrhosis. It has been observed that a threshold level of hepatic iron concentration of approximately 12 times the upper limit of normal is required for the production of significant fibrosis.17  This set of circumstances was encountered when hepatic iron concentration was measured in livers explanted in the course of orthotopic liver transplantation.16  In several cases, the hepatic iron index was mildly elevated above the level of 1.9 and the amount of iron present was considered insufficient to have caused cirrhosis. The explanation for the mild siderosis observed was uncertain but earlier liver biopsies, which were available in some cases, clearly showed cirrhosis to be present before excess iron deposition occurred, thus providing strong evidence against the diagnosis of hemochromatosis.

The occurrence of siderosis in conjunction with alcoholic liver disease presents a common and often vexing problem. As indicated above, serum iron parameters are often elevated in this setting and stainable hepatic iron may appear greater than the hepatic iron concentration would suggest. The hepatic iron concentration, however, rarely exceeds two to three times the upper limit of normal in alcoholic liver disease.20  On the other hand, alcoholic patients with heavy iron overload (for example, >5 times normal hepatic iron concentration) have been shown to have genetic hemochromatosis of the homozygous form.12  Alcohol likely works in synergy with iron to aggravate tissue injury and may give rise to cirrhosis in the hemochromatotic earlier than would be expected with iron overload alone.17,21,22 

The rest of the story . . . .

  1. The iron was reported semi-quantitatively as mild zone 1 hepatocellular hemosiderosis, with a comment that formal quantitation was not pursued but that the quantity would likely have been in the 3000-5000 micrograms of iron/g dry weight liver range.
  2. Active participation in the evaluation of iron overload is usually appreciated by the clinician and can help minimize costs of the work-up. A phone call was made and there was a discussion about optimum work-up with the clinician.
  3. The following tests (with results) were performed:
    • Serum Ferritin 1031
    • Serum iron 122, iron binding capacity 256, 48% saturation
    • HFE genotyping – C282Y/H63D compound heterozygote

The final impression is that this patient has hereditary hemochromatosis with a potentially clinically significant genotype but relatively mild penetrance. While she may never have suffered morbidity from the iron, a phlebotomy program was undertaken. Screening of first degree relatives with serum iron saturation assay was advised.

References

  1. von Recklinhausen FD, Uber hemochromatose. Tagelb Versamml Natur Artze Heidelberg 1889;62:324-325.
  2. Sheldon JH. Haemochromatosis. London, Osford University Press, 1935.
  3. Davis WD, Arrowsmith WR. The effect of repeated phlebotomomies in hemochromatosis; report of three cases. J Lab Clin Med 1952;39:526-532.
  4. Scheuer PJ, Williams R, Muir AR. Hepatic pathology in relatives of patients with hemochromatosis. J Pathol Bacteriol 1962;84:53-64.
  5. Simon M, Pawlotsky Y, Bourel M, Fauchet R, Genetet B. Hemochromatose idiopathique: maladie associee a l'antigene tissulaire HLA 3. Nouv Presse Med 1975;19:1432.
  6. Feder JN, Gnirke A, Thomas W, et al. A novel MHC class-1-like gene is mutated in patients with hereditary hemochromatosis. Nat Genet 1996;14:399-408.
  7. Edwards CQ, Ajioka RS, Kushner JP. Hemochromatosis: a genetic definition. In, Hemochromatosis. Genetics, pathophysiology, diagnosis, and treatment. Barton JC, Edwards CQ, eds. Cambridge University Press, 2000; 8-11.
  8. Fairbanks VF. Population Genetics. In, Hemochromatosis. Genetics, pathophysiology, diagnosis, and treatment. Barton JC, Edwards CQ, eds. Cambridge University Press, 2000; 42-50.
  9. Deugnier YM, Loreal O, Turlin B, et al. Liver pathology in genetic hemochromatosis: A review of 135 homozygous cases and their biochemical correlations. Gastroenterology 1992; 102:2050-9
  10. Deugnier YM, Guyader D, Crantock L, et al. Primary liver cancer in genetic hemochromatosis: A clinical, pathogenetic study of 54 cases. Gastroenterology 1993; 104:228-34.
  11. Niederau C, Fischer R, Purschel A, Stremmel W, Haussinger D, Strohmeyer G. Long-term survival in patients with hereditary hemochromatosis. Gastroenterology 1996; 110:1107-19.
  12. Powell LW, Fletcher LM, Halliday JW. Distinction between haemochromatosis and alcoholic siderosis. In: Hall P, ed. Alcoholic Liver Disease. 2nd ed. 1995:199-216.
  13. Charlton RW, Jacobs P, Seftel H, Bothwell TH. Effect of alcohol on iron absorption. British Med J 1964; 2:1427-9.
  14. Celda A, Rudolf H. Donath A. Effect of a single ingestion of alcohol on iron absorption. Am J Hematol 1978; 5:225-37.
  15. Duane P, Raja KB, Simpson RJ, Peters TJ. Intestinal iron absorption in chronic alcoholics. Alcohol and Alcoholism 1992; 27:539-44.
  16. Ludwig JL, Hashimoto E, Porayko MK, Moyer TP, Baldus WP. Hemosiderosis in cirrhosis: A study of 447 native livers. Gastroenterology 1997; 112:882-8.
  17. Bassett ML, Halliday JW, Powell LW. Value of hepatic iron measurements in early hemochromatosis and determination of the critical iron level associated with fibrosis. Hepatology 1986; 6:24-9.
  18. Deugnier YM, Turlin B, Powell LW, et al. Differentiation between heterozygotes and homozygotes in genetic hemochromatosis by means of a histological hepatic iron index: A study of 192 cases. Hepatology 1993; 19:30-4.
  19. Olynyk J, Hall P, Sallie R, Reed W, Shilkin K, Mackinnon M. Computerized measurement of iron in liver biopsies: A comparision with biochemical iron measurement. Hepatology 1990;12:26-30.
  20. LeSage GD, Baldus WP, Fairbanks VF, et al. Hemochromatosis: Genetic or alcohol-induced? Gastroenterology 1983; 84:1471-7.
  21. Irving MG, Halliday JW, Powell LW. Association between alcoholism and increased hepatic iron stores. Alcoholism: Clinical and Experimental Research 1988; 12:7-13.
  22. Adams PC, Agnew S. Alcoholism in hereditary hemochromatosis revisited; prevalence and clinical consequences among homozygous siblings. Hepatology 1996; 23:724-7.