Search

Although guidelines are available for hereditary hemochromatosis, a high percentage of the recommendations within them are not shared between the different guidelines. Our main aim is to provide an objective, simple, brief, and practical set of recommendations about therapeutic aspects of HFE hemochromatosis for p.Cys282Tyr (C282Y/C282Y) homozygous genotype, based on the published scientific studies and guidelines, in a form that is reasonably comprehensible to patients and people without medical training. This final version was approved at the Hemochromatosis International meeting on 12th May 2017 in Los Angeles.

The aim of this epidemiologic population survey was to assess the penetrance of the most frequent hemochromatosis (HFE) gene variants in ethnic Danish men. A cohort of 6,020 men aged 30-53 years was screened for HFE C282Y, H63D, and S65C variants by restriction fragment length polymorphism analysis. Subsequently, iron status markers (serum transferrin saturation, serum ferritin) were analyzed in 1,452 men. The C282Y allele was present in 5.6%, H63D in 12.8%, and S65C in 1.8% of the men. We found 23 out of 6,020 (0.38%) C282Y homozygotes, of whom two had been treated with phlebotomy. Among untreated C282Y homozygotes (n = 21) with available iron status markers (transferrin saturation n = 18, ferritin n = 16), 89% had elevated transferrin saturation >or=50%, 94% had elevated ferritin >or=300 microg/L, and 88% had elevation of both iron status markers; seven out of 16 (44%) had ferritin values >800 microg/L. One C282Y homozygote had normal iron status markers possibly due to nonexpressivity. Among C282Y/H63D compound heterozygotes (n = 66), 23% had elevated transferrin saturation, 27% elevated ferritin, and 9% elevation of both iron status markers. Among H63D/H63D homozygotes (n = 74), 15% had elevated transferrin saturation, 19% elevated ferritin, and 5.4% elevation of both iron status markers. Among C282Y/wild type (wt) heterozygotes (n = 255), 9% had elevated transferrin saturation, 9% elevated ferritin, and 1.2% elevation of both iron status markers. Among H63D/wt heterozygotes (n = 600), 8% had elevated transferrin saturation, 12% elevated ferritin, and 2% elevation of both iron status markers. None of the men with the S65C variant displayed elevation of both iron status markers. In conclusion, this study demonstrates a high penetrance of the C282Y variant in Danish men followed by the H63D variant while the S65D variant had no significant impact on iron status markers.

Discovery of genetic variants of the HFE gene has made it possible to screen for hereditary hemochromatosis. However, genetic screening raises ethical, legal, social, and psychological questions, which need to be addressed. To assess the psychological impact on individuals undergoing genetic screening for hereditary hemochromatosis and to determine the effects of providing different levels of information to the participants. Male residents, aged 30-50 years (n = 10,993) were invited to a genetic screening for hereditary hemochromatosis. Carriers of the gene variants H63D, S65C, and C282Y were offered additional biochemical screening using serum ferritin and transferrin saturation. Psychological factors were evaluated through questionnaires before and after genetic and biochemical screening. According to genotype, participants were divided into three groups with different risks profiles for having/developing clinical hemochromatosis (at-risk, uncertain risk, no risk). Before completion of the questionnaires, 929 participants received only genetic information and 366 received both genetic and biochemical information. At-risk participants receiving only genetic information generally displayed negative reactions to the test result, whereas participants receiving both genetic and biochemical information were more satisfied and had fewer negative psychological reactions to the test result. Genetic screening is more readily accepted by subjects when genetic information is supported by biochemical measurements, especially in participants with a risk of clinical disease.

The aim of the study was to assess the frequencies of the hereditary hemochromatosis HFE mutations C282Y, H63D, and S65C in the population in the Faroe Islands. The series comprised 200 randomly selected blood donors of Faroese heritage. The frequency of the C282Y, H63D, and S65C mutations on the HFE gene was assessed by genotyping using the polymerase chain reaction (PCR) technique and calculated from direct allele counting. We found no C282Y homozygous subjects; 28 (14.0%) subjects were C282Y heterozygous and four subjects were C282Y/H63D compound heterozygous (2.0%). The C282Y allele frequency was 8.0% (95% CI 5.3-10.7%). The series contained three (1.5%) H63D homozygous subjects and 60 (30.0%) H63D heterozygous subjects. The H63D allele frequency was 17.5% (95% CI 13.8-21.2%). There were four (2.0%) S65C heterozygous subjects. The S65C allele frequency was 1.0% (95% CI 0.3-2.5%). The frequency of the C282Y mutation is high in Faroese blood donors, being close to and not significantly different from the frequencies reported in other Scandinavian countries: Denmark 5.7%, Norway 6.6%, Iceland 5.1%, and Sweden 6.1%. The frequency of the H63D mutation in Faroese subjects is significantly higher than the frequency in Denmark 12.8% (p=0.007), Iceland 10.9% (p=0.003), and Sweden 12.4% (p=0.015), but not from the frequency in Norway 11.2% (p=0.063). The frequency of the S65C mutation in Faroese subjects is not significantly different from the frequencies in Denmark 1.5% and Sweden 1.6%. Screening of larger groups of the Faroese population for HFE mutations especially C282Y should be considered in order to establish the penetrance.

The aim of the study was to assess the frequency of the C282Y and H63D mutations of the hemochromatosis gene (HFE) in ethnic Danes. The series comprised 2501 subjects (1284 men) of Danish heritage who were drawn at random from the Census Registry in age cohorts of 30, 40, 50, and 60 years. The frequency of the C282Y and H63D mutations was assessed on blood samples by genotyping using a polymerase chain reaction (PCR) technique. The HFE genotype distribution was in Hardy-Weinberg equilibrium (p=0.85). C282Y mutation: 9 subjects (0.36%) were homozygous and 265 subjects (10.6%) were heterozygous. H63D mutation: 40 subjects (1.6%) were homozygous and 584 subjects (23.4%) were heterozygous. C282Y/H63D compound heterozygosity was found in 36 subjects (1.4%). The C282Y allele frequency was 5.7% [95% confidence interval (CI) 5.0-6.3%] and the H63D allele frequency was 13.3% (95% CI 12.3-14.2%). In conclusion, the C282Y frequency is relatively high in the Danes, being close to the frequency in other Scandinavian countries, i.e., Iceland 5.1%, the Faroe Islands 6.6%, and Sweden 5.7%, but significantly lower than in Norway 6.6% (p=0.02). Also, the H63D frequency in Danes is close to and not significantly different from the frequency in Iceland 10.9%, Norway 11.2%, and Sweden 12.4%, but significantly lower than in the Faroe Islands 15.4% (p=0.046).

Aim: To assess the frequency of the C282Y and H63D mutations on the HFE gene in Danish patients with clinical hereditary haemochromatosis initially diagnosed by phenotypic methods., Methods: In the period 1950-1985, an epidemiological survey in Denmark identified 179 patients with clinical idiopathic haemochromatosis diagnosed by phenotypic methods (serum transferrin saturation, serum ferritin, liver biopsy and mobilisable body iron stores). In 32 unrelated patients, frozen blood samples were available for genetic analysis. In a subsequent series of 26 unrelated Danish patients, a phenotypic diagnosis of clinical idiopathic haemochromatosis was made before blood samples were taken for HFE genotyping. The total series consisted of 58 patients (40 men and 18 women) with a median age of 60 yrs (range 18-74). HFE genotyping was performed by the polymerase chain reaction (PCR) technique., Results: Among the patients, 55 of 58 (94.8%) were C282Y/C282Y homozygous. One 63-year-old woman (1.7%) was compound C282Y/H63D heterozygous. Two women (3.4%), aged 42 and 43 yrs were negative for both the C282Y and the H63D mutation., Conclusion: In the Danish population, homozygosity for the C282Y mutation appears to be the prevailing cause of clinically overt genetic haemochromatosis. This finding has implications both for the evaluation of patients with iron overload disorders and for the strategy in future population screening surveys.

Hereditary hemochromatosis has been recognized as a clinical disorder for more than 100 years. The common form of the disorder is caused by the Cys282Tyr mutation (C282Y) of the HFE gene. Hereditary hemochromatosis affects predominantly people of Northern European origin. The C282Y mutation probably occurred on a single chromosome carrying the ancestral hemochromatosis haplotype, which subsequently was spread by emigration and the founder effect. It has been estimated that the C282Y mutation appeared 60-70 generations ago. It was initially suggested that the ancestral C282Y mutation occurred within the Celtic group of peoples. However, we hypothesize that the distribution of the C282Y mutation in Europe is more consistent with an origin among the Germanic Iron Age population in Southern Scandinavia. From this area, the mutation could later be spread by the migratory activities of the Vikings. The aim of the present study was to evaluate the validity of these two hypotheses. Several arguments are in favor of the 'Viking hypothesis': first, the highest frequencies (5.1-9.7%) of the C282Y mutation are observed in populations in the Northern part of Europe, i.e. Denmark, Norway, Sweden, Faeroe Islands, Iceland, Eastern part of England (Danelaw) and the Dublin area, all Viking homelands and settlements. Second, the highest allele frequencies are reported among populations living along the coastlines. Third, the frequencies of the C282Y mutation decline from Northern to Southern Europe. Intermediate allele frequencies (3.1-4.8%) are seen in the populations in Central Europe, which is the original Celtic homeland. Low allele frequencies (0-3.1%) are recognized in populations in Southern Europe and the Mediterranean.

Aim: The frequencies of the hereditary hemochromatosis allele were compared for different populations assessed by previous phenotypic methods and the present genotypic methods., Methods: From a literature survey, the calculated hemochromatosis allele frequencies from 16 studies using phenotypic biochemical markers (threshold levels for transferrin saturation [range, 46%-70%] and serum ferritin [range, 164-700 microg/L]) were compared with allele frequencies of the Cys282Tyr mutation of the hemochromatosis gene reported in 19 genotypic studies., Results: Calculated phenotypic allele frequencies are high in Scandinavia: Iceland, 6.1% to 7.4%; Norway, 5.8%; central Sweden, 6.3% to 6.9%; Denmark, 6.1%. Frequencies are similarly high in Wales, Canada, Utah, South Africa, and Australia (range, 5.2%-9.8%). Frequencies are low in Finland (1.9%) and northern Italy (4.5%). Genotypic allele frequencies of the Cys282Tyr mutation are likewise high in Scandinavia. Frequencies are high in the United Kingdom and northern France and low in Finland, central Germany, northern Italy, and Greece. The phenotypic-genotypic ratios of the hemochromatosis homozygosity frequencies for the same geographic area were calculated. A ratio of 1.0 indicates that the 2 methods give similar results. In 3 studies, the ratio was above 1.0, the highest ratio of 1.67 being reported from Italy. In most studies the ratio was slightly below 1.0 (0.71-0.97). The lowest ratio was found in Finland (0.33)., Conclusion: In most studies there was good agreement between the hemochromatosis allele frequencies determined by phenotypic and genotypic methods. A high ratio (northern Italy) may indicate that phenotypic selection criteria were too loose and/or that causes of iron overload other than the Cys282Tyr mutation are frequent in the region. A low ratio (in Finland) may indicate phenotypic selection criteria that were too stringent and/or a low penetration rate of the mutation.

Analysis of the common C282Y and H63D mutations in the HFE gene is widely used to diagnose hereditary hemochromatosis (HH). The aim of this study was to evaluate the efficiency with which different hospitals and general practitioners select patients for HH genotype and to determine the distribution of HFE mutations in such patients. Nine hundred unrelated patients from Danish hospitals and general practitioners (group A) and 69 consecutive patients from a specialized liver unit (group B) were examined for HFE substitutions using multiplex real-time polymerase chain reaction. In group A we found 13.0% (0%) C282Y homozygotes, 5.8% (2.6%) H63D/C282Y compound heterozygotes and 1.9% (3.1%) S65C heterozygotes. The values for 420 Danish blood donors are shown in parentheses. The distribution of genotypes in group B was similar to that of the blood donors. Serum ferritin, transferrin iron saturation and pathological data were collected from 38 randomly selected C282Y homozygotes, 36 H63D/C282Y compound heterozygotes, 19 H63D heterozygotes, 17 S65C heterozygotes and 144 wild-types. All of the C282Y homozygotes and 28% of the compound heterozygotes were diagnosed as HH patients. There was no evidence of HH in the H63D homozygotes or S65C heterozygotes. Moreover, 7 wild-type patients, 2 C282Y heterozygote patients and one H63D heterozygote patient fulfilled the criteria for HH. The significant enrichment of HH among associated genotype samples submitted for HFE testing indicates that the clinical selection is generally adequate. However, the study showed substantial deviation in the selection efficiency among the various hospitals and general practitioners.

The object was to analyze, in a nationwide survey, the incidence and course of hereditary hemochromatosis in relation to the degree of iron overload and the presence of organ damage. The study included 179 Danish Caucasian patients with clinically overt hemochromatosis diagnosed between 1948 and 1985. A cohort of 158 patients was followed for a median of 8.5 years (range: 0.2-29.5). From 1951 to 1975, the yearly relative incidence rate was constant: 0.58/100,000 persons >20 years of age. From 1981 to 1985, the yearly relative incidence rate rose to 1.40/100,000 persons >20 years of age. Survival was reduced in the entire series when compared with a matched control population ( p<0.0001). There was a steady increase in survival from 1948 to 1985 ( p<0.002). Survival was significantly reduced in patients with liver cirrhosis and/or diabetes mellitus ( p<0.01). In contrast, survival in patients without cirrhosis or diabetes was similar to rates expected. Survival in patients with arthropathy was higher than in patients without joint affection ( p<0.004). Patients adequately treated with phlebotomy ( n=66) had a higher survival than inadequately treated patients ( n=62; p<0.0001). Adequately treated patients with cirrhosis and/or diabetes had better survival than inadequately treated patients with similar organ damage ( p<0.001). The main causes of death were hepatic failure due to cirrhosis (32.0%) and cirrhosis with liver cancer (23.1%). Sharpened diagnostic awareness has improved early diagnosis and increased the diagnostic frequency of clinical hemochromatosis. Adequate phlebotomy treatment was the major determinant of survival and markedly improved prognosis. Early detection and treatment of this common iron overload disorder is crucial and can completely prevent any excess mortality caused by hemochromatosis.

A 54 year-old previously healthy woman was admitted with staphylococcus aureus septicaemia. The patient had been treated with oral iron supplementation for two years due to fatigue. In the evaluation of postinfectious anaemia, serum transferrin saturation and serum ferritin were found persistently elevated with values of 74% and 950 micrograms/1, respectively. Hereditary haemochromatosis was suspected even though there was no history of liver disease or diabetes mellitus in the family. A bone marrow biopsy showed a normal content of haemosiderin iron. The liver biopsy revealed haemosiderosis, mainly located to the periportal hepatocytes, and fibrosis in the portal tracts. The HLA-type was A3, B7, B37. Over a period of ten months, a total of 3.9 g of iron was removed by venesection while S-ferritin declined to 31 micrograms/l. A sister to the proband had an identical HLA type, but normal iron status markers, either indicating heterozygosity or homozygosity with lack of penetrance. In preclinical hereditary haemochromatosis, early diagnosis and treatment is essential in order to prevent organ damage and to improve prognosis. Prophylactic screening is recommended. The identification of one homozygous subject in a Danish year-cohort of 60.000 persons costs approximately 40.000 Danish kroner (7.000 US+).

The purpose of the present study was: 1) to evaluate which of the four iron status markers (serum iron, serum transferrin, serum transferrin saturation, serum ferritin) displayed the highest discriminatory potential in the distinction between homozygous patients with hereditary haemochromatosis and heterozygous relatives, 2) to suggest optimum cut-off values for these iron status markers, and 3) to demonstrate how these cut-off values change if the expected utility from a correct diagnosis is incorporated into the analysis. The patients and relatives were found by a nation-wide epidemiological survey. The study population consisted of 162 patients with clinically overt hereditary haemochromatosis and 84 asymptomatic heterozygous relatives. The statistical evaluation was performed using receiver operating characteristic (ROC) curve analysis. The diagnostic power of the iron markers is expressed as the area under the ROC curve. The optimum cut-off value is at the point where the slope of the ROC curve is equal to one. Changes in the optimum cut-off value at varying expected utility from a correct classification was estimated by changing the scaling of the ROC diagram. Serum iron and serum transferrin had the smallest area under the ROC curve, and were both unsuitable as discriminators. Near complete discrimination was obtained with serum transferrin saturation and serum ferritin concentrations, displaying the largest area under the ROC curve (0.991 and 0.998). The optimum threshold value for transferrin saturation was 61%, and for serum ferritin concentration 800 micrograms/l. The transferrin saturation level reflects the presence of the haemochromatosis allele, whereas the serum ferritin concentration indicates the degree of iron overload.(ABSTRACT TRUNCATED AT 250 WORDS)

The purpose of this pedigree study, comprising 29 families with hereditary haemochromatosis (HH), was to evaluate the relationship between the genotype (G), based on HLA typing, and the phenotype, based on measurement of iron status markers (serum transferrin saturation and serum ferritin). Due to tight linkage between the HH locus and the HLA-A locus, 172 relatives of the 29 unrelated probands could be assigned into three groups: G0 who were considered to be normal (n = 53), G1 who were considered to be heterozygotes (n = 105), and G2 who were considered to be homozygotes (n = 14), according to whether they had no, one or two HLA haplotypes in common with the proband. A high serum transferrin saturation (> 60%) was present in 8/14 = 57.1% of the homozygotes, in 11/105 = 10.5% of the heterozygotes, and in 0/53 = 0% of the normals. Of the homozygotes, 8/14 = 57.1% had preclinical disease, 4/14 = 28.6% had clinically overt iron overload, while 2/14 = 14.3% had normal iron status markers. None of the heterozygotes had clinical evidence of iron overload. Analysis of HLA alleles and iron status markers suggested that 11/105 = 10.5% subjects initially classified as heterozygotes (G1) according to HLA typing should be reclassified as homozygotes because of abnormal iron status markers, explained by either: homozygous x heterozygous (n = 7) or heterozygous x heterozygous (n = 2) matings, HLA recombination (n = 1) or strongly abnormal iron status markers (n = 1).(ABSTRACT TRUNCATED AT 250 WORDS)

Transferrin (TF) subtypes were determined by isoelectric focusing in 51 unrelated Danish patients with hereditary haemochromatosis (HH) and in 847 normal subjects. The following TF phenotype frequencies were observed in HH patients and controls, respectively: TF*C1, 70.6% vs. 58.8%; TF*C2, 5.9% vs. 2.4%; TF*C3, 0% vs. 0.4%; TF*C1-2, 11.8% vs. 24.7%; TF*C1-3, 5.9% vs. 9.7%; TF*C2-3, 3.9% vs. 2.2%; TF*B-C1, 2.0% vs. 1.5%; TF*B-C2, 0% vs. 0.4%. None of these differences were statistically significant. There was no relationship between the TF subtypes and the clinical or paraclinical expression of disease in HH patients.

HLA-A and -B alleles in 74 Danish patients and 21 homozygous relatives with idiopathic haemochromatosis (IH) were compared with those in a sample of 1719 chromosomes from healthy Danish control subjects. The following alleles occurred with higher frequencies in IH compared to controls: A3: 53.6% vs. 15.1% (Pc less than 0.001); B7: 33.1% vs. 15.6% (Pc less than 0.001); B14: 6.9% vs. 3.0% (Pc greater than 0.05); B38: 5% vs. 0.9% (Pc greater than 0.05); B47: 4.0% vs. 0.4% (Pc greater than 0.05). Pedigree analyses disclosed 19 different haplotypes in IH subjects, compared to 286 haplotypes in controls. The following haplotypes occurred with higher frequency in IH compared to controls: A3,B5: 10.3% vs. 0.3% (Pc less than 0.001); A3,B7: 25.6% vs. 6.6% (Pc = 0.001); A3,B14: 3.4% vs. 0.6% (Pc greater than 0.05); A3,B47: 6.9% vs. 0.2% (Pc greater than 0.05). The major IH marker HLA-A3 was found in 56% of the haplotypes. The patterns of HLA-alleles associated with IH in Denmark show similarities to those in Central Europe, Australia, USA and Canada, being A3,B7 dominated and those in Central Sweden, England and Ireland, being A3,B14 dominated.

Iron status markers, serum iron, serum transferrin, transferrin saturation and serum ferritin were analysed in 162 homozygous patients with clinical haemochromatosis, in 12 homozygous relatives with preclinical haemochromatosis, in 84 heterozygous, and in 9 normal subjects. In the distinction between homozygous patients with clinical haemochromatosis and heterozygotes, transferrin saturation and serum ferritin showed the highest diagnostic efficiencies. A diagnostic efficiency of 0.97 was obtained with a transferrin saturation value of 60%. A discriminatory transferrin saturation value of greater than 50% correctly classified 173 out of the 174 individuals with preclinical + clinical haemochromatosis, whereas a discriminatory value of greater than 60% identified 158 out of the 162 patients with clinical haemochromatosis. All patients with clinical haemochromatosis had transferrin saturation values greater than 50%. 10 heterozygotes had transferrin saturation values greater than 50%, and 3 values greater than 60%. Normal subjects had transferrin saturation values less than 47%. A diagnostic efficiency of 0.99 was obtained using a discriminatory serum ferritin value of 800 micrograms/l. Patients with clinical haemochromatosis had higher ferritin (p less than 0.001) than subjects with preclinical haemochromatosis, who in turn had higher values than heterozygotes (p less than 0.001). Iron status markers in heterozygotes and normal subjects displayed no significant differences.

During the period 1950-1985, a total of 179 cases of clinically overt hereditary haemochromatosis (HH) were registered in Denmark, 140 males and 39 females. Median age at diagnosis was 55 years (range 29-81). Diagnostic approaches, symptoms and physical signs at discovery are described. All patients had grade 3-4 liver haemosiderin iron, and cirrhosis was present in 84%. Serum (S-) transaminase was elevated in 92%, S-alkaline phosphatase in 47% and S-bilirubin in 23%, while plasma prothrombin time was below normal in 34%. Females had higher alkaline phosphatase than males (p less than 0.05). Bone marrow haemosiderin iron (n = 81) showed no relation to iron status indicators and was unsuitable as a diagnostic tool. Skin biopsy (n = 56) was positive for haemosiderin iron in 67% and for melanin in 57%, but was of limited value in the assessment of HH. Arthropathy was registered in 44%; arthralgias and clinical joint abnormalities occurred more frequently in females than in males (p less than 0.05). Latent diabetes mellitus was found in 34% and overt diabetes in 55%, being more frequent in males than in females (p less than 0.05). Other endocrine abnormalities were seen in 66%. Cardiac failure was observed in 9% and abnormal ECG in 35%. Males had higher haemoglobin (p less than 0.0001) and S-iron (p less than 0.01) than females, while S-transferrin, transferrin saturation, S-ferritin and mobilizable iron stores showed no significant sex differences. Median transferrin saturation was 87% (range 52-100); values greater than 62% were observed in 96% of the patients. Median S-ferritin was 3,400 micrograms/l (800-12,700) and median iron stores 14.8 g (4.5-36.4).

Pedigree studies were performed based on one Faroese and four Danish probands with overt idiopathic hemochromatosis (IH). The study consisted of HLA typing and determination of biochemical iron status indicators (serum transferrin saturation, serum ferritin). In total, 130 persons were evaluated. The screening identified 6 homozygous (h/h) subjects with preclinical IH, 46 heterozygous (h/n), and 8 normal (n/n) subjects, while 39 subjects were classified as normal or heterozygous (n/h?). One family demonstrated both a homozygous x heterozygous as well as a heterozygous x heterozygous mating. Recombination between the HLA region and IH locus occurred possibly in three subjects in three different families. The significance of detailed screening in families with probands with IH is discussed.

HLA-A, -B, -C and -DR antigens were determined in 70 unrelated Danish patients with idiopathic haemochromatosis. The frequencies of HLA-A and -B antigens compared to 1967 normal control subjects and the relative risk values (RR) were: A3, 80.0% vs. 26.9% (P less than 0.0001), RR = 10.9; B7, 60.0% vs. 26.8% (P less than 0.0001), RR = 4.1; B14, 10.0% vs. 4.5% (P = 0.03), RR = 2.4; B47, 4.3% vs. 0.5% (P less than 0.0001), RR = 9.7; A3, B7, 51.4% vs. 12.2% (P less than 0.0001), RR = 7.6; A3, B14, 10.0% vs. 1.4% (P less than 0.0001), RR = 7.7; A3, B47, 4.3% vs. 0.5% (P less than 0.0001), RR = 9.7. Six patients (8.6%) possessed none of these four typical antigens. There was no association between disease and the frequencies of HLA-C and HLA-DR antigens. The pattern of HLA-antigens associated with haemochromatosis in Denmark shows similarities to those reported both in Germany, being HLA-A3, B7 dominated, and in Brittany, Great Britain and Central Sweden, being HLA-A3, B14 dominated.

HLA-antigens were determined in 21 unrelated patients with idiopathic haemochromatosis and in eight siblings and 13 children of the probands. The prevalences of HLA-A3, B7, and B14 in patients compared to 1967 healthy control subjects were: A3, 76.2% versus 26.9% (p less than 0.0001); B7, 57.1% versus 26.8 (p less than 0.001); B14, 9.5% versus 4.5% (n.s.); A3 and B7, 42.9% versus 12.2% (p less than 0.0001); A3 and B14, 9.5% versus 1.4% (p less than 0.001). Siblings (n = 3) that were HLA-identical with the proband were considered to be homozygotes for the haemochromatosis allele and presented with preclinical haemochromatosis. Siblings and children (n = 17) having only one HLA-haplotype in common with the proband were considered to be heterozygotes. Biochemical markers for haemochromatosis (transferrin saturation and serum ferritin) were higher in homozygous than in heterozygous subjects (p less than 0.0001). The results confirm the association between the HLA-A and B loci and the haemochromatosis gene. HLA-typing is a valuable tool in the identification of the haemochromatosis genotype in a family, and it is an adjunct to the biochemical screening procedure in relatives of patients with this iron overload disorder.