Hereditary hemochromatosis is a genetically heterogeneous disease. In northern European populations, the majority of hereditary hemochromatosis cases are attributed to C282Y and H63D mutations of the HFE gene that lies on chromosome 6 (hemochromatosis type 1 [HFE1 hemochromatosis]) (1). Recently, 3 newly identified mutations of the transferrin receptor 2 gene in hereditary hemochromatosis have been described (hemochromatosis type 3 [HFE3 hemochromatosis]) (2).
Juvenile hemochromatosis (hemochromatosis type 2 [HFE2 hemochromatosis]) is a distinct entity both clinically and genetically (3). Only 20 families have been described in the literature to date (4). The responsible gene has not yet been identified, but linkage analysis studies have located the genetic locus on chromosome 1q21 (5). In all types of hemochromatosis, iron accumulation leads to cirrhosis, diabetes mellitus, cardiac dysfunction, and hypogonadotrophic hypogonadism, the latter two of which are more severe in juvenile hemochromatosis.
Arthritis is a well-studied clinical feature of hereditary hemochromatosis (6), but the pathogenesis of arthritis in hereditary hemochromatosis remains a matter of controversy (7). Most clinical studies have included only patients with hemochromatosis type 1, and in some the exact genetic background of the subjects is unclear (8). The reported frequency of arthritis in patients with hereditary hemochromatosis has varied from 40% to 70% (9). The arthritis may precede other symptoms, and involvement of the second and third metacarpophalangeal (MCP) joints is characteristic. It has been asserted that arthritic involvement is absent in juvenile hemochromatosis (7). This study was undertaken to evaluate bone and joint involvement as well as osteopenia in patients with genetically proven juvenile hemochromatosis.
PATIENTS AND METHODS
- Top of page
- PATIENTS AND METHODS
Eight Greek patients (2 men and 6 women) from 5 unrelated families, with an age range of 19–39 years (mean 26 years) at the time of diagnosis, were studied. At the time of this study 2 patients had died of cardiac failure; the data on these patients were retrieved from archived clinical records, and thus were incomplete.
The diagnosis of juvenile hemochromatosis was established by the following criteria: 1) early onset of clinical symptoms (usually in the second decade of life), 2) severity of iron overload as assessed by serum ferritin level, transferrin saturation, liver biopsy findings, and amount of iron mobilized by phlebotomies, 3) wild-type HFE genotype, and 4) linkage analysis studies of family members proving linkage to markers of chromosome 1q. The C282Y and H63D HFE mutations were studied on genomic DNA using polymerase chain reaction–based tests and restriction enzyme digestion with Rsa I and Mbo I. Chromosome 1q linkage analysis was performed by studying the intrafamiliar segregation of D1S442, D1S2344, D1S498, D1S1156, and GATA13C08 markers, mapping within the critical region (10).
All patients studied were evaluated by a rheumatologist for musculoskeletal symptoms, articular involvement, and clinical manifestations of rheumatic diseases. Radiography of the affected joints was performed based on the clinical manifestations and findings. A complete review of the patients' history was performed, including age at presentation of juvenile hemochromatosis manifestations, presenting symptoms, age at arthropathy onset if applicable, amount of iron mobilized by phlebotomies, and history of hormone replacement therapy.
To exclude other concomitant rheumatic diseases or metabolic disorders that are associated with articular manifestations or osteoporosis, erythrocyte sedimentation rate, C-reactive protein, complete blood cell count, uric acid, alkaline phosphatase, fasting serum glucose, thyroid-stimulating hormone, parathyroid hormone (PTH), calcium, phosphorus, rheumatoid factor (RF), and antinuclear antibodies were measured, and serum protein electrophoresis and liver and renal function tests were performed. Hypogonadism was diagnosed by clinical and laboratory assessment, including determination of the presence of symptoms such as amenorrhea in women and sexual dysfunction and infertility in men, confirmed by low values of follicle-stimulating hormone (FSH), luteinizing hormone (LH), estradiol, or testosterone.
Bone mineral density (BMD) was evaluated by dual-photon absorptiometry of the lumbar spine. Osteopenia was defined as a T score between −1 and −2.5, and osteoporosis as a T score below −2.5.
- Top of page
- PATIENTS AND METHODS
None of the patients carried mutations of the HFE gene. Furthermore, the intrafamilial segregation of marker alleles located within the juvenile hemochromatosis critical region at chromosome 1q was in accordance with linkage of the disease to the HFE2 locus. Thus, all of our patients fulfilled clinical and genetic criteria for the diagnosis of juvenile hemochromatosis.
Seven of the 8 patients with juvenile hemochromatosis reported articular symptoms, arthralgias, and/or arthritis, whereas none of the patients had symptoms of periarticular joint involvement or enthesopathy. Two of 6 patients reported diffuse bone pain (information regarding this symptom was not available for the other 2 patients). Clinical, laboratory, and radiologic features of the patients with juvenile hemochromatosis are shown in Table 1.
Table 1. Clinical, laboratory, and radiologic features of 8 Greek patients with juvenile hemochromatosis*
|Patient, sex/family||Presenting symptom||Age at disease onset||Age at arthritis/arthralgia onset||MCP joint||Other joints involved||Radiologic abnormalities found in MCP joints||BMD, T score||Hypogonadism||Iron removed by phlebotomy, gm||Liver disease/hepatic siderosis (I–IV)||Hormone replacement therapy|
|1, M/I||Ejaculation failure||21||23||+||Knees||+||−2.7||+||10.1||Fibrosis (III–IV)||+|
|2, F/II||Arthritis||39†||28||+||Shoulder, lumbar spine, MTP||+||−2.8||+||>8.2||Fibrosis (III–IV)||−|
|4, F/III||Amenorrhea||33||45||+||−||+||−2.08||+||9.5||Fibrosis (III–IV)||+|
|5, M/IV||Infertility||25||NA||+||NA||NA||NA||+||4.0||Fibrosis (III–IV)||+|
|8, F/V||Amenorrhea||19||NA||NA||NA||NA||NA||+||NA||Cirrhosis (III–IV)||−|
The age at arthropathy onset ranged from 20 to 45 years (mean 29.6). In 2 cases arthropathy preceded other symptoms of juvenile hemochromatosis.
The involved joints were knees (1 patient), lumbar spine (1 patient), shoulder (2 patients), metatarsophalangeal joint (1 patient), and MCP joints (6 patients). The arthritis that mainly affected the MCP joints was mildly deforming, causing pain on motion with functional impairment of the hands, resembling inflammatory arthropathy. The acute attacks of the affected joints were usually of short duration.
Radiologic findings in the affected MCP joints, especially the second and the third, were hook-like radial osteophytes, cystic lesions with sclerotic walls, joint space narrowing, and osteoporosis. Only 1 patient (patient 2) had chondrocalcinosis (Figure 1).
Figure 1. Radiographs of the hands of patients with juvenile hemochromatosis A, Posteroanterior view of the hands of patient 1, showing cystic lesions of the wrist bones (particularly on the left), flattening of the heads of the fourth and fifth metacarpophalangeal (MCP) joints, and narrowing of the joint space. Cystic lesions on the head of the left second metacarpal bones and flattening of the heads of the metacarpal bones of both hands are also seen. B, Posteroanterior view of the hands of patient 2, showing marked loss of joint space. A characteristic hook-like radial osteophyte is present on the head of the right fifth metacarpal bone. There is mild flattening of the first and second metacarpal heads, as well as amorphous calcification in the soft tissue of the second proximal interphalangeal joint on the right hand, and small subchondral cysts in the second and fifth MCP heads.
Download figure to PowerPoint
Levels of FSH, LH, testosterone, or estradiol were abnormal since all of the patients had hypogonadotrophic hypogonadism, and levels of liver enzymes that were moderately elevated in 7 of 8 patients. RF was positive in 1 patient (patient 5). Findings of other laboratory tests were within normal limits.
At the time of evaluation all living patients (6 of 8) were iron depleted (serum ferritin <50 ng/μl). Arthritis was progressive in 5 of 6 patients, despite iron depletion. One patient reported improvement of his articular symptoms after phlebotomy therapy.
BMD was measured in 6 patients. Two of them had T scores below −2.5 and were considered to be osteoporotic, and 2 were osteopenic. As indicated above, all patients had symptomatic hypogonadotrophic hypogonadism, and 6 had received hormone replacement therapy.
- Top of page
- PATIENTS AND METHODS
Arthropathy is one of the main clinical manifestations of hereditary hemochromatosis and has been reported to occur in ∼50% of patients (9). To date there has not been a published study on musculoskeletal manifestations of genetically proven juvenile hemochromatosis. In our study, 7 of 8 juvenile hemochromatosis patients had arthropathy, with a pattern like that seen in hereditary hemochromatosis. In the latter, arthritis starts at the small joints of the hands (second and third MCP joints). The initial symptoms are arthralgias after exercise, progressive stiffness, and restriction of MCP joint flexion. Other joints that may be involved are the elbows, shoulders, hips, knees, wrists, and lumbar spine.
The differential diagnosis of hemochromatosis arthritis includes degenerative arthropathy, calcium pyrophosphate dihydrate (CPPD) crystal deposition arthropathy, and rheumatoid arthritis (RA). One of our patients had positive RF, but neither the radiologic findings (i.e., absence of periarticular demineralization as seen in RA) nor the clinical and laboratory evaluation were confirmative of RA.
The pathogenesis of hemochromatosis arthritis has not yet been clarified. Articular cartilage biopsies in patients with hemochromatosis have demonstrated the presence of iron in chondrocytes, as seen by electron microscopy (11). Iron salts promote the nucleation of CPPD crystals and inhibit their removal from the joints (9). However, the absence of correlation between the iron deposits and the radiologic findings, and the progression of the arthropathy independent of iron depletion therapy, provide evidence against a major pathogenetic role of iron in hereditary hemochromatosis arthritis (7).
Some authors have claimed that the absence of arthritis in juvenile hemochromatosis is additional evidence that factors other than iron may be responsible for the arthropathy in hereditary hemochromatosis (7). In contrast, our results indicate that arthropathy is indeed a feature of juvenile hemochromatosis. Juvenile hemochromatosis is genetically distinct from HFE1 hemochromatosis and is linked to a genetic locus that maps to chromosome 1q. This is evidence against a putative role of HFE mutations and their possible metabolic and immunologic effects in the development of arthropathy, as suggested by others (7).
Furthermore, the prevalence of arthritis in our small group of juvenile hemochromatosis patients was higher (87.5%) than that reported for the classic HLA-linked adult form (∼50%). It is well known that juvenile hemochromatosis is a more severe disorder in which iron accumulates earlier in life. This is consistent with the observation that experimental iron loading of rabbits induced cartilage degeneration only when initiated early in life (12). As has been suggested previously (4), it is possible that iron accumulation at a younger age preferentially affects certain tissues (4). Thus, we assume that iron accumulation at an early age may play a pathogenetic role in hemochromatosis arthritis.
In terms of the pathogenesis of osteoporosis in adult patients with hereditary hemochromatosis, at least 3 different mechanisms have been proposed: hypogonadism, liver disease, and iron overload (13). Hypoparathyroidism or elevation of PTH 44–68 levels, which has been reported to correlate with iron overload as assessed by ferritin concentrations (13), is another putative mechanism. Hypoparathyroidism was excluded in our patients, but the putative role of PTH 44–68 was not evaluated. It is possible that the underlying mechanism responsible for osteoporosis or osteopenia in 4 of our patients with juvenile hemochromatosis was the hypogonadotrophic hypogonadism, since none of the patients who were evaluated for osteoporosis had histologic proof or clinical evidence of cirrhosis.
Arthralgias are the most frequent and longest-lasting manifestation of hemochromatosis, especially in juvenile hemochromatosis. The progression of the arthropathy is not modified by treatment, and patients with juvenile hemochromatosis have to cope with persistent arthralgias beginning at a younger age. Further prospective studies, with larger numbers of patients, are needed in order to more strongly establish any correlations between arthropathy and other clinical features of juvenile hemochromatosis.
In conclusion, juvenile hemochromatosis should be considered when evaluating the cause of arthropathy in young patients when other rheumatic diseases have been excluded. We have shown that arthropathy may even precede the other symptoms of juvenile hemochromatosis. As has been suggested by others (14), serum iron studies may be useful in young patients with unexplained hypogonadotrophic hypogonadism and/or arthropathy.