|Home | About | Journals | Submit | Contact Us | Français|
To investigate the prevalence of antibodies to cyclic citrullinated peptide (anti‐CCP) and rheumatoid factor in patients with hereditary haemochromatosis (HHC) and to evaluate their diagnostic reliability in distinguishing HHC‐associated arthropathy from rheumatoid arthritis.
Anti‐CCP antibodies and rheumatoid factor levels were determined by ELISA in sera from 87 patients with HHC homozygous for the C282Y mutation of the HFE gene, 31 patients with rheumatoid arthritis and 162 healthy controls.
Of the 87 patients with HHC, 32 (36.8%) had joint involvement. Anti‐CCP antibodies were detected in only 1 patient (1.1%) with HHC, who had no joint disease, and in (1.2%) healthy controls. In total, 18 (58.1%) patients with rheumatoid arthritis displayed anti‐CCP reactivity (p<0.001). Rheumatoid factor was detected in 10 (11.5%) patients with HHC compared with 7 (4.3%) healthy control subjects (p=0.03) and 21 of 31 (65.6%) patients with rheumatoid arthritis.
Testing for anti‐CCP antibodies discriminates HHC arthropathy from rheumatoid arthritis, as these patients were consistently anti‐CCP negative. Thus, HHC arthropathy should be considered in the differential diagnosis of CCP‐negative arthritis.
Hereditary haemochromatosis (HHC) is most commonly caused by the C282Y mutation in the HFE gene1 and is characterised by progressive iron accumulation in various organs including joints.2 Arthropathy is the most frequent condition affecting health‐related quality of life in patients with HHC. HHC arthropathy typically affects the second and third metacarpophalangeal joints, but larger joints such as the wrists, knees, hips, shoulders or ankles may also be affected.2 HHC arthropathy can resemble inflammatory arthritis mimicking rheumatoid arthritis, particularly at its most common site, the second and third metacarpophalangeal joints.3
Differential diagnosis between rheumatoid arthritis and HHC arthropathy can be challenging for several reasons. (1) HHC arthropathy can present as true synovitis with symmetrical swelling of metacarpophalangeal joints, which is a clinical presentation that is difficult to distinguish from rheumatoid arthritis. (2) The presence of rheumatoid factor, one of seven American College of Rheumatology (ACR) diagnostic criteria for rheumatoid arthritis, has been observed in HHC arthropathy.4,5 (3) Radiographic changes in HHC arthropathy frequently involve hook‐like osteophytes; however, these lesions are not consistent and, particularly when inflammatory changes dominate, subchondral bone erosions and joint space narrowing can occur, which resemble the radiographic changes found in rheumatoid arthritis. (4) There is no correlation between the extent of iron deposition and the severity of clinical, histological or radiographic changes in the joints of people with HHC arthropathy.6 (5) Ferritin, a marker of iron overload, may be raised in other inflammatory conditions such as the various rheumatic diseases.7 Hence, diagnostic pitfalls can arise in differentiating rheumatoid arthritis from HHC arthropathy, delaying appropriate treatment.
Based on the similarities between HHC arthropathy and rheumatoid arthritis , we sought potential markers to differentiate between these diseases. Anti‐cyclic citrullinated peptide (anti‐CCP) antibodies are highly specific for rheumatoid arthritis, with specificities ranging from 95 to 99%.8 Anti‐CCP antibodies are directed against proteins containing the amino acid citrulline, are likely to play a role in immunopathogenesis of rheumatoid arthritis9 and have proven helpful in distinguishing rheumatoid arthritis from other rheumatic diseases. We therefore determined rheumatoid factor and anti‐CCP in patients with HHC‐associated arthropathy, aiming to examine the value of rheumatoid factor and anti‐CCP in discriminating HHC arthropathy from rheumatoid arthritis.
The study included 87 patients with HHC homozygous for the C282Y HFE mutation (47 men, 40 women; mean (SD) age 46.0 (17.8) and 50.1(21.9) years, respectively). Joint involvement in patients with HHC was defined as 1 tender and/or swollen joints in the absence of trauma during the previous 2 months, or a history of synovectomy and/or joint replacement for arthritis.
Control groups consisted of 31 patients (8 men, 23 women, mean (SD) age 46.5 (11.3) years) fulfilling ACR criteria for rheumatoid arthritis recruited from the local rheumatology clinic and 162 healthy controls (91 men, 71 women; 52.5 (5.7) years). Healthy controls were recruited from the same Central European Caucasian background as the patients and did not show clinical or biochemical signs of rheumatological, metabolic, autoimmune, infectious or malignant disease.
In patients and healthy controls, PCR‐based HFE gene‐mutation analysis was performed as described previously.1 Antibodies against CCP and rheumatoid factor (Euroimmun Medizinische Labordiagnostika AG, Lübeck, Germany) were measured by ELISA microplate techniques. Anti‐CCP antibodies were considered positive at a cut‐off level of 5 rheumatology units (RU)/ml according to the manufacturerś instructions. For rheumatoid factor, a concentration >20 U/ml was considered positive according to the manufacturerś instructions and a value >50 U/ml was considered a high‐titre rheumatoid factor according to ACR criteria for rheumatoid arthritis.10
Data were summarised as means (SD) and tested for differences by Student's t test or analysis of variance. Skewed data as assessed by the Kolmogorov–Smirnoff test are presented as median and range. Comparison of these groups or groups with unequal variances, as assessed by Levine's test, was carried out using the Mann–Whitney U test or Kruskal–Wallis H test. Categorical data were summarised by frequencies and analysed by χ2 or Fisher's exact test. Significance was set at p=0.05 (two‐tailed). SPSS V.10.0 (SPSS Inc., Chicago, IL, USA) was used for all analyses.
HFE genotyping of 31 patients with rheumatoid arthritis showed that 2 (6.5%) were heterozygous for the C282Y mutation and 4 (12.9%) were heterozygous for the H63D mutation. None of these patients was compound heterozygous or homozygous for the C282Y or H63D mutation. Of the healthy controls, 13 (8.0%) and 22 (13.6%) were heterozygous for the C282Y and H63D mutations, respectively, but none was compound heterozygous or homozygous for either mutation. HFE mutation carrier rates in rheumatoid arthritis and HC patients were comparable to carrier rates of the local Central European population.1
Of 87 C282Y HFE homozygotes, 32 (37%) presented with arthropathy. Most of these patients (n=29; 91%) showed symmetrical joint involvement and 21 (58%) displayed polyarticular symmetrical joint involvement affecting at least two joints on each side. In total, 18 (56%) patients showed classical radiological and clinical involvement of the metacarpophalangeal joints, and 17 (53%) patients had undergone orthopaedic surgery because of progressive inflammatory arthritis or osteoarthritis. Mean (SD) age at time of surgery was 50.2 (6.5) years. Of the 17 patients, 10 (59%) were men who had received bilateral hip or knee replacement before the age of 50. Moreover, 9 patients (28%) had involvement of the talocrural joints, two of which had required surgery.
Patients with HHC did not differ from healthy controls with respect to anti‐CCP antibodies. Anti‐CCP antibodies were positive in one (1.1%) patient with HHC and two healthy control subjects (1.2%). The anti‐CCP‐positive patient with HHC did not display any joint symptoms. In contrast, 18 (58%) patients with rheumatoid arthritis had detectable anti‐CCP autoantibodies (p<0.001 compared with patients with HHC and healthy controls) (table 11).).
Positive rheumatoid factor (>20 IU/ml) was detected in 22 (25%) patients with HHC, which was significantly higher than in healthy controls (22 subjects, 13%, P=0.021). All patients with rheumatoid arthritis tested positive for rheumatoid factor >20 U/ml (p<0.001 compared with healthy controls and patients with HHC). Using a higher cut‐off point (>50 IU/ml) in accordance with ACR guidelines,10 10 (11.5%) of the 87 patients with HHC tested positive for rheumatoid factor, compared with 7 healthy controls (4.3%, p=0.03) and 21 of the patients with rheumatoid arthritis (65.6%, p<0.001). Thus, with a cut‐off level >50 U/ml, the sensitivity of positive rheumatoid factor was 65.7% and the specifity was 95.7%, respectively. Among patients with rheumatoid factor >50 U/ml, patients with rheumatoid arthritis tended to have higher rheumatoid factor titres than healthy controls and HHC, although this difference was not statistically significant. Titres of patients with HHC and healthy controls with elevated rheumatoid factor (>50 U/ml) were similar (figure 11).). There was no difference between patients with HHC with or without arthropathy in prevalence and titre of rheumatoid factor.
This study evaluates the diagnostic utility of anti‐CCP antibody and rheumatoid factor in a cohort of patients with HHC. Owing to the comparable prevalence of HHC and rheumatoid arthritis in Caucasian populations and overlaps in clinical presentation, HHC arthropathy may mimic rheumatoid arthritis and could lead to misdiagnosis.4,5,11,12 The virtual absence of anti‐CCP antibodies in the HHC cohort provides a further piece of evidence for the diagnostic usefulness of anti‐CCP antibodies in patients with CCP‐negative arthritis. Anti‐CCP antibodies were virtually absent in HHC, whether or not HHC was associated with joint disease. The only patient who tested positive for anti‐CCP antibodies in our HHC cohort did not display any sign of joint disease at the time when the study was performed. It is possible that this patient will develop rheumatoid arthritis later in life, as anti‐CCP antibodies may occur years before disease onset.13 Thus, HHC arthropathy should only be considered in the differential diagnosis of CCP‐negative arthropathy.
Development of anti‐CCP antibodies is considered an immunopathological process with a high specificity for rheumatoid arthritis. Citrullination of proteins is found in a wide range of inflammatory tissues, suggesting that this process is inflammation‐dependent rather that disease‐dependent.14 Autoimmunity against citrullinated proteins, however, depends on an immune system capable of recognising these modified proteins as foreign to initiate an immune reaction against citrulline residues. Thus, in the case of HHC arthropathy, citrullination of proteins may occur in the context of inflammation but lack the support from the immune system to recognise these residues and initiate an immune response.
The significantly increased prevalence of rheumatoid factor (11.5% vs. 4.3%, p=0.03) in HHC suggests that the use of rheumatoid factor to discriminate HHC arthropathy from rheumatoid arthritis is limited. Thus, even rheumatoid factor‐positive patients presenting with signs of arthritis may reflect HHC rather than rheumatoid arthritis. This clearly favours the use of anti‐CCP in discriminating these two disease entities. HHC arthropathy appears also relevant for differential diagnosis of undifferentiated arthritis. Based on recent findings in cohorts of patients with undifferentiated arthritis, anti‐CCP antibodies are one of the risk predictors for progression into rheumatoid arthritis.15 Thus, in case of undifferentiated arthritis only those patients with absent anti‐CCP antibodies should be considered as potential cases of HHC arthropathy.
In conclusion, we report a high prevalence of positive rheumatoid factor in a cohort of patients with HHC caused by HFE C282Y mutations. The prevalence of rheumatoid factor was not related to the presence of joint disease in patients with HHC. In contrast, anti‐CCP antibodies were found only in patients with rheumatoid arthritis, but were virtually absent in HHC. Therefore, HHC arthropathy should only be considered as a differential diagnosis of anti‐CCP negative arthritis.
ACR - American College of Rheumatology
CCP - cyclic citrullinated peptide antibodies
HHC - hereditary haemochromatosis