The topic of HCV infection and arthritis has attracted more attention from the rheumatologists than the infectious diseases specialists, yet the prevalence of HCV among patients who have arthritis is comparatively low, whereas the prevalence of arthritis among patients who have HCV infection is comparatively high. Data on the first topic is scant. Maillefert and colleagues  in France found that of 309 patients who had rheumatoid arthritis according to criteria by the American College of Rheumatology (ACR) (formerly the American Rheumatism Association [ARA]), 2 had anti-HCV antibodies and only 1 of them was HCV-RNA positive. Subsequent data by the same group  again showed that of 232 patients seen for initial evaluation of various recent-onset inflammatory arthritides only 6 had anti-HCV antibodies detected. Interestingly, despite this relatively low prevalence, the authors found that it is not infrequent for rheumatolo-gists to screen their patients for HCV . Conversely, data on the prevalence of arthritis among patients who had HCV infection is more abundant but varies widely from 2% to 27% . This wide range is because of important differences in study designs and by itself makes estimation of the true prevalence of HCV arthritis difficult. Rosner and coworkers  underscore these limitations and note that many published studies do not specify the types of arthritis, whether joint involvement is acute or chronic, and whether arthritis is associated with cryoglobulinemia. They also noted that only three of seven studies reviewed by them were prospective and only two of them had a pre-investigation definition of arthritis. In addition, selection bias may be a major problem because most studies include patients selected by attending rheumatology clinics and in that regard they may not be representative of the overall population of HCV-infected subjects. In support of this notion, the prevalence of arthritis in a group of patients who had HCV selected at a unit of infectious diseases was much lower—only 2.2% . Most of these observational studies are not standardized and have flawed study designs, limitations that make generalizations and comparisons difficult. If the true prevalence of arthritis in HCV infection is not well defined or depends on too many extraneous factors, an association of these two variables is even more difficult to measure.
Are arthritic disorders more frequent among patients who have HCV than among comparable noninfected individuals? This case-control approach requires standardized definition of cases and the use of well-matched controls. Because of convenience many studies have used this approach yet few meet these criteria. Recent publications highlight the uncertainties. In 2005, Giordano and colleagues  retrospectively analyzed a cohort of 265 patients who had chronic HCV infection admitted to a unit of infectious diseases over a 3-year period and investigated the prevalence of immune and autoimmune disorders. All patients met diagnostic criteria for chronic HCV infection (ie, positive anti-HCV antibodies, presence of HCV-RNA, and inflammation on liver biopsy). The prevalence of arthritis was 4% in subjects who did not have cirrhosis and 2% in patients who had cirrhosis. The study had no control group but the authors indicated that the prevalence of rheumatoid arthritis (diagnosed by criteria of the ARA, 1987) in their study group (2.2%) was similar to the prevalence of rheumatoid arthritis (0.5%-3%) found in the general population. Barbosa and coworkers  looked at the issue from the opposite side. They prospectively analyzed a cohort of 367 patients who had rheumatic diseases, seen in a rheumatology clinic in Brazil, and assessed the seroprevalence of HCV infection. Systemic lupus erythematosus was seen in 47% of their patients, rheumatoid arthritis in 24.2%, and other diagnoses in 28.1%. The overall seroprevalence of HCV in the total cohort was 1.9%, which was similar to the background rate (2.2%, 95% CI: 1.6%-2.8%) illustrated in a previous study performed in blood donors in the same geographic area ; again, no concurrent control group was similarly evaluated. When analysis was restricted to patients who had a diagnosis of rheumatoid arthritis, the concurrent HCV infection increased slightly to 3 of 89 (3.4%, 95% CI: 0.7%-9.5%)—higher, but not enough for a statistically significant difference. Finally, Ramos-Casals and colleagues  in a multicenter study evaluated two groups of patients who had systemic autoimmune diseases, one with and the other without HCV infection. Compared with the HCV-negative group, those who had HCV were older and more likely to be male. They also had more frequent vasculitis, cryoglobulinemia, and neoplasia, but their frequency of articular symptoms (84 of 180 [47%] versus 97 of 180 [54%]) was not different.
Does treatment of HCV infection improve the associated arthritis? Antiviral treatment of HCV is relatively new and consequently the available information is scarce. Information comes mostly from anecdotal reports or small series of retrospective, nonrandomized interventions with limited follow-up and sometimes concomitant use of antirheumatic medications [21,22], all limitations that make interpretation of results difficult. Zucker-man and coworkers  described a favorable response to antiviral therapy. They reported that of 25 patients who had HCV arthritis unresponsive to various antiinflammatory medications, including nonsteroidal antiinflammatory drugs (NSAIDs) and corticosteroids, 17 (67%) had complete (44%) or partial response (32%) of arthritis symptoms following therapy with interferon (IFN)-a for a median duration of 12 months. Although clinical improvement was noted in most of the patients studied, the role of antiviral therapy is unclear because a virologic remission and normalization of transaminases was seen in only 13 of 28 (46%) patients. Nissen and co-workers  recently reported a less favorable response, however. The investigators retrospectively evaluated 21 patients who had known chronic HCV infection followed in a rheumatology department. Nineteen of the patients had arthritis and 13 of them had received IFN-a either as monotherapy or combined with ribavirin. Following therapy, 4 had no change in the rheu-matologic manifestation, 6 had deterioration, and only 2 had improvement.
Clinical data: specificity of the syndrome
Even if difficult to detect on epidemiologic studies, is there a type of arthritis specific to HCV infection? The rheumatologic literature seems to agree on two main types of presentation [13,14]. The first, less common (seen in one third of cases) is associated with cryoglobulinemia and described as asymmetrical pauciarticular typically involving the medium-sized and large joints . Cryoglobulinemia is not seen only with HCV infection and not all patients who have HCV have cryoglobulins, but there is a strong association between the two entities. Mixed cryoglobulinemia can be observed in up to half of HCV infected patients and—depending on the geographic location—40% to 100% of people who have cryoglobulinemia may have HCV [25,26]. Most patients who have mixed cryoglobulinemia are asymptomatic and only 10% may have symptoms, such as cutaneous manifestations (vasculitis), peripheral neuropathy, liver involvement, renal involvement, sicca syndrome, and B-cell lymphoproliferative disorders. Polyarthralgias and myalgias are frequent in cryoglobulinemia syndrome but arthritis is relatively rare .
The second, more common type of arthritis (two thirds of cases) closely resembles rheumatoid arthritis. It is described as polyarticular symmetrical primarily involving small joints, and usually is positive for rheumatoid factor; as such many of these cases fulfill the ACR criteria for rheumatoid arthritis . It differs from rheumatoid arthritis, however, in that it affects a different population (older, higher proportion male), it does not progress to erosion or destruction of the joint, there are no subcutaneous nodules, and only half of them show elevation of the erythrocyte sedimentation rate . Because both HCV-related arthritis and rheumatoid arthritis can be positive for rheumatoid factor, to try to distinguish them several studies have evaluated the use of other antibodies, such as anti-cyclic citrullinated peptide antibodies (anti-CCP) and antikeratin antibodies (AKA). Van Boe-kel and colleagues  in 2002 reviewed the pertinent literature and concluded that anti-CPP had the most value for the diagnosis of rheumatoid arthritis. This conclusion was supported by Bombardieri and coworkers  who specifically evaluated the value of anti-CCP antibodies to distinguish between a cohort of patients who had HCV arthritis and patients who had rheumatoid arthritis. Although rheumatoid factor was present in 37.5% of 8 and 90% of 30 patients who had HCV arthritis or rheumatoid arthritis, respectively, anti-CCP antibodies were not detected in any of the 8 patients who had HCV arthritis (although 3 of them had chronic polyarthritis indistinguishable from rheumatoid arthritis), whereas it was found in 23 of the 30 patients (77%) who fulfilled ACR criteria for rheumatoid arthritis. In addition, the investigators retrospectively analyzed the sera of 10 patients who had chronic HCV infection who presented with symmetric polyarticular involvement who later developed an erosive pattern consistent with rheumatoid arthritis, and found that 60% of them had anti-CCP antibodies; these patients were believed to have a concomitant diagnosis of chronic hepatitis C and rheumatoid arthritis. Similarly, Girelli and colleagues  reported that rheumatoid factor was seen in 32 (91%) of 35 patients who had rheumatoid arthritis and 12 (86%) of 14 patients who had HCV-related arthritis, whereas anti-CPP was seen in 12 (71%) and none, respectively. Cojocaru and coworkers  evaluated AKA and found that these antibodies were more prevalent in a cohort of patients diagnosed with probable rheumatoid arthritis (64% or 18 of 28) compared with another group of patients diagnosed with symmetric polyarthritis or polyar-thralgia associated with HCV infection (9% or 3 of 31); both groups of patients were rheumatoid factor positive. These data provide evidence supporting the concept that HCV arthritis is distinct from rheumatoid arthritis and that both entities can be differentiated based on a combination of serologic markers; further large-scale studies are warranted, however.
Biologic evidence: plausibility
What could be the mechanisms behind HCV arthritis and is there any evidence to support them? Olivieri and coworkers  proposed three potential mechanisms for the occurrence of HCV-associated arthritis: (1) direct synovial tissue invasion by HCV, (2) autoimmune synovial response to HCV, and, (3) immune complex deposition/mixed cryoglobulinemia. They suggest these factors may not be mutually exclusive but rather act synergis-tically to the overall effect of arthritic syndromes observed in HCV. Regarding viral tissue invasion, the limited data available show that HCV-RNA has been detected, but only in few of the evaluated patients, and only in synovial fluid but not in synovial membrane [33,34]. It is not clear whether this represents free virus, infected tissue, or infected lymphocytes in transit in the affected joint. Serum autoantibodies have been reported extensively in chronic HCV infection [13,35-38]. The HCV core antigen is known to carry domains that mimic self-antigens , a fact that could explain the high prevalence of autoantibodies in chronic HCV infection. It is unknown, however, whether these autoantibodies play any role in the pathogenesis of HCV arthritis. The syndrome with the strongest link to chronic HCV infection is mixed cryoglobulinemia [9,40]. Cryoglobulins are more likely to cause poly-arthralgias than frank arthritis , however, and as common as cryoglobulins are in HCV infection they cannot explain the pathogenesis of most cases because patients who have HCV arthritis usually do not have serum cryo-globulins. The literature to support cryoglobulin deposition in the synovium is scant and aspirated joint fluid from patients who have arthralgias or arthritis and cryoglobulinemia is typically inflammatory and sterile .
Extrahepatic symptoms during chronic HCV infection are common and varied. Arthritis can be seen either as part of autoimmune processes (eg, associated with cryoglobulinemia) or independently. Whether there is a type of arthritis particular to HCV is not clear from epidemiologic studies but, in our judgment, is best suggested by description of a syndrome similar to rheumatoid arthritis but less aggressive that can be differentiated by sero-logic markers. Whether the manifestation is specifically attributable to HCV infection or rather is the nonspecific result of a chronic inflammatory process is not clear at all. Better designed studies are needed. In particular there is a need for large, collaborative, prospective studies, with standardized definitions and appropriately matched controls, that look at the frequency and type of arthritic complaints among patients who have HCV infection.
Was this article helpful?