oratory protocol, we studied all patients in search of systemic diseases and rheumatic symptoms. Patients were asked about the presence of widespread pain,.
British Journal of Rheumatology 1997;36:981±985
FIBROMYALGIA-ASSOCIATED HEPATITIS C VIRUS INFECTION J. RIVERA, A. DE DIEGO,* M. TRINCHET$ and A. GARCIÂA MONFORTE Rheumatology Unit, *Hepatology Unit and $General Surgery Department, Hospital General Universitario Gregorio MaranÄoÂn, c/Doctor Esquerdo 46, 28007 Madrid, Spain SUMMARY The objective was to determine whether there might be an association between hepatitis C virus (HCV) chronic infection and ®bromyalgia (FM). We determined the prevalence of HCV infection in 112 FM patients, in comparison with matched rheumatoid arthritis (RA) patients from the out-patient clinic of a teaching tertiary care general hospital. Furthermore, we looked for evidence of FM in 58 patients diagnosed with chronic hepatitis due to HCV, compared with matched surgery clinic patients. HCV antibodies were determined by enzyme-linked immunosorbent assay (ELISA) and recombinant immunoblot assay (RIBA). Serum RNA of HCV (HCV-RNA) was determined by polymerase chain reaction. In the group of FM patients, HCV antibodies were found by ELISA in 17 (15.2%) patients and in six (5.3%) of the RA controls (P < 0.05). RIBA was positive in 16 and indeterminate in one of the FM patients. Serum HCV-RNA was found in 13 of these FM patients. In eight (47%) FM patients, alanine aminotransferase (ALT) was normal, although HCV-RNA was detected in four (50%) of them. In the group of patients with chronic hepatitis due to HCV, all patients had HCV antibodies and the presence of HCV-RNA in serum. Within these patients, 31 (53%) had diuse musculoskeletal pain, while six (10%) ful®lled FM diagnostic criteria. In the control group, 13/58 (22%) had diuse musculoskeletal pain (P < 0.001), whereas only one female patient (1.7%) ful®lled FM criteria (P < 0.05). Serum ALT was 51.7 238.4 in FM patients, whereas it was 122 2 76.3 in patients with HCV chronic hepatitis but without FM (P < 0.001). There were no statistical dierences in autoimmune markers between patients with and without FM. These data suggest that there exists an association between FM and active HCV infection in some of our patients. FM is not associated with liver damage or autoimmune markers in these patients. HCV infection should be considered in FM patients even though ALT elevations were absent. KEY WORDS: Hepatitis C virus, Fibromyalgia, Infection, Rheumatoid arthritis.
infect lymphocytes and other blood mononuclear cells with active replication in these cells [16, 17]. Several autoimmune diseases and many immunological alterations have been associated with HCV chronic infection [18]. Mixed cryoglobulinaemia is the most common immunological alteration associated with HCV infection and diuse musculoskeletal pain is not so rare in this autoimmune disorder [19]. The objective of this study is to determine whether there might be an association between chronic infection by HCV and FM. For this purpose, we determined the prevalence of HCV infection in patients with FM, in comparison with matched RA patients. Furthermore, we looked for evidence of FM in patients diagnosed with chronic hepatitis due to HCV, compared with matched surgery clinic patients.
THE clinical syndrome of ®bromyalgia (FM) is characterized by diuse musculoskeletal pain, fatigue, morning stiness and sleep disturbance. The pathophysiological mechanisms of this disorder are still unknown, but it has been suggested that its origin can be multifactorial [1]: systemic diseases like rheumatoid arthritis (RA) [2] and systemic lupus erythematosus [3], physical trauma like motor vehicle accidents [4] and psychological disorders [5] have coexistent FM symptoms. It has been postulated that chronic viral infections may trigger FM [6]. In recent years, FM has been associated with some chronic infections like Lyme disease [7±9], human immunode®ciency virus (HIV) infection [10, 11] and parvovirus infection [12], although the pathogenetic mechanisms by which such a virus can induce FM have not been extensively studied. The recently described hepatitis C virus (HCV) [13] has shown a great tendency to chronicity and infection may persist in >80% of the patients with active viral replication during long periods of time [14]. This long-lasting chronic infection leads to liver cirrhosis in 20% of cases and to hepatocarcinoma in some of these patients [15]. HCV not only infects hepatocytes, but it may also
MATERIALS AND METHODS Subjects Patients with the clinical syndrome of FM were newly and consecutively collected from the outpatient rheumatology clinic of a teaching tertiary care hospital. Only those patients who met the criteria of the American College of Rheumatology (ACR) for the diagnosis of FM were included in the study [20]. Patients with systemic diseases, in¯ammatory rheumatic diseases, HIV infection and hepatitis B virus (HBV) infection were excluded. A disease control group formed by patients with RA (sex and age matched) also attending the rheum-
Submitted 17 October 1996; revised version accepted 24 February 1997. Correspondence to: J. Rivera, Abrego 19, 18D, Pozuelo de AlarcoÂn, 28224 Madrid, Spain.
# 1997 British Society for Rheumatology 981
982
BRITISH JOURNAL OF RHEUMATOLOGY VOL. 36 NO. 9
atology out-patient clinic was used to compare the frequency of positive HCV serology. All RA patients satis®ed ACR diagnostic criteria [21] and were selected from a population of 252 RA patients in whom HCV serology had been determined (data not published). Matching was done ®rst by sex and second by age (2 2 yr). Among the patients with HCV chronic hepatitis who were studied at the hepatology unit of the hospital for initial treatment, a group of consecutive patients was selected for the study. All these patients had HCV antibodies and active infection by HCV, as shown by the presence of viral RNA in serum. Again, patients with HIV and HBV infection were excluded. Data collection was carried out before interferon-alpha (IFN-a) therapy was started. To compare the prevalence of FM in this population, a group of patients, matched for age and sex (as described above), who were attending the general surgery unit of the hospital, was used as a control. All these patients were clinically stable, not febrile nor acutely ill. Most of them were on a waiting list for programmed abdominal surgery (gallbladder stones and abdominal hernias). Socioeconomic status was similar in all patients and controls. Methods Following a previously established clinical and laboratory protocol, we studied all patients in search of systemic diseases and rheumatic symptoms. Patients were asked about the presence of widespread pain, sleep disturbances, morning stiness, fatigue and arthritis. Clinical history of systemic diseases, hepatitis and previous history of elevation of liver enzymes were also evaluated. A complete physical examination was carried out in each patient including Schirmer's tear test. The 18 tender points of the ACR classi®cation study of FM [20] were examined by digital pressure by the same investigator (JR) in all patients and controls. All data were collected prospectively for patients and controls. Laboratory determinations were routinely performed in the clinical laboratories of the hospital by the same laboratory personnel in a blind manner. These included complete blood count, erythrocyte sedimentation rate, serum creatinine, aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline phosphatase, glucose, total serum proteins, albumin, gamma globulins, urinalysis, cryoglobulins, rheumatoid factor, antinuclear antibodies (ANA), and serum C3 and C4 levels. A commercial third-generation enzyme-linked immunosorbent assay (ELISA) for the detection of HCV antibodies (Ortho HCV3.0, Raritan, NJ, USA) was used and con®rmed by a commercial second-generation recombinant immunoblot assay (RIBA) (Abbott, Chicago, IL, USA). The RNA of HCV (HCV-RNA) was detected in serum by ampli®cation with the polymerase chain reaction of the 5'-untranslated region of the HCV genome using primers that de®ne a sequence
within this highly conserved region (Amplicor, Roche Diagnostic Systems, Branchburg, NJ, USA). HBV surface antigen was determined by ELISA (Abbott) and the presence of HIV antibodies was determined by ELISA (Abbott) and con®rmed by Western blot (Sano® Diagnostics Pasteur, Chaska, MN, USA). Statistical analysis Student's two-tailed t-test was performed for comparing independent numerical variables, and w2 test and Fisher's exact test for comparing proportions. RESULTS A total of 112 patients with the diagnosis of FM (mean age 51.43 2 11.08 yr; range 25±78 yr; sex F: 109/M: 3) were included in the study and compared with 112 age- and sex-matched RA patients (mean age 51.28 211.20 yr; range 24±76 yr; sex F: 109/M: 3). In 17 FM patients (15.2%) and in six RA patients (5.3%), anti-HCV antibodies were detected by ELISA (P < 0.05) (95% CI 1.9, 17.6). The presence of HCV antibodies was con®rmed by RIBA in 16/17 (94%) FM patients and was indeterminate in one. HCV-RNA was detected in 13 of these patients and absent in four of them. RIBA was positive in 4/6, indeterminate in one and negative in one of the RA patients. HCV-RNA was detected in two RA patients. HCV-RNA was signi®cantly more frequent among patients with FM than RA (P < 0.01) (95% CI 2.5, 17.13), but the rate of positive RIBAs and positive tests for HCV-RNA among patients with positive HCV ELISA tests did not dier between FM and RA patients. Mean ALT levels in the 17 FM patients who were HCV positive were increased up to 51.76 238.36 U/l (normal value 0±40 U/l). Only in three (18%) patients were ALT levels more than twice the normal range, in six (35%) patients between one and two times, and in eight (47%) patients ALT levels were normal. HCV-RNA was present in four (50%) of these eight HCV-positive FM patients with normal ALT levels. In this group of 17 HCV-positive FM patients, the presence of chronic hepatitis or active infection by HCV had not been diagnosed prior to the study in nine (53%) of the patients because they showed a small increase of liver enzymes only occasionally. In the group of patients with HCV chronic hepatitis who were followed at the hepatology unit of the hospital, 58 patients (mean age 46.22 2 12.88 yr; range 22±70 yr) were included. In this group, 31 (53%) of the patients complained of diuse arthralgias, six (10%) of them ful®lled FM diagnostic criteria and in 14 (24%) there were seven or more tender points with diuse arthralgias. Sex distribution showed a preponderance of males (M: 32/F: 26). Five of 26 (19%) females and 1/32 (3%) males ful®lled FM diagnostic criteria. Objective arthritis was present in one HCV chronic hepatitis patient. It was a chronic, erosive and sym-
983
RIVERA ET AL.: FIBROMYALGIA-ASSOCIATED HCV
TABLE I Comparison of the characteristics and analytical alterations between the groups of ®bromyalgia patients with positive HCV serology (FMHCV+), the group of patients with chronic hepatitis with less than seven tender points and the group of chronic hepatitis patients who had seven or more tender points Chronic hepatitis patients
Number Sex (F/M) Age ALT (U/l) Anaemia Leucopenia Thrombopenia Low complement$ ANA Cryoglobulins Rheumatoid factor
FM-HCV+
P
Less than 7 tender points
17 17/0 61.2929.25 51.76238.36 6% 0% 6% 50% 29% 9% 13%
*** *** *** ns * ** ns ns ns ns
44 15/29 43.662 13.04 122.002 76.35 0% 7% 27% 32% 24% 21% 16%
P
More than 7 tender points
P(@)
*** *** ns ns * ns ns ns ns ns
14 11/1 55.7826.38 133.422142.33 14% 28% 35% 33% 28% 21% 15%
ns * *** ns * ** ns ns ns ns
P(@) compares the FM-HCV+ group with the group of chronic hepatitis patients with >7 tender points. $Low complement was de®ned by serum levels of C3 < 83 mg/dl and/or C4 < 20 mg/dl. *P < 0.1. **P < 0.05. ***P < 0.001.
metrical polyarticular arthritis aecting small and large joints, consistent with the diagnosis of RA. Among the control group of sex- and age-matched patients attending the general surgery unit of the hospital, only 1/58 (1.72%), a female patient, ful®lled FM diagnostic criteria (P < 0.05) (95% CI ±1.6, 18.8), while 13/58 (22%) had diuse arthralgias (P < 0.001) (95% CI 12.5, 49.4). The patients in the group of HCV chronic hepatitis were divided into two groups. One group was composed of 44 patients who had less than seven tender points of FM, and the other was composed of 14 patients who had seven or more tender points (six of them met the FM criteria). We studied their laboratory alterations and compared these two populations with the group of 17 HCV-positive FM patients, as shown in Table I. Chronic hepatopathy, determined by ALT elevation, and the presence of portal hypertension with hypersplenism, determined by the presence of thrombopenia, were signi®cantly more severe in both groups of patients with HCV chronic hepatitis, suggesting that the patients with FM and HCV infection have mild liver disease. Unfortunately, liver biopsy was not performed in patients with FM and HCV infection. However, immunological alterations were not signi®cantly dierent between groups (Table I). Age and sex were similar in patients with FM and those patients with chronic hepatitis who had more than seven tender points, and signi®cantly dierent (P < 0.001) in the group of patients with less than seven tender points. No statistically signi®cant dierences were found in past history of major surgery and blood transfusions or other major organ involvement between patients with HCV-positive FM and patients with HCV chronic hepatitis with less than seven tender points. Sicca syndrome was present in 18% of FM patients
and in 21% of HCV chronic hepatitis patients with less than seven tender points. This dierence was not statistically signi®cant. DISCUSSION In this study, we have found that up to 15% of our patients with FM bear HCV antibodies, a signi®cant number in comparison with the 5.3% who were found to be HCV positive in the RA patient control group. This is a considerably high proportion if we take into account that the prevalence of HCV antibodies in blood donors in our demographic area is 0.75% [22], similar to that found in other countries, i.e. between 0.5 and 1.5% [23]. Another striking feature in this study is that 16/17 (94%) HCV-positive FM patients were RIBA positive, which points to a true contact with HCV. The presence of a positive RIBA may suggest an actual or a past infection by the virus, but the high speci®city of RIBA used in this work rules out the possibility of false-positive results, as can be seen with ELISA in other diseases [24]. By contrast, in blood donors it has been shown that only 33% of those who are HCV ELISA positive are also RIBA positive, with a high proportion of RIBA negative and indeterminate, showing that there exist false-positive results by ELISA in healthy populations [2]. The high prevalence of HCV antibodies in the control group of RA patients is also striking. Two possible explanations exist. First, some authors have previously found that there exists an increase in falsepositive results with ELISA in RA patients, probably due to the presence of hypergammaglobulinaemia or rheumatoid factors in serum [25]. This could be the case in our RA group because only 4/6 patients were positive for RIBA, while two were negative or indeterminate. A second possible explanation is that
984
BRITISH JOURNAL OF RHEUMATOLOGY VOL. 36 NO. 9
HCV infection might have a pathogenetic role in RA. It is known that in RA there is a higher incidence of antibodies to some viruses like Epstein±Barr [26] or parvovirus [27]. However, the role of these viruses in the pathogenesis of RA is not completely understood and we do not know whether the presence of these antibodies may be the consequence of an increased susceptibility of RA patients to viral infections or whether they play an active role in the aetiology of RA. This could be the case for HCV infection in RA patients. In any case, in our work, six patients are not sucient to draw conclusions and further studies about the real prevalence of HCV infection and its role in the pathogenesis of RA are needed. (These are currently in progress.) The selection of a control group of RA patients instead of a blood donor group was found to be more appropiate since the former constitutes a disease control group, while the latter individuals have been self-selected and have been screened by history at the time of blood donation. In the HCV-positive FM population, active infection by HCV has been proved by the presence of HCV-RNA in 13 of the 17 (76%) patients. In the rest, the absence of serum HCV-RNA does not rule out the presence of active infection by the virus, because HCV replication in cells is not continuous and HCV-RNA appears intermittently in serum [15]. In this sense, some authors have shown that the absence of HCV-RNA in serum is sometimes accompanied by its presence in peripheral blood mononuclear cells [17]. Chronic hepatitis was remarkably not very severe in FM patients based on the liver function test results. Normal values of ALT were found in 050% of the patients, although HCV-RNA was detected in 50% of these cases. Unfortunately, liver biopsy was not performed in patients with FM and HCV infection, and we cannot conclude whether chronic liver disease was milder in patients with FM. This means that FM can exist without the presence of alterations in liver enzymes. Active HCV infection of lymphocytes and monocytes may co-exist with minimal or complete absence of liver eects [15, 22]. Among patients with HCV chronic hepatitis, who were attending at the hepatology unit of the hospital, diuse musculoskeletal pain and FM symptoms were higher than expected. A signi®cant number of 10% of these patients ful®lled the FM diagnostic criteria compared with 1.72% of the control group. The prevalence of FM in the overall population is around 2% with a clear female tendency [28]. A female preponderance was also found in our work. It has been described that HCV infection may trigger autoimmune disorders [18, 19]. In this study, we have found the presence of cryoglobulins, rheumatoid factor, low levels of serum complement and ANA in a high proportion of patients, but there were no statistical dierences between patients with and without FM. Other studies did not ®nd immunological alterations in patients with FM [29]. Recently, Middleton
et al. [3] have shown a high prevalence of FM symptoms in patients with systemic lupus erythematosus, but the authors fail to show any association between FM and levels of immunological activity in these patients. Two pathogenetic mechanisms have been proposed for explaining how an infection can trigger FM [6]. First, direct infection of host tissues or in¯ammatory mediators released during infection could induce FM. At present, some in¯ammatory products or cytokines, like IFN-a [30], have been related to FM, although no systematic search has been carried out [31]. Secondly, another theory proposes that the stress and anxiety produced by the knowledge of an infectious disease could trigger FM, as occurs in Lyme disease [8]. In our work, however, 53% of patients with FM did not know that they suered from HCV infection or chronic hepatitis, although half of them showed evidence of active viral replication. These data suggest that the presence of HCV in the organism could trigger FM. The best way to prove the relationship between HCV infection and FM is by demonstrating the correlation between improvement of FM symptoms and HCV clearance from serum after treatment. However, at present, the best tried and tested treatment for HCV infection has been recombinant IFN-a and this cytokine only produces a sustained clearance of HCV in a minority of treated patients [32]. On the other hand, IFN-a therapy can trigger FM symptoms in some patients [30]. Here we found that 15% of our patients with FM may have an HCV infection. In the remaining 85%, other known and unknown viruses that can produce chronic infection in the host, with or without clinical manifestations, could be responsible for FM symptoms. We think that the presence of such viruses should be sought in patients with FM symptoms. In summary, these data suggest that an association exists between active infection by HCV and FM in some patients. With respect to the physiopathological mechanisms involved, it seems that there is no relationship with the severity of the associated chronic hepatopathy or with the immunological alterations produced by HCV infection. Age and female sex are important factors in the appearance of FM. Also, the presence of active infection by HCV is more likely to trigger FM than the stress and anxiety produced by the disease. REFERENCES 1. Hench PK. Evaluation and dierential diagnosis of ®bromyalgia. Rheum Dis Clin North Am 1989;15: 19±29. 2. Wolfe F, Cathey MA, Kleinheksel SM. Fibrositis (®bromyalgia) in rheumatoid arthritis. J Rheumatol 1984;11:814±8. 3. Middleton GD, McFarlin JE, Lipsky PE. The prevalence and clinical impact of ®bromyalgia in systemic lupus erythematosus. Arthritis Rheum 1994;37:1181±8. 4. Green®eld S, Fitzcharles MA, Esdaile JM. Reactive
RIVERA ET AL.: FIBROMYALGIA-ASSOCIATED HCV
5.
6. 7. 8. 9. 10. 11.
12. 13.
14.
15.
16.
17.
18.
®bromyalgia syndrome. Arthritis Rheum 1992;35:678± 81. Hudson JI, Goldenberg DL, Pope HG, Keck PE, Schlesinger L. Comorbidity of ®bromyalgia with medical and psychiatric disorders. Am J Med 1992;92: 363±7. Goldenberg DL. Do infections trigger ®bromyalgia? Arthritis Rheum 1993;36:1489±92. Dinerman H, Steere AC. Lyme disease associated with ®bromyalgia. Ann Intern Med 1992;117:281±5. Hsu VM, Patella SJ, Sigal LH. `Chronic Lyme disease' as the incorrect diagnosis in patients with ®bromyalgia. Arthritis Rheum 1993;36:1493±500. Steere AC, Taylor E, McHugh GL, Logigian EL. The overdiagnosis of Lyme disease. J Am Med Assoc 1993;269:1812±6. Buskila D, Gladman DD, Langevitz P, Urowitz S, Smythe HA. Fibromyalgia in human immunode®ciency virus infection. J Rheumatol 1990;17:1202±6. Simms RW, Zerbini CAF, Ferrante N, Anthony J, Felson DT, Craven DE. Fibromyalgia syndrome in patients infected with human immunode®ciency virus. Am J Med 1992;92:368±74. Leventhal LJ, Naides SJ, Freundlich B. Fibromyalgia and parvovirus infection. Arthritis Rheum 1991;34: 1319±24. Choo QL, Kuo G, Weiner AJ, Overby LR, Bradley DW, Houghton M. Isolation of a cDNA clone derived from blood-borne non-A, non-B viral hepatitis genome. Science 1989;244:359±62. Esteban JI, LoÂpez Talavera JC, Genesca J, Madoz P, Viladomiu L, MunÄiz E et al. High rate of infectivity and liver disease in blood donors with antibodies to hepatitis C virus. Ann Intern Med 1991;115:443±9. Alter MJ, Margolis HS, Krawczynski K, Judson FN, Mares A, Alexander WJ et al. The natural history of community acquired hepatitis C in the United States. N Engl J Med 1992:327:1899±905. Zignego AL, Macchia D, Monti M, Thiers V, Mazzetti M, Foschi M et al. Infection of peripheral mononuclear blood cells by hepatitis C virus. J Hepatol 1992;15: 382±6. Navas S, Castillo I, Bartolome J, Marriott E, Herrero M, CarrenÄo V. Positive and negative hepatitis C virus RNA strands in serum, liver and peripheral mononuclear cells in anti-HCV patients: relation with the liver lesion. J Hepatol 1994;21:182±6. Pawlotsky JM, Ben Yahia M, Andre C, Voisin MC, Intrator L, Deforges L et al. Immunological disorders
19. 20.
21.
22
23. 24. 25. 26. 27.
28. 29.
30. 31. 32.
985
in C virus chronic active hepatitis: A prospective case control study. Hepatology 1994;19:841±8. Agnello V, Chung RT, Kaplan LM. A role for hepatitis C virus infection in type II cryoglobulinemia. N Engl J Med 1992;327:1490±5. Wolfe F, Smythe HA, Yunus MB, Bennett RM, Bombardier C, Goldenberg DL et al. The American College of Rheumatology 1990 criteria for the classi®cation of ®bromyalgia. Arthritis Rheum 1990;33: 160±72. Arnett FC, Edworthy SM, Bloch DA et al. The American Rheumatism Association 1987 revised criteria for the classi®cation of rheumatoid arthritis. Arthritis Rheum 1988;31:315±24. Prieto M, Olaso V, Verdu C, CoÂrdoba J, Gisbert C, RayoÂn M et al. Does the healthy hepatitis C virus carrier state really exist? An analysis using polymerase chain reaction. Hepatology 1995;22:413±7. van der Poel CL, Cuypers HT, Reesink HW. Hepatitis C virus six years on. Lancet 1994;344:1475±9. Alter HJ. New kit on the block: Evaluation of second generation assays for detection of antibody to the hepatitis C virus. Hepatology 1992;15:350±3. Borque L, Elena A, Maside C, Rus A, Del Cura J. Rheumatoid arthritis and hepatitis C virus antibodies. Clin Exp Rheumatol 1991;9:617±9. Depper JM, Zvai¯er NJ. Epstein-Barr virus. Its relationship to the pathogenesis of rheumatoid arthritis. Arthritis Rheum 1981;24:755±60. Cohen BJ, Buckley MM, Clewley JP, Jones VE, Puttick A, Jacoby RK. Human parvovirus infection in early rheumatoid and in¯ammatory arthritis. Ann Rheum Dis 1986;45:832±8. Wolfe F, Ross K, Anderson J, Russel IJ, Hebert L. The prevalence and characteristics of ®bromyalgia in the general population. Arthritis Rheum 1995;38:19±28. Yunus MB, Hussey FX, Aldag JC. Antinuclear antibodies and connective tissue disease features in ®bromyalgia syndrome: A controlled study. J Rheumatol 1993;20:1557±60. Middleton GD, McFarlin JE, Lee W, Lipsky PE. Eect of alpha interferon on pain thresholds and ®bromyalgia. Arthritis Rheum 1994;37:S214 (Abstract). Wallace DJ, Bowman RL, Wormsley SB, Peter JB. Cytokines and immune regulation in patients with ®brositis. [Letter] Arthritis Rheum 1989;32:1334±5. Tine F, Magrin S, Craxi A, Pagliari L. Interferon for non-A, non-B chronic hepatitis: a meta-analysis of randomized clinical trials. J Hepatol 1991;13:192±9.
