Viruses can act as adjuvants in the development of autoimmunity, non-specifically stimulating the innate immune response (mast cells, dendritic cells, Toll-like receptors and complement receptors).1 Many viruses have been involved in the pathogenesis of different inflammatory arthropathies.2–5 Although joint manifestations have been reported in relation to rubella, HTLV-1, parvovirus B192 and hepatitis B and C viruses,1 there are few publications dealing with the joint condition produced by Epstein–Barr virus (EBV). We present a series of patients with inflammatory joint changes secondary to that virus.

Between February 2011 and January 2012, 6 patients with joints affected by EBV were treated in our section. The characteristics are shown in Table 1. The clinical onset had taken place within the preceding 3 months, without prodromes or extra-articular manifestations. In the 3 patients with arthritis, the synovial fluid had inflammatory features, without microcrystals, and the cultures were negative. Liver and renal function, electrolytes, creatine phosphokinase (CPK), and urine and blood tests were normal. All the patients tested negative for rheumatoid factor (RF), anti-citrullinated protein antibodies, antinuclear antibodies, anti-extractable nuclear antigen antibodies, human leukocyte antigen B27 and 2-step Mantoux test. Serological tests for hepatitis B and C viruses, parvovirus B19, human immunodeficiency virus, herpes simplex virus types 1 and 2, cytomegalovirus, EBV and varicella zoster virus, showed no evidence of current infection. The radiographic study of thorax, hands, feet and knees was normal.

Epstein–Barr virus belongs to the herpes virus family. Nearly 98% of the population aged 40 years or over has been infected by this virus at some time in their lives.1,6,7 It causes infectious mononucleosis and is associated with B-cell, T-cell and Hodgkin's lymphomas and nasopharyngeal carcinomas. The publications that deal with the effects of the virus on joints focus only on arthralgia and monoarthritis of the knee in relation to infectious mononucleosis.8

In acute infection, plasma concentrations of EBV anti-viral capsid antigen (VCA) IgM increase rapidly, decreasing a few months later. Anti-VCA IgG antibodies persist for life, with stable titers.1 The results have been interpreted in an appropriate clinical context, as positivity for anti-VCA IgM can be found in infections by other viruses of the same family. It is necessary to rule out other autoimmune and/or infectious diseases, since it can become positive in cases of strong immune response. In our patients, we defined the case as secondary to EBV infection in the presence of inflammatory joint involvement with positivity for VCA IgM, after excluding other diseases and infection by a virus of the same family. There is certain controversy concerning the time necessary for a test for IgM to become negative.

The serological determination of viruses like EBV, which are capable of producing clinically relevant joint involvement, may be key in the diagnosis of undifferentiated arthritides (term used when a patient does not meet the criteria for any particular rheumatic disease),9 as it enables the identification of the etiology or factor that triggered the condition. In situ hybridization of RNA and DNA makes it possible to detect the presence of cytomegalovirus, parvovirus B19, EBV and other viruses of the herpes family in the synovial fluid of patients with different forms of arthritis.9 This supports a possible major role of these viruses in inflammatory arthropathies.3 Moreover, there are publications that argue in favor of the possible role of EBV as a triggering factor in rheumatoid arthritis10 or juvenile idiopathic arthritis.

We report the first series of patients with inflammatory joint manifestations associated with acute EBV infection, excluding other diagnostic alternatives. Our patients were women, several of childbearing age. Disease presentation appeared to be seasonal (winter and spring). The patients with symmetrical polyarthritis required treatment with disease-modifying antirheumatic drugs; thus, that form of presentation seems to indicate a poor prognosis, with a trend toward chronicity. Larger series will be necessary to properly characterize this joint involvement.

The authors declare they have no conflicts of interest.