Absence of anti-cyclic citrullinated peptide antibodies in erosive osteoarthritis: further serological evidence of the disease as a subset of osteoarthritis

Erosive osteoarthritis (EOA) is considered to be a rare subset of osteoarthritis (OA) characterised by destructive changes involving the proximal interphalangeal and distal interphalangeal joints.1–3 Laboratory findings are usually negative even though a slight increase of erythrocyte sedimentation rate may occur.2,3 Radiologically, central erosions and the “gull wing” deformity characterise the disorder.4 Synovial pathology shows changes consistent with both rheumatoid arthritis (RA) and OA.1,2 Histological examination of synovium from patients with EOA joints shows lining cell hyperplasia, lymphocytic infiltration, and pannus formation, features indistinguishable from those of RA.1 The relationship between EOA and classical OA is controversial, as some authors consider it to be a separate disease entity, some regard it as one end of the spectrum of OA, and some regard it as an interface between OA and RA.1,4 A relationship between EOA and RA was first suggested by Ehrlich,5 who noted the superimposition of clinical, laboratory, and imaging findings of RA in 62 of 170 patients initially diagnosed with EOA. Moreover, in the early stages of the disease, the differential diagnosis between EOA and other arthritis, such as RA or psoriatic arthritis, may pose a challenge, requiring a number of laboratory tests and investigations.3,6–8 It is well known that anti-cyclic citrullinated peptide antibodies (anti-CCP) are highly specific for RA and good predictors of radiographic joint damage9; for that reason this study aimed at detecting these autoantibodies in serum samples of patients with EOA, in order to ascertain their clinical usefulness and, possibly, to contribute to the clarification of the relationship of EOA with classical OA and RA. On the one hand, positive results may support the hypothesis that EOA is an interface between OA and RA, but, on the other hand, negative results may support the hypothesis that EOA is one end of the spectrum of OA. We evaluated 32 patients with EOA showing typical EOA radiographic findings.6 The control group included 35 patients affected by nodal OA of the hands (NOA) fulfilling American College of Rheumatology criteria10 and 50 healthy subjects. All the patients were examined for exclusion of psoriatic arthritis, RA, undifferentiated spondyloarthropathies, gout, and pseudogout. In addition, 45 patients with RA were examined to test the sensitivity and specificity of our anti-CCP enzyme linked immunosorbent assay (ELISA) kit (table 1)