The presence of cartilage matrix glycoprotein in serum as determined by immunolocation analysis is not a sensitive indicator of "early" osteoarthritis of the knee.

Widespread effort is being devoted to the search for a serologic "marker" that could aid in the early diagnosis of osteoarthritis and in following the progression of the disease in response to treatment. It is obvious that such a marker would have its greatest utility in patients with mild or early osteoarthritis. CMGP is a disulfide-bonded 550,000 dalton cartilage matrix glycoprotein with a half-life of only 48 to 72 hours that has been found, through immunolocation analysis, in the serum of dogs with experimentally induced osteoarthritis and in the synovial fluid of patients with osteoarthritis but not other types of arthritis. To determine whether detection of CMGP in serum might be of value in identifying patients with "early" osteoarthritis, we examined serum samples from 26 patients with knee pain who had articular cartilage lesions of osteoarthritis at arthroscopy but whose knee radiographs were normal or showed only mild or moderate osteoarthritis. CMGP was identified by immunolocation analysis with specific antibodies. Eleven patients (42%) were seropositive for CMGP. In two, the degenerative cartilage lesions visualized at arthroscopy were mild (grade 2); in the other nine they were more severe (grade 3 or 4). However, 10 of the 15 seronegative patients also had grade 3 or 4 cartilage degeneration. Thus, this serum assay for CMGP was often negative in this group of patients in the presence of well-defined cartilage degeneration.

Synovial inflammation in patients with early osteoarthritis of the knee.

While synovitis is common in advanced osteoarthritis (OA), its prevalence and severity in patients with early or mild OA are uncertain. In our study synovial biopsies from patients with arthroscopic evidence of OA whose radiographs were normal, or showed only mild/moderate changes of OA, were examined to determine the prevalence and severity of lining cell proliferation and mononuclear cell infiltration. Synovitis was present in only 16 of 29 patients (55%) who underwent arthroscopy because of chronic knee pain and were found to have OA; no synovium from 50% of the 22 patients in this group with full thickness cartilage ulceration showed infiltration with mononuclear cells. Similarly, no evidence of synovitis was seen in biopsies from 7 of 14 additional patients with OA who did not have knee pain but who underwent arthroscopy to evaluate joint instability. An association was seen between synovial mononuclear cell infiltration and thickness of the synovial lining cell layer (p less than 0.03), but lining cell hyperplasia was found in samples from only 12% of the patients with OA in our series. The severity of OA cartilage lesions was unrelated to severity of synovitis and no topographic relationship was found between cartilage ulceration and synovitis.