Antiperinuclear factor in chronic juvenile arthritis.

Sir: Nesher et al recently described the first comprehensive series considering antiperinuclear factor in juvenile chronic arthritis (JCA). They found an overall positivity of 34%; in patients with the polyarticular type of the disease 16/28 patients were positive.' Our results are at variance with these data: in a group of 313 patients only three were positive, all of them children with polyarticular onset (table). Thus our results are more in line with a recently reported Czechoslovakian series (Bardfeld R, IVth Prague international pediatric rheumatology symposium, 1992). Our material consisted of 49 fresh and 264 frozen serum samples. One of the former and two of the latter were positive. As antiperinuclear factor is predominantly, if not exclusively, of the IgG class, sample preservation would seem unlikely to have influenced the results. A more plausible explanation for the discrepancy lies in the immunofluorescence system. A major confounding factor causing variability in antiperinuclear factor results is the variation in substrate sensitivity, between different donors, of the buccal cells that contain the antigen. This drawback has barred antiperinuclear factor from coming into general use despite the long history of the test.2 We used a recently described improved technique that includes detergent treatment of the cells, which is reported to minimise, albeit not to eliminate, donor differences.3 We tested the serum samples at a standard dilution of 1:5,4 which is also the titre we recorded for the WHO rheumatoid factor reference preparation that has been proposed by Feltkamp et al as the reference standard for antiperinuclear factor, too.5 Sixty per cent of serum samples from adult patients with rheumatoid arthritis tested in parallel by this technique were positive (manuscripts in preparation). In conclusion, antiperinuclear factor seems to be a specific but insensitive marker for the JCA subset with polyarticular onset that resembles adult rheumatoid arthritis. It contributes to the evidence for a basic difference between JCA in general and adult rheumatoid arthritis. The need for a common reference standard in future studies is obvious.