OBJECTIVE
Antiperinuclear factor (APF) is an autoantibody detected in >50% of patients with rheumatoid arthritis (RA); it shows a specificity of roughly 90%. We investigated the possible role of APF as a prognostic marker in RA.
METHODS
A series of 103 patients with RA who fulfilled the 1987 American College of Rheumatology criteria (88 women and 15 men; mean age 55.5 yrs, mean disease duration 9 yrs) were prospectively followed. Sixteen variables were assessed in each patient at inclusion and over a 3 year period. APF was determined by indirect immunofluorescence assay using human buccal mucosal cells as substrate. APF assays were done at entry and at the end of followup without knowledge of the clinical status of the patients. Mann-Whitney U, chi-squared tests, variance analysis, and kappa index were used for statistical analysis.
RESULTS
Eighty of 103 patients completed followup. APF was detected in 40 of 80. At inclusion, APF correlated with the visual analog scale (VAS) of pain (p = 0.02). However, patients who showed APF positivity at entry had a less favorable course than APF negative individuals, as shown by a worse VAS of well being (p = 0.01), Ritchie index (p = 0.01), number of painful joints (p = 0.03), grip strength (p = 0.01), C-reactive protein (p = 0.04), and Health Assessment Questionnaire score (p = 0.03) at the end of the study. In addition, APF positive patients showed a worse radiological course (p = 0.03).
CONCLUSION
Our results suggest APF is a possible marker of poor prognosis in RA.