Clinical relevance of C-reactive protein in ankylosing spondylitis and evaluation of the NSAIDs/coxibs' treatment effect on C-reactive protein

Objective. To evaluate the clinical relevance of CRP in AS and investigate the treatment effect of NSAIDs/coxibs on CRP. Methods. Data of 851 patients with painful axial AS participating in two randomized controlled trials evaluating coxibs (celecoxib) and NSAIDs (naproxen/ketoprofen) were used. The clinical relevance of CRP in AS was studied: the number of patients with abnormal CRP at baseline was expressed as a percentage, and the correlation among CRP, patient (age and gender) and disease [Bath AS Disease Activity Index (BASDAI), Bath AS Functional Index (BASFI)] characteristics was investigated. The standardized response mean (SRM) over placebo was used to detect the treatment effect of NSAIDs/coxibs on CRP. Results. The 851 enrolled patients (73% of whom were men) had an active disease, as assessed through functional impairment. Mean (s.d.) baseline CRP was 14 (19) (median 7.2) mg/l; 61% of patients had an increased CRP. The CRP levels were correlated positively with male sex, BASDAI and BASFI and negatively with age (multivariate analysis). There was a significant difference in the changes of CRP between placebo and NSAIDs/coxibs group (P = 0.003; SRM over placebo: 0.25). Such a treatment effect was of higher magnitude in the subgroup of patients with increased CRP and usually of lower magnitude than the treatment effect of other outcome variables such as BASDAI and BASFI. Conclusions. Increased CRP was frequently observed in patients with painful axial AS and was correlated both with activity and functional severity of the disease. The treatment effect of NSAIDs/coxibs was relevant in the subgroup of patients with increased CRP at baseline.