Background Recently published treatment recommendations indicate that disease activity, measured using the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI)should be assessed at least twice over a period of at least 4 weeks and should be consistently ≥4 prior to access to anti-tumor necrosis factor (anti-TNF) agents for patients with AS refractory to NSAID therapy (ref). The required period is a minimum of 3 months in many countries thereby causing a mandatory delay in treatment. However, neither of these requirements/recommendations are evidence-based and it is unclear to what degree disease activity changes or how patients respond to further NSAID therapy once they are considered candidates for anti-TNF by the rheumatologist.
Objectives To assess disease activity in patients with AS who are considered NSAID refractory and candidates for anti-TNF during follow up and to determine the extent of variation in disease activity and factors which influence variability.
Methods We reviewed consecutive patients considered candidates for anti-TNF who were followed systematically in a longitudinal cohort after this treatment decision had been made. Local access to anti-TNF therapy requires that patients have a BASDAI ≥4 on two occasions at least 8 weeks apart despite at least two different NSAID agents. Demographic and ASAS core set outcome parameters recommended for longitudinal follow up are collected at baseline, 8-12 weeks, and every 6 months thereafter.
Results We assessed 141 patients (101 male) of mean (SD) age 45.0 (14.0) years and mean (SD) disease duration of 19.8 (13.2) years who had a mean (SD) interval between the two assessments of 4.6 (1.9) months and median (range) of 4 (2-10) months. As per the access criteria all patients had a BASDAI ≥4 at baseline. The mean (SD) BASDAI at baseline was 7.1 (1.4) and 7.2 (1.3) at the second assessment. The mean (SD) change in BASDAI between the two assessments was 0.6 (1.2). Only one patient had a BASDAI <4 at the second assessment, which was conducted at 9 months after the first. We assessed by correlation analysis age, sex, disease duration, interval of time between the two assessments, and baseline disease activity (BASDAI) as factors that might influence the BASDAI at the second assessment. Only the baseline BASDAI was significantly correlated (r = -0.52, p<0.0001). This negative correlation likely reflects regression to the mean so that a reduction is observed in patients with a very high BASDAI at baseline and an increase is observed in those with a BASDAI near 4.
Conclusions The majority of patients deemed candidates for anti-TNF due to failure of NSAID therapy demonstrate little change in BASDAI after further change in NSAID therapy even over prolonged time frames (4-10 months). Any change in BASDAI in patients refractory to NSAIDs is likely due to regression to the mean.
Van der Heijde et al. Ann Rheum Dis 2011;70: 905
Disclosure of Interest None Declared