Physical function in ankylosing spondylitis is independently determined by both disease activity and radiographic damage of the spine

To study the relationship between disease activity, radiographic damage and physical function in patients with ankylosing spondylitis (AS) PATIENTS AND METHODS: Baseline and 2-year data of the Outcome in Ankylosing Spondylitis International Study (OASIS)(217 patients) were used. Physical function was expressed by the Bath Ankylosing Spondylitis Functional Index (BASFI) and the Dougados Functional Index (DFI); disease activity by the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) and by erythrocyte sedimentation rate and C-reactive protein; and structural damage by the modified Stoke Ankylosing Spondylitis Spine Score (mSASSS). Syndesmophyte- and non-syndesmophyte sum cores, and numbers of affected (bridged) vertebral units were derived from the mSASSS. Univariate correlations were calculated on baseline values using the Spearman rank correlation. Multivariate associations were investigated by generalised estimating equations (GEE) on baseline and 2-year data. mSASSS correlated moderately well with BASFI (Spearman's r = 0.45) and DFI (r = 0.38). BASDAI correlated well with BASFI (r = 0.66) and DFI (r = 0.59). Correlation coefficients for mSASSS versus BASFI and DFI decreased by increasing levels of BASDAI, being zero at the highest quintile of BASDAI. GEE showed that both BASDAI and mSASSS independently and significantly helped to explain either BASFI or DFI. Results were similar for syndesmophyte sum score, non-syndesmophyte sum score, number of affected VUs or number of VUs with bridging. The lumbar part of the mSASSS contributed similarly to the cervical part in explaining BASFI/DFI. Physical function impairment in AS is independently caused by patient-reported disease activity and the level of structural damage of the lumbar and cervical spine. Syndesmophytes and other radiographic abnormalities contribute to physical function impairment