The natural history of ankylosing spondylitis as defined by radiological progression

Radiological status is an important objective endpoint in the assessment of ankylosing spondylitis (AS). We investigated the disease development of AS using radiological change.

The existing radiographs (n = 2,284) of 571 AS patients attending the Royal National Hospital for Rheumatic Diseases were scored retrospectively using the Bath Ankylosing Spondylitis Radiology Index. (1) Progression of disease was initially examined cross sectionally. Univariate analysis was used to examine factors associated with joint involvement. (2) Progression of disease was then examined longitudinally for patients with films at time of symptom onset. (3) Rate of progression of radiological change was calculated using longitudinal data of 2 sets of radiographs taken 10 years apart (patient number = 54). The results from this were used to extrapolate backwards to age at first radiological change.

(1) Progression to cervical spine disease was a function of: disease duration, severity of hip and lumbar involvement, and a history of iritis (p < 0.001). Lumbar involvement was associated with disease duration, age now, and severity of cervical and hip involvement (p < 0.001). Hip involvement was a marker for cervical disease and associated with disease duration (p < 0.001). (2) Longitudinal analysis revealed marked variation among patients with a slow general rate of progression. (3) The progression of AS over any 10 year period is linear [first 10 years = 30% (SD 0.3) of potential change, 10-20 yrs = 40% (SD 0.3) change, 20-30 yrs = 35% (SD 0.4) change (p = 0.5)]. Backward extrapolation suggests that the approximate time of first radiological change is at the age of 8 years. CONCLUSION. (1) AS is a linearly progressive disease with about 35% change every 10 years. Spinal involvement is largely an expression of disease duration while the hips become involved in about 25% of individuals and may predict a more severe outcome for the cervical spine. (2) Backward extrapolation shows that the disease process may start as young as 8 years of age. However, the time interval between the disease trigger and radiological change remains unknown.