The objective of this work was to study the associations between ankylosing spondylitis (AS) and clinical vertebral and nonvertebral fractures. Data from a large population-based public health database in Spain, Sistema d'Informació per al Desenvolupament de l'Investigació en Atenció Primària (SIDIAP), were used in this parallel cohort study. All participants registered in SIDIAP on January 1, 2006, were screened to identify those with a diagnosis of AS. Five age-matched, gender-matched, and general practice surgery-matched controls were selected for each patient with AS. All participants were followed until December 31, 2011, transfer out date, or death date. Fractures during this time were classified as vertebral or nonvertebral. Adjustment was made for potential confounders (tobacco smoking, alcohol consumption, body mass index, and use of oral steroids). Of 4,920,353 eligible patients in SIDIAP, 6474 AS patients with matched controls (n = 32,346) were available. A higher proportion of patients with AS versus controls had clinical vertebral (0.86% versus 0.41%) and nonvertebral (3.4% versus 2.7%) fractures. Adjusted Cox regression models showed an increased risk of clinical vertebral (hazard ratio [HR] 1.93; 95% confidence interval [CI], 1.39 to 2.68; p < 0.001) and nonvertebral (HR 1.19; 95% CI, 1.02 to 1.39; p = 0.03) fractures among patients with AS. However, the observed increased risks were apparent only in those not on regular nonsteroidal anti-inflammatory drugs (NSAIDs). There were no interactions with inflammatory bowel disease, psoriasis, or previous back pain. Patients with AS are at increased risk of vertebral and nonvertebral clinical fractures, independently of various risk factors. Regular use of NSAIDs appears to eliminate the excess fracture risk related to AS, but the mechanisms involved are unknown.
Keywords: BONE; ELECTRONIC HEALTH RECORDS; EPIDEMIOLOGY; FRACTURE; SPONDYLITIS; SPONDYLOARTHROPATHY.
© 2014 American Society for Bone and Mineral Research.