Objective: Juvenile-onset ankylosing spondylitis (JoAS) and adult-onset ankylosing spondylitis (AoAS) are subtypes of ankylosing spondylitis (AS) that may have different clinical outcome. We compared cohorts of JoAS and AoAS in terms of: (1) clinical characteristics; (2) clinical outcomes; (3) proceeding to AS-related orthopedic surgery; (4) type of orthopedic surgery. Methods: A retrospective cohort study was conducted of all AS patients attending a teaching hospital. Demographics, clinical parameters, and history of AS-related orthopedic surgery to the spine, root or peripheral joints were recorded. Differences between surgery for JoAS and AoAS patients and effects of covariates were assessed using logistic regression and survival analyses. Results: 553 AS patients were studied; 162 JoAS and 391 AoAS cases. After adjusting for significant covariates, AoAS cases were less likely to have: proceeded to surgery (odds ratio, OR 0.31; p<0.001); had a hip procedure (resurfacing or arthroplasty; OR 0.374; p=0.001); had a hip arthroplasty (OR 0.43; p=0.01). Significant differences were also observed when comparing Kaplan-Meier survival curves (p=0.001) and using Cox proportional hazards regression (p=0.002). A history of smoking was not associated with surgery. AS cases with older age at symptom onset were far less likely to have surgery than those with younger onset, in a non-linear manner. Conclusion: JoAS are more likely than AoAS cases to proceed to hip arthroplasty, but equally likely to have hip resurfacing and hip arthroplasty revision / re-revisions. Smoking was not associated with the risk of orthopedic surgery. Orthopedic surgery was unlikely after 40 years of disease in both subsets. © 2014 American College of Rheumatology.