Background/Purpose:

Patients experiencing ankylosing spondylitis (AS) symptoms ²16 years-of-age are classified as juvenile-onset AS (JoAS), whilst those ³17 years adult-onset AS (AoAS). We compared JoAS and AoAS cases with respect to proceeding to orthopedic surgery, date and type of surgery, and revisions.

Methods:

A cross-sectional observational cohort study was conducted of all AS patients attending a secondary care rheumatology hospital since 1970.

The following were recorded: sex, age at symptom onset, age at diagnosis, HLA-B27 genotype status, smoking history, peripheral arthritis, psoriasis, uveitis, enthesitis, inflammatory bowel disease, family history of spondyloarthropathy (SpA), and AS-related orthopedic surgery.

Univariate analyses used t-tests and Fisher’s exact tests for continuous and categorical variables, respectively. Multivariate analyses were performed using logistic regression with variable selection based on stepwise selection.

Results:

143 JoAS were compared with 411 AoAS patients. At assessment, JoAS were slightly younger than AoAS cases (mean difference 2.9 years), and had slightly longer disease duration since diagnosis (26.0 vs.19.3 years). No statistically significant differences were found between the two groups in terms of sex distribution, HLA-B27 positivity, psoriasis, enthesitis or uveitis.

JoAS cases (n=29/143) were more likely to have had AS-related surgery than AoAS cases (n=29/411) even after adjusting for longer time since diagnosis (p=0.017).

Direct comparisons were then made of JoAS and AoAS cases proceeding to surgery (Table 1). No significant difference was found in the disease durations of JoAS and AoAS surgical cases to first surgery (diagnosis to first surgery, p=0.458; or symptom onset to first surgery, p=0.724).

The likelihood of second surgery (of any type) in AoAS cases was significantly less than for JoAS cases (adjusted analysis: OR 0.277; 95% CI 0.078-0.980; p=0.047).

Considering the likelihood of having a second surgery in the AS surgical cohort, smoking (OR 6.559 for ever having smoked vs. never smoked; p=0.097) and HLA-B27 positivity (OR 1.002; p=0.054) were significant predictors.

Considering the likelihood of having a hip procedure (arthroplasty or resurfacing), family history of SpA (OR 0.999; p=0.05) and age of symptom onset (OR 0.997; p=0.09) were significant predictors.

Most AoAS cases had their hip arthroplasty by 20 years disease duration (n=22), thereafter the procedure was infrequent (n=2 of 36 potential native hips).

Conclusion:

This is the largest reported study of detailed AS-related orthopedic surgery data comparing JoAS with AoAS cases. JoAS cases were more likely to proceed to surgery, and have had a second surgery, even after adjusting for a longer disease duration. Smoking and HLA-B27 positivity were predictors of a second surgery; age of symptom onset and family history of SpA were predictors of hip surgery.

Table 1.