Background/Purpose: Acute anterior uveitis (AAU) is the most common
extra-articular manifestation in patients with ankylosing spondylitis (AS),
developing in 20-30% of these patients during the course of their disease.

Our
aim was to study the characteristics and risk factors associated with AAU in
our cohort of AS patients.

Methods: From our longitudinal observational cohort, 716
patients with AS (meeting modified New York Criteria) and at least 2 years
follow up were included and who were followed from January 2006 to November
2013. Patients with (AAU+) and without (AAU-) uveitis were compared. Uveitis
flare rates were calculated per year. T-test, Chi-squared tests and logistic
regression were used where appropriate.

Results: Of the 716 patients, 225 (31.4%) had a reported  diagnosis
of AAU at their baseline clinic visit. Patients with AAU+ were older compared
to the AAU- group (mean age of 42.6 vs 37.9 years; p< 0.001). AAU
started after the onset of back pain in the majority of patients with only
10.5% of patients reporting onset of AAU before onset of AS related back pain.
Patients with AAU had higher HLA-B27 (B27) prevalence (91.8% vs 82.1%,
p<0.05). In the multivariate analysis (MVA),  AAU was independently
associated with age,  B27, psoriasis, IBD and elevated CRP.  B27
(OR=2.66 [95% CI=1.44-4.9]), psoriasis (OR=2.36 [95% CI=1.41-3.97]) and IBD
(OR=2.25 [95% CI=1.27-4]) are the strongest independent predictors of AAU.
 Within the AS/IBD group (n=86), 43% of these patients had a history of
AAU, of which 81% were B27 positive.

In patients with AAU
, there was a trend towards more peripheral arthritis and enthesitis. The BASMI
score was higher in patients with AAU (3.3 vs 2.7, P<0.05), however there
was no association found on MVA. There was no significant difference found between
the two groups in terms of BASDAI score, hypertension, diabetes, previous
history of myocardial infarctions and smoking history.

There
was no difference found between NSAID use at baseline (66.1% vs. 66.5%,
p=0.90). Patients with AAU were more frequently treated with DMARDs (26% vs.
16.5%, p<0.01). Sulfasalazine was used more frequently in the AAU+ group
(14.2% vs. 7.9%, p<0.01). Use of biologics was similar at baseline (22.2%
vs. 18.4%, p=0.23). 

Conclusion: In our cohort of AS patients, an increased frequency
of HLA-B27 was seen in AAU+ AS. AAU+ AS is associated with psoriasis and IBD.
The psoriasis and IBD association is independent of HLA-B27, suggesting an
interaction of other genetic as well as environmental factors. At baseline
DMARD use was associated with AAU, likely reflecting the association
with peripheral joint disease.