Article Text

FRI0441 Infection risk in ankylosing spondylitis: results from a longitudinal observational cohort
  1. D. Wallis1,
  2. A. Thavaneswaran1,
  3. N. Haroon1,
  4. R. Ayearst1,
  5. R. Inman1
  1. 1Toronto Western Hospital, Toronto, Canada


Background Long term data on infection risk in ankylosing spondylitis (AS) are sparse. Anti-TNF therapy is increasingly used in AS, with infection being the most important adverse event. In rheumatoid arthritis (RA), serious infections are a common adverse event of anti-TNF therapy with an incidence of 4-10 per 100 patient-years (pys)(1).

Objectives To investigate the frequency of infections in AS and to identify factors predisposing to infection.

Methods Data were extracted from a longitudinal observational database for patients meeting Assessment of SpondyloArthritis international Society (ASAS) criteria for axial SpA who had at least two annual standardized protocols completed. Infection rates were calculated and multivariate logistic regression was performed to investigate the association of independent variables with infection.

Results 449 patients (73% male) were included in the analysis. 241 infections (13 serious, defined as requiring IV antibiotics or hospitalization) occurred during 1698 pys of follow-up. Infection type was recorded in 151 cases and infection site in 182 cases. 81.5% were bacterial, 16.6% were viral and 2.0% fungal. The most common sites of infection were upper respiratory tract (14.5%), lung (13.7%) and skin (10.8%).

The overall incidence of infection was 14 per 100 pys and of serious infection 0.8 per 100 pys. The overall incidence of infection was higher in patients on biologic drugs than in patients not on a biologic drug (19 per 100 pys vs. 12 per 100 pys, p=0.0047). However, the incidence of serious infection was not increased for patients on biologic drugs compared to those not on a biologic drug (0.5 per 100pys vs. 1 per 100pys, p =0.271).

For patients who developed an infection while taking a biologic drug, data were analyzed with respect to the timing of infection (n=60). 5/60 (8.3%) developed the infection in the first 6 months of biologic use, 9/60 (15.0%) developed the infection 7-12 months after starting the biologic, and 46/60 (76.7%) developed the infection after the first 12 months of biologic use.

In the univariate analysis DMARD use (ever), biologic use (ever), increasing comorbidity score and history of hospitalization were associated with infection. Only DMARD use (OR 1.69 [1.07-2.68], p=0.03) and hospitalization (OR 2.14 [1.40-3.28], p=0.0005) remained significant in the multivariate analysis. When analyzing individual comorbidities, a history of mucous membrane ulcers was associated with infection. There was no association with age, disease duration, smoking, glucocorticoid use or disease activity or severity as reflected by BASDAI and BASFI respectively.

Conclusions The serious infection rate in AS is this observational cohort is low when compared to rates reported in RA. Biologic use is not a significant risk factor for serious infection. The lower rate of infection in patients with AS compared to that published for RA may reflect several factors: the younger age and lower frequency of comorbidities in AS, differing immunogenetic mechanisms in the respective diseases, or differences in treatment.


  1. Curtis JR, Patkar N, Xie A, Martin C, Allison JJ, Saag M, et al. Risk of serious bacterial infections among rheumatoid arthritis patients exposed to tumor necrosis factor alpha antagonists. Arthritis Rheum. 2007 Apr;56(4):1125-33.

Disclosure of Interest None Declared

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.