Arthritis & Rheumatism, Volume 63,
November 2011 Abstract Supplement

Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Chicago, Illinois November 4-9, 2011.

Profiling NSAID Responders in Ankylosing Spondylitis.

Alnaqbi1,  Khalid A., Haroon2,  Nigil, Shen3,  Hua, Cook3,  Richard J., Carty2,  Adele, Inman4,  R. D.

Toronto Western Hospital-EW 1–420, Toronto, ON
Toronto Western Hospital, Toronto, ON
University of Waterloo, Waterloo, ON
The Arthritis Program, Toronto Western Hospital and Division of Rheumatology, Toronto Western Hospital and University of Toronto, Toronto, ON


There is a substantial subset of ankylosing spondylitis (AS) patients in whom symptoms are well controlled with NSAID therapy alone. This is an important domain, since failure to achieve symptomatic control with NSAIDS is the operational prerequisite for biolologic therapy in AS. The current study was undertaken to characterize that subset of AS patients who are responsive to NSAID therapy.


The longitudinal prospective database of an AS cohort (patients was analyzed on the basis of response to NSAIDs. NSAID-responders (NS-R) were defined as having a BASDAI<4 on the last clinic visit while be treated only with NSAIDs. NSAID-nonresponders (NS-NR) were defined as having a BASDAI>4 on the last clinic visit while be treated only with NSAIDs OR as having required biologic treatment. AS patients met the modified New York criteria for AS and demographic, clinical and radiographic variables including BASDAI, BASFI, BASG, metrology, CRP, ESR, extra- articular manifestations, and mSASSS scores were compared to see if NS-R and NS-NR could be differentiated.


In the cohort of 524 AS patients there were 136 NS-R and 388 NS-NR.

Clinical assessment (Table) indicated that NS-R patients had milder disease as reflected in: (i) better spinal mobility (ii) lower BASFI scores (iii) absence of peripheral joint involvement. However, this was not reflected in differences in mSASSS scores.

Comparative Features of NSAID-responders vs NSAID-nonresponders

Variable (mean ±SD)NS-RNS-NRp
Occiput to wall, cm2.7 (5.5)4.9 (7.8)<0.001
Lateral spine flexion, cm16.8 (5.6)12.8 (6.1)<0.001
Chest expansion, cm5.1 (2.4)4.1 (2.2)<0.001
BASFI1.9 (1.8)4.6 (2.8)<0.001
BASDAI2.6 (1.6)5.7 (2.2)<0.001
mSASSS16.9 (23.4)15.9 (21.9)0.8
CRP10.9 (12.5)13.2 (21.6)0.2
ESR13.1 (14.5)15.5 (15.6)0.2
Disease duration, yr7.5 (9.5)9.3 (10.1)0.06
Age, yr39.3 (14.0)39.7 (12.8)0.8

The following parameters examined failed to identify differentiating features between NS-R and NS-NR: age, gender, HLA-B27 status, use of methotrexate or sulfasalazine, concomitant uveitis or psoriasis or IBD, CRP or ESR. There was a trend toward shorter disease duration but this was not statistically significant.


In our AS cohort, 26% of patients were NS-R, achieving a satisfactory symptomatic state with NSAID therapy alone. Of interest, the radiographic severity was no different in NS-R vs NS-NR, suggesting a disconnect between symptoms and radiographic damage. While NS-R patients manifested a milder disease profile, there was no clinical feature which predicted response to NSAID, raising the possibility that genetic factors may dictate this difference.

To cite this abstract, please use the following information:
Alnaqbi, Khalid A., Haroon, Nigil, Shen, Hua, Cook, Richard J., Carty, Adele, Inman, R. D.; Profiling NSAID Responders in Ankylosing Spondylitis. [abstract]. Arthritis Rheum 2011;63 Suppl 10 :1330

Abstract Supplement

Meeting Menu