Arthritis & Rheumatism, Volume 60,
October 2009 Abstract Supplement

The 2009 ACR/ARHP Annual Scientific Meeting
Philadelphia October 16-21, 2009.

Axial Metrology Measurement and Functional Status in Ankylosing Spondylitis

Sieper1,  J., Singh2,  Amitabh, Freundlich2,  Bruce, Koenig2,  Andrew S., Li3,  Wenzhi

Rheum/Charite Hosp, Berlin, Germany,
Wyeth Research, Collegeville, PA,
Wyeth Research, Philadelphia, PA


It has been suggested that evaluation of spinal mobility using axial metrology could be used to assess disease modification in ankylosing spondylitis (AS). Although axial metrology has been shown to have a good correlation with radiography, its relationship with functional status has not been systematically evaluated. To assess the relationship between spinal mobility measured using axial metrology and functional status.


In a randomized, 16-week, double-blind, randomized, multicenter, active comparator [sulfasalazine (n=187)] study to assess the efficacy and safety of etanercept (50 mg once weekly (n=379)] in patients with AS, spinal mobility was measured by Bath Ankylosing Spondylitis Metrology Index (BASMI: range 0–10). Functional status was measured with Bath Ankylosing Spondylitis Functional Index (BASFI: range 0–100). We estimated how the end of study (Week 16) changes (d) in functional status would be related to d in disease activity, d in spinal mobility, age, gender and disease duration. The relationships were estimated with a linear regression model with change in BASFI as the dependent variable. Three alternative scoring algorithms have been proposed for BASMI: BASMI 2, BASMI 10 and BASMI Linear. Although the range of score is unaffected, BASMI Linear has been found to be most sensitive to change, followed by BASMI 10 and BASMI 2. Accordingly, three separate regression models were estimated using these different scoring algorithms for BASMI.


The study population was predominantly male (75%) with an average age of 40.7y and disease duration of 7.6y. The baseline BASMI 2, BASMI 10, and BASMI Linear scores were 3.66, 4.08, and 4.10 respectively. The table below presents the standardized beta coefficients for the 3 regression models (I: BASMI 2, II: BASMI 10, III: BASMI Linear)

 Regression Models
 I (N=519)II (N=519)III (N=519)
d BASDAI0.777+0.761+0.757+
d BASMI 2, d BASMI 10, d BASMI Linear-0.089+-0.115+-0.128
Disease Duration-0.034-0.033-0.033
Gender (Male=1)0.0200.0210.020
Adj R20.640.650.65

Each of the 3 models was able to explain about 64% of the variation in BASFI. Only d in disease activity and d in spinal mobility were significantly associated with BASFI. d in BASFI was independent of disease duration, gender and age. Apart from disease activity, changes in spinal mobility had the largest impact on functional status.


This study demonstrated that changes in spinal mobility are significantly associated with changes in functional status in patients with AS. Because BASMI has shown good correlations with radiography, the potential for using BASMI as a surrogate for disease modification should be further examined.

To cite this abstract, please use the following information:
Sieper, J., Singh, Amitabh, Freundlich, Bruce, Koenig, Andrew S., Li, Wenzhi; Axial Metrology Measurement and Functional Status in Ankylosing Spondylitis [abstract]. Arthritis Rheum 2009;60 Suppl 10 :1767
DOI: 10.1002/art.26841

Abstract Supplement

Meeting Menu