Arthritis & Rheumatism, Volume 62,
November 2010 Abstract Supplement

Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Atlanta, Georgia November 6-11, 2010.


Reliability of Radiographic Scoring Methods in Axial Psoriatic Arthritis.

Chandran3,  Vinod, Biagioni3,  Bradley, Cook1,  Richard, Eder3,  Lihi, Wakhlu3,  Anupam, Li3,  Michael, Shen1,  Hua

Department of Statistics and Actuarial Science, University of Waterloo
Toronto Western Hospital, Toronto, ON, Canada
University of Toronto Psoriatic Arthritis Clinic, Toronto Western Hospital

Background:

More than a third of patients with psoriatic arthritis (PsA) have axial arthritis. Although there are similarities, important differences exist between axial PsA (AxPsA) and Ankylosing Spondylitis (AS). The Bath AS Radiology Index (BASRI), modified Stokes AS Spinal Score (mSASSS) and Radiographic AS Spinal Score (RASSS) were developed to score AS, and Psoriatic Arthritis Spondylitis Radiology Index (PASRI) to score AxPsA. Our aim was to determine the reliability of these scoring systems in AxPsA.

Methods:

A computerized scoring module that facilitated scoring radiographic features of axial arthritis on vertebrae, sacroiliac and cervical facet joints, and calculated total scores for all the available scoring methods for AS and AxPsA was developed. Four assessors all rheumatologists blinded to patient information were provided with an e-handbook and one hour training session. Spinal radiographs of 18 patients with AS (satisfying modified NY criteria) and 40 patients with AxPsA (defined as at least unilateral grade 2 sacroiliitis and/or inflammatory back pain and/or restricted spinal mobility) were duplicated, patient identifiers removed, and order randomized. Radiographs were read by the four assessors individually, data entered into the module and scores obtained. Intraclass correlation coefficient (ICC) estimates of the inter- and intra-assessor reliability of scores for each method were obtained. A consensus score for AxPsA patients was also obtained.

Results:

The intra-and inter- assessor reliability for the radiographic scoring methods in AS are given in table 1.

Table 1. Inter-class and intra-class correlation coefficients and 95% confidence intervals of radiographic scores in AS (N=18).

Scoring SystemReliabilityICC95% CI
Sacroiliitis grade (NY criteria)Between Assessor0.7950.646, 0.905
 Within Assessor0.9080.837, 0.957
BASRI-spineBetween Assessor0.8610.746, 0.938
 Within Assessor0.9550.917, 0.98
mSASSSBetween Assessor0.8570.735, 0.938
 Within Assessor0.9820.967, 0.992
RASSSBetween Assessor0.7480.562, 0.888
 Within Assessor0.9610.929, 0.983
PASRIBetween Assessor0.9260.858, 0.968
 Within Assessor0.9850.971, 0.993

All methods had excellent inter- and intra- assessor reliability in AS. Scoring was completed by the assessors within an average duration of 7 minutes. The 40 patients with AxPsA (24 males, age 54 years, disease duration 18 years, mean cervical rotation of 61 degrees, Occiput to wall distance of 2.3 cm, Schober's test of 4.7 cm, Domjan lumbar lateral flexion of 15.5 cm, actively inflamed joint count of 5 and damaged joint count of 16 at the time of radiographic assessment) by consensus read had mean (± standard deviation) BASRI-spine score of 3.98 (2.38), mSASSS of 8 (13.4), RASSS 6.54 (14.1) and PASRI score of 12 (12.3) units. The intra- and inter- assessor reliability for the radiographic scoring methods in AxPsA are given in table 2.

Table 2. Inter-class and intra-class correlation coefficients and 95% confidence intervals of radiographic scores in AxPsA (N=40).

Scoring SystemReliabilityICC95% CI
Sacroiliitis grade (NY criteria)Between Assessor0.670.544, 0.785
 Within Assessor0.8070.726, 0.875
BASRI-spineBetween Assessor0.5210.375, 0.669
 Within Assessor0.7710.688, 0.845
mSASSSBetween Assessor0.6520.488, 0.798
 Within Assessor0.9110.862, 0.949
RASSSBetween Assessor0.680.513, 0.823
 Within Assessor0.9020.845, 0.946
PASRIBetween Assessor0.880.817, 0.927
 Within Assessor0.9170.874, 0.95
Only PASRI had excellent intra- and inter-rater reliability in AxPsA.

Conclusions:

Available radiographic scoring systems have at least moderate inter- and intrarater reliability when applied to AxPsA. The PASRI has the highest reliability in AxPsA and performs well in AS.

To cite this abstract, please use the following information:
Chandran, Vinod, Biagioni, Bradley, Cook, Richard, Eder, Lihi, Wakhlu, Anupam, Li, Michael, et al; Reliability of Radiographic Scoring Methods in Axial Psoriatic Arthritis. [abstract]. Arthritis Rheum 2010;62 Suppl 10 :1954
DOI: 10.1002/art.29719

Abstract Supplement

Meeting Menu

2010 ACR/ARHP