Article Text

FRI0245 How Reliable is the Scoring of Msasss in Clinical Practice in Centers Participating in Desir? Comparison with the Gold Standard Central Reading
  1. M. de Hooge1,
  2. P. Claudepierre2,
  3. A. Feydy3,
  4. M. Reijnierse1,
  5. A. Saraux4,
  6. M. Dougados3,
  7. D. van der Heijde1
  1. 1LUMC, Leiden, Netherlands
  2. 2Henri Mondor Teaching Hospital, Créteil
  3. 3Descartes University, Côchin Hospital, Paris
  4. 4CHU de la Cavale Blanche, Brest Cedex, France


Background Spinal X-rays are considered as gold standard for assessing structural damage in the spine in AS, and a scoring system, the modified Stoke Ankylosing Spondylitis Spinal Score (mSASSS) is the preferred assessment method. In clinical studies and therapeutic trials, the mSASSS scoring process is usually done by ≥1 trained readers. In daily practice, the ability of rheumatologists and radiologists to adequately use the mSASSS without a specific training is unknown. In addition, it is not known what the impact would be in studies when using the score of multiple readers in various centres as in daily practice, instead of using a centralized scoring with a few trained readers

Objectives To compare the results of the mSASSS of the local reading of baseline spinal X-rays to the centralized reading as the gold standard

Methods Patients aged 18-50 with recent chronic back pain (≥3 months, ≤3 years) from 25 participating centers in France were included in the DESIR (Devenir des Spondylarthropathies Indifferenciées Récentes) -cohort (n=708). All available baseline X-rays of cervical and lumbar spine were scored by the local radiologist/rheumatologist who might have access to clinical data, according to the mSASSS scoring method. In addition, 2 well-calibrated centralized readers independently scored the same X-rays, blinded for any other data. In case the centralized readers disagreed, an experienced radiologist served as adjudicator. Agreement between the 2 centralized readers, and between the local and centralized scores was calculated (Kappa; percentage agreement). To calculate the agreement between readers a cut-off of ≥1 for mSASSS was used. When comparing centralized readers with local readers the mSASSS of the centralized readers was combined

Results Patients with complete X- ray data (n=664) were included in these analyses. The large majority of patients had a normal mSASSS both scored by the central and local readers. The agreement between the 2 centralized readers was 89.3% with a kappa of 0.50 (see Table 1). Comparing the local readings with the centralized scores there was an agreement in 72.2% of the cases with a kappa of 0.19. The local readers scored an mSASSS ≥1 in 169 cases, while this was in 119 cases if scored by central readers

Table 1

Conclusions The agreement between two trained central readers is better than between central and local readers. Local readers overestimate damage in the spine in comparison to the gold standard of central reading

Disclosure of Interest None declared

DOI 10.1136/annrheumdis-2014-eular.1522

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.