AB0269 COMPARISON BETWEEN DIFFERENT DISEASE ACTIVITY SCORES IN ELDERLY ONSET RHEUMATOID ARTHRITIS. (June 2019)

• Record Type: Journal Article
• Authors: Martínez-Morillo, Melania; Prior-Español, águeda; Brandy-Garcia, Anahy; Holgado, Susana; Aparicio-Espinar, María; Gifre, Laia; Riveros-Frutos, Anne; Sangüesa-Gomez, Clara; Camins-Fàbregas, Jordi; Casafont-Solé, Ivette; Nack, Annika; Olive, Alejandro; Soria, Lourdes Mateo
• Is Part Of: Annals of the rheumatic diseases. Volume 78(2019)Supplement 2
• Page Start: 1592
• Page End: 1592
• Publication Date: 2019-06
• Subjects: Rheumatism -- Periodicals; 616.723005
• DOI: 10.1136/annrheumdis-2019-eular.3453
• Languages: English
• ISSNs: 0003-4967

Abstract

Abstract : Background: Elderly onset rheumatoid arthritis (EORA) has several peculiarities. Which activity index should we use in this age group is still an unresolved issue. The use of ESR in DAS28 is widespread, but it is well known that ESR could increase with age, overestimating activity. DAS28-PCR or SDAI, which uses PCR, or CDAI that does not use analytical data, could be more suitable alternatives. Objectives: To describe disease activity scores evolution in EORA and to analyse the correlation and concordance between them. Methods: Longitudinal observational study in naïve treatment EORA patients (ACR/EULAR 2010 criteria). Study visits were carried out basally (pre-treatment) and at month 1, 3, 6 and 12. DAS28 VSG, DAS28 PCR, SDAI and CDAI were calculated at each visit, and ACR/EULAR remission index at 12 months. The same rheumatologist made all visits and explorations. Correlations were analysed using the Spearman or Pearson correlation index (p value ≤ 0.001). The concordance was evaluated with the kappa index. The statistical study was performed with Stata 15.1. Results: 45 patients with EORA were enrolled. Table 1 and 2 summarize the clinical and analytical characteristics. All the scores had a very good linear correlation both at baseline and at follow-up (correlation score: 0.83-0.98). A moderate concordance was observed from the baseline visit between DAS28-VSG and DAS28-PCR (k=0.43-0.52, p <0.001), which decrease to low at 12 months (k=0.32, p <0.001). A concordance of 100% was observed at baseline analysing DAS28-VSG and SDAI (k=1; p <0.001), but worsened during follow-up being moderate (k=0.40-056; p<0.001). A very good concordance was observed comparing DAS28-VSG to CDAI at the baseline visit (k=0.88, p<0.001), which decreased to good-moderate during follow-up (k=0.40-0.63, p<0.001). Between DAS28-PCR and SDAI a moderate concordance was observed up to 6 months (k=0.43-0.57, p <0.001), which worsened to low at 12 months (k=0.32, p<0.001). The concordance between DAS28-PCR and CDAI was low-moderate (k=0.26-0.48, p <0.001). On the other side, the agreement was very good between SDAI and CDAI throughout the follow-up (k=0.86-0.93, p <0.001). The ACR/EULAR index had a very low concordance with DAS28-PCR (k=0.17, p <0.001), moderate with DAS28-VSG (k=0.49, p<0.001) and the results are good with SDAI and CDAI (k=0.64, p<0.001). Conclusion: The correlation between the different scores was good. However, the concordance between all the scores decreases as time passes, according to a higher percentage of patients in remission or low activity. This low-moderate concordance is demonstrated even between DAS28-PCR and DAS28-VSG, and when these are compared with CDAI and SDAI. CDAI and SDAI are the only ones that maintain a good concordance between them, even when the number of patients in remission or low activity increases. ACR/EULAR only maintains an acceptable concordance with the most restrictive scores, SDAI and CDAI. By contrast, agreement is very low with DAS28, especially by DAS-PCR. Disclosure of Interests: None declared