Scoring radiographic damage with two methods in South African rheumatoid arthritis patients

Abstract

INTRODUCTION: The many radiographic damage scoring methods in rheumatoid arthritis (RA) patients such as the Sharp van der Heijde (SHS) method are difficult to implement in a clinical setting. The Simple Erosion and Narrowing Score (SENS) method as well as the Scott modification of the Larsen scoring method are both easier to perform and reliable in early RA. OBJECTIVES: To evaluate the level of agreement between two scoring methods, the modified Larsen and SENS, and to correlate these scoring systems between radiographic damage, disease activity and disability in early disease modifying antirheumatic drug (DMARD) naïve RA patients METHOD: Baseline digital radiographs of hands and feet of RA patients attending the Arthritis clinic of the Chris Hani Baragwanath Academic Hospital were independently scored by 2 radiologists using the modified Larsen and SENS methods. All patients were DMARD naive with a symptom duration of ≤ 2years. Baseline clinical data, disease activity as calculated by the simplified and the clinical disease activity index and physical disability as measured by the health assessment questionnaire was obtained from the METEOR database. The Spearman covariant assessed the correlation between the two scoring systems, and between radiological scores and disease activity and physical function. RESULTS: There is strong correlation (r= 0.892; p-value 0.00) between the SENS and the modified Larsen scoring methods in patients with early RA. There is no significant correlation v between the modified Larsen score and the SENS and disease activity measures, (CDAI (pvalue 0.479) and SDAI (p-value 0.746)) and (CDAI p-value 0.77; SDAI p-value 0.86)) respectively. There is no correlation between the modified Larsen score and the SENS with the HAQ, (r = -0.168; p-value 0.104) and (r= -0.101; p-value 0.332) respectively. CONCLUSIONS: Both the SENS and the modified Larsen scoring method are simple to perform and have a high level of agreement in early RA. There is no correlation between either scoring methods with disease activity or physical function.