Reliability and sensitivity to change of a simplification of the Sharp/van der Heijde radiological assessment in rheumatoid arthritis.

Abstract

To determine the reliability and sensitivity to change of a simplified radiological scoring method [simple erosion narrowing score (SENS)] for rheumatoid arthritis (RA). SENS was compared to the Sharp/van der Heijde score (SHS) as a gold standard. Sets of seven radiographs of hands and feet were taken of 20 RA patients with a wide spectrum of radiological damage. For 14 patients, these seven radiographs were taken during a follow-up period of 5 yr, and for six patients during a follow-up of 10 yr. Each set of radiographs was scored twice by the same observer (DvdH). Erosions and joint space narrowing were scored with SHS (range 0-448) in 32 and 30 joints in the hands, respectively, and both in 12 joints in the feet. SENS gives a score of 1 if there is any erosion in a joint and also 1 if there is any narrowing in the joint (range 0-86). In each case, SENS was derived from SHS. To analyse data, generalizability theory and repeated measurements ANOVA were used. The overall reliability coefficient was 0.81 for SHS and 0.80 for SENS. Intra-observer reliability [intraclass correlation coefficient (ICC)] was 0.99 and 0.98 for SHS and SENS, respectively. The ICC for the sensitivity to change was 0.84 for SHS and 0.88 for SENS. The smallest detectable difference (SDD) could be determined for both methods. The presence of progression based on this SDD was very comparable between the two methods. The measurement properties of SENS are good and comparable to SHS. This makes SENS suitable for use in clinical practice and in large (epidemiological) studies, especially in the first years of disease.