Three Groups in the 28 Joints for Rheumatoid Arthritis Synovitis – Analysis Using More than 17,000 Assessments in the KURAMA Database

Chikashi Terao,Tsuneyo Mimori,Daisuke Kawabata,Motomu Hashimoto,Kosaku Murakami,Ryo Yamada,Fumihiko Matsuda,Koichiro Ohmura,Noriyuki Yamakawa,Naoichiro Yukawa,Hiroyuki Yoshitomi,Moritoshi Furu,Takao Fujii,Takashi Usui,Hajime Yoshifuji,Hiromu Ito,Ran Nakashima,Keiichi Yamamoto,Bernhard Kaltenboeck

doi:10.1371/journal.pone.0059341

Abstract

Rheumatoid arthritis (RA) is a joint-destructive autoimmune disease. Three composite indices evaluating the same 28 joints are commonly used for the evaluation of RA activity. However, the relationship between, and the frequency of, the joint involvements are still not fully understood. Here, we obtained and analyzed 17,311 assessments for 28 joints in 1,314 patients with RA from 2005 to 2011 from electronic clinical chart templates stored in the KURAMA (Kyoto University Rheumatoid Arthritis Management Alliance) database. Affected rates for swelling and tenderness were assessed for each of the 28 joints and compared between two different sets of RA patients. Correlations of joint symptoms were analyzed for swellings and tenderness using kappa coefficient and eigen vectors by principal component analysis. As a result, we found that joint affected rates greatly varied from joint to joint both for tenderness and swelling for the two sets. Right wrist joint is the most affected joint of the 28 joints. Tenderness and swellings are well correlated in the same joints except for the shoulder joints. Patients with RA tended to demonstrate right-dominant joint involvement and joint destruction. We also found that RA synovitis could be classified into three categories of joints in the correlation analyses: large joints with wrist joints, PIP joints, and MCP joints. Clustering analysis based on distribution of synovitis revealed that patients with RA could be classified into six subgroups. We confirmed the symmetric joint involvement in RA. Our results suggested that RA synovitis can be classified into subgroups and that several different mechanisms may underlie the pathophysiology in RA synovitis.

Highlights

Rheumatoid arthritis (RA) is the most frequent inflammatory arthritis worldwide affecting 0.5 to 1% of the population [1]
We showed that affected rates of the 28 joints greatly vary in RA patients, and that RA patients could be classified into subgroups based on the distribution of joint synovitis
Joint symptoms especially joint swelling is known to correlate with future joint damage [3]

Summary

Introduction

Rheumatoid arthritis (RA) is the most frequent inflammatory arthritis worldwide affecting 0.5 to 1% of the population [1]. All of the three indices are shown to be well correlated with future joint destruction [7,9] These three methods include the same 28 joints for evaluation of disease activity, namely, bilateral wrist, 1st to 5th metacarpal (MCP) joints and proximal interphalangeal (PIP) joints, elbow, shoulder, and knee joints. There are several reports of successful prediction of joint damage using a reduced number of joints for evaluation by ultrasonography [11,12]. These reports raise the possibility that some of the 28 joints are less frequently involved, and are less informative for disease activity.

Methods

Results

Conclusion