Article Text
Abstract
Objective To determine how the European League Against Rheumatism (EULAR) definition of erosive disease (erosion criterion) contributes to the number of patients classified as rheumatoid arthritis (RA) according to the 2010 American College of Rheumatology/EULAR RA classification criteria (2010 RA criteria) in an early arthritis cohort.
Methods Patients from the observational study Norwegian Very Early Arthritis Clinic with joint swelling ≤16 weeks, a clinical diagnosis of RA or undifferentiated arthritis, and radiographs of hands and feet were included. Erosive disease was defined according to the EULAR definition accompanying the 2010 RA criteria. We calculated the additional number of patients being classified as RA based on the erosion criteria at baseline and during follow-up.
Results Of the 289 included patients, 120 (41.5%) fulfilled the 2010 RA criteria, whereas 15 (5.2%) fulfilled only the erosion criterion at baseline. 118 patients had radiographic follow-up at 2 years, of whom 6.8% fulfilled the 2010 RA criteria and only one patient fulfilled solely the erosion criterion during follow-up.
Conclusion Few patients with early arthritis were classified as RA based on solely the erosion criteria, and of those who did almost all did so at baseline.
- early rheumatoid arthritis
- epidemiology
- outcomes research
- disease activity
Statistics from Altmetric.com
Introduction
The aim with the 2010 American College of Rheumatology (ACR)/European League Against Rheumatism (EULAR) rheumatoid arthritis (RA) classification criteria (2010 RA criteria) was to identify patients at high risk for developing persistent erosive and/or inflammatory disease in the early stage of the disease. Erosions were not considered for inclusion in the scoring system by the ACR/EULAR working group because patients with erosions typical of RA were presumed to have prima facie evidence of RA.1 Later, a task force suggested that the presence of ≥3 joints with typical erosions is sufficient to classify patients as having RA based on erosive disease on radiographs alone (erosion criterion).2 3
The main objective of this study was to assess to what degree the EULAR definition of erosive disease contributes to the number of patients classified as RA according to the 2010 RA criteria.
Methods
Setting and patient selection
The current analyses were based on data from the observational, prospective Norwegian Very Early Arthritis Clinic study, including patients with ≥1 clinically swollen joint of ≤16 weeks’ duration.4 The cohort included 1118 patients (age 18–75 years) between years 2004 and 2010 with study visits at baseline and after 3, 6, 12 and 24 months. Patients with joint swelling due to trauma, osteoarthritis, crystal arthritis or septic arthritis were excluded. The study was approved by the Regional Ethics Committee of Southern Norway.
In this current study patients with a clinical diagnosis other than RA or undifferentiated arthritis (UA) at baseline were excluded. We included the remaining patients with radiographs of hands and feet at baseline (online supplementary file 1). A subset of these patients had radiographs both at baseline and at 2 years of follow-up (online supplementary file 2, figure S2). The 2010 RA criteria were retrospectively applied at baseline and cumulatively at follow-up visits.
Supplementary file 1
Supplementary file 2
Data collection and radiographic assessment
The full data collection has been described elsewhere.4 In the current study, conventional hand and feet radiographs were performed at baseline and at 24 months in patients included from year 2007 and onwards.
A trained reader, blinded to patient characteristics, scored radiographs of hands and feet according to the van der Heijde modified Sharp score method.5 The time order of the radiographs was known. We defined erosive disease according to the EULAR definition accompanying the 2010 RA criteria: ‘Erosive disease for use in the 2010 RA criteria is defined when an erosion (defined as a cortical break) is seen at at least three separate joints at any of the following sites: the proximal interphalangeal (PIP) joints, metacarpophalangeal (MCP) joints, the wrist (counted as one joint) and the metatarsophalangeal (MTP) joints on radiographs of both hands and feet’.3 The wrist includes the carpometacarpal(CMC) bone, trapezium, scaphoid, lunate, radial and ulnar bone. The 1st interphalangeal (IP1) joint of the feet is included in the MTP joints.
Analyses
The number of patients being classified as RA based on the 2010 RA criteria and on the erosion criterion (without fulfilling the 2010 RA criteria), as well as the number of patients with ≥1 and ≥2 erosive joints, were calculated at baseline and during follow-up. We also evaluated the distribution of erosive joints. Additionally, incident erosions during follow-up in patients without baseline erosions were determined.
Results
Fulfilment of the 2010 RA criteria and the EULAR definition of erosive disease (erosion criterion)
The baseline characteristics of the 289 patients included in the current study are shown in online supplementary file, table S1. One hundred and twenty (41.5%) patients fulfilled the 2010 RA criteria at baseline, of whom 49 (40.8%) patients had ≥1 erosive joint and 17 (14.2%) fulfilled the erosion criterion. Of the remaining 169 patients not fulfilling the 2010 RA criteria, 15 patients fulfilled the erosion criterion, 27 had ≥2 erosive joints and 55 patients had ≥1 erosive joint.
One hundred and eighteen patients had 2-year radiographic follow-up data. Presence/absence of erosions at baseline and follow-up in relation to the fulfilment of the 2010 RA criteria is shown in figure 1. Of all 118 patients, 8 additional patients (6.8%) fulfilled the 2010 RA criteria during follow-up, while 6 fulfilled the erosion criterion alone, of whom 5 patients at baseline.
Characteristics of patients with erosions
Table 1 presents the baseline characteristics of patients not fulfilling and patients fulfilling the 2010 RA criteria. The 15 patients who fulfilled the erosion criterion but not the 2010 RA criteria were all seronegative. They also had numerically shorter duration of joint swelling, were more often men and had fewer involved joints than those who fulfilled the 2010 RA criteria. One of these 15 patients received the final clinical diagnosis RA, while the rest of the patients were diagnosed with UA (n=10), reactive arthritis (n=1), chondrocalcinosis (n=1) or osteoarthritis (n=2). Furthermore, 2 of the 15 patients were treated with disease modifying antirheumatic drugs (DMARDs) (1 patient with RA and 1 patient with reactive arthritis).
Distribution of erosive joints at baseline in patients not fulfilling the 2010 RA criteria and incident erosions during follow-up
As shown in table 2, the MTP and PIP joints were the most frequently affected joints at baseline. Of the 169 patients not fulfilling the 2010 RA criteria, 40 patients had ≥1 hand erosion, 28 had ≥1 foot erosion and 13 had erosions in both hands and feet at baseline. Among patients with no baseline erosions (n=74), 13 (17.6%) developed erosions during follow-up (PIP joints n=3, MCP n=4, wrist n=4, MTP joints n=12), of whom 7 fulfilled the 2010 RA criteria at baseline, and 2 patients fulfilled both the 2010 RA criteria (at baseline) and the erosion criterion (during follow-up).
Discussion
Nearly 42% of the 289 patients fulfilled the 2010 RA criteria at baseline, and only additional 6.8% (eight patients) did so during the 2-year follow-up. In total, 16 patients were classified as RA based on the erosion criterion (15 at baseline and 1 during follow-up). We obtained nearly the same percentage fulfilling the erosion criterion at baseline as observed in the two cohorts used to define the erosion criterion (5.2% vs 3.3%, respectively).3
Finding patients with ≥3 erosive joints at baseline is surprising considering the short duration of joint swelling, that is, ≤16 weeks. A proportion of patients have erosions at first presentation,6 and this has been found to be a predictor for severe destructive disease in patients with early RA.7 8 However, data from our study regarding DMARD use and final clinical diagnosis indicate that the majority of patients fulfilling only the erosion criterion at baseline were false-positives.
To our knowledge, few studies have looked at the role of the erosion criterion in classifying patients with early RA. Le Loët et al9 studied 310 patients with early arthritis (median symptom duration of 128 days), and no patients with a 2010 RA criteria score <6 fulfilled the erosion criterion at baseline. Patients in our study had a higher mean Disease Activity Score-28 at baseline than the French patients, which could have contributed to the discrepancy in the erosive findings. The patients in the present study who solely fulfilled the erosion criterion at baseline were all seronegative, had fewer involved joints and more often male compared with patients fulfilling the 2010 RA criteria. Most of these are expected, as being seropositive and having more involved joints are included as points in the 2010 RA criteria.
The MTP joints in the feet had the highest occurrence of erosions both at baseline and during follow-up. A previous study of patients with UA with erosive joints at baseline suggested that presence of erosions in the joints of the feet was slightly more predictive for developing RA than erosions in the hand joints.10
The task force that defined the erosion criterion concluded that the specificity of a cut-off of ≥2 erosive joints would be too low.3 Our results show that using ≥2 erosive joints as cut-off would have increased the number of patients being classified as RA with 27 (16.0%) at baseline and another 4 patients during follow-up. Although the baseline characteristics of patients with ≥2 and ≥3 were quite similar, the low number of patients with radiographic follow-up data makes it difficult to consider the consequences of having ≥2 erosive joints as cut-off regarding disease course and outcome.
A limitation of our study is the low rate of radiographic follow-up. We believe that many patients declined or were not referred to radiographic follow-up because they were feeling healthy. Another limitation is the small number of patients with erosive joints in general, which precludes meaningful statistical comparisons. Additionally, if our inclusion criteria had allowed for longer duration of joint swelling, the proportion of patients developing ≥3 erosive joints might have been larger, as RA often has an insidious onset. Ideally, the baseline radiographs should also have been read separately; however, we do not think this has had a major impact on the results because the reader was unaware of the purpose of scoring the radiographs. Additionally, there were only a few patients developing incident erosions.
In conclusion, few patients were classified as RA based on the erosion criterion without fulfilling the 2010 RA criteria. Of those who did, almost all did so at baseline; thus, our results suggest that follow-up radiographs in patients with early UA might be of limited value for classifying patients with RA. Furthermore, data regarding DMARD use and clinical diagnosis indicate that despite having erosions at baseline, patients with UA may end up with other clinical diagnoses than RA.
Supplementary file 3
Acknowledgments
The authors thank the patients, the physicians and study nurses, and the laboratory and radiology staff for participating in the NOR-VEAC study.
References
Footnotes
Handling editor Gerd R Burmester
Contributors GHB, ESN, MDM, DvdH, TKK and EL contributed to the study design, including formulation of the research questions. GHB, ESN, MDM and EL were responsible for analysing and interpreting the data, and drafting the manuscript. All authors critically revised the manuscript and approved the final version.
Competing interests None declared.
Patient consent Obtained.
Ethics approval The Regional Ethics Committee of Southern Norway.
Provenance and peer review Not commissioned; externally peer reviewed.