Abstract
Objective. In rheumatoid arthritis (RA), radiographic progression may occur despite clinical remission. This may be explained by subclinical inflammation. Magnetic resonance imaging (MRI) provides a greater sensitivity than clinical examination and radiography for assessing disease activity. Our objective was to determine the MRI characteristics of RA patients in clinical remission or low disease activity (LDA) state.
Methods. Databases from 6 cohorts were collected from 5 international centers. RA patients in clinical remission according to Disease Activity Score28-C-reactive protein (DAS28-CRP < 2.6; n = 213) or LDA-state (2.6 ≤ DAS28-CRP < 3.2; n = 81) with available MRI data were included. MRI were assessed according to the OMERACT RA MRI scoring system (RAMRIS).
Results. Patient characteristics: 70% women, median age 55 (interquartile range, IQR 43–63) years, disease duration 2.3 (IQR 0.7–5.1) years, DAS28-CRP 2.2 (IQR 1.8–2.6), Simplified Disease Activity Index, SDAI, 3.9 (IQR 1.9–6.5), Clinical Disease Activity Index, CDAI, 3.1 (IQR 1.5– 5.8), rheumatoid factor/anti-cyclic citrullinated peptide positivity 57%/54%, presence of radiographic erosions: 66%. Wrist and metacarpophalangeal MRI (MCP-MRI) data were available for 287 and 241 patients, respectively. MRI inflammatory activity in wrist and/or MCP joints was observed in the majority [synovitis: 95%, bone edema (osteitis): 35%] of patients. The median (IQR) RAMRIS score was 6 (3–9) for synovitis and 0 (0–2) for osteitis. Synovitis and osteitis were not less frequent in DAS28 clinical remission (synovitis/osteitis 96%/35%) than LDA (91/36). A trend towards lower frequencies of osteitis in patients in SDAI and CDAI remission was observed.
Conclusion. Subclinical inflammation was identified by MRI in the majority of RA patients in clinical remission or LDA state. This may explain structural progression in such patients. Further work is required to understand the place of modern imaging in future remission criteria.
In the last few years, remission in rheumatoid arthritis (RA) has become an achievable therapeutic goal1. It is generally defined as the absence of inflammatory activity based upon clinical criteria. Recent studies have shown that radiographic structural progression may occur despite clinical remission or low disease activity (LDA) state2,3,4. Modern imaging methods such as magnetic resonance imaging (MRI) provide the potential to improve the evaluation of disease activity beyond clinical examination5. A recent study has demonstrated that MRI-detected “subclinical inflammation” may be present in LDA patients and is related to subsequent radiographic progression2,6.
The objective of this study was to determine the MRI characteristics of patients with RA in clinical remission or LDA state.
MATERIALS AND METHODS
Patients
Anonymized databases from 6 different cohorts were collected from 5 centers (Table 1). For the purposes of this combined cohort [Outcome Measures in Rheumatology Clinical Trials (OMERACT) Rheumatoid arthritis Acceptable Disease Activity State (ORAS) cohort], RA patients (American College of Rheumatology 1987 criteria) in clinical remission [defined as Disease Activity Score28-C-reactive protein (DAS28-CRP) < 2.6] or LDA state (defined as 2.6 ≤ DAS28-CRP < 3.2) with available MRI data were included. Patients could be treated by disease-modifying antirheumatic drugs and/or biologics. Clinical data [age, sex, disease duration in years, treatment, tender joint count, swollen joint count, patient visual analog scale (VAS) global assessment, physician VAS global assessment] and laboratory tests (erythrocyte sedimentation rate, CRP, rheumatoid factor, anti-cyclic citrullinated peptide antibody status) at baseline were collected. DAS28-CRP, the Clinical Disease Activity Index (CDAI), the Simplified Disease Activity Index (SDAI), proportions of patients in clinical remission (DAS28 < 2.6, CDAI ≤ 2.8, SDAI ≤ 3.3), and LDA state (2.6 ≤ DAS28 < 3.2, 2.8 < CDAI 10, 3.3 < SDAI ≤ 11) for the different composite indices were calculated.
Characteristics of the individual studies.
MRI acquisition and scoring
An overview of MRI acquisition data in the different cohorts is provided in Table 1. MRI of unilateral wrist and/or metacarpophalangeal (MCP) joints 2–5 was acquired. Synovitis, erosion, and osteitis were defined and scored semiquantitatively according to the OMERACT RAMRIS7,8,9,10,11. By adding scores from the individual joint regions, MRI sum scores for synovitis (wrist: 0–9/MCP: 0–12/wrist+MCP: 0–21), osteitis (0–45/0–24/0–69), and bone erosion (0–150/0–80/0–230) were calculated.
Statistical analyses
A single database was compiled from the individual cohort data. Data were then analyzed using SAS software, version 9.1. Descriptive statistics and nonparametric tests were used.
RESULTS
Patient characteristics
Included in the study were 294 RA patients, of which 213 were in clinical remission and 81 in LDA state. Patient characteristics are summarized in Table 2. Thirty-nine percent of patients had a disease duration less than 1 year. Clinical disease activity was low, with median (IQR) DAS28-CRP 2.2 (1.8–2.6) (n = 294), SDAI 3.9 (1.9–6.5) (n = 188), and CDAI 3.1 (1.5–5.8) (n = 188). According to DAS28-CRP, 72% and 28% of these patients were in remission and LDA state, respectively. SDAI and CDAI definitions of remission appeared more stringent: Of the 188 patients with data available for calculating both DAS28-CRP, CDAI, and SDAI indices, 75%, 47%, and 45% of the patients were in remission and 25%, 49%, and 52% were in LDA but not remission according to DAS28-CRP, CDAI, and SDAI, respectively. Four percent and 3% of the patients were in neither LDA nor remission according to CDAI and SDAI, respectively.
Characteristics of the patients.
MRI findings
Wrist and MCP-joint MRI data were available for 287 patients and 241 patients, respectively. MRI inflammatory activity was observed in the majority of patients, as synovitis and osteitis were observed in wrist and/or MCP joints in 95% and 35% of the patients, respectively (Table 3). Figure 1 illustrates the frequency of synovitis and osteitis in wrist and MCP joints in patients in clinical remission. The median (IQR) RAMRIS score (total wrist and MCP) for synovitis was 6 (3–9) and for osteitis was 0 (0–2). Synovitis and osteitis were more commonly observed in the wrist (90% and 31%) than MCP (81% and 16%) joints. The second MCP joint was the most frequently involved MCP joint for both synovitis and osteitis (Table 3). Bone erosion was present in 90% of the patients and more frequently observed in wrist (89%) than in MCP (60%) (Table 3).
Frequencies of synovitis, osteitis, and bone erosion in patients with clinical remission (Disease Activity Score 28 < 2.6).
MRI characteristics of patients in clinical remission or low disease activity state according to Disease Activity Score 28-C-reactive protein (DAS28-CRP). No statistically significant differences were observed for presence (Fisher’s exact test) or score (Mann-Whitney U test) of any MRI feature.
No difference was observed between MRI characteristics of patients in clinical remission or LDA, regardless of DAS28-CRP (Table 3), CDAI, or SDAI definitions. While no difference was observed for MRI synovitis with regard to the different remission criteria, a tendency for lower frequencies of osteitis was observed when SDAI/CDAI criteria, as compared to DAS28 criteria, were used (Table 4).
MRI synovitis and osteitis in RA patients fulfilling different clinical remission criteria (n = 188).
Patients with early RA had significantly lower RAMRIS osteitis and erosion scores than those with established RA: the median (IQR) RAMRIS osteitis score was 0 (0–0) for early RA versus 0 (0–2) for established RA (p < 0.001; Figure 2) and the RAMRIS erosion score 2 (0–5) for early RA versus 11 (7–19) for established RA (p < 0.0001). No differences in synovitis scores were observed.
Magnetic resonance imaging (MRI) osteitis in patients with early RA and late RA in clinical remission or low disease activity state. MCP: metacarpophalangeal joint; RAMRIS: Rheumatoid Arthritis MRI scoring system.
DISCUSSION
Clinical remission and LDA are considered realistic therapeutic targets in RA, and treatment should be adjusted if these targets are not achieved in individual patients1. Assessments from imaging modalities are currently not part of the criteria for remission and LDA, which are currently defined solely according to clinical criteria. The present data from multiple cohorts demonstrate clearly that MRI inflammation is frequent both in patients in clinical remission and in patients in LDA state. Inflammatory activity was observed in the majority of patients, with synovitis and osteitis in wrist and/or MCP joints in 95% and 35% of the patients, respectively. These results confirm, in a much larger cohort, the results previously reported from one of the included studies6.
MRI showed no significant difference in signs of inflammation when patients in clinical remission were compared with patients with LDA without remission, whatever the criteria used (DAS28, CDAI, SDAI). A tendency towards lower frequencies of MRI osteitis in patients in remission according to SDAI or CDAI compared to DAS28 was observed in both wrist and MCP joints, whereas no difference in frequencies of synovitis was observed whatever the remission criteria employed. SDAI and CDAI remission criteria are often considered more stringent than DAS28 remission criteria, and the observed tendency towards less osteitis in patients in SDAI/CDAI remission supports this notion. The available data also suggest that MRI could be a useful external reference method for different clinical remission criteria.
Osteitis and bone erosion, but not synovitis, were observed less frequently in patients with early RA compared to those with established RA. As osteitis is known to be an independent predictor of subsequent bone erosion12,13,14, one could speculate that clinical remission in established RA is associated with a higher risk of subsequent erosive progression than clinical remission in early RA. However, longitudinal studies of clinical, MRI, and radiographic data are needed to clarify the relative importance of the observed findings.
Recent studies have demonstrated that progression of joint damage may occur in patients in clinical remission2,3,4. Subclinical MRI inflammation may contribute to structural progression, and further multivariate analyses of longitudinal data are required to determine the strongest predictive factors in these patients fulfilling criteria for remission. The aim would be to determine an MRI-acceptable inflammatory disease activity state, where structural progression will not occur.
Our study presents some limitations. Patients from different cohorts having different study designs and using different MRI units were included. To homogenize the data analyzed in the study, DAS28-CRP, CDAI, and SDAI were recalculated based on the raw clinical data from the original cohorts. The original MRI assessments from the individual cohorts were used in the analyses, as re-reading of all MRI data was not feasible due to the high number of examinations. Nevertheless, most of the readers had previously participated in RAMRIS multireader exercises, which showed satisfactory interreader reliability.
In summary, MRI identified inflammatory activity in the majority of patients in clinical remission or LDA state. MRI showed no differences between patients in clinical remission and LDA, despite a tendency to lower osteitis scores in patients in CDAI and SDAI remission. Further work is required to understand the place of modern imaging in future remission criteria.
Acknowledgment
Physicians, study nurses, and patients who have contributed to the described RA cohorts are acknowledged for their contributions.
