Article Text
Abstract
Objectives Treatment of rheumatoid arthritis (RA) should aim at full remission. The aims of this study were to define: (1) how many patients reached ultrasound power Doppler (US-PD) remission in a cohort of patients with early RA (ERA) compared with longstanding RA (LSRA); (2) possible predictors of US-PD remission; and (3) how many patients with and without US-PD remission relapsed after 1 year of follow-up in ERA and LSRA.
Methods 48 patients with ERA and 46 with LSRA with disease activity score <1.6 underwent US assessment. Six hand and wrist joints were studied for active synovitis. 56.2% of patients with ERA and 50.0% of those with LSRA fulfilled American College of Rheumatology (ACR) remission criteria.
Results 43.7% of patients with ERA and 17.4% of those with LSRA had no evidence of synovitis at US evaluation. Using a stricter clinical definition of remission (ie, ACR criteria), US evaluation confirmed clinical remission in 66.7% of patients with ERA and 26.1% of those with LSRA. Early disease was predictive of clinical US remission. 20.0% of patients with RA who had a negative PD signal at the US evaluation had a flare during the 12-month follow-up period compared with 47.1% of patients who had a positive PD signal.
Conclusion US-PD remission occurs in half of patients with ERA and in a minority of patients with LSRA in clinical remission. Early disease seems to be the major determinant of full remission.
Statistics from Altmetric.com
Introduction
In rheumatoid arthritis (RA), active synovitis appears to be the primary cause for structural joint damage and disease progression.1 The goal of modern therapeutic strategies is to eliminate synovitis and establish a state of remission. Randomised controlled clinical trials have confirmed there are differences in the chances of achieving remission in longstanding RA (LSRA) compared with early RA (ERA).2,–,4 We previously compared two cohorts of patients with ERA and LSRA to characterise the clinical and autoimmune features and did not find clearcut differences between them.5
There is evidence of joint damage progression despite clinical remission.6,–,8 This could reflect inadequate sensitivity of traditional clinical and laboratory approaches to detect active synovitis.
Recent studies have shown a superior sensitivity of ultrasound (US) compared with clinical evaluation for detecting inflammation and synovial hypertrophy (SH),9,–,12 and the power Doppler (PD) technique helps to distinguish active from inactive synovitis.13 14
The objectives of this study were to define:
▶ how many patients reached US-PD remission in a cohort of patients with ERA compared with patients with LSRA;
▶ possible predictors of US-PD remission;
▶ how many patients with and without US-PD remission relapsed after 1 year of follow-up.
Methods
US evaluation was performed in 48 consecutive patients with ERA and 46 with LSRA in stable clinical remission for at least 6 months (disease activity score (DAS) <1.6)15; 66.7% of the patients with ERA had reached clinical remission on monotherapy with methotrexate (MTX) while 33.3% were receiving a combination with anti-tumour necrosis factor α (TNFα). All patients with LSRA were receiving a disease-modifying antirheumatic drug (DMARD) plus anti-TNFα therapy. The DAS value was evaluated every 3 months until the 12th month from US evaluation to identify patients who maintained persistent clinical remission. A control group of 20 sex- and age-matched healthy subjects was also studied.
Bilaterally, six joint regions—chosen from those easiest to explore by US and most frequently involved in RA—were assessed using multiplanar circumferential grey scale and PD-US techniques: namely, the second and third metacarpophalangeal and proximal interphalangeal joints and the wrist. In particular, in the wrist we evaluated the radiocarpal joint, between the distal radius and the proximal carpal row and the synovitis around the triangular fibrocartilage which extends from the ulnar aspect of the distal radius to the base of the ulnar styloid. Two overall US joint scores for SH and PD signals were calculated as a sum of the scores obtained from each joint and separately for SH and PD. A commercially available real-time scanner (Logiq5; General Electric Medical Systems, Kyunngi, Korea) with a 7–12 MHz linear array transducer was used.
We considered three US scenarios:
▶ ‘active synovitis’: SH with PD signal (PD+/SH+);
▶ ‘US remission’: no SH or PD signal (PD−/SH−);
▶ ‘inactive synovitis’: SH without PD signal (PD−/SH+) (figure 1 and figure S1 in online supplement).
All analyses were carried out using SPSS 15.0 (Chicago, Illinois, USA).
Full details of the patients and methods are given in the online supplement.
Results
ERA and LSRA populations were comparable for all demographic, clinical and laboratory characteristics except for disease duration and treatment (table 1 and additional online text file). Three of the 20 healthy subjects had evidence of mild SH and a low PD signal in one wrist.
At the time of US assessment, all patients had reached a persistent DAS <1.6; only 56.2% of the patients with ERA and 50% of those with LSRA also satisfied ACR remission criteria.16 Despite EULAR-DAS remission, on musculoskeletal US examination 43.7% patients with ERA were in ‘US remission’ compared with 17.4% of patients with LSRA; 41.7% of the patients with ERA and 30.4% of those with LSRA had an increased PD signal; and 14.6% of the patients with ERA and 52.2% of those with LSRA had ‘inactive synovitis’ (table 2). Moreover, the patients with LSRA had higher PD and SH scores than those with ERA (3.0±3.1 vs 0.5±1.7 and 4.9±3.3 vs 1.9±2.5, respectively). In both RA cohorts, grey scale and PD-US scores correlated with swollen joint count at the time of US evaluation (r=0.25, p=0.05 and r=0.21, p=0.40, respectively) and with disease duration (r=0.47, p<0.0001 and r=0.48, p<0.0001, respectively).
US confirmed ACR remission in 66.7% of patients with ERA compared with 26.1% of patients with LSRA, while 14.8% of patients with ERA and 8.7% of patients with LSRA had SH with an increased PD signal and 18.5% of patients with ERA and 65.2% of those with LSRA had ‘inactive synovitis’.
Predictors of ultrasonographic remission
When considering both RA cohorts, ‘US remission’ (PD−/SH−) and the absence of a PD signal (PD−/SH− and PD−/SH+) were associated with early disease (OR 7.6, 95% CI 2.3 to 25.8 and OR 15.0, 95% CI 4.9 to 45.7, respectively); the same result was seen for patients in ACR remission (OR 7.7, 95% CI 2.0 to 29.5).
Twelve-month follow-up
Twelve months after US evaluation, 70.2% of patients with RA had stable DAS values <1.6 (table 1); 63.6% were patients with ERA while 36.4% were patients with LSRA. Of the patients with RA who had no PD activity at the US evaluation (PD−/SH− and PD−/SH+), 20.0% had a clinical flare during the 12-month follow-up period compared with 47.1% of patients with a PD signal at US evaluation (p=0.009).
Considering US scores, the mean PD and SH scores in patients who maintained clinical remission at 12 months were 1.1±2.3 and 2.6±2.8, respectively, compared with 3.1±3.4 and 5.2±3.5 in patients who had DAS >1.6 at 12 months. The only variable statistically associated with sustained remission was early disease (OR 13.6, 95% CI 3.4 to 54.0).
Discussion
The aims of this study were threefold:
▶ to determine whether US-PD remission occurred in ERA and LSRA in clinical remission with the same frequency;
▶ to define possible predictors of remission; and
▶ to define how many patients with and without US-PD remission relapsed after 1 year of follow-up in the two subsets.
Two cohorts of patients with ERA and LSRA in stable clinical remission were studied with grey scale and PD-US techniques, evaluating the presence of synovial inflammation, to verify if this approach could improve the accuracy of disease activity assessment in each cohort. In particular, we examined whether clinical and US remission could coincide in the same patient or whether there were differences and, if so, which. A number of reports have suggested a disparity in remission between clinical and US evaluation in ERA and LSRA cohorts.6 17 Brown and Saleem showed US-PD positivity in 63% and 48% of patients, respectively.6 9 Interestingly, in the study by Saleem et al9 the percentage of patients satisfying the criteria for full remission (clinical plus US) was 10% in the subcohort on MTX plus TNFα blocker and 16% in the subcohort on DMARDs only. No evidence is available for patients with ERA compared with LSRA.
Our data show that ‘US remission’ was present in 43.7% of patients with ERA but only in 17.4% of patients with LSRA. This is of paramount importance since we obtained the majority of remissions with MTX alone in patients with ERA while all the patients with LSRA were receiving MTX or other DMARDs plus TNFα blockers. ‘US remission’ can therefore be achieved in a substantial number of patients with ERA with MTX alone. However, the data show that clinical remission may be different in different phases of RA, suggesting the need for imaging support in the definition of true remission. In fact, a minority of patients with ERA and >50% of those with LSRA had US ‘inactive synovitis’; the US difference between the two cohorts, despite comparable DAS remission levels, was the presence of an inactive SH which was seen in the majority of patients with LSRA (52.2%) and only in 14.6% of those with ERA. Synovial tissue can become chronically thickened without a PD signal and is less reversible in established LSRA. The main conclusion is that LSRA is characterised by chronic thickened synovial tissue.
When we defined clinical remission with stricter criteria such as the ACR criteria, only 14.8% of patients with ERA and 8.7% of those with LSRA had ‘active synovitis’; the difference in SH was confirmed in the two groups. Stricter criteria therefore guarantee more chances of achieving long-term remission.
These results suggest good but not total sensitivity of traditional clinical approaches to assessing the true remission status. Furthermore, our results indicate that the only independent prognostic factor of full remission is early disease, highlighting the importance of early diagnosis.
We found that the percentage of patients with stable clinical remission after a long follow-up period was twofold higher in patients with ERA than in those with LSRA. The clearcut message is that early remission can be achieved and maintained with more stability in patients with ERA than in those in late remission.
Cost differences are important for any healthcare system. From an economic perspective, these results are important because they suggest that, in ERA, DMARD treatment allows complete control of the disease, avoiding the use of biological drugs in a significant proportion of patients.
Only 20% of patients with RA who were in ‘US remission’ flared during the 12-month follow-up period compared with 47.1% of patients who had a PD signal at US evaluation. US-PD remission therefore indicates more stable disease once clinical remission is persistent.
In conclusion, despite the fact that this study was not performed with a high-quality machine, we suggest a combined approach of clinical and US evaluation should be used to define true remission in patients with RA. A prospective study comparing the clinical approach with the combined approach to define when to stop treatment may give further insight into the real value of US in the definition of RA remission in clinical practice.
References
Footnotes
-
Funding GFF has received consulting fees and speaking fees from Wyeth, Roche, Abbott and BMS (less than $10 000 each).
-
Competing interests None.
-
Ethics approval This study was conducted with the approval of the Catholic University ethics committee and informed consent was obtained from all patients before study entry.
-
Provenance and peer review Not commissioned; externally peer reviewed.