Article Text
PDF
Abstract
Background Conventional radiographs of hands and feet are used to depict structural damage in rheumatoid arthritis (RA). This is also commonly done in clinical practice in symptomatic patients at risk for RA (clinically suspect arthralgia (CSA)), but its rationale is unclear. We aimed to investigate the prevalence of radiographic erosive disease in patients with CSA and its progression over time.
Methods Patients with symptomatic arthralgia of the Leiden CSA cohort were studied during 2-year follow-up or until development of inflammatory arthritis (IA). Erosive disease was defined according to the radiologist, or according to the RA-specific erosive definition in light of the American College of Rheumatology (ACR)/European Alliance of Associations for Rheumatology (EULAR) 2010 RA criteria. Serial radiographs were evaluated according to the Sharp van der Heijde Scoring method (SHS) and radiographic erosive progression was determined. Additionally, it was evaluated if baseline erosive disease associated with IA development. Analyses were stratified for anticitrullinated protein antibody status.
Results 1497 radiographs of hands and feet of 749 patients with CSA were studied. Median SHS-erosion score at baseline was 0 (IQR 0–1). RA-specific erosive disease was present in 1.7% according to the radiologist, and 2.5% according to the ACR/EULAR criteria. No patients with CSA progressed ≥5 SHS-erosion points during follow-up. Erosive disease at CSA onset was not associated with IA development (HR 0.98 (95% CI 0.40 to 2.44)).
Conclusions At CSA onset, radiographic erosive disease is rare. In addition, it is rarely progressive within the CSA phase and not predictive for IA development. Therefore, for clinical practice, routinely made radiographs of hands and feet (such as regularly done at RA diagnosis) can be omitted in the at-risk stage of arthralgia.
- Arthritis, Rheumatoid
- Prevalence
- Outcome Assessment, Health Care
Data availability statement
Data are available upon reasonable request. Data can be obtained from the corresponding author upon reasonable request.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Traditionally, conventional radiographs are used to detect structural joint damage and its progression in patients with rheumatoid arthritis (RA).
WHAT THIS STUDY ADDS
Radiographic erosive disease and erosive progression in hands and feet are rare in the ‘at-risk’ phase of clinically suspect arthralgia, both in patients who do and do not develop inflammatory arthritis.
Erosive disease at onset of clinically suspect arthralgia is not predictive for the development of inflammatory arthritis.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Routinely made radiographs of the hands and feet in patients with clinically suspect arthralgia do not have clinical value for RA-specific erosive disease.
Introduction
Conventional radiography (CR) of hands and feet is used in both research and clinical practice to detect structural joint damage. Radiographs are made at diagnosis, among others to identify patients with a worse prognosis, and also over time, to monitor treatment efficacy. Its use is supported by the European Alliance of Associations for Rheumatology (EULAR) guidelines for rheumatoid arthritis (RA) that state that CR is the first-line imaging technique for identification of joint damage.1 Patients with arthralgia at risk for RA, for example, patients with clinically suspect arthralgia (CSA) or with musculoskeletal symptoms and anticitrullinated protein antibody (ACPA) positivity are increasingly recognised.2 In these patients in clinical practice, radiographs are often also conducted, similarly as is traditionally done in RA. However, the value of radiographs at first presentation and during follow-up in this symptomatic at-risk stage is insufficiently known.
To date, only one study reported on the value of CR conducted in the at-risk phase of arthralgia: in 418 ACPA-positive individuals with musculoskeletal symptoms, radiographic erosions occurred infrequently and this frequency was not higher in persons who developed clinical inflammatory arthritis (IA).3 However, the rate of erosive progression over time in the symptomatic at-risk stage of arthralgia remains unstudied in the total patient group. Thus, longitudinal studies on the value of CR in the CSA population are still absent.
Therefore, more data are needed about the value of CR of hands and feet in the risk stage of arthralgia. If it is true that radiographic damage is rare and not informative in patients with CSA regarding the course of RA development, performing CR would no longer be necessary in daily practice for this clinical implication. That would save money and effort. We therefore aimed to study the clinical value of radiographs in the symptomatic at-risk phase of RA by outlining the prevalence of erosive disease and the progression of erosive disease over time or until development of RA.
Methods
Patients
All patients consecutively included in the Leiden Clinically Suspect Arthralgia cohort (CSA Leiden) between April 2012 and June 2022 were studied. Patients were included in the case of recent onset arthralgia (<1 year) of the small joints (metacarpophalangeal (MCP), proximal interphalangeal (PIP), wrist or metatarsophalangeal (MTP) joints) that were considered clinically suspect for development of RA according to the expert opinion of the rheumatologist. Exclusion criteria were clinical arthritis at physical examination or another more likely explanation for the symptoms than evolving RA (eg, osteoarthritis or fibromyalgia), according to the treating rheumatologist. Patients were included regardless of ACPA status, which was determined after inclusion and considered positive as anti-CCP2 ≥7 U/mL (EliA CCP, Phadia, the Netherlands). The CSA Leiden cohort is described previously in more detail and in online supplemental file 1.4 Patients did not use disease-modifying antirheumatic drugs (DMARDs) during follow-up. A flow chart is provided in online supplemental file 2. All patients provided written informed consent.
Supplemental material
Radiographs and definition of erosive disease at baseline
First, baseline dorsoplantar radiographs of hands and feet were scored for radiographic damage via the Sharp van der Heijde Scoring method (SHS) by one trained reader (QAD) blinded for clinical data, with excellent intrareader reliability (intra-class correlation coefficient (ICC)=0.98).5–7 This ICC was calculated by rescoring the radiographs of 46 random patients with CSA 1 month after the initial scoring was completed. The ICC of 0.98 was thereafter calculated from both scores, indicating consistent scoring on different occasions. The SHS method determines radiographic damage based on presence of erosions and joint space narrowing (JSN) and assesses the PIP, MCP, MTP joints and the bones of the wrist; this method is explained in more detail in online supplemental file 3. Radiographs at baseline were also evaluated as part of regular care by a radiologist. Erosive disease at baseline was defined via two separate definitions. Erosive disease was present if (1) the radiologist described presence of erosive disease and (2) if ≥3 joints were involved according to the SHS, in line with what is considered RA-specific erosive disease in light of the ACR/EULAR 2010 RA criteria.8 9
Radiographs and erosive disease over time
Patients were followed during 2-years follow-up, or until development of IA, which was determined by a trained physician via physical examination at the outpatient clinic. Follow-up dorsoplantar radiographs of hands and feet were made at 12 and 24 months, or on development of IA. Some patients were additionally included in the placebo arm of the TREAT EARLIER trial, described elsewhere; of these patients, additional radiographs were made at 4 months.10 Radiographs were scored chronologically together with the baseline radiographs with known time order to ensure the highest sensitivity to detect change over time and to decrease variability.11 12 The minimal clinically important difference (MCID) of radiographic damage is unknown for patients with CSA. Therefore, a difference of ≥5 SHS points (the MCID for early RA) was followed to define radiographic progression.13
Statistics
First, prevalence of erosive disease was assessed via descriptives at baseline, while applying above-mentioned definitions. Second, progression of erosive disease was outlined via cumulative probability plots. Progression was determined by calculating the delta SHS-erosion score between the first and last radiograph during follow-up. Finally, the predictive value of erosive disease at baseline for the development of IA was analysed via Cox regression analyses adjusted for age. Analyses were stratified for ACPA positivity, and cumulative probability plots were stratified for development of IA. As sensitivity analyses, the prevalence and progression of radiographic damage were analysed while measured via the total SHS score, thus also considering JSN.
Results
Patients
749 patients with CSA with in total 1497 radiographs of hands and feet were studied. The majority of patients were female (78%), mean age was 44 years, tender joint count-68 was 5 and 24% of the patients were autoantibody positive (table 1).
Baseline characteristics of study population
Erosive disease is rare at CSA onset
Baseline erosions measured via SHS were rare; median SHS erosion score was 0 (IQR 0–1). Also, JSN and total SHS scores were low (respectively 1 (0–2), 1 (0–3)). Similarly, few patients had erosive disease scored via two definitions; 1.7% of patients had erosive disease according to the evaluation of the radiologist, whereas 2.5% fulfilled the definition for RA-specific erosive disease in light of the ACR/EULAR 2010 RA criteria (figure 1). After stratification for ACPA status, percentages of patients with erosive disease according to the two definitions were similar in ACPA-positive and ACPA-negative patients (respectively, 2.9% vs 1.6% by the radiologist (p=0.32) and 2.0% vs 2.6% by the ACR/EULAR definition (p=0.68)).
Radiographic erosive disease of hands and feet is rare in patients with CSA. Prevalence of erosive disease. (A) Percentage of patients with erosive disease according to the radiologist. (B) Percentage of patients who fulfil the definition of radiographic RA-specific erosive disease in light of the American College of Rheumatology/European Alliance of Associations for Rheumatology 2010 RA criteria.8 9 CSA, clinically suspect arthralgia; RA, rheumatoid arthritis; SHS, Sharp van der Heijde Score.
Progression of erosive disease is rare in patients with CSA
Change of erosion score is depicted via cumulative probability plots. In the total study population, no patients had erosive progression ≥5 SHS points, regardless of ACPA status or development of IA (figure 2). Also when studying progression of the total SHS score, thus including JSN, progression of ≥5 SHS points was rare in the total population (4.0%), in ACPA-positive and ACPA-negative patients (respectively, 5.2%, 3.8%) and in patients who did and did not develop IA (respectively, 3.7%, 4.1%) (online supplemental file 4).
Erosive progression in patients with CSA is scarce. Cumulative probability plots depicting the change of SHS-erosion score between the baseline radiographs and the last radiographs performed (Y-axis), for the total study population (A), stratified for ACPA status (B) and stratified for development of IA (C). Every symbol represents a single patient. ACPA, anti-citrullinated protein antibody; CSA, clinically suspect arthralgia; IA, inflammatory arthritis; MCID, minimal clinically important difference; SHS, Sharp van der Heijde Score.
Erosive disease in patients with CSA is not predictive for IA development
109 (15%) patients developed IA during 2-year follow-up. Baseline SHS erosion score was not predictive for development of IA (HR 0.87, 95% CI 0.70 to 1.08). Also, having erosive disease at baseline according to either of the two definitions was not predictive for IA development (0.98, 95% CI 0.40 to 2.44). Finally, as sensitivity analyses, when studying total baseline SHS score, there was no effect on the development of IA (1.02, 0.97 to 1.07). Again, ACPA stratification resulted in similar results (online supplemental file 5).
Discussion
In clinical practice, conventional radiographs of hands and feet are performed in patients with CSA, similarly to patients with RA. However, the diagnostic and prognostic clinical value of radiographs in this at-risk patient group is unknown. In our study in symptomatic patients at risk for RA development, radiographic erosive disease and progression of erosions during follow-up were rare. In addition, baseline erosive disease was not predictive for development of IA. Therefore, we suggest that radiographs of hands and feet in patients with CSA can be omitted in clinical practice.
The question remains why the few erosions observed in this population are not related to RA development, despite their known value in patients with RA. Multiple considerations were made in our choice to use two definitions for erosive disease. The SHS method was not designed for the symptomatic at-risk phase or for clinical practice but for RA and research purposes and was therefore possibly unsuitable as measure of our primary outcome as it may not detect very small progression of erosions. The ACR/EULAR definition of RA-specific erosive disease was more specific, which is an advantage when studying pre-RA patients. However, in patients with CSA, the test value of this definition is unknown. Furthermore, the radiologist defining erosive disease forms a clinical assessment that is presumably more in line with clinical practice. On the other hand, this could be slightly more subjective. In summary however, both definitions provided similar results, making our conclusion robust. Also, as expected the ACPA-positive patients had slightly more erosive disease according to the radiologist and higher median total SHS than ACPA-negative patients (2.9% vs 1.6%, 2 vs 1), providing face validity. Finally, the prevalence of erosions in the healthy population is unknown. Also, it is unexplored whether the erosions seen in the CSA phase have a similar pathophysiological base as in RA or are more alike ‘erosion-like’ findings in the healthy population. If we speculate that erosions can be found in 1%–2% of asymptomatic people, this could imply that the erosions found in this study may be aspecific.
In addition to baseline scores, long-term data showed that erosive progression was scarce. Due to lacking previous literature on clinically important progression of radiographic damage, we followed the MCID (5 SHS points) known from patients with RA as cut-off to quantify progression. It is unknown if this MCID can be used in CSA. By scoring radiographs of 30 patients for a second time, the smallest detectable change score of 0.6 SHS-erosion points was calculated.14 So, progression of >0.6 SHS points could be reliably detected above the measurement error and can be assumed to be true progression. 20% of patients showed progression of SHS-erosion score above this strict cut-off, of whom the majority (>70%) increased only 1 SHS-erosion point, which clinical relevance should be disputed. Also, total SHS-erosion score and definitions of erosive disease were not associated with IA development. Thus, similar results of baseline and longitudinal data again underline robustness.
Previous research showed that MRI-detected erosions in patients with CSA are also not predictive for development of RA.15 These MRI-detected erosions were presumably even smaller than radiographic erosions given the higher sensitivity of MRI. So despite differences in sensitivity between both imaging modalities, erosions depicted by both modalities do not seem to be of clinical relevance.
The SHS method was initially developed during an era where much more extensive structural damage was present in patients with RA.16 Fortunately, DMARD therapies and treat-to-target protocols have tremendously improved the prognosis of patients with RA and have decreased the amount of structural damage in patients with early RA.17 18 High SHS scores are thus increasingly rare in RA, and as observed here, very rare in the phase preceding clinically apparent arthritis.
In general, radiographs depict not only erosions but can also be used to visualise, for example, fractures, osteoarthritis and calcium pyrophosphate depositions.19 20 As CSA is the symptomatic at-risk phase of RA and per definition other diseases than imminent RA, such as osteoarthritis, are an unlikely explanation of the symptoms, other findings than erosions were not evaluated in this study.
A possible limitation of this study is that the radiographs were scored via SHS by one reader. Adding a second reader could have improved the accuracy of the scores. However, all results are in line with previous research and the majority of radiographs showed no damage, so a second reader presumably would have scored similarly. Furthermore, the radiographs were also evaluated by a radiologist which enhanced the credibility of our readings.
The longitudinal design is a strength of this study. For the first time, radiographs of hands and feet of multiple subsequent study visits during the symptomatic at-risk phase of RA were studied.
The role of imaging in CSA is mainly reserved for ultrasound and MRI in order to detect subclinical joint inflammation; thus, future research should focus on these modalities.21 22 CR does not contribute in patients with arthralgia at risk of RA, as long as classified RA has not manifested.
In conclusion, radiographs of hands and feet have no clinical diagnostic or prognostic value in symptomatic patients at risk for RA. Currently, general recommendations for imaging of patients with CSA are lacking and clinicians tend to extrapolate RA recommendations to the preclinical phase. When guidelines are composed for symptomatic patients at risk for RA, we suggest that radiographs should be omitted.
Data availability statement
Data are available upon reasonable request. Data can be obtained from the corresponding author upon reasonable request.
Ethics statements
Patient consent for publication
Ethics approval
The research protocol for the Leiden CSA cohort (P11.210) was approved by the medical ethical committee of the Leiden University Medical Center. Participants gave informed consent to participate in the study before taking part.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Contributors All authors designed the study. QAD collected and analysed the data. All authors contributed to the interpretation of the data and the writing of the manuscript. All authors approved the final version of the manuscript. All authors were guarantor.
Funding This work was supported by the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (starting grant, agreement no. 714312), and by the Dutch Arthritis Society. The funder had no role in analysis and interpretation of the data, or writing of the manuscript.
Competing interests AvdH is an editorial board member for RMD open. Otherwise none.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.