Article Text
Abstract
Objective The aim of this study was to assess the predictive value of tenderness in the absence of swelling with consideration of other potential risk factors for subsequent radiographic progression in rheumatoid arthritis (RA) and psoriatic arthritis (PsA).
Methods Clinical and sonographic (grey scale and power Doppler (PD)) examination of 22 joints of the hand were performed in patients with RA and PsA. The impact of tenderness on progression after 2 years was analysed in non-swollen joints for RA and PsA separately with multilevel mixed logistic regression analysis.
Results We included 1207 joints in 55 patients with RA and 352 joints in 18 patients with PsA. In RA, tenderness was associated with radiographic progression after 2 years (model 2: OR 1.85 (95% CI 1.01 to 3.27), p=0.047), although the association of PD (OR 2.92 (95% CI 1.71 to 5.00), p<0.001) and erosions (OR 4.74 (95% CI 2.44 to 9.23), p<0.001) with subsequent structural damage was stronger. In PsA, we found a positive but not significant association between tenderness and radiographic progression (OR 1.72 (95% CI 0.71 to 4.17), p=0.23). In contrast, similarly to RA, erosions (OR 4.62 (95% CI 1.29 to 16.54), p=0.019) and PD (OR 3.30 (95% CI 1.13 to 9.53), p=0.029) had a marked effect on subsequent structural damage.
Conclusion Our findings imply that tenderness in non-swollen joints in RA is associated with subsequent damage. In both diseases, additional risk factors, such as sonographic signs for synovitis and baseline radiographic damage are associated with radiographic progression.
- Arthritis, Psoriatic
- Arthritis, Rheumatoid
- Ultrasonography
Data availability statement
Data are available upon reasonable request. De-identified, coded data will be made available from the corresponding author upon reasonable request.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
In contrast to joint swelling, the association between tenderness without swelling and inflammation is weaker to not existent.
The association of tenderness with radiographic progression is not clear.
WHAT THIS STUDY ADDS
Tenderness is associated with radiographic progression after 2 years in rheumatoid arthritis while a positive trend was observed in psoriatic arthritis.
Positive power Doppler signals and baseline radiographic damage are risk factors in tender non-swollen joints for subsequent progression.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Additional risk factors could be assessed and considered for treatment decisions in patients with many tender but few swollen joints.
Introduction
Rheumatoid arthritis (RA) and psoriatic arthritis (PsA) are chronic systemic autoimmune diseases with synovitis as their primary hallmark. They present with joint swelling and tenderness as well as systemic features like fatigue and elevated acute phase reactants.1 2 Without adequate treatment, synovitis leads to structural damage characterised by bone erosions and cartilage loss.3 4 The treatment goal is to reach remission and thus prevent development or progression of structural damage.
Joint swelling is widely regarded as a clinical sign of synovitis and is associated with radiographic progression.5 6 Many studies have confirmed the association of joint swelling and signs of synovitis on various imaging modalities both in RA7–9 and in PsA.10 11 In contrast to swollen joints, a number of recent studies reported weaker or even a lack of association of tenderness with imaging signs of inflammation in RA8 9 12 and PsA.13 A recently published study has shown that tenderness was not associated with sonographic signs of synovitis in long-standing RA and PsA and that in both, tenderness was better explained by existing radiographic damage and malalignment.13
The development of clinical PsA and RA is commonly preceded by a phase of non-specific musculoskeletal symptoms including tenderness.14 15 In contrast to patients with long-standing disease, there is a significant association between sonographic signs of synovitis and tenderness in early RA and early PsA.13 This is consistent with the classification criteria for RA, which include consideration of tender joints, in the presence of at least one swollen joint.16 Also the classification criteria for PsA do not specify that tenderness should not be considered when determining the presence of ‘inflammatory’ arthritis.17
Some studies reported poor association between tenderness and radiographic progression in RA on the patient level.18–20 Data on the association of tenderness and progression of structural damage on joint level is limited. One study that evaluated patients with early RA reported the association of tenderness with progression of joint damage on the joint level.5 In another study, tenderness was associated with radiographic progression only with coexisting swelling or sonographic signs for synovitis.18 Thus, the literature regarding the relationship between joint tenderness and progression of damage, especially on joint level remains conflicting and therefore, it is unclear whether tenderness may be considered a sign of inflammation in inflammatory arthritides.
To elaborate the question further, we wished to assess whether tenderness is associated with progression in non-swollen joints after 2 years of follow-up on the level of individual joints in both RA and PsA. As a secondary aim, we wanted to assess predictive factors associated with radiographic progression in the presence of tenderness in joints that are not clinically swollen at baseline.
Methods
Patients
Consecutive patients with RA classified according to the American College of Rheumatology/European League Against Rheumatism (ACR/EULAR) criteria16 and those with PsA classified according to the ClASsification for Psoriatic ARthritis (CASPAR) criteria17 were recruited from the outpatient clinic if they had at least one tender joint. This recruitment involved ‘all-comers’ independent of disease duration and current or prior treatment. All patients were included in an observational database where standard variables including joint counts are routinely assessed every 3–6 months and prospectively documented. Details of the database have been published elsewhere.21 22 Therapy with conventional synthetic/biological/targeted synthetic disease-modifying anti-rheumatic drugs (cs/b/tsDMARDs) was assessed at baseline and for the time period until the follow-up X-ray.
Clinical and laboratory examination
In all patients, laboratory examination of rheumatoid factor and anti-cyclic citrullinated peptide (CCP) antibodies was performed at baseline. Clinical examination (28 joint count for RA, 66 swollen joint count (SJC) and 68 joint count for PsA, as well as pain, evaluator global assessment and patient global assessment using visual analogue scales) were performed by health professionals blinded to sonographic data. Additionally, the Clinical Disease Activity Index (CDAI)23 and the Clinical Disease Activity index for Psoriatic Arthritis (CDAPSA)24 were calculated in patients with RA and PsA, respectively. Joints which had undergone replacement or other types of surgery were excluded from both the clinical and imaging analyses. Standard radiographic assessment of hands and feet were acquired at baseline. Patients were followed up by regular outpatient visits and visits in case of disease flares, including tender joint counts (TJC) and SJCs every 3–6 months up to the follow-up X-ray (see under Structural Damage below).
Ultrasound examination
On the day of the above clinical examination, which was the day of inclusion into the study, patients underwent an ultrasound examination of 22 joints of both hands: bilateral wrist, metacarpophalangeal (MCP) 1–5 joints and proximal interphalangeal (PIP) 1–5 joints. The examinations were performed on one of two high-end ultrasound units: Esaote MyLab Twice or GE Logiq E9 with high-frequency linear transducers (6–18 MHz and 10–15 MHz, respectively) by two rheumatologist experts in sonography (IG and PM), with >3 years of experience in musculoskeletal ultrasound who were blinded to clinical and radiographic data. Longitudinal and transverse scans were performed on the dorsal aspect of each joint using both B-mode (grey scale (GS)) and power Doppler (PD) flow according to the recent EULAR standardised procedures.25 Hands were examined palms down resting on the examination table, with the wrist and MCP joints in neutral position and the PIP joints in slight flexion. Copious amounts of gel were used and special care taken not to compress potential Doppler signal. Doppler gain was adjusted to the level just below random noise; pulse repetition frequency was set between 0.5 and 0.8 MHz. GS and PD were both scored semi-quantitatively (0–3) and as a combined score (Global EULAR-Outcome Measures in Rheumatology (OMERACT) Synovitis Score) according to the OMERACT definitions for sonographic pathology.26 27 In case of conflicting grades on the transverse and longitudinal scans, the higher grade was selected. Interobserver reliability was assessed as described previously16; briefly the two US experts reviewed and assessed recorded sonographic images of 15 patients each with RA and PsA, respectively. The agreement was found to be excellent: 0.75 (95% CI 0.71 to 0.78) and 0.9 (95% CI 0.88 to 0.91) for GS and PD, respectively.13
Structural damage
Radiographs of the hands were performed at the time of inclusion (±1 year for ethical reasons) and a follow-up radiograph was performed 2 (±1) years after the first radiograph was obtained. MCP and PIP joints as well as wrists were scored on X-rays for erosions and joint space narrowing (JSN) according to the Sharp/van der Heijde method28 by an experienced reader (GS). Radiographic progression was defined as an increase of the erosion or JSN score of ≥128 for each joint. In order to assess the probability of any joint damage, a joint was regarded as damaged in the presence of either erosions (scored 1 or higher) or JSN (scored 1 or higher).
Statistical analysis
Patient characteristics are described as mean with standard deviation (SD) or percentages with frequencies. For further calculations, only the 22 joints of the hand (wrists, PIPs and MCPs) were included. Clinically swollen joints and joints with prior surgery were excluded from the study.
The association of tenderness at baseline with radiographic progression (yes/no) after 2 years of follow-up was modelled using multilevel mixed logistic regression. Patient-level was used as a random effect in all three models to consider the fact that joints belonging to the same patient can not be considered to be independent. Baseline tenderness as well as the covariates below were used as fixed effects. Three models were fitted. Model 1 (crude model) was adjusted for sex and age only, model 2 (main model) was further adjusted for disease duration, baseline PD and baseline erosions and model 3 (extended model) was additionally adjusted for baseline JSN, baseline SJC (28 for RA and 66 for PsA) as well as seropositivity (for RA).
We assessed the robustness of our estimates in further sensitivity analyses. First, in order to consider the swelling status of the joint between visits, we performed a sensitivity analysis in which we repeated the mixed-effect logistic regression including covariates from model 2, this time excluding joints, which became swollen between the two X-rays. To further examine whether the probability to develop swelling during follow-up (follow-up swelling) is higher in non-swollen joints which were tender at baseline as compared with those which were not, we performed a mixed-effect logistic regression including model 2 covariates and follow-up swelling as dependent variable.
Second, to evaluate potential systematic differences arising from the fact that radiographic progression in joints belonging to the same joint group or to the same hand is likely less independent as compared with progression in random joints, models (model 2) were fitted for the right and left hand, as well as for wrists, and only finger joints (MCPs and PIPs) separately in random effect meta-analyses.
P values of <0.05 were regarded as significant.
Results
Patient characteristics
After exclusion of 3 joints due to prior surgery and 144 joints due to swelling at baseline, we included 1063 joints in 55 patients with RA for further analyses. In PsA, 352 joints in 18 patients were included after exclusion of 44 swollen joints.
Baseline characteristics of included patients and joints are stated in detail in table 1.
Association of tenderness with radiographic progression in RA and PsA
In RA, radiographic progression was observed in 105/901 (11.7%) non-tender non-swollen joints compared with 28/162 (17.3%) in tender non-swollen joints. Baseline tenderness was significantly associated with radiographic progression after 2 years in non-swollen joints. The OR was 1.79 (95% CI 1.04 to 3.06, p=0.035) in the sex-adjusted and age-adjusted model (model 1, crude model) and 1.85 (95% CI 1.01 to 3.27, p=0.047) after adjustment for baseline erosions, PD and disease duration (model 2, main model) showing that the odds for progressing are 1.85 times greater for a joint of a patient with RA that has been tender 2 years prior compared with a joint that was not tender. After further adjustment for baseline JSN, SJC and seropositivity, (model 3, extended model) the OR was 1.72 (95% CI 0.94 to 3.20, p=0.08) (table 2, figure 1). In PsA, the association between baseline tenderness and radiographic progression was also consistently positive in all models although not significant (eg, model 2 OR 1.72 (95%CI 0.71 to 4.17), p=0.23,) (table 2, figure 1).
Association of covariates with radiographic progression in RA and PsA
In RA, non-swollen joints with positive PD signal showed radiographic progression in 32/110 (29.1%) joints compared with 94/836 (11.2%) joints without detected PD signal. Progression was seen in 23/60 (38.3%) non-swollen joints with erosions compared with 110/1003 (11.0%) non-swollen joints without erosions at baseline.
In the mixed-model logistic regression, joints with a positive PD at baseline (OR 2.92 (95% CI 1.71 to 5.00), p<0.001) and those with erosions at baseline (OR 4.74 (95% CI 2.44 to 9.23), p<0.001) were more likely prone to radiographic progression in RA (table 3, figure 2). In the extended model (model 3), also baseline JSN (OR 1.73 (95% CI 1.07 to 2.81), p=0.027) showed a significant association with radiographic progression (online supplemental table 1).
Supplemental material
In PsA, 12/29 (41.4%) joints with PD signal at baseline showed structural progression compared with 36/236 (15.3%) PD negative joints. Among joints with erosions at baseline, progression was seen in 19/26 (73.1%) joints compared with 33/326 (10.1%) joints without erosions at baseline.
Similar to RA, baseline erosions (OR 4.62 (95% CI 1.29 to 16.54), p=0.019) and PD (OR 3.30 (95% CI 1.13 to 9.53), p=0.029) were significantly associated with subsequent structural damage (table 3, figure 2) in PsA. In the calculation with model 3 covariates, only baseline erosions (OR 4.27 (95% CI 1.15 to 15.83), p=0.03) showed a significant association with radiographic progression (online supplemental table 1).
Swelling over time, tenderness and radiographic progression in RA and PsA
Because it is not unreasonable that a tender joint may develop swelling over time, in our first sensitivity analysis we wanted to take into consideration the time period between the X-rays with regard to swelling status. Follow-up swelling was more frequent in joints which were tender at baseline compared with those which were not, both in RA (44/158, 28.2% vs 112/875, 12.8%) and in PsA (13/80, 16.3% vs 25/272, 9.2%). We found a significant association of baseline tenderness with follow-up swelling in RA (model 2: tenderness OR 4.46 (95% CI 2.45 to 8.13), p<0.001) but only a trend in PsA (model 2: tenderness OR 1.81 (95% CI 0.72 to 4.57), p=0.21).
When performing the main model excluding joints with detected swelling between the X-rays, we did not find a significant association between tenderness and swelling, neither in RA (OR 1.16 (95% CI 0.52 to 2.60), p=0.71) nor in PsA (OR 1.11 (95% CI 0.42 to 2.94), p=0.83) (online supplemental table 2).
Joint type and radiographic progression in RA and PsA
In our second sensitivity analysis we considered potential homogeneity in radiographic progression of joints belonging to the same joint group or hand as detailed in Methods. The random-effect meta-analysis did not show significant heterogeneity between joints of the wrist versus finger joints, nor between those of the right versus left hand, respectively, in RA or PSA (p>0.05 I² statistic), (online supplemental figures 1, 2).
Discussion
This study revealed a significant role of baseline tenderness in subsequent radiographic damage after 2 years of follow-up in RA. In PsA this association did not reach significance. While previous studies have shown that joint swelling impacts radiographic progression in RA,5 29 studies assessing the association of joint tenderness T and radiographic progression on patient level reported conflicting results.19 20 Only few studies investigated the impact of tenderness on structural progression on the joint level in RA.5 18 30 Cheung et al found tenderness to be associated with radiographic progression only in combination with clinical swelling or positive PD signal.18 Indeed, although tenderness was associated with radiographic progression in joints without swelling at baseline, this association was no longer significant after exclusion of joints which became swollen over time. Furthermore, baseline tenderness was significantly associated with swelling over time in RA, while we found a positive but not significant association in PsA. This might suggest that the association between baseline tenderness and progression after 2 years of follow-up is likely driven by subsequent swelling and not by tenderness alone. A mediation analysis using a larger prospective cohort would be necessary to explore the mediating/driving role of swelling on the association between tenderness and radiographic progression on joint level.
In the current study, positive PD signal was found to be a risk factor for structural progression in tender non-swollen joints. This finding is in line with that of previous studies in RA, which have shown that sonographic signs of synovitis, especially PD are predictive of radiographic progression.31–33 Baseline erosions were also reported to be predictive,34–36 and this was also confirmed in our study. Indeed, in the multilevel mixed logistic regression analysis, the association with subsequent structural damage was even higher for PD and erosion as compared with that for tenderness.
In a recently published study,13 we showed that tenderness is associated with signs of inflammation on imaging in early but not in long-standing RA. Tenderness was also predictive of radiographic progression in RA in the current study, although disease duration was not significantly associated with progression. Interestingly, in contrast to published literature,6 higher SJC were not predictive for structural progression in individual tender, non-swollen joints in our study. However, published studies assessed the patient-level, rather than joint-level association between SJC and radiographic progression. While patient-level variables such as SJC may have an effect on patient level, they should be considered carefully when interpreting the distribution of damage on the level of individual joints. Furthermore, the extended model included several covariates known or thought to be associated with one another (eg, JSN and erosion), which likely led to overadjustment.
Similar to RA, correlation between tenderness and radiographic progression was positive in all models in PsA, however none of these associations were significant. Given the fact that periarticular inflammation such as enthesitis is a hallmark of the disease, tenderness in PsA may indeed be explained by factors which are not known to play a major role in RA. Such periarticular inflammation may not necessarily have an impact on the development or progression of articular erosions or JSN. Furthermore, the PsA group was smaller than the RA group which might have influenced our results. However, PD as well as baseline erosions were significantly associated with subsequent radiographic progression also in PsA. Data on risk factors for radiographic progression in PsA are scarce. In one study, SJC but not TJC was predictive for subsequent damage.37 Another study found a low association between TJC and radiographic progression, the latter showing a higher association with SJC.38 Similar to RA, studies which investigated predictors for radiological damage in PsA on patient level report conflicting results.38–40 To our knowledge, there is no study which assessed structural damage as a risk factor for subsequent radiographic progression on joint level in PsA.
In general, care should be taken when considering the effect of patient-level variables (eg, disease duration, gender) which do not vary at the joint level and thus cannot formally explain the variability of an outcome measured at the joint level. One important aspect to consider when investigating joint-level association is the fact that joints belonging to the same patient, hand or joint group cannot be regarded as independent with respect to structural outcomes. We tried to overcome this by including the patient as a random effect. Furthermore, although our patient number was limited, in order to address potential systematic differences between location of joints, we performed a meta-analysis with different subgroups of joints, which revealed no significant heterogeneity with regard to radiographic progression.
Our study is not without limitations. The relatively small number of patients in the PsA group may have limited our power to detect a significant association between tenderness and progression. These findings will need to be re-confirmed in future studies with a larger number of patients. Although we assessed a large number of joints, the number in some subgroups was small even in RA, for example, joints with detected erosions at baseline. Furthermore, we assessed the predictive value of the clinical examination at only a single time point. We tried to overcome this limitation by adding a sensitivity analysis excluding joints with detected swelling in the time period between the X-rays. However, in everyday clinical practice, treatment decisions are often based on a single visit, which is why we believe that assessing the predictive value of factors at one time point is clinically relevant.
In conclusion, our study suggests that tenderness in non-swollen joints is associated with subsequent damage in RA, while this association was not significant in PsA. In both diseases, baseline damage and positive PD signal showed a stronger association with radiographic progression when compared with that of tenderness. Furthermore, after exclusion of joints which became swollen during follow-up, the association with tenderness was no longer significant. These results suggest that in the absence of additional factors, such as previous damage, and in particular inflammation, as indicated by positive PD signal or the development of subsequent clinical swelling, tenderness in a non-swollen joint may not be a risk factor for radiographic progression. Recently, two randomised controlled studies concluded that the systematic use of ultrasound to reach sonographic remission is not justified41 42 and clinical composite scores should be prioritised. Regular systemic sonographic examination for all patients with RA and PsA is not feasible. However, in patients with high disease activity scores with many tender joints but lacking or with few swollen joints, treatment decisions should be made carefully. In this setting, modern, sensitive imaging techniques such as musculoskeletal ultrasound might be helpful.
Data availability statement
Data are available upon reasonable request. De-identified, coded data will be made available from the corresponding author upon reasonable request.
Ethics statements
Patient consent for publication
Ethics approval
The study was approved by the Ethics Committee of the Medical University of Vienna (no: 1159/2021). Participants gave informed consent to participate in the study before taking part.
Acknowledgments
The work of IG during the study period was supported partly by a grant from the Austrian Science Fund (KLI514B30). Data of this study was presented at the ACR convergence 2022: Gessl I et al., Tenderness and radiographic progression in rheumatoid arthritis and psoriatic arthritis [abstract]. Arthritis Rheumatol. 2022.
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
Handling editor Kimme L Hyrich
Twitter @Stiddyo
Contributors IG: Substantial contributions to the conception or design of the work, or the acquisition, analysis or interpretation of data. Drafting the work or revising it critically for important intellectual content. CH: Substantial contributions to the analysis or interpretation of data. Drafting the work or revising it critically for important intellectual content. TD: Substantial contributions to the conception or design of the work, or the acquisition, analysis or interpretation of data. Drafting the work or revising it critically for important intellectual content. MD: Substantial contributions to the conception or design of the work, or the acquisition, analysis or interpretation of data. MH: Drafting the work or revising it critically for important intellectual content. VK: Substantial contributions to the conception or design of the work, or the acquisition, analysis or interpretation of data. MP: Substantial contributions to the conception or design of the work, or the acquisition, analysis or interpretation of data. PS: Drafting the work or revising it critically for important intellectual content. GS: Substantial contributions to the conception or design of the work, or the acquisition, analysis or interpretation of data. MZ: Substantial contributions to the conception or design of the work, or the acquisition, analysis or interpretation of data. JSS: Drafting the work or revising it critically for important intellectual content. DA: Drafting the work or revising it critically for important intellectual content. PM: Drafting the work or revising it critically for important intellectual content and final approval of the version published. PM is guarantor for the study.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests JSS is Editorial Board Member (Editor-in-Chief), the other authors have no competing interests to declare.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
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.