Forefoot inflammation in recent-onset ACPA-positive and ACPA-negative RA: clinically similar, but different in underlying inflamed tissues

Abstract

Objectives

Although joint swelling is traditionally interpreted as synovitis, recent imaging studies showed that there is also inflammation of tenosynovium and intermetatarsal bursae in the forefoot. We aimed to increase our understanding of differences and similarities regarding forefoot involvement between ACPA-positive and ACPA-negative rheumatoid arthritis (RA) at diagnosis. Therefore, we (1) compared metatarsophalangeal (MTP) joint counts, walking disabilities and inflamed tissues between ACPA groups and (2) studied associations of joint swelling/tenderness and walking disabilities with underlying inflamed tissues within ACPA groups.

Methods

171 ACPA-positive and 203 ACPA-negative consecutively diagnosed patients with RA had a physical joint examination (swollen joint count-66/tender joint count-68), filled a Health Assessment Questionnaire including the domain walking and underwent MRI of the MTP joints at diagnosis. Synovitis, tenosynovitis, osteitis and intermetatarsal bursitis (IMB) were assessed. Findings in age-matched healthy controls were applied to define abnormalities on MRI.

Results

While ACPA-negative RA patients had more swollen joints (mean SJC 8 vs 6 in ACPA-positives, p=0.003), the number of swollen MTP joints was similar (mean 1 in both groups); walking disabilities were also equally common (49% vs 53%). In contrast, inflamed tissues were all more prevalent in ACPA-positive compared with ACPA-negative RA. Within ACPA-positive RA, IMB was associated independently with MTP-joint swelling (OR 2.6, 95% CI 1.4 to 5.0) and tenderness (OR 3.0, 95% CI 1.8 to 5.0). While in ACPA-negatives, synovitis was associated independently with MTP-joint swelling (OR 2.8, 95% CI 1.4 to 5.8) and tenderness (OR 2.5, 95% CI 1.3 to 4.8). Tenosynovitis contributed most to walking disabilities.

Conclusions

Although the forefoot of ACPA-positives and ACPA-negatives share clinical similarities at diagnosis, there are differences in underlying inflamed tissues. This reinforces that ACPA-positive and ACPA-negative RA are different entities.

Correspondence to Dennis A. Ton; [email protected]

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Synovium is naturally present around tendons and intermetatarsal bursae of the forefoot, which may be inflamed in rheumatoid arthritis (RA) patients resulting in tenosynovitis and intermetatarsal bursitis.

• Although differences in risk factors and severity of the disease course have suggested that ACPA-positive and ACPA-negative RA are different disease entities, both groups are indistinguishable at physical joint examination at diagnosis. We hypothesised that this ‘apparent paradox’ can be explained by differences in inflamed tissues underlying forefoot involvement in ACPA-positive and ACPA-negative RA.

WHAT THIS STUDY ADDS

• This study adds that, despite clinical and functional similarities at diagnosis in ACPA-positive and ACPA-negative RA, ACPA-positive RA has more prevalent and more severe inflamed tissues in the forefoot.

• Within ACPA-positive RA, intermetatarsal bursitis contributes independently to swelling and tenderness at physical joint examination, while synovitis does within ACPA-negative RA.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• This study provided first-time evidence that forefoot-related disease burden in ACPA-positive and ACPA-negative RA is attributed to differences in inflamed tissues.

• This increases our understanding of the pathophysiology, and differences therein, of ACPA-positive and ACPA-negative RA.

Introduction

The metatarsophalangeal (MTP) joints in the forefoot are a preferential location for inflammation related to rheumatoid arthritis (RA). Swelling of these MTP joints is traditionally interpreted as synovitis. However, anatomical and histological studies on the forefoot showed that juxta-articular tissues, that is, the tendon sheaths and intermetatarsal bursae, also have a synovial lining similar to those of synovial joints.^{1 2} In addition, the tendon sheaths and intermetatarsal bursae may be inflamed in RA, resulting in tenosynovitis and intermetatarsal bursitis (IMB) that can be detected by imaging modalities, such as MRI.^{1 3} Both inflamed tissues have a high specificity for RA compared with healthy controls.^{3 4} Although MRI studies in recent years have examined the role of tenosynovitis, many of these studies involved the hands.^5–9 However, the forefeet are at least as important to study since forefeet inflammation may lead to functional impairment.^{10 11} Moreover, the recently discovered juxta-articular synovial tissues tenosynovitis and IMB both occur in the forefoot and their relationship with forefoot-related disease burden has not been extensively studied in RA.

RA can be differentiated into two subgroups: autoantibody-positive and autoantibody-negative RA. Both groups have differences in genetic and environmental risk factors. Differences in treatment response and severity of disease course further substantiate the notion of different disease entities.¹² Nevertheless, RA is a clinical diagnosis, classically defined as a symmetric polyarthritis of small joints that can be accompanied by morning stiffness. Furthermore, studies on the involvement of swollen and tender hand and foot joints of ACPA-positive and ACPA-negative RA have reported similarities at diagnosis.^{13 14} Hence, there is an ‘apparent paradox’ of clinical similarities in disease presentation and a wealth of data suggesting different pathophysiological trajectories in the development of ACPA-positive and ACPA-negative RA.^{12 15–17}

Knowing that IMB and tenosynovitis were recently discovered as inflamed tissues,^{1 3} the question arises as to whether the forefoot is truly similar between ACPA-positive and ACPA-negative RA. We hypothesise that, along with differences in risk factors and long-term outcomes, forefoot involvement does differ between ACPA-positive and ACPA-negative RA at diagnosis. We aimed to (1) compare the involvement of swollen and tender forefoot joints, walking disabilities and inflamed tissues of the forefoot (synovitis, tenosynovitis, IMB, osteitis) at diagnosis between ACPA-positive and ACPA-negative RA and (2) study associations of joint swelling, joint tenderness and walking disabilities with underlying joint inflammation within both ACPA groups. We, therefore, performed a large, cross-sectional study of clinical, physical functioning and MRI parameters in recent-onset ACPA-positive and ACPA-negative RA patients.

Methods

Patients and study design

Consecutively diagnosed patients with RA included in the Leiden Early Arthritis Clinic (EAC) between June 2013 and September 2020, and who underwent MRI at inclusion, were studied. The Leiden EAC is a population-based inception cohort that was started in 1993 where newly consecutively presenting patients with recent-onset arthritis, who were naive to disease-modifying antirheumatic drugs (DMARDs), are studied. Its study design has been extensively described previously.¹⁸ Patients were included in the Leiden EAC if clinical arthritis was determined at physical examination by a rheumatologist and symptom duration was <2 years. Since the start of the cohort, general practitioners were informed of the relevance of early referral, and patients referred with suspicion of early arthritis were seen with priority, generally within 2 weeks. Contrast-enhanced MRI of the forefoot was added to the study protocol in June 2013. RA was defined as a clinical diagnosis of RA plus fulfilment of the 1987 and/or 2010 American College of Rheumatology/European Alliance of Associations for Rheumatology RA classification criteria within the first year of follow-up.^{19 20} An overview of how patients were included in this study is shown in online supplemental figure S1.

Baseline assessments

At the baseline visit, serum samples as part of routine diagnostic care were obtained (including C reactive protein: increased if >10 mg/L, ACPA: positive if >7 IU/mL and IgM rheumatoid factor (RF): positive if >3.5 IE/mL). At inclusion, patients underwent physical joint examination to assess swollen (SJC) and tender joint counts (TJCs) and filled in the Health Assessment Questionnaire Disability Index (HAQ) to assess functional disability. The physical joint examination was performed by trained research nurses and included TJC of 68 joints (TJC-68) and SJCs of 66 joints (SJC-66).²¹ Regular consensus exercises led by a rheumatologist were held to maintain interobserver agreement among the research nurses. Walking disabilities were evaluated using the domain walking of the HAQ. The domain walking consists of questions 4A: ‘Are you able to walk outdoors?’ and 4B: ‘Are you able to walk five steps?’. Possible answers were without difficulty, with some difficulty, with much difficulty and unable to do. The highest component of both questions was used for analysis, as described previously.²² The presence of any walking disabilities was defined as scoring 1 (some difficulty) or higher in this domain.

MRI scanning and evaluation

At the baseline visit, unilateral MRI of the MTP joints of the most painful side or dominant side (in case of symmetrical symptoms) were performed to assess underlying inflamed tissues of the forefoot. Patients were requested not to use any non-steroidal anti-inflammatory drugs for at least 24 hours before they underwent MRI. Gadolinium contrast-enhanced MRIs of the MTP1-5 were performed on average within 6 days after the first presentation (1.5-Tesla extremity MRI, GE Healthcare, Madison, Wisconsin, USA). Axial T1-weighted fast spin-echo with fat suppression and coronal T1-weighted fast spin-echo with fat suppression sequences were acquired after intravenous injection of gadolinium contrast. MRIs were assessed on synovitis, tenosynovitis, osteitis and IMB. Synovitis and osteitis were scored according to the OMERACT Rheumatoid Arthritis MRI Scoring System (RAMRIS).²³ Tenosynovitis was scored according to the definition of Haavardsholm et al.²⁴ IMB was defined as contrast enhancement in the superior intermetatarsal space on two consecutive slices in the axial and coronal plane, with or without rim enhancement, as described previously.^{25 26} Scoring of RAMRIS inflammation (ie, synovitis, tenosynovitis and osteitis) was done by medical doctors who had priorly trained on >400 MRI scans. Interreader intraclass correlation coefficients (ICCs) were >0.95 for RAMRIS inflammation.²⁷ Scoring of IMB was done by a musculoskeletal radiologist with >20 years of experience and a trained medical doctor with >2 years of experience in RAMRIS scoring (inter-reader and intrareader ICCs were 0.90 and 0.85).²⁶ All readers were blinded for clinical data and independently assessed the MRIs. An inflammatory lesion was considered present if there was agreement between both readers at the same location. More details on the MRI protocol and scoring are described in online supplemental data S2. Due to logistic reasons, IMB was assessed in a subset of consecutively included patients with RA, suggesting missingness completely at random. No differences in baseline characteristics were observed between the subset of patients with RA with and without IMB assessment (online supplemental table S3). Since contrast enhancement in the joints, bones and intermetatarsal bursae may also occur in the symptom-free population, IMB and RAMRIS inflammation were considered abnormal if the dorsoplantar diameter (IMB) or the grade of RAMRIS inflammation was present in <5% of age-matched healthy controls at the same location (online supplemental table S4).^{3 28}

Statistical analyses

Patient characteristics at diagnosis including joint counts at MTP level, walking disabilities and inflamed tissues were collected and compared. Normally and non-normally distributed continuous data, and count or categorical data were compared using unpaired t-tests, Mann-Whitney U tests and χ² tests, respectively. Joint-level analyses of underlying inflamed tissues with swelling/tenderness of adjacent MTP joints were performed using multivariable generalised estimating equations. Ordinal logistic regression was performed to assess the association of underlying inflamed tissues with the severity of walking disabilities. Corresponding ORs were reported where appropriate. ACPA stratification was performed for all analyses. All inflamed tissues were added as independent variables in the multivariable models since they often co-occur. However, osteitis was not added in the models with joint swelling as a dependent variable since it does not directly contribute to swelling due to its alignment by cortical bone. There was little missing data for the dependent variables (2.4% for joint swelling/tenderness at MTP level and 6.9% for walking disabilities). Baseline characteristics did not differ between patients with and without data on walking disabilities, and therefore, we assumed that these data were missing completely at random (online supplemental table S5).

Two sensitivity analyses were performed to assess the robustness of our results. Joint-level analyses of swelling with underlying inflamed tissues were repeated excluding MTP1, as the MTP1 joint is a predilection site for other diseases (eg, osteoarthritis and gout). Second, analyses assessing the association of joint swelling with underlying inflamed tissues were repeated with stratification for autoantibody-positivity (defined as ACPA- and/or RF-positivity) versus autoantibody-negativity (ACPA and RF both negative). Two-sided p values <0.05 were considered statistically significant. Statistical analyses were conducted in SPSS V.25 (IBM).

Results

Patient characteristics

Baseline characteristics are presented in table 1. Overall, the mean age was 60 years, 64% were female and the median symptom duration at diagnosis was 11 weeks. Of the 376 patients with RA, 171 (46%) were ACPA-positive; their mean SJC-66 and TJC-68 were 6 and 8, respectively. Among the 203 (54%) ACPA-negative patients, the mean SJC-66 and TJC-68 were 8 and 11, respectively.

Table 1

Characteristics of ACPA-positive and ACPA-negative RA patients

	All RA n=376	ACPA-positive n=171	ACPA-negative n=203	P value ACPA+ vs ACPA−
Age in years, mean±SD	60±14	58±15	62±14	0.002
Female, n (%)	239 (64)	107 (63)	131 (65)	0.70
Symptom duration, weeks, median (IQR)	11 (6–28)	12 (5–32)	9 (6–26)	0.22
IgM RF-positivity (>3.5 IE/mL), n (%)	193 (51)	143 (84)	50 (25)	<0.001
Increased CRP (>10 mg/L), n (%)	182 (48)	79 (46)	103 (51)	0.36
SJC-66, mean±SD	7±6	6±6	8±6	0.003
Swollen MTP joints, mean±SD	1±2	1±2	1±2	0.29
TJC-68, mean±SD	10±8	8±8	11±8	<0.001
Tender MTP joints, mean±SD	2±3	2±3	2±3	0.49
Any walking disability, n (%)	192 (51)	84 (49)	107 (53)	0.52
Severity of walking disability, n (%)				0.80
No difficulty	158 (42)	75 (44)	83 (41)
Some difficulty	134 (36)	59 (35)	75 (37)
Much difficulty	49 (13)	20 (12)	28 (14)
Impossible to walk	9 (2)	5 (3)	4 (2)
>1 inflamed tissue at MTP level, n (%)
Synovitis	170 (45)	94 (55)	75 (37)	<0.001
Tenosynovitis	150 (40)	78 (46)	71 (35)	0.02
Osteitis	110 (29)	75 (44)	34 (17)	<0.001
Intermetatarsal bursitis*	133 (55)	76 (66)	56 (44)	<0.001
Grade >2 inflammation at MTP level, n (%)
Synovitis	49 (13)	33 (20)	15 (8)	<0.001
Tenosynovitis	9 (2)	6 (4)	3 (2)	0.052
Osteitis	70 (19)	50 (30)	19 (10)	<0.001
Total RAMRIS inflammation score at MTP level, median (IQR)	1.5 (0.5–4.5)	4.5 (1.5–8.5)	2 (0.5–5.5)	<0.001

*Assessed in 242 patients with RA (116 ACPA-positive and 126 ACPA-negative patients).

ACPA, anti-citrullinated peptide antibodies; CRP, C reactive protein; MTP, metatarsophalangeal; RA, rheumatoid arthritis; RAMRIS, Rheumatoid Arthritis MRI Scoring System; RF, rheumatoid factor; SJC-66, 66 swollen joint count; TJC-68, 68 tender joint count.

ACPA-positive and ACPA-negative RA share clinical and functional similarities at MTP level, but ACPA-positive RA patients have more inflamed joint tissues

Although ACPA-positive patients had a lower total number of swollen joints (6 vs 8, p=0.003), the number of swollen MTP joints was similar in both groups (mean 1, p=0.29). Similar results were found for the number of tender MTP joints (mean 2 in both groups, p=0.49). As depicted in figure 1A, the distribution of MTP joint swelling and tenderness did not differ between ACPA-positive and ACPA-negative RA. In both groups, MTP 2 and 3 were predilection sites for joint swelling/tenderness. In addition, no differences were found regarding the frequency of walking disabilities in ACPA-positive and ACPA-negative RA (49% vs, 53%, p=0.52, respectively).

Figure 1. Frequencies (%) of swollen and tender joints (A) and inflamed joint tissues detected by MRI (B) of the MTP joints, showing no differences in clinical signs between ACPA-positive and ACPA-negative RA, yet there are more inflamed tissues in ACPA-positive RA. The depicted frequencies of joint counts are the average of both sides. I: extensor tendon of the first MTP joint, II: synovium of the first MTP joint, III: bone marrow of the first MTP joint, IV: flexor tendon of the first MTP joint, V: intermetatarsal bursa of the first intermetatarsal space. Exact percentages are shown in online supplemental table S6 .

When studying the frequency of inflamed tissues at the MTP level, differences were observed. ACPA-positive patients more often had >1 inflamed tissue detected by MRI. Synovitis was present in 55% of ACPA-positive RA, while in ACPA-negative RA, it was present in 37% of the patients (p<0.001). Similarly, tenosynovitis was more often present in ACPA-positive compared with ACPA-negative RA (46% vs 35%, p=0.02). Osteitis occurred more frequently in ACPA-positive than in ACPA-negative RA as well (44% vs 17%, p<0.001). IMB was highly present in both groups, however, it occurred more frequently in ACPA-positive RA (66% vs 44% in ACPA-negative RA, p<0.001). Inflamed joint tissues were not only more common in ACPA-positive RA, but the inflammation was also more severe in this group than in ACPA-negative RA (table 1). When studying the distribution of inflamed joint tissues, IMB (all locations except for the first intermetatarsal space) and osteitis were more often found in ACPA-positive RA (figure 1B).

Thus, regarding the forefoot, ACPA-positive and ACPA-negative RA share clinical and functional similarities when considering the prevalence and distribution of involved MTP joints, however, inflamed tissues in the forefoot are more frequent and severe in ACPA-positive RA.

Joint swelling/tenderness and severity of walking disabilities in relation to underlying inflamed tissues at MRI

Joint-level analyses were performed to assess the association of underlying inflamed tissues with adjacent joint swelling/tenderness (table 2). In univariable analyses of ACPA-positive RA, joint swelling was associated with synovitis (OR 3.6, 95% CI (2.2, 6.0)), as well as tenosynovitis (OR 3.0, 95% CI (1.7, 5.2)) and IMB (OR 3.5, 95% CI (2.0, 6.2)). Likewise, joint tenderness was associated with synovitis, tenosynovitis, osteitis and IMB.

Table 2

Association of presence of MRI-detected inflamed joint tissues with adjacent clinical signs of the MTP joints in ACPA-positive and ACPA-negative RA patients

Joint swelling	Univariable OR (95% CI)	P value	Multivariable OR (95% CI)	P value
ACPA-positive
Synovitis	3.6 (2.2 to 6.0)	<0.001	2.4 (1.2 to 4.6)	0.01
Tenosynovitis	3.0 (1.7 to 5.2)	<0.001	1.7 (0.8 to 3.3)	0.15
IMB	3.5 (2.0 to 6.2)	<0.001	2.6 (1.4 to 5.0)	0.003
ACPA-negative
Synovitis	2.8 (1.5 to 5.0)	<0.001	2.8 (1.4 to 5.8)	0.005
Tenosynovitis	1.6 (0.8 to 3.0)	0.16	0.9 (0.4 to 1.9)	0.83
IMB	1.8 (0.9 to 3.4)	0.08	1.3 (0.7 to 2.5)	0.44

Joint tenderness	Univariable OR (95% CI)	P value	Multivariable OR (95% CI)	P value
ACPA-positive
Synovitis	2.6 (1.5 to 4.5)	0.001	1.4 (0.7 to 2.7)	0.39
Tenosynovitis	2.3 (1.3 to 3.8)	0.002	1.4 (0.8 to 2.5)	0.26
Osteitis	2.5 (1.5 to 4.4)	<0.001	1.7 (0.9 to 3.3)	0.11
IMB	3.4 (2.1 to 5.7)	<0.001	3.0 (1.8 to 5.0)	<0.001
ACPA-negative
Synovitis	2.9 (1.6 to 5.3)	<0.001	2.5 (1.3 to 4.8)	0.007
Tenosynovitis	1.7 (0.8 to 3.2)	0.14	0.9 (0.4 to 1.8)	0.76
Osteitis	2.1 (0.9 to 5.0)	0.09	1.2 (0.4 to 3.1)	0.77
IMB	1.8 (1.0 to 3.4)	0.047	1.4 (0.8 to 2.6)	0.29

ORs were calculated using GEE and represent joint swelling/tenderness as dependent variables. Multivariable joint-level analyses with swelling as dependent variable were performed in 547 MTP joints of ACPA-positive and in 624 MTP joints of ACPA-negative RA patients. Multivariable joint-level analyses with tenderness as dependent variable were performed in 541 MTP joints of ACPA-positive and 617 MTP joints of ACPA-negative RA patients. Osteitis was not added in the models for joint swelling since it cannot contribute directly to swelling due to its alignment by cortical bone.

ACPA, anti-citrullinated peptide antibodies; GEE, generalised estimating equation; IMB, intermetatarsal bursitis; MTP, metatarsophalangeal; RA, rheumatoid arthritis.

In univariable analyses of ACPA-negative RA, joint swelling was associated with synovitis (OR 2.8, 95% CI (1.5, 5.0)). Joint tenderness was associated with synovitis (OR 2.9, 95% CI (1.6, 5.3)) and IMB (OR 1.8, 95% CI(1.0, 3.4).

Since the inflamed tissues frequently occur together, multivariable analyses were performed. In ACPA-positive RA, synovitis and IMB were independently associated with joint swelling (OR for synovitis: 2.4, 95% CI (1.2, 4.6) and OR for IMB: 2.6, 95% CI (1.4, 5.0)). Also, IMB was the only joint tissue that was independently associated with joint tenderness (OR 3.0, 95% CI (1.8, 5.0)). In ACPA-negative RA in contrast, only synovitis contributed to joint swelling (OR 2.8, 95% CI (1.4, 5.8)) and joint tenderness (OR 2.5, 95% CI (1.3, 4.8)).

Subsequently, we studied which underlying inflamed tissues were associated with the severity of walking disabilities. In both subgroups, tenosynovitis was the only inflamed tissue associated with the severity of walking disabilities (OR for ACPA-positive RA: 1.3, 95% CI (1.0, 1.6) and OR for ACPA-negative RA: 1.1, 95% CI (1.0, 1.3)) (table 3).

Table 3

Association of MRI-detected inflamed joint tissues scores with severity of walking disabilities in ACPA-positive and ACPA-negative RA patients

	Univariable OR (95% CI)	P value
ACPA-positive
Synovitis	1.1 (1.0 to 1.3)	0.16
Tenosynovitis	1.3 (1.0 to 1.6)	0.02
Osteitis	1.1 (1.0 to 1.1)	0.13
IMB lesions	1.3 (0.9 to 1.7)	0.15
ACPA-negative
Synovitis	1.0 (0.9 to 1.2)	0.61
Tenosynovitis	1.1 (1.0 to 1.3)	0.054
Osteitis	1.1 (0.9 to 1.3)	0.46
IMB lesions	1.0 (0.8 to 1.3)	0.88

ORs were calculated using ordinal logistic regression and represented the severity of walking disabilities as dependent variable.

ACPA, anti-cyclic peptide antibodies; IMB, intermetatarsal bursitis; RA, rheumatoid arthritis.

Sensitivity analyses

The analyses of associations of underlying inflamed tissues with joint swelling were repeated without the MTP1 joint since it is a predilection site for other diseases (eg, gout and osteoarthritis). Then, results from univariable analyses in ACPA-positive RA were unchanged (online supplemental table S7). In multivariable analysis, IMB was the only inflamed tissue that was associated with joint swelling in ACPA-positive RA (OR 2.4, 95% CI (1.2, 4.8)). In ACPA-negative RA, results from univariable and multivariable analyses were similar, only synovitis was independently associated with joint swelling (OR 2.2, 95% CI (1.0, 4.9)).

In a second sensitivity analysis, patients were stratified for autoantibody-positivity or autoantibody-negativity (ACPA and/or RF positive vs double negative) instead of ACPA-positivity. This revealed similar findings in both patient groups (online supplemental table S8). Altogether, IMB was the inflamed tissue that was most robustly associated with joint inflammation at physical examination in ACPA-positive RA, while this was synovitis in ACPA-negative RA.

Discussion

We aimed to improve the comprehension of similarities and differences in forefoot involvement between ACPA-positive and ACPA-negative RA since well-known data on differences in risk factors and long-term outcomes suggest that both ACPA groups are subsets or subentities of RA, while patients with RA are indistinguishable at diagnosis based on their clinical phenotypes. Moreover, since recent imaging studies described tenosynovitis and IMB in the forefoot,^{1 3} we hypothesised that a thorough investigation on tissue level with imaging would identify differences in forefoot involvement between ACPA-positive and ACPA-negative RA. This large cross-sectional MRI study on consecutively diagnosed ACPA-positive and ACPA-negative RA patients is the first to explain the paradox by showing similarities in forefoot joint swelling, joint tenderness, walking disabilities at diagnosis and differences in underlying inflammation. ACPA-positive RA patients had more inflamed forefoot joint tissues on MRI at diagnosis. Moreover, not synovitis but IMB was associated independently with joint swelling and joint tenderness in ACPA-positive RA.

In the past decades, various risk factors related to genetic variations and environmental factors were identified that contribute to RA development. The strongest associations of genetic variation and environmental factors concerned shared epitope alleles and smoking in relation to development of ACPA-positive RA, while typical risk factors for development of ACPA-negative RA remain to be elucidated.^{29 30} Likewise, although the treatment of RA has substantially improved over time, ACPA-positive RA patients remain to have a more severe disease course, which is, for example, illustrated by a lower frequency of sustained DMARD-free remission.^{12 31} These observations have led to the notion that ACPA-positive and ACPA-negative RA are different disease entities. In spite of these underlying differences, ACPA-positive and ACPA-negative RA patients share similar clinical phenotypes at diagnosis regarding joint swelling/tenderness at the MTP joints and walking disabilities, which is in line with previous findings.^{13 32} However, by using MRI, the most sensitive imaging modality for detecting inflammation,³³ we were able to demonstrate that ACPA groups do differ in terms of underlying inflammation at diagnosis, a finding that could not have been made on the basis of physical joint examination alone. All inflamed tissues (ie, synovitis, tenosynovitis, osteitis and IMB) occurred more frequently in ACPA-positive RA than in ACPA-negative RA patients. Moreover, more ACPA-positive patients had an inflamed tissue that was more severe (grade 2 or higher according to RAMRIS). Our findings, therefore, reinforce the notion that ACPA-positive and ACPA-negative RA are two different disease entities based on differences in underlying inflammation.

The impact of clinical practice on the current findings is related to the physician’s understanding of which inflamed tissues contribute to the clinical phenotype of RA at forefoot level. Our data suggests that, in addition to synovitis, the recently discovered inflamed tissue IMB also contributes to joint swelling and joint tenderness, especially in ACPA-positive RA. The forefeet are notoriously difficult to examine with physical examination and swelling and tenderness may be differently explained in ACPA-positive and ACPA-negative RA. A possible way how IMB contributes to joint swelling and tenderness could be due to dorsal distension of severe IMB which is located nearby to the inflamed intracapsular synovial tissue or nearby the joint surface (figure 2). Our data suggest that IMB can contribute either alone or together with synovitis to perceived swelling or tenderness of MTP joints. Previous data have shown that intra-articular as well as juxta-articular inflamed tissues respond similarly on DMARD-treatment in ACPA-positive and ACPA-negative RA, hence, the observed findings at forefoot level may not affect treatment effectiveness.¹³ Nevertheless, understanding the tissues that relate to abnormalities at physical examination of forefeet in ACPA-positive and ACPA-negative RA patients may be relevant for rheumatologists in clinical practice.

Figure 2. T1-weighted fat-saturated MRI of the forefoot of two ACPA-positive RA patients showing concomitant IMB with other inflamed joint tissues. (A) Axial T1-weighted fat-saturated MRI of a patient in their 50s with swelling of MTP2-3 at physical examination. Intermetatarsal bursitis (IMB) is present at the first, second and third intermetatarsal spaces (arrows). Dorsal distension (dumbbell sign) due to severe IMB is seen in the second intermetatarsal space. Concomitant presence of synovitis (arrowheads) and flexor tenosynovitis (dotted arrows) is seen at MTP3. (B) Axial T1-weighted fat-saturated MRI of a patient in their 20s with swelling of MTP2 and 4 at physical examination. IMB is present at the third intermetatarsal space (arrow) and synovitis is present at MTP2 (arrowheads).

The finding of more severe MRI-detected joint inflammation in ACPA-positive RA compared with ACPA-negative RA was interpreted as relating to differences in RA subgroup and subgroup entities. It was unlikely that environmental risk factors (BMI, smoking) disturbed these findings. BMI was not correlated to the total MRI inflammation scores in both ACPA-positive and ACPA-negative RA (data are not shown). Likewise, current smokers did not have statistically significant differences in total inflammation scores compared with non-smokers (data are not shown).

While in daily clinical practice imaging is primarily used as a diagnostic tool, imaging was here used for research purposes to increase the understanding of inflamed tissues in the forefoot. MRI has the advantage that it is sensitive to detect intra-articular inflammation such as synovitis and osteitis, but also juxta-articular inflammation such as tenosynovitis and IMB. Therefore, MRI has been ideally suited to also be used as a means to investigate intra-articular and juxta-articular tissues in detail, which in combination with clinical findings has allowed us to increase our current understanding of the clinical phenotype of RA.

There were some limitations to this study. First, we assumed that the strongest associations in multivariable analyses equated to those contributing most to joint swelling and tenderness. However, synovitis and tenosynovitis often occur together and the absence of significance in statistical analyses does not preclude that these inflamed tissues do contribute to joint swelling. For instance, tenosynovitis located at the dorsal side may contribute to joint swelling or tenderness, although statistically the strongest associations were found for synovitis and IMB. Second, physical joint examination of the MTP joints is known to be notoriously challenging and poorly reproducible, possibly due to other subcutaneous tissues which may be evaluated as swelling such as fat and peripheral oedema.^{34 35} Therefore, regular consensus exercises were held to maintain interobserver agreement.³⁶ However, ACPA status was unknown at first physical examination and possible difficulties with reproducibility of examination of the forefeet will, therefore, not have influenced the currently observed differences between ACPA-positive and ACPA-negative RA. Another potential limitation is that we studied the HAQ domain walking. However, the HAQ was validated for integrated use of all domains and not for single domains.²² An alternative foot impairment measurement might have been more sensitive to assess walking disabilities, however, these data were not available.^{11 37}

A major strength of our study is that we used MRI to investigate inflamed tissues. MRI is the most sensitive and reproducible imaging modality to detect inflammation.³⁸ Furthermore, our study is one of the largest MRI studies to assess forefoot involvement that also considered all known inflamed tissues of the forefoot (osteitis, synovitis, tenosynovitis and IMB). Finally, since subtle inflammation-like features can also be present on MRI of healthy people, especially at higher age, and therefore, unrelated to RA, we performed an age-matched correction based on findings in healthy symptom-free controls to reduce false-positive findings detected by MRI. This addresses the problem of non-specific MRI findings. However, if there are still signal intensities at MRI related to, for instance, the position or mechanical load on the feet, it is unlikely that these non-RA-related MRI abnormalities differ between ACPA-positive and ACPA-negative RA. The influence on comparisons between ACPA-positive and ACPA-negative RA at diagnosis may, therefore, be limited.

Given the findings of our study, several important questions related to underlying pathophysiology remain to be answered. Whether inflammation of synovial tissues at the MTP joints is different between ACPA-positive and ACPA-negative RA at the cellular level is subject to future studies. This is interesting as studies on synovium of knee, elbow, ankle, wrist, metacarpophalangeal, proximal interphalangeal, and MTP joints have already revealed that ACPA-positivity is associated with increased lymphocyte infiltration.^{39 40} Ideally, the presence of cellular differences is investigated through histopathological studies in intra-articular located synovial tissue as well as in tenosynovium and the synovial lining from the intermetatarsal bursae of the MTP joints. However, due to the small sizes of these tissues in the forefoot, performing biopsies in these areas will be challenging.

In conclusion, this large MRI study on consecutively diagnosed ACPA-positive and ACPA-negative RA patients demonstrated that, despite sharing a similar clinical phenotype, there is a difference regarding underlying inflamed tissues at the forefoot level. While IMB was the major inflamed tissue that contributed to joint swelling and tenderness in ACPA-positive RA, only synovitis contributed to joint swelling and tenderness in ACPA-negative RA. These findings reinforce the notion that ACPA-positive and ACPA-negative RA are different disease entities based on their differences in underlying inflamed tissues.

We thank Gerrit Kracht for preparing the MR images of this manuscript. This work was presented during a poster presentation at the EULAR European Congress of Rheumatology 2024 (POS0621).

Data availability statement

Data are available on reasonable request.

Ethics statements

Patient consent for publication

Not applicable.

Ethics approval

This study involves human participants and was approved by Commisie Medische Ethiek Leiden University Medical Center (B19.008). Participants gave informed consent to participate in the study before taking part.

Footnote

Contributors All authors contributed to the conception and study design. DAT analysed the data. All authors contributed to the interpretation of the data. DAT wrote the first version of the manuscript and AvdH revised it critically. All authors read and approved the final submitted manuscript. DAT and AvdH are the guarantors for this manuscript.

Funding The research leading to these results has received funding from the Dutch Arthritis Foundation and the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (starting grant, agreement No 714312).

Disclaimer The funding source had no role in the design and conduct of the study; collection, management, analysis and interpretation of the data; preparation, review or approval of the manuscript; or decision to submit the manuscript for publication.

Competing interests None declared.

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.

References

¹ Dakkak YJ, Jansen FP, DeRuiter MC, et al. Rheumatoid Arthritis and Tenosynovitis at the Metatarsophalangeal Joints: An Anatomic and MRI Study of the Forefoot Tendon Sheaths. Radiology 2020; 295: 146–54. doi:10.1148/radiol.2020191725

² Theumann NH, Pfirrmann CW, Chung CB, et al. Intermetatarsal spaces: analysis with MR bursography, anatomic correlation, and histopathology in cadavers. Radiology 2001; 221: 478–84. doi:10.1148/radiol.2212010469

³ Dakkak YJ, Niemantsverdriet E, van der Helm-van Mil AHM, et al. Increased frequency of intermetatarsal and submetatarsal bursitis in early rheumatoid arthritis: a large case-controlled MRI study. Arthritis Res Ther 2020; 22: 277. doi:10.1186/s13075-020-02359-w

⁴ Rogier C, Hayer S, van der Helm-van Mil A. Not only synovitis but also tenosynovitis needs to be considered: why it is time to update textbook images of rheumatoid arthritis. Ann Rheum Dis 2020; 79: 546–7. doi:10.1136/annrheumdis-2019-216350

⁵ Filippucci E, Gabba A, Di Geso L, et al. Hand tendon involvement in rheumatoid arthritis: an ultrasound study. Semin Arthritis Rheum 2012; 41: 752–60. doi:10.1016/j.semarthrit.2011.09.006

⁶ Navalho M, Resende C, Rodrigues AM, et al. Bilateral evaluation of the hand and wrist in untreated early inflammatory arthritis: a comparative study of ultrasonography and magnetic resonance imaging. J Rheumatol 2013; 40: 1282–92. doi:10.3899/jrheum.120713

⁷ Nieuwenhuis WP, Krabben A, Stomp W, et al. Evaluation of magnetic resonance imaging-detected tenosynovitis in the hand and wrist in early arthritis. Arthritis Rheumatol 2015; 67: 869–76. doi:10.1002/art.39000

⁸ Rubin DA. MRI and ultrasound of the hands and wrists in rheumatoid arthritis. I. Imaging findings. Skeletal Radiol 2019; 48: 677–95. doi:10.1007/s00256-019-03179-z

⁹ Wakefield RJ, O’Connor PJ, Conaghan PG, et al. Finger tendon disease in untreated early rheumatoid arthritis: a comparison of ultrasound and magnetic resonance imaging. Arthritis Rheum 2007; 57: 1158–64. doi:10.1002/art.23016

¹⁰ Dakkak YJ, Wouters F, Matthijssen XME, et al. Walking Disabilities in Association With Tenosynovitis at the Metatarsophalangeal Joints: A Longitudinal Magnetic Resonance Imaging Study in Early Arthritis. Arthritis Care Res (Hoboken) 2022; 74: 301–7. doi:10.1002/acr.24452

¹¹ Hooper L, Bowen CJ, Gates L, et al. Prognostic indicators of foot‐related disability in patients with rheumatoid arthritis: Results of a prospective three‐year study. Arthritis Care Res (Hoboken) 2012; 64: 1116–24. doi:10.1002/acr.21672

¹² Matthijssen XME, Niemantsverdriet E, Huizinga TWJ, et al. Enhanced treatment strategies and distinct disease outcomes among autoantibody-positive and -negative rheumatoid arthritis patients over 25 years: A longitudinal cohort study in the Netherlands. PLoS Med 2020; 17: e1003296. doi:10.1371/journal.pmed.1003296

¹³ van der Helm-van Mil AHM, Verpoort KN, Breedveld FC, et al. Antibodies to citrullinated proteins and differences in clinical progression of rheumatoid arthritis. Arthritis Res Ther 2005; 7: R949–58. doi:10.1186/ar1767

¹⁴ Weman L, Salo H, Kuusalo L, et al. Initial presentation of early rheumatoid arthritis. PLoS One 2023; 18: e0287707. doi:10.1371/journal.pone.0287707

¹⁵ Floudas A, Canavan M, McGarry T, et al. ACPA Status Correlates with Differential Immune Profile in Patients with Rheumatoid Arthritis. Cells 2021; 10: 647. doi:10.3390/cells10030647

¹⁶ Khidir SJH, Toes REM, van Mulligen E, et al. Is tooth extraction as proxy for periodontal disease related to the development of RA? Lessons from a longitudinal study in the at-risk stage of clinically suspect arthralgia. Ann Rheum Dis 2024; 83: 1227–9. doi:10.1136/ard-2024-225688

¹⁷ Matthijssen XME, Niemantsverdriet E, Le Cessie S, et al. Differing time-orders of inflammation decrease between ACPA subsets in RA patients suggest differences in underlying inflammatory pathways. Rheumatology (Sunnyvale) 2021; 60: 2969–75. doi:10.1093/rheumatology/keaa658

¹⁸ de Rooy DPC, van der Linden MPM, Knevel R, et al. Predicting arthritis outcomes--what can be learned from the Leiden Early Arthritis Clinic? Rheumatology (Oxford) 2011; 50: 93–100. doi:10.1093/rheumatology/keq230

¹⁹ Arnett FC, Edworthy SM, Bloch DA, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis & Rheum 1988; 31: 315–24. doi:10.1002/art.1780310302

²⁰ Aletaha D, Neogi T, Silman AJ, et al. 2010 Rheumatoid arthritis classification criteria: An American College of Rheumatology/European League Against Rheumatism collaborative initiative. Arthritis & Rheum 2010; 62: 2569–81. doi:10.1002/art.27584

²¹ Scott DL, Houssien DA. Joint assessment in rheumatoid arthritis. Br J Rheumatol 1996; 35 Suppl 2: 14–8. doi:10.1093/rheumatology/35.suppl_2.14

²² Bruce B, Fries JF. The Stanford Health Assessment Questionnaire: dimensions and practical applications. Health Qual Life Outcomes 2003; 1: 20. doi:10.1186/1477-7525-1-20

²³ Østergaard M, Peterfy CG, Bird P, et al. The OMERACT Rheumatoid Arthritis Magnetic Resonance Imaging (MRI) Scoring System: Updated Recommendations by the OMERACT MRI in Arthritis Working Group. J Rheumatol 2017; 44: 1706–12. doi:10.3899/jrheum.161433

²⁴ Haavardsholm EA, Ostergaard M, Ejbjerg BJ, et al. Introduction of a novel magnetic resonance imaging tenosynovitis score for rheumatoid arthritis: reliability in a multireader longitudinal study. Ann Rheum Dis 2007; 66: 1216–20. doi:10.1136/ard.2006.068361

²⁵ Dakkak YJ, Boer AC, Boeters DM, et al. The relation between physical joint examination and MRI-depicted inflammation of metatarsophalangeal joints in early arthritis. Arthritis Res Ther 2020; 22: 67. doi:10.1186/s13075-020-02162-7

²⁶ van Dijk BT, Dakkak YJ, Matthijssen XME, et al. Intermetatarsal Bursitis, a Novel Feature of Juxtaarticular Inflammation in Early Rheumatoid Arthritis Related to Clinical Signs: Results of a Longitudinal Magnetic Resonance Imaging Study. Arthritis Care Res (Hoboken) 2022; 74: 1713–22. doi:10.1002/acr.24640

²⁷ Dakkak YJ, Matthijssen XME, van der Heijde D, et al. Reliability of Magnetic Resonance Imaging (MRI) Scoring of the Metatarsophalangeal Joints of the Foot according to the Rheumatoid Arthritis MRI Score. J Rheumatol 2020; 47: 1165–73. doi:10.3899/jrheum.190258

²⁸ Mangnus L, van Steenbergen HW, Reijnierse M, et al. Magnetic Resonance Imaging-Detected Features of Inflammation and Erosions in Symptom-Free Persons From the General Population. Arthritis Rheumatol 2016; 68: 2593–602. doi:10.1002/art.39749

²⁹ Deane KD, Demoruelle MK, Kelmenson LB, et al. Genetic and environmental risk factors for rheumatoid arthritis. Best Pract Res Clin Rheumatol 2017; 31: 3–18. doi:10.1016/j.berh.2017.08.003

³⁰ Padyukov L, Seielstad M, Ong RTH, et al. A genome-wide association study suggests contrasting associations in ACPA-positive versus ACPA-negative rheumatoid arthritis. Ann Rheum Dis 2011; 70: 259–65. doi:10.1136/ard.2009.126821

³¹ Seegobin SD, Ma MHY, Dahanayake C, et al. ACPA-positive and ACPA-negative rheumatoid arthritis differ in their requirements for combination DMARDs and corticosteroids: secondary analysis of a randomized controlled trial. Arthritis Res Ther 2014; 16: R13. doi:10.1186/ar4439

³² Cader MZ, Filer AD, Buckley CD, et al. The relationship between the presence of anti-cyclic citrullinated peptide antibodies and clinical phenotype in very early rheumatoid arthritis. BMC Musculoskelet Disord 2010; 11: 187. doi:10.1186/1471-2474-11-187

³³ Woodworth TG, Morgacheva O, Pimienta OL, et al. Examining the validity of the rheumatoid arthritis magnetic resonance imaging score according to the OMERACT filter-a systematic literature review. Rheumatology (Oxford) 2017; 56: 1177–88. doi:10.1093/rheumatology/kew445

³⁴ Salaffi F, Filippucci E, Carotti M, et al. Inter-observer agreement of standard joint counts in early rheumatoid arthritis: a comparison with grey scale ultrasonography--a preliminary study. Rheumatology (Oxford) 2008; 47: 54–8. doi:10.1093/rheumatology/kem286

³⁵ Sokka T, Pincus T. Joint counts to assess rheumatoid arthritis for clinical research and usual clinical care: advantages and limitations. Rheum Dis Clin North Am 2009; 35: 713–22. doi:10.1016/j.rdc.2009.10.004

³⁶ Cheung PP, Dougados M, Andre V, et al. Improving agreement in assessment of synovitis in rheumatoid arthritis. Joint Bone Spine 2013; 80: 155–9. doi:10.1016/j.jbspin.2012.07.014

³⁷ Helliwell P, Reay N, Gilworth G, et al. Development of a foot impact scale for rheumatoid arthritis. Arthritis Rheum 2005; 53: 418–22. doi:10.1002/art.21176

³⁸ Colebatch AN, Edwards CJ, Østergaard M, et al. EULAR recommendations for the use of imaging of the joints in the clinical management of rheumatoid arthritis. Ann Rheum Dis 2013; 72: 804–14. doi:10.1136/annrheumdis-2012-203158

³⁹ Humby F, Lewis M, Ramamoorthi N, et al. Synovial cellular and molecular signatures stratify clinical response to csDMARD therapy and predict radiographic progression in early rheumatoid arthritis patients. Ann Rheum Dis 2019; 78: 761–72. doi:10.1136/annrheumdis-2018-214539

⁴⁰ van Oosterhout M, Bajema I, Levarht EWN, et al. Differences in synovial tissue infiltrates between anti–cyclic citrullinated peptide–positive rheumatoid arthritis and anti–cyclic citrullinated peptide–negative rheumatoid arthritis. Arthritis & Rheum 2008; 58: 53–60. doi:10.1002/art.23148