DOI:
Original papers
Ankle involvement in rheumatoid arthritis – a comparison of inflammatory signs on musculoskeletal ultrasound and magnetic resonance imaging
Luminița Enache
1, Claudiu C. Popescu
1,2, Mihaela C. Micu
3, Adriana Cojocaru
4, Victoria- Cristina Șuța
5, Maria Șuța
5, Cătălin Codreanu
1,21“Dr. Ion Stoia” Clinical Centre of Rheumatic Diseases, Bucharest, 2“Carol Davila” University of Medicine and Pharmacy, Bucharest, 3Rheumatology Division, 2nd Rehabilitation Department, Rehabilitation Clinical Hospital, Cluj Napoca, 4Magnetic Resonance Imaging Department, Affidea Romania SRL, Cluj-Napoca, 5“Ovidius” University of Constanta, Faculty of Medicine, Constanta, Romania
Received 02.05.2019 Accepted 10.06.2019 Med Ultrason
2019, Vol. 21, No 3, 265-272
Corresponding author: Victoria-Cristina Șuța MD, PhD
“Ovidius” University of Constanța, Faculty of Medicine, 124th Mamaia Boulevard, Constanta, Romania
E-mail: [email protected]
Introduction
Rheumatoid arthritis (RA) is a chronic inflammatory disease that affects synovial joints, causing irreversible joint deformities which lead to important disability. Clin-
ical examination for synovitis detection is an important tool in the diagnosis and evaluation of RA, but it has low accuracy and reproducibility [1]. Radiological examina- tion reveals late bone destructive changes without being able to evaluate soft parts that can be affected from the onset of the disease [2]. New imaging techniques, mus- culoskeletal ultrasound (US) and magnetic resonance imaging (MRI), can identify early articular and peri-ar- ticular inflammatory lesions, playing an important role in diagnosis, therapeutic decisions and monitoring of the disease [3-6].
The ankle and foot are affected by RA in more than 90% of patients during the disease course [7]: 20% of pa- Abstract
Aim: To evaluate the frequency of tibiotalar and subtalar joints together with extensor, flexor and peroneal tendons in- flammatory lesions in rheumatoid arthritis (RA) patients by using ultrasound (US) and magnetic resonance imaging (MRI).
Material and methods. Fifty RA patients and 25 healthy subjects were prospectively included. All patients and controls underwent clinical examination (to screen for swollen and/or tender ankles) and ankle US and MRI (to screen for synovial hypertrophy – SH, tenosynovitis and power Doppler – PD signals). The imaging tests were compared using overall agree- ment, positive agreement, Cohen’s κ, sensitivity, specificity and positive likelihood ratio. Results. The subtalar joint had the highest frequency of US-detected SH (30%), as well as positive PD signals (10%). Regarding US joint effusion, the tibiotalar joint recorded the highest frequency (44%). The most frequent US tenosynovitis was detected in the tibialis posterior tendon (40%). Compared to MRI, US evaluation of tibiotalar joints had very good agreement and large effect on detection probability for both SH and effusion (kappa 0.84, positive likelihood ratio 21.1). Compared to MRI, the sensitivity and specificity for US joint involvement ranged between 72.0-88.5% and 82.4-95.8%, and for tenosynovitis were 33.3-78.6% and 85.2-100%, re- spectively. Compared to asymptomatic RA patients (n=25), those with at least one symptomatic ankle (n=25) had significantly higher frequencies of both SH and effusion in all the evaluated structures. Conclusion: US has high sensitivity and specificity in detecting RA inflammatory lesions in the ankle and rearfoot, in very good agreement with MRI. The high frequency of ankle inflammatory lesions in RA should result in increased interest in the imaging evaluation of these structures.
Keywords: rheumatoid arthritis; musculoskeletal ultrasound; magnetic resonance imaging; synovitis; tenosynovitis DOI: 10.11152/mu-2038
tients have structural changes from disease onset, while 50% start displaying these changes in the first 6 years of illness [8]. Clinical ankle examination has low accuracy and it is influenced by many local factors (e.g. adjacent anatomical structures, deformities, obesity and venous insufficiency). In current practice, MRI is rarely used be- cause of high costs, limited accessibility, motion artefacts and exposure to contrast agents. The superiority of US and MRI to clinical examination in early detection of in- flammatory synovial lesions has been proved by numer- ous studies, especially in the hands [9,10]. Regarding the ankles, the literature is scarce and the few articles which research this area have focused especially on the forefoot [11,12]. However, despite the destructive potential of RA joint involvement, the ankle remains an anatomical re- gion neglected in current practice. In particular, the ankle is not included in composite RA activity scores and there is little data from studies on ankle and rear-foot imaging, despite poor clinical examination [13-19].
Therefore, this study aims to evaluate the frequency of ankle and rear-foot inflammatory lesions (synovitis and tenosynovitis) in RA patients and to compare the US and MRI (as a reference standard) findings.
Materials and methods Patients selection
In this prospective study, we evaluated 50 patients di- agnosed with RA (fulfilling the 2010 ACR/EULAR crite- ria [20]) and 25 healthy controls subjects with no current or history of joint disease. RA patients were recruited in 2018 in random order of presentation from the out- patient clinic. All patients and controls underwent clini- cal examination, laboratory tests, US and MRI evalua- tion (contrast MRI for RA patients and native MRI for controls). Exclusion criteria included age below 18 years, severe deformity of ankle joints, ankle surgery/trauma, MRI contraindications, local complex regional pain syn- drome, fibromyalgia, pregnancy, glucocorticoids (pulse- therapy, intramuscular, intra-articular injections) in the month prior to study inclusion. Oral doses of glucocorti- coids of ≤10 mg prednisone equivalent which were sta- ble in the month prior to study inclusion were allowed.
All patients and healthy controls gave written informed consent prior to enrolment in the study. The study was approved by the local Ethics Committee.
Clinical examination and laboratory data
All patients and control subjects were clinically eval- uated by the same rheumatologist, blinded to laboratory and imaging evaluations. The clinical examination of the ankles was made bilaterally and consisted of inspection, palpation and active movement of each compartment
from ankle and subtalar region, in order to detect signs of inflammation (pain and swelling). Additionally, the rheu- matologist noted concomitant local pathology: hallux valgus, pes planus, chronic venous insufficiency. Current conditions that could influence ankle symptoms (i.e. dia- betes mellitus, chronic venous insufficiency, obesity and exclusion criteria) were retrieved from the patient’s med- ical history. Laboratory data included rheumatoid factor (RF) and anti-citrullinated protein antibodies (ACPA).
US examination
The US scans were performed and interpreted by a single rheumatologist with more than 7 years of expe- rience, on the same day with clinical examination and blinded to clinical and imaging results, using the EULAR ankle technique [21]. An Esaote MyLabTwice machine, equipped with a 12-18 MHz linear transducer was used.
Both ankles were scanned. Tibiotalar joints (TTJ) with their anterior and posterior recesses and subtalar joints (STJ) from medial, lateral and posterior aspects were as- sessed. The following tendons were examined: tibialis anterior (TA), extensor hallucis longus (EHL), extensor digitorum longus (EDL), tibialis posterior (TP), flexor digitorum longus (FDL), flexor hallucis longus (FHL), peroneus longus (PL) and brevis (PB). The US examina- tion included initial grey scale (GS), followed by power Doppler (PD) evaluation. Power Doppler settings were constant (gain just below noise level, 750 Hz pulse rep- etition frequency, 8-10 MHz Doppler frequency, low wall filter). The following pathologic findings were re- corded: synovial hypertrophy (SH) and effusion at joint/
tendon level. Their definitions and interpretation were made according to OMERACT recommendations [22].
For quantifying these pathologic findings, the following semi-quantitative scales (0-3) were used: for joint SH, the initial score developed by Szkudlarek et al [23,24], taking into account the latest EULAR-OMERACT rec- ommendations [25], which were verified on large joints.
For joints evaluated from multiple windows, the highest grade of SH was recorded. Joint effusion was quantified with a dichotomous score (present/absent). For tenosyno- vitis quantification, we applied the OMERACT recom- mended score [26].
MRIA trained radiologist performed contrast MRI in RA patients and native MRI in controls (for ethical reasons), in the same day/ the next 2 days for each patient. The involved ankle was scanned in patients with unilateral symptomatic ankle. In RA patients with bilateral symp- tomatic/ asymptomatic ankles and also in healthy con- trols, the right ankle and rearfoot were evaluated. MRI was performed with a 1.5 Tesla General Electric Optima 450 WGEM machine, using a high definition dedicated
8-channel transmission antenna, on patients positioned in dorsal decubitus with their examined ankle in a neutral position. The examination protocol included: a) native acquisition proton-density-weighted fast spin echo (FSE) in axial, coronal and sagittal planes, T2-weighted FSE in coronal plane andT1-weighted FSE in sagittal plane with slice thickness of 4 mm, field of view (FOV) of 19 cm and matrix of 320 x 224 mm; b) acquisition of pre-con- trast 3D T1-weighted spoiled gradient recalled (SPGR;
liver acquisition with volume acceleration sequence - LAVA) in axial plane with slice thickness of 2 mm and FOV of 19 cm and matrix of 228 x 224 mm; two post- contrast acquisitions: and early 3D T1-weighted SPGR (LAVA) in axial plane with slice thickness of 2 mm and a late T1-weighted SPGR (LAVA) in a sagittal plane with a slice thickness of 3 mm. Contrast was administered intra- venously (0.1 mmols of gadolinium-diethylenetriamine penta-acetic acid per kilogram of body weight) only in RA patients. For this subgroup, SH and effusion were evaluated separately for both joints (TTJ, STJ with their anterior and posterior recesses) and tendons (anterior, lateral, medial and posterior ankle compartments). These inflammatory lesions were defined according to OMER- Table I. General characteristics of rheumatoid arthritis patients (n=50)
Age (years) 55(17)
Women (%) 42(84)
Disease duration (years) 6(11)
Ankle symptoms (%) 25(50)
Rheumatoid factor positive (%) 34(68)
ACPA positive (%) 40(80)
NSAIDs (%) 5(10)
GC (%) 10(20)
GC dose (mg/day PE) 10(5)
MTX (%) 20(40)
MTX dose (mg/week) 10(10)
> 1 csDMARDs (%) 7(14)
tsDMARDs (%) 1(2)
bDMARDs (%) 17(34)
Diabetes mellitus (%) 7(14)
Hallux valgus (%) 7(14)
Obesity (%) 6(12)
CVI-LL (%) 11(22)
Pes planus (%) 3(6)
Continuous variables are expressed as “median (interquartile range)” and categorical variables are expressed as “number (per- centage of group)”. ACPA – anti-citrullinated protein antibodies;
b/cs/tsDMARD – biological or conventional synthetic or targeted synthetic disease-modifying anti-rheumatic drug; CVI-LL – chronic venous insufficiency of lower limbs; GC – glucocorticoids;
PE – prednisone equivalent; MTX – methotrexate
ACT recommendations [27]. Grading of synovitis was done using the RAMRIS semi-quantitative score [28].
Tenosynovitis was graded using a separate semi-quanti- tative method [29].
Statistics
Nominal variables are reported as “absolute frequen- cy (percent of subgroup)”. All recorded continuous vari- ables were non-normally distributed and were reported as
“median (inter-quartile range)”. Differences between di- chotomous groups were evaluated using Mann Whitney U and χ2 tests. US’s performance compared to MRI was evaluated with: overall agreement (OA), positive agree- ment (PA), Cohen’s κ (kappa; strength of agreement:
<0.2 poor, 0.21-0.40 fair, 0.41-0.60 moderate, 0.61-0.80 good and >0.80 very good [30]); sensitivity, specificity and positive likelihood ratio (PLR; effect on increasing probability of involvement detection: PLR>10 large, PLR=5-10 moderate, PLR<5 small [31]). To compare ul- trasound to MRI regarding SH/tenosynovitis grading, the cross-tabulation was reduced to a two-by-two table by grouping “absent-minimal” (grade 1 SH being difficult to appreciate as pathological) and “moderate-severe” (grade 2 and 3). The statistical tests were considered significant if p<0.05 and were done with IBM SPSS Statistics ver- sion 22.0 for Windows (Armonk, NY, IBM Corp.).
Results
General characteristics
The study included 50 RA patients and 25 controls.
Demographic and laboratory data of RA patients are de- tailed in Table I.
Frequency of US inflammatory lesions in RA STJ had the highest frequency of US-detected SH (30% versus 28% in TTJs), as well as PD signals (10%
versus 6% in TTJs). Regarding joint effusion, TTJ re- corded a higher frequency (44% versus 38 % in STJ).
The most frequent tenosynovitis was detected in TP ten- don sheath (40%), with 38% SH and 34% positive PD signal. Most frequently, US detected effusion in the TP (28%). The posterior recess of STJ proved to be the most reliable for SH and effusion detection. The lateral STJ window offered the best view for PD signal detection (Table II, Table III).
Frequency of MRI inflammatory lesions in RA Table II presents the summary of lesions detected on MRI, which proved to be superior to US in joint SH de- tection. In MRI, both TTJ and STJ recorded the same fre- quency of involvement (32%); in the case of TTJ, the dif- ference between the two imaging methods being higher.
MRI was also superior to US in detecting joint effusion (50% in TTJ and 42% in STJ). Regarding tenosynovitis,
MRI remained superior to US, recording TP tenosynovi- tis in 46% of RA patients. Similar to US, MRI detected effusion most frequently in TTJ (50%) and TP (26%).
US compared to MRI in RA
Compared to MRI, US evaluation of TTJs had very good agreement for both SH and effusion (fig 1), includ- ing SH grading. However, STJs evaluation produced moderate to good agreement and moderately increased the chance of detection of both SH and effusion. In tendons, the highest agreement compared to MRI was achieved for overall US involvement (SH and effu- sion) of EDL (Table IV), while the lowest performance of US was recorded for overall involvement of TA and FHL.Symptomatic versus asymptomatic ankles
Compared to asymptomatic RA patients (n=25), those with at least one symptomatic ankle (n=25) had significantly lower median age and significantly higher frequencies of both SH and effusion in all the evalu- ated structures. PD signal was more frequently detect- ed in symptomatic patients, both in joints (24% versus 0%, p=0.009) and tendons (84% versus 12%, p<0.001) (Table IV). These two groups did not differ significantly regarding the frequency of positive RF, ACPA, treatment regime and recorded comorbidity (p>0.1). In the asymp- tomatic RA subgroup, there were 10 (20%) patients with normal US, 11 (22%) with normal MRI and 6 (12%) with normal aspect in both imaging techniques.
US compared to MRI in controls
The control group (n=25) had a median age of 53 years and included 21 women (84%). Unlike US, native MRI cannot differentiate effusion from SH, detecting
only distension of joint capsules and tendon sheaths. In the control group US revealed effusion in 5 (20.0%) TTJs and in 9 (36.0%) STJs, with very good agreement with MRI (kappa = 0.82 and 0.83 respectively). Regarding tendon involvement, US also detected only intra-tendon sheath effusion in the TP (5 cases), FHL (5 cases) and FDL (4 cases) tendons, with good and very good agree- ment with MRI (κ>0.63, PLR>11). US detected no ankle/
tendon PD signals.
Table II. Ultrasound–MRI comparison on RA ankle joint involvement (n=50 patients)
US (n) MRI (n) OA (%) PA (%) κ Se (%) Sp (%) PLR
overall (effusion and SH)
TTJ 24 26 92 95.8 0.84* 88.5 95.8 21.1
STJ 22 25 78 81.8 0.56* 72.0 84.0 4.5
any 29 33 80 89.7 0.58* 78.8 82.4 4.5
effusion presence
TTJ 22 25 90 95.5 0.80* 84.0 96.0 21.0
STJ 19 21 80 78.9 0.58* 71.4 86.2 5.2
SH presence
TTJ 14 16 92 92.9 0.81* 81.2 97.1 28.0
STJ 15 16 86 80.0 0.67* 75.0 91.2 8.5
grade 2 and 3 SH
TTJ 7 5 96 71.4 0.81* 100 95.6 22.7
STJ 9 2 86 22.2 0.32* 100 85.4 6.8
κ’s significance: * p<0.007; US – ultrasound; MRI – magnetic resonance imaging; OA – overall agreement; PA – positive agreement;
PLR – positive likelihood ratio; RA – rheumatoid arthritis; Se – sensitivity; SH – synovial hypertrophy; Sp – specificity; STJ – subtalar joint;
TTJ – tibiotalar joint.
Fig 1. Longitudinal US image (A) and sagittal T1 weighted post-gadolinium 3D LAVA MRI (B) at anterior recess of tibio- talar joint reveal synovial hypertrophy (arrow heads) and joint effusion (*) in an RA patient. Longitudinal US image (C) and sagittal T1 weighted fat saturated post-gadolinium MRI image (D) of tibialis posterior tendon reveal tenosynovitis (arrows) with synovial proliferation and peri- and intra-tendineous PD signals in an RA patient.
Table III. The frequency of positive PD signals in RA patients (n=50)
joint cases TTJ STJ pSTJ lSTJ mSTJ any
3(6) 5(10) 0 4(8) 1(2) 6(12)
tendon cases TA EHL EDL PL PB TP FDL FHL any
3(6) 2(4) 4(8) 10(20) 7(14) 17(34) 8(16) 4(8) 24(48)
The results are expressed as number (%). EDL – extensor digitorum longus; EHL – extensor hallucis longus; FDL – flexor digitorum longus;
FHL – flexor hallucis longus; l – lateral; m – medial; p – posterior; PB – peroneus brevis; PL – peroneus longus; RA – rheumatoid arthritis;
TA – tibialis anterior; TP – tibialis posterior; STJ – subtalar joint; TTJ – tibiotalar joint.
Table IV. Ultrasound-MRI comparison on ankle tendon involvement in RA (n=50 patients)
US(n) MRI(n) OA(%) PA(%) κ Se(%) Sp(%) PLR
overall TA 3 9 90 100 0.45* 33.3 100 -
EHL 2 5 94 100 0.55* 40.0 100 -
EDL 7 8 94 85.7 0.80* 75.0 97.6 31.3
PL 14 14 88 78.6 0.70* 78.6 91.7 9.5
PB 11 11 88 72.7 0.65* 72.7 92.3 9.4
TP 20 23 78 80.0 0.55* 69.6 85.2 4.7
FDL 12 15 76 83.3 0.65* 66.7 94.3 11.7
FHL 8 19 78 100 0.47* 42.1 100 -
Any 29 28 86 86.2 0.72* 89.3 81.8 4.9
effusion presence TA 3 4 98 100 0.85* 75.0 100 -
EHL 1 0 - - - - - -
EDL 6 4 92 50.0 0.56* 75.0 93.5 11.5
PL 9 9 80 44.4 0.32* 44.4 87.8 3.6
PB 6 3 86 16.7 0.16# 33.3 89.4 3.1
TP 14 13 82 64.3 0.54* 69.2 86.5 5.1
FDL 8 4 88 37.5 0.44* 75.0 89.1 6.9
FHL 7 14 86 100 0.59* 50.0 100 -
SH presence TA 3 8 90 100 0.50* 37.5 100 -
EHL 2 5 94 100 0.55* 40.0 100 -
EDL 5 6 94 80.0 0.69* 66.7 97.7 29.0
PL 11 11 96 90.9 0.88* 90.9 97.4 35.0
PB 9 11 88 77.8 0.63* 63.6 94.9 12.5
TP 19 20 82 78.9 0.62* 75.0 86.7 5.6
FDL 9 13 84 77.8 0.54* 53.8 94.6 10.0
FHL 5 11 88 100 0.57* 45.5 100 -
grade 2 and 3 SH TA 3 5 96 100 0.73* 60.0 100 -
EHL 1 0 - - - - - -
EDL 5 4 94 60.0 0.63* 75.0 95.7 17.4
PL 7 10 86 71.4 0.51* 50.0 95.0 10.0
PB 3 2 94 33.3 0.37* 50.0 95.8 11.9
TP 14 14 96 92.9 0.90* 92.9 97.2 33.2
FDL 4 5 90 50.0 0.39* 40.0 95.6 9.1
FHL 5 17 76 100 0.36* 29.4 100 -
κ’s significance: * p≤0.023; # non-significant; - incalculable (division by 0 or all ratings are the same for MSUS and/or MRI); US – ultra-sound; MRI – magnetic resonance imaging fndings; OA – overall agreement; PA – positive agreement; PLR – positive likelihood ratio; RA – rheumatoid arthritis; Se – sensitivity; SH – synovial hypertrophy; Sp – specificity; tendons: EDL – extensor digitorum longus;
EHL – extensor hallucis longus; FDL – flexor digitorum longus; FHL – flexor hallucis longus; PB – peroneus brevis; PL – peroneus longus;
TA – tibialis anterior; TP – tibialis posterior.
Discussions
Taking into account that there is scarce literature on the frequency of inflammatory lesions in the ankles of RA patients, our study brings important information to this subject. The results have revealed a very good agree- ment (both in joints and tendons) between US and MRI findings.
Concerning the frequency of inflammatory lesions, our study found similar data to previous studies performed on subgroups of patients (symptomatic or asymptomatic ankles) [13,19,32,33]. In the only study that evaluated ankle using US regardless of the presence of symptoms, Alsawaidi et al [34] reported higher frequency of TTJ and lower frequency of TP and FDL involvement.
The literature contains few reports of US compari- son to MRI in the evaluation of ankle involvement in RA patients. In a study published in 1996, Lehtinen et al [15] reported high concordance between US and high field MRI findings but their results are difficult to be compared with our results due to the technological dif- ferences of imaging machines used and by differences in study design. Wakefield et al [19] reported the same pattern of US performance in detecting ankle joint and tendon involvement, generally with lower sensitivity and specificity of TTJ and tendon involvement compared to our observations. Taken together, these two studies and our results seem to confirm the overall good performance of US compared to MRI in evaluation RA involvement of ankle structures.
Concerning the agreement between US and MRI, Premkumar et al [35] reported sensitivities and specifi- cities above 80% for US detection of TP and FDL in- volvement compared to MRI. We report slightly lower sensitivities and higher specificities of US compared to MRI for the detection of anomalies in these tendons, in a homogenous and larger group of RA patients. However, the results of this study are not directly comparable to our observations, since the authors evaluated patients with different forms of spondyloarthritis, which are known to generate important structural abnormalities of tendons [36]. Scheel et al [37] found low agreement of US with MRI on overall involvement of TTJs (57% compared to 92% in our study), STJs (57% compared to 78% in our study) and extensor tendons (36% compared to 90% in our study) and good agreement on overall involvement of flexors (86% compared to 82% in our study) and peroneus tenosynovitis (71% compared to 88% in our study).
Unlike previous studies, for a more accurate compari- son between US and MRI, we differentiated by US effu- sion from SH in joints and tendon sheaths. This allowed
us to understand the low agreements recorded in some structures, regarding the overall assessment. For exam- ple, in the case of TA, EHL and FHL, the low agreement is caused by low sensitivity of US to detect what MRI identify as grade 1 synovial hypertrophy. More so, joint by joint analysis (PA=100% in all three tendons) under- lines that all the pathologic findings detected by US were confirmed by MRI. In addition, the latter detected more minimal proliferation in the tendon sheath, possibly over- estimating true pathological lesions. We included PD sig- nals in the US examination, increasing the US specificity for SH detection. Also, special attention was paid to the posterior foot (posterior recess of TTJ and STJ), which was not included in the cited articles.
We note the low incidence of PD in the ankle joints, which can be explained by the low sensitivity of PD for large joints and deep anatomic areas. The same low PD frequency in TTJ was also noted by other authors [13,34,38,39], who proposed to further scan the lateral and medial aspects of TTJ’s anterior recess in order to increase PD sensitivity. Regarding STJ, PD signals were detected only in the medial and lateral recesses.
By including healthy controls in our study, we aimed to identify truly pathological findings. The presence of effusion in ankle joints/ tendon sheaths in normal indi- viduals draws attention to other possible causes, not nec- essarily inflammatory.
The interpretation of our results needs to be per- formed in light of several study limitations: we did not perform an inter- and intra-observer agreement; conven- tional radiography was not done and neither bone dam- age evaluation; not all MRI scans were in the same day as clinical and US evaluations. For ethical reasons, controls were evaluated by native MRI, in order to mitigate poten- tial risks of contrast substance.
In conclusion US has high sensitivity and specificity in detecting RA inflammatory lesions in the ankle and rearfoot, generally in very good agreement with MRI.
For the detection of PD signals, a more elaborate exami- nation of the ankle may be necessary, as well as a stand- ardized ankle and rearfoot evaluation protocol. Although ankles are not routinely evaluated in clinical practice despite the destructive potential and implicitly the func- tional deficit of RA, the high frequency of inflammatory lesions at this level should result in increased interest in the imaging evaluation of these structures.
References
1. Hanson MT, Keiding S, Lauritzen SL, Manthorpe R, So- rensen SF, Wiik A. Clinical assessment of disease activity in rheumatoid arthritis. Scand J Rheumatol 1979;8:101-105.
2. Klarlund M, Ostergaard M, Jensen KE, Madsen JL, Skjodt H, Lorenzen I. Magnetic resonance imaging, radiography, and scintigraphy of the finger joints: one year follow up of patients with early arthritis. The TIRA Group. Ann Rheum Dis 2000;59:521-528.
3. Baker JF, Conaghan PG, Emery P, Baker DG, Ostergaard M. Validity of early MRI structural damage end points and potential impact on clinical trial design in rheumatoid ar- thritis. Ann Rheum Dis 2016;75:1114-1119.
4. Emery P, van der Heijde D, Ostergaard M, et al. Explorato- ry analyses of the association of MRI with clinical, labora- tory and radiographic findings in patients with rheumatoid arthritis. Ann Rheum Dis 2011;70:2126-2130.
5. Naredo E, Collado P, Cruz A, et al. Longitudinal power Doppler ultrasonographic assessment of joint inflamma- tory activity in early rheumatoid arthritis: predictive value in disease activity and radiologic progression. Arthritis Rheum 2007;57:116-124.
6. Wakefield RJ, Gibbon WW, Emery P. The current status of ultrasonography in rheumatology. Rheumatology (Oxford) 1999;38:195-198.
7. Brenton-Rule A, Dalbeth N, Menz HB, Bassett S, Rome K.
Foot and ankle characteristics associated with falls in adults with established rheumatoid arthritis: a cross-sectional study. BMC Musculoskelet Disord 2016;17:22.
8. Baan H, Drossaers-Bakker W, Dubbeldam R, van de Laar M. We should not forget the foot: relations between signs and symptoms, damage, and function in rheumatoid arthri- tis. Clin Rheumatol 2011;30:1475-1479.
9. Naredo E, Bonilla G, Gamero F, Uson J, Carmona L, Laffon A. Assessment of inflammatory activity in rheumatoid ar- thritis: a comparative study of clinical evaluation with grey scale and power Doppler ultrasonography. Ann Rheum Dis 2005;64:375-381.
10. Schmidt WA, Schicke B, Ostendorf B, Scherer A, Krause A, Walther M. Low-field MRI versus ultrasound: which is more sensitive in detecting inflammation and bone damage in MCP and MTP joints in mild or moderate rheumatoid arthritis? Clin Exp Rheumatol 2013;31:91-96.
11. Szkudlarek M, Narvestad E, Klarlund M, Court-Payen M, Thomsen HS, Ostergaard M. Ultrasonography of the metatarsophalangeal joints in rheumatoid arthritis: com- parison with magnetic resonance imaging, conventional radiography, and clinical examination. Arthritis Rheum 2004;50:2103-2112.
12. Wakefield RJ, Kong KO, Conaghan PG, Brown AK, O’Connor PJ, Emery P. The role of ultrasonography and magnetic resonance imaging in early rheumatoid arthritis.
Clin Exp Rheumatol 2003;21(5 Suppl 31):S42-S49.
13. Gutierrez M, Pineda C, Salaffi F, et al. Is ankle involvement underestimated in rheumatoid arthritis? Results of a multi- center ultrasound study. Clin Rheumatol 2016;35:2669-2678.
14. Joshua F, Lassere M, Bruyn GA, et al. Summary findings of a systematic review of the ultrasound assessment of syno- vitis. J Rheumatol 2007;34:839-847.
15. Lehtinen A, Paimela L, Kreula J, Leirisalo-Repo M, Taavit- sainen M. Painful ankle region in rheumatoid arthritis.
Analysis of soft-tissue changes with ultrasonography and MR imaging. Acta Radiol. 1996;37:572-577.
16. Luukkainen RK, Saltyshev M, Koski JM, Huhtala HS. Re- lationship between clinically detected joint swelling and effusion diagnosed by ultrasonography in metatarsophalan- geal and talocrural joints in patients with rheumatoid arthri- tis. Clin Exp Rheumatol 2003;21:632-634.
17. Maillefert JF, Dardel P, Cherasse A, Mistrih R, Krause D, Tavernier C. Magnetic resonance imaging in the assess- ment of synovial inflammation of the hindfoot in patients with rheumatoid arthritis and other polyarthritis. Eur J Ra- diol 2003;47:1-5.
18. Szkudlarek M, Terslev L, Wakefield RJ, et al. Summary Findings of a Systematic Literature Review of the Ultra- sound Assessment of Bone Erosions in Rheumatoid Arthri- tis. J Rheumatol. 2016;43:12-21.
19. Wakefield RJ, Freeston JE, O’Connor P, et al. The optimal assessment of the rheumatoid arthritis hindfoot: a compara- tive study of clinical examination, ultrasound and high field MRI. Ann Rheum Dis 2008;67:1678-1682.
20. Aletaha D, Neogi T, Silman AJ, et al. 2010 rheumatoid ar- thritis classification criteria: an American College of Rheu- matology/European League Against Rheumatism collabo- rative initiative. Ann Rheum Dis 2010;69:1580-1588.
21. Backhaus M, Burmester GR, Gerber T, et al. Guidelines for musculoskeletal ultrasound in rheumatology. Ann Rheum Dis 2001;60:641-649.
22. Wakefield RJ, Balint PV, Szkudlarek M, et al. Musculoskel- etal ultrasound including definitions for ultrasonographic pathology. J Rheumatol 2005;32:2485-2487.
23. Szkudlarek M, Court-Payen M, Jacobsen S, Klarlund M, Thomsen HS, Ostergaard M. Interobserver agreement in ultrasonography of the finger and toe joints in rheumatoid arthritis. Arthritis Rheum 2003;48:955-962.
24. Szkudlarek M, Court-Payen M, Strandberg C, Klarlund M, Klausen T, Ostergaard M. Power Doppler ultrasonography for assessment of synovitis in the metacarpophalangeal joints of patients with rheumatoid arthritis: a compari- son with dynamic magnetic resonance imaging. Arthritis Rheum 2001;44:2018-2023.
25. Terslev L, Naredo E, Aegerter P, et al. Scoring ultrasound synovitis in rheumatoid arthritis: a EULAR-OMERACT ultrasound taskforce-Part 2: reliability and application to multiple joints of a standardised consensus-based scoring system. RMD Open 2017;3:e000427.
26. Naredo E, D’Agostino MA, Wakefield RJ, et al. Reliabil- ity of a consensus-based ultrasound score for tenosynovitis in rheumatoid arthritis. Ann Rheum Dis 2013;72:1328- 1334.
27. Ostergaard M, Peterfy C, Conaghan P, et al. OMERACT Rheumatoid Arthritis Magnetic Resonance Imaging Stud- ies. Core set of MRI acquisitions, joint pathology defini- tions, and the OMERACT RA-MRI scoring system. J Rheumatol 2003;30:1385-1386.
28. Ostergaard M, Peterfy CG, Bird P, et al. The OMER- ACT Rheumatoid Arthritis Magnetic Resonance Imaging (MRI) Scoring System: Updated Recommendations by the
OMERACT MRI in Arthritis Working Group. J Rheumatol 2017;44:1706-1712.
29. Haavardsholm EA, Ostergaard M, Ejbjerg BJ, Kvan NP, Kvien TK. Introduction of a novel magnetic resonance imaging tenosynovitis score for rheumatoid arthritis: reli- ability in a multireader longitudinal study. Ann Rheum Dis 2007;66:1216-1220.
30. McHugh ML. Interrater reliability: the kappa statistic. Bio- chem Med (Zagreb) 2012;22:276-282.
31. Altman DG. Some common problems in medical research.
In: Altman DG (ed.). Practical statistics for medical re- search. London: Chapman & Hall; 1991:404-405.
32. Hammer HB, Kvien TK. Ultrasonography shows signifi- cant improvement in wrist and ankle tenosynovitis in rheu- matoid arthritis patients treated with adalimumab. Scand J Rheumatol 2011;40:178-182.
33. Suzuki T, Okamoto A. Ultrasound examination of symp- tomatic ankles in shorter-duration rheumatoid arthritis patients often reveals tenosynovitis. Clin Exp Rheumatol 2013;31:281-284.
34. Alsuwaidi M, Ehrenstein B, Fleck M, Hartung W. Asymp- tomatic Versus Symptomatic Ankle Joints in Rheumatoid
Arthritis: A High-Resolution B-Mode and Power Dop- pler Ultrasound Study. Arthritis Care Res (Hoboken) 2016;68:861-864.
35. Premkumar A, Perry MB, Dwyer AJ, et al. Sonography and MR imaging of posterior tibial tendinopathy. AJR Am J Roentgenol 2002;178:223-232.
36. Hoenen-Clavert V, Rat AC, Loeuille D, et al. Inflammatory and structural evaluation in spondyloarthritis: magnetic resonance imaging analysis of axial and peripheral involve- ment. J Rheumatol 2007;34:762-768.
37. Scheel AK, Schmidt WA, Hermann KG, et al. Interob- server reliability of rheumatologists performing muscu- loskeletal ultrasonography: results from a EULAR “Train the trainers” course. Ann Rheum Dis 2005;64:1043- 1049.
38. Suzuki T. Power Doppler ultrasonographic assessment of the ankle in patients with inflammatory rheumatic diseases.
World J Orthop 2014;5:574-584.
39. Șerban O, Bădărînză M, Fodor D. The relevance of ultra- sound examination of the foot and ankle in patients with rheumatoid arthritis - a review of the literature. Med Ultra- son 2019;21:175-182.
