Acta Radiológica 37 (1996) 572-577 Copyright © Acta Radiológica 1996

Printed in Denmark ■ AU rights reserved

ACTA RADIOLÓGICA
ISSN 0284-1851

PAINFUL ANKLE REGION IN RHEUMATOID ARTHRITIS

Analysis of soft-tissue changes with ultrasonography and MR imaging

A. LEHTINEN , L. PAIMELA , J. KREULA , M. LEIRISALO-REPO and M. TAAVITSAINEN

Departments of 'Radiology, and 'Medicine, University Central Hospital, and 2Rheumatology, Helsinki City Hospital, Helsinki,
Finland.

Abstract
Purpose: To establish the diagnostic usefulness of ultrasonography (US) and MR Key words: Ankle; rheumatoid arthritis;
imaging in patients with rheumatoid arthritis (RA) suffering from prolonged pain in the joints; diseases; MR; ultrasound; radio­
ankle region, where plain radiography did not demonstrate any changes. graphy; comparative studies.
Material and Methods: Seventeen patients were studied with 0.1 T MR imaging and
with high-frequency US. Talocrural and subtalar joints (including talonavicular joints), Correspondence: Ari Lehtinen,
and medial, lateral, and extensor tendons and their synovial sheaths were examined by Radiology, University Central Hospital,
MR and US. Haartmaninkatu 4,
Results: Abnormal findings were found by MR imaging and US in altogether 76% FIN-00290, Helsinki, Finland.
of the patients, by MR alone in 53%, and by US alone in 59%. In 41% of the patients, FAX *358-0-47 14 404.
lesions were demonstrated only by one method. Talocrural and subtalar joints were the
most often affected sites (41% each), followed by the peroneus tendon (23%). In the Accepted for publication 12 January
joints, the abnormal findings were interpreted as synovitis; in the tendon areas, mostly 1996.
as tenosynovitis. MR and US were highly significantly concordant (p<0.0001), but cor­
relation with clinical features was poor.
Conclusion: In patients with pain in the ankle region, US and MR imaging can con­
tribute to the diagnosis and localization of the abnormality when the plain radiography
is normal. Easily available and inexpensive US can be recommended as the first imag­
ing method after plain radiography. Some divergence seems to exist between US and
MR, and in complicated cases both methods are recommended.

The ankle region is commonly affected in rheuma­
toid arthritis (RA) (17). Significant disability may
arise due to persistent damage to tarsal, subtarsal, or
midtarsal joints. In clinical assessment, joint tender­
ness and joint swelling are important for determina­
tion of therapeutic interventions. The evaluation of
painful ankle can, however, be problematic, and
findings at plain radiography may not be sufficient
to provide diagnostic information. Early changes of
the cartilage, bones, and tendons of the ankle area
can usually not be diagnosed with conventional ra­
diography. For the clinician, accurate localization of
the inflammation in a painful ankle is necessary for
the determination of adequate treatment.
Ultrasonography (US) has turned out to be a reli­
able method in the diagnosis of soft-tissue lesions
(6, 10, 14). The tendons and their synovial sheaths,
as well as the articular synovium and inflammatory
fluid contents around the tendons and in the joint
spaces, are well demonstrated (3, 8, 9). US is a sen­
sitive method in detecting intra-articular effusion
and synovitis also in subtalar and midtarsal joints
(11).
MR imaging is well suited for the examination of
intra- and extra-articular soft tissues (4, 18, 19).
Synovial proliferation and small effusions can be
shown and quantified by MR (15, 16). In previous
studies of patients with RA, MR has proved to be an
effective method, especially in early phases of the
disease, to detect changes not visible on conven­
tional radiographs (5).
The purpose of the present study was to evaluate

572

Downloaded from acr.sagepub.com at CORNELL UNIV on August 6, 2015

 US AND MR OF PAINFUL ANKLE IN RHEUMATOID ARTHRITIS

Table 1 were performed by experienced rheumatologists.

Clinical characteristics and radiologie findings in patients with All patients had normal plain anteroposterior and
rheumatoid arthritis lateral radiographs of the ankle, usually performed
Number of patients 17
within 2 months prior to the US and MR examina­
Female/male 13/4 tion, which were performed on the same day by 2
Mean age, years 51.5 independent radiologists, who were unaware of the
Range 27-72 other study result or of the clinical symptoms of the
Mean duration of disease, years 9.5 patients.
Range 0.5-40
The MR examinations were performed with a 0.1
RF-positive* 15/17
Antirheumatic treatment
Auranofin 3/17
Intramuscular gold 3/17 Table 3
Methotrexate 4/17 Correlation between clinical findings and US and MR findings
Sulphasalazine 6/17 in the ankle region
Oral corticosteroids 4/17
Erosions Subtalar joint (including the ta lonavicular joint)
In hands 14/17 Pat. Clinical us MR
In feet 15/17 1
In ankles 0/17
+/+ - -
2 +/- - -
* Rheumatoid factor; Waaler-Rose >1:64. 3 +/+ - -
4 -/- - -
5 +/- - -
Table 2 6 -/- - -
Correlation between clinical findings and US and MR findings 7 +/+ synovitis -
in the ankle region 8 +/+ - -
9 +/+ synovitis -
Talocrural joint 10 +/- - -
Pat. Clinical US MR 11 +/- synovitis synovitis
12 +/+ - -
1 -/- - 13 +/+ synovitis -
2 -/- - 14 +/- - -
3 -/- synovitis 15 +/+ synovitis synovitis
4 -/- - 16 +/+ synovitis synovitis
5 -/- - synovitis 17 +/+ synovitis synovitis
6 +/+ -
7 -/- -
8 -/- - synovitis Table 4
9 -/- - Correlation between clinical findings and US and MR findings
10 -/- - synovitis in the ankle region
11 -/- synovitis
12 -/- - Tendons (peroneus communis, peroneus longus, peroneus brevis,
13 -/+ - tibialis posterior, flexor hallucis longus)
14 -/- - Pat. Clinical US MR
15 -/- synovitis
16 +/+ synovitis synovitis 1 - tenosynovitis (pc, pi, pb) -
17 -/- - 2
- normal. + clinically abnormal finding (pain/swelling). 3
4 + - -
5
6
the diagnostic usefulness of US and MR imaging in 7 - : tenosynovitis (pi)
8
RA patients with prolonged pain in the ankle region 9 — _ tendinitis (pi)
but without changes at plain radiography, and to 10 - - -
correlate the findings with clinical assessment. 11 - tenosynovitis (pi) -
12 + - -
13 + - tenosynovitis (pi, pb, fhl)
Material and Methods 14 - tenosynovitis (pi) -
15 - - -
Seventeen patients with prolonged pain in the ankle 16 - - -
region were examined. All patients had RA accord­ 17 + tenosynovitis (tp) tenosynovitis (tp)
ing to the ACR criteria (1). The clinical data of the pc =:peroneus communis, pi = peroneus long;us, pb = peroneus brevis,
patients are shown in Table 1. The clinical studies

573

Downloaded from acr.sagepub.com at CORNELL UNIV on August 6, 2015

 A. LEHTINEN ET AL.
pathological normal
3 4 MR
US 4
^
Wfk ¡11 3
Fig. 1. Pathological and normal findings at US and MR exami­
nations.
T MR unit (Merit, Picker). Gradient-echo sequences
were used: T2-weighted (TR/TE 1500/60 ms) and
Tl-weighted (150/20 ms) images in sagittal and
coronal orientation, and Tl-weighted (150/20 ms)
sagittal images after i.v. injection of contrast me­
dium (gadolinium dimeglumine, Gd-DTPA) at a
dosage of 0.1 mmol/kg. The slice thickness was 5.0
or 6.0 mm, the matrix size 256x256, and the field of
view 307x307 mm or 333x333 mm to get the whole
foot inside the examination area. The flip angle was
90°, and the imaging time 7 min 41 s in Tl-
weighted images and 12 min 48 s in T2-weighted
images.
The US examinations were performed with high-
resolution US equipment (Acusón 128 XP/10, 7.0-
MHz linear probe, or Aloka SSD 650, 7.5-MHz li­
near probe). A stand-off pad (Kitecko, 3M, St.
Louis, MO) was used to achieve an optimal distance
from the probe to the surface of the skin and to ob­
tain a sufficient view of the examination area. The
574
Downloaded from acr.sagepub.com at CORNELL UNIV on August 6, 2015
asymptomatic ankle was examined first; as no ab­
normalities were observed, it was used as compari­
son. Findings were also compared concerning soft-
tissue diagnostics with US (3, 6, 8, 9). The scans
were registered with a laser printer. US examina­
tions were performed by 2 radiologists (A. L. and
M. T.), both with 5 years' experience in soft-tissue
diagnostics by US.
In joints, synovial thickening and/or increased
synovial fluid was interpreted as synovitis; in ten­
don areas, tendinous thickening and/or increased
synovial fluid around the tendon was interpreted as
tenosynovitis.
In MR and US examinations, the same anatomic
regions were examined, i.e., the talocrural joint, the
subtalar joint (including talonavicular joint), and the
medial, lateral, and extensor tendons and their syno­
vial sheaths. Thickening of the synovium, synovial
fluid contents, tendon pathology, and pathological
fluid spaces were also registered.
In joint cavities, MR and US findings were classi­
fied into 3 grades: grade 0 (normal) = the thickness
of synovium and/or articular fluid together <2 mm,
grade 1 = from 2 mm to <4 mm, and grade 2 = >4
mm.
In tendons, the findings were classified into 3
grades: grade 0 (normal) = the thickness of syn­
ovium and/or synovial fluid together <lmm, grade 1
= from 1 mm to <3 mm, and grade 2 = >3 mm.
Statistical analysis. Correlation between US and
MR images was made by determining a p-value
with the contingency coefficient (C) with df=4 and
chi square.
Results
At clinical examination, all patients had pain and
most of them had swelling in the ankle region (Ta-
Fig. 2. Synovitis in talocrural joint, a) A sagit­
tal T2-weighted MR image. A large area of in­
creased signal intensity, indicating synovitis, is
seen in the joint (—»). b) US image of the same
patient in the sagittal section. The symptomatic
joint S shows a large hypoechoic synovial
thickening (+ - +).
 US AND MR OF PAINFUL ANKLE IN RHEUMATOID ARTHRITIS

Patient's Sin exti

!PH PWB§ jyjü^gK1

¡ s ; ·>*■■■">

Fig. 3. Synovitis in talonavicular joint, a) Sagittal T2-weighted MR image. A large area of increased signal intensity, indicating syno­
vitis, is seen in the joint (—»). b) US of the same patient. The asymptomatic joint (RT = right) shows moderate synovitis ( Î Î ) . On the
left (LT), symptomatic joint, a large hypoechoic area of synovial thickening is seen ( Î Î ) .

bles 2-4), usually in medial or lateral subtalar areas
(Table 3). MR showed abnormalities (synovitis and/
or tenosynovitis, no bone erosions) in 9 cases and
was normal in 8 cases (Fig. 1). US showed abnor­
malities in 10 cases and was normal in 7 cases (Fig.
1). Both MR images and US showed simultaneous
abnormalities in 6 cases, were normal in 4 cases,
and divergent in 7 cases. Either examination method
showed abnormalities in 13 cases; 11 patients were
interpreted as normal (Fig. 1).
In the talocrural joint, MR imaging and US
showed abnormalities in 4 cases each, each method
showed abnormalities in 6 cases, and both together
in one case (Fig. 2). Findings by both techniques to­
gether were normal in 10 cases (Table 2).
In the subtalar joint (including the talonavicular
joint), MR showed abnormalities in 4 cases and US
alone in 7 cases. Both techniques together showed
abnormalities in 4 cases (Fig. 3). Both methods to­
gether showed normal findings in 10 cases (Table
3).
In the tibialis posterior tendon, MR and US to­
gether showed tenosynovitis in one case. In the
flexor hallucis longus tendon, MR showed abnor­
malities in one case (Table 4).
In the peroneus tendons, MR and US alone were
abnormal in 3 cases each (Fig. 4). In no case was ab­
normality detected by both methods; each method
showed abnormalities in 6 cases. Peroneus longus
was the most often affected tendon (Table 4).
Talocrural and subtalar joints were the most often
affected of the examined sites; 41% (7/17 each) of
the patients had abnormal findings with either MR
or US. Of the tendon areas, the peroneus longus ten-

Fig. 4. Tenosynovitis. a) Sagittal T2-weighted MR image. An area of increased signal intensity (bright), indicating fluid, is seen
around the peroneus longus tendon (dark). Arrow points to the inflamed tendon, b) Transversal US image of tenosynovitis. Around
the peroneus longus tendon (—►), a hypoechoic zone of synovial thickening (+ - +) is seen. A hypoechoic synovial thickening is also
seen around the peroneus brevis tendon ( > ) .

575

Downloaded from acr.sagepub.com at CORNELL UNIV on August 6, 2015

 A. LEHTINEN ET AL.
don was most often affected; 35% (6/17) of the pa­
tients had abnormal findings. In flexor tendons,
there were only a few abnormal findings (2/
17=12%), and in extensor tendons there were no ab­
normal findings.
In the assessment of radiologie changes (grades
0-2), MR and US were significant (contingency co­
efficient C=0.608, df=4, chi square=49.8,
p<0.0001).
In the talocrural joint and tendon areas, the corre­
lation between clinical examination and imaging
was poor. In the subtalar joint (including the talona-
vicular joint), the clinical examination showed ab­
normality (pain and/or swelling) in 88% (15/17),
but US and/or MR showed inflammation in only
about half of the cases (47%) (Table 3).
Discussion
Involvement of the ankle region is common in RA.
Nearly 60% of patients with long-standing RA
show significant involvement of the ankle, and
about 30% of all surgery performed on RA pa­
tients involves the ankle and the hind foot (2). Per­
sistent pain and/or swelling is often a diagnostic
and therapeutic problem. Synovitis, bursitis, and
tenosynovitis may develop in any tendon sheath.
The early diagnosis of synovitis in the ankle area
is important to allow the prompt initiation of treat­
ment. The clinical features of ankle inflammation
are, however, neither specific nor sensitive, and
plain radiography of the ankle and midtarsal joints
may be informative only in the late stages of joint
affection when there is already permanent joint de­
struction.
Although the present study consisted of a rela­
tively small number of patients, the results indicate
that both US and MR are useful in the evaluation of
synovitis in a painful ankle without any detectable
changes on plain radiography. Correlation of the
findings by the imaging methods with clinical ex­
amination was less distinct. Obesity and extra-
articular swelling may hinder the diagnostics of in­
flammation in the ankle area. However, an impor­
tant clinical application in RA is the evaluation of
chronic tenosynovitis, also in order to determine the
right time for surgical intervention. US and MR
seem to facilitate the judgment of the condition of
tendons in the inflamed ankle. With US, some addi­
tional abnormal changes were found when com­
pared with MR (10 vs. 9), but there was discrepancy
in 7 cases between the 2 techniques. As there is no
gold standard method, it is not possible to compare
the diagnostic value of US and MR imaging. How­
ever, a soft-tissue analysis to detect synovitis/teno-
576
Downloaded from acr.sagepub.com at CORNELL UNIV on August 6, 2015
synovitis can be better made with imaging than with
palpation.
The use of US in the diagnostics of RA has been
limited. Recently we applied US in the diagnostics
of enthesopathy in patients with spondylarthro-
pathies (13). US has also been used in the examina­
tion of metacarpophalangeal joints, where it showed
its accuracy compared to conventional radiography,
especially in evaluation of early changes (7). Re­
cently KOSKI (12) found US to be a more objective
procedure than clinical examination in diagnosing
plantar tenosynovitis of the forefoot in early arthri­
tis. In another study by KOSKI (11), US proved to be
a rapid and sensitive method for detecting effusion
in the subtalar and midtarsal joints. It can also be
used as an accurate method in detecting abnormali­
ties in the joint cavities and synovial sheaths of ten­
dons in the ankle (11).
One disadvantage of US is that the method is
very dependent on the radiologist's skills, and, be­
cause the examination is dynamic, the visual data on
the prints are usually not as informative as in MR
images.
In conclusion, both US and MR imaging add to
the diagnostic accuracy of the hind foot in RA pa­
tients. The information can be readily applied in
clinical situations when the optimal therapeutic
strategy (local corticosteroid injection or synovec-
tomy/tenosynovectomy) is considered. US, being
easily available and less costly than MR examina­
tion, should be the first imaging technique after
plain radiography to be used for detection of soft-
tissue changes in the ankle. Guided by the US find­
ings, local injections can also be applied at the same
time. However, some divergence seems to exist be­
tween US and MR, and in complicated cases both
methods may be needed.
REFERENCES
1. ARNETT F. C „ EDWORTHY S. M., BLOCH D. A. et al.: The
American Rheumatism Association 1987 revised criteria
for the classification of rheumatoid arthritis. Arthritis
Rheum. 31 (1988), 315.
2. AYERS D. C. & SHORT W. H.: Arthritis surgery. Clin. Exp.
Rheumatol. 11 (1993), 75.
3. COAKLEY F. V., SAMANTA A. K. & FINLAY D. B.: Ultrasono-
graphy of the tibialis posterior tendon in rheumatoid arthri­
tis. Br. J, Rheumatol. 33 (1994), 273.
4. FELDMAN F., STARON R. B. & HARAMATI N.: Magnetic reso­
nance imaging of the foot and ankle. In: Rheumatic dis­
ease. Clinics of North America, 17:3. Imaging of rheumatic
diseases, p. 617. Edited by B. N. Weissman. W. B. Saun-
ders, Philadelphia 1991.
5. FOLEY-NOLAN D., STACK J. P., RYAN M. et al.: Magnetic res­
onance imaging in the assessment of rheumatoid arthritis -
a comparison with plain film radiographs. Br. J. Rheumatol.
30(1991), 101.
 US AND MR OF PAINFUL ANKLE IN RHEUMATOID ARTHRITIS
6. FORNAGE B. D. & RIFKIN M. D.: Ultrasound examination of
tendons. Radiol. Clin. North Am. 26 (1988), 87.
7. G R A S S I W . , TlTTARELLI E . , PlRANI O . , AVALTRONI D . & CER-
VINI C : Ultrasound examination of metacarpophalangeal
joints in rheumatoid arthritis. Scand. J. Rheumatol. 22
(1993), 243.
8. VAN HOLSBEECK M. & INTROCASO J. H.: Sonography of ten­
dons. In: Musculoskeletal ultrasound, p. 57. Edited by M.
van Holsbeeck & J. H. Introcaso. Mosby Year Book, St.
Louis 1991.
9. VAN HOLSBEECK M. & INTROCASO J. H.: Sonography of large
synovial joints. In: Musculoskeletal ultrasound, p. 143. Ed­
ited by M. van Holsbeeck & J. H. Introcaso. Mosby Year
Book, St. Louis 1991.
10. KAPLAN P. A., MATAMOROS A. JR & ANDERSON J. C.: Sono­
graphy of the musculoskeletal system. AJR 155 (1990),
237.
11. KOSKI J. Μ.: Ultrasonography of the subtalar and midtarsal
joints. J. Rheumatol. 20 (1993), 1753.
12. KOSKI J. M.: Detection of plantar tenosynovitis of the fore­
foot by ultrasound in patients with early arthritis. Scand. J.
Rheumatol. 24(1995), 312.
13. LEHTINEN A., TAAVITSAINEN M. & LEIRISALO-REPO M.: Sono-
Downloaded from acr.sagepub.com at CORNELL UNIV on August 6, 2015
graphic analysis of enthesopathy in the lower extremities of
patients with spondylarthropathy. Clin. Exp. Rheumatol. 12
(1994), 143.
14. MARTINOLI C , DERCHI L. E., PASTORINO C , BERTOLOTTO M.
& SILVESTRI E.: Analysis of echotexture of tendons with
US. Radiology 186 (1993), 839.
15. ÖSTERGAARD M . , GlDEON P., H E N R I K S E N O . & LoRENZEN I.:
Synovial volume - a marker of disease severity in rheuma­
toid arthritis? Quantification by MRI. Scand. J. Rheumatol.
23 (1994), 197.
16. SIMS R. E. & GENANT H. K.: Magnetic resonance imaging
of joint disease. Radiol. Clin. North Am. 24 (1986), 179.
17. VAINIO K.: The rheumatoid foot. A clinical study with path­
ological and roentgenological comments. Ann. Chir. Gy-
naecol. (1956), Suppl. 1, p. 45.
18. WEISS K. L., BELTRAN J. & LUBBERS L. M.: High-field M R
surface-coil imaging of the hand and wrist. Part II. Patho­
logic correlations and clinical relevance. Radiology 160
(1986), 147.
19. WEISS K. L., BELTRAN J., SHAMAM O. M., STILLA R. F. &
LEVEY M.: High-field MR surface-coil imaging of the hand
and wrist. Part I. Normal anatomy. Radiology 160 (1986),
143.
577