AJR 2004; 182:337-340
© American Roentgen Ray Society
Arthrofibrosis Associated with Total Knee Arthroplasty: Gray-Scale and Power Doppler Sonographic Findings
Jens G. Boldt1,2,
Urs K. Munzinger1,
Marco Zanetti3 and
Juerg Hodler3
1 Department of Orthopedic Surgery, Schulthess Clinic, Lenggstrasse 2, Zurich
CH-8008, Switzerland.
2 Present address: Department of Orthopedic Surgery, St. Vinzenz Hospital,
Schloss Strasse 85, Düsseldorf D-40477, Germany.
3 Department of Radiology, Orthopedic University Hospital Balgrist, Forchstrasse
340, Zürich CH-8008, Switzerland.
Received April 14, 2003;
accepted after revision August 12, 2003.
Address correspondence to J. Hodler.
Abstract
OBJECTIVE. The objective of this study was to determine gray-scale
and power Doppler sonographic findings in patients with arthrofibrosis
associated with total knee arthroplasty.
SUBJECTS AND METHODS. From a consecutive cohort of more than 3,000
mobilebearing total knee arthroplasties, 44 cases (1.5%) with arthrofibrosis
were identified, of which 38 were recruited for a clinical and sonographic
investigation. A control group of 38 patients with a well-functioning total
knee arthroplasty was matched. Synovial hypertrophy, presence of
neovascularity, patellar tendon thickness, and extent of effusion were
assessed.
RESULTS. Synovial membrane thickness was significantly (p
< 0.001) increased in the arthrofibrosis group (medial, 3.4 mm; lateral,
3.0 mm; suprapatellar, 3.1 mm) when compared with the control group (medial,
2.0 mm; lateral, 2.0 mm; suprapatellar, 1.9 mm). When a cutoff of 3.0 mm was
used, sonography had a sensitivity of 84% and a specificity of 82% for
detecting arthrofibrosis. Neovascularity (rated as grades 0–3) of the
synovial membrane and Hoffa's fat pad was significantly (p 
0.003) more pronounced in the arthrofibrosis group (medial, 1.1; lateral, 1.2;
suprapatellar, 1.0; Hoffa's fat pad, 1.1) than in the control group (medial,
0.1; lateral, 0.3; suprapatellar, 0.2; Hoffa's fat pad, 0.1). No significant
difference was seen between study groups with regard to the amount of joint
effusion at three locations and with regard to patellar tendon thickness.
CONCLUSION. Synovial membrane thickening and neovascularity are
characteristic sonographic findings for the diagnosis of arthrofibrosis
associated with total knee arthroplasty.
Introduction
Chronic pain and stiffness occurring after total knee arthroplasty are
frustrating for both the patient and the surgeon
[1]. Such symptoms are often
attributed to arthrofibrosis, a joint abnormality characterized by synovial
hypertrophy and capsular thickening. The differential diagnosis includes
infection, pigmented villonodular synovitis, reflex sympathetic dystrophy, and
complex regional pain syndrome. Arthrofibrosis has been treated with
physiotherapy; long-term peridural anesthetics; knee manipulation under
anesthesia; arthroscopic débridement; and open procedures, including
revision surgery with exchange of prosthetic components
[2]. The prevalence of
arthrofibrosis after mobile-bearing knee arthroplasty appears to be variable,
ranging from 1% to 17% [3]. To
our knowledge, the sonographic appearance of arthrofibrotic total knee
arthroplasty has not been described in the literature. The purpose of this
study was to determine the gray-scale and power Doppler sonographic findings
in patients with arthrofibrosis associated with total knee arthroplasty.
Subjects and Methods
Patients
The study population was recruited from a cohort of more than 3,000
mobile-bearing low contact stress (LCS knee, DePuy, Leeds, England) total knee
arthroplasties that had been performed in one orthopedic hospital since 1988.
The polyethylene bearing in this knee system allows free rotation without
translational mobility. The component design of the LCS prosthesis and the
surgical technique have remained unchanged during the past 24 years.
All patients with total knee arthroplasty performed at our department are
routinely followed up clinically and with standard radiography after 1 week, 6
weeks, 6 months, 1 year, 5 years, and whenever complications occur.
Forty-nine knees (1.6%) were diagnosed with arthrofibrosis on the basis of
the following criteria: reduced range of motion (maximal range of motion of
< 80°, an extension deficit of > 10°, or a total range of motion
of < 80°); knee pain with no known reasons; palpable synovial
thickening; an interval of more than 12 months after the last surgical
intervention (to exclude physiologic postoperative changes); and absence of
infection, hemarthrosis, and extraarticular abnormalities (including deep vein
thrombosis). Clinical and laboratory tests or joint aspiration was performed
whenever appropriate. This assessment was performed by two experienced senior
knee surgeons who each perform more than 250 knee arthroplasties per year.
Five of the 49 patients had died at the time of this investigation. Six
patients could not be motivated to attend the follow-up examinations for
personal reasons (n = 2), distance between home and the hospital
(n = 2), illness not relating to total knee arthroplasty (n
= 1), and revision surgery performed at another institution (n = 1).
Therefore, 38 (86%) of the surviving 44 total knee arthroplasties were
included in the study. The investigation was approved by the responsible
institutional review board and informed consent was obtained from all
patients.
The mean age of the patients with arthrofibrosis was 65 years (range,
49–76 years). The range of motion of the knees that underwent surgery
was 40–80° (mean, 70°). The 38 knees had been treated by one to
five manipulations (mean, 1.9 manipulations) with the patient under anesthesia
at the time of follow-up. At least one surgical revision had been performed in
36 (95%) of the 38 knees, including open débridement in 26 (72%),
arthroscopic débridement in 15 (42%), bearing exchange in six (17%),
tibial component revision in four (11%), resurfacing of a primarily
nonoperated patella in three (8%), repositioning of the tibial tuberosity in
three (8%), and patellar component removal in one (3%) knee.
Control Subjects
A control group of 38 individuals with well-functioning LCS knee
arthroplasties were matched with regard to age, sex, body mass index, and
follow-up period. Inclusion criteria for the control group were range of
motion of more than 100°, lack of peri- or postoperative complications,
and excellent or good clinical results according to a published clinical knee
arthroplasty score [4]. During
a period of 8 months, all patients who were scheduled for a routine follow-up
visit at least 1 year after surgery were invited to enter the control group
until the appropriate number of the matched group was obtained. None of the
eligible patients rejected study participation. The mean age of the control
patients was 67 years (range, 54–77 years). Fourteen patients were men
and 24 were women. The range of motion in the control group was
100–135° (mean, 115°). Age, sex, body mass index, diagnosis, and
type of prosthesis were not significantly different in the arthrofibrosis and
control groups.
Sonography
The patients were examined by one of two consultant musculoskeletal
radiologists who have worked at the same institution for 8 years; both have
more than 10 years of experience in sonography of the musculoskeletal systems.
Before the start of the study, they examined a few patients not included in
the investigation to make sure that the same imaging technique and image
criteria were used. The radiologists were unaware of the patients' symptoms.
The radiologists were asked not to talk to the patient about the knee during
the examination but rather about technical aspects of sonography. The patients
were placed in the supine position with the knees extended (90° flexion)
for examination of the patellar ligament. A 7.5-MHz linear transducer on an
Elegra scanner (Siemens Medical Solutions, Erlangen, Germany) was used.
Measurements were obtained during the examination with the electronic calipers
of the sonographic equipment. The output on the screen (to the nearest 1/10 of
a millimeter) was used for calculations. Such numbers probably overstate the
precision of sonographic measurements but do not introduce a bias in favor of
either patient group. For the power Doppler examinations, the color gain was
adjusted until no color signal was present deep in relation to the cortical
bone of the femoral condyle. The pulse repetition frequency was maintained at
1,250 Hz in all cases. Hard copies and digital (JPEG) images were obtained for
documentation.
Synovial hypertrophy (in millimeters), neovascularity of synovial membrane
and Hoffa's fat pad (grades 0–3), patellar tendon thickness (in
millimeters), and the extent of effusion (in millimeters squared) were
assessed. Synovial hypertrophy and neovascularity were separately assessed in
the suprapatellar, medial, and lateral parapatellar recesses. When synovial
thickness was variable in a specific compartment, the thickest part was
measured. The amount of effusion was measured in the same three compartments.
From the largest transverse (craniocaudal for the suprapatellar recess) and
anteroposterior diameters, a cross-sectional area was calculated with the 2D
ellipsoid formula (
x a x b / 4, with
a and b representing the two measured diameters). This is
only an estimation of the amount of effusion. However, because of the
complicated form of the synovial recesses, the true volume would have been
complicated to calculate, presumably without changing the message.
Neovascularity was graded as 0 for not detectable, 1 for minimal (one or two
vessels visible in the field of view), 2 for moderate (three or more vessels
in field of view, flow signal covering < 50% of the section even in the
most vascularized parts of the synovial membrane), and 3 for prominent (three
or more vessels in the field of view, flow signal covering > 50% of the
section in the most vascularized parts of the synovial membrane). The same
grading was used for vascularity of Hoffa's fat pad. The anteroposterior
diameter of the patellar ligament was measured at mid distance between the
patellar and tibial insertions.
Statistical Analysis
Continuous data in both groups were normally distributed (one-sample
Kolmogorov-Smirnov test), and an unpaired Student's t test was
performed. For the grading of neovascularity, Mann-Whitney tests were used.
Significance was set at the 5% level.
Results
Clinical Findings
The control group had a mean clinical score of 91.2 points (range,
81–100 points), and the arthrofibrosis group
(Fig. 1) had a mean clinical
score of 31.5 points (range, 15–51 points). Correcting osteotomy before
total knee arthroplasty was a significant (p < 0.05) risk factor
for developing arthrofibrosis in total knee arthroplasty with eight patients
(18%) in the arthrofibrosis group versus one (3%) in the control group. In
both groups, nine knees each had undergone arthroscopic meniscal surgery
before total knee arthroplasty. This potential risk factor, therefore, appears
not to be associated with arthrofibrosis. Other potential risk factors were
rarely encountered. Therefore, their importance cannot be determined on the
basis of our study. Three of 38 patients in the arthrofibrosis group had
poliomyelitis affecting the lower limbs compared with no individual in the
control group. An additional three patients in the arthrofibrosis group had
previously undergone open reduction and internal fixation of an intraarticular
fracture compared with no fracture in the control group.
View larger version (137K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1. —49-year-old man 28 months after total right knee
arthroplasty. Photograph shows typical clinical appearance of arthrofibrosis,
with diffuse swelling of knee. On examination, swelling is firm, not painful,
and not fluctuating.
|
|
Sonography
Results are presented in Table
1. Synovial thickness was significantly increased in the
arthrofibrosis group (Fig. 2)
in comparison with the control group (Fig.
3). When a cutoff of 3.0 mm was used, sonography had a sensitivity
of 84% and a specificity of 82% for detecting arthrofibrosis. Synovial
membrane and Hoffa's fat pad hypervascularity were significantly more
pronounced in the arthrofibrosis group than in the control group. Thirty-three
knees had a vascularity of grade 2 or more
(Fig. 2) in at least two
regions versus three in the control group
(Fig. 3). Differences with
regard to joint effusion and patellar tendon thickness were not significant. A
significantly (p < 0.04) increased synovial thickness (mean, 3.5
mm) was seen in patients who had previously been treated with open
débridement compared with those with previous arthroscopic
débridement (mean, 2.9 mm). However, neovascularity of the synovial
membrane was not significantly different for these two groups.
View larger version (122K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2. —58-year-old woman 26 months after total right knee
arthroplasty. Sonography shows increased synovial hypertrophy associated with
arthrofibrosis. Transverse power Doppler sonogram through medial parapatellar
recess of right knee shows prominent (grade 3) neovascularity (curved
arrow) in ventral and medial parts of thickened synovial membrane
(between straight arrows). F = femoral condyle.
|
|
View larger version (133K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3. —69-year-old man in control group who has well-functioning
knee replacement 25 months after total right knee arthroplasty. Transverse
sonogram through medial parapatellar recess of right knee shows normal
thickness of synovial membrane (arrow points to anterior synovial
membrane, calipers measure anterior [x – x] and posterior [+
– +] synovial membranes). F = femoral condyle.
|
|
Discussion
Arthrofibrosis is a serious complication of total knee arthroplasty that
requires revision surgery in as many as 17% of cases
[3]. Arthrofibrosis is
clinically characterized by serious functional impairment, pain, and
considerable patient dissatisfaction. Arthrofibrosis responds poorly to
treatment, which may include physiotherapy; long-term peridural anesthesia
( 2 weeks); closed manipulation; arthroscopic débridement; and open
procedures, including revision surgery with exchange of prosthetic components.
The etiology of arthrofibrosis is a subject of debate. Biochemical and
immunologic factors, genetically determined predisposition, reflex sympathetic
dystrophy, complex regional pain syndrome, metal-related allergy, lowgrade
infection, and mechanical factors have been discussed
[1,
5–8].
Histopathologic analysis has shown massive connective tissue proliferation
with deposition of disordered matrix proteins and increased expression of type
VI collagen in the subsynovia as well as around capillary walls
[6–8].
According to our data, sonography can differentiate total knee arthroplasty
patients with arthrofibrosis from those with an uncomplicated postoperative
course on the basis of synovial membrane thickness and neovascularity.
Therefore, sonography may be useful in the assessment of the postoperative
knee. Previously published papers have not focused on the specific diagnosis
of arthrofibrosis but support the potential role of postoperative sonography.
Grobbelaar et al. [9] have
emphasized both the economic and diagnostic potential of power Doppler
sonography in assessing synovial structures. Giovagnorio et al.
[10] have shown that power
Doppler sonography of the knee is useful for identification of synovial
thickening and hypervascularity and that it differentiates joint capsule from
synovial membrane and effusion. Sonography has also been used for the
qualitative assessment, grading, and follow-up of a number of synovial and
other joint abnormalities
[11–19].
We acknowledge that the precision and reproducibility of the sonographic
measurements and gradings have not been assessed. We believe, however, that
any errors would be identical for both groups of patients and would not change
our conclusions. At the institution of the orthopedic coauthors, sonography is
used for surgical planning in patients with arthrofibrosis. The amount,
diameter, and localization of hypertrophic tissues in the knee are relevant
when deciding whether to perform open versus arthroscopic surgery. In
addition, because synovial and capsular tissue cannot easily be differentiated
intraoperatively, sonography assists in tissue-sparing synovectomy, thereby
avoiding injury to capsular tissue.
In conclusion, synovial membrane thickening and neovascularity are
characteristic sonographic findings for the diagnosis of arthrofibrosis
associated with total knee arthroplasty.
References
- Ries MD, Badalamente M. Arthrofibrosis after total knee
arthroplasty. Clin Orthop2000; 380:177
–183
- Lindenfeld TN, Wojtys EM, Husain A. Surgical treatment of
arthrofibrosis of the knee. Instr Course Lect2000; 49:211
–221[Medline]
- Kaper BP, Smith PN, Bourne RB, Rorabeck CH, Robertson D.
Medium-term results of a mobile bearing total knee replacement.
Clin Orthop1999; 367:201
–209
- Buechel FF. A simplified evaluation system for the rating of the
knee function. Orthop Rev1982; 11:97
–101
- Dennis DA, Komistek RD, Walker SA, Cheal EJ, Stiehl JB. Femoral
condylar lift-off in vivo in total knee arthroplasty. J Bone Joint
Surg Br 2001;83:33
–39
- Zeichen J, van Griensven M, Albers I, Lobenhoffer P, Bosch U.
Immunohistochemical localization of collagen VI in arthrofibrosis.
Arch Orthop Trauma Surg1999; 119:315
–318
- Delcogliano A, Franzese S, Branca A, Magi M, Fabbriciani C. Light
and scan electron microscopic analysis of cyclops syndrome: etiopathogenic
hypothesis and technical solutions. Knee Surg Sports Traumatol
Arthrosc 1996;4:194
–199[Medline]
- Romanos G, Schroter-Kermani C, Hinz N, Bernimoulin JP.
Immunohistochemical distribution of the collagen types IV, V, VI and
glycoprotein laminin in the healthy rat, marmoset (Callithrix
jacchus) and human gingivae. Matrix1991; 11:125
–132[Medline]
- Grobbelaar N, Bouffard JA. Ultrasonography of the knee: a pictorial
review. Semin Ultrasound CT MR2000; 21:231
–274[Medline]
- Giovagnorio F, Martinoli C, Coari G. Power Doppler ultrasonography
in knee arthritis: a pilot study. Rheumatol Int2001; 20:101
–104[Medline]
- Walther M, Harms H, Krenn V, Radke S, Faehndrich TP, Gohlke F.
Correlation of power Doppler ultrasonography with vascularity of the synovial
tissue of the knee joint in patients with osteoarthritis and rheumatoid
arthritis. Arthritis Rheum2001; 44:331
–338[Medline]
- Rasmussen OS. Ultrasonography of tendons. Scand J Med
Sci Sports 2000;10:360
–364[Medline]
- Schmidt WA, Volker L, Zacher J, Schlafke M, Ruhnke M, Gromnica-Ihle
E. Colour Doppler ultrasonography to detect pannus in knee joint synovitis.
Clin Exp Rheumatol2000; 18:439
–444[Medline]
- Troltzsch M. Color Doppler study in patients with rheumatoid
arthritis and scleroderma [in German]. Z Rheumatol1994; 53:2
–6[Medline]
- Gratz S, Gobel D, Behr TM, Herrmann A, Becker W. Correlation
between radiation dose, synovial thickness, and efficacy of
radiosynoviorthesis. J Rheumatol1999; 26:1242
–1249[Medline]
- Gray SD, Kaplan PA, Dussault RG. Imaging of the knee: current
status. Orthop Clin North Am1997; 28:643
–658[Medline]
- Starok M, Lenchik L, Trudell D, Resnick D. Normal patellar
retinaculum: MR and sonographic imaging with cadaveric correlation.
AJR 1997;168:1493
–1499[Abstract/Free Full Text]
- Fiocco U, Cozzi L, Rubaltelli L, et al. Long-term ultrasonographic
follow-up of rheumatoid and psoriatic proliferative knee joint synovitis.
J Rheumatol1996; 35:155
–163
- Rubaltelli L, Fiocco U, Cozzi L, et al. Prospective
ultrasonographic and arthroscopic evaluation of proliferative knee joint
synovitis. J Ultrasound Med1994; 13:855
–862[Abstract]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
