Abstract
Background: There is a lack of data on the functional effect of open
hip synovectomy in a large number of patients with juvenile rheumatoid
arthritis evaluated with a validated assessment tool.
Methods: Between 1985 and 1997, sixty-seven open hip-joint
synovectomies were carried out in fifty-six patients with juvenile rheumatoid
arthritis. Fifty-five hips (82%) had radiographic changes that were stage III
or higher according to the system of Larsen et al. Hip function was evaluated
preoperatively and after a mean of fifty months with the Merle
d'Aubigné hip score.
Results: Sixty-five (97%) of the sixty-seven hips were available for
follow-up. The mean total Merle d'Aubigné hip score (and standard error
of the mean) was significantly improved from 9.5 ± 2.5 points at
baseline to 16.3 ± 1.0 points at the time of follow-up (p < 0.001).
The individual scores for pain, mobility, and walking ability were
significantly increased as well (all p < 0.001). Eighty-five percent of the
hips were observed to have a very great or great improvement in function. A
concomitant soft-tissue release was performed in seven hips, and nine hips
required surgical dislocation. Surgical complications included two superficial
wound hematomas that did not require intervention; osteonecrosis of the
femoral head was not observed. Five hips required total hip arthroplasty
during the follow-up period. Thus, the survival rate for the hips was 94% at a
mean of four years following the synovectomy.
Conclusions: Open hip synovectomy in patients with juvenile
rheumatoid arthritis is a safe procedure that can improve hip-joint function
for up to five years.
Level of Evidence: Therapeutic Level IV. See Instructions
to Authors for a complete description of levels of evidence.
Hip-joint involvement is a frequent cause of walking disability in patients
with juvenile rheumatoid arthritis. Clinical characteristics of the rheumatoid
hip are pain, a limited range of motion, and rapid-onset cartilage destruction
resulting in secondary changes in other lower-limb joints and the spine.
Nonoperative treatment options include physical therapy, anti-inflammatory
medication, and local or systemic corticosteroid
therapy1. Surgical
treatment may consist of isolated soft-tissue release such as tenotomy,
capsulotomy, and/or open hip-joint synovectomy.
The outcome of open hip-joint synovectomy has not been investigated
systematically, to our knowledge. In 1990, Ovregard et al. reported on 514
synovectomies in children, but they provided little information regarding the
surgical approach or perioperative care associated with the twelve hip-joint
synovectomies done in that
series2. In their
article on forty-one synovectomies, including ten in the hip, Jacobsen et al.
expressed doubt that there were any benefits from the
operation3. Heimkes
and Stotz noted reduced pain and an improved range of motion two years after
synovectomy in six
hips4. In 1974,
Schwagerl reported reduced pain and enhanced walking ability in a series in
which twenty-one hip-joint synovectomies had been performed in combination
with soft-tissue releases; however, he did not perform a standardized
evaluation of hip-joint
function5. In
summary, we are aware of no prior report on hip synovectomies in a large
number of patients with juvenile rheumatoid arthritis that included a
postoperative standardized functional evaluation.
The aim of our study was to evaluate the mid-term results of hip-joint
synovectomy in patients with juvenile rheumatoid arthritis. Our hypothesis was
that synovectomy improves hip-joint function in these patients.
From 1985 to 1997, sixty-seven open hip-joint synovectomies in fifty-six
patients with juvenile rheumatoid arthritis were done at the Department of
Orthopedic and Trauma Surgery, Wichernhaus, Rummelsberg, Schwarzenbruck,
Germany. Eleven patients had bilateral surgery. The patient subgroups are
presented in Figure 1. We
operated on thirty-five female patients with a mean age of 13.7 years (range,
nine to twenty-two years) and twenty-one male patients with a mean age of 16.3
years (range, twelve to twenty-one years). The mean duration of the disease
was 6.0 years (range, 178 days to twenty years). None of the patients had had
previous surgery on the affected hip. All patients had received individually
optimized medical therapy consisting of nonsteroidal anti-inflammatory drugs,
systemic corticosteroids, and disease-modifying medication (methotrexate,
sulfasalazine, and/or biologicals). All patients had active hip synovitis that
could not be controlled medically and some degree of joint destruction.
Synovectomy was recommended on the basis of persistent synovitis of the hip
demonstrated on ultrasound investigation, pain, joint effusion, and a limited
range of motion of the hip. Weight-bearing anteroposterior radiographs of the
affected hip were made both before the surgery and at the time of follow-up
and were used to grade arthritic changes of the hip joint with a
semiquantitative ordinal score based on standard reference radiographs, as
described by Larsen et
al.6. The Larsen
grading system ranges from 0 (absence of arthritic changes) to V (mutilating
joint destruction).
Surgical Technique
The surgery was carried out with the patient under general anesthesia. The
patient was positioned supine, and the lower extremities were draped free
during the procedure. A Smith-Petersen anterior iliofemoral hip approach was
used to obtain adequate exposure and occasionally allow dislocation of the
hip. The tensor fasciae latae and the gluteus medius and minimus muscles were
detached subperiosteally and were reflected distally on the outer surface of
the ilium. The straight and reflected heads of the rectus femoris muscle were
detached.
A T-shaped capsular incision was performed with one limb located anterior
and parallel to the acetabular rim and the second located parallel to the
femoral neck. The hypertrophic synovium was separated from the anterior aspect
of the capsule and was resected first with the hip in external rotation and
then with it in internal rotation (Fig.
2-A). The posterior aspect of the synovium was visualized by
flexing and rotating the hip and applying gentle traction, allowing the
acetabular fossa to come into view. The hip was dislocated when there was
markedly hypertrophic synovial tissue involving the acetabular fossa that was
not readily accessible otherwise. Dislocation was carried out by flexing,
externally rotating, and adducting the hip. The ligament teres was resected
during dislocation, and the synovectomy was completed
(Fig. 2-B). Extreme positions
were avoided to minimize compression of the medial femoral circumflex artery.
Femoral head cysts were thoroughly curetted and débrided to remove
synovial tissue. The acetabular labrum as well as the posterior aspect of the
capsule was left intact. After completion of the synovectomy, the femoral head
was reduced and the capsule was closed. A suction drain was inserted. The
tensor fasciae latae and the gluteus medius and minimus muscles were
reattached to the iliac crest, and the edges of the sartorius and tensor
fasciae latae were sutured to the rectus femoris. The subcutaneous tissue was
reapproximated, and the skin was closed.
Soft-Tissue Release
If a flexion contracture of >20° was present when the hip was
initially examined with the patient under general anesthesia, a soft-tissue
release was performed. The reflected head of the rectus femoris was reattached
at the anterior inferior iliac spine with the hip in full extension, and the
straight head of the rectus femoris was not reattached. If necessary, the
iliopsoas was lengthened at the lesser trochanter.
Postoperative Clinical Care
Postoperatively, the hips were positioned in full extension and neutral
rotation until the suction drain was removed two days after the surgery. Both
the patients and the parents were instructed by a physical therapist to begin
active and passive exercises immediately after removal of the drain. In order
to allow immediate mobilization, an individually optimized analgesic regimen
consisting of paracetamol, nonsteroidal anti-inflammatory drugs, and opiates
was administered. Neither cast immobilization nor traction was used. Full
weight-bearing was allowed, initially with crutches. Heparin was given, until
full mobilization, to patients who were more than twelve years of age and had
a body weight of >90 lb (>40.8 kg).
Hip function was assessed, with the scoring method described by Merle
d'Aubigné et
al.7, preoperatively
and after a mean duration of follow-up of fifty months (range, twenty-seven to
sixty-six months). This scoring method includes evaluation of pain, the
ability to walk, and mobility, with each parameter scored on a scale ranging
from 0 points (greatest limitation) to 6 points (absence of limitation).
Summing of the points for pain and walking ability allows the estimation of
hip function, and the difference between the preoperative and postoperative
functional states reflects the functional improvement achieved by the surgical
procedure (see Appendix). All of the patients were operated on by the same
surgeon (A.S.) and were assessed by that surgeon preoperatively and at the
time of follow-up. Patients were screened for osteonecrosis of the femoral
head by radiographic evaluation at the time of follow-up.
The local ethics committee approved the conduct of the study. Written
informed consent for us to analyze the records, which remained anonymous, was
obtained preoperatively from the patients and/or at least one of the
parents.
For statistical analyses, data were expressed as the mean and standard
error of the mean. Data from paired samples (preoperative and postoperative
Merle d'Aubigné scores) were subjected to the multigroup Friedman test
and, subsequently, to the nonparametric Wilcoxon test with use of the SPSS
program (version 11.0; SPSS, Chicago, Illinois). P values of <0.01 were
considered significant. The Kaplan-Meier survival curve was calculated from
the individual follow-up data for each patient, with use of total hip
replacement as the end point. We used GraphPad Prism 5 (GraphPad Software, San
Diego, California) for this calculation.
Before the surgery, the arthritis was rated as Larsen stage III or worse in
fifty-five hips (82%) and as Larsen stage I or II in twelve (18%). At the time
of follow-up, eleven hips showed evidence of worsening radiographically, with
four hips changing from Larsen stage I to II; three hips, from Larsen stage II
to III; two hips, from Larsen stage III to IV; and two hips, from Larsen stage
IV to V. With the numbers available, no significant difference could be
observed between the subgroup with mild radiographic changes (Larsen stage I
or II) and the subgroup with worse radiographic changes (Larsen stage III, IV,
or V) with regard to progression of the disease.
The mean blood loss was approximately 300 mL, and the surgery usually
required less than sixty minutes. During the open hip synovectomy, seven of
the sixty-seven hips were treated with a concomitant soft-tissue release (of
the rectus femoris muscle in five and of the iliopsoas muscle in two). Nine of
the sixty-seven hips had to be dislocated to remove extensive synovitis in the
acetabular fossa. No superficial or deep wound infections, neural injuries, or
cases of femoral head necrosis were noted radiographically at the time of
follow-up. A superficial hematoma, which did not require further treatment,
developed in two patients.
Sixty-five (97%) of the sixty-seven hips were followed for a mean of fifty
months (range, twenty-seven to sixty-six months). Two children had moved
abroad. Of the original sixty-seven hips, five hips (7%) in five patients
required total hip arthroplasty after a median of thirty-nine months (range,
twenty-seven to forty-four months). Four of these five hips had had Larsen
stage-III or IV arthritis at the time of the synovectomy. According to the
Kaplan-Meier analysis, with total hip arthroplasty as the end point, the
survival of the operatively treated hip joints was 97% at three years after
the surgery, 94% at four years, and 87% at five years
(Fig. 3).
The mean preoperative Merle d'Aubigné hip score was 9.5 ± 2.5
points. The mean values for pain, mobility, and walking ability were 2.0
± 1.3 points, 4.4 ± 1.0 points, and 3.0 ± 1.1 points,
respectively. At the time of final follow-up, the mean total Merle
d'Aubigné score was significantly improved to 16.3 ± 1.0 points
(p < 0.001). Moreover, the scores for pain, mobility, and walking ability
were significantly improved to 5.7 ± 0.4 points, 5.4 ± 0.5
points, and 5.0 ± 0.7 points, respectively (all p < 0.001)
(Table I). Synovectomy led to a
significant improvement in hip function in both patients with early-stage
disease (Larsen stage 0, I, or II) and those with late-stage disease (Larsen
stage III, IV, or V).
There was a very great functional improvement, from the preoperative to the
postoperative status, in thirty-seven cases (57%), a great improvement in
eighteen (28%), a fair improvement in eight (12%), and a failure in two (3%).
Taken together, a very great or great improvement was observed in fifty-five
hips (85%) (Fig. 4).
This study demonstrates a clear-cut functional benefit of hip-joint
synovectomy in patients with juvenile rheumatoid arthritis, supporting
previous publications advocating synovectomies in general in patients with
juvenile rheumatoid
arthritis2,4,8,9.
Since persistent synovitis accelerates joint destruction, an isolated
soft-tissue release of a hip-joint contracture does not seem appropriate,
although this approach has been proposed
previously10. In
1985, Jacobsen et al. found no improvement in terms of pain relief or the
range of motion after late synovectomy in patients with juvenile rheumatoid
arthritis3. Their
study comprised ten hips and thirty-one knees. The mean duration of follow-up
of their patients who had a hip synovectomy was 5.4 years, which is comparable
with that in our study, but that report did not provide any detail on the
surgical procedure or the perioperative care.
We chose to perform the synovectomy through the anterior iliofemoral
surgical approach, which is advantageous when concomitant soft-tissue release
is required. This approach has been proven to be safer than the posterior
approach with regard to preservation of the blood supply to the femoral
head11. In our
patients, exposure of the acetabulum with the anterior approach was usually
sufficient to perform the synovectomy, and extensive detachment of the gluteus
medius was not needed. Dislocation was indicated in nine hips in which
contracted soft tissues did not allow enough exposure to perform the
synovectomy in the acetabular fossa. Dislocation was carried out, following
anterior capsulotomy, by hip flexion, external rotation, and adduction. Ganz
et al.12 proposed
the use of a surgical dislocation procedure through a lateral approach,
including the trochanteric flip osteotomy technique, in an adult hip. With the
anterior approach, a trochanteric osteotomy to protect the deep branch of the
medial femoral circumflex artery is not necessary, so the risks of nonunion or
heterotopic ossification can be avoided. To prevent osteonecrosis of the
femoral head, the posterior aspect of the capsule is kept intact in order to
preserve its blood supply through the intracapsular branch of the medial
femoral circumflex artery. In addition, we avoided extreme positioning of the
hip when we dislocated it, in order to preserve femoral head perfusion.
Avoiding these pitfalls may be sufficient to avoid causing osteonecrosis of
the femoral head, as we did not observe osteonecrosis in our nine patients
treated with surgical dislocation.
In the current literature, hip arthroscopy has been suggested for joint
débridement in patients with
osteoarthritis13,
for treatment of acetabular labral
tears14, or as a
therapeutic option for septic arthritis of the
hip15,16.
Hip arthroscopy has been proposed as a diagnostic technique for patients with
juvenile rheumatoid
arthritis17, but it
has not been proven effective for hip-joint synovectomy. We believe that an
open anterior approach is more appropriate because concomitant soft-tissue
release or hip-joint dislocation can be performed.
In the knees of adult patients with rheumatoid arthritis, the combination
of arthroscopic synovectomy and radiation synovectomy has proven to be
superior to either single
method18. This may
be due to the fact that the quantity of synovial tissue that is removable with
arthroscopic knee joint synovectomy varies among different anatomical regions
of the knee
joint19, leaving
residual inflamed tissue as a possible source of recurring synovitis. To our
knowledge, there are no reports on the combination of surgical synovectomy and
radiation synovectomy in patients with juvenile rheumatoid arthritis, perhaps
for the following reasons. First, radiation therapy should be restricted to
adults because of its possible adverse effects in growing children. Second,
hip-joint synovectomy is performed as an open procedure, and all aspects of
the synovium are readily accessible with the anterior approach. Therefore, we
do not recommend radiation synovectomy after open hip-joint synovectomy in
patients with juvenile rheumatoid arthritis.
We believe that hip-joint synovectomy is effective, even in patients with
late-stage articular damage. A reduction in pain has been observed following
late-stage synovectomy in adult
hips4,
wrists20, and
knees21 as well as
after knee synovectomy in patients with late-stage juvenile rheumatoid
arthritis22. Those
findings suggest that this procedure can be effective as a late option to
reduce pain in severely affected arthritic joints. Given that pain and limited
motion persist without synovectomy, our procedure can be considered superior
to the natural course of the disease in terms of pain and mobility, although
our data do not clearly prove this. A limitation of our work was that we were
unable to compare our radiographic findings with those documenting the natural
course of the disease, and there are no data available with which to compare
the late clinical results of synovectomy in patients who have juvenile
rheumatoid arthritis with the natural course of this disease.
Even if a late synovectomy can lead to improvement, the goal still should
be to perform the synovectomy early, when there is no or mild joint
destruction. A team of pediatric rheumatologists and orthopaedic surgeons
should monitor the clinical symptoms of pain, effusion, and limited motion
carefully in order not to miss the best opportunity for early synovectomy. In
our opinion, synovectomy can provide its "joint-preserving" effect
only in the early stages of rheumatic diseases. We are not aware of any
clinical studies comparing the effect of early synovectomy with the natural
course of joint degeneration or untreated disease. Such investigations should
include the need for total hip arthroplasty as a primary study end point.
Because the hip survival rates after total hip arthroplasty are still limited
in young patients, a delay in joint replacement achieved by hip synovectomy
would be most important for patients with juvenile rheumatoid arthritis.
Many of our patients had already exceeded the point where they were
candidates for early synovectomy—i.e., they had late-stage joint
degeneration. Although the medical treatment of juvenile rheumatoid arthritis
has changed markedly over the last decade, many patients are still not treated
by pediatric rheumatologists and therefore may not receive the best possible
medication. Furthermore, long-term remission is not regularly achieved in
patients with juvenile rheumatoid arthritis in spite of these medical
advances, so progressive joint damage occurs despite competent medical
treatment1.
The general benefit of synovectomy in patients with juvenile rheumatoid
arthritis has been
doubted3, but our
data support the need for hip synovectomy as a powerful tool to reduce pain
and improve hip function even in patients with distinct joint damage. These
findings are in line with those of several studies on the effect of
synovectomy in patients with rheumatoid arthritis of the
knee21,22,
the glenohumeral
joint23, and the
elbow24. There is
also some previous evidence that synovectomy performed when joint destruction
is in its early stages has a better functional outcome than late-stage
synovectomy18,25.
However, synovectomy has not yet been proven to have a real joint-preserving
effect in patients with rheumatic diseases, and joint deterioration may
progress in spite of the
synovectomy3,25,26.
Anteroposterior radiographs were available for all sixty-five hips both
preoperatively and at the time of follow-up. Progressive joint-space narrowing
was seen in eleven hips (17%). With the numbers available, no significant
difference could be observed between the subgroups with Larsen stage-I or II
radiographic changes and those with Larsen stage-III, IV, or V changes. In
general, our radiographic data indicate that joint deterioration may progress
after hip synovectomy, but the clinical improvement does not seem to correlate
with the radiographic progression. However, our midterm data do not allow us
to estimate if the clinical improvement is stable over time or if radiographic
changes may precede a loss of hip function. Nonetheless, the pain reduction
and clear-cut improvement in hip function, as demonstrated in this midterm
study, substantially improve the quality of life for patients with juvenile
rheumatoid arthritis.
In summary, patients with juvenile rheumatoid arthritis with persistent
symptoms of hip synovitis should be considered to be candidates for early or
late open synovectomy. In our experience, this procedure has been beneficial
in terms of pain relief and hip function.
A table showing the details of the Merle d'Aubigné scoring method is
available with the electronic versions of this article, on our web site at
(go to
the article citation and click on "Supplementary Material") and on
our quarterly CD-ROM (call our subscription department, at 781-449-9780, to
order the CD-ROM). ?
Hashkes PJ, Laxer RM. Medical treatment
of juvenile idiopathic arthritis. JAMA.
2005;294:
1671-84.2941671
2005
[PubMed][CrossRef]
Ovregard T, Hoyeraal HM, Pahle JA,
Larsen S. A three-year retrospective study of synovectomies in children.
Clin Orthop Relat Res.
1990;259:
76-82.25976
1990
[PubMed]
Jacobsen ST, Levinson JE, Crawford AH.
Late results of synovectomy in juvenile rheumatoid arthritis. J Bone
Joint Surg Am. 1985;67:
8-15.678
1985
Heimkes B, Stotz S. [Results of late
synovectomy of the hip in juvenile chronic arthritis]. Z
Rheumatol. 1992;51:
132-5. German.51132
1992
Schwagerl W. [Synovectomy and myotomy at
the hip-joint in rheumatoid arthritis]. Z Orthop Ihre Grenzgeb.
1974;112:
1210-8.1121210
1974
[PubMed]
Larsen A, Dale K, Eek M. Radiographic
evaluation of rheumatoid arthritis and related conditions by standard
reference films. Acta Radiol Diagn (Stockh).
1977;18:
481-91.18481
1977
[PubMed]
Merle d'Aubigné R, Cauchoix J,
Ramadier JV. Evaluation chiffrée de la fonction de la hanche.
Application àl'étude des résultats des opérations
mobilisatrices de la hanche. Rev Chir Orthop.
1949;35:
541-8.35541
1949
Rydholm U, Elborgh R, Ranstam J,
Schroder A, Svantesson H, Lidgren L. Synovectomy of the knee in juvenile
chronic arthritis. A retrospective, consecutive follow-up study. J Bone
Joint Surg Br. 1986;68:
223-8.68223
1986
Maenpaa H, Kuusela P, Lehtinen J,
Savolainen A, Kautiainen H, Belt E. Elbow synovectomy on patients with
juvenile rheumatoid arthritis. Clin Orthop Relat Res.
2003;412:
65-70.41265
2003
[PubMed][CrossRef]
Swann M, Ansell BM. Soft-tissue release
of the hips in children with juvenile chronic arthritis. J Bone Joint
Surg Br. 1986;68:
404-8.68404
1986
Swiontkowski MF, Thorpe M, Seiler JG,
Hansen ST. Operative management of displaced femoral head fractures:
case-matched comparison of anterior versus posterior approaches for Pipkin I
and Pipkin II fractures. J Orthop Trauma.
1992;6:
437-42.6437
1992
[PubMed][CrossRef]
Ganz R, Gill TJ, Gautier E, Ganz K,
Krügel N, Berlemann U. Surgical dislocation of the adult hip: a technique
with full access to the femoral head and acetabulum without the risk of
avascular necrosis. J Bone Joint Surg Br.
2001;83:
1119-24.831119
2001
[PubMed][CrossRef]
Sozen YV, Hepgur G, Kilicoglu O,
Yazicioglu O. [The effectiveness of arthroscopic debridement and lavage
treatment in osteoarthritis of the hip: preliminary results]. Acta
Orthop Traumatol Turc. 2004;38:
96-103. Turkish.3896
2004
Santori N, Villar RN. Acetabular labral
tears: result of arthroscopic partial limbectomy. Arthroscopy.
2000;16:
11-5.1611
2000
[PubMed][CrossRef]
Kim SJ, Choi NH, Ko SH, Linton JA, Park
HW. Arthroscopic treatment of septic arthritis of the hip. Clin Orthop
Relat Res. 2003;407:
211-4.407211
2003
[CrossRef]
Bould M, Edwards D, Villar RN.
Arthroscopic diagnosis and treatment of septic arthritis of the hip joint.
Arthroscopy. 1993;9:
707-8.9707
1993
[PubMed][CrossRef]
Holgersson S, Brattström H,
Mogensen B, Lidgren L. Arthroscopy of the hip in juvenile chronic arthritis.
J Pediatr Orthop. 1981;1:
273-8.1273
1981
[PubMed][CrossRef]
Klug S, Wittmann G, Weseloh G.
Arthroscopic synovectomy of the knee joint in early cases of rheumatoid
arthritis: follow-up results of a multicenter study.
Arthroscopy. 2000;16:
262-7.16262
2000
[PubMed][CrossRef]
Carl HD, Klug S, Seitz J, Swoboda B,
Kinne RW. Site-specific intraoperative efficacy of arthroscopic knee joint
synovectomy in rheumatoid arthritis. Arthroscopy.
2005;21:
1209-18.211209
2005
[PubMed][CrossRef]
Hanff G, Sollerman C, Elborgh R,
Pettersson H. Wrist synovectomy in juvenile chronic arthritis (JCA).
Scand J Rheumatol. 1990;19:
280-4.19280
1990
[PubMed][CrossRef]
Jüsten HP, Pommer S, Leeb I,
Pilhofer C, Kosslinger E, Wessinghage D. [Long-term results of open knee
synovectomy in later cases of rheumatoid arthritis]. Z
Rheumatol. 1999;58:
201-6. German.58201
1999
[CrossRef]
Häfner R, Pieper M. [Arthroscopic
synovectomy of the knee joint in chronic juvenile arthritis]. Z
Rheumatol. 1995;54:
165-70. German.54165
1995
Smith AM, Sperling JW, O'Driscoll SW,
Cofield RH. Arthroscopic shoulder synovectomy in patients with rheumatoid
arthritis. Arthroscopy.
2006;22:
50-6.2250
2006
[PubMed][CrossRef]
Nemoto K, Arino H, Yoshihara Y, Fujikawa
K. Arthroscopic synovectomy for the rheumatoid elbow: a short-term outcome.
J Shoulder Elbow Surg.
2004;13:
652-5.13652
2004
[PubMed][CrossRef]
Kerschbaumer F, Kandziora F, Herresthal
J, Hertel A, Hor G. [Combined arthroscopic and radiation synovectomy in
rheumatoid arthritis]. Orthopäde.
1998;27: 188-96.
German.27188
1998
[PubMed]
Paus AC, Dale K. Arthroscopic and
radiographic examination of patients with juvenile rheumatoid arthritis before
and after open synovectomy of the knee joint. A prospective study with a
5-year follow-up. Ann Chir Gynaecol.
1993;82:
55-61.8255
1993
[PubMed]