total knee arthroplasty in juvenile ...isbn 952-91-7512-4 (paperback) isbn 952-10-1959-x (pdf)...
TRANSCRIPT
Rheumatism Foundation Hospital, Heinola, Finland
TOTAL KNEE ARTHROPLASTY IN JUVENILE RHEUMATOID ARTHRITIS
A clinical and radiological study
CHRISTER LYBÄCK
ACADEMIC DISSERTATION
To be presented, with the permission of the Faculty of Medicine of the University of Helsinki, for public discussion in the auditorium of the Töölö
Hospital, Helsinki University Central Hospital, on September 29th, 2004, at 12 o`clock noon
University of Helsinki
Helsinki 2004
Supervised by Docent Eero Belt MD, PhD University of Tampere, Finland
and
Docent Matti Lehto MD, PhD University of Tampere, Finland
Reviewed by Docent Hannu Miettinen MD, PhD University of Kuopio, Finland
and
Docent Jorma Pajamäki MD, PhD University of Tampere, Finland
Opponent Professor (h.c.) Hannu Pätiälä MD, PhD University of Helsinki, Finland
Cover design by painter Kimmo Tolin
ISBN 952-91-7512-4 (paperback) ISBN 952-10-1959-X (PDF)
Helsinki 2004 Yliopistopaino
This work is dedicated to the memory of my father Sven and to Alcoholics Anonymous
5
ABSTRACT
The aim of the study was to analyse the results of 77 total knee arthroplasties (TKA) in 52
patients with juvenile rheumatoid arthritis (JRA) using the non-constrained AGC prosthesis.
Patients were operated on between the years 1985 and 1995. The mean age at the onset of
disease was 7.8 years and at the time of surgery 32.7 years. Follow-ups were conducted 3
months, 1, 4, and 8 years after the operation. Additionally, an interview was arranged at the
end of 1998, a mean of 7.3 (3-13) years after surgery.
Secondary amyloidosis (SA) was documented in 14 (11 women) of these 52 patients. All
procedures were performed using the classical method and the anteromedial parapatellar
approach. The patella was resurfaced for 18 of 52 patients [in 23 of 77 knees (30%)]. All-
polyethylene components were used in three knees (13%) and metal-backed components in
20 knees (87%). During 1967-1991 a total of 37 knee synovectomies (30 knees) were
performed on 23 out of these 52 patients [mean age was 20 (range 4 – 42) years at the time of
synovectomy]. A total of 13 temporary epiphyseal staplings of the knee due to valgus
deformity (12 knees) or leg length discrepancy (1 knee) had been performed on 9 out of the
52 patients in 1965-1982 [mean age was 8 (range 4 – 16) years at the time of epiphyseal
stapling]. In addition to TKA, 65 hip arthroplasties were performed on 36 patients, 13 elbow
arthroplasties on 9 patients, 8 shoulder arthroplasties on 7 patients, and 12
metacarpophalangeal implants were installed into 8 patients.
When revision surgery or death of the patient were the end points, the overall survival of
TKA was 99% (95% confidence interval, 92 - 100) during the 3-13 (mean 7.3) years of
follow-up. Of the 71 knees, 53 (75%) were subjectively excellent, 18 (25%) fair and none
was poor. There were no deep infections during the follow-up. Two patients died during the
follow-up, the cause of death was unknown in one and traumatic in the other. One knee with
a preoperative valgus angulation of 33º was revised four years after implantation because of
loosening of both components.
The presence of SA did not impair the survival of knee arthroplasty. The duration of the
disease and the age of patients with SA at the time of surgery did not differ significantly from
those without SA and there were no differences regarding patient satisfaction or functional
ability. Thus, the results of TKA in the patients with JRA + SA were excellent.
6
Anterior knee pain was present in 14 of 30 patients (47%) who had an unreplaced patella and
in two of 18 patients (11%) who had patella resurfacing. There was no revision surgery or
component failures of the patella during the follow-up. Patient satisfaction was higher among
the patients who had a replaced patella, and there were no complications related to
resurfacing. Patella infera occurred commonly but there was no connection between patella
infera and anterior knee pain.
The need for stapling due to valgus deformity in childhood was associated with early disease
onset and a need for TKA in early adulthood. The age of the patients at the time of TKA who
had no history of surgical knee procedures was higher than in other subgroups. Similarly, the
age in the subgroup who had undergone synovectomy was significantly higher than in the
patients who had had epiphyseal stapling.
This study is one of the largest published on TKA in JRA. The AGC total knee prosthesis
combined with patellar resurfacing yielded excellent results in this specific patient group.
7
ACKNOWLEDGEMENTS
The present study was carried out at the Department of Surgery, Rheumatism Foundation
Hospital, in cooperation with the Helsinki University, faculty of medicine during 1997-2004.
I am deeply grateful to all patients who devotedly participated in the follow-up examinations
and the telephone inquiry at the end of 1998.
I am indebted to Hannele Kalske BLL, Director of the Rheumatism Foundation Hospital,
Professor (h.c.) Martti Nissilä MD, PhD, and Docent Markku Hakala MD, PhD, Chief
Physician of the Rheumatism Foundation Hospital, for providing working facilities. I am also
grateful to Professor (h.c.) Martti Hämäläinen MD, PhD, the father of the Heinola
arthroplasty follow-up studies.
I express my profound gratitude to my supervisors Docent Eero Belt MD, PhD and Docent
Matti Lehto MD, PhD for their encouragement, guidance and useful advice. I am especially
grateful to Docent Eero Belt. His encouragement in initiating this study was crucial, and he
has untiringly and with an optimistic sense of humour supported my work during these years.
Without his support this study would not have been possible.
I am greatly indebted to Docent Anneli Savolainen MD, PhD for her positive attitude,
expertise and guidance during this study.
I thank Docent Hannu Miettinen MD, PhD and Docent Jorma Pajamäki MD, PhD, the
official reviewers of this thesis, for their constructive criticism and useful advice.
I thank Hannu Kautiainen BA for aid in the statistical work and the graphics as well as for
useful comments in the process of this undertaking. The help of Docent Robert Paul MD,
PhD, Certified Translator, in revising the English of this thesis is also gratefully
acknowledged.
I owe my warm thanks to my other co-workers Docent Markku Kauppi MD, PhD, Janne
Lehtinen MD, PhD and Carl Lybäck CM.
8
I am indebted to all the staff of the Rheumatism Foundation Hospital for their contribution,
especially to the Departments of Surgery and Radiology. I am also grateful to the staff of the
archives and scientific library. I thank Mrs Anna-Maija Linnakko and Mrs Eva Riihikorpi for
their practical help with my correspondence and other arrangements.
My very special thanks go to my close friend, artist Kimmo Tolin, who has designed the
cover picture of the thesis.
Finally, my profoundest gratitude belongs to my dear wife Paula and my children Viktor,
Wille and Calle for their admirable patience and support during this work.
This research was financially supported by ESR-project PaTu of the Rheumatism Foundation
Hospital, the Research Fund of Orthopedic and Traumatology, and The Fund of Pär Slätis.
Helsinki, August 2004
Christer Lybäck
9
CONTENTS
ABSTRACT 5
ACKNOWLEDGEMENTS 7
ABBREVIATIONS 11
LIST OF ORIGINAL PUBLICATIONS 12
1. INTRODUCTION 13
2. REVIEW OF THE LITERATURE 15
2.1. Anatomy of the knee 15
2.2. Juvenile rheumatoid arthritis 17
2.2.1. Diagnostic criteria and classification 17
2.2.2. Clinical manifestations of JRA 19
2.2.3. Prognosis 21
2.2.4. Secondary amyloidosis 22
2.3. Clinical examination of the knee joint 23
2.4. Radiological examination 24
2.4.1. Plain radiography and rheumatoid changes of the knee joint 24
2.4.2. Magnetic resonance imaging and ultrasonography in JRA 26
2.5. Conservative treatment of JRA 26
2.6. Surgical treatment of JRA 28
2.6.1. Synovectomy 28
2.6.2. Stapling procedure 30
2.6.3. Total knee arthroplasty 31
2.6.3.1. Total knee arthroplasty: operative technique 32
2.6.3.2. Development of knee prostheses 33
2.6.3.3. Patellar resurfacing in TKA 35
2.6.3.4. Results of TKA 36
2.6.3.5. Complications of TKA 39
3. PURPOSE OF THE STUDY 41
4. PATIENTS 42
10
5. METHODS 45
5.1. Surgical techniques 45
5.2. Follow-up 45
5.3. Radiography 46
5.4. Statistical methods 48
6. RESULTS 49
6.1. Study I 49
6.2. Study II 52
6.3. Study III 54
6.4. Study IV 57
7. DISCUSSION 59
7.1. Survival of knee replacement in patients with JRA 59
7.2. Patellar resurfacing in patients with JRA 60
7.3. Patella infera in patients with JRA 61
7.4. TKA and secondary amyloidosis in patients with JRA 62
7.5. Synovectomy and epiphyseal stapling in patients with JRA 62
7.6. Knee surgery in the future 64
8. CONCLUSIONS 65
9. REFERENCES 66
10. ORIGINAL PUBLICATIONS (I-IV) 78
11
ABBREVIATIONS
AGC Anatomically graduated components
ARA American Rheumatism Association
(today ACR, American College of Rheumatology)
AS ankylosing spondylitis
CI confidence interval
CMC carpometacarpal
CRP C-reactive protein
DMARD disease modifying antirheumatic drug
ESR erythrocyte sedimentation rate
EULAR European League Against Rheumatism
Hb hemoglobin
HLA human leukocyte antigen
ILAR International League of Associations of Rheumatologists
JAS juvenile ankylosing spondylitis
JCA juvenile chronic arthritis
JIA juvenile idiopathic arthritis
JRA juvenile rheumatoid arthritis
KSS Knee Society Scores
MCP metacarpophalangeal
MIS minimally invasive surgery
MRI magnetic resonance imaging
NSAID non-steroidal anti-inflammatory drug
OA osteoarthrosis
RA rheumatoid arthritis
RF rheumatoid factor
RFH Rheumatism Foundation Hospital
SA secondary amyloidosis
SD standard deviation
TKA total knee arthroplasty
US ultrasonography
12
LIST OF ORIGINAL PUBLICATIONS
This thesis is based on the following original communications:
I. Lybäck CO, Belt EA, Hämäläinen MMJ, Kauppi MJ, Savolainen HA, Lehto MUK:
Survivorship of AGC knee replacement in juvenile chronic arthritis: A follow-up of 77 knees
up to 13 years. J Arthroplasty 2000; 15:166-170.
II. Lybäck CO, Belt EA, Kauppi MJ, Savolainen HA, Kautiainen HJ, Lehto MUK: Knee
replacement is not contraindicated in the presence of secondary amyloidosis in patients with
juvenile chronic arthritis. J Rheumatol 2000; 27:2059-2061.
III. Lybäck CO, Lehto MUK, Hämäläinen MMJ, Belt EA: Patellar resurfacing reduces pain
after total knee arthroplasty for juvenile rheumatoid arthritis. Clin Orthop 2004; 423:152-
156.
IV. Lybäck CO, Belt EA, Savolainen HA, Lehtinen JT, Lybäck CC, Lehto MUK: Previous
synovectomy or epiphyseal stapling and the influence on knee replacement in juvenile
chronic arthritis. Int Orthop 2004; 28:134-137.
13
1. INTRODUCTION
The cause of JRA (see 2.2.1. for nomenclature) is unknown. It is widely agreed that the
etiology of chronic juvenile arthritis comprises a combination of genetic and environmental
factors - especially infections - of which very little is known. The prevalence of JRA is about
0.1% - about ten times less than RA - and its incidence varies in Scandinavian population-
based studies from 10 to 23/100 000 among children <16 years of age (Kunnamo et al. 1986,
Andersson-Gäre and Fasth 1995b, Andersson-Gäre 1999, Kaipiainen-Seppänen and
Savolainen 2001). Population studies have confirmed that there is female predominance over
males at a ratio of 3:2. It has been recognized that each JRA subtype has a different
prognosis. In the study by Oen et al. (2002) the probability of remission at 10 years after
onset was 37% for patients with systemic, 47% with pauciarticular, 23% with RF-
polyarticular and 6% with RF+ polyarticular JRA.
A multiprofessional team is of crucial for the successful treatment of JRA patients. The
treatment consists of a range of items: pharmacotherapy, intra-articular injections, physical
and occupational therapy (including orthoses and other aids for daily life), psychosocial
support, vocational guidance, nutritional guidance, surgery etc.
The most frequently affected site in JRA is the knee joint which is involved in two thirds of
patients at disease onset (Ansell 1977). However, any joint can be affected by JRA (Jacobs
and Hui 1977). Surgical treatment of the knee joint of patients with JRA consists of
epiphyseal stapling, synovectomy and total joint replacement. Replacement surgery in
patients with JRA is very challenging and requires experience and surgical skill. Careful
preoperative planning is crucial in the treatment of these patients. Surgery is technically
especially demanding since the bones are small, there is often osteoporosis and the soft tissue
may be severely involved. Often, custom-made components are needed (Scott 1990).
There is very seldom need for total knee arthroplasty (TKA) in patients under 16 years old
(Hämäläinen 1995), and it is in many cases appropriate to postpone arthroplasty to
adolescence or adulthood (Scott 1990). In this way, the young patient will probably be better
motivated and understand the reason for the operation and postoperative rehabilitation.
Hämäläinen (1995) urged caution in surgical treatment of the knee in patients with JRA.
14
TKA is, nevertheless, indicated in patients with JRA when marked joint destruction is
present and pain or deformity compromises function (Ranawat et al. 1983, Scott et al. 1984,
Carmichael and Chaplin 1986).
The main attention in the present study was paid to the outcome of TKA in patients with
JRA.
15
2. REVIEW OF THE LITERATURE
2.1. Anatomy of the knee
The knee is a complex joint with three major compartments: the patellofemoral joint, the
medial tibiofemoral joint and the lateral tibiofemoral joint; the two latter compartments
contain the fibrocartilaginous menisci. The osseous structures of the knee joint consist of the
patella, the distal femoral condyles, and the proximal tibial plateaus. Two flat surfaces,
plateaus or condyles, of the proximal end of the tibia articulate with the femoral condyles.
The patellofemoral groove is located anteriorly between the femoral condyles. The patella is
a triangular sesamoid bone with a smaller medial and a larger lateral articular facet. The knee
is principally a hinge joint, although some rotation occurs during flexion-extension.
Fig. 1. Undestroyed knee joint of 12-year old female with JRA. The same knee is shown in Fig 2 when the patient was 20 years old and there was marked destruction.
16
The knee is stabilized by the following structures: the articular capsule, the
musculotendineous units (including the quadriceps mechanism, gastrocnemius, medial and
lateral hamstring groups, popliteus, and iliotibial band), the medial and lateral collateral
ligaments, which provide medial and lateral stability, and the anterior and posterior cruciate
ligaments, which provide anteroposterior support and rotatory stability. Both the medial and
the lateral meniscus participate in stability. They are especially important rotatory stabilizers.
The knee joint is surrounded by a large synovial membrane which extends up to the
suprapatellar pouch beneath the quadriceps femoris muscle. The following bursae are found
in the knee area: the superficial prepatellar bursa, the superficial and deep infrapatellar
bursae, the pes anserine bursa, and the posterior medial semimembranous and posterolateral
gastrocnemius bursae.
All three compartments of the knee joint are covered by articular cartilage, which absorbs
transmitted forces thereby protecting the underlying bone and enables low-friction and high-
velocity movement between the bones, whereas the menisci are important in reducing stress
on the articular cartilage. A healthy knee joint contains about 2.5 ml of synovial fluid.
Synovial fluid is essential for the nutrition and lubrication of the articular cartilage and
tendons.
Knee motion is a combination of flexion, extension and rotation. The tibia rotates internally
on flexion of the knee and externally on the femur on extension. The knee flexes when the
hamstring muscles contract and extends when the quadriceps femoris muscle contracts. The
normal range of motion of the knee joint is from slight hyperextension (~ 5 degrees) to 140
grades of flexion. The joint permits rotation, which ranges from 5 to 25 degrees when the
joint is flexed. No rotation is possible when the knee is in full extension. Rotation of the tibia
on the femur is provided by the biceps femoris on external rotation, whereas the popliteus
and semitendinosus muscles are involved in internal rotation.
The anatomical axis of the knee is the angle between the axis of the shaft of the femur and
tibia, and normally the anatomical axes of the femur and the tibia form a valgus angle of 6 ±
2 degrees. The mechanical axis is a line traversing the center of the hip joint to the center of
the ankle joint. This mechanical axis should project through the center of the knee joint.
17
Throughout the period of growth, the skeleton changes in size, shape and microstructure. The
epiphyseal plate, physis, is a cellular zone in which new cells are being formed. Factors that
influence growth at the physis include thyroxine, growth hormone, and testosterone.
Estrogens suppress the growth rate by increasing calcification of the matrix. This is a
prerequisite for epiphyseal closure. In general, the physes fuse earlier in girls than in boys. A
significant finding is that glucocorticoids affect physeal closure indirectly by suppressing
growth hormone release (Sledge 1968, Balogh and Kunin 1971). Most (67%) of the length
of the lower limb is derived from the physes of the distal femur and proximal tibia (Ogden
1982).
2.2. Juvenile rheumatoid arthritis
Chronic arthritis is much less frequent in children than in adults. The juvenile form of
rheumatoid disease has to be held distinct from adult rheumatoid arthritis (RA), which is a
much more uniform disease and has usually a less favourable prognosis. The juvenile form
is a group of diseases with different manifestations and prognoses. What they have in
common is chronic arthritis in one or several joints. The course of the disease differs from
patient to patient, often with alternating remissions and exacerbations. In some instances the
disease clears completely and sometimes - fortunately seldom - its activity is relentless,
leading to serious impairment or even death.
2.2.1. Diagnostic criteria and classification
The disease has been called juvenile chronic arthritis (JCA) in Europe and juvenile
rheumatoid arthritis (JRA) on the other side of the Atlantic. The diagnostic criteria are only
slightly different (Brewer et al. 1977, Wood 1978) see Table 1. A task group has been
working during the last decade to achieve universal diagnostic criteria. These seem to gain
acceptance and are under validation in several countries (Miranda et al. 1997, Berntson et al.
2001). The new name is juvenile idiopathic arthritis (JIA). In the present thesis the
nominations JRA and JCA are used, because the patients were diagnosed before the new
criteria were created. Due to the nature of the criteria it is not always possible to apply them
retrospectively. The diagnosis is based on the clinical presentation and exclusion of other
diseases with a clear etiology. Based on differences in clinical presentation, the disease is
divided into subtypes: this is made after six months of disease duration. The disease was
initially divided into the three types by onset: systemic, polyarticular and oligoarticular
18
(pauciarticular). Division into subgroups is important because they represent different
diseases with different clinical presentation and prognosis. The systemic form is defined by a
daily high spiking fever with or without a rash or other organ involvement (rash,
hepatomegaly, splenomegaly, lymphadenopathy or pericarditis) (Southwood and Woo 1995).
In the polyarticular disease form five or more joints and in the oligoarticular form maximally
four joints are affected by six months of disease duration. The only laboratory value included
in all three sets of criteria is the rheumatoid factor (RF).
Table 1. Current sets of diagnostic criteria for prolonged arthritis in children.
EULAR ARA ILAR
Name of disease group
Juvenile chronic arthritis (JCA)
Juvenile rheumatoid arthritis (JRA)
Juvenile idiopathic arthritis (JIA)
JAS, IBD, JpsA Included (listed separately)
Excluded
Duration of symptoms necessary
for diagnosis
3 months 6 weeks 6 weeks
Age of patient at disease onset
0-15 years 0-15 years 0-15 years
Exclusion of other diseases
Yes Yes Yes
Division into onset subtypes at 6
months´ disease duration
Yes Yes Yes
Abbreviations: JAS = juvenile ankylosing spondylitis; IBD = arthropathy associated with inflammatory bowel disease; JpsA = juvenile psoriatic arthropathy.
The new criteria by ILAR includes seven categories: systemic, seropositive and seronegative
polyarthritis, oligoarthritis, enthesitis-related arthritis, psoriatic arthritis and other (Petty et al.
1998). The aim of this classification is to define homogeneous groups according to their
clinical and biologic features, which facilitates scientific collaboration.
19
International collaboration is needed because some of the subtypes are so rare that the
collection of any reasonable number of patients necessitates monitoring of several
nationwide populations.
The character of the disease can change at any time after the initial six months. In these cases
- which present up to 25 per cent of all - in addition to the above onset type also the disease
course type is given. The course type is, in fact, more important for the prognosis than the
onset type (Cassidy et al. 1986).
The prevalence of the different subtypes of JRA differ to some extent by studies. The
distribution is not the same in the western countries as in the Far East. There are also
differences if the study population consist of hospital patients or open care patients.
According to Andersson-Gäre and Fasth 1992 (EULAR), the proportion of systemic JRA is 5
% of all JRA patients, of polyarticular 35%, of oligoarticular 55%, of JAS 1.5% and of JpsA
3%. Lantto and von Wendt 1985 (ARA) reported 3% systemic JRA, 31% polyarticular and
66% oligoarticular JRA.
2.2.2. Clinical manifestations of JRA
The cardinal symptom of JRA is joint stiffness in the mornings and after other prolonged
inactivity. Inflamed joints are tender and may be painful, especially at night. Young children
are not able to express their pain verbally and the presence of pain may only be suspected by
careful observation of the child and by a careful history by the parents. Although pain is one
of the cardinal symptoms of arthritis, it is absent in 14% of patients with juvenile disease
(Sherry et al. 1990). Often the child does not complain of pain at rest but pain on motion of a
joint may be present (Laaksonen and Laine 1961). Fever, anorexia, weight loss, growth
retardation and fatigue are present in all subtypes of JRA, but they are more common in
patients with the systemic or seropositive type of disease (Lovell and Walko 1989).
Large joints, such as the knees, wrists and ankles, are more prominently involved than are the
small joints of patients with JRA. The arthritis is usually asymmetric, except in seropositive
polyarthritis. The most common site is the knee joint which is involved in two thirds of the
patients during the course of the disease (Ansell 1977). However, every single joint - even
the cricoarytenoid - may be affected in JRA (Jacobs and Hui 1977).
20
The affected joints are warm and swollen. Swelling results from periarticular soft tissue
inflammation and intra-articular effusion. An active, long-standing inflammation of the joint
destroys articular cartilage and affects the ligaments, tendons, and bone. The soft tissue and
osseous damage is due to enzymes within the synovial fluid, and from granulation tissue or
pannusformation.
In addition to arthritis, extra-articular manifestations in JRA occur (Table 2) (Sury and
Vesterdal 1968). Disturbances of growth are characteristic features of children with severe
arthritis. These may be caused by glucocorticoid treatment, but the disease itself may also
blunt growth. Ansell and Bywaters (1956) reported that a long duration of active disease is a
significant factor for reducing of linear growth of children who had not received
glucocorticosteroid drugs. Local growth disturbances may cause deformities, and if this
involves the knee, valgus deformity and unequal leg length may result (Vostrejs and Hollister
1988).
Adults with JRA have an increased risk of osteopenia (French et al. 2002) and osteoporosis
(McDonagh 2001). However, osteopenia and osteoporosis have their origins in childhood
and adolescence (Cassidy 1999). Osteoporosis in JRA is multifactorial in origin and there are
several causes of reduced bone mineral density (BMD) in these patients, e.g., high disease
activity, glucocorticoid use, insufficient nutrition, and reduced physical activity (McDonagh
2001). Kotaniemi et al. (1999) observed that body size, physical activity, and calcium intake
are significant positive predictors, and disease activity and glucocorticoid use are significant
negative predictors of BMD in JRA. Mäenpää et al. (2002) found that patients with JRA are
at high risk of developing stress fractures.
The most common extra-articular disease manifestation of JRA is uveitis (Cassidy et al.
1989, Kotaniemi et al. 2001). In a Finnish study, severe visual impairment was observed in
1.2% of patients who had had JRA for 15 years or longer. In addition, 2% of patients had
impaired vision because of glucocorticoid-induced cataract (Ylijoki 1998).
21
Table 2. Extra-articular manifestations by JRA-subgroups. The percentage devotes the estimated frequency of occurrence of the respective manifestations.
Polyarthritis (%) Oligoarthritis (%) Systemic disease (%)
Fever 30 0 100Rheumatoid rash 2 0 95 Rheumatoid nodules 10 0 5 Hepatosplenomegaly 10 0 85 Lymphadenopathy 5 0 70 Chronic uveitis 5 20 <1 Pericarditis 5 0 35 Pleuritis 1 0 20 Abdominal pain 1 0 10
2.2.3. Prognosis
Each JRA subtype has a different prognosis. In the study by Oen K et al. (2002) the
probability of remission at 10 years after onset was 37% for patients with systemic, 47% with
pauciarticular, 23% with RF- polyarticular, and 6% with RF+ polyarticular. The probability
of relapse in different subtypes varied from 30 to 100% at 15 years. The probability of
arthroplasty varied from 13 to 57% after 15 years of active disease.
The presence of RF affects the outcome negatively (Andersson-Gäre and Fasth 1995b).
Morling et al. (1985) found an increased frequency of HLA B27 in all subgroups of JRA.
Savolainen et al. (1998) observed that patients positive for HLA B27 are overrepresented
among the most severe cases of JRA, e.g., those who develop amyloidosis or who need
arthroplasty. HLA B27 positive patients underwent arthroplasty on average 2.9 years earlier
than HLA B27 negative patients. RF positive polyarthritis is associated with marked
disability in adulthood. A poor outcome of patients with the systemic form of JRA correlates
with markers of disease activity, such as fever and polyarticular involvement, within the first
6 months of disease onset. The risk of joint destruction in oligoarthritis correlates with the
severity of arthritis within the first 2 years (Prieur et al. 2001).
22
JRA continues into adulthood in approximately 50 per cent of patients, but fewer than 10%
are severely disabled or handicapped (Minden et al. 2002). The exact number of patients with
disease relapse in adulthood is not known, but such instances are not uncommon.
Koivuniemi and Leirisalo-Repo (1999) compared the prognostic factors and outcome of JRA
with adult-onset RA and observed that significantly more patients with JRA than RA (60%
versus 23%) were in remission at the end of the follow-up. The long-term prognosis is better
in patients with JRA than in those with RA.
Ylijoki (1998) followed up 264 patients with JRA for 25 years and found that 27% of
children with systemic disease and 25% with polyarthritis had severe limitations and were
unable to attend regular school or were unemployed. In contrast, 90% of the children in the
pauciarticular group remained undisabled or had only slight impairment. The JRA-related
mortality in the systemic group was 14%, in the polyarticular 10% and in the pauciarticular
arthritis group 2%. Secondary amyloidosis was the most important cause of death, and was
characteristic of patients with systemic disease.
2.2.4. Secondary amyloidosis
In amyloidosis extracellular insoluble fibrillar proteins, amyloid, accumulate in tissues and
organs. Serum amyloid-A protein (SAA) is the putative precursor of the amyloid-A (AA)
protein which forms the fibrils in secondary amyloidosis (SA). The precise mechanism of
amyloidogenesis is not known. Secondary amyloidosis may occur as a complication of
chronic inflammation. The concentration of SAA is greatly elevated in patients with RA and
JRA, and correlates with the primary disease activity (De Beer et al. 1982). A high serum
SAA protein concentration in JRA may not correlate with the presence of SA (Filipowicz-
Sosnowska et al. 1978). Confirmation of the diagnosis requires a biopsy (Tiitinen et al.
1988). Gertz and Kyle 1991 found that biopsies from renal, gastric, rectal, fat, and bone
marrow were positive for amyloid in 100%, 94%, 82%, 58%, and 46% of tested patients with
SA, respectively.
SA occurs in children with systemic JRA (Prieur and Chedeville 2001). SA is one of the
most deleterious complications of inflammatory joint diseases. The typical clinical
manifestations of SA in patients with JRA are proteinuria and renal failure.
23
Helin et al. (1995) reported that SA is the most common cause of nephrotic syndrome in
patients with RA.
SA is not distributed evenly by geography. In Europe the occurrence of SA is much higher,
3.2-11%, than in the USA/Americas, 1.8% (Schnitzer and Ansell 1977, Rostropowicz-
Denisewicz and Matdyk 1977, Stoeber 1981). The 25-year incidence of SA in patients with
JRA is 3.3 % in Finland (Ylijoki 1998). In a study from Norway, approximately 2% of the
patients develop amyloidosis in 10 years (Flatø et al. 1998).
The incidence of SA is declining both in patients with JRA and RA in Finland (Savolainen
and Isomäki 1993, Laiho et al. 1999). The reduced incidence of SA has been attributed to
improved medical treatment directed toward of the underlying inflammatory disease (Husby
1985, Möttönen et al. 1999, Sokka et al. 1999). Effective treatment with suppression of
inflammation should alleviate the course of SA and reduce the extent of organ complications
(David et al. 1993).
The attitude towards TKA in patients with JRA and SA has been conservative due to
technical challenges (Rydholm et al. 1985). SA carries a high complication rate and a
shortened lifespan (Suzuki et al. 1992, Kageyama et al. 1998). In the study reported by
Lehtimäki et al. (1999), SA appears to reduce the survival of hip arthroplasty in patients with
RA. It is, however, worth noting that the replacements in that series were performed between
1971 and 1991, i.e., before the time of intensified medical treatment. Hämäläinen et al.
(1995) found in a study of survival of Townley knee prostheses similar results in a follow-up
of 11 years. Survival was 82.4% in patients with SA and 95.5% in those without SA.
2.3. Clinical examination of the knee joint
Patients with JRA have often changes in several joints – not only in the knee. It is, therefore,
important to examine the patient as a whole taking into account all joints and the patient’s
general condition. It is essential to obtain a detailed description of the patient’s symptoms.
For pediatric patients it is important to get a history of the parents, as well.
24
A full knee examination includes an observation of the patient both standing and walking.
Deviations (valgus and varus deformation) of the knees are best seen with the patient in the
standing position. The patient should also be observed walking for evidence of gait
abnormalities. Swelling or muscle atrophy may be noted on inspection, as may erythema of
the knee area.
The knee should first be gently palpated, which allow identification of swelling, tenderness,
warmth, and effusion of the knee joint. The range of motion should be recorded in maximal
extension and flexion of the knee joint. Patellar stability should also be assessed. Collateral
ligamentous stability is tested by application of valgus and varus stress to the knee. The
stability of cruciate ligaments is tested with anterior and posterior drawer tests, Lachman test
and pivot shift test. Radiography is mandatory. Quality of bone, deformities, grade of
erosions, position and shape of patella can be evaluated on the radiograph. The anatomical
and mechanical axes may be evaluated.
2.4. Radiological examination
The radiographic features in JRA depend on disease duration and severity. Erosions appear
late and not as often as in adult-onset disease. It takes usually at least two years after onset
before joint narrowing or erosive changes appear in plain radiographs (Still 1897, Sairanen
1958). Soft-tissue swelling is usually confined to joints with evident synovial thickening.
Joint effusions are an early sign and are easily detected clinically and may be demonstrated
by ultrasonography (US) and magnetic resonance imaging (MRI).
2.4.1. Plain radiography and rheumatoid changes of the knee joint
The radiographic features of JRA can be divided into early and late manifestations. The early
radiographic features are articular soft-tissue swelling, widened joint space due to effusions,
juxtacortical osteopenia, and metaphyseal sclerotic lines which may imply growth arrest
around the adjacent joint. Articular erosions, pseudocysts, synovial cysts, deformities, and
subluxations appear late. The joint and epiphyseal cartilage and ossification center cartilages
are not radiographically visible, and erosions are thus detectable only after ossification. Joint
cartilage destruction leads to narrowing of the joint spaces.
Osteoporosis due to long-standing glucocorticoid therapy may produce stress or insufficiency
fractures (Mäenpää et al. 2002).
25
Fig. 2. Destroyed knee joint of 20-year old female with JRA. The same knee at the age of 12 years is shown in Fig 1.
The valgus deformity of the knee joint is a common finding in patients with JRA (Isacson et
al. 1987). Radiographs have also suggested an association between the valgus deformity of
the ankles and valgus deformity of the knees in patients with RA (Keenan et al. 1991). The
varus or valgus deformity in the knee joint can be determined on the anteroposterior
radiograph taken with the patient in the standing position. If the mechanical axis of the lower
limb lies to the lateral side of the knee center, the knee is in valgus alignment. In varus
alignment, the mechanical axis lies to the medial side of the knee center. The amount of
varus or valgus deformity in terms of degrees can be calculated by measuring the mechanical
axis of the femur and the mechanical axis of the tibia; the latter runs from the center of the
tibial plateau to the center of the ankle joint. The angle formed between these separate
mechanical axes is the degree of varus or valgus deformity.
Insall and Salvati (1971) were the first to publish measurements of the height of the patella.
They measured the length of the patellar tendon (LT) and the length of the patella (LP) on the
lateral radiograph of the knee joint. The ratio is now called the Insall-Salvati ratio (LT/LP)
and it is used to assess the vertical position of the patella.
26
2.4.2. Magnetic resonance imaging and ultrasonography in JRA
MRI has proved to be very useful for imaging of articular surfaces, subchondral bone
changes and edema in arthritis. Articular surface defects and osteochondral lesions are best
defined by contrast MRI (Chandnani et al. 1991). Synovitis as well as other soft tissue
changes, such as tenosynovitis, tendinitis, enthesitis, joint effusions, and ligament and tendon
tears can be visualized by MRI (Ostergaard and Szkudlarek 2001).
Lindegaard et al. (2001) compared a low field dedicated extremity magnetic resonance
imaging (E-MRI) system with radiography and clinical examination in patients with newly
diagnosed, untreated RA. E-MRI detected 57 bone erosions, radiography only six. A total of
51% of the joints with no clinical signs of synovitis showed synovial hypertrophy on E-MRI.
Ostergaard et al. (1995) compared the sensitivity of US with MRI for demonstration of the
synovial membrane, bone erosions, effusions and Baker's cysts. US demonstrated joint
effusions and Baker's cysts reliably, but showed a lack of sensitivity for demonstration of the
synovial membrane and bone erosions. The clinical value of US is limited, because the
weight-bearing areas are inaccessible.
MRI is a valuable tool for diagnosing and determining the prognosis of patients with RA and
for disease activity and disease severity (Ostergaard et al. 1995, Ostergaard and Szkudlarek
2001). MRI is a non-invasive procedure that complements arthroscopy for evaluating
diseases of the knee (Burk et al. 1990).
2.5. Conservative treatment of JRA
The aim of treating JRA is to diminish disease activity, to prevent joint deformities,
contractures, growth disturbances and other complications, and to relieve symptoms.
Treatment of JRA patients requires a multiprofessional team. The treatment consists of a
wide range of items: pharmacotherapy, intra-articular injections, physical and occupational
therapy (including orthoses and other aids for daily life), psychosocial support, vocational
guidance, nutritional guidance, surgery etc.
27
Non-steroidal anti-inflammatory drugs (NSAIDs) are introduced as soon as possible, even
before the diagnosis is confirmed. Glucocorticoids may be used systemically or as local
injections. Disease modifying antirheumatic drugs (DMARDs) and their combinations are
used to reduce inflammatory activity. Methotrexate is the gold standard for the management
of moderate to severe polyarthritis (Giannini et al. 1992). Other DMARDs such as
hydroxychloroquine, gold, sulfasalazine, leflunomide and cyclosporin have also a role in the
treatment of JRA. Chlorambucil is useful for treating JRA patients with amyloidosis.
Because of the severe side effects its use is warranted only in grave situations (Savolainen
1998). The novel drugs, so called biological response modifiers which selectively block
individual factors in the immunologic processes, were not yet used by the patients
participating in the present study.
The Finnish anti-rheumatic treatment policy has become more intensive over the last decade,
and patients are often treated with combinations of different DMARDs (Möttönen et al.1999,
Sokka et al. 1999). Intra-articular glucocorticoids are commonly used in children with
oligoarticular JRA and as a useful adjunct to DMARD therapy in children with other forms
of JRA (Honkanen et al. 1993, Murray and Lovell 2002). Ravelli et al. (2001) analyzed the
effect of intra-articular glucocorticoid injections in 94 JRA patients with synovitis of knee
joint. At 6 months after the injection, 65 (69%) patients showed a sustained complete clinical
response, i.e., no clinical evidence of synovitis. Their findings indicate that patients with JRA
who have a high ESR are more likely to benefit from intra-articular gluccocorticoid injection
of the knees.
Rheumatoid synovitis in the knee joint can be treated with intra-articular osmic acid or
yttrium-90 (radiation) injections. Sheppeard et al. (1981) found osmic acid to be more
effective than yttrium-90. However, both therapies are well tolerated by patients with RA and
should be considered as an alternative to operative synovectomy. Nissilä (1975) compared
the effect of osmic acid to that of glucocorticoids and found osmic acid more potent than
glucocorticoid alone. Intra-articular osmic acid injections seem to be more effective in the
treatment of rheumatoid knee joint synovitis at an early stage of joint destruction (Nissilä et
al. 1977, Sheppeard and Ward 1980).
28
Similar results were obtained by Kastner and Wessel (1977) who used intra-articular
Varicocid (sodium salts of fish oil fatty acids) in the treatment of RA patients. After 12
months, 46% of the treated knees were symptomfree, and 33% showed considerable
improvement. This type of therapy is non-toxic.
2.6. Surgical treatment of JRA
When surgery is contemplated, it is important to differentiate between the three main types of
JRA because each has a different prognosis. In rheumatological orthopedics good results are
highly dependent on the function of an effective multiprofessional team (Pahle 1996).
Prophylactic measures against joint contractures are of great importance. Procedures are
often multiple and require careful preparation, including psychological support, nutritional,
medical and orthopedic care, and should be followed by intensive and focused physiotherapy.
Special training in surgical techniques is necessary. It is also necessary to realize the
involvement of the whole connective tissue and the multiple organ affections in this disease.
A centralized unit with an experienced staff mastering the problems of anesthesia,
medication, intraoperative bloodtransfusion and physiotherapy before and after surgery may
yield good results in the management of this disease (Pahle 1996).
2.6.1. Synovectomy
An important and interesting question is whether synovectomy can slow down the
destruction of the knee joint and thus influence the need and timing of arthroplasty.
The joint most frequently affected in JRA is the knee. This holds true for all subtypes of JRA
and early in the course of disease a good two-thirds manifest as a knee arthritis (Ansell 1977,
Rydholm et al. 1985). Larsen et al. (1977) presented a radiographic evaluation of RA.
Standard reference films are used to grade joint damage on a scale of 0-5. The Larsen grades
have also been applied to knee destruction in JRA to divide synovectomies into early (grade
0-2) and late (grade 3-5) synovectomies.
Laine (1964) found that early synovectomy of the knee joint had a moderating influence on
the disease activity of RA. Pätiälä (1975) reported that knee synovectomy does not prevent
destruction but seems to retard its progression. The same observation was seen in late
(Larsen 3-4) synovectomies. In that series the subjective contentment of patients was good in
29
71% and poor in 12% of the patients. Jensen et al. (1991) compared early and late
synovectomy in a retrospective study including 44 patients (55 knees). Early synovectomy
(Larsen 0-2) showed significant pain relief and improvement of knee function, effusion was
reduced and the range of motion was unchanged. They concluded that early synovectomy is
indicated after 3 to 6 months if medical treatment has failed. Late synovectomy was regarded
as a palliative procedure: it postponed TKA by 5 to 10 years. A recent long-term (6-16 years)
study of arthroscopic synovectomy in RF-positive RA patients showed that the range of
motion was maintained or improved by surgery in 73% of the patients, but radiographic
evidence of degenerative change was seen in all knees (Gibbons et al. 2002).
Cathepsins, like proteinases, play a major role in destroying the synovial tissue and cartilage
matrix in patients with JRA and RA (Taubert et al. 2002). The activity of these pathological
enzymes is reduced in the synovial tissue after knee synovectomy (Myllylä et al. 1983, Paus
et al. 1990). These findings support the clinical observation that early synovectomy may
retard destruction of the knee joint cartilage in patients with RA.
Because of the different prognosis between the three main types of JRA, it is important that
the indications for open or arthroscopic synovectomy are carefully considered (Adamec et al.
2002). Knee synovectomy carries a poor outcome in patients with an overall high
inflammatory activity and polyarticular or systemic disease (Rydholm et al. 1986, Adamec et
al. 2002). Progressive joint destruction is more common in younger patients - those with
systemic or polyarticular disease, and those with highly active disease at the time of
operation. Recurrence of synovitis is more frequent in patients who are synovectomized
when the disease activity is high (Rydholm et al. 1986). Some authors consider the presence
of acute systemic disease as a contraindication to synovectomy (Adamec et al. 2002). It is
prudent to perform the surgical intervention, when disease activity is low.
Synovectomy reduces pain, swelling and disease activity of the joints of patients with JRA
(Meijers et al. 1983, Rydholm et al. 1986, Kvien et al. 1987, Matsui et al. 1989, Ochi et al.
1991, Ogilvie-Harris and Basinski 1991, Vilkki et al. 1991). Whether joint cartilage
destruction is prevented by synovectomy is a more controversial question. Arthroscopic
studies have shown that cartilage destruction is frequently more advanced than can be
deduced from radiography (Rydholm 1986, Adamec et al. 2002). Radiographic deterioration
30
(average, 6 years follow-up after operation) seems to continue in the joints operated on if
there are significant radiographic changes at the time of operation (late synovectomy).
In joints without marked radiographic changes at the time of operation (early synovectomy),
joint deterioration continues in patients with polyarticular disease but less so in patients with
oligoarthritic disease (Jacobsen et al. 1985). Pahle et al. (1996) performed 528
synovectomies over 13 years in a controlled study of open knee synovectomies. Recurrence
of inflammation was rarely seen when synovectomy was early and radical. If there are clear
signs of plastic synovitis and skeletal lesions by radiography and if synovitis recurs open
synovectomy may be indicated (Adamec et al. 2002).
Arthroscopic synovectomies are more difficult to perform in children than in adults because
of the knees in children are more “tight” (Adamec et al. 2002). Arthroscopic synovectomy is
considered to be a safe procedure, and a good alternative for open synovectomy in selected
pediatric patients (Vilkki et al. 1991). Today most knee synovectomies are performed
arthroscopically (Cohen and Jones 1987, Hafner and Pieper 1995, Ayral et al. 1997, Dirienzo
et al. 1997, Takagi et al. 2001).
2.6.2. Stapling procedure
Chronic synovitis of the knee joint causes growth disturbances and results in valgus
deformity of the knee or in leg length discrepancy (Laine and Mikkelsen 1968, Rydholm et
al. 1987). During the growth period these deformities can be treated by stapling of the
epiphyses to the distal femur and sometimes also to the proximal tibia (Blount and Clarke
1949).
Temporary epiphyseal arrest with staples in the management of knee malalignment was
introduced in 1949 by Blount and Clarke (Blount and Clarke 1949). The first temporary
arrest of epiphyseal growth was performed on May 17th 1956 at Rheumatism Foundation
Hospital (RFH) in Finland for the management of knee valgus deformity.
Fluoroscopic localization of all growth plates and careful preservation of the periosteum is
necessary in order to avoid staple mislocation which would lead to premature closure of the
growth plate (Mielke and Stevens 1996). In management of the valgus deformity of the knee
31
the technique consists of inserting one or two staples per physis medially at the distal femoral
epiphysis. The knee with leg length discrepancy is stapled at the distal femoral epiphysis
both medially and laterally, and in difficult deformities the proximal tibial epiphysis may also
be stapled (Blount and Clarke 1949, Skyttä et al. 2003a, Skyttä et al. 2003b).
In a recent study involving correction of leg length discrepancy in a total of 76 knees in
patients with JRA, the mean preoperative leg length discrepancy was 18 mm (Skyttä et al.
2003a,b). After 19 patients had stopped growing, an endpoint analysis was done. It turned out
that the average leg length discrepancy of 20.4 mm was corrected by 16.4 mm at the time of
staple removal, and by 15.5 mm after a follow-up of 57 months. The mean age of the patients
at the time of stapling was 12 years and the duration of stapling 20 months. Most patients had
oligoarthritic JRA (Skyttä et al.2003a,b). Most JRA patients with the valgus deformity are
affected by aggressive systemic arthritis and polyarthritis (Skyttä et al., personal
communication).
In a study on the effect of stapling on the angular deformities of the knee Mielke and Stevens
(1996) reported that the anatomic angle and mechanical axis improved in all patients. The
mean age at stapling was 6 years, and the average follow-up 3 years. In a study on medial
physeal stapling Fraser et al. (1995) found that at skeletal maturity, 85% of the patients had
excellent or good leg alignment, and correction had occurred within one year. Temporary
epiphyseal stapling is a safe and effective method for managing angular deformities in
patients with JRA (Zuege et al. 1979, Arden 1983, Rydholm et al. 1987).
2.6.3. Total knee arthroplasty
Joint replacement surgery in patients with JRA is very challenging and requires experience
and surgical skill. Careful preoperative planning is crucial in the treatment of these patients.
Since the patients have small bones and often also osteoporosis and severe soft tissue disease,
surgery is technically demanding and requires often the use of custom-made components
(Scott 1990). Individual planning of the postoperative treatment is also important because
problems with other joints may arise. A rheumatoid patient with a severe handicap demands
postoperatively three-fold more help than the average surgical patient (Jonsson and Larson
1991). Pre- and postoperative serial casts have been used to correct severe flexion deformity
32
of the knee. Still, postoperative manipulation under general anesthesia is sometimes required,
in addition, for some JRA patients (Sarokhan et al.1983).
The timing and order of operations require careful planning. The operative procedure is more
demanding if destruction, deformity, contracture and instability are advanced. Therefore,
TKA should be performed before appearance of too advanced changes in the knee joint.
When both the hip and the knee are involved, hip arthroplasty should usually be done first
(Scott 1990, Vahvanen 1991).
TKA is indicated in patients with JRA if there is marked joint destruction and pain or
deformity compromises function (Ranawat et al. 1983, Scott et al. 1984, Carmichael and
Chaplin 1986).
2.6.3.1. Total knee arthroplasty: operative technique
The anteromedial parapatellar approach is often used for TKA. This approach allows
retracting the patella laterally and allows a good view of the anterior compartment of the
knee joint. Sometimes wider access to the joint is needed, and if this is the case, the incision
can be extended proximally or the proximal part of the incision can be extended obliquely
and medially into the vastus medialis muscle. If standard medial parapatellar arthrotomy is
inadequate, a quadriceps snip is helpful. Occasionally exposure will still be limited and a
tibial tubercle osteotomy can be helpful (Clarke 2003).
The anterolateral approach is an alternative, but it is not used widely because displacing the
patella medially is more difficult than laterally. However, Keblish (2003) has successfully
used this approach in TKA for fixed valgus deformity.
The valgus deformity is common in patients with JRA and ligament asymmetry with
longstanding deformities is often seen. In the valgus knee the lateral collateral ligament is
shorter than the medial and requires releasing of the lateral capsule and iliotibial band from
the femur. This procedure is necessary to obtain a good soft tissue balance. Sometimes
release of the tight posterolateral corner is needed. In marked valgus deformity the medial
structures are carefully preserved (non-touch).
33
For good long-term results the prosthetic components must be installed accurately. This goal
is achieved by a good surgical technique and experience. A proper prosthesis design and
instrumentation are also important. There are several manufacturers of knee implants with
different designs and instrumentations; both intra- and extramedullar guidance are provided.
The accuracy of the classical, surgeon-controlled operative technique has been questioned as
computer-assisted navigation and implantation of total knee prostheses has become possible.
The navigation system allows a significant improvement in the quality of implantation of a
total knee prosthesis in comparison to that obtained with classical, surgeon-controlled
instrumentation, and the long-term outcome may be improved (Jenny and Boeri 2001).
2.6.3.2. Development of knee prostheses
In 1861, Ferguson reported his experience of resection arthroplasty of the arthritic knee joint.
Two years later Verneuil performed the first interposition arthroplasty of the knee. He
inserted a flap of joint capsule between the resected joint surfaces to prevent them from
growing together. Later, many other surgeons tried similar procedures and used skin, muscle,
fat etc. as interposition material. The success of the arthroplasties was poor. In the 1940s,
metallic hemiarthroplasty of the femur and later of the tibia were developed without success.
The first attempts to replace both the femoral and the tibial articular surfaces were made in
the 1950s; hinged implants were used and the rate of complications was high (Tooms 1987).
Modern joint replacement surgery of the knee was developed in the beginning of 1970s when
bi- and tricompartmental prostheses were designed. These prostheses usually consist of a
metallic femoral component and a metal-backed high-density polyethylene tibial component
(Tooms 1987). Recently, an all-polyethylene tibial component has been shown to yield a
similar clinical and radiographic outcome in the short-to-medium term as the metal-backed
tibial component (Gioe and Bowman 2000, Forster 2003).
The cruciate-retaining (CR) and posterior-stabilized (PS) total knee prosthesis represents a
prosthetic design for a wide variety of knee deformities. For correction of severe knee
deformities, an alternative to the hinge implant is the total condylar III (TC III) prosthesis
(Johnson and Johnson, Braintree, MA, USA), which can substitute for a deficient collateral
ligament. The TC III is a semi-constrained, unlinked knee prosthesis that provides medial
34
and lateral stability through a rectangular tibial post and a high femoral box. The major
indications for the TC III prosthesis is medial ligament loss, severe valgus or combined
deformities.
There is the 20 years experience of the New Jersey low-contact stress (LCS) mobile-bearing
TKA (DePuy Orthopaedics Inc, Warsaw, Ind, USA). Mobile-bearing knee arthroplasty
(MBKA) may have advantages compared with conventional fixed-bearing TKA by allowing
unconstrained axial rotation, which can offer greater articular conformity without an
increased risk of loosening due to increased axial torque. Increased articular conformity
minimizes polyethylene contact stress, thereby reducing linear wear and the risk of
subsurface fatigue failure. Despite these advantages, the long-term clinical results obtained
with current mobile-bearing devices are similar to those obtained with well-designed fixed-
bearing TKA prostheses, with no data suggesting their superiority. The disadvantages of
MBKA are risk of dislocation or breakage of bearing, soft-tissue impingement and a steep
technique learning curve. It is essential that flexion and extension gaps are controlled to
maintain contact pressure on such bearings to avoid problems of subluxation or dislocation
(Vertullo et al. 2001).
Huang et al. (2003) reported in their long-term study of 495 LCS mobile-bearing knee
prostheses that the survival was 88% at 15 years. The survival rate was 83% for the
meniscal-bearing prostheses and 92% for the rotating-platform prostheses. The LCS mobile-
bearing knee prosthesis was not superior to fixed-bearing knees, especially for the meniscal-
bearing design in prevention of polyethylene failure or revision.
Ranawat et al. (2004) compared the fixed-bearing press-fit condylar Sigma (PFC Sigma)
(DePuy Orthopedics, Warsaw, IN, USA) TKA to the recently introduced rotating-platform
version of the same design in a study including 26 patients. At an average follow-up time of
46 months for the fixed-bearing side and 16 months for the rotating-platform side, no
significant differences emerged in terms of knee preference, knee pain, range of motion,
overall satisfaction, or Knee Society Scores (KSS). No revisions, subluxations, dislocations,
or infections occurred. The results of the PFC Sigma rotating-platform, posterior-stabilized
total knee, showed excellent patient satisfaction at 1-year follow-up and were comparable
with the clinical and radiographic results of the fixed-bearing version.
35
2.6.3.3. Patellar resurfacing in TKA
Resurfacing of the patella in total knee arthroplasty (TKA) is still controversial (Holt and
Dennis 2003, Burnett and Bourne 2004), and in many reports it is not routinely
recommended (Barrack et al. 2001, Muller and Wirz 2001). In some reports resurfacing of
the patella has routinely been recommended in association of TKA (Kolettis and Stern 1992,
Kajino et al. 1997).
Patients with JRA form a special group that differs in many respects from other candidates
for knee replacement. There are several morphological differences between JRA and adult
onset RA (Ansell 1977, Hyman and Gregg 1991). Many authors consider that patellar
resurfacing should always be performed in patients with JRA (Sarokhan et al.1983, Scott
1990, Hyman and Gregg 1991).
The outcome of TKA in patients with JRA was first reported in the literature in 1983
(Sarokhan et al.1983). There were 29 knee arthroplasties which were followed for a mean of
five years. Four of six reoperations (18%) were performed due to patellofemoral pain
requiring subsequent patellar resurfacing, and the procedure was successful in all of them. In
a series of 22 cementless total knee arthroplasties in JRA presented by Boublik et al. (1993),
one reoperation was performed due to failure of a metal-backed patellar component. Dalury
et al. (1995) published a follow-up evaluation of TKA in patients less than 45 years of age.
The series comprised 103 knees in 67 patients followed for a mean of 7.2 years. Fifty-eight
percent of the patients had RA and 29 % JRA. The patella was resurfaced in all patients with
inflammatory joint disease. The patellar component of two knees was revised, one for
loosening and one for an acute avulsion fracture of the quadriceps insertion.
The risk of complications has been reported to be high, especially with the use of metal-
backed patellar implants (Bayley et al. 1988, Martin et al. 1997, Schai et al. 1998, Berger et
al. 2001, Crites and Berend 2001, Kraay et al. 2001). Typically, prosthesis failure has been
caused by wear, polyethylene dissociation, metal-induced synovitis or fracture (Bayley et al.
1988). Metal debris may play a role in the pathogenesis of late prosthesis infection (Petrie et
al. 1998). Schroder et al. (1996) reported similar results with metal-backed components in a
series of 51 AGC knee replacement in patients with RA.
36
2.6.3.4. Results of TKA
Several analyses of the survival of TKA have been performed during last years. Follow-up
needs to be started already at the time of installation of the prosthesis and is concluded at the
time of reoperation of any part of prosthesis, death of the patient or some other agreed end-
point. In this way, the outcome of TKA can assessed be reliably. Most of studies comprise
mixed patient materials, and thus all the TKAs performed with variable diagnoses are
included.
In 2001, 4909 primary TKAs were performed in Finland. Osteoarthrosis (OA) was the most
common indication for operation (about 85%), and approximately 400 in patients with RA.
At the same time, 406 revision knee arthroplasties were performed. During the years 1992-
2001 the three most commonly used cemented knee prostheses for OA and RA in Finland
were Duracon (Howmedica, Rutherford, NJ, USA), AGC V2 (Biomet, Warsaw, IN, USA)
and PFC (DePuy Orthopedics, Warsaw, IN, USA) (Nevalainen et al. 2003). Correspondingly,
AGC, Duracon and Kinemax (Howmedica, Rutherford, NJ, USA) were most-favored knee
prostheses in Sweden in 1988-1997 (Robertsson et al. 2001). In the Swedish knee
arthroplasty register loosening appeared to be the commonest reason for primary revision
(44%). Infection was the indication in 15% of all revisions in OA and 22% in patients with
RA (Robertsson et al. 2001).
A survival analysis of 11,606 primary TKAs carried out between January 1, 1978, and
December 31, 2000, was performed in the Mayo Clinic. Survival was 91% at 10 years (2943
knees), 84% at 15 years (595 knees), and 78% at 20 years (104 knees) following surgery.
Prosthetic survival at 10 years was 83% in patients with less than 55 years of age compared
with 94% in those over 70 years of age, 90% in those with OA compared with 95% for those
with RA, and 91% in those with retention of the posterior cruciate ligament compared with
76% in those with substitution of the posterior cruciate ligament (a posterior stabilized
prosthesis). Survival at 10 years was 92% for the non-modular metal-backed tibial
components, 90% for the modular metal-backed tibial components, and 97% for the all-
polyethylene tibial components. Survival at 10 years was 92% for the prostheses fixed with
cement compared with 61% for those fixed without cement (Rand et al. 2003).
37
Osteolysis may occur in association with several types of total knee prostheses. Osteolysis in
association with TKA may be underestimated on routine radiographs because it typically
occurs in cancellous bone of the distal femur and proximal tibia, which has a low
radiodensity, and because it is often obscured on radiographs by the metallic prosthesis.
Osteolysis of the distal femur is best seen on lateral radiographs and often involves the
posterior condyles. Osteolysis of the tibia occurs often along the access tracks, such as
screws or around well-fixed implants (Berry 2004, Nadaud et al. 2004).
Cemented TKAs have yielded good long-term results. In the study by Rodriguez et al. (1996)
104 cemented total condylar knee arthroplasties with polyethylene tibial components were
performed on 67 patients with RA. At the 15-year follow-up, prosthetic survival was 91%.
The results were good also with respect to pain relief and functional ability. Laskin (1990)
followed 80 TKAs with a total condylar prosthesis in patients with RF-positive RA. A
cumulative prostheses survival of 75% was achieved at 10 years. The major indications for
revision were loosening of the tibial component or late bacteremic seeding from another site.
Pain and radiographic evidence of loosening were considered as indications of failure.
Partio et al. (1993) reported a prosthesis survival of 85 % of Townley-prosthesis at 10 years
in patients with RA. Hämäläinen et al. (1995) reported a rate of 96% for Townley-prosthesis
(704 TKAs) at 8 years and 84% at 15 years in patients with chronic arthritis. The AGC
prosthesis (798 prostheses) gave a survival of 95 % at 8 years. From 1971 through 1987,
9200 (2876 in patients with RA) TKAs were performed at the Mayo Clinic. The probability
of an implant being in situ was 97% at both 5 and 10 years when these four variables were
present: primary TKA, a diagnosis of RA, an age of 60 years or more, and use of a condylar
prosthesis with a metal-backed tibial component and patellar resurfacing (Rand and Ilstrup
1991).
The influence of age on the prosthesis survival following TKA has been investigated in
previous studies. Dalury et al. (1995) examined the success of TKA (103 knees) in patients
under age 45. Eighty seven per cent of the patients had a diagnosis of RA or JRA. The
average follow-up period was 7.2 years. The results demonstrate that the success of TKA in
this patient population is comparable to those for TKA in the elderly. Similar results had
been achieved in a 5- to 18-year follow-up study of cemented total knee arthroplasty of
patients 55 years old or younger (Gill et al. 1997).
38
Nafei et al. (1996) identified a clear difference of survival rate of knee prostheses between
OA and RA patients in a 12-year long-term follow-up study of 348 primary TKA. The
diagnosis was OA in 184 cases and RA in 164 cases. The endpoint was defined as prosthesis
not in situ. The variables considered were age, sex, body mass index, and diagnosis. The
overall cumulative prosthesis survival rate was 92%. The survival rate in the OA group was
significantly higher (97%) than the RA group (87%). None of the other variables affected
survival rate significantly.
The medium-term results have also been good with cementless fixation of the implant.
Schröder et al. (1996) followed 51 cementless AGC knees in patients with RA. The
cumulative success rate after 4 to 5 years was 97%. One tibial component was revised due to
aseptic loosening. They stated that the survival rate of TKA in RA may be longer than in OA
because of lower physical activity in patients with RA. Partio and von Bonsdorff (1994)
examined a mixed patient population (91 OA and 46 RA patients) in whom 153 cementless
Synatomic endoprostheses (DePuy Orthopedics, Warsaw, IN, USA) had been implanted.
The prosthesis survival rate was 99 % at 5 years and 95 % at 6 years. Nielsen et al. (1992)
reviewed 103 cementless AGC 2000 TKAs in unselected OA and RA patients over a follow-
up period of 3 years. In terms of prosthesis survival the cumulative success rate was 97%.
When pain and radiographic loosening were also considered, the success rate was 91%. They
emphasized the importance of a good primary prosthetic fit specially at the tibial side.
According to a report from the Swedish knee arthroplasty register the tibial components
inserted without cement fixation run an increased risk of revision while the risk is not
significantly affected by whether or not the patella has been resurfaced (Robertsson et al.
2001).
There are only a few studies regarding TKA for only JRA patients. Carmichael and Chaplin
(1986) studied 13 patients with 25 TKAs who were followed for an average of 61 months.
No revisions, infections, or loosenings occurred. Ranawat et al. (1983) studied 29 TKAs
among JRA patients for a mean follow-up of 36 months. The overall results were excellent in
13 knees, good in 11, fair in 3, and poor in 2. Of the two failures, one was due to deep
infection and the other to severity of disease and poor patient co-operation.
Sarokhan et al. (1983) followed 29 TKA for an average of 5 years in patients with JRA. The
patients' age at the time of surgery averaged 23 years. The preoperative deformities of the
39
knees ranged from 20° of varus angulation to 35° of valgus. The average preoperative flexion
deformity was 23° and the arc of motion averaged 45°. The mean arc of motion increased by
34°, and in all except one knee the angular deformity had been corrected from 0 to 10° of
valgus angulation. Custom-made components were required for 12 of the 29 knees.
Postoperative manipulation was required for 21 knees. The complications included one late
deep infection and one posterior tibial subluxation. No prostheses had required revision for
loosening.
2.6.3.5. Complications of TKA
Many complications may follow knee replacement surgery: infection (early and late,
superficial and deep), dislocation of the endoprosthesis, wear and loosening of the
endoprosthesis, periprosthetic fracture, malposition, hematoma, neural injury, and thrombo-
embolism.
RA is considered to be a predisposing risk factor for surgical infections (Blackburn and
Alarcon 1991). An increased susceptibility to infections is due to impaired general health
status and immune suppressive medication, such as glucocorticoids. Delayed wound healing
occurs frequently (about 20 %) in patients with RA (Hämäläinen 1985, Partio et al. 1993).
Wound edge necrosis is not uncommon (Hämäläinen et al. 1995). Most complications are
mild and do not require revision of the arthroplasty.
The most common causes for revision surgery in TKA are late infections and aseptic
osteolysis (Rodriguez et al. 1996). Hämäläinen et al. (1995) found that the femoral
components stay quite reliably both in cemented and in cementless series. In contrast only 69
% of the cementless tibial components and 84 % of cemented tibial components were
successfully in situ after 4 years of follow-up.
Hvid et al. (1987) studied a series of 119 total condylar knee arthroplasties in patients with
RA. Sixteen complications, mostly minor, occurred in 14 patients. There were no early
infections. Two prostheses needed to be removed for deep infection, after 3 and 5 years,
respectively. One patient required patellectomy for avascular necrosis. Tibial radiolucencies
were noted in 76% of cases; in 29% they were significant, 2 mm or more.
40
In the study by Laskin (1990) tibial radiolucencies were noted in 85% at 10-years of follow-
up. In this study 19 of the 80 TKAs needed revision. The major indication for revision was
loosening of the tibial component.
41
3. PURPOSE OF THE STUDY
The aim of this thesis was:
1. to investigate the survival of the AGC prosthesis and the factors contributing to it in
JRA.
2. to analyze the clinical and radiographic outcome of AGC total knee arthroplasty
at an intermediate-term follow-up time in patients with JRA.
3. to investigate the influence of secondary amyloidosis on the results of TKA in
patients with JRA.
4. to evaluate the appearance of anterior knee pain in patients with JRA who had
undergone TKA either with the replaced or unreplaced patella.
5. to evaluate the influence of previous synovectomy and epiphyseal stapling on the
timing of TKA in patients with JRA.
42
4. PATIENTS
Data from the same patients were used in all Studies (I-IV).
During the years 1985 – 1995 a total of 77 knee replacements were performed on 52 patients
(45 female) who fulfilled the criteria of JRA (Brewer et al.1977) at the Rheumatism
Foundation Hospital, Heinola, Finland. The mean age at disease onset was 7.8 (1.5 – 16)
years and at the time of operation 32.7 (16 – 64) years. The mean duration of arthritis was 24
(range 10-56) years. Details of the patients are presented in Table 3 (Studies I-IV).
Table 3. Demographic data of 52 patients with juvenile rheumatoid arthritis before knee arthroplasty.
Male Female Total
Number of patients 45 52 7
Number of operated knees
11 66 77
Mean age at time of operation in years (SD)
30 (7) 32 (11) 32 (10)
Mean weight of patients in kg (SD)
60 (14) 50 (11) 52 (12)
Mean length of patients in cm (SD)
161 (12) 154 (11) 155 (11)
Disease onset type (N) - systemic - polyarthritis - oligoarthritis* - JpsA
2 4 1 -
4 34 6 1
6 38 7 1
RF-positive patients (N) 4 15 19
HLA B27-positive patients (N)**
3/6 16/40 19/46
ESR, median (range) 35 (4-94) 33 (7-84) 33 (4-94)
*By the time of TKA the disease type of all the seven patients with oligoarthritis had changed: with five women into polyarthritis and with one woman and one man into spondyloarthritis. **in six patients HLA B 27 had not been checked
43
In Study II patients with SA were studied. In 14 (11 women) of the total population of 52
patients, SA was verified by Congo red staining of tissue samples. In these patients the level
of ESR (mm/h) was 49 (14-84) and of C-reactive protein (mg/l) 46 (12-112). Demographic
data of these 14 patients (22 knees) are presented in Table 4 (Study II). At the time of surgery
of the first knee, the disease-modifying anti-rheumatic drug (DMARD) treatment consisted
of methotrexate in 4, aurothiomalate in 2, and sulfasalazine, azathioprine and clorambucil,
each in 1 patient. They were combined with non-steroidal anti-inflammatory drugs (NSAIDs)
and prednisolone in 6 cases and 2 had prednisolone alone. Thirteen of these 14 patients had
previously been treated with several DMARDs and/or cytotoxics. Pharmacoterapy continued
during the follow-up.
Table 4. Demographic data on the 14 patients (22 knees) with JRA + SA before knee arthroplasty (if bilateral arthroplasty, at the time of the first knee operation).
Variable Mean (range)
Patients (N, female / male) 11 / 3
Age (years) 33 (22 – 54)
Duration of disease, years 25 (10 – 47)
Weight (kg) 50 (32 – 65)
Body mass index (kg/m²) 20 (16 – 24)
ESR (mm/h) 49 (14 – 84)
CRP (mg/l) 46 (12 – 112)
Hemoglobin (g/l) 124 (107 – 155)
44
In Study III the effect of patellar resurfacing on pain reduction after TKA in JRA was
examined and data on 18 patients (23 knees) with resurfaced patella were collected.
In Study IV 52 patients with JRA were divided into three subgroups and the effect of
previous synovectomy and stapling on TKA was studied. The demographic data of the
subgroups are presented in Table 5.
Table 5. Demographic data of 52 patients with JRA and subsequent TKA. Distribution by previous surgery.
No previous knee surgery
Previous synovectomy
Previous epiphyseal
stapling
Both previous synovectomy and stapling*
Patients (N) 24 23 9 4 Female / male 22/2 18/5 8/1 3/1
Knees (N) 39 30 13 5
Mean age at the onset of 6.7 11.3 3.5 3.9 JRA (years), range 1.5-16 1.5-16 1.5-8 1.5-7
Mean disease duration 27.0 21.7 19.8 16.8 before TKA (years), range 10-56 10-37 13-28 13-23
Mean age at the time of 34 31 23 21 TKA (years), range 16-64 18-45 18-30 18-26
Mean preoperative 29 32 36 36 CRP (mg/ml), range 0-80 0-112 4-85 0-73 at the time of TKA
Mean length of the 153 164 146 146 patients (cm), range 130-167 150-174 130-169 136-161 at the time of TKA
Mean weight of the 51 59 40 44 patients (kg), range 28-72 47-78 32-72 34-56 at the time of TKA
*Figures are also included in the groups “previous synovectomy” and “previous epiphyseal stapling”
45
5. METHODS
5.1. Surgical techniques
All patients were operated on using the anteromedial parapatellar approach and by insertion
of the AGC prosthesis. Standard or small components were implanted into 72 knees and
custom-made implants were installed into 5 knees. The arthroplasties were performed using
nonconstrained components with careful soft tissue balancing and filling prevailing bone
defects with cancellous bone from resections (either blocks or morselized bone). The
cementless fixation technique was favored because of the young age of the patients and
following the prevailing general treatment recommendations. The patella was resurfaced in
18 patients, and in 23 of 77 knees (30%), using all-polyethylene components in three knees
and metal-backed components in 20 knees.
During the years 1967-1991 a total of 37 knee synovectomies (30 knees) were performed on
23 out of these 52 patients. All procedures were performed surgically using the classical
method and anteromedial parapatellar approach. The mean age of the patients was 11 (1.5 -
16) years at the time of disease onset and 20 (4 - 42) years at the time of synovectomy.
A total of 13 temporary epiphyseal staplings of the knee due to valgus deformity (12 knees)
or leg length discrepancy (1 knee) were performed on 9 out of 52 patients during the years
1965-1982. The mean age was 4 (1.5 - 8) years at the time of disease onset and 8 (4 - 16)
years at the time of epiphyseal stapling. The knees with valgus deformity were stapled
medially (two vitallium staples) at the distal femoral epiphysis and additional temporary
arrest of the medial proximal tibial epiphysis was used in difficult deformities or in cases
with the risk of imminent closure of epiphyses. The knee with leg length discrepancy was
stapled at the distal femoral epiphysis both medially and laterally.
5.2. Follow-up
Knees were radiographed immediately after surgery, and patients were controlled clinically
after three months, and clinically and radiographically after 1, 4 and 8 years. In connection
with the clinical control the functional status of the patient and knee was assessed and the
range of motion and stability of the knee were checked.
46
A total of 38 patients attended the 4-year follow-up and 10 patients the 8-year follow-up.
Data on the replacement of other joints in these patients were also recorded (Table 6).
Table 6. Number of arthroplasties performed for other joints than knee in the 52 patients with JRA.
Additionally, at the end of the year 1998 a telephone inquiry was performed 3 to 13 years
(7.3 mean) after surgery. In this interview the subjective satisfaction of the patients and the
longevity of the replaced knees were recorded. Two patients could not be contacted and two
patients had died during the follow-up. Therefore, information was collected on 48 patients
(71 replaced knees).
5.3. Radiography
Weight-bearing anteroposterior (ap) radiographs and lateral and tangential views of the
patients´ knees were taken preoperatively, during the immediate postoperative period, and
after 1, 4 and 8 years postoperatively. Preoperatively, the grade of destruction in the knees
was assessed by the method of Larsen et al. (1977).
Radiolucency around the components and position of the components were measured on the
ap and lateral radiographs. In the evaluation, the following variables were measured using a
tangential perspective view (skyline) (Fig. 3 A-C): lateral patellar displacement (LPD)
(Laurin et al. 1979), sulcus angle (SA) (Brattström 1964), sulcus width (SW) and patellar
width (PW) (Kujala et al. 1989). Patellar thickness (PT), length of the patella (LP), and
length of the patellar tendon (LT) were measured on the lateral radiographs (Fig. 4).
Arthroplasties / Patients
Hip 65 /36
Shoulder 8 / 7
Elbow 13 / 9
MCP 12 / 8
CMC I 4 / 3
47
The assessment of the position of the patella was based on the Insall-Salvati index. The
patellar tilt angle (Fig. 5) was measured in the group of patients who had resurfaced patellas
(Grelsamer et al. 1993).
Fig. 3 and 4. Schematic illustrations of skyline and lateral radiograph of knee joint.
48
Fig. 5. Radiograph of metal-backed patellar implant with a 13º patellar tilt angle.
5.4. Statistical methods
Survival analysis was performed using the standard Kaplan-Meier technique with confidence
intervals (Study I).
Differences in the two patient subgroups (patients with JRA and SA and those without SA)
were assessed using the Fisher exact test (Study II).
In the statistical analysis of pain the difference between resurfaced and unreplaced patellas
was assessed using the Fisher-Freeman-Halton Test (Study III).
Differences between the subgroups were tested by analysis of variance (Anova) (Study IV).
49
6. RESULTS
6.1. Study I
The aim of this study was to analyze the survival of the prostheses and results of 77 knee
arthroplasties in 52 patients with JRA who received the nonconstrained AGC prosthesis.
Cemented fixation was used in 4/77 knees. Bone grafts were needed in 15 knees. The patella
was resurfaced with an all-polyethylene component in 3 of 77 knees and with a metal-backed
component in 20 of 77 knees. Moulded tibial components were used in 14 knees and modular
components in 63 knees. Custom-made components were used in 5 knees. Synovectomy had
been performed previously in 30 of 77 knees.
Of 71 knees, 53 (75%) were subjectively excellent, 18 (25%) were fair, and none were poor
in the interview 7.3 (mean) years after surgery. Subjective satisfaction, functional ability, and
pain condition of the operated knees are presented in Table 7. The walking distance in meters
was on average 2,660 (50-10,000 meters).
Table 7. Subjective satisfaction and functional ability of 48 patients with JRA based on an interview performed 3-13 years after TKA *
*Of the 52 patients, 2 died and 2 were not reached. **One knee was excellent and the other fair in 2 patients.
Knees PatientsSubjective contentment Excellent 53 36** Fair 18 14 Poor 0 0 Pain at rest 0 0
Pain on movement No pain 53 32 Mild pain 17 15 Moderate pain 1 1 Hard pain 0 0
Walking aids No aid 34 Crutches or stick 9 Wheelchair out of doors 5
Ability to climb stairs Normal 14 One step at time 28 Does not use stairs 6
50
No deep infections were encountered during the follow-up. Radiolucent lines of 1.0 to 1.5
mm were recorded under 14 tibial components (cementless) but there was no obvious
subsidence and no clinical symptoms. No migration or radiolucencies were detected on the
femoral components.
One knee with preoperative valgus angulation of 33º (Fig. 6 A-C) was revised four years
after implantation because of marked instability and loosening of the components. Two
patients died during the follow-up: the cause of death was unknown in one (the oldest patient
in the series, died at the age of 71 years, 6 years after surgery) and by accident in the other.
Fig. 6 A-C. Patient with knee with valgus angle of 33º before primary arthroplasty (A), and before (B) and after (C) revision arthroplasty due to instability and loosening of prosthetic components.
A B C
51
The overall survival of TKA was 99% (95% CI 92 - 100) during the 13 years follow-up,
when revision surgery or death of the patient were end points. The life-table is presented in
Table 8.
Table 8. Life-table for overall survival of 77 AGC knee arthroplasties in 52 patients with JRA.
Years after operation
No. kneesat start
Failure* Withdrawn** Survival at beginning of interval (%)
0-1 77 0 0 100 1-2 77 0 0 100 2-3 77 0 0 100 3-4 77 0 3 100 4-5 74 1 9 99 5-6 64 0 16 99 6-7 48 0 16 99 7-8 32 0 7 99 8-9 25 0 6 99 9-10 19 0 4 99 10-11 15 0 7 99 11-12 8 0 4 99 12-13 4 0 4 99
*Number of knees revised.
**Number of knees whose potential follow-up time has ended without revision during that interval.
52
6.2. Study II
The aim of Study II was to evaluate whether the presence of SA affects prosthesis survival
after TKA in patients with JRA. In the series of 52 patients one knee was revised four years
after implantation because of marked instability and loosening of the components, but SA
was never documented in this patient. Two patients without SA died during the follow-up.
The mean follow-up of the 14 patients with SA was 7.2 (3.8-10.8) years after knee
arthroplasty. SA was documented a mean of 5.4 years preoperatively (range 14.5 years
before to 6 months after TKA of the first knee). Duration of JRA and the age of the patients
with SA at the time of surgery did not differ significantly from those of the patients without
SA, nor did subjective satisfaction or functional ability (Table 9). Most SA patients were
very satisfied with the operative results, they were able to walk without aids, but had still
some difficulties in climbing stairs. The patients with SA did not have any major renal
complications as massive proteinuria or uremia. There was no difference in the survival rate
after TKA between the patients with JRA and SA and those without SA.
53
Table 9. Subjective satisfaction and functional ability of 14 patients with JRA+SA and 34 patients with JRA but no SA after knee arthroplasty. Data are based on an interview 3 – 13 years after TKA.
Variable Patients with JRA + SA Patients with JRA - SA Walking distance (km), median (range) 1 (0.2 –6.0) 1.8 (<0.1 – 10.0) Subjective satisfaction excellent, n (%) 10 (72) 24 (71) fair, n (%) 4 (28) 10 (29) Poor, n (%) 0 (0) 0 (0) Pain at rest, n (%) 0 (0) 0 (0) Pain on movement no pain, n (%) 10 (72) 22 (65) mild pain, n (%) 4 (28) 11 (32) moderate pain, n (%) 0 (0) 1 (3) hard pain, n (%) 0 (0) 0 (0) Walking aids no aid, n (%) 10 (72) 24 (71) crutches or stick, n (%) 3 (21) 6 (17) wheel chair out of doors, n (%) 1 (7) 4 (12) Ability to climb stairs normal, n (%) 3 (21) 11 (32) one step at a time, n (%) 10 (72) 18 (53) does not use stairs, n (%) 1 (7) 5 (15)
Differences in the two patient subgroups were not statistically significant.
54
6.3. Study III
Study III was performed to elucidate in detail the changes of patella and the patellar tendon
in patients with JRA. The appearance of postoperative anterior knee pain with or without
replacement of the patella was also evaluated. The frequency of patella infera was also
documented as was its impact on anterior knee pain.
The patella was resurfaced in 18 of 52 patients, and in 23 of 77 knees (30%). All-
polyethylene components were used in three knees and metal-backed components in 20
knees (87%). No revision surgery or component failures of the patella were recorded during
the follow-up. Mild radiolucency (radiolucent lines < 1.5 mm) around the patellar component
was observed in five knees with metal-backed components and in one knee with an all-
polyethylene implant. Marked radiolucency (> 1.5 mm) was seen in one knee with a metal-
backed component.
Preoperatively most of the knees (54 of 77; 70%) had a low-riding patella – patella infera
(Fig. 7 A-B) with an Insall-Salvati ratio <1.0, but none had a connection between patella
infera and anterior knee pain. The index was <0.9 in 46 (60%), and <0.8 in 28 knees (36%).
Postoperatively (3 - 13 years after surgery) the corresponding figures were <1 in 54 knees
(70%), <0.9 in 48 knees (62%), and <0.8 in 33 of 77 knees (43%) (Table 10). The length of
the patella was unchanged during the follow-up. Lateral patellar displacement was seen in 17
knees preoperatively and in 23 knees postoperatively. Medial displacement was observed in
11 knees preoperatively and in five knees postoperatively. In the remaining knees the
patellas were in a neutral position (Table 10). The mean preoperative Larsen grade was 4.2
(range 3-5).
55
Fig. 7 A-B. Severe patella infera (A) before surgery (Insall-Salvati index 0.56) and (B) after surgery.
A B
Anterior knee pain was present in 14 of 30 patients (47%) with an unreplaced patella and in
two of 18 patients (11%) with patella resurfacing (25% of the knees: in 33% of the
unreplaced patella group, and in 9% with resurfaced patella (p=0.05)). Only two of 18
painful knees (11%) (both with a metal-backed patella) had a replaced patella. Altogether
53 (75%) out of 71 knees - or 32 (67%) out of 48 patients - were painless. Thirty-four of 48
patients (71%) did not need any walking aids. Mild anterior knee pain was present in 17
knees and moderate pain in one.
Only one revision arthroplasty was needed. This was because of instability, but no revision
procedures were needed for the replaced patellas, and no later resurfacing was done for the
unreplaced patellas. Radiolucencies around the patellar component were present in 13% of
the knees. Patient satisfaction was higher with the replaced patella, and no complications
related to resurfacing occurred. Patella infera was common.
56
Table 10. Radiographic parameters of patella and patellar tendon of 77 knees in 52 patients with JRA.
Parameter Preoperative Postoperative
Insall-Salvati index (LT/LP), mean, (range) 0.89 (0.56-1.51) 0.89 (0.47-1.90) Sulcus angle (degrees) 145 (128-160) not measured Sulcus width (mm) 38.3 (27-49) not measured Lateral patellar tilt angle, number 15*
(degrees, mean [range] ) not measured 8.8 (3-20) Patellar thickness (mm) 21 (12-29) not measured Patellar width (mm) 46 (30-60) 46 (30-60) Lateral patellar displacement, number 17 23
(mm, mean [range] ) 9.1 (2-25) 6.3 (2-13) Medial patellar displacement, number 11 5
(mm, mean [range] ) 4.8 (2-10) 4.6 (2-8)
* Five of the 23 replaced patellas were neutral and three had a medial tilt.
57
6.4. Study IV
The purpose of Study IV was to investigate the impact of previous knee synovectomy or
stapling of knee epiphyses on the timing of TKA in patients with JRA. Seventy-seven AGC
TKAs were performed on 52 patients with JRA. TKAs were divided into three subgroups:
knees with previous synovectomy (PS), knees with previous stapling of epiphyses (ES), and
knees without previous surgery (NS).
In the NS subgroup (39 knees, 28 patients) the mean duration of arthritis was 27 (10-56)
years, and the mean age of the patients at the time of TKA 34 (16-64) years, in the PS
subgroup (30 knees, 23 patients) 22 (10-37) years and 31 (18-45) years, and in the ES
subgroup (13 knees, 9 patients) 20 (13-28) years, and 23 (18-30) years, respectively. Four
patients (4 knees) underwent both epiphyseal stapling and knee synovectomy. The mean age
of these patients was 4 (1.5-7) years at the disease onset, and TKAs were performed at the
mean age of 21 (18-26) years. One previous synovectomy had been performed on 26 knees
(19 patients) and the age of these patients at the time of TKA was 33 (18-45) years.
Resynovectomies (2-3 synovectomies) had been performed on four knees (4 patients). The
mean age of these patients was 24 (18-35) years at the time of TKA. There was a significant
difference between the subgroups with respect to age at disease onset (p<0.001) and age of
patients at the time of TKA (p=0.006) and with disease duration (p=0.032).
The need for stapling due to valgus deformity in childhood paralleled with early disease
onset and with early arthroplasty in adulthood. The age of the patients at TKA without
previous surgery was higher than in the other subgroups. The distribution of subgroups and
the temporal connection between synovectomies and temporary epiphyseal arrest and TKA
are presented in Fig 8.
58
Fig. 8. Subgroups of patients with JRA with respect to the ages at disease onset and at the time of TKA. Knees with both synovectomy and epiphyseal stapling are included only in the subgroup of knees with epiphyseal stapling. In the subgroup of knees with previous stapling, one knee with leg length discrepancy was excluded; as well as in the subgroup of knees with previous synovectomy those with several (>2) synovectomies.
59
7. DISCUSSION
In Finland, two thirds of the patients with JRA are referred to the Rheumatism Foundation
Hospital at some phase of disease and the treatment of most of them also monitored there.
This percentage has not changed significantly during last decades. The RFH is specialized on
the surgical treatment of these patients and has gained great experience in this challenging
patient category. Knee surgery of these patients in Finland has been centralized to the RFH,
and this makes it possible to collect a comprehensive series of patients.
The patients of the present study do not - except for the mean age of patients at disease onset
(7 years) - present a typical subtype distribution in our country. Patients positive for RF
(37%) or HLA B27 (41%) are exceptionally numerous which is not surprising because both
factors are associated with active disease (Andersson-Gäre and Fasth 1995, Savolainen et al.
1998). In this series there are two typical patient types, which are candidates for TKA. Most
of these patients have usually seropositive polyarthritis; and a few (in this material two)
whose disease in the beginning affects relentlessly only few joints, and who subsequently
developed spondyloarthopathy.
Deformities and the functional deficits due to JRA may affect adversely the patient´s
physical and emotional development (Bisla et al. 1976, Sarokhan et al.1983). Patients are
young and enter a period of their lives in which education and socialization are crucial and
they have high demands on ambulation (Sarokhan et al.1983, Cage et al. 1992). Only
restorative surgery is able to keep them active in society. Patients may have multiple joint
involvement and they often have to undergo several joint reconstructions. In addition to TKA
70% of the patients in the present series had also undergone hip replacement during follow-
up, 17% elbow arthroplasty and 13% shoulder arthroplasty.
7.1. Survival of knee replacement in patients with JRA
The outcome of TKA in patients with JRA was first reported in a series of 29 knee
arthroplasties by Sarokhan et al. 1983. Six had to be reoperated, four of six reoperations were
done because patellofemoral pain subsequently required patellar resurfacing. In the study by
Boublik et al. (1993) one reoperation was done because of failure of a metal-backed patellar
component. Dalury et al. (1995) published a follow-up of total knee arthroplasty in patients
60
younger than 45 years. There were 103 knees, and 29 % of the patients had JRA. The patella
was replaced in all patients with inflammatory joint disease. In two knees, the patellar
component was subsequently revised, one for loosening and one for an acute avulsion
fracture of the quadriceps insertion.
The failure and complication rates of knee arthroplasty in patients with JRA have been
relatively high in the early reports (Sarokhan et al. 1983, Ranawat et al. 1983, Rydholm et al.
1985, Cage et al. 1992), and the attitude toward TKA in these patients has been cautious
(Rydholm et al. 1985). Surgery is indicated for correction of deformity, for improving the
range of motion and for relieving pain. Knee surgery in patients with JRA is demanding and
requires careful preoperative planning and often the use of custom made components. The
present study on 77 TKAs in 52 patients with JRA seems to be the most extensive study on
this subject in the literature. In this patient series, only one revision was performed due to
loosening of both the tibial and the femoral components in a patient who had marked
preoperative valgus deformity. In retrospect, this revision might have been avoided with the
use of more constrained components and an exact correction of the mechanical axis. Patients
with JRA tend to have low body weight and they usually strain their replaced joints only
moderately due to disability and poor function of other joints (Rydholm et al. 1985), and
reduced general condition. This may enhance the survival of arthroplasty (Ranawat et al.
1983). The durability of the implant has great demands, however, because the life expectancy
in these patients may exceed 40 years (Cage et al. 1992). Modern development of implants
has improved the long-term results of TKA in patients with JRA. Our results from the non-
constrained AGC prosthesis after a follow-up period of 13 years are encouraging, but a
longer follow-up period is still needed.
7.2. Patellar resurfacing in patients with JRA
In some reports resurfacing of the patella has been recommended in connection with knee
arthroplasty of patients with JRA (Sarokhan et al. 1983, Hyman and Gregg 1991). In the
present series patellar resurfacing was performed only in 30% of operations to minimize the
complication rate; the risk of complication was especially high with the use of metal-backed
patellar implants (Bayley et al. 1988, Petrie et al. 1998, Berger et al. 2001, Crites and Berend
2001, Kraay et al. 2001). Typically, the failure has been caused by wear, polyethylene
dissociation, and metal-induced synovitis or fracture (Bayley et al. 1988). It also has been
61
shown that metal debris may play a role in the pathogenesis of late prosthesis infection
(Petrie et al. 1998). Schroder et al. (1996) reported similar results with metal-backed
components in their series of 51 AGC TKA in patients with RA.
Whether to resurface the patella or not in connection with TKA is a dilemma for many
orthopedic surgeons. Two studies have shown that routine resurfacing has not yielded real
benefit in patients with OA (Barrack et al. 2001, Muller and Wirz 2001). However, the
situation is different among patients with RA, and resurfacing is often preferred in these
patients (Sledge and Walker 1984, Kajino et al. 1997). Patients with JRA are in many aspects
a unique group of patients (Ansell 1977, Sarokhan et al. 1983, Swan 1990, Hyman and Gregg
1991). There are strong advocators of routine resurfacing of the patella in patients with JRA
(Sarokhan et al. 1983, Scott 1990, Hyman and Gregg 1991). In the current series patient
satisfaction was higher among those who had their patella replaced, and there were no
complications related to resurfacing. These results imply that resurfacing the patella in
patients with JRA is advisable.
7.3. Patella infera in patients with JRA
Patella infera may occur after knee fractures (Mariani et al. 1994), high tibial osteotomy
(Okamoto et al. 1993), reconstruction of the anterior cruciate ligament (Dandy and Desai
1994) and, sometimes, after TKA (Koshino et al. 1990, Weale et al. 1999). Patella infera is
suggested to exist when the Insall-Salvati index is reduced by 10% or more (Koshino et al.
1990, Noyes et al. 1991, Weale et al. 1999). Schlenzka and Schwesinger (1990) assessed the
variation of this index in a healthy adult population and reported that the distribution is large:
from 0.72 to 1.38. Fern et al. (1992) reported a significant relationship between patella infera
and anterior knee pain in patients with RA, and recommended selective resurfacing of the
patella only in patients with a low-riding patella. Patella infera occurs rarely, except as a
consequence of surgery, but it is often encountered in association with JRA (Scott 1990).
In the current study, the patella was preoperatively low-riding in most patients. The Insall-
Salvati index was <1 in 70% of the knees. Patella infera reflects problems with the extensor
system and predisposes to flexion restrictions. Severe patella infera may cause perioperative
difficulties, and osteotomy of the tibial tubercle may be necessary (Cameron and Jung 1988).
Patella infera also favors resurfacing and thinning of the patella in JRA. Moreover, we
62
assessed whether patella infera had impact on the anterior knee pain, but there was no
connection between the patella infera and anterior knee pain.
7.4. TKA and secondary amyloidosis in patients with JRA
The pharmacological treatment of patients with chronic arthritic diseases has become more
intensive during recent decade, and patients are often treated with combinations of DMARD.
Intra-articular glucocorticoid injections are liberally used. It appears that effective treatment
reduces the progressiveness of SA (David et al. 1993). The patients with SA in the present
study did not have any major renal complications or massive proteinuria.
In the study reported by Lehtimäki et al. (1999), SA appeared to impair the survival of hip
arthroplasty in patients with RA. The arthroplasties were performed between the years 1971
and 1991, i.e. mainly before the period of intensified medical treatment. In the present study
there was no difference in the survival rate after TKA between the patients with JRA with SA
and those without SA. If the patients with JRA and SA are treated effectively, reconstructive
surgery with endoprosthetic joint replacement yields excellent results and the patients
remains active in the society. The duration of the disease and the age of patients with SA at
the time of surgery did not differ significantly from patients without SA and no differences
were found in the subjective satisfaction or functional ability, either. Thus, the results of
TKA in patients with JRA and SA appeared to be as good as of the patients with no signs of
SA.
7.5. Synovectomy and epiphyseal stapling in patients with JRA
Patients with severe JRA are frequently young when severe knee involvement is actual, and
before TKA every effort should be made to postpone the need of TKA. These efforts include
effective pharmacological treatment, synovectomy and temporary epiphyseal arrest with
stapling.
The need for synovectomy in JRA has been declining because of improved medical treatment
of JRA. If conservative treatment with intra-articular glucocorticoids and anti-rheumatic
drugs has failed after 3-6 months, surgical synovectomy is indicated (Jensen et al. 1991). In
the current study, all synovectomies were performed with the classical open technique.
63
However, today most synovectomies are performed artroscopically, as is the case for the
RFH. The condition of the knee joint before synovectomy is of utmost importance. The
outcome of synovectomy depends on the preoperative stage of joint destruction.
Synovectomy is early when it is performed on a joint with grade 0-2 involvement according
Larsen (Larsen et al. 1977) and late if the grade is 3-5. Jensen et al. (1991) reported the
results of early and late synovectomy in 55 knees in patients with RA. Early synovectomy
showed significant pain relief and improvement of knee function. Late synovectomy was
classified as a palliative intervention; the need for TKA was usually postponed by 5 to 10
years but there was no evidence of retarding of destruction of bone.
Temporary epiphyseal arrest with staples was introduced in 1949 by Blount and Clarke for
the management of knee malalignment and in 1956 at the RFH in Finland. By the end of
1999, a total of 119 patients had undergone the procedure at the RFH (Skyttä et al. 2003a).
In many of the patients in the present study disease activity had been high which had resulted
in arthritic joint destruction and often in the need for early arthroplasty. A local growth
disturbance, characterized by a valgus deformity, reflects the severity of disease. Patients are
frequently short and their body weight is low. The present study indicates that short stature of
a patient is associated with disease severity and local growth disturbances and thus with a
young age of patients at the time of TKA.
The present study shows that the need for stapling in pediatric patients who have a valgus
deformity of the knee is associated with arthroplasty in early adulthood. Stapling itself does
not cause joint destruction but is a sign of severe disease. The age of the patients in the
present study was higher at the time of arthroplasty, if there was no history of surgical
procedures on the knee joint compared to other subgroups. Similarly, the age in the subgroup
with a history of synovectomy was significantly higher than in patients with epiphyseal
stapling. Only one revision was performed among these patients. No epiphyseal stapling was
performed on this knee, and retrospectively the survival of TKA might have been enhanced
by temporary epiphyseal arrest in childhood. Temporary epiphyseal stapling is a safe and
effective method for the management of the valgus deformity of the knee of patients with
JRA (Arden 1983, Rydholm et al. 1987).
64
7.6. Knee surgery in the future
Pharmacotherapy for treating JRA has developed rapidly during the last decade, but whether
it will have impact on the development of joint destruction is still unknown. Improved
medication may prevent difficult deformities and general impairment in the future, but the
need for joint reconstructions will exist for years to come.
In the early 1990s the concept of minimally invasive surgery (MIS) was introduced through
the unicondylar prosthesis. Later the same technique was introduced for TKA. Tria (2003)
performed 120 minimally invasive TKAs and early results show that MIS produces better
early motion, less blood loss, less pain, and a shorter hospital stay than the standard TKA
with no compromise in accuracy. However, MIS has not been employed for treating patients
with JRA.
In the future, the use of navigation instrumentation (Jenny and Boeri 2001) and mini-
exposures may improve the outcome of TKA and shorten the postoperative recovery period,
and new materials have been developed for hip surgery and these may be utilized also for
TKAs. In the present study excellent results were achieved using the classical knee
replacement technique and non-constrained components. Careful and skilful surgery (all
operations were performed by senior orthopedic surgeons) and a knowledge of the
characteristics of JRA have contributed to these excellent results.
65
8. CONCLUSIONS
1. The overall survival for TKA was 99% (95% CI, 92 - 100) during the 13 years of
follow-up when revision surgery or death of the patient were end points. Patients with
JRA have a low body weight and they usually strain only moderately their replaced
joints due to reduced mobility of other joints and a compromised general condition.
This probably enhances the success of knee arthroplasty.
2. Of 71 knees, 53 (75%) were subjectively excellent, 18 (25%) fair, and none were
poor according to patient interviews. Only one revision was performed necessitated
by loosening of both the tibial and the femoral components. Radiolucent lines of 1.0
to 1.5 mm were seen under 14 tibial components (cementless) but without obvious
subsidence or clinical symptoms. No migration or radiolucencies were detected on the
femoral components. Mild radiolucency (radiolucent lines < 1.5 mm) around the
patellar component was observed in five knees with metal-backed components and in
one knee with an all-polyethylene implant. Marked radiolucency (> 1.5 mm) was seen
in one knee with a metal-backed component.
3. The satisfaction and the functional ability of the patients with SA did not differ
significantly from those of the patients without SA. Most SA patients were very
satisfied with the operative results. The patients did not develop any major organ
failure, like severe uremia or massive proteinuria. The results of TKA in the patients
with JRA and SA was excellent.
4. Patellar resurfacing reduced anterior knee pain among the patients in this study.
Patient satisfaction was higher in the patellar resurfacing group than in the
unresurfaced group.
5. The need for stapling to treat valgus deformity in childhood was associated with early
disease onset and a need for arthroplasty in early adulthood. The age of the patients
who had no previous knee surgery by the time of TKA was higher than in other
subgroups. Similarly, the average age of the patients in the subgroup who had
undergone synovectomy was significantly higher than in the epiphyseal stapling
subgroup.
66
9. REFERENCES
Adamec O, Dungl P, Kasal T, Chomiak J (2002): Knee joint synovectomy in treatment of juvenile idiopathic arthritis. Acta Chir Orthop Traumatol Cech 69: 350-356.
Andersson-Gäre B (1999): Juvenile arthritis – Who gets it, where and when ? A review of current data on incidence and prevalence. Clin Exp Rheumatol 17: 367-374.
Andersson-Gäre B, Fasth A (1995): The natural history of juvenile chronic arthritis: a population based cohort study: II. outcome. J Rheumatol 22: 308-319.
Andersson-Gäre B, Fasth A (1992): Epidemiology of juvenile chronic arthritis in Southwestern Sweden: a 5-year prospective study. Pediatrics 90: 950-958.
Ansell BM, Bywaters EGL (1956): Growth in Still´s disease. Ann Rheum Dis 15: 295-319.
Ansell BM (1977): Joint manifestations in children with juvenile chronic polyarthritis. Arthritis Rheum 20 (Suppl): 204-206.
Ansell BM (1987): Juvenile chronic arthritis. Scand J Rheumatol 16(suppl66): 47-50.
Arden GP (1983): Surgical treatment of Still’s disease (juvenile chronic arthritis). Ann Acad Med Singapore 12: 174-184.
Ayral X, Bonvarlet JP, Simonnet J, Amor B, Dougados M (1997): Arthroscopy-assisted synovectomy in the treatment of chronic synovitis of the knee. Rev Rhum Engl Ed 64: 215-226.
Balogh K, Jr, Kunin AS (1971): The effect of cortisone on the metabolism of epiphyseal cartilage. Clin Orthop 80: 208-215.
Barrack RL, Bertot AJ, Wolfe MW, Waldman DA, Milicic M, Myers L (2001): Patellar resurfacing in total knee arthroplasty: A prospective, randomised, double-blind study with five to seven years of follow-up. J Bone Joint Surg 83A: 1376-1381.
Bayley JC, Scott RD, Ewald FC, Holmes Jr GB (1988): Failure of the metal-backed patellar component after total knee replacement. J Bone Joint Surg 70A: 668-674.
Berger RA, Lyon JH, Jacobs JJ, Barden RM, Berkson EM, Sheinkop MB, Rosenberg AG, Galante JO (2001): Problems with cementless total knee arthroplasty at 11 years followup. Clin Orthop 392: 196-207.
Berntson L, Fasth A, Andersson-Gäre B, Kristiansson J, Lahdenne P, Marhaug G, Nielsen S, Pelkonen P, Svensson E (2001): Construct validity of ILAR and EULAR criteria in juvenile idiopathic arthritis: a population based incidence study from the Nordic countries. J Rheumatol 12: 2737-2743.
67
Berry DJ (2004): Recognizing and identifying osteolysis around total knee arthroplasty. Instr Course Lect 53: 261-264.
Bisla RS, Inglis AE, Ranawat CS (1976): Joint replacement surgery in patients under thirty. J Bone Joint Surg 58A: 1098-1104.
Blackburn W D, Alarcon G S (1991): Prosthetic joint infections. Arthritis Rheum 34: 110-117.
Blount WP, Clarke GR (1949): Control of bone growth by epiphyseal stapling: a preliminary report. J Bone Joint Surg 31A: 464-478.
Boublik M, Tsahakis PJ, Scott RD (1993): Cementless total knee arthroplasty in juvenile onset rheumatoid arthritis. Clin Orthop 286: 88-93.
Brattström H (1964): Shape of the intercondylar groove normally and in recurrent dislocation of patella: A clinical and x-ray anatomical investigation. Acta Orthop Scand 68: 1-148.
Brewer Jr EJ, Bass J, Baum J, Cassidy JT, Fink C, Jacobs J, Hanson V, Levinson JE, Schaller J, Stillman JS (1977): Current proposed revision of JRA criteria. Arthritis Rheum 20: 195-199.
Burk DL Jr, Mitchell DG, Rifkin MD, Vinitski S (1990): Recent advances in magnetic resonance imaging of the knee. Radiol Clin North Am 28: 379-393.
Burnett RS, Bourne RB (2004): Indications for patellar resurfacing in total knee arthroplasty. Instr Course Lect 53: 167-186.
Cage DJN, Granberry WM, Tullos HS (1992): Long-term results of total arthroplasty in adolescents with debilitating polyarthropathy. Clin Orthop 283: 156-162.
Cameron HU, Jung YB (1988): Patella baja complicating total knee arthroplasty: A report of two cases. J Arthroplasty 3: 177-180.
Carmichael E, Chaplin DM (1986): Total knee arthroplasty in juvenile rheumatoid arthritis. A seven-year follow-up study. Clin Orthop 210: 192-200.
Cassidy JT (1980): Juvenile rheumatoid arthritis. p. 1279. In Kelly WN, Harris ED, Ruddy S, Sledge CB (eds): Textbook of rheumatology. Philadelphia, WB Saunders.
Cassidy JT, Levinson JE, Bass JC, Baum J, Brewer EJ Jr, Fink CW, Hanson V, Jacobs JC, Masi AT, Schaller JG, et al. (1986): A study of classification criteria for a diagnosis of juvenile rheumatoid arthritis. Arthritis Rheum 29: 274-281.
Cassidy JT, Levinson JE, Brewer JE (1989): The development of classification criteria for children with juvenile rheumatoid arthritis. Bull Rheum Dis 38: 1-7.
Cassidy JT (1999): Osteopenia and osteoporosis in children. Clin Exp Rheumatol 17: 245-250.
68
Chandnani VP, Ho C, Chu P, Trudell D, Resnick D (1991): Knee hyaline cartilage evaluated with MR imaging: a cadaveric study involving multiple imaging sequences and intraarticular injection of gadolinium and saline solution. Radiology 178: 557-561.
Clarke HD (2003): Tibial tubercle osteotomy. J Knee Surg 16: 58-61.
Cohen S, Jones R (1987): An evaluation of the efficacy of arthroscopic synovectomy of the knee in rheumatoid arthritis: 12-24 month results. J Rheumatol 14: 452-455.
Crites BM, Berend ME (2001): Metal-backed patellar components: A brief report on 10-year survival. Clin Orthop 388: 103-104.
Dalury DF, Ewald FC, Christie MJ, Scott RD (1995): Total knee arthroplasty in a group of patients less than 45 years of age. J Arthroplasty 10: 598-602.
Dandy DJ, Desai SS (1994): Patellar tendon length after anterior cruciate ligament reconstruction. J Bone Joint Surg 76B: 198-199.
David J, Vouyiouka O, Ansell BM, Hall A, Woo P (1993): Amyloidosis in juvenile chronic arthritis: a morbidity and mortality study. Clin Exp Rheumatol 11: 85-90.
De Beer FC, Mallya RK, Fagan EA, Lanham JG, Hughes GR, Pepys MB (1982): Serum amyloid-A protein concentration in inflammatory diseases and its relationship to the incidence of reactive systemic amyloidosis. Lancet 2: 231-234.
Dirienzo G, Osti L, Merlo F (1997): Our experience in the treatment of rheumatoid knee by arthroscopic synovectomy. Chir Organi Mov 82: 275-278.
Fern ED, Winson IG, Getty CJ (1992): Anterior knee pain in rheumatoid patients after total knee replacement. J Bone Joint Surg 74B: 745-748.
Filipowicz-Sosnowska AM, Roztropowicz-Denisiewicz K, Rosenthal CJ, Baum J (1978): The amyloidosis of juvenile rheumatoid arthritis--comparative studies in Polish and American children. I. Levels of serum SAA protein. Arthritis Rheum 21: 699-703.
Flatø B, Aasland A, Vinje 0, Forre O (1998): Outcome and predictive factors in juvenile rheumatoid arthritis and juvenile spondyloarthropathy. J Rheumatol 25: 366-375.
Forster MC (2003): Survival analysis of primary cemented total knee arthroplasty: which designs last? J Arthroplasty 18: 265-270.
Fraser RK, Dickens DR, Cole WG (1995): Medial physeal stapling for primary and secondary genu valgum in late childhood and adolescence. J Bone Joint Surg 5B: 733-735.
French AR, Mason T, Nelson AM, Crowson CS, O'Fallon WM, Khosla S, Gabriel SE (2002): Osteopenia in adults with a history of juvenile rheumatoid arthritis. A population based study. J Rheumatol 29: 1065-1070.
Gertz MA, Kyle RA (1991): Secondary systemic amyloidosis: response and survival in 64 patients. Medicine (Baltimore) 70: 246-256.
69
Giannini EH, Brewer EJ, Kuzmina N, Shaikov A, Maximov A, Vorontsov I, Fink CW, Newman AJ, Cassidy JT, Zemel LS (1992): Methotrexate in resistant juvenile rheumatoid arthritis. Results of the U.S.A.-U.S.S.R. double-blind, placebo-controlled trial. The Pediatric Rheumatology Collaborative Study Group and The Cooperative Children's Study Group. N Engl J Med 326: 1043-1049.
Gibbons CE, Gosal HS, Bartlett J (2002): Long-term results of arthroscopic synovectomy for seropositive rheumatoid arthritis: 6-16 year review. Int Orthop 26: 98-100.
Gill GS, Chan KC, Mills DM (1997): 5- to 18-year follow-up study of cemented total knee arthroplasty for patients 55 years old or younger. J Arthroplasty 12: 49-54.
Gioe TJ, Bowman KR (2000): A randomized comparison of all-polyethylene and metal-backed tibial components. Clin Orthop 380: 108-115.
Grelsamer RP, Bazos AN, Proctor CS (1993): Radiographic analysis of patellar tilt. J Bone Joint Surg 75B: 822-824.
Hafner R, Pieper M (1995): Arthroscopic synovectomy of the knee joint in chronic juvenile arthritis. Z Rheumatol 54: 165-170.
Hämäläinen M (1985): Knee replacement arthroplasty in patients with rheumatoid arthritis. Academic Dissertation. The Publications of Kuopio University (Medical Faculty).
Hämäläinen M, Kammonen M, Repo A, Ikävalko M, Lehtimäki M, Belt E, Kautiainen H (1995): Polven tekonivelleikkaus tulehduksellisissa nivelsairauksissa. Suom Ortop ja Traumat 18: 368-375.
Hämäläinen M. Lapsireumaatikon kirurginen hoito 1995: Suom Ortop ja Traumat 18: 317-325.
Helin HJ, Korpela MM, Mustonen JT, Pasternack AI (1995): Renal biopsy findings and clinicopathologic correlations in rheumatoid arthritis. Arthritis Rheum 38: 242-247.
Holt GE, Dennis DA (2003): The role of patellar resurfacing in total knee arthroplasty. Clin Orthop 416: 76-83.
Honkanen VE, Rautonen JK, Pelkonen PM (1993): Intra-articular glucocorticoids in early juvenile chronic arthritis. Acta Paediatr 82: 1072-1074.
Huang CH, Ma HM, Lee YM, Ho FY (2003): Long-term results of low contact stress mobile-bearing total knee replacements. Clin Orthop 416: 265-270.
Husby G (1985): Amyloidosis and rheumatoid arthritis: Clin Exp Rheumatol 3: 173-180.
Hvid I, Kjaersgaad- Andersen P, Wethelund J-O, Sneppen O (1987): Knee arthroplasty in rheumatoid arthritis. Four- to six-year follow-up study. J Arthroplasty 2: 233-239.
Hyman BS, Gregg JR (1991): Arthroplasty of the hip and knee in juvenile rheumatoid arthritis. Rheum Dis Clin North Am 17: 971-983.
70
Insall J, Salvati E (1971): Patella position in the normal knee joint. Radiology 101: 101-104.
Isacson J, Broström L-Å, Allander E, Czechowskij J (1987): Radiological findings in the rheumatoid knee joint in a seventeen-year follow-up. Scand J Rheumatol 16: 153-159.
Jacobs JC, Hui RM (1977): Cricoarytenoid arthritis and airway obstruction in juvenile rheumatoid arthritis. Pediatrics 59: 292-294.
Jacobsen ST, Levinson JE, Crawford AH (1985): Late results of synovectomy in juvenile rheumatoid arthritis. J Bone Joint Surg 67A: 8-15.
Jenny JY, Boeri C (2001): Computer-assisted implantation of total knee prostheses: a case-control comparative study with classical instrumentation. Comput Aided Surg. 6: 217-220.
Jensen CM, Poulsen S, Ostergren M, Hansen KH (1991): Early and late synovectomy of the knee in rheumatoid arthritis. Scand J Rheumatol 20: 127-131.
Jonsson B, Larson S-E (1991): Functional improvement and costs of hip and knee arthroplasty indestructive rheumatoid arthritis. Scand J Rheumatol 20: 351-357.
Kageyama Y, Miyamoto S, Ozeki T et al. (1998): Outcomes for patients undergoing one or more total hip and knee arthroplasties. Clin Rheumatol 17: 130-134.
Kaipiainen-Seppänen O, Savolainen A (2001): Changes in the incidence of juvenile rheumatoid arthritis in Finland. Rheumatology 40: 928-932.
Kajino A, Yoshino S, Kameyama S, Kohda M, Nagashima S (1997): Comparison of the results of bilateral total knee arthroplasty with and without patellar arthroplasty for rheumatoid arthritis. J Bone Joint Surg 79A: 570-574.
Kastner P, Wessel G (1977): Chemical synovectomy with Varicocid in rheumatoid arthritis--further results. Scand J Rheumatol 6: 28-32.
Keblish PA (2003): The lateral approach for total knee arthroplasty. J Knee Surg 16: 62-68.
Keenan MA, Peabody TD, Gronley JK, Perry J (1991): Valgus deformities of the feet and characteristics of gait in patients who have rheumatoid arthritis. J Bone Joint Surg 73A: 237-247.
Koivuniemi R, Leirisalo-Repo M (1999): Juvenile chronic arthritis in adult life: a study of long-term outcome in patients with juvenile chronic arthritis or adult rheumatoid arthritis. Clin Rheumatol 18: 220-226.
Kolettis GT, Stern SH (1992): Patellar resurfacing for patellofemoral arthritis. Orthop Clin North Am 23: 665-673.
Koshino T, Ejima M, Okamoto R, Morii T (1990): Gradual low riding of the patella during postoperative course after total knee arthroplasty in osteoarthritis and rheumatoid arthritis. J Arthroplasty 5: 323-327.
71
Kotaniemi A, Savolainen A, Kroger H, Kautiainen H, Isomaki H (1999): Weight-bearing physical activity, calcium intake, systemic glucocorticoids, chronic inflammation, and body constitution as determinants of lumbar and femoral bone mineral in juvenile chronic arthritis. Scand J Rheumatol 28: 19-26.
Kotaniemi K, Kautiainen H, Karma A, Aho K (2001): Occurrence of uveitis in recently diagnosed juvenile chronic arthritis: a prospective study. Ophthalmology 108: 2071-2075.
Kraay MJ, Darr OJ, Salata MJ, Goldberg VM (2001): Outcome of metal-backed cementless patellar components: The effect of implant design. Clin Orthop 392: 239-244.
Kujala UM, Österman K, Kormano M, Nelimarkka O, Hurme M, Taimela S (1989): Patellofemoral relationships in recurrent patellar dislocation. J Bone Joint Surg 71B: 788-792.
Kunnamo I, Kallio P, Pelkonen P (1986): Incidence of arthritis in urban Finnish children. A prospective study. Arthritis Rheum 29: 1232-1238.
Kvien TK, Pahle JA, Hoyeraal HM, Sandstad B (1987): Comparison of synovectomy and no synovectomy in patients with juvenile rheumatoid arthritis. A 24-month controlled study. Scand J Rheumatol 16: 81-91.
Laaksonen AL, Laine V (1961): A comparative study of joint pain in adult and juvenile rheumatoid arthritis. Ann Rheum Dis 20: 386-387.
Laiho K, Tiitinen S, Kaarela K, Helin H, Isomäki H (1999): Secondary amyloidosis has decreased in patients with inflammatory joint disease in Finland. Clin Rheumatol 18: 122-123.
Laine H, Mikkelsen OA (1968): Epiphyseal stapling in juvenile rheumatoid gonarthritis. Acta Rheumatol Scand 14: 317-322.
Laine V (1964): Indications and contraindications for orthopedic surgery in rheumatoid arthritis with a special reference to the early synovectomy. Rhumatismes Inflammatoires Chroniques (France): 217-220.
Lantto R, von Wendt L (1985): Juvenile rheumatoid arthritis in Northern Finland. Abstracts of the second International Symposium on Inflammatory Connective Tissue Diseases in Childhood and Adolescence. Prague.
Larsen A, Dale K, Eek M (1977): Radiographic evaluation of rheumatoid arthritis and related conditions by standard reference films. Acta Radiol Diagn 18: 481-491.
Laskin RS (1990): Total condylar knee replacement in patients who have rheumatoid arthritis: ten-year follow-up study. J Bone Joint Surg 72A: 529-535.
Laurin CA, Dussault R, Levesque HP (1979): The tangential x-ray investigation of the patellofemoral joint: X-ray technique, diagnostic criteria and their interpretation. Clin Orthop 144: 16-26.
72
Lehtimäki MY, Kautiainen HJ, Lehto MUK, Hämäläinen MMJ (1999): Charnley low-friction arthroplasty in rheumatoid patients. A survival study up to 20-years. J Arthroplasty 14: 657-661.
Lindegaard H, Vallo J, Horslev-Petersen K, Junker P, Ostergaard M (2001): Low field dedicated magnetic resonance imaging in untreated rheumatoid arthritis of recent onset. Ann Rheum Dis 60: 770-776.
Lovell DJ, Walco GA (1989): Pain associated with juvenile rheumatoid arthritis. Pediatr Clin North Am 36: 1015-1027.
Mäenpää HM, Soini I, Lehto MU, Belt EA (2002): Insufficiency fractures in patients with chronic inflammatory joint diseases. Clin Exp Rheumatol 20: 77-79.
Mariani PP, Del Signore S, Perugia L (1994): Early development of patella infera after knee fractures. Knee Surg Sports Traumatol Arthrosc 2: 166-169.
Martin SD, McManus JL, Scott RD, Thornhill TS (1997): Press-fit condylar total knee arthroplasty: 5- to 9-year follow-up evaluation. J Arthroplasty 12: 603-614.
Matsui N, Taneda Y, Ohta H, Itoh T, Tsuboguchi S (1989): Arthroscopic versus open synovectomy in the rheumatoid knee. Int Orthop 13: 17-20.
McDonagh JE (2001): Osteoporosis in juvenile idiopathic arthritis. Curr Opin Rheumatol 13: 399-404.
Meijers KA, Valkenburg HA, Cats A (1983): A synovectomy trial and the history of early knee synovitis in rheumatoid arthritis. A multicentre study. Rheumatol Int 3: 161-166.
Mielke CH, Stevens PM (1996): Hemiepiphyseal stapling for knee deformities in children younger than 10 years: a preliminary report. J Pediatr Orthop 16: 423-429.
Minden K, Niewerth M, Listing J, Biedermann T, Bollow M, Schontube M, Zink A (2002): Long-term outcome in patients with juvenile idiopathic arthritis. Arthritis Rheum. 46: 2392-2401.
Miranda M, Toso M, Lira L, Norambuena X, Quevedo F (1997): Classification criteria for idiopathic arthritides of childhood (ILAR´s proposal) in Chilean children. Rev Rhum (Engl Ed) 64: 218S.
Morling N, Friis J, Heilmann C, Hellesen C, Jakobsen BK, Jörgensen B, Pedersen FK, Svejgaard A, Thomsen K (1985): HLA antigen frequencies in juvenil chronic arthritis. Scand J Rheumatol 13: 209-216.
Möttönen T, Hannonen P, Leirisalo-Repo M, Nissilä M, Kautiainen H, Korpela M, Laasonen L, Julkunen H, Luukkainen R, Vuori K, Paimela L, Blåfield H, Hakala M, Ilva K, Yli-Kerttula U, Puolakka K, Järvinen P, Hakola M, Piirainen H, Ahonen J, Palvimäki I, Forsberg S, Koota K, Friman C (1999): Comparison of combination therapy with single-drug therapy in early rheumatoid arthritis: a randomised trial. FIN-RACo trial group. Lancet 353: 1568-1573.
73
Muller W, Wirz D (2001): The patella in total knee arthroplasty: Does it matter? 750 LCS total knee replacements without resurfacing of the patella. Knee Surg Sports Traumatol Arthrosc 9: 24-26.
Murray KJ, Lovell DJ (2002): Advanced therapy for juvenile arthritis. Best Pract Res Clin Rheumatol 16: 361-378.
Myllylä T, Peltonen L, Puranen J, Korhonen LK (1983): Consequences of synovectomy of the knee joint; clinical, histopathological and enzymatic changes and changes in 2 components of complement. Ann Rheum Dis 42: 28-35.
Nadaud MC, Fehring TK, Fehring K (2004): Underestimation of osteolysis in posterior stabilized total knee arthroplasty. J Arthroplasty 19: 110-115.
Nafei A, Kristensen O, Knudsen HM, Hvid I, Jensen J (1996): Survivorship analysis of cemented total condylar knee arthroplasty. A long-term follow-up report on 348 cases. J Arthroplasty 11: 7-10.
Nevalainen J, Keinonen A, Mäkelä A (2003): The 2000-2001 Implant Yearbook on Orthopaedic Endoprostheses. The Finnish Arthroplasty Register.
Nielsen PT, Hansen EB, Rechnagel K (1992): Cementless total knee arthroplasty in selected cases of osteoarthritis and rheumatoid arthritis: 3-year follow-up study of 103 cases. J Arthroplasty 7: 137-143.
Nissilä M (1975): Osmic acid treatment for rheumatoid synovitis. Ann Clin Res 7: 202-204.
Nissilä M, Isomäki H, Koota K, Larsen A, Raunio K (1977): Osmic acid in rheumatoid synovitis. A controlled study. Scand J Rheumatol 6: 158-160.
Noyes FR, Wojtys EM, Marshall MT (1991): The early diagnosis and treatment of developmental patella infera syndrome. Clin Orthop 265: 241-252.
Ochi T, Iwase R, Kimura T, Hirooka A, Masada K, Owaki H, Wakitani S, Murata N, Ono K (1991): Effect of early synovectomy on the course of rheumatoid arthritis. J Rheumatol 18: 1794-1798.
Oen K, Malleson PN, Cabral DA, Rosenberg AM, Petty RE, Cheang M (2002): Disease course and outcome of juvenile rheumatoid arthritis in a multicenter cohort. J Rheumatol 29: 1989-1999.
Ogden JA (1982): Skeletal injury in the child. Lea & Febiger, Philadelphia. p. 57.
Ogilvie-Harris DJ, Basinski A (1991): Arthroscopic synovectomy of the knee for rheumatoid arthritis. Arthroscopy 7: 91-97.
Okamoto R, Koshino T, Morii T (1993): Shortening of patellar ligament and patella baja with improvement of quadriceps muscle strength after high tibial osteotomy. Bull Hosp Jt Dis 53: 21-24.
74
Ostergaard M, Szkudlarek M (2001): Magnetic resonance imaging of soft tissue changes in rheumatoid arthritis wrist joints. Semin Musculoskelet Radiol 5: 257-274.
Ostergaard M, Court-Payen M, Gideon P, Wieslander S, Cortsen M, Lorenzen I, Henriksen O (1995): Ultrasonography in arthritis of the knee. A comparison with MR imaging. Acta Radiol 36: 19-26.
Pahle JA (1996): Orthopaedic management of juvenile chronic arthritis (JCA). Z Rheumatol 55: 376-387.
Partio E, von Bonsdorff H, Hatanpää S, Lehto M U K (1993): Anatomic total knee (Townley) arthroplasty in the rheumatoid and osteoarthritic knee: a clinical and radiological study with follow-up from 3.5 to 10 years. J Orthop Rheumatol 6: 91-95.
Partio E, von Bonsdorff H (1994): Cementless Synatomic total knee arthroplasty in rheumatoid and osteoarthritic knees: a clinical and radiological review with follow-up from 4 to 7 years. J Orthop Rheumatol 7: 155-163.
Pätiälä H (1975): On health and economic impact of rheumatoid arthritis in Finland. Dissertation. The publications of Tampere University (Medical Faculty).
Paus AC, Mellbye OJ, Forre O (1990): Immunohistopathologic findings in synovial biopsies before and after synovectomy in patients with chronic inflammatory joint diseases and their relation to clinical evaluation. A prospective study of biopsies taken from areas selected by arthroscopy. Scand J Rheumatol 19: 269-279.
Paus AC, Dale K (1993): 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 82: 55-61.
Petrie RS, Hanssen AD, Osmon DR, Ilstrup D (1998): Metal-backed patellar component failure in total knee arthroplasty: A possible risk for late infection. Am J Orthop 27: 172-176.
Petty RE, Southwood TR, Baum J, Bhettay E, Glass DN, Manners P, Maldonado-Cocco J, Suarez-Almazor M, Orozco-Alcala J, Prieur A-M (1998): Revision of the proposed classification criteria for juvenile idiopathic arthritis: Durban, 1997. J Rheumatol 25: 1991-1994.
Prieur AM, Chedeville G (2001): Prognostic factors in juvenile idiopathic arthritis. Curr Rheumatol Rep 3: 371-378.
Ranawat CS, Bryan WJ, Inglis AE (1983): Total knee arthroplasty in juvenile arthritis. Arthritis Rheum 26: 1140-1144.
Ranawat AS, Rossi R, Loreti I, Rasquinha VJ, Rodriguez JA, Ranawat CS (2004): Comparison of the PFC sigma fixed-bearing and rotating-platform total knee arthroplasty in the same patient: short-term results. J Arthroplasty 19: 35-39.
Rand JA, Ilstrup DM (1991): Survivorship analysis of total knee arthroplasty. Cumulative rates of survival of 9200 total knee arthroplasties. J Bone Joint Surg 73A: 397-409.
75
Rand JA, Trousdale RT, Ilstrup DM, Harmsen WS (2003): Factors affecting the durability of primary total knee prostheses. J Bone Joint Surg 85A: 259-265.
Ravelli A, Manzoni SM, Viola S, Pistorio A, Ruperto N, Martini A. (2001): Factors affecting the efficacy of intraarticular corticosteroid injection of knees in juvenile idiopathic arthritis. J Rheumatol 28: 2100-2102.
Robertsson O, Knutson K, Lewold S, Lidgren L (2001): The Swedish knee arthroplasty register 1975-1997. An update with special emphasis on 41,223 knees operated on in 1988-1997. Acta Orthop Scand 72: 503-513.
Rodriguez JA, Saddler S, Edelman S, Ranawat CS (1996): Long-term results of total knee arthroplasty in class 3 and 4 rheumatoid arthritis. J Arthroplasty 11: 141-145.
Rostropowicz-Denisiewicz K, Matdyk E (1977): Amyloidosis: Eular Bull (3). Basel, Eular Publ.
Rydholm U (1986): Arthroscopy of the knee in juvenile chronic arthritis. Scand J Rheumatol 15: 109-112.
Rydholm U, Boegård T, Lidgren L (1985): Total knee replacement in juvenile chronic arthritis. Scand J Rheumatol 14: 329-335.
Rydholm U, Elborgh R, Ranstam J, Schroder A, Svantesson H, Lidgren L (1986): Synovectomy of the knee in juvenile chronic arthritis. A retrospective, consecutive follow-up study. J Bone Joint Surg 68B: 223-228.
Rydholm U, Brattström H, Bylander B, Lindgren L (1987): Stapling of the knee in juvenile chronic arthritis. J Pediatr Orthop 7: 63-68.
Sairanen E (1958): On rheumatoid arthritis in children: clinicoroentgenological study. Acta Rheum Scand 2(suppl): 1-79.
Sarokhan AJ, Scott RD, Thomas WH, Sledge CB, Ewald FC, Cloos DW (1983): Total knee arthroplasty in juvenile rheumatoid arthritis. J Bone J Surg 65A: 1071-1080.
Savolainen A, Isomäki H (1993): Decrease in the number of deaths from secondary amyloidosis in patients with juvenile rheumatoid arthritis. J Rheumatol 20: 1201-1203.
Savolainen A, Lehtimäki M, Kautiainen H, Aho K, Anttila P (1998): HLA B27: A prognostic factor in juvenile chronic arthritis. Clin Rheumatol 17: 121-124.
Schai PA, Thornhill TS, Scott RD (1998): Total knee arthroplasty with the PFC system: results at a minimum of ten years and survivorship analysis. J Bone Joint Surg 80B: 850-858.
Schlenzka D, Schwesinger G (1990): The height of the patella: An anatomical study. Eur J Radiol 11: 19-21.
Schnitzer TJ, Ansell BM (1977): Amyloidosis in juvenile chronic polyarthritis. Arthritis Rheum 20(2 Suppl): 245-252.
76
Schroder HM, Aaen K, Hansen EB, Nielsen PT, Rechnagel K (1996): Cementless total knee arthroplasty in rheumatoid arthritis: a report on 51 AGC knees followed for 54 months. J Arthroplasty 11: 18-23.
Scott RD (1990): Total hip and knee arthroplasty in juvenile rheumatoid arthritis. Clin Orthop 259: 83-91.
Scott RD, Sarokhan AJ, Dalziel R (1984): Total hip and total knee arthroplasty in juvenile rheumatoid arthritis. Clin Orthop 182: 90-98.
Sheppeard H, Ward DJ (1980): Intra-articular osmic acid in rheumatoid arthritis: five years' experience. Rheumatol Rehabil 19: 25-29.
Sheppeard H, Aldin A, Ward DJ (1981): Osmic acid versus yttrium-90 in rheumatoid synovitis of the knee. Scand J Rheumatol 10: 234-236.
Sherry DD, Bohnsack J, Salmonson K, Wallace CA, Mellins E (1990): Painless juvenile rheumatoid arthritis. J Pediatr 116: 921-923.
Skyttä E, Savolainen HA, Kautiainen HJ, Lehtinen JT, Belt EA (2003a): Treatment of leg lenght discrepancy with temporary epiphyseal stapling in children with juvenile idiophatic arthritis during 1957-99. J Pediatr Orthop 23: 378-380.
Skyttä E, Savolainen HA, Kautiainen HJ, Lehtinen JT, Belt EA (2003b): Long-term results of leg length discrepancy treated with temporary epiphyseal stapling in children with JIA. Clin Exp Rheumatol 2: 669-671.
Sledge CB (1968): Biochemical events in the epiphyseal plate and their physiologic control. Clin Orthop 61: 37-47.
Sledge CB, Walker PS (1984): Total knee arthroplasty in rheumatoid arthritis. Clin Orthop 182: 127-136.
Sokka TM, Kaarela K, Mottonen TT, Hannonen PJ (1999): Conventional monotherapy compared to a “sawtooth” treatment strategy in the radiographic progression of rheumatoid arthritis over the first eight years. Clin Exp Rheumatol 17: 527-532.
Southwood TR, Woo P (1995): Juvenile chronic arthritis. Baill Clin Rheumatol 9: 331-353.
Still FG (1897): On a form of chronic joint disease in children. Med Chir Trans 80: 47.
Sury B, Vesterdal E (1968): Extra-articular lesions in juvenile rheumatoid arthritis. A survey based upon a study of 151 cases. Acta Rheumatol Scand 14: 309-316.
Suzuki M, Miyamoto S, Ikuta Y, Kageyama Y, Kushida K, Inoue T (1992): Amyloidosis associated with rheumatoid arthritis after total joint replacement. Ryumachi 32: 432-436 (Abstract in English).
Swan M (1990): The surgery of juvenile chronic arthritis. Clin Orthop 259: 70-75.
77
Takagi T, Koshino T, Okamoto R (2001): Arthroscopic synovectomy for rheumatoid arthritis using a holmium:YAG laser. J Rheumatol 28: 1518-1522.
Taubert H, Riemann D, Kehlen A, Meye A, Bartel F, John V, Brandt J, Bache M, Wurl P, Schmidt H, Weber E (2002): Expression of cathepsin B, D and L protein in juvenile idiopathic arthritis. Autoimmunity 35: 221-224.
Tiitinen S, Myllykangas R, Helin H, Kaarela K (1988): Modern trends in the diagnosis of secondary amyloidosis. Scand J Rheumatol 17 (Suppl 67): 30-31.
Tooms R.E. (1987): Arthroplasty of ankle and knee in Campbell`s operative orthopaedics 7th ed, Crenshaw A.H. (ed.), Mosby Company, pp 1152-1153.
Tria AJ Jr. (2003): Advancements in minimally invasive total knee arthroplasty. Orthopedics 26(Suppl8): 859-863.
Vahvanen V (1991): Reumakirurgisten toimeenpiteiden ajoittaminen. Duodecim 107: 557-562.
Vertullo CJ, Easley ME, Scott WN, Insall JN. (2001): Mobile bearings in primary knee arthroplasty. J Am Acad Orthop Surg 9: 355-364.
Vilkki P, Virtanen R, Mäkela AL (1991): Arthroscopic synovectomy in the treatment of patients with juvenile rheumatoid arthritis. Acta Univ Carol [Med] (Prague) 37: 84-86.
Vostrejs M, Hollister JR (1988): Muscle atrophy and leg length discrepancies in pauciarticular juvenile rheumatoid arthritis. Am J Dis Child 142: 343-345.
Weale AE, Murray DW, Newman JH, Ackroyd CE (1999): The length of the patellar tendon after unicompartmental and total knee arthroplasty. J Bone Joint Surg 81B: 790-795.
Wood PHN (1978): Nomenclature and classification of arthritis in children. In: Munthe E (ed): The care of rheumatic children. EULAR Monograph series No. 3. Eular Publishers, Basle: 47-50.
Ylijoki H (1998): Re-evaluation of the factors determining the longterm clinical outcome of juvenile rheumatoid arthritis (in Finnish with English summary). Turku: The Social Insurance Institution, Finland, Studies in Social Security and Health 31: 1-135. ISBN 951-669-447-0.
Zuege RC, Kempken TG, Blount WP (1979): Epiphyseal stapling for angular deformity at the knee. J Bone Joint Surg 61A: 320-329.
78
10. ORIGINAL PUBLICATIONS (I-IV)
Permission to reprint the original publications granted by the following publishers:
Elsevier, Health Sciences Division, Philadelphia, USA ( I )
The Journal of Rheumatology Publishing Company Limited, Toronto, Canada ( II )
Lippincott Williams & Wilkins, Baltimore, USA ( III )
Springer-Verlag GmbH, Heidelberg, Germany ( IV )