Mid-Term Results of Unicompartmental Knee Arthroplasty in Medial Compartment Osteoarthritis
Yusuf erdem1*, Cagri neyisci1, Bulent karslioglu2, Zafer atbasi3, Cemil yildiz1 and Mustafa basbozkurt1
1Orthopedics and Traumatology Department, Gulhane Training and Research Hospital, Turkey
2Orthopedics and Traumatology Department, Okmeydani Training and Research Hospital, Turkey
3Orthopedics and Traumatology Department, Guven Hospital, Turkey
Submission: January 30, 2019;Published: February 21, 2019
*Corresponding author: Yusuf Erdem, Orthopedics and Traumatology Department, Gulhane Training and Research Hospital, Turkey
How to cite this article: Yusuf erdem, Cagri neyisci, Bulent karslioglu, Zafer atbasi, et al. Mid-Term Results of Unicompartmental Knee Arthroplasty in
Medial Compartment Osteoarthritis. Ortho & Rheum Open Access J 2019; 13(4): 555869.
Background: The popularity of unicompartmental knee arthroplasty (UKA) for the treatment of isolated compartment osteoarthritis of the knee has risen over the past two decades. Currently, UKA covers a considerable amount of all knee arthroplasties worldwide. The aim of this study was to present the clinical and radiological outcomes of unicondylar knee arthroplasty in patients with medial compartment osteoarthritis.
Material and Methods: Between January 2010 and January 2014, unicondylar knee prosthesis was applied to 46 (3 men, 43 women) patients’ 56 knees with isolated knee medial gonarthrosis. The mean age of the patients was 54 (range 42 to 73), the body mass index was 26.3 (range 24 to 29) and the follow-up period was 48 months (range 20-84), respectively. patients were evaluated with Oxford Knee Score (OKS), Knee Society Score (KSS) and pain score (VAS) before and after surgery and evaluated radiologically by Oxford Radiological Evaluation Criteria.
Results: Compared to the preoperative values, the knee flexion increased from 116° to 123° (p<0.001). Statistically significant increase in OKS and KSS, and decrease in VAS was obtained postoperatively (p<0.001). All of the components were aligned within the acceptable ranges radiologically. In one patient, there have been an insert dislocation at 6th weeks postoperatively which was initially managed with closed reduction and then revised due tto redislocation. No major complications occurred including infection, deep vein thrombosis, pulmonary emboli, and neurovascular injury.
Conclusion: The mid-term clinical and radiological outcomes of UKA were excellent in this study, and our results demonstrate that UKA in proper indication is effective with considerable success in the treatment of medial compartmental knee osteoarthritis.
There is still controversy about the best treatment options for patients with medial compartment-involved knee osteoarthritis (OA). In cases of symptomatic medial compartment knee OA when conservative treatment is insufficient, correcting osteotomies such as high tibial osteotomy (HTO) and distal femoral osteotomy, total knee arthroplasties (TKA) or unicompartmental knee arthroplasties (UKA) are both utilized for surgical treatment [1-3]. Though TKA has been considered the gold standart of operative intervention for degenerative knee joint, many authors have shown better long term results for the latter choice, UKA, as well as good knee kinematics and function with lower complication rates. Beside this, preserving the healthy joint structures of the knee and replacing the degenerative compartment alone are the main reason for its
rising popularity and led the technique to further increase of the applied frequency [3-5].
Since the first design was introduced by McKeever in 1957 , UKA technique had been developed with more anatomical implants and minimal invasive approach over the years. Marmor reported %10 revision rates in two-years follow-up and %65 survivorship with first design implants in UKA at a mean follow-up of ten years [7,8], while Goodfellow’s early results with Oxford phase II was reported by Murray as more than 90 % survivorship at ten years . Increased success had been achieved by understanding the isolated anteromedial arthritis which is identified as bone cartilage loss in the anterior and mid portion of the medial compartment in association with intact ligamentous structures and normal lateral compartment cartilage [4,10] 17. In the late 1990s, more anatomical implants
were manufactured and superior outcomes were yielded with
these anatomical designs and mini-invasive surgical techniques.
In a review, Khanna reported that 10-year survivorships have
been increased ranging from %91 to %100 in most series with
%93 survival rate in fifteen years . In some similar studies,
UKA survivorship, with the use of most recent implants, has
demonstrated greater than %90 (ranging from %83 to %95)
for more than 15-year follow-up [12,13]. These scientific
publications, even with good long-term survival rates by the
designer surgeons, induced the resurgence in UKA’s popularity
over osteotomies and total knee artroplasties (TKA).
The purpose of this study was to assess the clinical and
radiologic results, survival rates and complications of 56 knees
with medial compartment OA treated consecutively with UKA
A written informed consent was obtained from each patient.
The study protocol was approved by the Institutional Review
Board. The study was conducted in accordance with the principles
of the Declaration of Helsinki. 56 consecutive knees of 46
patients who underwent UKA due to isolated medial compartment
osteoarthritis between January 2010 and January 2014 in
our clinic were included with no loss to follow up. 43 patients
(%93) were female and 3 (%7) were male. The mean age was
54 years (range, 42 to 73 years), mean follow-up was 48 months
(range, 20 to 84 months) and mean BMI was 26.3 (range, 24 to
29). Patients demographic data are summarized in Table 1. All
surgeries were performed and the data was collected by the senior
surgeon. Patients were assessed clinically and functionally
by using Oxford Knee Score (OKS) , Knee Society Score
(KSS)  preoperatively and at further follow-ups postoperatively.
Also patient satisfaction regarding pain and perception of
knee normality were assessed by subjective evaluation which is
categorized as very satisfied, satisfied, uncertain and dissatisfied.
Range of motion of the operated knee is recorded pre and
postoperatively. Prosthesis were assessed using the Oxford radiological
After medical history and physical examination, serial radiographs
including anteroposterior (AP) standing up-right and
lateral (L) knee graphies at 30° flexion and Rosenberg views (PA
10° caudal knee graphy in knee slightly flexion) are routinely taken.
The inclusion criterias for this study were:
i. Anteromedial knee pain on one finger test with medial
ii. Intact lateral compartment (no loss of cartilage on
lateral condyles weight-bearing surfaces and no meniscal
tear) and intact anterior cruciat ligament some of which
were revealed by magnetic resonance imaging (MRI) scans
iii. Varus deformity lower than 15° or correctable varus
deformity at 20° flexion
iv. Flexion contracture lower than 15°
Furthermore patellofemoral arthritis classified as lower
than outerbridge grade 4, the patients’s age, weight and activity
level were not contraindications to the operation. However
patients with a history of surgery for osteoarthritis and previous
fractures around the affected knee were excluded from the study.
We used the templates of implants on anteroposterior and
lateral x-rays to predict the implant size preoperatively. Also
the weight, height, and BMI of each patient had been recorded
preoperatively. Preoperative patient information booklet was
given to each patient including informations about the operation
and medical care they will require during perioperative period.
Patients were placed in the supine position after combined
(spinal+epidural) anesthesia in the operating room. The thigh
was held in special leg holder to allow minimum 120 degrees
knee flexion during the procedure (Figure 1). After the medial
parapatellar incision, arthrotomy was performed. Appropriate
bone cuts were done and implants were placed (Figure 2). UKA
was performed both knees in 10 patiens simultaneously, while
implanted unilaterally in 36 patients. Mobile-bearing implants
in 56 knees (Biomet, Warsaw, IN, USA) were used. Jones
bandage, Ranawat coctail and tranexamic acid were applied
All patients had the same physiotherapy with starting Continuous
Passive Motion (CPM) at the same day after operation
and mobilized with crutches the day after surgery. Follow-up clinical
and radiological interventions were planned at 6 weeks, 3
months, 6 months and 1 year postoperatively at our outpatient
clinic. Further follow-ups were then set at yearly intervals. All
data was analyzed statistically using SPSS ver.22.0. For statistical
analysis the t test of comparison of means for paired data was
used, and p<0.05 was considered significant.
Preoperative mean range of motion was increased from
116° (range, 100-126°) flexion to 123° flexion (range 115-
135°), whereas there was no limitation in knee extension preor
postoperatively. The mean KSS was increased from 54.6 ± 7.5
to 90.1 ± 6, the mean OKS increased from 24.1 ± 3.2 to 52.8 ±
4.6, and the mean VAS was decreased from 7.96 ± 1.02 to 2.29
± 0.89 (p<0.001). Regarding the radiological assessment, no
femoral or tibial component showed radiological loosening.
Postoperatve short to mid-term radiographic measurements
showed that the position of the femoral components was within
acceptable ranges with a mean of 1.8° varus (range 7° varus to 3°
valgus), whereas the position of the tibial component was within
acceptable ranges with a mean of 0.5° varus (range 2° varus
to 4°valgus) and a mean slope of 84.5° (range 79°-88°). All of
the tibial components, except an overflow of 1mm. in three and
2mm. in one, showed full congruency with medial plane.
Preoperative and postoperative scores were summarized
in Table 2, and radiological scores were summarized in Table
3. During follow-up, 1 male patient had insert dislocation at 6
weeks postoperatively. İnitially, insert was replaced with the
manuplation of closed reduction under sedative anesthesia.
He returned to his pre-dislocation level of activity, however he
had insert dislocation again after 4 weeks of reduction. 4-mm.
insert was replaced with a thicker and more anatomical 6-mm
one. Furthermore, 2 female patients had persistent anterior
and medial knee pain postoperatively without any sign of insert
dislocation or implant-related problems, however complete
pain relief was achieved with continous physiotherapy at 1 year
postoperatively. None of the paients had deep vein trombosis,
infection, implant loosening, osteolysis, implant-related fracture
or opposite compartment osteoarthritis. The overall satisfaction
in this study was noted as very satisfied in 36 patients’ 42 knees
(75 %), satisfied in 6 patients’ 9 knees (16 %), uncertain in 3
patients’ 4 knees (7 %), and dissatisfied in one patients’ knee (2
%). The overall survival rate of implants were %100 at 5 years
The survival rate after UKA depends on many factors. In
the current study, the survival rate of prosthesis was 100 %
at 4-year follow-up in patients with the mean age of 54 years,
results of which is found better than the aforomentioned studies.
In similar studies, Cepni et al.  reported 95.6 % survival rate
of UKA implants at 5.5 years with the mean age of 61 ± 7.3 years,
while Tadros et al.  reported 93 % survival rate at a mean
follow-up of 4.7 years with the mean age of older than 57.9
years, Clement et al.  reported 87.7 % survival rate at 5 years
with a mean age of 69.5 years, and Pandit et al.  97.3% at
seven years with a mean age of 66.4 years . In these studies,
younger age (<60 years) was accepted a predictor of failure, at
the end they explain the reason as younger and males expect
greater improvement in knee function than the prosthesis can
offer. However, our study group has disproved those scientific
Many complications dedicated to UKA has been reported
such as progression of lateral compartment osteoarthritis,
periprosthetic fracture (mostly medial tibial plateau), and
polyethylene insert dislocation of mobile implants [8,11-13,19-
21]. All these problems occurred due to non-anatomical implant
designs, inappropriate patient selection, overcorrection and
malposition of the implants. Depending on the increase in the
incidence of complications, more anatomical and minimally
invasive implant designs were produced by manufacturers
which contributed less soft tissue injury, more bone stock
preservation, early postoperative rehabilitation due to less
blood loss and pain, and shorter hospital stay resulting with
rapid recovery [2,10,22,23]. Bearing dislocations in mobile
design implants are considered higher and the most important
cause of this complication would be inappropriate gap balancing
between flexion and extension. Lewold et al. reported that the
reason of the bearing dislocation is mostly attributed to implant
malposition and soft tissue imbalancing . Other possible
reasons are the posterior impingement by remaining meniscus,
osteophytes and excessive release of medial collateral ligament,
respectively . In our study, all implants were mobile, however
we had only one insert dislocation as a major complication.
Probably the reason of our complication is considered the use
of early-designed non-anatomical mobile bearing implants. İn
terms of higher incidence of dislocation, Oxford group developed
a new anatomical bearing which had an extended length of
medial wall to protect further dislocations. Moreover the new
implant increased the amount of rotation that the bearing had
to undergo, and the anteromedial corner of the bearing has been
reduced to decrease anterior overhang in extension.
In the literature, there are some reports about this subject
which was managed with revision surgeries. Clement et al.
reported 3 revisions due to unexplained persistant symptomatic
knee pain in the second postoperative year. They emphazised
that unnecessarily revisions before two-year follow-up may
result higher early revision rates . In a similar study, Lewold
et al. reported 6 revisions of 1000 UKA due to unexplained
pain three of which were reported as unhappy after the second
operations. In our study, two patients had unexplained anterior
knee pain after the operation which was managed by further
physiotherapy at postoperative first year.
In our study we had also 100% survivorship with the careful
preoperative planning and patient selection at 4 years follow-up.
Supporting our results, Andrew et al. reported %91 survival for
the Oxford mobile implants by following the correct indications
in appropriate selected patients at 20 years follow-up , and
Bell et al. reported 100% survivorship at a mean of 24 months
follow-up with the accurate preoperative plan and patient
selection . Comparison of the survivorship and functional
results of mobile and fixed bearing implants is another subject
that should be discussed. Arthroplasty registries suggested
no conclusive advantage of one bearing design over another,
however this is an individual subject that should be discussed.
Inoue et al.  reported that some laxity should be provided
in medial collateral ligament (MCL) in fixed bearing implants,
whereas there must be appropriate tension in MCL in mobile
bearing implants to prevent dislocation. We considered this
feature in our study which resulted without any revision.
In a study, Aldinger et al.  reported higher wear and
failure rates of fixed bearings than mobile bearings. In a
review by Ko et al.  while the progression of arthritis and
dislocation were the main reasons of complications for the
mobile bearing implants, wear was the only main complication
requiring reoperation with fixed bearing implants. There is a
limited follow-up of comparative studies between fixed and
mobile bearing implants in terms of survivorship, however they
found no difference [29,30]. On the other hand, In a long-term
study, comparing the results of fixed and mobile bearing designs
of UKA, Parratte et al. found no difference in survivorship and
function between two bearing designs . The meniscal
mobile bearing moves to posterior with the knee fully flexed
which permits and restores isometric function of the ligaments.
Regarding this subject, we therefore used only mobile-bearing
To our knowledge, there are no reports on the outcomes of
the radiological positioning and congruency of the UKA. Optimal
positioning between components in terms of tibial slope, medial
fit, varus/valgus, and flexion/extension angle resulting neutral alignment of the knee was achieved in all cases. Few limitations
should be noted. First, the study was retrospective with the
risk of attendence of selection bias. Second, 3 of 46 patients,
included in this study, were male, so we could not compared
the gender demografic, clinical and radiological effects on
survivorship. Lustig et al.  reported that the gender effect on
outcome of UKA has no difference in terms of range of motion,
radiologic progression of arthritis and alignment. There is no
comprehensive study about gender effect on results of UKA,
there may be further investigations requiring this subject.
Third this study has a short- to midterm follow-up ,thus we
can comment on whether our intervention provides outcomes
superior to those provided by nonoperative treatment, TKA or
HTO or if the surgical procedure provides an advantage over
these procedures. More prospective randomised trial would be
the best method to determine this.
However we can comment on performing the surgeries experienced
team could provide significant contribution to achieve
the success. Because the level of experience of the surgeon is
considered a key factor to the overall survival of the implants
and satisfactory outcomes in many scientific reports. Also It is
reported that low volume UKA performing centers have caused
higher revision rates in many reports [17,20,33,34]. Bini et al.
 evaluated the surgeon volume effects on revision rates and
declared the yearly volume of less than twelve UKA had been
significant risk for failure. In a high volume study reported by
Baker et al., 23400 UKA were evaluated and concluded that high
volume centers and surgeons specialized in UKA showed superior
outcomes, and the minimum number of UKA’s per surgeon
should be more than thirteen per year . Nevertheless, we
are encouraged by our short to midterm results using UKA for
the treatment of medial compartment osteoarthritis of the knee,
with complete satisfaction with the outcome of their surgery at
Our study results suggest that UKA in patients with isolated
medial compartmental osteoarthritis can yield promising
results if it remains faithful to the proper patient selection and
surgical technique. However, we recommend large-scale and
long-term, prospective, clinical studies to confirm the efficacy of