Unicompartmental Knee Arthroplasty
Nanne P. Kort, MD, PhD
N.P.Kort,1 Dr. R.Deutman,2 Dr. J.J.A.M.van Raay,2 Prof. J.R. van Horn3
1 Orthopaedic surgeon
6130 MB Sittard
2 Orthopaedic surgeon
9700 RM Groningen
3 Orthopaedic surgeon
University Medical Center Groningen
9713 GZ Groningen
Correspondence: N.P. Kort
The function and survival rate of unicompartmental knee prostheses for patients with severe gonarthrosis have improved in recent years thanks to developments in design, instrumentarium and surgical technique. A medial unicompartmental knee prosthesis may be indicated for patients with arthrosis of the medial tibiofemoral compartment. Prerequisites are an intact anterior cruciate ligament, an intact lateral compartment, a correctable varus axis and sufficient flexion in the knee. Contraindications are inflammatory arthropathies and a recent episode of septic arthritis. Relative contraindications are old age, excess body weight, patellofemoral arthrosis and chondrocalcinosis. A unicompartmental knee prosthesis can be placed via a small parapatellar incision. Postoperative recovery is quicker than following the classic open approach, while knee function after five years is comparable. Knee functions also seem to be better following a medial unicompartmental knee prosthesis than after a valgusing tibial head osteotomy. The latter is still preferred for young active patients. Medial unicompartmental knee prostheses fail in 6-8% of patients, in which case revision to a total knee prosthesis would be the treatment of choice. In the long term, unicompartmental prostheses with a mobile bearing become loose less often than comparable prostheses with a fixed bearing.
Arthrosis of the knee is the most common chronic joint condition (1). The cause is often unknown, and prevalence increases with age. The most important clinical characteristic is pain, when starting movement as well as with certain loads. Limitations can come up in daily functioning. There tends to be a short period of stiffness after inactivity or night rest. The arthrosis can occur in the entire knee or it can be limited to the medial or lateral tibiofemoral compartment or the patellofemoral compartment. In the latter cases there can be a genu varum (O leg) or a genu valgum (X leg). Patients with gonarthrosis who despite conservative therapy sustain severe symptoms and limitations in daily functioning can be referred to an orthopaedic surgeon.
About 6% of patients with gonarthrosis meet the criteria used for placement of an unicompartmental knee prosthesis (2). This technique has been used since the 1950s as an operative treatment method for unicompartmental arthrosis of the knee. In the initial period there was only a tibial component made of acrylic or metal (3,4). In the 1970s a metal femoral component was added to the polyethylene tibial component (5). Poor short-term results are reported over this period (6-8).
In recent years there have been many developments in the design (among other things with a mobile or fixed bearing), instrumentarium and operative techniques of unicompartmental knee prosthesis. This has soundly improved the function and survival rate of the prosthesis (9-12). Although the concept of replacing only the affected part of the knee seems attractive, the role of the unicompartmental knee prosthesis is controversial (13).
This article discusses the medial unicompartmental knee prosthesis; it is the most described in the medical literature, given that a lateral unicompartmental knee prosthesis is placed relatively infrequently. To determine its current role in the gamut of treatment possibilities for unicompartmental knee replacement for medial compartment gonarthrosis, it must be compared to alternative operative treatment methods such as tibial head osteotomy and total knee replacement.
The success of a unicompartmental knee prosthesis depends strongly on, among other factors, a strict indication. Too wide an application will worsen results (14,15). A medial unicompartmental knee prosthesis is indicated for patients with a symptomatic, isolated unicompartmental deviation such as primary arthrosis (Figure 1), posttraumatic degenerative arthrosis and osteonecrosis of the medial side of the knee. The anterior cruciate ligament must be intact because absence of this ligament is related to degeneration of the other compartments and to a fixated varus malformation (16). No obvious degenerative deviations may exist in the lateral compartment. It should be possible to correct the varus axis passively in order to restore stability of the knee (16). The unicompartmental knee prosthesis only has a minimal possibility of correcting a flexion contracture, for which reason placing the prosthesis on a flexion contracture of more than 15 degrees is not recommended. The knee should be able to flex at least 95 degrees, and in some types at least 110 degrees. This flexion is important in order to attain optimal position of the guidance equipment during the operation to position the femoral component. Contraindications are inflammatory arthropathies (related to release of the prosthesis), a previously performed tibial head osteotomy with a genu valgum (placement of the prosthesis increases the genu valgum) and recent septic arthritis (2,16,17). Advanced age, a high body weight, patellofemoral arthrosis and chondrocalcinosis need not be contraindications. Nevertheless, these factors are considered as such by some specialists (18-21).
The definitive decision to place a unicompartmental or total knee prosthesis should be made preoperatively (22). Even if clinically and radiologically the patient seems to be a good candidate for unicompartmental knee prosthesis, contraindications for the procedure can be found perioperatively, such as lack of an anterior cruciate ligament or clear degenerative characteristics in the lateral compartment.
Until 1999, unicompartmental knee prostheses were placed with the classic ‘open’ approach as is done with a total knee prosthesis, in which the patella is turned over laterally. For five years now it has been possible to place the unicompartmental prosthesis via a small medial parapatellar incision (23-25). This approach does not require turning over the patella, and causes less damage to the extensor apparatus. Recovery can be quicker, morbidity more limited and hospital stay shorter (24). Early functional recovery (climbing stairs, lifting the stretched leg and flexion up to 70 degrees) goes twice as fast compared to the conventional open approach (23,25). Mid- to long-term results (five years) show no difference between the open and the minimally invasive approach in the final flexion and extension of the knee and knee scores (24). No differences can be seen radiologically either (25).
Tibial head osteotomies have been used for a long time as treatment for patients with medial compartmental gonarthrosis. The load axis is corrected with an osteotomy of the proximal tibia (valgusing), causing higher loading of the lateral compartment. An osteotomy saves the joint, no prosthetic material is introduced and as much bone as possible is spared, but obtaining the ideal valgus position of the knee postoperatively is technically difficult. Besides, chances of postoperative complications are greater than after placing a unicompartmental prosthesis (26).
Comparative retrospective research shows results of medial unicompartmental knee replacement to be better than those of osteotomy (26-28). After a unicompartmental knee arthroplasty there were good results for 76% of the knees, compared to 43% following an osteotomy (27). No late degeneration was seen in the lateral compartment after placement of a unicompartmental knee prosthesis. There is better mobility and less pain after a unicompartmental knee arthroplasty than after an osteotomy (28), as well as quick full loading, easy rehabilitation and fewer perioperative complications (29). The walking pattern is considerably better after placement of a unicompartmental knee prosthesis than after a tibial head osteotomy (30,31).
The combination of young age and a high activity level remains a relative contraindication for placement of a prosthesis. A valgusing tibial head osteotomy may be preferable, deferring placement of a total knee prosthesis (26).
Revision of a tibial head osteotomy as well as of a medial unicompartmental knee prosthesis into a total knee prosthesis can be successful (32,33). A failed unicompartmental prosthesis does cause more bone loss, hence more osseous reconstructions are needed in revisions. Revision of a valgusing tibial head osteotomy into a total knee prosthesis has greater chances of wound-healing disorders due to non-overlapping scars than placing a unicompartmental knee prosthesis.
Revision of a medial unicompartmental knee prosthesis is done in 6-8% of prostheses in ten years. The advice is not to place a new unicompartmental knee prosthesis (34). After revision interventions with placement of a new unicompartmental knee prosthesis, three times more re-revisions are seen after five years of follow-up than after a revision with placement of a total knee arthroplasty (26% versus 7%) (34).
In contrast with a total knee prosthesis, in a unicompartmental knee prosthesis both cruciate ligaments are spared, which allows preserving the normal kinetics of the knee (35,36). In the area of knee function, unicompartmental knee arthroplasty gives better results than total knee replacement (36,37). In a prospective study among 23 patients on whom during one hospital stay a unicompartmental knee prosthesis was placed on one side and a total knee prosthesis on the other, eight of the patients indicated preferring unicompartmental knee arthroplasty because the knee felt the most natural, 12 patients felt no difference and three patients preferred the total knee prosthesis (36). In a study in which 94 patients with a total of 102 knees with an indication for unicompartmental knee prosthesis were randomised for a unicompartmental or a total knee prosthesis, those patients with a unicompartmental knee prosthesis had less perioperative morbidity and a quicker functional recovery, and were released from the hospital two days earlier (38). After five years, the unicompartmental group had more patients who assessed the result as excellent (89% versus 83%), as well as a larger number of knees with more than 120 degrees of flexion (69 versus 17%). There was no difference in revision percentages (0 versus 0%).
The unicompartmental knee prosthesis with a mobile polyethylene bearing was first introduced in 1978 (figures 2 & 3) (39). The mobile bearing allows a maximal contact surface between the articulating prosthetic parts, thus minimising polyethylene wear to 0.03 mm per year. A movement pattern that is as natural as possible ensues, thanks to the free mobility of this bearing. The long-term results are good. The designers describe a cumulative survival percentage of the prosthesis of up to 99.1% after five years (9) and 97.7% after ten years (10). Loosening or revision is used as results measure for the survival percentage. The 15-year survival rate was 94% in an independent prospective but nonrandomised study (40). These good results do not appear to be always reproducible though (41,42). Varying results are described for unicompartmental knee prosthesis with a fixed bearing (figures 4 & 5). Two recent prospective studies give a 10-year survival rate of 98% (11) and 94% (12). A revision percentage of 18% to 28% is reported (18,43,44). In a prospective, comparative but nonrandomised study with two groups of patients, in which one group got a mobile and the other a fixed bearing, survival after 11 years with loosening or revision as results measure was 93% for the prosthesis with the fixed bearing and 99% for the prosthesis with the mobile bearing (45). There was a statistically significant greater loosening of the tibial plateau in the fixed version. The mobile-bearing version presented no problems with the tibial component; there was however more lateral arthrosis after ten years, but the difference with the other group was not statistically significant.
In recent years there have been many developments on the design, instrumentarium and operative techniques for unicompartmental knee prostheses. These have had positive effects on short- and long-term results (8-12). Without prospective randomised studies it is not possible to express a preference for a prosthesis with a mobile or with a fixed bearing. Compared to tibial head osteotomy, recovery is quicker and there are fewer complications. Compared to a total knee prosthesis, faster functional recovery and better knee function are seen. Keeping the cruciate ligament allows normal mobility of the knee.
The described complications and revisions of unicompartmental knee arthroplasty can be explained by a combination of factors. A potential problem could be that fewer unicompartmental knee prostheses are placed than total knee prostheses. Whoever operates must have sufficient experience in terms of establishing indications and surgical techniques (46). Unicompartmental knee arthroplasty requires strict patient selection and a meticulous operative technique to attain good long-term results. As long as these conditions are met, unicompartmental knee prostheses have a place in the operative treatment possibilities of medial compartmental gonarthrosis.
X-rays of a knee with arthrosis in the medial compartment
Assembled components of a unicompartmental knee prosthesis with a mobile bearing
Postoperative X-rays of a unicompartmental knee prosthesis with a mobile bearing
(A) anteroposterior view (B) lateral view
Postoperative X-rays of a unicompartmental knee prosthesis with a fixed bearing
(A) anteroposterior view (B) lateral view