REVIEW ARTICLE Unicompartmental Knee Arthroplasty

Hong HKJOS Kong Journal of Orthopaedic Surgery 2003;7(1):68-75. Cheung WY, Chiu KY, Tang WM REVIEW ARTICLE Unicompartmental Knee Arthroplasty Cheung WY, 1 Chiu KY, 2 Tang WM 2 1 Department of Orthopaedics and Traumatology, and 2 Division of Joint Replacement Surgery, Department of Orthopaedic Surgery, Queen Mary Hospital, The University of Hong Kong, Hong Kong ABSTRACT Unicompartmental knee arthroplasty is a procedure that resurfaces only 1 compartment of the tibio-femoral joint. The procedure is indicated for patients who have unicompartmental osteoarthritis of the knee with mild deformity. It is contraindicated for inflammatory arthritis such as rheumatoid and gouty arthritis and for knees with significant deformity or a ruptured anterior cruciate ligament. Compared with high tibial osteotomy, unicompartmental knee arthroplasty has fewer perioperative complications, faster rehabilitation, better knee function, and longer survival. Compared with tricompartmental knee replacement, unicompartmental knee arthroplasty produces close to original knee kinematics, better range of motion and ambulatory function, and higher patient satisfaction. Unicompartmental knee arthroplasty is also easier to revise with better outcome when compared with tricompartmental knee replacement. Ninety percent to 98% 10-year survival has been reported for unicompartmental knee arthroplasty. In view of these advantages, there has been a resurgence of interest in the procedure in the USA, Europe, and Asia. Key Words: Complications, Degenerative changes, Knee arthroplasty, Rehabilitation!!"#$%!"#$!%&!"#$%&#$'(!")*+,-./+0123456789: ;<=><!"?.!"#$%" &'()*'+,)-$%./0123456789:;<=>?@ABCDEF!"#$%&'()*+,-./0$12*3$45'67*8$9:;<=* >?@A45'!"# $%&'()*+,-./01!234,56789:;<=!>?@ABC8D3EF!"#$%&'()*)+,-./0!123456789:;<=!>?@AB3CDEFD!"# INTRODUCTION Unicompartmental knee arthroplasty (UKA) is a procedure that only resurfaces the portion of the knee that is severely involved with degenerative changes, while allowing relatively normal articular surfaces to remain in situ. The early results with this procedure were not encouraging. Only 80% of patients had excellent or good results and the 4-year survival rate was only 80%. 1 However, with advances in surgical technique and implant design, and stricter patient selection for the procedure, 90% to 98% 10-year survival rates have been reported. 2 With such great improvements in unicompartmental arthroplasty, its role in treating osteoarthritis of the knee needs to be redefined. This paper reviews the history, evolution, advantages, indications, contraindications, and results for UKA. Correspondence: Dr WY Cheung, Department of Orthopaedic Surgery, Queen Mary Hospital, 102 Pokfulam Road, Pokfulam, Hong Kong. Tel: (852) 2855 4654; Fax: (852) 2817 4392. 68 2003 Hong Kong Orthopaedic Association & Hong Kong College of Orthopaedic Surgeons.

Unicompartmental Knee Arthroplasty HISTORY AND EVOLUTION OF UNICOMPARTMENTAL KNEE ARTHROPLASTY In 1973, Marmor reported the earliest experience using unicompartmental knee replacement. 1 The Marmor prosthesis was an unconstrained design with an all-polyethylene tibial component. Marmor initially cemented the tibial component onto cancellous bone within a cortical rim as an in-lay prosthesis. After a 10-year follow-up, he found that such small tibial components had a high incidence of subsidence, causing early failures, and recommended the widest tibial component be used to allow the prosthesis to rest on the peripheral cortical rim. 3 Also, he found that the use of 6 mm polyethylene was a significant risk factor for excessive wear, loosening, and subsequent revision, and the use of polyethylene of more than 6 mm was recommended. 3 Another similar design was the St Georg sledge prosthesis designed by Engelbrecht in 1971. 4 Similar to the Marmor prosthesis, fracture of the surrounding cortex and subsidence of the tibial component was seen, and the initial inlay design was abandoned for a wider component. All polyethylene tibial components have been shown to deform by creep or cold flow and thereby possibly break up the cement-bone interface leading to micromotion and later clinical loosening. 5 Metal backing was introduced to eliminate the effect of creeping. Both the Marmor and the St Georg sledge prosthesis were offered with metal backed tibial components in the mid-1980s. Another similar design was the Robert-Brigham unicompartmental knee. 6 Unfortunately, metal backing decreases the thickness of polyethylene available for any given composite thickness of the prosthesis, ranging from 2 mm to 4 mm. More tibial bone has to be sacrificed so as to insert a polyethylene more than 6 mm thick. Another concern was the small contact area between the components in unconstrained designs that could create high point loading of the polyethylene surface causing excessive delamination and wear. 7 To deal with this problem, a concept of meniscal bearing was introduced by Goodfellow and O Connor in 1978. 8 The Oxford Meniscal Bearing knee was then designed to offer a large contact area by a congruent femoromeniscal articulation, while the sliding of the meniscal component on a flat metal tray on the tibia reduced the shearing forces to the tibial bone-cement interface. Unfortunately, mobile bearing prostheses can be technically demanding to implant, bearings can dislocate, and more bone stock must be sacrificed from the tibial side than is necessary for all-polyethylene prostheses. The choice of prosthesis offered by the industry today is vast. So far there is no evidence to support one prosthesis being better than the others. In addition to the design of the prosthesis, it is obvious today that many other factors such as selection of patients, individual and collective learning, introduction of new techniques and instruments, and rehabilitation also have a significant influence on the outcome. ADVANTAGES OF UNICOMPARTMENTAL KNEE ARTHROPLASTY While there has been general agreement that a young and active patient with unicompartmental arthritis is best treated by high tibial osteotomy (HTO) and an elderly patient with more extensive disease should undergo total knee arthroplasty (TKA), the treatment of the middle-age to older patient with disease confined to 1 compartment is in dispute. For this group of patients, isolated unicompartmental replacement has potential advantages when compared with proximal tibial osteotomy or total knee arthroplasty. COMPARISON WITH PROXIMAL TIBIALl OSTEOTOMY When compared with osteotomy, unicompartmental replacement has fewer perioperative complications and a higher success rate. Broughton reported a higher incidence of postoperative wound complications and deep vein thromboseis among patients in the high tibial osteotomy group compared with the UKA group. 9 Also, 76% of the patients undergoing UKA had good results and only 7% required revision compared with 43% in the osteotomy group with good results and 20% requiring revision at an average of 7.8 years after operation. 9 Marked differences have been reported in the rehabilitation course after the 2 operations. Objective measurements of muscle torque showed better results for patients 6 months after unicompartmental replacement than for patients 1 year after proximal tibial osteotomy. Patients in the arthroplasty group also showed an increased duration of single limb support and maximal gait velocity. 10 Patients who need bilateral knee treatments may also do better from a rehabilitation standpoint because the operation can be performed simultaneously or staged during the same hospital admission. By contrast, patients undergoing 69

HKJOS osteotomy may require 3 to 6 months between operations and a recovery period of 1 year or more. 11 Conversion of proximal tibial osteotomy to total knee arthroplasty is also known to be a difficult task. COMPARISON WITH TRICOMPARTMENTAL REPLACEMENT Unicompartmental arthroplasty, unlike tricompartmental arthroplasty, can preserve nearly-normal knee kinematics by preserving the anterior and posterior cruciate ligaments, the patellofemoral joint, and the opposite tibio-femoral compartment. In a review of patients treated with unicompartmental replacement in 1 knee and bicompartmental or tricompartmental replacement in the other knee, the majority of patients believed the unicompartmental knee to be the more normal knee. 12 Patients with a unicompartmental replacement also had better range of motion and ambulatory function compared with those with a tricompartmental replacement. 13 Another advantage of unicompartmental replacement over tricompartmental replacement is that revision surgery is easier for the unicompartmentally-replaced knee because of maintenance of bone stock. Two early reports of patients undergoing revision of unicompartmental arthroplasty did not support this advantage, claiming that augmentation with special components or bone grafts were often necessary and the results were no better than revision of standard total knee arthroplasties. 14,15 However, these deficiencies are frequently the result of poor surgical techniques or prostheses that invade the bone stock unnecessarily. With modern techniques, using surface replacements on the femoral and tibial sides has made the procedure as conservative as it is in theory and has overcome this problem. In a study by Levine et al, the results of revision of failed unicompartmental replacements performed with modern bone-sparing components were similar to those published for primary knee arthroplasty. 16 Bohm et al also reported good results after revision surgeries for failed UKAs. 17 In their series of 35 unicompartmental knee revisions converted to total knee arthroplasties, only 12 knees had bone defects requiring autogenous bone grafting. Eighty percent of patients had good-to-excellent results 4 years after the revision. Similarly, McAuley et al showed an average knee society knee score of 81 points for 39 unicompartmental knee revisions to total knee arthroplasties 5 years after the operation. 18 Cheung WY, Chiu KY, Tang WM INDICATIONS AND CONTRAINDICATIONS Traditionally, the ideal candidate for unicompartmental knee replacement has been a patient with unicompartmental osteoarthritis with a physiological age greater than 60 years and a sedentary lifestyle. 4 In view of the reduced postoperative complications, better functional outcome, and better survival when compared with high tibial osteotomy, UKA is indicated for middle-aged patients with arthritis. Schai et al reported 28 patients aged between 37 and 60 years who underwent UKA for unicompartmental knee osteoarthritis. 19 All patients had good postoperative knee functions and only 2 knees required revision at an average follow-up of 4 years. 19 Another group of potential candidates comprises osteoarthritic octagenerians undergoing their first and last arthroplasty. Advantages include faster surgery, faster recovery, less blood loss, and a less expensive prosthesis when compared with tricompartmental arthroplasty. Scott reported survival of all except one unicompartmental knee in 42 octagenarians after 5 to 10 years of follow-up. 4 In addition to using age group for selecting appropriate candidates, the disease process should have no inflammatory component, for example rheumatoid or gouty arthritis, which may lead to early failure of the opposite compartments. Obese patients with a body mass index of more than 30 or weight more than 180 pounds are also contraindicated. Patients with only mild knee deformity should be accepted for UKA. Ideally the patient should have a flexion contracture of less than 15, flexion range greater than 90, valgus and varus deformity less than 15 and 10, respectively, and intact anterior cruciate ligaments. Whether unicompartmental knee arthroplasty is contraindicated in patients with patellofemoral joint degeneration is still controversial. 11 SURGICAL TECHNIQUES The surgical technique is critical to the success of the procedure. Details of implanting procedures vary with the different prostheses. However some principles have to be observed in order to achieve good results. 4 Insall and medial parapatellar approaches are commonly used for unicompartmental knee replacement. Some surgeons recommend a lateral parapatellar approach for lateral compartment arthroplasty. This will provide excellent exposure of the lateral compartment 70

Unicompartmental Knee Arthroplasty Figure 1 The femoral component should be placed at the central part of the femoral condyle and must not impinge on the tibial spine. but may make bicondylar replacement more difficult if this is deemed necessary and the surgeon is not familiar with this approach. The size of the femoral component used should most accurately reproduce the anteroposterior dimension of the femoral condyle. It should be placed in the centre of the mediolateral dimension of the femoral condyle, measured after removal of peripheral and intercondylar osteophytes (Figure 1). The compnent should also extend far enough anteriorly to cover the weight-bearing surface that comes in contact with the tibia in full extension. The leading edge of the femoral component must be countersunk into this junction to prevent patellar impingement during flexion of the knee (Figure 2). The tibial component should be positioned on the tibia so that, with the knee correctly aligned, this component is directly under the femoral component in the mediolateral dimension and the articulating surfaces of the 2 components are rotationally congruent. Figure 2 The femoral component should cover the whole weightbearing surface and the anterior edge should be counter-sunk to prevent patella impingement. Viewed from the front, the line of resection of the tibial plateau should be within 5 of a right angle to the longitudinal axis of the tibia (Figure 3). Viewed from the side, 3 to 5 of posterior slope is usually appropriate. The thickness of the tibial component should be governed by the soft tissue tension. Ideally, the compnent should replace the worn tibial plateau to its normal height after resection (Figure 4). Overcorrection and undercorrection leads to early failure. 20-23 After the medial compartment replacement, the medial joint space should open up 1 to 2 mm when valgus stress is applied with the knee in full extension. The same principles apply to replacement of the lateral compartment. A minimally invasive technique is becoming more popular in unicompartmental knee replacement. With better instrumentation, UKAs can be done through a 3 inch incision from the proximal border of the patella to the proximal tibia (Figure 5). Advocates of this technique cite many advantages, including fewer extensor mechanism problems, faster rehabilitation, earlier discharge from hospital, and cost saving. 24 Webb reported patients who received UKAs using a 71

HKJOS Cheung WY, Chiu KY, Tang WM Figure 3 The tibial transverse cut should be perpendicular to the long axis of the tibia and the medial longitudinal cut should preserve the cruciate ligaments. minimally invasive technique recovered 2-fold more quickly than with the standard approach. 25 Price reported a 98% 10-year survival rate for the Oxford unicompartmental prostheses implanted using the minimally invasive technique, and patients with the minimally invasive UKA recovered faster than patients with conventional arthroplasties. 26 However, one of the biggest drawbacks of this new technique involves the revision procedure. The skin around the knee is extremely sensitive to multiple approaches and surgeons may have significant trouble incorporating the incision from the minimally invasive UKA into the revision procedure. The consequences of this approach and the subsequent outcome of the revision surgery therefore remain to be determined. RESULTS FOR UNICOMPARTMENTAL KNEE ARTHROPLASTY Marmor, a long-time proponent of UKA, has published extensively on the subject. 1,3,27 In 1979, he reported short-term results for 56 patients who underwent UKA Figure 4 The tibial component should restore the height of the tibial plateau and soft tissue tension of the knee. with his prosthesis. 27 Good to excellent results were found in 80% of patients. The range of motion averaged 112 and the rate of revision was 20% at 4 years after the operation. Most revisions were associated with the use of tibia components only 6 mm thick. 27 In 1988, reporting on the same group of patients plus others with a minimum of 10 years follow-up, the author noted that satisfactory results had declined to 63% and reported a 30% revision rate. 3 Of the 21 failures in the series, 9 occurred because of the use of 6 mm polyethylene that was only 4 mm at minimum thickness. Furthermore, 6 failures were attributable to inclusion of patients who would now be considered to have disease too severe for unicompartmental arthroplasty. Degeneration of the uninvolved compartment was the cause for revision in only 2 patients, both at 9.5 years after surgery. 3 It should be noted that these results are from the early experience with UKA. Technical modifications recommended by Marmor, including careful patient selection, the use of a tibial guide to ensure accurate bone cuts, placement of the tibial component on the cortical rim to prevent early 72

Unicompartmental Knee Arthroplasty Most series published since the 1980s have shown improved results. Eighty percent to 98% of patients have good-to-excellent results with implant survival rates ranging from 85% to 98% at 10 years (Table 1). The majority of these studies defined failure as the occurrence of a revision procedure. In 1991, Scott et al reported an 85% 10-year survival rate with the Brigham Prostheses. 28 In 1992, Capra and Fehring claimed a 94% survival rate at 10 years using Marmor and Zimmer Compartmental II Prostheses. 29 Three other reports from 1993 to 1996 using the Marmor design showed 10-year survival rates of 91%, 92%, and 93%. 2,30,31 Data regarding the Oxford UKA were reported by Murray et al. 32 Their 1998 series of 144 medial compartment UKAs had 98% 10-year survival. A series of 62 patients by Berger et al documented 98% 10-year survival using the Miller-Galante prostheses. 20 More recent studies in 2002 by Argenson et al 33 and Perkins and Gunckle 34 also showed 94% and 97% 10-year survival with the Miller-Galante prostheses. Figure 5 An incision extends from the proximal border of the patella to the proximal tibia for minimally invasive unicompartmental knee arthroplasty. In = minimally invasive unicompartmental knee arthroplasty; JL = joint line; P = patella; T = tibial tuberosity. subsidence, and use of a tibial component of more than 6 mm thickness so as to decrease the rate of polyethylene wear, led to better outcomes. However, not all prostheses produce good results. The porous-coated anatomic (PCA) unicompartmental knee (Howmedica, Rutherford, USA) was proven to be a failure. Bergenudd reported 108 PCA unicompartmental knee arthroplasties with follow-up from 3 to 9 years 30% required revision within an average of 39 months after the surgery. 35 In 2002, Skyrme et al reported even worse results. 36 Of 26 PCA UKAs, 42% required revision with a mean revision time of 38.4 months after the surgery. The majority of the failures were due to femoral loosening, polyethylene wear, or a combination of both. Such failures were due to the poor quality of the polyethylene Table 1 Summary of unicompartmental knee outcome studies. Study Year Prosthesis No. of knees Average age (years) Survival rates (years) 10 15 20 Marmor 1988 Marmor 228 59 70% Scott et al 28 1991 Brigham 100 71 85% Capra and Fehring 29 1992 Marmor, Zimmer II 52 63 94% Heck et al 30 1993 Marmor, Zimmer I and II 294 68 91% Deshmukh and Scott 2 1994 Marmor 68 65 92% Weale and Newman 37 1994 St Georg Sledge 42 80 90% 88% Cartier et al 31 1996 Marmor 60 65 93% Ansari et al 21 1997 St Georg Slege 461 70 96% Tabor 23 1998 Marmor 67 61 84% 79% Murray et al 32 1998 Oxford 144 70 98% Squire et al 38 1999 Marmor 140 71 90% 84% Berger et al 20 1999 Miller-Galante 62 68 98% Argenson et al 33 2002 Miller-Galante 160 66 94% Perkins and Gunckle 34 2002 Miller-Galante 40 65 97% 73

HKJOS Cheung WY, Chiu KY, Tang WM and poor prosthesis design, leading to high contact stress between the femoral and tibial components. 36 The PCA unicompartmental prosthesis is no longer manufactured, since 1991. The second decade results are less promising, with studies often reflecting a rapid decline. Fifteen-year survivals have been reported in only 3 series, and were 88%, 37 79%, 23 and 90%. 38 The first of this series used the St Georg sledge prosthesis, whereas the other 2 used the Marmor design. The third series reports the only 20-year survival cited as 84%. 38 From these few studies, second decade survival of UKA would seem to be somewhat inferior to that reported for tricompartmental knee arthroplasty. UNICOMPARTMENTAL KNEE ARTHROPLASTY IN HONG KONG Similar to other Asian countries, there is not a long track record of UKA in Hong Kong. Few orthopaedic surgeons in Hong Kong have training in unicompartmental knee arthroplasty. Also, local patients tend to present late and the knee involvement is usually too advanced for UKA to be indicated. This may explain the low incidence of unicompartmental knee athroplasty in Hong Kong. In the near future, however, with better patient acceptance and more surgeons mastering this technically demanding procedure, it is expected that more unicompartmental knee arthroplasties will be performed in the region. CONCLUSION Unicompartmental arthroplasty was pioneered by Marmor in the 1970s for treatment of unicompartmental arthritis of the knees. The technique had the disadvantages of unacceptably high failure rates and difficulties with revision in the initial period. However, with better patient selection, implant design, and surgical techniques, 10-year survival comparable to total knee arthroplasty has been achieved. When compared with high tibial osteotomy in the treatment of middle-aged patients with arthritis, UKA has fewer perioperative complications, faster rehabilitation, better knee function, and longer survival. When compared with tricompartmental knee arthroplasty in the treatment of patients with arthritis in the older age group, UKA has better knee kinematics, better ambulatory function, higher patient satisfaction, and is easier to revise. In view of these advantages, UKA is a good option for treatment of unicompartmental knee arthritis in selected patients and it is becoming more popular in the USA, Europe, and Asia. REFERENCES 1. Marmor L. The modular knee. Clin Orthop 1973;94:242-248. 2. Deshmukh RV, Scott RD. Unicompartmental knee arthroplasty: long term results. Clin Orthop 2001;392:272-278. 3. Marmor L. Unicompartmental arthroplasty of the knee with a minimum ten-year follow-up period. Clin Orthop 1988; 228:171-177. 4. Scott RD. Unicompartmental total knee arthroplasty. In: Insall JN, Scott RD, editors. Surgery of the knee. 3rd ed. Edinburgh: Churchill Livingstone; 2001:1621-1628. 5. Ryd L, Lindstrand A, Stenstrom A, Selvik G. Cold flow reduced by metal backing. An in vivo roentgenstereophotogrammetric analysis of unicompartmental tibial components. Acta Orthop Scand 1990;61:21-25. 6. Scott RD, Santore RF. Unicondylar unicompartmental replacement for osteoarthritis of the knee. J Bone Joint Surg Am 1981;63:536-544. 7. Blunn GW, Joshi AB. Wear in retrieved condylar knee arthroplasty. A comparison of wear in different designs of 280n retrieved condylar knee prostheses. J Arthroplasty 1997;12:281-290. 8. Goodfellow J, O Connor J. The mechanics of the knee and prosthesis design. J Bone Joint Surg Br 1978;60-B:358-369. 9. Broughton NS, Newman JH, Baily RA. Unicompartmental replacement and high tibial osteotomy for osteoarthritis of the knee. A comparative study after 5-10 year follow-up. J Bone Joint Surg 1986;68B:447-452. 10. Ivarsson I, Gillquist J. Rehabilitation after high tibial osteotomy and unicompartmental arthroplasty. Clin Orhtop 1991:266:139-144. 11. Scott RD. Unicompartmental knee arthroplasty. AAOS instructional course lectures 1993;309-314. 12. Laurencin CT, Zalicof SB, Scott RD, Ewald FC. Unicompartmental versus total knee arthroplasty in the same patient: a comparative study. Clin Orthop 1991;273:151-156. 13. Rougraff BT, Heck DA, Gibson AE. A comparison of tricompartmental and unicompartmental arthroplasty for the treatment of gonarthrosis. Clin Orthop 1991;273:157-164. 14. Barrett WP, Scott RD. Revision of failed unicondylar unicompartmental knee arthroplasty. J Bone Joint Surg 1987; 69A:1328-1335. 15. Padgett DE, Stern SH, Insall JN. Revision total knee arthroplasty for failed unicompartmental replacement. J Bone Joint Surg 1991;73A:186-190. 16. Levine WN, Ozuna RM, Scott RD, Thornhill TS. Conversion of failed modern unicompartmental arthroplasty to total knee arthroplasty. J Arthroplasty 1996;11:797-801. 17. Bohm I, Landsiedl F. Revision surgery after failed unicompartmental knee arthroplasty. A study of 35 cases. J Arthroplasty 2000;15:982-989. 18. McAuley JP, Engh GA, Ammeen DJ. Revision of failed unicompartmental knee arthroplasty. Clin Orthop 2001; 74

Unicompartmental Knee Arthroplasty 392:279-282. 19. Schai PA, Sun JT, Thornhill TS, Scott RD. Unicompartmental knee arthroplasty in middle-aged patients. J Arthroplasty 1998;13:365-372. 20. Berger RA, Nedeff DD, Barden RM, et al. Unicompartmental knee arthroplasty Clinical experience at 6- to 10- year followup. Clin Orthop 1999;367:50-60. 21. Ansari S, Newman JH, Ackroyd CE. St Georg sledge for medial compartment knee replacement. 461 arthroplasties followed for 1-17 years. Acta Orthop Scand 1997;68:430-434. 22. Ridgeway SR, McAuley JP, Ammeen DJ, Engh GA. The effect of alignment of the knee on the outcome of unicompartmental knee replacement. J Bone Joint Surg 2002; 84-B:351-355. 23. Tabor OB. Unicompartmental arthroplasty: a long-term follow-up study. J Arthroplasty 1998;13:373-379. 24. Lavernia CJ, Burke WV, Sadun A. Limited exposure unicondylar arthroplasty: hype or hope? Curr Opin Orthop 2001; 12:13-17. 25. Webb JM, Topf H, Dodd CA, Goodfellow JW, Murray DW. Minimally invasive Oxford unicompartmental knee replacement. J Bone Joint Surg 1999;81-B (Suppl III):29. 26. Price A, Webb J, Topf H, Dodd C, Goodfellow J, Murray D. Oxford unicompartmental knee replacement with a minimally invasive technique. J Bone Joint Surg (Br) 2000;82-B (Suppl I):24-25. 27. Marmor L. Marmor modular knee in unicompartmental disease: minimum 4-year follow-up. J Bone Joint Surg 1979; 61A:347-353. 28. Scott RD, Cobb AG, McQueary FG, Thornhill TS. Unicompartmental knee arthroplasty: eight to 12-year follow-up evaluation with survivorship analysis. Clin Orthop 1991;271: 96-100. 29. Capra S W, Fehring TK. Unicondylar arthroplasty: A survivorship analysis. J Arthroplasty 1992;7:247-251. 30.Heck DA, Marmor L, Gibson A, Rougraff BT. Unicompartmental knee arthroplasty. A multicentre investigation with long term follow-up evaluation. Clin Orthop 1993;286: 154-159. 31. Cartier P, Sanouiller JL, Grelsamer RP. Unicompartmental knee arthroplasty surgery. 10-year minimum follow-up period. J Arthroplasty 1996;11:782-788. 32. Murray DW, Goodfellow JW, O Connor JJ. The Oxford medial unicompartmental arthroplasty: A ten-year survival study. J Bone Joint Surg 1998;80B:983-989. 33. Argenson JN, Chevrol-Benkeddache Y, Aubaniac JM. Modern unicompartmental knee arthroplasty with cement. A three to ten-year follow-up study. J Bone Joint Surg 2002; 84A:2235-2239 34. Perkins TR, Gunckle W. Unicompartmental knee arthroplasty 3- to 10-year results in a community hospital setting. J Arthroplasty 2002;17:293-297. 35. Bergenudd H. Porous-coated anatomical unicompartmental knee arthroplasty in osteoarthritis. A 3- to 9-year follow-up study. J Arthroplasty 1995;10 (Suppl):8-13. 36. Skyrme AD, Mencia MM, Skinner PW. Early failure of the porous-coated anatomic cemented unicompartmental knee arthroplasty. A 5- to 9-year follow-up study. J Arthroplasty 2002;17:201-205. 37. Weal AE, Newman JH. Unicompartmental arthroplasty and high tibial osteotomy for osteoarthritis of the knee: a comparative study with a 12 to 17-year follow-up period. Clin Orthop 1994;302:134-137. 38. Squire MW, Callaghan JJ, Goetz DD, Sullivan PM, Johnson RC. Unicompartmental knee replacement. A minimum 15 year follow-up study. Clin Orthop 1999;367:61-72. The Authors CHEUNG Wai-Yuen, MRCSE, Department of Orthopaedics and Traumatology, Queen Mary Hospital, Pokfulam, Hong Kong. CHIU Kwong-Yuen, FRCSE, FHKAM (Orth Surg), Division of Joint Replacement Surgery, Department of Orthopaedic Surgery, The University of Hong Kong, Pokfulam, Hong Kong. TANG Wai-Man, FRCSE, FHKAM (Orth Surg), Division of Joint Replacement Surgery, Department of Orthopaedic Surgery, The University of Hong Kong, Pokfulam, Hong Kong. 75