Patellar Resurfacing in Total Knee Arthroplasty: Does Design Matter?: A Meta-Analysis of 7075 Cases

Pavlou, George BSc, MRCS; Meyer, Carl MScOrth, FRCS; Leonidou, Andreas MRCS; As-Sultany, Mohammed BSc, MRCS; West, Robert DPhil, FIMA; Tsiridis, Eleftherios MD, MSc, PhD, FRCS

Journal of Bone & Joint Surgery - American Volume:
doi: 10.2106/JBJS.J.00594
Scientific Articles
Abstract

Background: Patellar resurfacing in total knee arthroplasty remains controversial. The aim of this study was to compare outcomes following total knee arthroplasty with patellar resurfacing with those following total knee arthroplasty without patellar resurfacing. We also sought to identify any correlation between outcomes and prosthetic design.

Methods: Eighteen Level-I randomized controlled trials with a cumulative sample size of 7075 knees (3463 in the resurfacing group and 3612 in the non-resurfacing group) satisfied the inclusion criteria. In the primary analysis, patellar resurfacing total knee arthroplasty was compared with non-resurfacing total knee arthroplasty, with use of reoperation rates, incidence of anterior knee pain, and functional scores as outcome measures. The secondary analysis focused on comparing patella-friendly and non-patella-friendly total knee arthroplasty designs with regard to the same three outcome measures.

Results: No significant differences were found between the resurfacing and non-resurfacing groups with regard to the incidence of anterior knee pain. A higher rate of reoperations was observed in the non-resurfacing group. Analysis of homogeneous data comparing patella-friendly with non-patella-friendly total knee arthroplasty designs demonstrated no differences in the incidence of reoperations.

Conclusions: No evidence was found to suggest that either patellar resurfacing or the prosthetic design affects the clinical outcome of a total knee arthroplasty. The higher incidence of reoperations in the non-resurfacing group may be attributed to the fact that secondary patellar resurfacing adds a surgical option for the treatment of anterior knee pain following total knee arthroplasty, thus artificially increasing the rate of reoperations in the non-resurfacing group.

Level of Evidence: Therapeutic Level I. See Instructions to Authors for a complete description of levels of evidence.

Author Information

1Robert Jones Agnes Hunt Orthopaedic Hospital, Gabowen, Oswestry SY10-7AG, United Kingdom

2Academic Department of Orthopaedics, Division of Surgery, Aristotle University Medical School, Unversity Campus GR 54124, Thessaloniki, Greece. E-mail address for E. Tsiridis: etsiridis@doctors.org.uk

3Royal United Hospital, Combe Park, Bath BA1 3NG, United Kingdom

4Institute of Health Sciences, Centre for Epidemiology and Biostatistics, Charles Thackrah Building, 101 Clarendon Road, Leeds University, Leeds LS2 9L, United Kingdom

Article Outline

There is controversy regarding whether patellar resurfacing should be performed routinely in total knee arthroplasty. The ongoing debate centers on the incidence of postoperative anterior knee pain and the risk of future reoperations. Some authors advocate routine resurfacing of the patella1-4 to reduce the risk of postoperative anterior knee pain and the need for revision surgery. In contrast, several authors have stated that the clinical results of arthroplasties with and without resurfacing do not differ5-8. Potential advantages of not resurfacing the patella include the lower cost of a procedure requiring fewer implants9, decreased operative time10, and a lower risk of complications associated with the patellofemoral articulation. Complications of patellar resurfacing include patellar fracture, osteonecrosis, loosening of the patellar button, polyethylene wear, instability, and resulting problems with patellar tracking and the extensor mechanism11-17. In view of this ongoing debate, the concept of selective resurfacing has evolved to identify patients who are likely to benefit from this procedure3,6,18,19. However, to our knowledge, there have been no randomized controlled trials to support this concept. Although factors such as body weight, the quality of the patellar cartilage as assessed intraoperatively, sex, age, and the presence of anterior knee pain preoperatively have been thoroughly investigated, discrepancies remain within the literature.

The meta-analysis reported by Pakos et al.20 in 2005, which included 1223 patients in ten prospective randomized controlled trials, suggested that patellar resurfacing reduces the risk of reoperations and anterior knee pain. This analysis, however, did not take into account the design of the total knee prostheses that had been used and may therefore have introduced a discrepancy between the studies examined. The design elements of the anterior flange of the femoral component, the relative conformity of the patella to the prosthetic trochlear groove, improvement in the distal extent of the trochlear groove, and the radius of curvature of the femoral component are thought to be vitally important when considering the relationship between the patellar articulation and anterior knee pain21-25. A Letter to The Editor regarding the meta-analysis by Pakos et al.26 suggested that the design of the prosthesis may well make a difference in the incidence of anterior knee pain.

The current meta-analysis will expand on the work carried out by Pakos et al.20 by including 7075 total knee arthroplasty cases from eighteen randomized controlled trials and furthering the focus to investigate the effect of total knee arthroplasty design on postoperative outcomes.

Back to Top | Article Outline

Materials and Methods

The thorough literature review involved searching MEDLINE27, Ovid28, EMBASE29, and the Cochrane Register30 databases for English-language articles on prospective Level-I randomized controlled trials comparing total knee arthroplasties with and without patellar resurfacing. The terms used in the search were total knee arthroplasty, patella, resurfacing, and anterior knee pain; the Boolean indicators were OR and AND. In an attempt to identify all relevant literature, the reference lists of electronically identified papers were manually searched to identify articles published up to December 31, 2009. Additional strategies to identify all pertinent literature included a manual search of the tables of contents of issues from 1990 to 2008 of four major orthopaedic journals (The Journal of Bone and Joint Surgery (American and British volumes), The Journal of Arthroplasty, and Clinical Orthopaedics and Related Research) and of the bibliographic sections in two major orthopaedic textbooks (Campbell's Operative Orthopaedics31 and Orthopaedic Knowledge Update 832). Whenever reports pertained to the same cohort of patients, the study with the longest follow-up period was retained.

Back to Top | Article Outline
Variables

The variables of the meta-analyses consisted of total knee arthroplasties with patellar resurfacing, those without patellar resurfacing, those with a patella-friendly design, and those with a non-patella-friendly design21-23,25. The primary analysis was a comparison of resurfacing and non-resurfacing total knee arthroplasties, and the second analysis compared patella-friendly and non-patella-friendly total knee arthroplasties. The same postoperative outcome measures were used in the two analyses.

Back to Top | Article Outline
Data Extraction and Outcome Definition

All investigators were involved in data extraction and reached a consensus on all extracted items. The postoperative outcome measures that were used included anterior knee pain, rates of reoperations to treat anterior knee pain or patellofemoral maltracking (reoperations to treat infection or aseptic loosening were not included), and knee scores if they had been recorded. Different types of knee scores were identified, including the Knee Society Score33, the Hospital for Special Surgery Score7, the Bristol Knee Score34, and the International Knee Society Score9. The documented durations of follow-up after which knee scoring was performed varied among the papers analyzed. The mean duration of follow-up ranged from twelve months to 120 months.

With regard to the second analysis concerning the impact of prosthetic design, each total knee arthroplasty in the extracted randomized controlled trials was classified as patella-friendly or non-patella-friendly according to the conformity of the patella against the femoral component. This decision was based on analysis of the technical monographs on the total knee prostheses in question and data reported by Ma et al.21, Mahoney et al.22, D'Lima et al.25, and Wang et al.23. The most important denominator for the patella-friendly design was improvement in extensor mechanism function, which is associated with features that increase the length of the extensor mechanism moment arm. This has the effect of decreasing joint reaction forces and contact stresses, which results in less patellar pain22,23,25. A relatively posterior flexion-extension axis lengthens the extensor mechanism moment arms and therefore improves extensor mechanism function22. This biomechanical principle is similar to that recognized in total hip arthroplasty, where increasing femoral offset lengthens the abductor lever arm and decreases the joint reaction force. Single-radius knee designs facilitate maintenance of ligamentous isometry and improve the function of the extension mechanism22,23,25. The single-radius knee design is based on the premise that, for the knee flexion/extension axis, there is only one axis location and subsequently a single radius of rotation. If the knee extension/flexion axis is more posterior, as in a single-radius design, the quadriceps moment arm is increased, thus reducing joint reaction forces25,35,36. Additional desirable features include a lateralized proximal trochlear design to maintain congruency in extension; a deepened patellar groove to allow for deep flexion37-41; a more distal extent of the trochlear groove42; and conformity of the anterior flange groove, which is defined as the lower angle of intersection between the patella and the groove21. Any total knee prosthesis satisfying the above criteria was defined as patella-friendly. The AMK (Anatomic Modular Knee) and Duracon implants exhibited a lateralized trochlear groove; the NexGen, Profix, Duracon, and Scorpio designs featured a deepened patellar groove; and the Profix and Scorpio prostheses also exhibited a deepened distal extent of the trochlea. Thus, all of those designs were categorized as patella-friendly. The Miller-Galante-II, PFC (press-fit condylar), and Insall-Burstein-II implants did not exhibit any important anatomic features for the patella-femoral articulation and thus were considered to be non-patella-friendly.

Back to Top | Article Outline
Analysis of Data

Statistical analysis was conducted with use of STATA/SE, version 11.0 (StataCorp, College Station, Texas), employing the “metan” command. Reoperation data for all variables (patella resurfacing versus non-resurfacing and patella-friendly versus non-patella-friendly total knee arthroplasties) were sufficiently homogeneous, as determined by the chi-square test, to allow for meta-analysis. Significance was defined by an overall α error of p < 0.05. A random effects meta-analysis with the DerSimonian-Laird method was employed to reflect variation among studies43. This method allows for the combination of studies for meta-analysis with use of the random effects model. On the other hand, analysis of data on anterior knee pain and postoperative knee scores revealed heterogeneity, and it was deemed methodologically inappropriate to combine these results into a meta-analysis. Summary measures are, however, provided. Instead a systematic review, which focused purely on differences between studies, was conducted for each of these outcomes. For each meta-analysis, an I2 value that indicates the heterogeneity of the studies is reported. This is the proportion of variation observed that is attributed to the studies rather than to the patients. The significance of this proportion (I2) is tested with use of the chi-square test for heterogeneity.

Back to Top | Article Outline
Source of Funding

No external funding was received in preparation of this manuscript.

Back to Top | Article Outline

Results

Literature Search

Eighteen prospective randomized controlled trials were identified and found to be eligible for inclusion4,7-9,18,44-56; five9,45,46,55,56 of these studies were published subsequent to the meta-analysis by Pakos et al.20. Therefore, a cumulative sample comprising 7075 total knee arthroplasties—3463 patellar resurfacing and 3612 non-resurfacing total knee arthroplasties—were included in the current analysis (Table I). The Miller-Galante-II prosthesis was analyzed in five randomized controlled trials4,8,46,50,55, and the AMK prosthesis was evaluated in three18,44,54. A variety of total knee arthroplasty designs, including the Insall-Burstein-II49, NexGen56, Profix45, PFC52,53, PCA (Porous Coated Anatomic)7, PCS (posterior cruciate-sparing) kinematic modular51, Duracon47, and Scorpio9,48 implants, were analyzed in the other randomized controlled trials. Regarding the functional score outcome, the Knee Society Score was used in eleven randomized controlled trials4,8,18,44-47,52,53,55,56, the Hospital for Special Surgery Score was used in two7,49, and the International Knee Society Score was used in one9. Therefore, to perform an adequate comparison of this outcome, we could include only randomized controlled trials in which the Knee Society Score had been used. The Knee Society Score was first described by Insall et al. in 198933, and since its introduction it has been used extensively in the literature to record patient outcome following total knee arthroplasty. While a recently identified study was initially deemed eligible for inclusion57, further analysis revealed that it was based on the same cohort of patients that had been evaluated in an earlier study, by the same group, that had already been included in the current meta-analysis8. However, the results of this study will be summarized in the discussion section.

Back to Top | Article Outline
Analysis of Resurfacing Versus Non-Resurfacing Total Knee Arthroplasty
Meta-Analysis of Homogeneous Data
Back to Top | Article Outline
Incidence of reoperations:

Sixteen randomized controlled trials, with a total of 2041 total knee arthroplasty cases, were included4,7-9,18,44-47,49-55. Of these procedures, 921 included patellar resurfacing and 1120 involved no resurfacing (Fig. 1). The combined results showed that patellar resurfacing reduced the incidence of reoperations as denoted by a relative ratio of 0.52 (95% confidence interval [CI], 0.32 to 0.87; p = 0.012).

Back to Top | Article Outline
Systematic Review of Heterogeneous Data
Back to Top | Article Outline
Incidence of anterior knee pain:

Sixteen randomized controlled trials were eligible for inclusion4,7-9,18,44-47,49-53,55,56; these studies comprised 6925 total knee arthroplasty cases, of which 3388 included patellar resurfacing and 3537 did not (Fig. 2). As a result of the heterogeneity among the included trials, as determined with the chi-square test for heterogeneity (p = 0.000; I2 = 80.2%), a meta-analysis was not conducted, although summary data are provided. One source of heterogeneity could be variations in objective measurement of pain among the studies. Seven of the trials demonstrated no difference in the incidence of postoperative anterior knee pain between resurfacing and non-resurfacing procedures8,9,44,45,50,51,55, five of them favored resurfacing4,47,52,53,56, and one did not18.

Back to Top | Article Outline
Functional knee scores:

Eleven randomized controlled trials with a total of 6416 total knee arthroplasty cases were eligible for analysis4,8,18,44-47,52,53,55,56. Of these procedures, 3208 included resurfacing of the patella and 3208 did not. The presence of heterogeneity meant that pooling of the data was not inferred (Fig. 3). However, summary measures are provided. The systematic review of all eleven trials demonstrated no difference in functional outcome scoring between resurfacing and non-resurfacing total knee arthroplasties. As only eleven of the studies were available, no further subgroup analysis was conducted on the basis of the duration of follow-up to assess the functional knee score outcome.

Back to Top | Article Outline
Analysis of Patella-Friendly Versus Non-Patella-Friendly Total Knee Arthroplasty
Meta-Analysis of Homogeneous Data
Back to Top | Article Outline
Incidence of reoperations:

Since the calculated relative ratios for the risk of a reoperation were 0.56 (95% CI, 0.21 to 1.51) and 0.46 (95% CI, 0.21 to 1.01) in the patella-friendly and non-patella-friendly groups (Fig. 1), the two meta-analyses demonstrated no difference between the groups.

Back to Top | Article Outline
Systematic Review of Heterogeneous Data
Back to Top | Article Outline
Incidence of anterior knee pain:

The data for the patella-friendly and non-patella-friendly total knee arthroplasty designs (Fig. 2) were heterogeneous and thus were not combined, but summary data are available. The heterogeneity was attributed to variation among studies as determined by the I2 result (80.2%). No differences in the incidence of anterior knee pain between groups could be discerned. In the non-patella-friendly group, two of the largest trials52,53 had results that were significantly different from those in the rest of the study sample. These trials indicated contrary conclusions with respect to anterior knee pain.

Back to Top | Article Outline
Functional knee scores:

Analysis comparing the functional knee scores in the trials in which Knee Society Scores had been used to assess the results of arthroplasties done with a patella-friendly or non-patella-friendly prosthesis revealed heterogeneity. Although the heterogeneity of data precluded any statistical comparison, some cumulative evidence from the current work suggests better functional outcomes in the non-resurfaced group (Fig. 3). Four of the five studies in which the investigators had analyzed patella-friendly knee designs suggested an improved functional outcome after arthroplasties without resurfacing18,44,45,56, while the cumulative evidence regarding non-patella-friendly designs suggested a better functional outcome with resurfacing4,8,46,52,53,55.

Back to Top | Article Outline

Discussion

The current meta-analysis expands the work by Pakos et al.20 through the inclusion of eight additional randomized controlled trials that had since become available, raising the number of cases by nearly sevenfold, from 1223 to 7075, and increasing the power of the analysis. In addition, the total knee prosthesis design—i.e., either patella-friendly or non-patella-friendly—was introduced as a variable and analyzed to identify its effect on postoperative outcome measures.

This study offers evidence suggesting that resurfacing of the patella reduces the incidence of subsequent reoperations for cumulative problems associated with the patellofemoral joint. However, no difference was demonstrated regarding the incidence of anterior knee pain between the resurfacing and non-resurfacing groups. The later result contradicts the findings of the study by Pakos et al.20, who showed a lower incidence of anterior knee pain in the resurfacing group; however, this observation was limited by the low number of cases included in their study.

The important conclusion from the current study is that no evidence was found to suggest that either patellar resurfacing or prosthetic design affects the clinical outcome of a total knee arthroplasty. The higher incidence of reoperations in the non-resurfacing group may be attributed to a clinical speculation that anterior knee pain originates from the articulation of the native patella with the metallic femoral component. It is therefore expected that a patellar replacement would add a surgical option with which to treat a patient with anterior knee pain following total knee arthroplasty, thus artificially increasing the rate of reoperations in the non-resurfacing group. On the other hand, the lack of a difference in the incidence of anterior knee pain between the resurfacing and non-resurfacing groups suggests a similar conclusion. It is therefore conceivable that the articulation between the resurfaced or non-resurfaced patella and the metal femoral component is not the single cause of postoperative anterior knee pain but instead the cause is the overall balance of the knee and the tracking of the patella. This was confirmed indirectly in the current study, as no difference was found between the patella-friendly and non-patella-friendly total knee arthroplasty designs. Since neither was considered superior, it raises the hypothesis that the available designs are not of relevance to the occurrence of anterior knee pain. Future designs may address this issue, but it remains certain that accurate knee soft-tissue balancing and subsequent patellofemoral tracking are the keys to achieving an optimal postoperative result. We speculate that, if the surgeon is not entirely satisfied with the overall soft-tissue balance, it is preferable to perform patellar resurfacing and reduce the chance of a reoperation.

It is important to consider several limitations to the current study. Meta-analysis as a statistical tool in medical research allows the combination of large numbers to increase the power of a study. It is, however, not without problems. A meta-analysis relies on the quality of the trials that it summates. We analyzed only Level-I randomized controlled trials. Publication bias due to the lower rates of acceptance of studies with non-significant findings has been widely documented58.

The classification of the prostheses as patella-friendly or non-patella-friendly was based on a number of parameters derived from the literature and deemed to be objective. This method may not be suitably sensitive or specific to identify subtle differences between various designs. Since the current study included a majority of older-generation implants, it does not reflect substantial improvements that may be seen in currently available, newer, and more anatomical implants. Furthermore, we could not account for differing surgical techniques, total knee replacement alignment, and ligament balancing, which are all surgeon-specific factors and may vary among individuals and among centers. We also did not differentiate among the plethora of available patellar button implants, which vary in design. We did not conduct subgroup analysis on the basis of these design factors, but we acknowledge that the design of the patellar prosthesis may influence outcome.

A weakness of the available literature is that randomized controlled trials conducted to compare resurfacing with non-resurfacing are likely to produce results that are specific to the implant being addressed. Therefore, assumptions cannot be made about other implants in general. Implant-specific randomized controlled trials and National Registry analyses may become necessary. Furthermore, a randomized controlled trial specifically addressing the concept of selective resurfacing in total knee arthroplasty would contribute greatly to the orthopaedic literature. That said, trials analyzing prostheses with similar design profiles may be comparable. However, we acknowledge that these would not be as accurate as the methods mentioned above.

Investigation performed at the Academic Department of Orthopaedics, Division of Surgery, Aristotle University Medical School, Thessaloniki, Greece

A commentary by Robert B. Bourne, MD, FRCSC, is linked to the online version of this article at jbjs.org.

Back to Top | Article Outline

References

1. Ranawat CS. The patellofemoral joint in total condylar knee arthroplasty. Pros and cons based on five-to ten-year follow-up observations. Clin Orthop Relat Res. 1986;205:93–9.
2. Insall J Scott WN Ranawat CS. The total condylar knee prosthesis. A report of two hundred and twenty cases. J Bone Joint Surg Am. 1979;61:173–80.
3. Picetti GD 3rd McGann WA Welch RB. The patellofemoral joint after total knee arthroplasty without patellar resurfacing. J Bone Joint Surg Am. 1990;72:1379–82.
4. Wood DJ Smith AJ Collopy D White B Brankov B Bulsara MK. Patellar resurfacing in total knee arthroplasty: a prospective randomized trial. J Bone Joint Surg Am. 2002;84:187–93.
5. Soudry M Mestriner LA Binazzi R Insall JN. Total knee arthroplasty without patellar resurfacing. Clin Orthop Relat Res. 1986;205:166–70.
6. Keblish PA Varma AK Greenwald AS. Patellar resurfacing or retention in total knee arthroplasty. A prospective study of patients with bilateral replacements. J Bone Joint Surg Br. 1994;76:930–7.
7. Feller JA Bartlett RJ Lang DM. Patellar resurfacing versus retention in total knee arthroplasty. J Bone Joint Surg Br. 1996;78:226–8.
8. Barrack RL Bertot AJ Wolfe MW Waldman DA Milicic M Myers L. Patellar resurfacing in total knee arthroplasty. A prospective, randomized, double-blind study with five to seven years of follow-up. J Bone Joint Surg Am. 2001;83:1376–81.
9. Epinette JA Manley MT. Outcomes of patellar resurfacing versus nonresurfacing in total knee arthroplasty: a 9-year experience based on a case series of Scorpio PS knees. J Knee Surg. 2008;21:293–8.
10. Ong K Lau E Manley M Kurtz SM. Patient, hospital, and procedure characteristics influencing total hip and knee arthroplasty procedure duration. J Arthroplasty. 2009;24:925–31.
11. Huang CH Liau JJ Ho FY Lin CY Young TH Cheng CK. Polyethylene failure of the patellar component in New Jersey low-contact stress total knee arthroplasties. J Arthroplasty. 2005;20:202–8.
12. Goldberg VM Figgie HE 3rd Inglis AE Figgie MP Sobel M Kelly M Kraay M. Patellar fracture type and prognosis in condylar total knee arthroplasty. Clin Orthop Relat Res. 1988;236:115–22.
13. Grace JN Rand JA. Patellar instability after total knee arthroplasty. Clin Orthop Relat Res. 1988;237:184–9.
14. Huang CH Lee YM Lai JH Liau JJ Cheng CK. Failure of the all-polyethylene patellar component after total knee arthroplasty. J Arthroplasty. 1999;14:940–4.
15. Lombardi AV Jr Engh GA Volz RG Albrigo JL Brainard BJ. Fracture/dissociation of the polyethylene in metal-backed patellar components in total knee arthroplasty. J Bone Joint Surg Am. 1988;70:675–9.
16. Lynch AF Rorabeck CH Bourne RB. Extensor mechanism complications following total knee arthroplasty. J Arthroplasty. 1987;2:135–40.
17. Mochizuki RM Schurman DJ. Patellar complications following total knee arthroplasty. J Bone Joint Surg Am. 1979;61:879–83.
18. Bourne RB Rorabeck CH Vaz M Kramer J Hardie R Robertson D. Resurfacing versus not resurfacing the patella during total knee replacement. Clin Orthop Relat Res. 1995;321:156–61.
19. Kim BS Reitman RD Schai PA Scott RD. Selective patellar nonresurfacing in total knee arthroplasty. 10 year results. Clin Orthop Relat Res. 1999;367:81–8.
20. Pakos EE Ntzani EE Trikalinos TA. Patellar resurfacing in total knee arthroplasty. A meta-analysis. J Bone Joint Surg Am. 2005;87:1438–45.
21. Ma HM Lu YC Kwok TG Ho FY Huang CY Huang CH. The effect of the design of the femoral component on the conformity of the patellofemoral joint in total knee replacement. J Bone Joint Surg Br. 2007;89:408–12.
22. Mahoney OM McClung CD dela Rosa MA Schmalzried TP. The effect of total knee arthroplasty design on extensor mechanism function. J Arthroplasty. 2002;17:416–21.
23. Wang H Simpson KJ Ferrara MS Chamnongkich S Kinsey T Mahoney OM. Biomechanical differences exhibited during sit-to-stand between total knee arthroplasty designs of varying radii. J Arthroplasty. 2006;21:1193–9.
24. Banks SA Hodge WA. Implant design affects knee arthroplasty kinematics during stair-stepping. Clin Orthop Relat Res. 2004;426:187–93.
25. D'Lima DD Poole C Chadha H Hermida JC Mahar A Colwell CW Jr. Quadriceps moment arm and quadriceps forces after total knee arthroplasty. Clin Orthop Relat Res. 2001;392:213–20.
26. Merchant AC. Patellar resurfacing in total knee arthroplasty. [Letter to The Editor] J Bone Joint Surg Am. Epub 17 Aug 2006. http://www.ejbjs.org/cgi/eletters/87/7/1438#1333 .
27. MEDLINE. http://library.nhs.uk. Accessed 2008 Oct 31.
28. Ovid. http://www.ovid.com. Accessed 2008 Oct 31.
29. EMBASE. http://library.nhs.uk. Accessed 2008 Oct 31.
30. UK Cochrane Centre. http://ukcc.cochrane.org/. Accessed 2008 Oct 31.
31. Crockarell J Jr Guyton J. Arthroplasty of the ankle and knee. In: Canale ST, editor. Campbell's operative orthopaedics. 10th ed. Philadelphia: Mosby; 2003.
32. Froimson M. Knee reconstruction and replacement. In: Vaccaro A, editor. Orthopaedic knowledge update 8. Rosemont, IL: American Academy of Orthopaedic Surgeons; 2005.
33. Insall JN Dorr LD Scott RD Scott WN. Rationale of the Knee Society clinical rating system. Clin Orthop Relat Res. 1989;248:13–4.
34. Newman JH Ackroyd CE Shah NA. Unicompartmental or total knee replacement? Five-year results of a prospective, randomised trial of 102 osteoarthritic knees with unicompartmental arthritis. J Bone Joint Surg Br. 1998;80:862–5.
35. Hollister AM Jatana S Singh AK Sullivan WW Lupichuk AG. The axes of rotation of the knee. Clin Orthop Relat Res. 1993;290:259–68.
36. Churchill DL Incavo SJ Johnson CC Beynnon BD. The transepicondylar axis approximates the optimal flexion axis of the knee. Clin Orthop Relat Res. 1998;356:111–8.
37. von Eisenhart-Rothe R Siebert M Bringmann C Vogl T Englmeier KH Graichen H. A new in vivo technique for determination of 3D kinematics and contact areas of the patello-femoral and tibio-femoral joint. J Biomech. 2004;37:927–34.
38. Tanzer M McLean CA Laxer E Casey J Ahmed AM. Effect of femoral component designs on the contact and tracking characteristics of the unresurfaced patella in total knee arthroplasty. Can J Surg. 2001;44:127–33.
39. Hsieh YF Draganich LF Ho SH Reider B. The effects of removal and reconstruction of the anterior cruciate ligament on the contact characteristics of the patellofemoral joint. Am J Sports Med. 2002;30:121–7.
40. Hinterwimmer S von Eisenhart-Rothe R Siebert M Welsch F Vogl T Graichen H. Patella kinematics and patello-femoral contact areas in patients with genu varum and mild osteoarthritis. Clin Biomech (Bristol, Avon). 2004;19:704–10.
41. Goodfellow J Hungerford DS Zindel M. Patello-femoral joint mechanics and pathology. 1. Functional anatomy of the patello-femoral joint. J Bone Joint Surg Br. 1976;58:287–90.
42. Kulkarni SK Freeman MA Poal-Manresa JC Asencio JI Rodriguez JJ. The patellofemoral joint in total knee arthroplasty: is the design of the trochlea the critical factor? J Arthroplasty. 2000;15:424–9.
43. DerSimonian R Laird N. Meta-analysis in clinical trials. Control Clin Trials. 1986;7:177–88.
44. Burnett RS Haydon CM Rorabeck CH Bourne RB. Patella resurfacing versus nonresurfacing in total knee arthroplasty: results of a randomized controlled clinical trial at a minimum of 10 years’ followup. Clin Orthop Relat Res. 2004;428:12–25.
45. Smith AJ Wood DJ Li MG. Total knee replacement with and without patellar resurfacing: a prospective, randomised trial using the Profix total knee system. J Bone Joint Surg Br. 2008;90:43–9.
46. Campbell DG Duncan WW Ashworth M Mintz A Stirling J Wakefield L Stevenson TM. Patellar resurfacing in total knee replacement: a ten-year randomised prospective trial. J Bone Joint Surg Br. 2006;88:734–9.
47. Schroeder-Boersch H Scheller G Synnatschke M Arnold P Jani L. [Patellar resurfacing. Results of a prospective randomized study]. Orthopade. 1998;27:642–50. German.
48. Kordelle J Schleicher I Kaltschmidt I Haas H Grüner MR. [Patella resurfacing in patients without substantial retropatellar knee pain symptoms?]. Z Orthop Ihre Grenzgeb. 2003;141:557–62. German.
49. Waikakul S Vanadurongwan V Bintachitt P. The effects of patellar resurfacing in total knee arthroplasty on position sense: a prospective randomized study. J Med Assoc Thai. 2000;83:975–82.
50. Barrack RL Wolfe MW Waldman DA Milicic M Bertot AJ Myers L. Resurfacing of the patella in total knee arthroplasty. A prospective, randomized, double-blind study. J Bone Joint Surg Am. 1997;79:1121–31.
51. Newman JH Ackroyd CE Shah NA Karachalios T. Should the patella be resurfaced during total knee replacement? Knee. 2000;7:17–23.
52. Partio E Wirta J. Comparison of patellar resurfacing and nonresurfacing in total knee arthroplasty: a prospective randomized study. J Orthop Rheum. 1995;8:69–74.
53. Waters TS Bentley G. Patellar resurfacing in total knee arthroplasty. A prospective, randomized study. J Bone Joint Surg Am. 2003;85:212–7.
54. Mayman D Bourne RB Rorabeck CH Vaz M Kramer J. Resurfacing versus not resurfacing the patella in total knee arthroplasty: 8-to 10-year results. J Arthroplasty. 2003;18:541–5.
55. Burnett RS Boone JL McCarthy KP Rosenzweig S Barrack RL. A prospective randomized clinical trial of patellar resurfacing and nonresurfacing in bilateral TKA. Clin Orthop Relat Res. 2007;464:65–72.
56. Tabutin J Banon F Catonne Y Grobost J Tessier JL Tillie B. Should we resurface the patella in total knee replacement? Experience with the Nex Gen prosthesis. Knee Surg Sports Traumatol Arthrosc. 2005;13:534–8.
57. Burnett RS Boone JL Rosenzweig SD Steger-May K Barrack RL. Patellar resurfacing compared with nonresurfacing in total knee arthroplasty. A concise follow-up of a randomized trial. J Bone Joint Surg Am. 2009;91:2562–7.
58. Sutton AJ Abrams KR Jones DR Sheldon TA Song F. Methods for meta-analysis in medical research. Chichester: John Wiley & Sons Ltd; 2000.

Disclosure: None of the authors received payments or services, either directly or indirectly (i.e., via his or her institution), from a third party in support of any aspect of this work. None of the authors, or their institution(s), have had any financial relationship, in the thirty-six months prior to submission of this work, with any entity in the biomedical arena that could be perceived to influence or have the potential to influence what is written in this work. Also, no author has had any other relationships, or has engaged in any other activities, that could be perceived to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitted by the authors of this work are available with the online version of this article at jbjs.org.

Copyright 2011 by The Journal of Bone and Joint Surgery, Incorporated