Routine Patellar Resurfacing During Total Knee Arthroplasty Is Not Cost-Effective in Patients Without Patellar Arthritis

The Fate of Unresurfaced Patellae in Contemporary Total Knee Arthroplasty: Early to Midterm Results

BACKGROUND

Leaving the patella unresurfaced in total knee arthroplasty (TKA) has increased significantly over the past decade in the United States, likely due to modern patella-friendly implants, complications with resurfacing, and the knowledge that historical studies were scientifically confounded. This study evaluated revision-free survivorship out to 8.5 years in a cohort of contemporary primary TKAs with patella-friendly femoral components and unresurfaced patellae.

METHODS

A total of 1,053 consecutive primary TKAs with unresurfaced patellae were retrospectively reviewed. A selective patellar nonresurfacing protocol was used for all cases. Kaplan-Meier survivorship estimates were calculated based on patellar revision and the latest follow-up. An aggressive lateral patellar facetectomy was performed in 78% (823 of 1,053) of cases. The cohort was 62% women and 43% American Society of Anesthesiologists physical status classification I or II with a mean age and body mass index of 65 years (range, 35 to 94) and 35 kg/m2 (range, 18 to 65), respectively.

RESULTS

A total of 4 (0.4%, 4 of 1,053) unresurfaced patellae were revised. Three were resurfaced as part of other procedures: 2 for global instability and one for aseptic loosening at a mean of 1.6 years; and one patella was resurfaced by an outside surgeon for unexplained pain. The all-cause revision-free survivorship estimate specifically related to the patella was 98.9% (95% confidence interval, 98 to 100) out to 8.5 years. No significant difference in survivorship was related to patellae with or without a lateral patellar facetectomy (99.5 versus 98.1%, P ≥ .191); however, 3 of 4 patellar revisions occurred in TKAs without a lateral patellar facetectomy (P = .035).

CONCLUSIONS

The results of this study demonstrate excellent revision-free survivorship related to unresurfaced patellae, particularly when a lateral facetectomy was performed. These early to midterm results using modern patella-friendly femoral components are promising and comparable to resurfaced patellae in the literature.

LEVEL OF EVIDENCE

IV.

Incidence of Anterior Knee Pain and Functional Difficulty in a Prospectively Recruited Cohort Following Total Knee Arthroplasty With Selective Patellar Resurfacing

BACKGROUND

The influence of patellar resurfacing on anterior knee pain, stair climbing, and functional activity outcomes following total knee arthroplasty (TKA) are incompletely understood. This study examined the influence of patellar resurfacing on patient-reported outcome measures (PROMs) relating to anterior knee pain and function.

METHODS

The Knee Injury and Osteoarthritis Outcome Score of Joint Replacement (KOOS, JR.) patient PROMs were collected preoperatively and at 12 months follow-up for 950 TKAs performed over 5 years. Indications for patellar resurfacing included Grade IV patello-femoral (PFJ) changes or mechanical PFJ findings during patellar trialing. Patellar resurfacing was performed on 393 (41%) of the 950 TKAs performed. Multivariable binomial logistic regressions were performed, using KOOS, JR. questions assessing pain during stair climbing, standing upright, and function during rising from sitting as surrogates for anterior knee pain. Independent regression models were performed for each of the targeted KOOS, JR. questions, with adjustment for age at surgery, sex, and baseline pain and function.

RESULTS

No association was observed between 12-month postoperative anterior knee pain or function with patellar resurfacing (P = .17, .97). Patients who had moderate or greater preoperative pain on stairs had an increased likelihood of postoperative pain and functional difficulty (odds ratio 2.3, P = .013), while males were 42% less likely to report postoperative anterior knee pain (odds ratio 0.58, P = .002).

CONCLUSION

Selective patellar resurfacing based on PFJ degeneration and mechanical PFJ symptoms produces similar improvement in PROMs for resurfaced and unresurfaced knees.

Selective Patella Resurfacing in Contemporary Cruciate Retaining and Substituting Total Knee Arthroplasty: A Matched Cohort Analysis

BACKGROUND

Leaving the patella unresurfaced in total knee arthroplasty (TKA) is increasing due to modern patella-friendly implants, awareness that complications are not uncommon with resurfacing, and knowledge that historical studies were scientifically confounded. This study examined the effect of selective patellar resurfacing on patient-reported outcome measures (PROMs) using modern implants and techniques in cohorts rigorously matched for demographics and osteoarthritis severity.

METHODS

A total of 166 TKAs performed without patellar resurfacing were case-control matched to 166 TKAs with patella resurfacing. Case-control matching was based on demographics, American Society of Anesthesiology Physical Status, comorbidities, and osteoarthritis severity. No significant differences were observed between cohorts for demographics (P ≥ .347), comorbidities (P ≥ .443), or radiographic osteoarthritis severity (P ≥ .078). Radiographic alignment and prospectively collected PROMs were evaluated preoperatively and at latest clinical follow-up.

RESULTS

Preoperatively, patellar tilt was less for the unresurfaced patella group (3 versus 4°, P = .003); however, postoperative patellar tilt was not different (3 versus 3°, P = .225). At a mean of 2.1 years follow-up (range, 1 to 7), University of California Los Angeles Activity Level was significantly higher for the unresurfaced patella group (6.3 versus 5.5, P = .002), but the mean group difference did not reach a minimal clinically important difference. There were no other significant differences in PROMs or reoperation rates between cohorts (P ≥ .135).

CONCLUSION

In contemporary cruciate retaining and substituting TKA designs, not resurfacing the patella in select patients may achieve comparable PROMs and re-operation rates; and potentially greater activity level compared to patella resurfacing at early follow-up.

LEVEL OF EVIDENCE

III.

Similar rate of return to sports activity between posterior-stabilised and cruciate-retaining primary total knee arthroplasty in young and active patient

PURPOSE

Cruciate-retaining and posterior-stabilised implant designs are available for primary total knee arthroplasty. However, whether the implant design is associated with a difference in the level of activity still remains unclear. This clinical trial compared posterior-stabilised and cruciate-retaining implants in sport-related patient-reported outcome measures, range of motion, rate of return to sport, and weekly time dedicated to sport in active adults. It was also hypothesised that in young and active patients both implants lead to a similar rate of return to sport in terms of hours per week, type of sport, and joint mobility.

METHODS

All patients were evaluated preoperatively and for a minimum of 36 months follow-up. The University of California Los Angeles activity scores, High-Activity Arthroplasty Score, and Visual Analogue Scale were administered preoperatively and at the last follow-up. The range of motion was investigated at admission and the last follow-up. Data concerning the hours per week dedicated to sports and the type of sport practiced were also collected at admission and at the last follow-up. The Kaplan-Meier Curve was performed to compare implant survivorship.

RESULTS

Data from 227 procedures (cruciate-retaining: 109, posterior-stabilised: 118) were prospectively collected. At the last follow-up, no difference was reported in The University of California Los Angeles activity scores (p = 0.6), High-Activity Arthroplasty Score (p = 0.1), Visual Analogue Scale (p = 0.9), flexion (p = 0.7) and extension (p = 0.4). No difference was found in the rate of return (p = 0.1) and weekly hours dedicated to sport (p = 0.3). The Kaplan-Meier curve evidenced no statistically significant difference in implant survivorship (p = 0.6).

CONCLUSIONS

At approximately five years of follow-up, no difference was reported between cruciate-retaining and posterior-stabilised implants in active adults in sport-related patient-reported outcomes measures, range of motion, pain, weekly time dedicated to sport, rate of return to sport, and implant survivorship.

LEVEL OF EVIDENCE

Level II, prospective study.

Patellar resurfacing in total knee arthroplasty leads to better isokinetic performance

BACKGROUND

For decades there have been concerns about patellar resurfacing (PR) in total knee arthroplasty (TKA) and the individual preference of the surgeon is still the main determinant of whether or not resurfacing is applied. According to preference, surgeons can be categorized in 3 main groups of those who usually, selectively, or rarely resurface. The aim of this prospective, randomized, controlled study was to compare the isokinetic performance and clinical outcome of TKAs with PR and without PR.

METHODS

A total of 50 patients scheduled to undergo TKA for primary osteoarthritis of the knee were randomly assigned to either the PR or non-PR groups. There were no significant differences between the groups in respect of age, BMI, gender and preoperative Knee Society Score (KSS) and isokinetic performance. Patients were evaluated at postoperative 3, 6, and 12 months with KSS and at 6 months and 1 year with isokinetic measurements.

RESULTS

The PR group had a higher mean score, especially in the functional component of KSS, but the difference was not statistically significant. Knee extension peak torque was significantly higher in the PR group at 6 months (p = 0.029) and 1 year (p = 0.004) postoperatively. There were no significant differences between the groups in respect of knee flexion peak torque values following TKA.

CONCLUSIONS

The results of this study demonstrated that PR during TKA is associated with better isokinetic performance and higher knee scores. These results support routine/usually resurfacing of the patella. For surgeons who selectively resurface the patella, the advantage of better isokinetic performance may be taking into consideration in favor of resurfacing the patella where they are undecided.

LEVEL OF EVIDENCE

Level I, therapeutic study.

Higher risk of knee arthroplasty during ten-year follow-up if baseline radiographic osteoarthritis involves the patellofemoral joint: a CHECK Cohort Study

Background

Patellofemoral OA is a strong risk factor for progression to generalized whole knee OA, but it is unknown whether involvement of the patellofemoral joint in early radiographic OA (ROA) is associated with risk of undergoing future knee arthroplasty. This is clinically relevant because patellofemoral OA likely requires a different treatment approach than tibiofemoral OA, and identifying prognostic factors for future arthroplasty might assist clinicians with prioritizing and guiding early interventions that could improve long-term outcomes. Therefore, we evaluated association of baseline patellofemoral or tibiofemoral ROA with undergoing knee arthroplasty over 10 years.

Methods

Using the multicenter Cohort Hip and Cohort Knee (CHECK) study, we acquired three views of radiographs in both knees of individuals aged 45–65 years with complaints of knee symptoms in at least one knee. From baseline radiographs, we categorized each knee as having one of four patterns of ROA: no ROA, isolated patellofemoral ROA, isolated tibiofemoral ROA, or combined ROA. We evaluated the 10-year relative hazard for undergoing going arthroplasty, based on baseline ROA pattern, using Cox proportional hazard models, adjusting for age, sex body mass index, and pain severity.

Result

Our sample (n = 842) included 671 (80%) women and had mean (SD) age 56 (5) years, and BMI 26.3 (4.0) kg/m². Arthroplasties were undertaken in 44/1678 knees. In comparison to having no ROA at baseline, adjusted hazard ratios (aHR) for arthroplasty were highest for combined ROA (aHR 14.2 [95% CI 5.8, 34.6]) and isolated patellofemoral ROA (aHR 12.7 [5.6, 29.0]). Isolated tibiofemoral ROA was not significantly associated with arthroplasty (aHR 2.9 [0.6, 13.6]).

Conclusions

In a sample of middle-aged individuals with complaints in one or both knees, the 10-year relative hazard for undergoing arthroplasty, compared to no ROA, was increased when OA involved the patellofemoral joint, regardless of whether it was isolated to the patellofemoral joint or occurred in combination with tibiofemoral OA. Further research is needed to confirm this association and to clarify the causal mechanism of this relationship. However, our results provide preliminary evidence that identifying patellofemoral ROA may be a clinically useful prognostic indicator in early knee OA.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12891-022-05549-6.

Osteoarthritic Severity in Unresurfaced Patellae Does Not Adversely Affect Patient-reported Outcomes in Contemporary Primary TKA

Introduction:

The degree of osteoarthritis (OA) acceptable to leave in a native patella during unresurfaced total knee arthroplasty (TKA) remains unknown. This study's purpose was to examine the effect of patellofemoral OA severity on patient-reported outcome measures (PROMs) in primary TKAs performed without patellar resurfacing.

Methods:

One hundred ninety-three primary TKAs performed without patellar resurfacing were retrospectively reviewed. Preoperative patellofemoral OA severity was graded on severity, marginal osteophytes, joint space narrowing, and chondral damage using accepted grading systems. Patellar tilt and tibiofemoral alignment were measured radiographically. PROMs were evaluated at a minimum of 1-year follow-up.

Results:

In multivariate regression, preoperative lateral patella Kellgren-Lawrence grade ≥2 was associated with superior change in Knee Society Score pain with level walking, higher absolute change in Knee Injury and Osteoarthritis Outcome Score for Joint Replacement (P ≤ 0.029), and knees ‟always feeling normal” (odds ratio [OR] 3.12; P = 0.005). Osteoarthritis Research Society International atlas grades and Outerbridge classification scores did not significantly influence PROMs.

Discussion:

Worse preoperative OA severity in the lateral patellar facet, graded with the Kellgren-Lawrence system, predicted superior knee-specific PROMs in patients with unresurfaced patellae after contemporary TKA. This observation supports the clinical finding that patients with more severe OA have optimized patient outcomes and highlights the minimal contribution of patella OA to knee function after primary TKA.

To Retain or Resurface the Patella in Primary Total Knee Arthroplasty: A Comparative Study

Objective  To evaluate and compare clinically and functionally patients undergoing primary total knee arthroplasty (TKA) with preservation or replacement of the patella.

Methods  In a cross-sectional study, the functional results were evaluated and compared, using the Western Ontario and McMaster Universities (WOMAC) and Lequesne scores, of 158 patients undergoing primary TKA (162 knees); in 81 knees the patella was submitted to arthroplasty and in 81 the joint surface of the patella was preserved.

Results  No significant differences were identified in terms of the Lequesne score (p = 0.585), global WOMAC score (p = 0.169), nor in terms of its subdivisions regarding stiffness (p = 0.796) and functional capacity (p = 0.190). There was a significant difference only in terms of the subdivision that evaluates pain in the WOMAC score, being lower in the group undergoing patellar arthroplasty (p = 0.036).

Conclusion  In the present study, there was no difference in functional assessment in patients who underwent or not patellar replacement during primary knee arthroplasty surgery. However, individuals in whom the patella was preserved reported more pain.

Management of the Patella During Total Knee Arthroplasty

»

The optimal management of the patella during total knee arthroplasty (TKA) remains a controversial topic with no consensus.

»

Patellar management techniques during TKA include patellar retention or resurfacing with or without patellaplasty, as well as circumferential denervation and partial lateral facetectomy (PLF).

»

Special considerations such as patient age, etiology of disease, implant design, and surgeon preference should be accounted for when determining optimal management.

»

We recommend that most patellae be circumferentially denervated, regardless of whether they are resurfaced, as the potential benefits outweigh the small associated risks.

»

Evidence demonstrates improved functional outcomes with the addition of PLF to nonresurfaced patellae. There is currently a paucity of evidence of this technique with respect to resurfaced patellae.

»

Patellar resurfacing adds additional costs and health-care resources that should be considered in the decision-making process.

Patellar cartilage loss does not affect early outcomes of total knee arthroplasty performed without patella resurfacing

BACKGROUND

The aim of this study was to compare the outcomes and cost economics of TKA without patella resurfacing in patients with and without patellar cartilage loss.

METHODS

Prospective case control study of 209 consecutive patients undergoing TKA without patella resurfacing. Patella cartilage status was documented intra-operatively: 108 patients had patella cartilage loss (mean age 70 ± 9.7, mean BMI 31 ± 6.2, 72 (67%) female) and 101 control patients did not (age 68 ± 9.2, BMI 31 ± 5.6, 52 (51%) female). The primary outcome measure was Oxford Knee Score (OKS) improvement at one year. Secondary outcomes included OKS, EQ-5D, anterior knee pain (AKP), Kujala scores and reoperation at 2-4 years. The cost to prevent secondary patella resurfacing was calculated.

RESULTS

There were more women in the patella cartilage loss group (67% Vs 51%, p = 0.037), but no other preoperative characteristics differed. There was no difference in OKS improvement between those with and without patella cartilage loss at 1 year (mean difference -1.03, -3.68 to 1.62 95%CI, p = 0.446) or 2-4 years (mean difference 1.52, -1.43 to 4.45 95%CI, p = 0.310). At 2-4 years there was no difference in AKP (14/87 with vs 17/80 without, p = 0.430) nor Kujala score (mean difference 2.66, -3.82 to 9.13 95%CI, p = 0.418). Routine patella resurfacing would have cost £58,311 to prevent one secondary resurfacing.

CONCLUSION

There was no difference in OKS, anterior knee pain, reoperation or Kujala scores up to 2-4 years between patients with and without patellar cartilage loss following TKA without patella resurfacing. Resurfacing for this indication would not have been a cost effective intervention.

Patellar facet ratio affects knee pain, stair climbing and stair descent after TKA without patellar resurfacing

PURPOSE

To determine whether knee pain or functional impairment after total knee arthroplasty (TKA) without patellar resurfacing are correlated with preoperative patellar morphology or postoperative patellar orientation. The hypotheses were that patellar shape, increased tilt and lateral displacement would be associated with pain and functional impairment.

METHODS

From a consecutive series of 152 knees that received a cemented postero-stabilized TKA, the Oxford Knee Score (OKS) and the Knee injury and Osteoarthritis Outcome Score (KOOS) were collected at a minimum follow-up of 12 months. Uni- and multi-variable linear regression analyses were performed to determine associations between the collected clinical scores and patient demographics and patellar morphology, measured from pre- and post-operative frontal, lateral and skyline view radiographs.

RESULTS

The OKS was 75 ± 23, whereas the KOOS pain, stair climbing, and descent were respectively 77 ± 24, 3.9 ± 1.1 and 3.8 ± 1.2. OKS was not associated with any radiographic outcomes, whereas KOOS pain was better for knees with larger medial patellar facets. The KOOS stair climbing and descent were also better for knees with larger medial patellar facets.

CONCLUSION

The findings of this study partly confirm the hypotheses that pain and functional impairments after TKA without patellar resurfacing are associated with patellar shape. No association was revealed between postoperative patellar orientation and function nor pain. Quantitative consideration of patellar congruency could therefore prevent pain and improve function after TKA without patellar resurfacing.

LEVEL OF EVIDENCE

Retrospective study, Level III.

Modelling the cost-effectiveness of total knee arthroplasty: A systematic review

Objective

Osteoarthritis causes a significant healthcare burden and the number of total knee arthroplasty (TKA) procedures is predicted to increase significantly in the coming years. We conducted a systematic review to assess the scope and quality of all current TKA cost-effectiveness analysis (CEA) studies, identify trends, and identify areas for improvement.

Methods

An electronic database search of MEDLINE, Embase, the CEA registry and Scopus was used to identify all CEA studies where TKA was used with a comparator. Studies were included from January 1, 1997 to February 2, 2020. The Quality of Health Economic Analysis Studies (QHES) instrument was used to assess their quality. Thirty-three studies were included that offered both a QALY and cost calculation. The main findings, incremental-cost effectiveness ratios and other important study characteristics were then ascertained, and trends identified.

Results

Certain surgical interventions were suggested to be more cost-effective than TKA. This included unicompartmental knee arthroplasty for unicompartmental osteoarthritis, computer-assisted TKA compared to conventional TKA, and resurfacing the patella compared to no resurfacing. TKA was more cost-effective compared to non-operative management regardless of specific patient variables.

Conclusions

The analyses of the CEAs included in the study have to be interpreted with caution. Overall, certain surgical methods within TKA and alternative methods to TKA appear to be favoured for treating particular knee osteoarthritic conditions due to their suggested greater cost-effectiveness but this should be interpreted within local contexts. Our results should help guide future policy-making as healthcare associated costs continue to rise.

Wiberg Type III patellae and J-sign during extension compromise outcomes of total knee arthroplasty without patellar resurfacing

BACKGROUND

Understanding the risk factors associated with postoperative pain and worse outcome can guide surgeons on whether primary patellar resurfacing is warranted during total knee arthroplasty (TKA). The purpose of this study was to determine whether clinical scores and pain after TKA without patellar resurfacing are correlated with patellar shape and postoperative patellar position and kinematics.

METHODS

Radiographs as well as anterior knee pain according to the Visual Analogue Scale (pVAS) were collected pre- and postoperatively for 100 knees aged 68 ± 7.7 years that received uncemented TKA without patellar resurfacing. At a minimum follow-up of 12 months the Forgotten Joint Score (FJS), the Oxford Knee Score (OKS) as well as the flexion range of motion and the presence of J-sign during active extension were recorded. Uni- and multivariable linear regression analyses were performed to determine associations between the collected clinical scores and patient demographic and radiographic data.

RESULTS

Postoperative OKS (79 ± 14.4) was worse for Wiberg Type III patellae (β = -11.4, P = .020, compared with Type II). Anterior pVAS (2 ± 2) was greater in knees with J-sign during extension (β = 2.8, P < .001). None of the other radiographic measurements (patellar tilt, congruence angle and lateral patellar displacement) were correlated with postoperative OKS or anterior pVAS.

CONCLUSION

Incongruent patellar shape (Wiberg Type III) is associated with worse clinical scores, and abnormal kinematics (J-sign) with increased pain after TKA without patellar resurfacing. The authors therefore recommend routine resurfacing for Wiberg Type III patellae, though further studies are required to confirm whether resurfacing truly improves clinical scores and pain in this subgroup.