- Email: [email protected]
Do Changes in Patellofemoral Joint Offset Lead to Adverse Outcomes in Total Knee Arthroplasty With Patellar Resurfacing? A Radiographic Review
The Journal of Arthroplasty xxx (2016) 1e6
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Original Article
Do Changes in Patellofemoral Joint Offset Lead to Adverse Outcomes in Total Knee Arthroplasty With Patellar Resurfacing? A Radiographic Review Jacob Matz, MD a, James L. Howard, MD, MSc, FRCSC a, David J. Morden, BMSc b, Steven J. MacDonald, MD, FRCSC a, Matthew G. Teeter, PhD a, c, Brent A. Lanting, MD, MSc, FRCSC a, * a
Division of Orthopedic Surgery, London Health Sciences Center, University Hospital, London, Ontario, Canada Schulich School of Medicine and Dentistry, Western University, London, Ontario, Canada c Surgical Innovation Program, Lawson Health Research Institute, London, Ontario, Canada b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 1 May 2016 Received in revised form 4 August 2016 Accepted 22 August 2016 Available online xxx
Background: Patellofemoral joint biomechanics contribute to anterior knee pain, instability, and dysfunction following total knee arthroplasty (TKA). Information about specific factors leading to anterior knee pain and dysfunction is currently limited. Changes in patellofemoral joint offset (PFO) refers to a mismatch between the preoperative and postoperative anteroposterior geometry of the patellofemoral joint. It remains unclear whether these changes lead to adverse outcomes in TKA. Methods: A retrospective radiographic review of 970 knees pre-TKA and post-TKA was completed to correlate the radiographic and clinical outcomes of changing the PFO using a posterior-stabilized single knee design with patellar resurfacing. Results: A total of 970 patients were reviewed. Postoperatively, the anterior femoral offset, anteroposterior femoral size, and anterior patellar offset were changed in 40%, 60%, and 71% of knees, respectively, compared to preoperative values. The Western Ontario and McMasters Osteoarthritis Index total score as well as subscale scores for pain and function were not significantly affected by an increase or decrease in PFO. Similarly, Knee Society Scores and range of motion were not significantly affected. Increased anterior patellar offset was, however, associated with increased postoperative patellar tilt. Postoperative patellar tilt was not correlated with adverse patient satisfaction scores or loss of range of motion. Conclusion: Changes in PFO (decreased, maintained, or increased) are common post-TKA and are not associated with a difference in clinical outcomes. Increases in anterior patellar offset led to increased patellar tilt, which was not associated with adverse patient satisfaction scores. © 2016 Elsevier Inc. All rights reserved.
Keywords: knee arthroplasty patellofemoral overstuffing patellofemoral offset patient reported outcomes
Despite significant advances in surgical technique, component design, and perioperative management in total knee arthroplasty (TKA), complications related to the patellofemoral joint (PFJ) continue to be a substantial source of patient morbidity, causing
One or more of the authors of this paper have disclosed potential or pertinent conflicts of interest, which may include receipt of payment, either direct or indirect, institutional support, or association with an entity in the biomedical field which may be perceived to have potential conflict of interest with this work. For full disclosure statements refer to http://dx.doi.org/10.1016/j.arth.2016.08.032. * Reprint requests: Brent A. Lanting, MD, MSc, FRCSC, Division of Orthopedic Surgery, London Health Sciences Center, University Hospital, Room C9-006, 339 Windermere Road, London, Ontario N6A 5A5, Canada. http://dx.doi.org/10.1016/j.arth.2016.08.032 0883-5403/© 2016 Elsevier Inc. All rights reserved.
anterior knee pain, instability, and dysfunction [1,2]. As the volume and patient demands for TKA increase, a greater understanding of the PFJ is required [3]. Following TKA, the patellofemoral joint offset (PFO) may be decreased, maintained, or increased. Changing the PFO results in a mismatch between the anteroposterior (AP) geometry of the host bones and the AP size of the femoral and patellar components. Changing the PFO may occur by placing a femoral component or a patellar component that is smaller or larger than the space created for the implant by the bone cuts. Translation of the femoral component may also affect the PFO. Recently, computer-based modeling combined with cadaveric knee experimentation demonstrated that knee flexion decreased
2
J. Matz et al. / The Journal of Arthroplasty xxx (2016) 1e6
Table 1 Demographic Data. Age (y) Gender (female/male) Side (right/left) Body mass index (kg/m2)
76 ± 9 607/363 512/458 33 ± 8
exponentially with increasing patellar thickness [4]. It was recommended to restore preoperative patellar thickness in order to maximize postoperative knee flexion. Other in vitro studies have demonstrated that a thicker patella or femoral components larger than the anterior condyle resected may have an adverse effect on contact forces, lead to increased shear forces, and contribute to abnormal patellofemoral motion [5-7]. Conceptually, this may result in early component loosening, increased wear, and anterior knee pain. Although not demonstrated in literature, decreasing the PFO may lead to quadriceps insufficiency, weakness, and instability. While some biomechanical studies have demonstrated the importance of reproducing the AP size of the host bone, limited clinical evidence exists to support this notion [8-10]. It is important to establish whether changes in PFO in resurfaced knees are associated with adverse satisfaction and patient-reported outcomes [11]. This study will provide comprehensive clinical evidence on the relationship between changing the PFO and outcomes in TKA. Materials and Methods A retrospective review was completed of 1374 primary TKA surgeries performed between 2004 and 2014 at a tertiary care medical center. The protocol was approved by our institutional review board. The review was limited to a single, posteriorstabilized implant with patellar resurfacing using an inlay technique (Genesis II, Smith & Nephew, Memphis, TN). The surgeries were performed by 1 of 6 fellowship-trained arthroplasty surgeons. Both anterior and posterior referencing systems were used. Patients with follow-up of less than 2 years were excluded from the study. Other exclusion criteria included incomplete data postoperatively, prior open knee surgery, prior fractures, neuromuscular conditions, and English as a second language.
Following exclusions, a total of 970 patients were included. The patient demographic data are outlined in Table 1. Standard preoperative and postoperative (1-3 years) lateral and skyline knee radiographs were reviewed and measurements taken. On the lateral radiograph, measurements were taken assessing anterior femoral offset and AP femoral size [9,10] (Figs. 1 and 2). The anterior femoral offset was measured between the anterior edge of the femoral cortex and the anterior aspect of the anterior femoral condyle. The AP femoral size was defined as the distance between the posterior condylar line and the anterior condylar line. In cases where a true lateral radiograph was not available, the midpoint between the 2 condyles was taken as the average measurement. On the skyline radiograph, anterior patellar offset and patellar tilt were measured. Anterior patellar offset was defined as the distance from the deepest part of the trochlear groove to the anterior cortex of the patella. The anterior femoral offset, AP femoral size, and anterior patellar offset were used to quantify the PFO. The patellar tilt was measured by drawing a line on the anterior aspect of the femoral condyles and another line along the posterior aspect of the articular surface of the patella [12]. The angle between the 2 lines defined the patellar tilt (Fig. 3). Calibration based on known component size or a calibration marker was performed for all radiographic measurements. Radiographic measurements were taken by 2 independent observers. Inter-rater correlation coefficients for each of the radiographic measurements in this study were all good/excellent (range, 0.7-0.98). To account for measurement error, changes in PFO within 1 mm from the preoperative measurement were classified as “maintained.” Changes in PFO greater than 1 mm in a positive or negative direction were classified as “increased” or “decreased” PFO, respectively. Given that cartilage thickness could not be determined from the radiographs, for the measurements of anterior femoral offset and AP femoral size, an additional sensitivity analysis was performed taking into account þ1mm, þ2mm, þ3mm, þ4mm of cartilage thickness. This was based on previous data showing that the average cartilage thickness of the distal femur is between 2.0 ± 0.5 mm [13] and 2.1 ± 0.6 mm [14]. Because anterior patellar offset was measured between boney surfaces, cartilage thickness did not play a role in this measurement. Patients completed the Western Ontario and McMasters Osteoarthritis Index (WOMAC) and Knee Society Score (KSS)
Fig. 1. Measurement of anterior femoral offset.
J. Matz et al. / The Journal of Arthroplasty xxx (2016) 1e6
3
Fig. 2. Measurement of AP femoral size. AP, anteroposterior.
questionnaires both preoperatively and in postoperative follow-up at 1-3 years. Statistical analysis was carried out using SPSS statistics version 20 (IBM, New York, NY). Statistical comparisons between groups were made using the Kruskal-Wallis test and pairwise comparisons using the Wilcoxon signed-rank test. Spearman correlation was also used. All P values were for 2-sided tests, and P values <.05 were considered statistically significant. A post hoc power analysis was performed given our sample size of 970. This analysis showed that with a significance level of 0.05, we had 80% power to detect any significant differences (G*Power 3.0 software, Heinrich Heine University, Dusseldorf, Germany). Results Postoperative anterior femoral offset was increased in 50.8%, maintained in 13.4%, and decreased in 35.6% of patients (Fig. 4). The total AP femoral size was increased in 43.6%, maintained in 9.8%, and decreased in 45.1% of patients (Fig. 5). Finally, anterior patellar offset was increased in 11.5%, maintained in 5.9%, and decreased in 82.4% of patients (Fig. 6). The magnitude of increase in anterior femoral offset, AP femoral size, and anterior patellar offset was, on average (standard deviation [SD]), 2.8 mm (1.5), 4.7 mm (4.0), and 3.1 mm (2.1), respectively. In knees where the PFO was decreased, anterior femoral offset, AP femoral size, or anterior patellar offset were decreased, on average (SD), 3.1 mm (1.7), 5.7 mm (3.7), and 5.5 mm (2.8), respectively. Overall, on average (SD), anterior femoral offset was unchanged (mean, 0 mm; SD, 2.9), AP femoral
size was decreased 0.7 mm (5.9), and anterior patellar offset was decreased 4.2 mm (3.8). Patient outcome scores and range of motion (ROM) were evaluated on the basis of PFO measurements. Increased, maintained, or decreased anterior femoral offset was not significantly associated with changes in postoperative WOMAC scores (P ¼ .681) or subscale scores, including pain (P ¼ .904), function (P ¼ .647), and stiffness (P ¼ .098). Similarly, postoperative KSS were not affected (P ¼ .885), or subscales for pain (P ¼ .735) or function (P ¼ .519). AP femoral size was not significantly associated with changes in WOMAC patient outcomes (P ¼ .589), and subscale WOMAC outcomes such as pain (P ¼ .259), function (P ¼ .316), and stiffness (P ¼ .316). Changes in AP femoral size were also not associated with changes in KSS (P ¼ .620), and subscale scores for pain (P ¼ .300) and function (P ¼ .440). Finally, changing the anterior patellar offset was not significantly associated with changes in WOMAC (P ¼ .490), pain (P ¼ .236), function (P ¼ .703), or KSS (P ¼ .619). Postoperative ROM was not affected by changes in anterior femoral offset, AP femoral size, or anterior patellar offset (P ¼ .690, P ¼ .883, P ¼ .060, respectively). Using a sensitivity analysis to account for cartilage thickness (accounting for þ1mm, þ2mm, þ3mm, and þ4mm cartilage thickness), patient-reported outcomes and ROM outcomes were compared between groups of increased, maintained, and decreased PFO variables. The analysis did not reveal any differences between the groups for anterior femoral offset and AP femoral size (Appendix A).
Fig. 3. Measurement of anterior patellar offset.
4
J. Matz et al. / The Journal of Arthroplasty xxx (2016) 1e6 20
15
Frequency (%)
Frequency (%)
20
10 5
15 10 5
0 ≤-10
-8
-6
-4
-2
0
2
4
6
8
≥10
0 ≤-10
∆Anterior Femoral Offset (mm)
-8
-6
-4
-2
0
2
4
6
8
≥10
∆Anterior Patellar Offset (mm) Fig. 4. Changes in anterior femoral offset (postoperative anterior femoral offset preoperative anterior femoral offset).
The mean and median postoperative patellar tilt were 6.7 and respectively. A positive correlation was found between increased anterior patellar offset and increased patellar tilt postoperatively (P < .01, r ¼ 0.210). Postoperative patellar tilt was not associated with changes in anterior femoral offset (P ¼ .275) or AP femoral size (P ¼ .602). The postoperative patellar tilt was not correlated with patient satisfaction scores, such as WOMAC (P ¼ .824) or KSS (P ¼ .061), as well as subscales for WOMAC pain (P ¼ .609), function (P ¼ .406), and stiffness (P ¼ .979). ROM was not correlated with postoperative patellar tilt (P ¼ .052).
6 ,
Discussion Patellofemoral complications remain some of the most challenging problems following TKA. The rate of anterior knee pain with modern designs and technique is between 10% and 23% [11,15]. Recent registry data suggest that up to 10% of TKA revision is secondary to patellofemoral pain alone [16]. Other complications are less frequent but present with significant morbidity [2]. It remains controversial whether changes in PFO lead to adverse outcomes in TKA. In the present study, our primary outcome was the influence of changes in the PFO on clinical patient-reported outcomes, with secondary outcomes of ROM and postoperative patellar tilt. Our results showed that there was no association between change in PFO and WOMAC or KSS, or ROM postoperatively. There was an association between increased postoperative PFO and increased postoperative patellar tilt. However, increased tilt was not correlated with adverse patient satisfaction scores. Our findings are in line with previous investigations. Beldman et al [10] recently evaluated the effect of overstuffing the PFJ on clinical outcomes or anterior knee pain in TKA without patellar resurfacing. They found no association between overstuffing
Frequency (%)
20
15
10
5
0 ≤-10
-8
-6
-4
-2
0
2
4
6
8
≥10
∆AP Femoral Size (mm) Fig. 5. Changes in AP femoral size (postoperative AP femoral size preoperative AP femoral size).
Fig. 6. Changes in anterior patellar offset (postoperative anterior patellar offset preoperative anterior patellar offset).
and anterior knee pain or patient-reported outcomes. Importantly, the sample size was limited (193 knees) and the study may have lacked adequate power to detect subtle trends in outcomes. Pierson et al [9] conducted a retrospective review of the effect of overstuffing the PFJ in resurfaced knees with 2 different knee designs, finding no adverse effects associated with overstuffing. To our knowledge, this was the first clinical study to challenge the importance of overstuffing. While they were able to quantify percent change in postoperative stuffing, the amount of overstuffing was not directly quantifiable. In the present study, the presence of calibration allowed us to draw outcomes based on directly quantifiable measures. Furthermore, the presence of a single knee design eliminated potential confounders. Whereas clinical studies have not shown an association between changes in PFO and outcomes, biomechanical studies demonstrated mixed results. In a cadaveric biomechanical study, Hsu et al [6] tested knee ROM with 3 different patella thickness levels, finding that ROM was not significantly affected by patellar thickness. On the other hand, a more recent biomechanical study found that increasing the patellar thickness exponentially decreased knee flexion [4]. This decrease, however, may not be clinically significant [4]. We found frequent changes in PFO postoperatively, with a tendency toward decreased PFO in our sample. The etiology for changes in PFO is multifactorial [4,6,17]. In the majority of knees, the magnitude of change in PFO was overall relatively small. Our findings are therefore likely only relevant for small changes and should not be extrapolated to extreme changes in PFO. Increased PFO was associated with increased patellar tilt in this cohort of patients. Previous investigations indicated that patellar tilt may increase component wear [18]. In addition, increased tilt may also contribute to patellar maltracking [19]. Maltracking has been associated with various complications affecting the PFJ, such as anterior knee pain and ROM limitation [19,20]. Our findings did not show an association between adverse outcomes and increased patellar tilt. Unlike the biomechanical studies showing increased loading and wear [18], other clinical studies have also not shown an association between increased tilt and adverse patient outcomes [21]. The present study has a number of limitations. The use of radiographs as opposed to an advanced imaging modality, such as magnetic resonance imaging, prevented us from taking cartilage thickness into account for our measurements. This is particularly important for anterior femoral offset and AP femoral size. Based on previous work by DeFrate et al [13] and Cohen et al [14], the average cartilage thickness of the distal femur is between 2.0 ± 0.5 mm and 2.1 ± 0.6 mm, respectively. Therefore, we used an additional sensitivity analysis to account for cartilage thickness of þ1mm, þ2mm, þ3mm, and þ4mm. The results of
J. Matz et al. / The Journal of Arthroplasty xxx (2016) 1e6
this analysis were consistent with the remainder of our results. The angle of the X-ray beam on the skyline view can affect the anterior patellar offset measurement. While some variability certainly existed in our data, the majority of radiographs are performed with a standardized technique in our high-volume center. In addition, our institution historically used the standard skyline view of the patella as opposed to the patellofemoral axial weight-bearing views, which have previously been shown to correlate with anterior knee pain [22]. Other limitations of this study are as follows. The patientreported outcome scores are susceptible to a ceiling effect, which may affect the results of the study. All surgeons used an inlay technique for patellar resurfacing, making the results potentially not generalizable for the onlay technique. Finally, while using a sample with a single knee design improves our internal validity, it may also limit the generalizability of our results to other designs. In conclusion, this study quantified changes in PFO following TKA using a large sample size and calibrated measurements. Our findings not only highlight the difficulty in maintaining the PFO postoperatively but also show that small changes in PFO may not adversely affect clinical outcomes post-TKA. Given the potential for increased contact forces at the PFJ, further studies are required to establish whether changes in PFO would lead to adverse implant wear implications. Acknowledgments The authors would like to thank Lyndsay Somerville for her advice regarding the statistical analysis. Appendix A. Supplementary Data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.arth.2016.08.032. References 1. Russell RD, Huo MH, Jones RE. Avoiding patellar complications in total knee replacement. Bone Joint J 2014;96-B(11 Supple A):84. 2. Schiavone Panni A, Cerciello S, Del Regno C, et al. Patellar resurfacing complications in total knee arthroplasty. Int Orthop 2014;38(2):313.
5
3. Losina E, Katz JN. Total knee arthroplasty on the rise in younger patients: are we sure that past performance will guarantee future success? Arthritis Rheum 2012;64(2):339. 4. Abolghasemian M, Samiezadeh S, Sternheim A, et al. Effect of patellar thickness on knee flexion in total knee arthroplasty: a biomechanical and experimental study. J Arthroplasty 2014;29(1):80. 5. Mihalko W, Fishkin Z, Krackow K. Patellofemoral overstuff and its relationship to flexion after total knee arthroplasty. Clin Orthop Relat Res 2006;449:283. 6. Hsu HC, Luo ZP, Rand JA, et al. Influence of patellar thickness on patellar tracking and patellofemoral contact characteristics after total knee arthroplasty. J Arthroplasty 1996;11(1):69. 7. Kawahara S, Matsuda S, Fukagawa S, et al. Upsizing the femoral component increases patellofemoral contact force in total knee replacement. J Bone Joint Surg Br 2012;94(1):56. 8. Koh JS, Yeo SJ, Lee BP, et al. Influence of patellar thickness on results of total knee arthroplasty: does a residual bony patellar thickness of
6
J. Matz et al. / The Journal of Arthroplasty xxx (2016) 1e6
Appendix
Carlage Thickness +1mm +2mm WOMAC Pain
p=0.399
p=0.343
Funcon Sffness Total
p=0.690 p=0.379 p=0.552
p=0.996 p=0.209 p=0.672
Pain Funcon Total Range of Moon
p=0.075 p=0.788 p=0.858 p=0.631
p=0.29 p=0.591 p=0.701 p=0.868
KSS
Fig. A1. Sensitivity analysis showing comparisons in patient-reported outcomes between implants with increased, maintained, and decreased anterior femoral offset with presumed cartilage thickness of þ1mm and þ2mm. KSS, Knee Society Score; WOMAC, Western Ontario and McMasters Osteoarthritis Index.
+1mm WOMAC Pain
Carlage Thickness +2mm +3mm
p=0.945
p=0.311
p=0.874
Funcon Sffness Total
p=0.899 p=0.915 p=0.884
p=0.993 p=0.600 p=0.455
p=0.752 p=0.932 p=0.804
Pain Funcon Total Range of Moon
p=0.139 p=0.394 p=0.971 p=0.139
p=0.139 p=0.394 p=0.756 p=0.392
p=0.114 p=0.066 p=0.208 p=0.710
+4mm p=0.217 p=0.292 p=0.723 p=0.262
KSS p=0.406 p=0.545 p=0.906 p=0.795
Fig. A2. Sensitivity analysis showing comparisons in patient-reported outcomes between implants with increased, maintained, and decreased AP femoral size with presumed cartilage thickness of þ1mm, þ2mm, þ3mm, and þ4mm. AP, anteroposterior.
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Original Article
Do Changes in Patellofemoral Joint Offset Lead to Adverse Outcomes in Total Knee Arthroplasty With Patellar Resurfacing? A Radiographic Review Jacob Matz, MD a, James L. Howard, MD, MSc, FRCSC a, David J. Morden, BMSc b, Steven J. MacDonald, MD, FRCSC a, Matthew G. Teeter, PhD a, c, Brent A. Lanting, MD, MSc, FRCSC a, * a
Division of Orthopedic Surgery, London Health Sciences Center, University Hospital, London, Ontario, Canada Schulich School of Medicine and Dentistry, Western University, London, Ontario, Canada c Surgical Innovation Program, Lawson Health Research Institute, London, Ontario, Canada b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 1 May 2016 Received in revised form 4 August 2016 Accepted 22 August 2016 Available online xxx
Background: Patellofemoral joint biomechanics contribute to anterior knee pain, instability, and dysfunction following total knee arthroplasty (TKA). Information about specific factors leading to anterior knee pain and dysfunction is currently limited. Changes in patellofemoral joint offset (PFO) refers to a mismatch between the preoperative and postoperative anteroposterior geometry of the patellofemoral joint. It remains unclear whether these changes lead to adverse outcomes in TKA. Methods: A retrospective radiographic review of 970 knees pre-TKA and post-TKA was completed to correlate the radiographic and clinical outcomes of changing the PFO using a posterior-stabilized single knee design with patellar resurfacing. Results: A total of 970 patients were reviewed. Postoperatively, the anterior femoral offset, anteroposterior femoral size, and anterior patellar offset were changed in 40%, 60%, and 71% of knees, respectively, compared to preoperative values. The Western Ontario and McMasters Osteoarthritis Index total score as well as subscale scores for pain and function were not significantly affected by an increase or decrease in PFO. Similarly, Knee Society Scores and range of motion were not significantly affected. Increased anterior patellar offset was, however, associated with increased postoperative patellar tilt. Postoperative patellar tilt was not correlated with adverse patient satisfaction scores or loss of range of motion. Conclusion: Changes in PFO (decreased, maintained, or increased) are common post-TKA and are not associated with a difference in clinical outcomes. Increases in anterior patellar offset led to increased patellar tilt, which was not associated with adverse patient satisfaction scores. © 2016 Elsevier Inc. All rights reserved.
Keywords: knee arthroplasty patellofemoral overstuffing patellofemoral offset patient reported outcomes
Despite significant advances in surgical technique, component design, and perioperative management in total knee arthroplasty (TKA), complications related to the patellofemoral joint (PFJ) continue to be a substantial source of patient morbidity, causing
One or more of the authors of this paper have disclosed potential or pertinent conflicts of interest, which may include receipt of payment, either direct or indirect, institutional support, or association with an entity in the biomedical field which may be perceived to have potential conflict of interest with this work. For full disclosure statements refer to http://dx.doi.org/10.1016/j.arth.2016.08.032. * Reprint requests: Brent A. Lanting, MD, MSc, FRCSC, Division of Orthopedic Surgery, London Health Sciences Center, University Hospital, Room C9-006, 339 Windermere Road, London, Ontario N6A 5A5, Canada. http://dx.doi.org/10.1016/j.arth.2016.08.032 0883-5403/© 2016 Elsevier Inc. All rights reserved.
anterior knee pain, instability, and dysfunction [1,2]. As the volume and patient demands for TKA increase, a greater understanding of the PFJ is required [3]. Following TKA, the patellofemoral joint offset (PFO) may be decreased, maintained, or increased. Changing the PFO results in a mismatch between the anteroposterior (AP) geometry of the host bones and the AP size of the femoral and patellar components. Changing the PFO may occur by placing a femoral component or a patellar component that is smaller or larger than the space created for the implant by the bone cuts. Translation of the femoral component may also affect the PFO. Recently, computer-based modeling combined with cadaveric knee experimentation demonstrated that knee flexion decreased
2
J. Matz et al. / The Journal of Arthroplasty xxx (2016) 1e6
Table 1 Demographic Data. Age (y) Gender (female/male) Side (right/left) Body mass index (kg/m2)
76 ± 9 607/363 512/458 33 ± 8
exponentially with increasing patellar thickness [4]. It was recommended to restore preoperative patellar thickness in order to maximize postoperative knee flexion. Other in vitro studies have demonstrated that a thicker patella or femoral components larger than the anterior condyle resected may have an adverse effect on contact forces, lead to increased shear forces, and contribute to abnormal patellofemoral motion [5-7]. Conceptually, this may result in early component loosening, increased wear, and anterior knee pain. Although not demonstrated in literature, decreasing the PFO may lead to quadriceps insufficiency, weakness, and instability. While some biomechanical studies have demonstrated the importance of reproducing the AP size of the host bone, limited clinical evidence exists to support this notion [8-10]. It is important to establish whether changes in PFO in resurfaced knees are associated with adverse satisfaction and patient-reported outcomes [11]. This study will provide comprehensive clinical evidence on the relationship between changing the PFO and outcomes in TKA. Materials and Methods A retrospective review was completed of 1374 primary TKA surgeries performed between 2004 and 2014 at a tertiary care medical center. The protocol was approved by our institutional review board. The review was limited to a single, posteriorstabilized implant with patellar resurfacing using an inlay technique (Genesis II, Smith & Nephew, Memphis, TN). The surgeries were performed by 1 of 6 fellowship-trained arthroplasty surgeons. Both anterior and posterior referencing systems were used. Patients with follow-up of less than 2 years were excluded from the study. Other exclusion criteria included incomplete data postoperatively, prior open knee surgery, prior fractures, neuromuscular conditions, and English as a second language.
Following exclusions, a total of 970 patients were included. The patient demographic data are outlined in Table 1. Standard preoperative and postoperative (1-3 years) lateral and skyline knee radiographs were reviewed and measurements taken. On the lateral radiograph, measurements were taken assessing anterior femoral offset and AP femoral size [9,10] (Figs. 1 and 2). The anterior femoral offset was measured between the anterior edge of the femoral cortex and the anterior aspect of the anterior femoral condyle. The AP femoral size was defined as the distance between the posterior condylar line and the anterior condylar line. In cases where a true lateral radiograph was not available, the midpoint between the 2 condyles was taken as the average measurement. On the skyline radiograph, anterior patellar offset and patellar tilt were measured. Anterior patellar offset was defined as the distance from the deepest part of the trochlear groove to the anterior cortex of the patella. The anterior femoral offset, AP femoral size, and anterior patellar offset were used to quantify the PFO. The patellar tilt was measured by drawing a line on the anterior aspect of the femoral condyles and another line along the posterior aspect of the articular surface of the patella [12]. The angle between the 2 lines defined the patellar tilt (Fig. 3). Calibration based on known component size or a calibration marker was performed for all radiographic measurements. Radiographic measurements were taken by 2 independent observers. Inter-rater correlation coefficients for each of the radiographic measurements in this study were all good/excellent (range, 0.7-0.98). To account for measurement error, changes in PFO within 1 mm from the preoperative measurement were classified as “maintained.” Changes in PFO greater than 1 mm in a positive or negative direction were classified as “increased” or “decreased” PFO, respectively. Given that cartilage thickness could not be determined from the radiographs, for the measurements of anterior femoral offset and AP femoral size, an additional sensitivity analysis was performed taking into account þ1mm, þ2mm, þ3mm, þ4mm of cartilage thickness. This was based on previous data showing that the average cartilage thickness of the distal femur is between 2.0 ± 0.5 mm [13] and 2.1 ± 0.6 mm [14]. Because anterior patellar offset was measured between boney surfaces, cartilage thickness did not play a role in this measurement. Patients completed the Western Ontario and McMasters Osteoarthritis Index (WOMAC) and Knee Society Score (KSS)
Fig. 1. Measurement of anterior femoral offset.
J. Matz et al. / The Journal of Arthroplasty xxx (2016) 1e6
3
Fig. 2. Measurement of AP femoral size. AP, anteroposterior.
questionnaires both preoperatively and in postoperative follow-up at 1-3 years. Statistical analysis was carried out using SPSS statistics version 20 (IBM, New York, NY). Statistical comparisons between groups were made using the Kruskal-Wallis test and pairwise comparisons using the Wilcoxon signed-rank test. Spearman correlation was also used. All P values were for 2-sided tests, and P values <.05 were considered statistically significant. A post hoc power analysis was performed given our sample size of 970. This analysis showed that with a significance level of 0.05, we had 80% power to detect any significant differences (G*Power 3.0 software, Heinrich Heine University, Dusseldorf, Germany). Results Postoperative anterior femoral offset was increased in 50.8%, maintained in 13.4%, and decreased in 35.6% of patients (Fig. 4). The total AP femoral size was increased in 43.6%, maintained in 9.8%, and decreased in 45.1% of patients (Fig. 5). Finally, anterior patellar offset was increased in 11.5%, maintained in 5.9%, and decreased in 82.4% of patients (Fig. 6). The magnitude of increase in anterior femoral offset, AP femoral size, and anterior patellar offset was, on average (standard deviation [SD]), 2.8 mm (1.5), 4.7 mm (4.0), and 3.1 mm (2.1), respectively. In knees where the PFO was decreased, anterior femoral offset, AP femoral size, or anterior patellar offset were decreased, on average (SD), 3.1 mm (1.7), 5.7 mm (3.7), and 5.5 mm (2.8), respectively. Overall, on average (SD), anterior femoral offset was unchanged (mean, 0 mm; SD, 2.9), AP femoral
size was decreased 0.7 mm (5.9), and anterior patellar offset was decreased 4.2 mm (3.8). Patient outcome scores and range of motion (ROM) were evaluated on the basis of PFO measurements. Increased, maintained, or decreased anterior femoral offset was not significantly associated with changes in postoperative WOMAC scores (P ¼ .681) or subscale scores, including pain (P ¼ .904), function (P ¼ .647), and stiffness (P ¼ .098). Similarly, postoperative KSS were not affected (P ¼ .885), or subscales for pain (P ¼ .735) or function (P ¼ .519). AP femoral size was not significantly associated with changes in WOMAC patient outcomes (P ¼ .589), and subscale WOMAC outcomes such as pain (P ¼ .259), function (P ¼ .316), and stiffness (P ¼ .316). Changes in AP femoral size were also not associated with changes in KSS (P ¼ .620), and subscale scores for pain (P ¼ .300) and function (P ¼ .440). Finally, changing the anterior patellar offset was not significantly associated with changes in WOMAC (P ¼ .490), pain (P ¼ .236), function (P ¼ .703), or KSS (P ¼ .619). Postoperative ROM was not affected by changes in anterior femoral offset, AP femoral size, or anterior patellar offset (P ¼ .690, P ¼ .883, P ¼ .060, respectively). Using a sensitivity analysis to account for cartilage thickness (accounting for þ1mm, þ2mm, þ3mm, and þ4mm cartilage thickness), patient-reported outcomes and ROM outcomes were compared between groups of increased, maintained, and decreased PFO variables. The analysis did not reveal any differences between the groups for anterior femoral offset and AP femoral size (Appendix A).
Fig. 3. Measurement of anterior patellar offset.
4
J. Matz et al. / The Journal of Arthroplasty xxx (2016) 1e6 20
15
Frequency (%)
Frequency (%)
20
10 5
15 10 5
0 ≤-10
-8
-6
-4
-2
0
2
4
6
8
≥10
0 ≤-10
∆Anterior Femoral Offset (mm)
-8
-6
-4
-2
0
2
4
6
8
≥10
∆Anterior Patellar Offset (mm) Fig. 4. Changes in anterior femoral offset (postoperative anterior femoral offset preoperative anterior femoral offset).
The mean and median postoperative patellar tilt were 6.7 and respectively. A positive correlation was found between increased anterior patellar offset and increased patellar tilt postoperatively (P < .01, r ¼ 0.210). Postoperative patellar tilt was not associated with changes in anterior femoral offset (P ¼ .275) or AP femoral size (P ¼ .602). The postoperative patellar tilt was not correlated with patient satisfaction scores, such as WOMAC (P ¼ .824) or KSS (P ¼ .061), as well as subscales for WOMAC pain (P ¼ .609), function (P ¼ .406), and stiffness (P ¼ .979). ROM was not correlated with postoperative patellar tilt (P ¼ .052).
6 ,
Discussion Patellofemoral complications remain some of the most challenging problems following TKA. The rate of anterior knee pain with modern designs and technique is between 10% and 23% [11,15]. Recent registry data suggest that up to 10% of TKA revision is secondary to patellofemoral pain alone [16]. Other complications are less frequent but present with significant morbidity [2]. It remains controversial whether changes in PFO lead to adverse outcomes in TKA. In the present study, our primary outcome was the influence of changes in the PFO on clinical patient-reported outcomes, with secondary outcomes of ROM and postoperative patellar tilt. Our results showed that there was no association between change in PFO and WOMAC or KSS, or ROM postoperatively. There was an association between increased postoperative PFO and increased postoperative patellar tilt. However, increased tilt was not correlated with adverse patient satisfaction scores. Our findings are in line with previous investigations. Beldman et al [10] recently evaluated the effect of overstuffing the PFJ on clinical outcomes or anterior knee pain in TKA without patellar resurfacing. They found no association between overstuffing
Frequency (%)
20
15
10
5
0 ≤-10
-8
-6
-4
-2
0
2
4
6
8
≥10
∆AP Femoral Size (mm) Fig. 5. Changes in AP femoral size (postoperative AP femoral size preoperative AP femoral size).
Fig. 6. Changes in anterior patellar offset (postoperative anterior patellar offset preoperative anterior patellar offset).
and anterior knee pain or patient-reported outcomes. Importantly, the sample size was limited (193 knees) and the study may have lacked adequate power to detect subtle trends in outcomes. Pierson et al [9] conducted a retrospective review of the effect of overstuffing the PFJ in resurfaced knees with 2 different knee designs, finding no adverse effects associated with overstuffing. To our knowledge, this was the first clinical study to challenge the importance of overstuffing. While they were able to quantify percent change in postoperative stuffing, the amount of overstuffing was not directly quantifiable. In the present study, the presence of calibration allowed us to draw outcomes based on directly quantifiable measures. Furthermore, the presence of a single knee design eliminated potential confounders. Whereas clinical studies have not shown an association between changes in PFO and outcomes, biomechanical studies demonstrated mixed results. In a cadaveric biomechanical study, Hsu et al [6] tested knee ROM with 3 different patella thickness levels, finding that ROM was not significantly affected by patellar thickness. On the other hand, a more recent biomechanical study found that increasing the patellar thickness exponentially decreased knee flexion [4]. This decrease, however, may not be clinically significant [4]. We found frequent changes in PFO postoperatively, with a tendency toward decreased PFO in our sample. The etiology for changes in PFO is multifactorial [4,6,17]. In the majority of knees, the magnitude of change in PFO was overall relatively small. Our findings are therefore likely only relevant for small changes and should not be extrapolated to extreme changes in PFO. Increased PFO was associated with increased patellar tilt in this cohort of patients. Previous investigations indicated that patellar tilt may increase component wear [18]. In addition, increased tilt may also contribute to patellar maltracking [19]. Maltracking has been associated with various complications affecting the PFJ, such as anterior knee pain and ROM limitation [19,20]. Our findings did not show an association between adverse outcomes and increased patellar tilt. Unlike the biomechanical studies showing increased loading and wear [18], other clinical studies have also not shown an association between increased tilt and adverse patient outcomes [21]. The present study has a number of limitations. The use of radiographs as opposed to an advanced imaging modality, such as magnetic resonance imaging, prevented us from taking cartilage thickness into account for our measurements. This is particularly important for anterior femoral offset and AP femoral size. Based on previous work by DeFrate et al [13] and Cohen et al [14], the average cartilage thickness of the distal femur is between 2.0 ± 0.5 mm and 2.1 ± 0.6 mm, respectively. Therefore, we used an additional sensitivity analysis to account for cartilage thickness of þ1mm, þ2mm, þ3mm, and þ4mm. The results of
J. Matz et al. / The Journal of Arthroplasty xxx (2016) 1e6
this analysis were consistent with the remainder of our results. The angle of the X-ray beam on the skyline view can affect the anterior patellar offset measurement. While some variability certainly existed in our data, the majority of radiographs are performed with a standardized technique in our high-volume center. In addition, our institution historically used the standard skyline view of the patella as opposed to the patellofemoral axial weight-bearing views, which have previously been shown to correlate with anterior knee pain [22]. Other limitations of this study are as follows. The patientreported outcome scores are susceptible to a ceiling effect, which may affect the results of the study. All surgeons used an inlay technique for patellar resurfacing, making the results potentially not generalizable for the onlay technique. Finally, while using a sample with a single knee design improves our internal validity, it may also limit the generalizability of our results to other designs. In conclusion, this study quantified changes in PFO following TKA using a large sample size and calibrated measurements. Our findings not only highlight the difficulty in maintaining the PFO postoperatively but also show that small changes in PFO may not adversely affect clinical outcomes post-TKA. Given the potential for increased contact forces at the PFJ, further studies are required to establish whether changes in PFO would lead to adverse implant wear implications. Acknowledgments The authors would like to thank Lyndsay Somerville for her advice regarding the statistical analysis. Appendix A. Supplementary Data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.arth.2016.08.032. References 1. Russell RD, Huo MH, Jones RE. Avoiding patellar complications in total knee replacement. Bone Joint J 2014;96-B(11 Supple A):84. 2. Schiavone Panni A, Cerciello S, Del Regno C, et al. Patellar resurfacing complications in total knee arthroplasty. Int Orthop 2014;38(2):313.
5
3. Losina E, Katz JN. Total knee arthroplasty on the rise in younger patients: are we sure that past performance will guarantee future success? Arthritis Rheum 2012;64(2):339. 4. Abolghasemian M, Samiezadeh S, Sternheim A, et al. Effect of patellar thickness on knee flexion in total knee arthroplasty: a biomechanical and experimental study. J Arthroplasty 2014;29(1):80. 5. Mihalko W, Fishkin Z, Krackow K. Patellofemoral overstuff and its relationship to flexion after total knee arthroplasty. Clin Orthop Relat Res 2006;449:283. 6. Hsu HC, Luo ZP, Rand JA, et al. Influence of patellar thickness on patellar tracking and patellofemoral contact characteristics after total knee arthroplasty. J Arthroplasty 1996;11(1):69. 7. Kawahara S, Matsuda S, Fukagawa S, et al. Upsizing the femoral component increases patellofemoral contact force in total knee replacement. J Bone Joint Surg Br 2012;94(1):56. 8. Koh JS, Yeo SJ, Lee BP, et al. Influence of patellar thickness on results of total knee arthroplasty: does a residual bony patellar thickness of
6
J. Matz et al. / The Journal of Arthroplasty xxx (2016) 1e6
Appendix
Carlage Thickness +1mm +2mm WOMAC Pain
p=0.399
p=0.343
Funcon Sffness Total
p=0.690 p=0.379 p=0.552
p=0.996 p=0.209 p=0.672
Pain Funcon Total Range of Moon
p=0.075 p=0.788 p=0.858 p=0.631
p=0.29 p=0.591 p=0.701 p=0.868
KSS
Fig. A1. Sensitivity analysis showing comparisons in patient-reported outcomes between implants with increased, maintained, and decreased anterior femoral offset with presumed cartilage thickness of þ1mm and þ2mm. KSS, Knee Society Score; WOMAC, Western Ontario and McMasters Osteoarthritis Index.
+1mm WOMAC Pain
Carlage Thickness +2mm +3mm
p=0.945
p=0.311
p=0.874
Funcon Sffness Total
p=0.899 p=0.915 p=0.884
p=0.993 p=0.600 p=0.455
p=0.752 p=0.932 p=0.804
Pain Funcon Total Range of Moon
p=0.139 p=0.394 p=0.971 p=0.139
p=0.139 p=0.394 p=0.756 p=0.392
p=0.114 p=0.066 p=0.208 p=0.710
+4mm p=0.217 p=0.292 p=0.723 p=0.262
KSS p=0.406 p=0.545 p=0.906 p=0.795
Fig. A2. Sensitivity analysis showing comparisons in patient-reported outcomes between implants with increased, maintained, and decreased AP femoral size with presumed cartilage thickness of þ1mm, þ2mm, þ3mm, and þ4mm. AP, anteroposterior.