N. Goyal et al. / The Journal of Arthroplasty xxx (2015) xxx–xxx
1
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Does Implant Design Influence the Accuracy of Patient Specific Instrumentation in Total Knee Arthroplasty? Nitin Goyal, BA a, Anay R. Patel, MD a, Mark A. Yaffe, MD a, Michael Y. Luo, MD a, S. David Stulberg, MD a,b a b
Northwestern Feinberg School of Medicine Department of Orthopaedic Surgery, Chicago, Illinois Northshore Orthopaedics, Chicago, Illinois
a r t i c l e
i n f o
Article history: Received 1 February 2015 Accepted 20 March 2015 Available online xxxx Keywords: total knee arthroplasty patient specific instrumentation implant design alignment preoperative planning
a b s t r a c t PSI software adjusts preoperative planning to accommodate differences in implant design. Such adjustments may influence the accuracy of intraoperative jig placement, bone resection, or component placement. Our purpose was to determine whether implant design influences PSI accuracy. 96 and 123 PSI TKA were performed by a single surgeon using two different implant systems and identical PSI software. Femoral coronal alignment outliers were greater for Implant 1 (23.9% Implant 1 vs. 13.4% Implant 2; P = 0.050). Tibial coronal alignment outliers were greater for Implant 2 (10.9% Implant 1 vs. 22.7% Implant 2; P = 0.025). There was no difference in overall mechanical axes. Differences in implant design can influence bone resection and component alignment. PSI software rationale must align with surgeons’ intraoperative goals. © 2015 Elsevier Inc. All rights reserved.
Patient specific instrumentation (PSI) is a technique for performing total knee arthroplasty (TKA) that utilizes preoperative magnetic resonance imaging (MRI) or computed tomography (CT) to generate a preoperative plan aimed at achieving surgeon-specified preferences. Rapid prototyping technology is used to generate customized guides for intraoperative cutting block placement with the goal of executing the preoperative plan. The evidence surrounding PSI in the literature has shown mixed results. Compared to conventional instrumentation, PSI has been shown in some studies to improve alignment [1–3], reduce length of surgery [4,5], and demonstrate cost effectiveness [6]. However, other studies have shown PSI to be comparable to conventional instrumentation with regard to improving alignment [4,5,7–14], reducing length of surgery [8], and demonstrating cost effectiveness [4]. The variable results within the literature regarding PSI accuracy may be attributable to the use of different PSI systems or implant systems among published studies. During the preoperative planning phase, PSI software fits a given implant onto a modeled knee. PSI software adjusts its preoperative planning to accommodate specific features of implant design. Due to the particularity of these planned adjustments, it is possible that the accuracy of a given PSI system with regard to intraoperative jig placement, bone resection, or component placement may vary when applied to different implant designs. Thus, the effect of implant design on PSI accuracy needs to be 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.2015.03.019. This study was performed at Northwestern Feinberg School of Medicine and approved by the Northwestern University Institute Review Board. Reprint requests: Nitin Goyal, BA, 600 N. McClurg Ct #1402A, Chicago, IL 60611.
established in order to understand whether the accuracy of a given PSI system is consistent across implant systems. The purpose of this study was to determine whether implant design influences PSI accuracy by comparing the accuracy with which a single PSI system achieved planned intraoperative and radiographic goals in two different implant systems differing in femoral posterior condyle thickness, femoral sizing increments, and tibial tray design. We hypothesized that the accuracy of jig placement, bone resection, and component placement would differ when applying the same PSI planning software to two different implant systems. Materials and Methods In this retrospective comparative study approved by the institutional review board, we evaluated a single experienced surgeon’s (SDS) initial 96 consecutive PSI TKA with Implant 1: Persona CR implant system (Zimmer, Warsaw, IN, USA) and initial 123 consecutive PSI TKA with Implant 2: NexGen CR Flex implant system (Zimmer). All patients during this period underwent PSI TKA unless unable to undergo preoperative MR scanning. All patients undergoing PSI TKA received the most current implant system available unless they had a nickel allergy, in which case the NexGen implant system was used. All TKA were performed with Zimmer Patient Specific Instruments, which utilizes an MR-based pin guide system. The Zimmer PSI, Persona, and NexGen systems are all FDA approved medical devices. Preoperative long-standing radiographs were obtained to evaluate overall preoperative mechanical axis. There were no demographic or preoperative radiographic differences between the implant groups (Table 1). Implant 1 includes a number of specific design changes compared to Implant 2 (Fig. 1), most notably femoral posterior condyle thickness, femoral component sizing increments, and tibial tray design. Femoral
http://dx.doi.org/10.1016/j.arth.2015.03.019 0883-5403/© 2015 Elsevier Inc. All rights reserved.
Please cite this article as: Goyal N, et al, Does Implant Design Influence the Accuracy of Patient Specific Instrumentation in Total Knee Arthroplasty?, J Arthroplasty (2015), http://dx.doi.org/10.1016/j.arth.2015.03.019
2
N. Goyal et al. / The Journal of Arthroplasty xxx (2015) xxx–xxx
Table 1 Demographic and Radiographic Data of Implant Groups. Demographic
Implant 1 (95% CI)
Implant 2 (95% CI)
P Value
Preoperative Mechanical Axisa (°) Age (Years)
5.7 (4.3–7.0) 66.5 (64.8–68.1) 31.0 (29.6–32.4) 67.4 98.9
4.2 (2.7–5.7) 67.3 (65.6–69.1) 30.8 (29.4–32.3) 64.4 100
0.150 0.479
Body Mass Index (kg/m2) Gender (% Female) Preoperative Diagnosis (% Osteoarthritis) a
0.878 0.662 0.264
Positive values indicate varus in the coronal plane; CI = confidence interval.
component posterior condyle thickness is 9 mm in Implant 1 compared to 11 mm in Implant 2. PSI software adjusts targeted posterior femoral resection accordingly. Femoral component sizes are in increments of 2 mm posteriorly in Implant 1 compared to 4 mm posteriorly in Implant 2. Tibial component design is anatomic for Implant 1 and symmetric for Implant 2. We determined in a separate study that in order to maximize tibial surface coverage, the symmetric tray must be frequently internally rotated while the anatomic tray can be maintained in neutral rotation. PSI software similarly attempts to maximize tibial surface coverage in both systems and therefore uses different rotational axes between the two implants. The Implant 1 tibial component is rotated according to the line joining the medial third of the tibial tuberosity and the middle of the PCL, while the Implant 2 tibial component is rotated according to the line running through the midspine point that is perpendicular to the line connecting the geometric centers of the medial and lateral tibia plateau. This rotational difference plans the Implant 2 component to be internally rotated relative to the Implant 1 in the PSI software (Fig. 2). MR images were uploaded to Materialise (Leuven, Belgium) software, which generated a preoperative plan according to surgeon preferences. Surgeon preferences were as follows: overall mechanical axis 0°, femoral coronal alignment 90° relative to femoral mechanical axis, tibial coronal alignment 90° relative to the tibial mechanical axis, femoral flexion 3° relative to femoral sagittal mechanical axis, and tibial posterior slope 4–7° relative to tibial sagittal mechanical axis. Following plan approval, femoral and tibial guides customized to patient anatomy were manufactured for intraoperative use.
Intraoperatively, PSI guides were fitted onto the femur and tibia to establish cutting block placement and subsequent bony resection. The depth of resections from the medial and lateral distal femur, posterior femur, and proximal tibia were measured at designated anatomic points with calipers to the nearest 0.5 mm by a surgical assistant blinded to planned resection depth. Discrepancy between initial resection and PSI-planned resection was calculated, accounting for a saw blade thickness of 1.3 mm. A recut was performed if measured resection fell short of planned resection by more than 2 mm. This technique of measuring resection depth has been validated by Bae et al, who confirmed that caliper-measured thickness of resected condyles after cartilage removal corresponds well with radiographic-measured thickness (average difference of 0.3°) [15]. Our method differed only in that we measured resected bone as well as cartilage to compare with PSI-planned resection, since the MR-based PSI software included cartilage resection in its planning of resection. The surgeon determined appropriate component sizes based on his intraoperative assessment. Component sizes were recorded and compared to PSI predicted sizes to determine sizing accuracy. The surgeon’s intraoperative goal for tibial rotation differed from that of the PSI planning software. While the PSI software attempted to maximize coverage and accordingly adjust rotation, the surgeon’s priority was to minimize component rotation mismatch in an effort to avoid excess postoperative pain [16] and suboptimal tracking. The PSI planning software did not have the option to align the tibial component with the femoral component. In accordance with his intraoperative rotational goal, the surgeon set tibial rotation by floating the tibial tray to align with the femoral component. In both systems, this was most accurately achieved when the tibial component was aligned with the line joining the medial third of the tibial tubercle and the PCL attachment. Thus, the Implant 2 tibial component had to be frequently externally rotated relative to the PSI planned rotation. Intraoperative accuracy was assessed according to discrepancy between actual and PSI-planned resection and component sizing Long-standing and lateral radiographs were obtained 4-weeks and 6months postoperatively to determine postoperative alignment. Overall mechanical axis was measured on both 4-weeks and 6-months postoperative radiographs. Femoral and tibial coronal alignment was measured on 4weeks postoperative long-standing radiographs. Femoral and tibial sagittal alignment was measured on 4-weeks postoperative lateral radiographs.
Fig. 1. Implant 1 and Implant 2 differ in femoral posterior condyle thickness (9 mm vs. 11 mm), femoral component sizing increments (2 mm vs. 4 mm), and tibial tray design (anatomic vs. symmetric).
Please cite this article as: Goyal N, et al, Does Implant Design Influence the Accuracy of Patient Specific Instrumentation in Total Knee Arthroplasty?, J Arthroplasty (2015), http://dx.doi.org/10.1016/j.arth.2015.03.019
N. Goyal et al. / The Journal of Arthroplasty xxx (2015) xxx–xxx
3
Table 2 Intraoperative Accuracy of PSI for Implant Groups. Measure Component Sizes (% Accurate) Femoral Component Within 1 Size (%) Tibial Component Within 1 Size (%) Within 2 Sizes (%) Difference between Planned and Actual Resection (mm) Medial Posterior Femur Lateral Posterior Femur Medial Distal Femur Lateral Distal Femur Medial Proximal Tibia Lateral Proximal Tibia
Implant 1 (95% CI)
Implant 2 (95% CI)
79.8 100.0 77.1 98.9 100.0
82.9 100.0 73.1 97.6 100.0
1.7 (1.4–2.1) 0.3 (−0.1–0.7) −0.9 (−1.3 to −0.5) −0.2 (−0.6–0.2) 1.6 (1.2–1.9) 0.5 (0.2–0.8)
1.0 (0.6–1.3) −0.1 (−0.5–0.2) −1.1 (−1.4 to −0.8) −0.2 (−0.5–0.1) 0.5 (0.2–0.8) −0.6 (−1.0 to −0.2)
P Value
0.631 0.949
*0.003 0.088 0.308 0.953 *b0.001 *b0.001
Positive values indicate over resection relative to plan.
alignment was significantly greater for Implant 2 (0.6° varus Implant 1 vs. 1.5° varus Implant 2; P = 0.002), as was the proportion of tibial coronal alignment outliers (10.9% vs. 22.7%; P = 0.025). Deviation from planned tibial slope differed significantly between groups (1.9° anterior Implant 1 vs. 0.5° anterior Implant 2; P = 0.002), but no significant difference was found in the corresponding proportion of outliers (45.3% Implant 1 vs. 45.0% Implant 2; P = 0.969). Fig. 2. PSI software, in an attempt to maximize coverage for the anatomic and symmetric tibial tray designs, plans the Implant 2 tibial tray to be internally rotated relative to the Implant 1.
The degree of deviation from PSI planned alignment was recorded. Achievement of PSI-planned alignment was assessed according to overall mechanical axis, proportion of mechanical axis outliers (outside 3° neutral), deviation from PSI-planned femoral and tibial coronal and sagittal alignment, and proportion of femoral and tibial component outliers (outside 3° from planned alignment). Statistical analysis was performed using Stata 10 Statistical Software (StataCorp, 2007. Stata Statistical Software: Release 10. College Station, TX: StataCorp LP). Unpaired Student’s t-tests were performed to assess differences in discrepancy between PSI-planed and actual resection, overall mechanical axis, and deviation from PSI-planned component alignment. Pearson’s chi-squared tests were performed to assess differences in component sizing accuracy and the proportion of alignment outliers. Results Medial posterior femoral resection relative to plan was significantly greater for Implant 1 in knees with preoperative varus deformity (2.0 mm over resection Implant 1 vs. 1.0 mm over resection Implant 2; P b 0.001). Medial and lateral tibial resection relative to plan was significantly greater for Implant 1 (medial tibia: 1.6 mm over resection Implant 1 vs. 0.5 mm over resection Implant 2; P b 0.001; lateral tibia: 0.5 mm over resection Implant 1 vs. 0.6 mm under resection Implant 2; P b 0.001). Sizing accuracy in the femur and tibia did not differ significantly between groups (Table 2). Overall, PSI component sizing accuracy was within 1 size in 100% of the femoral components and 98% of the tibial components. Overall mechanical axis did not differ significantly between groups (Table 3). No significant difference was found in the proportion of overall mechanical axis outliers both at 4-weeks (26.1% Implant 1 vs. 26.9% Implant 2; P = 0.896) and 6-months postoperative (11.6% Implant 1 vs. 10.5% Implant 2; P = 0.829). Component alignment differed significantly between groups in both the femur and tibia (Table 3). The proportion of femoral coronal alignment outliers was significantly greater for Implant 1 (23.9% Implant 1 vs. 13.4% Implant 2; P = 0.050). Deviation from planned tibial coronal
Discussion The increasing prevalence of total knee arthroplasty strongly incentivizes efforts to improve the surgical process with regard to accuracy and efficiency given the morbidity and expense of revision arthroplasty for failed primary TKA. Patient specific instrumentation offers the potential for improvements in TKA accuracy and efficiency compared to conventional instrumentation; however, currently published literature demonstrates mixed results to this effect [1,2,4,5,7–11]. To fully appreciate the efficacy of this new technology, it is critically important to understand if the accuracy of a given PSI system is consistent across implant systems. In this study, we sought to assess the accuracy with which PSI achieved planned intraoperative and radiographic goals in two different implant systems differing in femoral posterior condyle thickness, femoral sizing increments, and tibial tray design. We Table 3 Radiographic Achievement of PSI-Planned Alignment for Implant Groups. Measure Mechanical Axis 4-Weeks Postoperative Overall Mechanical Axis (°) Outliers (%) 6-Months Postoperative Overall Mechanical Axis (°) Outliers (%) Femoral Coronal Alignment Deviation from Plan (°) Outliers (%) Tibial Coronal Alignment Deviation from Plan (°) Outliers (%) Femoral Flexion Deviation from Plan (°) Outliers (%) Tibial Slope Deviation from Plan (°) Outliers (%)
Implant 1 (95% CI)
Implant 2 (95% CI)
P Value
1.3 (0.6–1.9)
0.6 (0.1–1.1)
0.105
26.1
26.9
0.896
0.9 (0.2–1.6)
0.9 (0.3–1.4)
0.880
11.6
10.5
0.829
0.5 (0.0–1.1) 23.9
0.0 (−0.3–0.4) 13.4
0.079 *0.050
0.6 (0.2–1.0) 10.9
1.5 (1.1–2.0) 22.7
*0.002 *0.025
0.7 (0.0–1.5) 34.7
1.4 (0.7–2.1) 40.8
0.197 0.361
−1.9 (−2.5 to −1.3) −0.5 (−1.1 to −0.1) *0.002 45.3 45.0 0.969
Positive values indicate varus in coronal plane, flexion or posterior slope in sagittal plane. Outlier defined as greater than 3° deviation from plan.
Please cite this article as: Goyal N, et al, Does Implant Design Influence the Accuracy of Patient Specific Instrumentation in Total Knee Arthroplasty?, J Arthroplasty (2015), http://dx.doi.org/10.1016/j.arth.2015.03.019
4
N. Goyal et al. / The Journal of Arthroplasty xxx (2015) xxx–xxx
hypothesized that the accuracy of jig placement, bone resection, and component placement would differ due to differences in implant design. Femoral bone resection and component alignment differed between implant groups. Medial posterior femoral resection was 1.0 mm greater in Implant 1 compared to Implant 2 in knees with preoperative varus deformity. We attribute this difference to difference in posterior femoral condyle thickness between implant groups. Target posterior femoral resection was 9 mm for Implant 1 and 11 mm for Implant 2. The thinner resection for Implant 1 may have caused increased blade interaction with sclerotic bone, leading to greater blade deflection. This difference may have been more prominent for knees in varus deformity especially at the posterior medial aspect of the femur. Femoral coronal alignment outliers were greater in Implant 1 compared to Implant 2. This difference may have been due to the increased medial posterior femur resection error in Implant 1, which may have influenced femoral rotation and consequently femoral coronal alignment due to presence of femoral flexion. Differences in femoral component intraoperative and radiographic accuracy can likely be attributed to intraoperative technical issues associated with posterior femoral condyle thickness as opposed to PSI software errors. Tibial bone resection and component alignment differed between implant groups. Medial and lateral tibial resection was approximately 1.1 mm greater in Implant 1 compared to Implant 2. We attribute this difference to the way that the manufactured jigs fit on the tibial surface. The tibial jigs were oriented according to the PSI planned rotation. Since the PSI software planned rotation differently between implant groups in response to the different tibial tray design, the jigs were oriented differently and thus fit differently onto the tibial surface. This difference in fit may have led to greater over resection relative to plan for Implant 1. Tibial coronal alignment error and the proportion of corresponding outliers were significantly greater in Implant 2 compared to Implant 1. We attribute the difference to the disagreement between the PSI-planned tibial rotational axis and the intraoperative tibial rotational axis for Implant 2. The surgeon’s intraoperative goal for tibial rotation was consistent between systems in attempting to minimize component rotational mismatch. However, the PSI software did not have such an option and instead attempted to maximize surface coverage, planning Implant 2 to be internally rotated as a consequence. The intraoperative rotation by the surgeon was found to frequently disagree with the PSI-planned Implant 2 tibial rotation, resulting in external rotation of the Implant 2 tibial component. This intraoperative discrepancy between planned and actual rotation did not occur when Implant 1 was used. Since the resection for Implant 2 was made with the cutting guide oriented toward the Implant 2 anteroposterior axis, included posterior slope, and the component was subsequently placed in a relatively externally rotated position, the ultimate result was a varus position of the Implant 2 tibial component. Differences in tibial component intraoperative and radiographic accuracy can likely be attributed to PSI software rationale for tibial rotation not matching the surgeon’s intraoperative goals. Despite the differences in bone resection and component placement between groups, no difference in overall postoperative mechanical axis was found between implant groups at 4-weeks (26.1% Implant 1 vs. 26.9% Implant 2) and 6-months postoperative (11.6% Implant 1 vs. 10.5% Implant 2). The decreased percentage of outliers between 4weeks and 6-months suggests possible incorrect positioning due to pain or soft-tissue swelling at 4-weeks that influences measurement of overall mechanical axis. Also possible is a difference in posture at 4weeks compared to 6-months. At 4-weeks postoperative, soft tissues may not have yet adjusted to the newly determined limb alignment afforded by the TKA, thus contributing to a posture that affects measurement of overall alignment at 4-weeks versus 6-months. Yaffe et al reported mechanical axis measurements at 4-weeks postoperative and 2-years postoperative for a cohort of patients, finding the 4-week mechanical axes to range from 4° valgus to 8° varus while the 6-month mechanical axes ranged from 2° valgus to 4° varus [17]. The effect of time postoperative on mechanical axis measurement should be further
studied. Despite the difference between measures at 4-weeks and 6months, no differences in overall mechanical axis were found at both time points between implant groups. This lack of difference may be due to increased femoral component outliers in Implant 1 and increased tibial component outliers in Implant 2 leading to the same proportion of overall limb alignment outliers. Sizing accuracy was found to be similar between implant groups, suggesting that implant design did not influence PSI software’s ability to predict component sizes in these two systems. Femoral component sizing increments different between implant groups, as did tibial tray shape and dimensions. However, despite these differences, PSI predicted component sizes with similar accuracy for both implant systems. The determinants of PSI sizing accuracy and error should be further evaluated. A limitation of this study is its retrospective design that potentially makes the groups difficult to compare. However, both implant groups were no different with regard to preoperative mechanical axis and demographic variables. The surgeon’s initial cases were chosen for each implant system in a consecutive fashion so that any learning curve associated with the use of each implant system would apply equally to both systems. A second limitation of this study is the inability to measure component rotation given the imaging we had available. Given the difference found in femoral resection between groups, it would be interesting to see how femoral component rotation would be affected. Additionally, the question raised by this study with regard to planned tibial rotation versus intraoperative tibial rotation could be further explored by measuring postoperative tibial component rotation with CT scans. It could be argued that the inclusion of a single PSI system and only two implant systems all developed by the same manufacturer limits the generalizability of our study. However, the use of a single PSI system and the similarities of the two implant systems with the exception of specific differences in femoral posterior condyle thickness, femoral component sizing increments, and tibial tray design is what allows us to attribute differences between groups to the specific differences in implant design. It is possible that implant systems differing in design features other than femoral posterior condyle thickness, femoral component sizing increments, and tibial tray design influence PSI accuracy differently or not at all. Further study with different implant systems is warranted. However, given the variability of planning between PSI software [17], implant systems should only be compared with the use of a single PSI software in order to be able to isolate the effects of specific aspects of implant design. Since a given implant systems is typically only utilized by its respective manufacturer’s PSI system, then only implant systems developed by a single manufacturer can be compared directly. Therefore, this study design allows for comparison of two implant systems while minimizing the potential variation provided by different PSI systems. Longstaff et al has shown that good coronal implant alignment is associated with better functional outcomes [18]. Within the literature, several investigators have reported different results with regard to component alignment achievement with PSI. Chen et al reported 7% coronal femoral alignment outliers and 10% coronal tibial alignment outliers with Zimmer PSI and NexGen LPS (Zimmer), Roh et al reported 4.8% coronal femoral outliers and 0% coronal tibial outliers with Signature (Biomet Inc, Warsaw, IN, USA) and Vanguard PS (Biomet), and Victor et al found 23% femoral outliers and 15% tibial outliers with a cohort of multiple PSI systems and implants. We found 24% and 13% femoral coronal alignment outliers and 11% and 23% tibial coronal alignment outliers with two different implant systems. Differences in implant design may account for some of the differences in the accuracy of component alignment reported using PSI. The present study suggests that differences in implant design can influence the accuracy of bone resection and component alignment for a given PSI system. Surgeons should be aware of these differences when comparing different implant systems using the same PSI technology. As a given implant design evolves, the accuracy of a PSI system should be validated concurrently. PSI software rationale must align with surgeons’ intraoperative goals.
Please cite this article as: Goyal N, et al, Does Implant Design Influence the Accuracy of Patient Specific Instrumentation in Total Knee Arthroplasty?, J Arthroplasty (2015), http://dx.doi.org/10.1016/j.arth.2015.03.019
N. Goyal et al. / The Journal of Arthroplasty xxx (2015) xxx–xxx
Acknowledgements We thank Heather McKinley MMS, PA-C for her valued contribution to this study.
References 1. Daniilidis K, Tibesku CO. A comparison of conventional and patient-specific instruments in total knee arthroplasty. Int Orthop 2013;38(3):503. 2. Ng VY, DeClaire JH, Berend KR, et al. Improved accuracy of alignment with patientspecific positioning guides compared with manual instrumentation in TKA. Clin Orthop Relat Res 2012;470(1):99. 3. Ivie CB, Probst PJ, Bal AK, et al. Improved radiographic outcomes with patient-specific total knee arthroplasty. J Arthroplast 2014;29(11):2100. 4. Nunley RM, Ellison BS, Ruh EL, et al. Are patient-specific cutting blocks cost-effective for total knee arthroplasty? Clin Orthop Relat Res 2012;470(3):889. 5. Nunley RM, Ellison BS, Zhu J, et al. Do patient-specific guides improve coronal alignment in total knee arthroplasty? Clin Orthop Relat Res 2012;470(3):895. 6. DeHaan AM, Adams JR, DeHart ML, et al. Patient-specific versus conventional instrumentation for total knee arthroplasty: peri-operative and cost differences. J Arthroplast 2014;29(11):2065. 7. Barrett W, Hoeffel D, Dalury D, et al. In-vivo alignment comparing patient specific instrumentation with both conventional and computer assisted surgery (CAS) instrumentation in total knee arthroplasty. J Arthroplast 2013;29(2):343.
5
8. Hamilton WG, Parks NL, Saxena A. Patient-specific instrumentation does not shorten surgical time: a prospective, randomized trial. J Arthroplast 2013;28(8 Suppl):96. 9. Roh YW, Kim TW, Lee S, et al. Is TKA using patient-specific instruments comparable to conventional TKA? A randomized controlled study of one system. Clin Orthop Relat Res 2013;471(12):3988. 10. Victor J, Dujardin J, Vandenneucker H, et al. Patient-specific guides do not improve accuracy in total knee arthroplasty: a prospective randomized controlled trial. Clin Orthop Relat Res 2013;472(1):263. 11. Barrack RL, Ruh EL, Williams BM, et al. Patient specific cutting blocks are currently of no proven value. J Bone Joint Surg Br 2012;94(11 Suppl A):95. 12. Marimuthu K, Chen DB, Harris IA, et al. A multi-planar CT-based comparative analysis of patient-specific cutting guides with conventional instrumentation in total knee arthroplasty. J Arthroplast 2013;29(6):1138. 13. Stronach BM, Pelt CE, Erickson JA, et al. Patient-specific instrumentation in total knee arthroplasty provides no improvement in component alignment. J Arthroplast 2014;29(9):1705. 14. Woolson ST, Harris AH, Wagner DW, et al. Component alignment during total knee arthroplasty with use of standard or custom instrumentation: a randomized clinical trial using computed tomography for postoperative alignment measurement. J Bone Joint Surg Am 2014;96(5):366. 15. Bae DK, Song SJ, Yoon KH, et al. Intraoperative assessment of resected condyle thickness in total knee arthroplasty. Knee Surg Sports Traumatol Arthrosc 2012;20(10):2039. 16. Bell SW, Young P, Drury C, et al. Component rotational alignment in unexplained painful primary total knee arthroplasty. Knee 2014;21(1):272. 17. Yaffe MA, Koo SS, Stulberg SD. Radiographic and navigation measurements of TKA limb alignment do not correlate. Clin Orthop Relat Res 2008;466(11):2736. 18. Longstaff LM, Sloan K, Stamp N, et al. Good alignment after total knee arthroplasty leads to faster rehabilitation and better function. J Arthroplast 2009;24(4):570.
Please cite this article as: Goyal N, et al, Does Implant Design Influence the Accuracy of Patient Specific Instrumentation in Total Knee Arthroplasty?, J Arthroplasty (2015), http://dx.doi.org/10.1016/j.arth.2015.03.019
1
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Does Implant Design Influence the Accuracy of Patient Specific Instrumentation in Total Knee Arthroplasty? Nitin Goyal, BA a, Anay R. Patel, MD a, Mark A. Yaffe, MD a, Michael Y. Luo, MD a, S. David Stulberg, MD a,b a b
Northwestern Feinberg School of Medicine Department of Orthopaedic Surgery, Chicago, Illinois Northshore Orthopaedics, Chicago, Illinois
a r t i c l e
i n f o
Article history: Received 1 February 2015 Accepted 20 March 2015 Available online xxxx Keywords: total knee arthroplasty patient specific instrumentation implant design alignment preoperative planning
a b s t r a c t PSI software adjusts preoperative planning to accommodate differences in implant design. Such adjustments may influence the accuracy of intraoperative jig placement, bone resection, or component placement. Our purpose was to determine whether implant design influences PSI accuracy. 96 and 123 PSI TKA were performed by a single surgeon using two different implant systems and identical PSI software. Femoral coronal alignment outliers were greater for Implant 1 (23.9% Implant 1 vs. 13.4% Implant 2; P = 0.050). Tibial coronal alignment outliers were greater for Implant 2 (10.9% Implant 1 vs. 22.7% Implant 2; P = 0.025). There was no difference in overall mechanical axes. Differences in implant design can influence bone resection and component alignment. PSI software rationale must align with surgeons’ intraoperative goals. © 2015 Elsevier Inc. All rights reserved.
Patient specific instrumentation (PSI) is a technique for performing total knee arthroplasty (TKA) that utilizes preoperative magnetic resonance imaging (MRI) or computed tomography (CT) to generate a preoperative plan aimed at achieving surgeon-specified preferences. Rapid prototyping technology is used to generate customized guides for intraoperative cutting block placement with the goal of executing the preoperative plan. The evidence surrounding PSI in the literature has shown mixed results. Compared to conventional instrumentation, PSI has been shown in some studies to improve alignment [1–3], reduce length of surgery [4,5], and demonstrate cost effectiveness [6]. However, other studies have shown PSI to be comparable to conventional instrumentation with regard to improving alignment [4,5,7–14], reducing length of surgery [8], and demonstrating cost effectiveness [4]. The variable results within the literature regarding PSI accuracy may be attributable to the use of different PSI systems or implant systems among published studies. During the preoperative planning phase, PSI software fits a given implant onto a modeled knee. PSI software adjusts its preoperative planning to accommodate specific features of implant design. Due to the particularity of these planned adjustments, it is possible that the accuracy of a given PSI system with regard to intraoperative jig placement, bone resection, or component placement may vary when applied to different implant designs. Thus, the effect of implant design on PSI accuracy needs to be 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.2015.03.019. This study was performed at Northwestern Feinberg School of Medicine and approved by the Northwestern University Institute Review Board. Reprint requests: Nitin Goyal, BA, 600 N. McClurg Ct #1402A, Chicago, IL 60611.
established in order to understand whether the accuracy of a given PSI system is consistent across implant systems. The purpose of this study was to determine whether implant design influences PSI accuracy by comparing the accuracy with which a single PSI system achieved planned intraoperative and radiographic goals in two different implant systems differing in femoral posterior condyle thickness, femoral sizing increments, and tibial tray design. We hypothesized that the accuracy of jig placement, bone resection, and component placement would differ when applying the same PSI planning software to two different implant systems. Materials and Methods In this retrospective comparative study approved by the institutional review board, we evaluated a single experienced surgeon’s (SDS) initial 96 consecutive PSI TKA with Implant 1: Persona CR implant system (Zimmer, Warsaw, IN, USA) and initial 123 consecutive PSI TKA with Implant 2: NexGen CR Flex implant system (Zimmer). All patients during this period underwent PSI TKA unless unable to undergo preoperative MR scanning. All patients undergoing PSI TKA received the most current implant system available unless they had a nickel allergy, in which case the NexGen implant system was used. All TKA were performed with Zimmer Patient Specific Instruments, which utilizes an MR-based pin guide system. The Zimmer PSI, Persona, and NexGen systems are all FDA approved medical devices. Preoperative long-standing radiographs were obtained to evaluate overall preoperative mechanical axis. There were no demographic or preoperative radiographic differences between the implant groups (Table 1). Implant 1 includes a number of specific design changes compared to Implant 2 (Fig. 1), most notably femoral posterior condyle thickness, femoral component sizing increments, and tibial tray design. Femoral
http://dx.doi.org/10.1016/j.arth.2015.03.019 0883-5403/© 2015 Elsevier Inc. All rights reserved.
Please cite this article as: Goyal N, et al, Does Implant Design Influence the Accuracy of Patient Specific Instrumentation in Total Knee Arthroplasty?, J Arthroplasty (2015), http://dx.doi.org/10.1016/j.arth.2015.03.019
2
N. Goyal et al. / The Journal of Arthroplasty xxx (2015) xxx–xxx
Table 1 Demographic and Radiographic Data of Implant Groups. Demographic
Implant 1 (95% CI)
Implant 2 (95% CI)
P Value
Preoperative Mechanical Axisa (°) Age (Years)
5.7 (4.3–7.0) 66.5 (64.8–68.1) 31.0 (29.6–32.4) 67.4 98.9
4.2 (2.7–5.7) 67.3 (65.6–69.1) 30.8 (29.4–32.3) 64.4 100
0.150 0.479
Body Mass Index (kg/m2) Gender (% Female) Preoperative Diagnosis (% Osteoarthritis) a
0.878 0.662 0.264
Positive values indicate varus in the coronal plane; CI = confidence interval.
component posterior condyle thickness is 9 mm in Implant 1 compared to 11 mm in Implant 2. PSI software adjusts targeted posterior femoral resection accordingly. Femoral component sizes are in increments of 2 mm posteriorly in Implant 1 compared to 4 mm posteriorly in Implant 2. Tibial component design is anatomic for Implant 1 and symmetric for Implant 2. We determined in a separate study that in order to maximize tibial surface coverage, the symmetric tray must be frequently internally rotated while the anatomic tray can be maintained in neutral rotation. PSI software similarly attempts to maximize tibial surface coverage in both systems and therefore uses different rotational axes between the two implants. The Implant 1 tibial component is rotated according to the line joining the medial third of the tibial tuberosity and the middle of the PCL, while the Implant 2 tibial component is rotated according to the line running through the midspine point that is perpendicular to the line connecting the geometric centers of the medial and lateral tibia plateau. This rotational difference plans the Implant 2 component to be internally rotated relative to the Implant 1 in the PSI software (Fig. 2). MR images were uploaded to Materialise (Leuven, Belgium) software, which generated a preoperative plan according to surgeon preferences. Surgeon preferences were as follows: overall mechanical axis 0°, femoral coronal alignment 90° relative to femoral mechanical axis, tibial coronal alignment 90° relative to the tibial mechanical axis, femoral flexion 3° relative to femoral sagittal mechanical axis, and tibial posterior slope 4–7° relative to tibial sagittal mechanical axis. Following plan approval, femoral and tibial guides customized to patient anatomy were manufactured for intraoperative use.
Intraoperatively, PSI guides were fitted onto the femur and tibia to establish cutting block placement and subsequent bony resection. The depth of resections from the medial and lateral distal femur, posterior femur, and proximal tibia were measured at designated anatomic points with calipers to the nearest 0.5 mm by a surgical assistant blinded to planned resection depth. Discrepancy between initial resection and PSI-planned resection was calculated, accounting for a saw blade thickness of 1.3 mm. A recut was performed if measured resection fell short of planned resection by more than 2 mm. This technique of measuring resection depth has been validated by Bae et al, who confirmed that caliper-measured thickness of resected condyles after cartilage removal corresponds well with radiographic-measured thickness (average difference of 0.3°) [15]. Our method differed only in that we measured resected bone as well as cartilage to compare with PSI-planned resection, since the MR-based PSI software included cartilage resection in its planning of resection. The surgeon determined appropriate component sizes based on his intraoperative assessment. Component sizes were recorded and compared to PSI predicted sizes to determine sizing accuracy. The surgeon’s intraoperative goal for tibial rotation differed from that of the PSI planning software. While the PSI software attempted to maximize coverage and accordingly adjust rotation, the surgeon’s priority was to minimize component rotation mismatch in an effort to avoid excess postoperative pain [16] and suboptimal tracking. The PSI planning software did not have the option to align the tibial component with the femoral component. In accordance with his intraoperative rotational goal, the surgeon set tibial rotation by floating the tibial tray to align with the femoral component. In both systems, this was most accurately achieved when the tibial component was aligned with the line joining the medial third of the tibial tubercle and the PCL attachment. Thus, the Implant 2 tibial component had to be frequently externally rotated relative to the PSI planned rotation. Intraoperative accuracy was assessed according to discrepancy between actual and PSI-planned resection and component sizing Long-standing and lateral radiographs were obtained 4-weeks and 6months postoperatively to determine postoperative alignment. Overall mechanical axis was measured on both 4-weeks and 6-months postoperative radiographs. Femoral and tibial coronal alignment was measured on 4weeks postoperative long-standing radiographs. Femoral and tibial sagittal alignment was measured on 4-weeks postoperative lateral radiographs.
Fig. 1. Implant 1 and Implant 2 differ in femoral posterior condyle thickness (9 mm vs. 11 mm), femoral component sizing increments (2 mm vs. 4 mm), and tibial tray design (anatomic vs. symmetric).
Please cite this article as: Goyal N, et al, Does Implant Design Influence the Accuracy of Patient Specific Instrumentation in Total Knee Arthroplasty?, J Arthroplasty (2015), http://dx.doi.org/10.1016/j.arth.2015.03.019
N. Goyal et al. / The Journal of Arthroplasty xxx (2015) xxx–xxx
3
Table 2 Intraoperative Accuracy of PSI for Implant Groups. Measure Component Sizes (% Accurate) Femoral Component Within 1 Size (%) Tibial Component Within 1 Size (%) Within 2 Sizes (%) Difference between Planned and Actual Resection (mm) Medial Posterior Femur Lateral Posterior Femur Medial Distal Femur Lateral Distal Femur Medial Proximal Tibia Lateral Proximal Tibia
Implant 1 (95% CI)
Implant 2 (95% CI)
79.8 100.0 77.1 98.9 100.0
82.9 100.0 73.1 97.6 100.0
1.7 (1.4–2.1) 0.3 (−0.1–0.7) −0.9 (−1.3 to −0.5) −0.2 (−0.6–0.2) 1.6 (1.2–1.9) 0.5 (0.2–0.8)
1.0 (0.6–1.3) −0.1 (−0.5–0.2) −1.1 (−1.4 to −0.8) −0.2 (−0.5–0.1) 0.5 (0.2–0.8) −0.6 (−1.0 to −0.2)
P Value
0.631 0.949
*0.003 0.088 0.308 0.953 *b0.001 *b0.001
Positive values indicate over resection relative to plan.
alignment was significantly greater for Implant 2 (0.6° varus Implant 1 vs. 1.5° varus Implant 2; P = 0.002), as was the proportion of tibial coronal alignment outliers (10.9% vs. 22.7%; P = 0.025). Deviation from planned tibial slope differed significantly between groups (1.9° anterior Implant 1 vs. 0.5° anterior Implant 2; P = 0.002), but no significant difference was found in the corresponding proportion of outliers (45.3% Implant 1 vs. 45.0% Implant 2; P = 0.969). Fig. 2. PSI software, in an attempt to maximize coverage for the anatomic and symmetric tibial tray designs, plans the Implant 2 tibial tray to be internally rotated relative to the Implant 1.
The degree of deviation from PSI planned alignment was recorded. Achievement of PSI-planned alignment was assessed according to overall mechanical axis, proportion of mechanical axis outliers (outside 3° neutral), deviation from PSI-planned femoral and tibial coronal and sagittal alignment, and proportion of femoral and tibial component outliers (outside 3° from planned alignment). Statistical analysis was performed using Stata 10 Statistical Software (StataCorp, 2007. Stata Statistical Software: Release 10. College Station, TX: StataCorp LP). Unpaired Student’s t-tests were performed to assess differences in discrepancy between PSI-planed and actual resection, overall mechanical axis, and deviation from PSI-planned component alignment. Pearson’s chi-squared tests were performed to assess differences in component sizing accuracy and the proportion of alignment outliers. Results Medial posterior femoral resection relative to plan was significantly greater for Implant 1 in knees with preoperative varus deformity (2.0 mm over resection Implant 1 vs. 1.0 mm over resection Implant 2; P b 0.001). Medial and lateral tibial resection relative to plan was significantly greater for Implant 1 (medial tibia: 1.6 mm over resection Implant 1 vs. 0.5 mm over resection Implant 2; P b 0.001; lateral tibia: 0.5 mm over resection Implant 1 vs. 0.6 mm under resection Implant 2; P b 0.001). Sizing accuracy in the femur and tibia did not differ significantly between groups (Table 2). Overall, PSI component sizing accuracy was within 1 size in 100% of the femoral components and 98% of the tibial components. Overall mechanical axis did not differ significantly between groups (Table 3). No significant difference was found in the proportion of overall mechanical axis outliers both at 4-weeks (26.1% Implant 1 vs. 26.9% Implant 2; P = 0.896) and 6-months postoperative (11.6% Implant 1 vs. 10.5% Implant 2; P = 0.829). Component alignment differed significantly between groups in both the femur and tibia (Table 3). The proportion of femoral coronal alignment outliers was significantly greater for Implant 1 (23.9% Implant 1 vs. 13.4% Implant 2; P = 0.050). Deviation from planned tibial coronal
Discussion The increasing prevalence of total knee arthroplasty strongly incentivizes efforts to improve the surgical process with regard to accuracy and efficiency given the morbidity and expense of revision arthroplasty for failed primary TKA. Patient specific instrumentation offers the potential for improvements in TKA accuracy and efficiency compared to conventional instrumentation; however, currently published literature demonstrates mixed results to this effect [1,2,4,5,7–11]. To fully appreciate the efficacy of this new technology, it is critically important to understand if the accuracy of a given PSI system is consistent across implant systems. In this study, we sought to assess the accuracy with which PSI achieved planned intraoperative and radiographic goals in two different implant systems differing in femoral posterior condyle thickness, femoral sizing increments, and tibial tray design. We Table 3 Radiographic Achievement of PSI-Planned Alignment for Implant Groups. Measure Mechanical Axis 4-Weeks Postoperative Overall Mechanical Axis (°) Outliers (%) 6-Months Postoperative Overall Mechanical Axis (°) Outliers (%) Femoral Coronal Alignment Deviation from Plan (°) Outliers (%) Tibial Coronal Alignment Deviation from Plan (°) Outliers (%) Femoral Flexion Deviation from Plan (°) Outliers (%) Tibial Slope Deviation from Plan (°) Outliers (%)
Implant 1 (95% CI)
Implant 2 (95% CI)
P Value
1.3 (0.6–1.9)
0.6 (0.1–1.1)
0.105
26.1
26.9
0.896
0.9 (0.2–1.6)
0.9 (0.3–1.4)
0.880
11.6
10.5
0.829
0.5 (0.0–1.1) 23.9
0.0 (−0.3–0.4) 13.4
0.079 *0.050
0.6 (0.2–1.0) 10.9
1.5 (1.1–2.0) 22.7
*0.002 *0.025
0.7 (0.0–1.5) 34.7
1.4 (0.7–2.1) 40.8
0.197 0.361
−1.9 (−2.5 to −1.3) −0.5 (−1.1 to −0.1) *0.002 45.3 45.0 0.969
Positive values indicate varus in coronal plane, flexion or posterior slope in sagittal plane. Outlier defined as greater than 3° deviation from plan.
Please cite this article as: Goyal N, et al, Does Implant Design Influence the Accuracy of Patient Specific Instrumentation in Total Knee Arthroplasty?, J Arthroplasty (2015), http://dx.doi.org/10.1016/j.arth.2015.03.019
4
N. Goyal et al. / The Journal of Arthroplasty xxx (2015) xxx–xxx
hypothesized that the accuracy of jig placement, bone resection, and component placement would differ due to differences in implant design. Femoral bone resection and component alignment differed between implant groups. Medial posterior femoral resection was 1.0 mm greater in Implant 1 compared to Implant 2 in knees with preoperative varus deformity. We attribute this difference to difference in posterior femoral condyle thickness between implant groups. Target posterior femoral resection was 9 mm for Implant 1 and 11 mm for Implant 2. The thinner resection for Implant 1 may have caused increased blade interaction with sclerotic bone, leading to greater blade deflection. This difference may have been more prominent for knees in varus deformity especially at the posterior medial aspect of the femur. Femoral coronal alignment outliers were greater in Implant 1 compared to Implant 2. This difference may have been due to the increased medial posterior femur resection error in Implant 1, which may have influenced femoral rotation and consequently femoral coronal alignment due to presence of femoral flexion. Differences in femoral component intraoperative and radiographic accuracy can likely be attributed to intraoperative technical issues associated with posterior femoral condyle thickness as opposed to PSI software errors. Tibial bone resection and component alignment differed between implant groups. Medial and lateral tibial resection was approximately 1.1 mm greater in Implant 1 compared to Implant 2. We attribute this difference to the way that the manufactured jigs fit on the tibial surface. The tibial jigs were oriented according to the PSI planned rotation. Since the PSI software planned rotation differently between implant groups in response to the different tibial tray design, the jigs were oriented differently and thus fit differently onto the tibial surface. This difference in fit may have led to greater over resection relative to plan for Implant 1. Tibial coronal alignment error and the proportion of corresponding outliers were significantly greater in Implant 2 compared to Implant 1. We attribute the difference to the disagreement between the PSI-planned tibial rotational axis and the intraoperative tibial rotational axis for Implant 2. The surgeon’s intraoperative goal for tibial rotation was consistent between systems in attempting to minimize component rotational mismatch. However, the PSI software did not have such an option and instead attempted to maximize surface coverage, planning Implant 2 to be internally rotated as a consequence. The intraoperative rotation by the surgeon was found to frequently disagree with the PSI-planned Implant 2 tibial rotation, resulting in external rotation of the Implant 2 tibial component. This intraoperative discrepancy between planned and actual rotation did not occur when Implant 1 was used. Since the resection for Implant 2 was made with the cutting guide oriented toward the Implant 2 anteroposterior axis, included posterior slope, and the component was subsequently placed in a relatively externally rotated position, the ultimate result was a varus position of the Implant 2 tibial component. Differences in tibial component intraoperative and radiographic accuracy can likely be attributed to PSI software rationale for tibial rotation not matching the surgeon’s intraoperative goals. Despite the differences in bone resection and component placement between groups, no difference in overall postoperative mechanical axis was found between implant groups at 4-weeks (26.1% Implant 1 vs. 26.9% Implant 2) and 6-months postoperative (11.6% Implant 1 vs. 10.5% Implant 2). The decreased percentage of outliers between 4weeks and 6-months suggests possible incorrect positioning due to pain or soft-tissue swelling at 4-weeks that influences measurement of overall mechanical axis. Also possible is a difference in posture at 4weeks compared to 6-months. At 4-weeks postoperative, soft tissues may not have yet adjusted to the newly determined limb alignment afforded by the TKA, thus contributing to a posture that affects measurement of overall alignment at 4-weeks versus 6-months. Yaffe et al reported mechanical axis measurements at 4-weeks postoperative and 2-years postoperative for a cohort of patients, finding the 4-week mechanical axes to range from 4° valgus to 8° varus while the 6-month mechanical axes ranged from 2° valgus to 4° varus [17]. The effect of time postoperative on mechanical axis measurement should be further
studied. Despite the difference between measures at 4-weeks and 6months, no differences in overall mechanical axis were found at both time points between implant groups. This lack of difference may be due to increased femoral component outliers in Implant 1 and increased tibial component outliers in Implant 2 leading to the same proportion of overall limb alignment outliers. Sizing accuracy was found to be similar between implant groups, suggesting that implant design did not influence PSI software’s ability to predict component sizes in these two systems. Femoral component sizing increments different between implant groups, as did tibial tray shape and dimensions. However, despite these differences, PSI predicted component sizes with similar accuracy for both implant systems. The determinants of PSI sizing accuracy and error should be further evaluated. A limitation of this study is its retrospective design that potentially makes the groups difficult to compare. However, both implant groups were no different with regard to preoperative mechanical axis and demographic variables. The surgeon’s initial cases were chosen for each implant system in a consecutive fashion so that any learning curve associated with the use of each implant system would apply equally to both systems. A second limitation of this study is the inability to measure component rotation given the imaging we had available. Given the difference found in femoral resection between groups, it would be interesting to see how femoral component rotation would be affected. Additionally, the question raised by this study with regard to planned tibial rotation versus intraoperative tibial rotation could be further explored by measuring postoperative tibial component rotation with CT scans. It could be argued that the inclusion of a single PSI system and only two implant systems all developed by the same manufacturer limits the generalizability of our study. However, the use of a single PSI system and the similarities of the two implant systems with the exception of specific differences in femoral posterior condyle thickness, femoral component sizing increments, and tibial tray design is what allows us to attribute differences between groups to the specific differences in implant design. It is possible that implant systems differing in design features other than femoral posterior condyle thickness, femoral component sizing increments, and tibial tray design influence PSI accuracy differently or not at all. Further study with different implant systems is warranted. However, given the variability of planning between PSI software [17], implant systems should only be compared with the use of a single PSI software in order to be able to isolate the effects of specific aspects of implant design. Since a given implant systems is typically only utilized by its respective manufacturer’s PSI system, then only implant systems developed by a single manufacturer can be compared directly. Therefore, this study design allows for comparison of two implant systems while minimizing the potential variation provided by different PSI systems. Longstaff et al has shown that good coronal implant alignment is associated with better functional outcomes [18]. Within the literature, several investigators have reported different results with regard to component alignment achievement with PSI. Chen et al reported 7% coronal femoral alignment outliers and 10% coronal tibial alignment outliers with Zimmer PSI and NexGen LPS (Zimmer), Roh et al reported 4.8% coronal femoral outliers and 0% coronal tibial outliers with Signature (Biomet Inc, Warsaw, IN, USA) and Vanguard PS (Biomet), and Victor et al found 23% femoral outliers and 15% tibial outliers with a cohort of multiple PSI systems and implants. We found 24% and 13% femoral coronal alignment outliers and 11% and 23% tibial coronal alignment outliers with two different implant systems. Differences in implant design may account for some of the differences in the accuracy of component alignment reported using PSI. The present study suggests that differences in implant design can influence the accuracy of bone resection and component alignment for a given PSI system. Surgeons should be aware of these differences when comparing different implant systems using the same PSI technology. As a given implant design evolves, the accuracy of a PSI system should be validated concurrently. PSI software rationale must align with surgeons’ intraoperative goals.
Please cite this article as: Goyal N, et al, Does Implant Design Influence the Accuracy of Patient Specific Instrumentation in Total Knee Arthroplasty?, J Arthroplasty (2015), http://dx.doi.org/10.1016/j.arth.2015.03.019
N. Goyal et al. / The Journal of Arthroplasty xxx (2015) xxx–xxx
Acknowledgements We thank Heather McKinley MMS, PA-C for her valued contribution to this study.
References 1. Daniilidis K, Tibesku CO. A comparison of conventional and patient-specific instruments in total knee arthroplasty. Int Orthop 2013;38(3):503. 2. Ng VY, DeClaire JH, Berend KR, et al. Improved accuracy of alignment with patientspecific positioning guides compared with manual instrumentation in TKA. Clin Orthop Relat Res 2012;470(1):99. 3. Ivie CB, Probst PJ, Bal AK, et al. Improved radiographic outcomes with patient-specific total knee arthroplasty. J Arthroplast 2014;29(11):2100. 4. Nunley RM, Ellison BS, Ruh EL, et al. Are patient-specific cutting blocks cost-effective for total knee arthroplasty? Clin Orthop Relat Res 2012;470(3):889. 5. Nunley RM, Ellison BS, Zhu J, et al. Do patient-specific guides improve coronal alignment in total knee arthroplasty? Clin Orthop Relat Res 2012;470(3):895. 6. DeHaan AM, Adams JR, DeHart ML, et al. Patient-specific versus conventional instrumentation for total knee arthroplasty: peri-operative and cost differences. J Arthroplast 2014;29(11):2065. 7. Barrett W, Hoeffel D, Dalury D, et al. In-vivo alignment comparing patient specific instrumentation with both conventional and computer assisted surgery (CAS) instrumentation in total knee arthroplasty. J Arthroplast 2013;29(2):343.
5
8. Hamilton WG, Parks NL, Saxena A. Patient-specific instrumentation does not shorten surgical time: a prospective, randomized trial. J Arthroplast 2013;28(8 Suppl):96. 9. Roh YW, Kim TW, Lee S, et al. Is TKA using patient-specific instruments comparable to conventional TKA? A randomized controlled study of one system. Clin Orthop Relat Res 2013;471(12):3988. 10. Victor J, Dujardin J, Vandenneucker H, et al. Patient-specific guides do not improve accuracy in total knee arthroplasty: a prospective randomized controlled trial. Clin Orthop Relat Res 2013;472(1):263. 11. Barrack RL, Ruh EL, Williams BM, et al. Patient specific cutting blocks are currently of no proven value. J Bone Joint Surg Br 2012;94(11 Suppl A):95. 12. Marimuthu K, Chen DB, Harris IA, et al. A multi-planar CT-based comparative analysis of patient-specific cutting guides with conventional instrumentation in total knee arthroplasty. J Arthroplast 2013;29(6):1138. 13. Stronach BM, Pelt CE, Erickson JA, et al. Patient-specific instrumentation in total knee arthroplasty provides no improvement in component alignment. J Arthroplast 2014;29(9):1705. 14. Woolson ST, Harris AH, Wagner DW, et al. Component alignment during total knee arthroplasty with use of standard or custom instrumentation: a randomized clinical trial using computed tomography for postoperative alignment measurement. J Bone Joint Surg Am 2014;96(5):366. 15. Bae DK, Song SJ, Yoon KH, et al. Intraoperative assessment of resected condyle thickness in total knee arthroplasty. Knee Surg Sports Traumatol Arthrosc 2012;20(10):2039. 16. Bell SW, Young P, Drury C, et al. Component rotational alignment in unexplained painful primary total knee arthroplasty. Knee 2014;21(1):272. 17. Yaffe MA, Koo SS, Stulberg SD. Radiographic and navigation measurements of TKA limb alignment do not correlate. Clin Orthop Relat Res 2008;466(11):2736. 18. Longstaff LM, Sloan K, Stamp N, et al. Good alignment after total knee arthroplasty leads to faster rehabilitation and better function. J Arthroplast 2009;24(4):570.
Please cite this article as: Goyal N, et al, Does Implant Design Influence the Accuracy of Patient Specific Instrumentation in Total Knee Arthroplasty?, J Arthroplasty (2015), http://dx.doi.org/10.1016/j.arth.2015.03.019
