Knee Surg Sports Traumatol Arthrosc DOI 10.1007/s00167-014-2839-2
KNEE
Improved positioning of the tibial component in unicompartmental knee arthroplasty with patient-specific cutting blocks M. L. Dao Trong • C. Diezi • G. Goerres N. Helmy
•
Received: 8 August 2013 / Accepted: 6 January 2014 Ó Springer-Verlag Berlin Heidelberg 2014
Abstract Purpose Unicompartmental knee arthroplasty (UKA) has recently regained popularity for the treatment of osteoarthritis of the knee. Numerous authors have cited alignment as an important prognostic factor in the survival of UKA. Limb alignment affects not only the longevity of UKA by influencing wear of polyethylene, but also affects the unreplaced contralateral compartment. Malpositioning of the components may result in unequal wear patterns, thus further leading to early failure and additionally influencing clinical outcome as well. However, there is a lack of techniques to assure a high accuracy of the implant positioning. Methods In this study, we investigated tibia component alignment of 28 medial UKAs implanted with patientspecific cutting blocks. Three patients were excluded due to bad imaging. Measurements of tibial component alignment from postoperatively computed tomography (CT) scans were compared to respective CT-based preoperative plannings to assess the accuracy of implant positioning. Results Our results show excellent high accuracy of tibial implant position in tibial varus/valgus (D 0.3° ± 1.7°), posterior slope (D 1.1° ± 2.6°) and external rotation (D 1.5° ± 3.3°). Conclusion We conclude that patient-specific cutting blocks improve the accuracy of tibia component positioning in unicompartmental knee arthroplasty. M. L. Dao Trong (&) C. Diezi N. Helmy Department of Orthopedic Surgery, Bu¨rgerspital Solothurn, Scho¨ngru¨nstrasse 42, 4500 Solothurn, Switzerland e-mail: [email protected] G. Goerres Department of Radiology, Bu¨rgerspital Solothurn, Scho¨ngru¨nstrasse 42, 4500 Solothurn, Switzerland
Level of evidence Level IV.
Case series with no comparison group,
Keywords Unicompartmental knee arthroplasty Computed tomography Alignment Rotation Patientspecific cutting blocks
Introduction Unicompartmental knee arthroplasty (UKA) has recently undergone resurgence in popularity. Surgical technique, approaches, instrumentation and implants have evolved dramatically. The advantages of the UKA over the total knee arthroplasty (TKA) include a preservation of bone stock because of reduced bone resection, retaining ligamentous stability due to anterior cruciate ligament (ACL) sparing techniques and thus maintenance of more normal joint kinematics, better proprioception, better range of motion (ROM), and faster recovery [16]. Careful patient selection, however, is crucial. Currently accepted indications for UKA are noninflammatory unicompartmental arthritis, contained mature osteonecrosis, C90° flexion, intact ACL, flexion contracture B10°, varus or valgus deformities that can be passively corrected to 5–7° of valgus, or outerbridge changes B1 in the opposite compartment or the patellofemoral joint. Numerous studies have reported 10-year survival rates for various devices of at least 85–90 %, which are still inferior to TKA survival rates [1, 20]. Other studies have described high early revision rates for tibial wear, femoral component loosening and progressive disease [13]. Nonetheless, it remains a legitimate alternative as an intermediate intervention before TKA. Additionally, a recent study regarding cost-effectiveness showed UKA to be most
123
Knee Surg Sports Traumatol Arthrosc
economical over TKA if survival rates are a minimum of 12 years [18]. Comparing subjective health status measured in SF-12 and WOMAC pain and physical function scores between UKAs and TKAs, Noticewala et al. [15] have shown a significantly larger improvement in patients who underwent UKA compared with patients with TKA. Several studies have shown the crucial role of limb alignment, varus/valgus malpositioning of the single components and tibial slope for polyethylene wear in implant failure of UKA [4, 6–9]. Therefore, computernavigated systems have been developed to improve the accuracy of alignment with satisfying results in the coronal and the sagittal plane [3, 5]. A recent meta-analysis of ten studies suggested a risk reduction in outliers with computer navigation compared with conventional methods, measuring radiological positioning of the femoral and tibial component in AP and lateral view and the mechanical axis [22]. However, rotational malpositioning has still been poorly described and investigated. Measurements of the rotational position can still only be performed through CT scans and have shown unsatisfying results for UKAs implanted through conventional techniques [2, 17], and furthermore inconsistent results for computer-navigated techniques [5, 14]. Patient-specific cutting blocks for TKA are thriving and have already been investigated on several different implants with inconsistent [10, 12, 19] results in limb Fig. 1 Demonstration of the measured angles through 3D CT reconstruction: measurement of a tibial rotation, b tibial varus/ valgus, c HKA, d tibial posterior slope
123
alignment and tibial slope. However, it seems to be an easy-to-learn instrumentation and technique, making it a promising strategy for the clinical use. Therefore, patientspecific cutting block techniques were also developed for UKA, which is technically more challenging, especially when using minimally invasive techniques. Minding the importance of correct implant position and concurrently overcoming the demanding technique, we conducted this study to investigate the accuracy of the tibial component positioning through patient-specific cutting blocks.
Materials and methods Twenty-eight knees were investigated, and 25 knees from 24 patients (14 females; mean age 70 ± 8.3 years) who received a medial UKA (MyKnee UNIÒ) in the period from October 2010 until March 2012 and agreed to undergo a postoperative CT scan were included. Three patients were excluded due to bad imaging. All patients were suffering from an isolated medial gonarthrosis. Exclusion factors were a flexion contracture [10°, varus or valgus deformities that could not be passively corrected to 5°–7° of valgus, bicondylar and/or patellofemoral osteoarthritis and polyarthritis. The unicompartmental knee prostheses were implanted with patient-specific cutting blocks for the tibial
Knee Surg Sports Traumatol Arthrosc Fig. 2 Measurement of the tibial component rotation after 3D CT reconstruction through the posterior tibial condylar line: a and b determination of the posterior tibial condylar line marked by the red circles. c Measurement of the rotation: angle (a) between the perpendicular and the posterior tibial condylar line and the tibial implant
component (MyKnee UNIÒ, Medacta Switzerland) through a minimally invasive approach. The preoperative planning was set for a maximal resection of 6 mm on the tibial side, and maximal slope was set to patients’ anatomy, however, not greater than 8°. Therefore, patients undergoing this procedure received a preoperative CT scan of the knee with scout scans of the ipsilateral hip and ankle for planning. Preoperative bone 3D models were created by segmenting the CT images using a thresholding tool, followed by a manual editing to refine the model (Fig. 1) using Mimics (materialize). Tibial preoperative planning was based on anatomical reference points, taking into account surgeon preferences. An appropriate planning including resection levels, implant positioning, resulting limb alignment, as well as the femoral/tibial implant sizes is proposed to the surgeon as additional information. After validation by the surgeon, a tibial patient-matched cutting block was provided for surgery. The vertical and horizontal tibial cut was performed through the patient-specific cutting block. The femoral cut was determined according to the tibial cut by a spacer and ligament-balanced technique. Both the tibial and femoral components were cemented implants. After surgery, the accuracy of the component positioning was assessed performing a CT postoperatively with the same protocol used as in the preoperative CT scan. Postoperative bone and implant 3D models were created using the same segmenting procedure of the preoperative phase. To combine preoperative planning with postoperative images, postoperative measures were based on the same anatomical reference points that were initially used in the preoperative analysis, double checking their positions (Fig. 2). Accuracy was assessed comparing the bone cut with the preoperative planning. The final implant position with respect to the planned position was evaluated as well. All measures
were approximated to 0.5 mm and 0.5° to maintain the same accuracy used to edit the preoperative planning. At first follow-up, 6-week post surgery, patients received a conventional X-ray control of the knee as well as long-leg standing X-rays for assessment of the hip–knee–ankle angle (HKA). This study was validated and approved by the ethical review committee in Aargau, Switzerland, reference number 2011/072 and has therefore been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki. All patients included, gave their informed consent prior to their inclusion in this study. Statistical analysis A mean tibial rotation of 2° suggested that a sample size of n = 28 was required. Statistical analysis was performed calculating the arrhythmic mean with standard deviation (SD) and the root mean square error (RMS). Mean differences between planned angles, final bone cut and final implant position were calculated using paired t test.
Results Preoperative measurement of HKA in the CT scans showed mean values of 175.4° ± 2.5° varus compared to postoperative HKA with 177° ± 2.8° varus (D 1.8° ± 3.3°; p = 0.02). The measurements of postoperative HKA on conventional X-rays showed comparable results with 177.3° ± 2.8° (n = 23). In two patients, postoperative measurement of HKA on X-ray was not possible due to missing follow-ups 6 weeks postoperatively. The differences between postoperative final bone cut compared to the
123
Knee Surg Sports Traumatol Arthrosc
preoperative planning showed for the tibial varus/valgus a mean of 0.5° ± 1.2° (2.9° ± 0.7° varus planned; n.s.), for the tibial posterior slope 0.7° ± 2.0° (4.6° ± 1.3° planned; n.s.), and for the tibial implant rotation a mean difference of 1.6° ± 3.5° external rotation (0° planned; p = 0.03). The differences between postoperative final implant position compared to the preoperative planning was very similar with 0.03° ± 1.7° (n.s.) for the tibial varus/valgus, 1.1° ± 2.6° (p = 0.03) for the tibial posterior slope and 1.5° ± 3.3° (p = 0.05) for the external rotation. Comparing these differences of final bone cut to those of final implant position, the difference for tibial varus/valgus was D 0.1° ± 1.6° (n.s.), for tibial slope D 0.4° ± 2.5° (n.s.) and for rotation D 0.2° ± 3.5° (n.s.) (Table 1). In all of the 25 cases, the implanted size of the tibial and femoral component was the same as planned preoperatively.
Discussion The here presented study shows that patient-specific instrumentation has the potential to improve the accuracy of the tibial component in unicompartimental knee arthroplasty, subsequently leading to a possibly better orientation of the femoral component, improved kinematics and increased longevity of the implant. UKA is an established method for treating single-compartment osteoarthritis. But, implanting techniques are still technically more demanding with chance for consecutive malalignment of the components, resulting in early-implant failure through either aseptic loosening, excessive polyethylene wear and/or disease progression in the opposite compartment—and thus short time to implant failure [4, 6– 9]. Several studies have investigated alternative techniques with computer navigation to improve component alignment showing inconsistent results for coronal and sagittal alignment compared with conventional methods [5, 21]. However, rotational alignment has yet been poorly investigated, and as to our knowledge, no study has examined the rotational alignment in UKAs implanted through patient-specific instrumentation so far. Servien et al. [17] investigated the tibial component rotation in UKAs
implanted through conventional ligament-adjusted methods in postoperative CT scans and described poor results (6.5° ± 5.1° external rotation for medial UKA, 7.3° ± 10.3° for lateral UKAs). A recent prospective pilot study with 13 patients investigating a computer-assisted minimally invasive UKA showed adequate positioning of the tibial and femoral component regarding flexion– extension and varus–valgus, but poor results for the component rotation with large variations from the target range in both components (femoral 8/10 [3° rotation, tibial 6/10 [3° rotation) [14]. In this study, using patient-specific cutting blocks planned and manufactured through preoperative CT scans, good results for accuracy of tibial component varus/valgus positioning, tibial slope and rotation with only very little deviation to the preoperative planning could be documented. We have never observed a positive posterior slope. For the tibial rotation, only the difference between final bone cut, not final implant position, and preoperative planning was statistically significant. Most probably, this has only little to no clinical significance. The same also applies inversely to the tibial varus/valgus. The HKA could be maintained with a mean difference of 1.8° ± 3.3° (p = 0.02), also statistically, but most likely not clinically significant. However, no complete correction of the varus alignment was observed corresponding to the aim for maintaining a slight varus or neutral alignment. Compared with previous studies for UKAs implanted with conventional techniques, rotation of the tibial component was superior with smaller variation from the planned range [14, 17]. The mean value of rotation was externally rotated, which could be explained through precaution while performing the vertical tibial cut in means not to injure the insertion of the ACL. When comparing the final implant position to the final bone cut, the differences observed could be due to different factors: cementing technique, intraoperative implant placement or errors associated with the CT-based 3D measurements. Compared with the recent study performed by Dunbar et al. [5] with good alignment and rotational results using dynamic tactile-guided UKAs (Table 2), our results showed a very small difference always less than 1°.
Table 1 Differences between postoperative final bone cut and postoperative final implant position compared to the preoperative planning and compared to each other given as mean ± SD CT measurements
Planned
D Final bone cut to planned
D Final implant position to planned
D Final bone cut to final implant position
Tibial varus/valgus
2.9° ± 0.7° varus
0.5° ± 1.2°
0.3° ± 1.7°
0.1° ± 1.6°
Tibial posterior slope
-4.6° ± 1.3°
0.7° ± 2.0°
1.1° ± 2.6°
0.4° ± 2.5°
External rotation
0°
1.6° ± 3.5°
1.5° ± 3.3°
0.2° ± 3.5°
123
Knee Surg Sports Traumatol Arthrosc Table 2 Comparison of our results in root mean square (RMS) with the study from Dunbar et al. [5] investigating dynamic tactile-guided unicompartmental knee arthroplasty
Dunbar et al. final implant position versus planning
MyKnee UNIÒ final implant position versus planning
MyKnee UNIÒ bone cut versus planning
Tibial varus/ valgus
1.5°
1.7°
1.3°
Tibial posterior slope
1.9°
2.8°
2.1°
3.7°
3.8°
External rotation
3°
Considering that our bone cuts were performed with an oscillating saw while Dunbar et al. used a motorized computer controlled reamer, we suggest that UKAs with patient-specific cutting blocks can provide good outcomes comparable with those with dynamic tactile-guided UKAs. A major limitation of this present study might be the evaluation of only the tibia alignment. We consciously limited our investigation to the tibia alignment because we strictly follow the principle of femoral component orientation according to the tibial cuts. Considering this ligament-oriented technique for retaining physiological stability, we anticipated a ligament-balanced orientation of the femoral cut and thus implant position compared to the tibial implant position, which would not be possible in a solely CT-based planning. Therefore, to allow the surgeon to control the knee balancing intraoperatively, patientspecific cutting blocks were only planned and manufactured for the tibial cut. The position of the femoral component depended on the surgeon’s preferences. Thus, we restricted our measurements to the tibia alignment as to only evaluate the accuracy of patient-specific cutting block technique. However, other manufacturers of patient-specific instrumentation for UKAs like, for example the SignatureTM Partial Knee (BIOMET), are offering patientspecific cutting blocks for both tibial and femoral side. The iUniÒ (ConforMIS) even produces patient-specific implants in addition to the patient-specific cutting blocks with consecutive higher production costs. Koeck et al. [11] investigated the iUniÒ and showed accurate results for implant position compared to the preoperative planning in the coronal and sagittal plane (both p \ 0.001). But, measurements were only done on conventional radiographs, so rotational alignment could not be assessed. The actual investigations were concentrated to the CT scan measurements of one particular prosthesis model; assessment of the clinical outcomes still needs to be carried out. Long-term results are yet missing and need to be assessed in the future. Findings and performance, as well as clinical consequences, may differ for the different available prosthesis designs. Further studies would be necessary for a closer understanding of particular mid-/long-term relevance of the above-mentioned findings in regard of consequence for implant survival and clinical outcome.
UKA has a huge impact on the patient’s well-being and has superior biomechanical properties compared to TKA. However, implantation technique remains challenging as instrumentation and computer-assisted surgery are dependent on surgical skills and high volume of implantations. With patient-specific cutting blocks, the technical challenge of correct implant positioning can be overcome and therefore may lead to a wider spread of UKA also among lowervolume surgeons, so that more patients profit from the better biomechanical properties in UKA compared with TKA.
Conclusion The results of the present study show accurate tibia implant position in UKAs with patients-specific cutting block technique and are comparable to those for navigated UKAs. Thus, patient-specific cutting block technique seems to be a promising strategy to optimize implant positioning for UKAs and therefore improving longevity of implant survival. Conflict of interest The authors declare a financial and inventor relationship with MedactaÒ Switzerland.
References 1. Ackroyd CE (2003) Medial compartment arthroplasty of the knee. J Bone Joint Surg Br 85(7):937–942 2. Campbell DG, Johnson LJ, West SC (2006) Multiparameter quantitative computer-assisted tomography assessment of unicompartmental knee arthroplasties. ANZ J Surg 76(9):782–787 3. Cobb J, Henckel J, Gomes P, Harris S, Jakopec M, Rodriguez F, Barrett A, Davies B (2006) Hands-on robotic unicompartmental knee replacement: a prospective, randomised controlled study of the acrobot system. J Bone Joint Surg Br 88(2):188–197 4. Diezi C, Wirth S, Meyer DC, Koch PP (2010) Effect of femoral to tibial varus mismatch on the contact area of unicondylar knee prostheses. Knee 17(5):350–355 5. Dunbar NJ, Roche MW, Park BH, Branch SH, Conditt MA, Banks SA (2012) Accuracy of dynamic tactile-guided unicompartmental knee arthroplasty. J Arthroplast 27(5):803–808 6. Engh GA, Dwyer KA, Hanes CK (1992) Polyethylene wear of metal-backed tibial components in total and unicompartmental knee prostheses. J Bone Joint Surg Br 74(1):9–17 7. Hernigou P, Deschamps G (2004) Posterior slope of the tibial implant and the outcome of unicompartmental knee arthroplasty. J Bone Joint Surg Am 86-A(3):506–511
123
Knee Surg Sports Traumatol Arthrosc 8. Hernigou P, Deschamps G (2004) Alignment influences wear in the knee after medial unicompartmental arthroplasty. Clin Orthop Relat Res 423:161–165 9. Hernigou P, Poignard A, Filippini P, Zilber S (2008) Retrieved unicompartmental implants with full pe tibial components: the effects of knee alignment and polyethylene thickness on creep and wear. Open Orthop J 2:51–56 10. Koch PP, Mu¨ller D, Pisan M, Fucentese SF (2013) Radiographic accuracy in TKA with a CT-based patient-specific cutting block technique. Knee Surg Sports Traumatol Arthrosc 21(10):2200–2205 11. Koeck FX, Beckmann J, Luring C, Rath B, Grifka J, Basad E (2011) Evaluation of implant position and knee alignment after patient-specific unicompartmental knee arthroplasty. Knee 18(5):294–299 12. Lustig S, Scholes CJ, Oussedik SI, Kinzel V, Coolican MR, Parker DA (2013) Unsatisfactory accuracy as determined by computer navigation of VISIONNAIRE patient-specific instrumentation for total knee arthroplasty. J Arthroplast 28(3):469–473 13. Mariani EM, Bourne MH, Jackson RT, Jackson ST, Jones P (2007) Early failure of unicompartmental knee arthroplasty. J Arthroplast 22(6 Suppl 2):81–84 14. Martinez-Carranza N, Weidenhielm L, Crafoord J, Hedstrom M (2012) Deviation between navigated and final 3-dimensional implant position in mini-invasive unicompartmental knee arthroplasty: a pilot study in 13 patients. Acta Orthop 83(6):625–628 15. Noticewala MS, Geller JA, Lee JH, Macaulay W (2012) Unicompartmental knee arthroplasty relieves pain and improves
123
16.
17.
18.
19.
20.
21.
22.
function more than total knee arthroplasty. J Arthroplast 27(8 Suppl):99–105 O’Rourke MR, Gardner JJ, Callaghan JJ, Liu SS, Goetz DD, Vittetoe DA, Sullivan PM, Johnston RC (2005) The John Insall Award: unicompartmental knee replacement. A minimum twenty-one-year follow-up, end-result study. Clin Orthop Relat Res 440:27–37 Servien E, Fary C, Lustig S, Demey G, Saffarini M, Chomel S, Neyret P (2011) Tibial component rotation assessment using CT scan in medial and lateral unicompartmental knee arthroplasty. Orthop Traumatol Surg Res 97(3):272–275 SooHoo NF, Sharifi H, Kominski G, Lieberman JR (2006) Costeffectiveness analysis of unicompartmental knee arthroplasty as an alternative to total knee arthroplasty for unicompartmental osteoarthritis. J Bone Joint Surg Am 88(9):1975–1982 Stronach BM, Pelt CE, Erickson J, Peters CL (2013) Patientspecific total knee arthroplasty required frequent surgeon-directed changes. Clin Orthop Relat Res 471(1):169–174 Swienckowski JJ, Pennington DW (2004) Unicompartmental knee arthroplasty in patients sixty years of age or younger. J Bone Joint Surg Am 86-A Suppl 1(Pt 2):131–142 Valenzuela GA, Jacobson NA, Geist DJ, Valenzuela RG, Teitge RA (2013) Implant and limb alignment outcomes for conventional and navigated unicompartmental knee arthroplasty. J Arthroplast 28(3):463–468 Weber P, Crispin A, Schmidutz F, Utzschneider S, Pietschmann MF, Jansson V, Mu¨ller PE (2013) Improved accuracy in computer-assisted unicondylar knee arthroplasty: a meta-analysis. Knee Surg Sports Traumatol Arthrosc 21(11):2453–2461
KNEE
Improved positioning of the tibial component in unicompartmental knee arthroplasty with patient-specific cutting blocks M. L. Dao Trong • C. Diezi • G. Goerres N. Helmy
•
Received: 8 August 2013 / Accepted: 6 January 2014 Ó Springer-Verlag Berlin Heidelberg 2014
Abstract Purpose Unicompartmental knee arthroplasty (UKA) has recently regained popularity for the treatment of osteoarthritis of the knee. Numerous authors have cited alignment as an important prognostic factor in the survival of UKA. Limb alignment affects not only the longevity of UKA by influencing wear of polyethylene, but also affects the unreplaced contralateral compartment. Malpositioning of the components may result in unequal wear patterns, thus further leading to early failure and additionally influencing clinical outcome as well. However, there is a lack of techniques to assure a high accuracy of the implant positioning. Methods In this study, we investigated tibia component alignment of 28 medial UKAs implanted with patientspecific cutting blocks. Three patients were excluded due to bad imaging. Measurements of tibial component alignment from postoperatively computed tomography (CT) scans were compared to respective CT-based preoperative plannings to assess the accuracy of implant positioning. Results Our results show excellent high accuracy of tibial implant position in tibial varus/valgus (D 0.3° ± 1.7°), posterior slope (D 1.1° ± 2.6°) and external rotation (D 1.5° ± 3.3°). Conclusion We conclude that patient-specific cutting blocks improve the accuracy of tibia component positioning in unicompartmental knee arthroplasty. M. L. Dao Trong (&) C. Diezi N. Helmy Department of Orthopedic Surgery, Bu¨rgerspital Solothurn, Scho¨ngru¨nstrasse 42, 4500 Solothurn, Switzerland e-mail: [email protected] G. Goerres Department of Radiology, Bu¨rgerspital Solothurn, Scho¨ngru¨nstrasse 42, 4500 Solothurn, Switzerland
Level of evidence Level IV.
Case series with no comparison group,
Keywords Unicompartmental knee arthroplasty Computed tomography Alignment Rotation Patientspecific cutting blocks
Introduction Unicompartmental knee arthroplasty (UKA) has recently undergone resurgence in popularity. Surgical technique, approaches, instrumentation and implants have evolved dramatically. The advantages of the UKA over the total knee arthroplasty (TKA) include a preservation of bone stock because of reduced bone resection, retaining ligamentous stability due to anterior cruciate ligament (ACL) sparing techniques and thus maintenance of more normal joint kinematics, better proprioception, better range of motion (ROM), and faster recovery [16]. Careful patient selection, however, is crucial. Currently accepted indications for UKA are noninflammatory unicompartmental arthritis, contained mature osteonecrosis, C90° flexion, intact ACL, flexion contracture B10°, varus or valgus deformities that can be passively corrected to 5–7° of valgus, or outerbridge changes B1 in the opposite compartment or the patellofemoral joint. Numerous studies have reported 10-year survival rates for various devices of at least 85–90 %, which are still inferior to TKA survival rates [1, 20]. Other studies have described high early revision rates for tibial wear, femoral component loosening and progressive disease [13]. Nonetheless, it remains a legitimate alternative as an intermediate intervention before TKA. Additionally, a recent study regarding cost-effectiveness showed UKA to be most
123
Knee Surg Sports Traumatol Arthrosc
economical over TKA if survival rates are a minimum of 12 years [18]. Comparing subjective health status measured in SF-12 and WOMAC pain and physical function scores between UKAs and TKAs, Noticewala et al. [15] have shown a significantly larger improvement in patients who underwent UKA compared with patients with TKA. Several studies have shown the crucial role of limb alignment, varus/valgus malpositioning of the single components and tibial slope for polyethylene wear in implant failure of UKA [4, 6–9]. Therefore, computernavigated systems have been developed to improve the accuracy of alignment with satisfying results in the coronal and the sagittal plane [3, 5]. A recent meta-analysis of ten studies suggested a risk reduction in outliers with computer navigation compared with conventional methods, measuring radiological positioning of the femoral and tibial component in AP and lateral view and the mechanical axis [22]. However, rotational malpositioning has still been poorly described and investigated. Measurements of the rotational position can still only be performed through CT scans and have shown unsatisfying results for UKAs implanted through conventional techniques [2, 17], and furthermore inconsistent results for computer-navigated techniques [5, 14]. Patient-specific cutting blocks for TKA are thriving and have already been investigated on several different implants with inconsistent [10, 12, 19] results in limb Fig. 1 Demonstration of the measured angles through 3D CT reconstruction: measurement of a tibial rotation, b tibial varus/ valgus, c HKA, d tibial posterior slope
123
alignment and tibial slope. However, it seems to be an easy-to-learn instrumentation and technique, making it a promising strategy for the clinical use. Therefore, patientspecific cutting block techniques were also developed for UKA, which is technically more challenging, especially when using minimally invasive techniques. Minding the importance of correct implant position and concurrently overcoming the demanding technique, we conducted this study to investigate the accuracy of the tibial component positioning through patient-specific cutting blocks.
Materials and methods Twenty-eight knees were investigated, and 25 knees from 24 patients (14 females; mean age 70 ± 8.3 years) who received a medial UKA (MyKnee UNIÒ) in the period from October 2010 until March 2012 and agreed to undergo a postoperative CT scan were included. Three patients were excluded due to bad imaging. All patients were suffering from an isolated medial gonarthrosis. Exclusion factors were a flexion contracture [10°, varus or valgus deformities that could not be passively corrected to 5°–7° of valgus, bicondylar and/or patellofemoral osteoarthritis and polyarthritis. The unicompartmental knee prostheses were implanted with patient-specific cutting blocks for the tibial
Knee Surg Sports Traumatol Arthrosc Fig. 2 Measurement of the tibial component rotation after 3D CT reconstruction through the posterior tibial condylar line: a and b determination of the posterior tibial condylar line marked by the red circles. c Measurement of the rotation: angle (a) between the perpendicular and the posterior tibial condylar line and the tibial implant
component (MyKnee UNIÒ, Medacta Switzerland) through a minimally invasive approach. The preoperative planning was set for a maximal resection of 6 mm on the tibial side, and maximal slope was set to patients’ anatomy, however, not greater than 8°. Therefore, patients undergoing this procedure received a preoperative CT scan of the knee with scout scans of the ipsilateral hip and ankle for planning. Preoperative bone 3D models were created by segmenting the CT images using a thresholding tool, followed by a manual editing to refine the model (Fig. 1) using Mimics (materialize). Tibial preoperative planning was based on anatomical reference points, taking into account surgeon preferences. An appropriate planning including resection levels, implant positioning, resulting limb alignment, as well as the femoral/tibial implant sizes is proposed to the surgeon as additional information. After validation by the surgeon, a tibial patient-matched cutting block was provided for surgery. The vertical and horizontal tibial cut was performed through the patient-specific cutting block. The femoral cut was determined according to the tibial cut by a spacer and ligament-balanced technique. Both the tibial and femoral components were cemented implants. After surgery, the accuracy of the component positioning was assessed performing a CT postoperatively with the same protocol used as in the preoperative CT scan. Postoperative bone and implant 3D models were created using the same segmenting procedure of the preoperative phase. To combine preoperative planning with postoperative images, postoperative measures were based on the same anatomical reference points that were initially used in the preoperative analysis, double checking their positions (Fig. 2). Accuracy was assessed comparing the bone cut with the preoperative planning. The final implant position with respect to the planned position was evaluated as well. All measures
were approximated to 0.5 mm and 0.5° to maintain the same accuracy used to edit the preoperative planning. At first follow-up, 6-week post surgery, patients received a conventional X-ray control of the knee as well as long-leg standing X-rays for assessment of the hip–knee–ankle angle (HKA). This study was validated and approved by the ethical review committee in Aargau, Switzerland, reference number 2011/072 and has therefore been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki. All patients included, gave their informed consent prior to their inclusion in this study. Statistical analysis A mean tibial rotation of 2° suggested that a sample size of n = 28 was required. Statistical analysis was performed calculating the arrhythmic mean with standard deviation (SD) and the root mean square error (RMS). Mean differences between planned angles, final bone cut and final implant position were calculated using paired t test.
Results Preoperative measurement of HKA in the CT scans showed mean values of 175.4° ± 2.5° varus compared to postoperative HKA with 177° ± 2.8° varus (D 1.8° ± 3.3°; p = 0.02). The measurements of postoperative HKA on conventional X-rays showed comparable results with 177.3° ± 2.8° (n = 23). In two patients, postoperative measurement of HKA on X-ray was not possible due to missing follow-ups 6 weeks postoperatively. The differences between postoperative final bone cut compared to the
123
Knee Surg Sports Traumatol Arthrosc
preoperative planning showed for the tibial varus/valgus a mean of 0.5° ± 1.2° (2.9° ± 0.7° varus planned; n.s.), for the tibial posterior slope 0.7° ± 2.0° (4.6° ± 1.3° planned; n.s.), and for the tibial implant rotation a mean difference of 1.6° ± 3.5° external rotation (0° planned; p = 0.03). The differences between postoperative final implant position compared to the preoperative planning was very similar with 0.03° ± 1.7° (n.s.) for the tibial varus/valgus, 1.1° ± 2.6° (p = 0.03) for the tibial posterior slope and 1.5° ± 3.3° (p = 0.05) for the external rotation. Comparing these differences of final bone cut to those of final implant position, the difference for tibial varus/valgus was D 0.1° ± 1.6° (n.s.), for tibial slope D 0.4° ± 2.5° (n.s.) and for rotation D 0.2° ± 3.5° (n.s.) (Table 1). In all of the 25 cases, the implanted size of the tibial and femoral component was the same as planned preoperatively.
Discussion The here presented study shows that patient-specific instrumentation has the potential to improve the accuracy of the tibial component in unicompartimental knee arthroplasty, subsequently leading to a possibly better orientation of the femoral component, improved kinematics and increased longevity of the implant. UKA is an established method for treating single-compartment osteoarthritis. But, implanting techniques are still technically more demanding with chance for consecutive malalignment of the components, resulting in early-implant failure through either aseptic loosening, excessive polyethylene wear and/or disease progression in the opposite compartment—and thus short time to implant failure [4, 6– 9]. Several studies have investigated alternative techniques with computer navigation to improve component alignment showing inconsistent results for coronal and sagittal alignment compared with conventional methods [5, 21]. However, rotational alignment has yet been poorly investigated, and as to our knowledge, no study has examined the rotational alignment in UKAs implanted through patient-specific instrumentation so far. Servien et al. [17] investigated the tibial component rotation in UKAs
implanted through conventional ligament-adjusted methods in postoperative CT scans and described poor results (6.5° ± 5.1° external rotation for medial UKA, 7.3° ± 10.3° for lateral UKAs). A recent prospective pilot study with 13 patients investigating a computer-assisted minimally invasive UKA showed adequate positioning of the tibial and femoral component regarding flexion– extension and varus–valgus, but poor results for the component rotation with large variations from the target range in both components (femoral 8/10 [3° rotation, tibial 6/10 [3° rotation) [14]. In this study, using patient-specific cutting blocks planned and manufactured through preoperative CT scans, good results for accuracy of tibial component varus/valgus positioning, tibial slope and rotation with only very little deviation to the preoperative planning could be documented. We have never observed a positive posterior slope. For the tibial rotation, only the difference between final bone cut, not final implant position, and preoperative planning was statistically significant. Most probably, this has only little to no clinical significance. The same also applies inversely to the tibial varus/valgus. The HKA could be maintained with a mean difference of 1.8° ± 3.3° (p = 0.02), also statistically, but most likely not clinically significant. However, no complete correction of the varus alignment was observed corresponding to the aim for maintaining a slight varus or neutral alignment. Compared with previous studies for UKAs implanted with conventional techniques, rotation of the tibial component was superior with smaller variation from the planned range [14, 17]. The mean value of rotation was externally rotated, which could be explained through precaution while performing the vertical tibial cut in means not to injure the insertion of the ACL. When comparing the final implant position to the final bone cut, the differences observed could be due to different factors: cementing technique, intraoperative implant placement or errors associated with the CT-based 3D measurements. Compared with the recent study performed by Dunbar et al. [5] with good alignment and rotational results using dynamic tactile-guided UKAs (Table 2), our results showed a very small difference always less than 1°.
Table 1 Differences between postoperative final bone cut and postoperative final implant position compared to the preoperative planning and compared to each other given as mean ± SD CT measurements
Planned
D Final bone cut to planned
D Final implant position to planned
D Final bone cut to final implant position
Tibial varus/valgus
2.9° ± 0.7° varus
0.5° ± 1.2°
0.3° ± 1.7°
0.1° ± 1.6°
Tibial posterior slope
-4.6° ± 1.3°
0.7° ± 2.0°
1.1° ± 2.6°
0.4° ± 2.5°
External rotation
0°
1.6° ± 3.5°
1.5° ± 3.3°
0.2° ± 3.5°
123
Knee Surg Sports Traumatol Arthrosc Table 2 Comparison of our results in root mean square (RMS) with the study from Dunbar et al. [5] investigating dynamic tactile-guided unicompartmental knee arthroplasty
Dunbar et al. final implant position versus planning
MyKnee UNIÒ final implant position versus planning
MyKnee UNIÒ bone cut versus planning
Tibial varus/ valgus
1.5°
1.7°
1.3°
Tibial posterior slope
1.9°
2.8°
2.1°
3.7°
3.8°
External rotation
3°
Considering that our bone cuts were performed with an oscillating saw while Dunbar et al. used a motorized computer controlled reamer, we suggest that UKAs with patient-specific cutting blocks can provide good outcomes comparable with those with dynamic tactile-guided UKAs. A major limitation of this present study might be the evaluation of only the tibia alignment. We consciously limited our investigation to the tibia alignment because we strictly follow the principle of femoral component orientation according to the tibial cuts. Considering this ligament-oriented technique for retaining physiological stability, we anticipated a ligament-balanced orientation of the femoral cut and thus implant position compared to the tibial implant position, which would not be possible in a solely CT-based planning. Therefore, to allow the surgeon to control the knee balancing intraoperatively, patientspecific cutting blocks were only planned and manufactured for the tibial cut. The position of the femoral component depended on the surgeon’s preferences. Thus, we restricted our measurements to the tibia alignment as to only evaluate the accuracy of patient-specific cutting block technique. However, other manufacturers of patient-specific instrumentation for UKAs like, for example the SignatureTM Partial Knee (BIOMET), are offering patientspecific cutting blocks for both tibial and femoral side. The iUniÒ (ConforMIS) even produces patient-specific implants in addition to the patient-specific cutting blocks with consecutive higher production costs. Koeck et al. [11] investigated the iUniÒ and showed accurate results for implant position compared to the preoperative planning in the coronal and sagittal plane (both p \ 0.001). But, measurements were only done on conventional radiographs, so rotational alignment could not be assessed. The actual investigations were concentrated to the CT scan measurements of one particular prosthesis model; assessment of the clinical outcomes still needs to be carried out. Long-term results are yet missing and need to be assessed in the future. Findings and performance, as well as clinical consequences, may differ for the different available prosthesis designs. Further studies would be necessary for a closer understanding of particular mid-/long-term relevance of the above-mentioned findings in regard of consequence for implant survival and clinical outcome.
UKA has a huge impact on the patient’s well-being and has superior biomechanical properties compared to TKA. However, implantation technique remains challenging as instrumentation and computer-assisted surgery are dependent on surgical skills and high volume of implantations. With patient-specific cutting blocks, the technical challenge of correct implant positioning can be overcome and therefore may lead to a wider spread of UKA also among lowervolume surgeons, so that more patients profit from the better biomechanical properties in UKA compared with TKA.
Conclusion The results of the present study show accurate tibia implant position in UKAs with patients-specific cutting block technique and are comparable to those for navigated UKAs. Thus, patient-specific cutting block technique seems to be a promising strategy to optimize implant positioning for UKAs and therefore improving longevity of implant survival. Conflict of interest The authors declare a financial and inventor relationship with MedactaÒ Switzerland.
References 1. Ackroyd CE (2003) Medial compartment arthroplasty of the knee. J Bone Joint Surg Br 85(7):937–942 2. Campbell DG, Johnson LJ, West SC (2006) Multiparameter quantitative computer-assisted tomography assessment of unicompartmental knee arthroplasties. ANZ J Surg 76(9):782–787 3. Cobb J, Henckel J, Gomes P, Harris S, Jakopec M, Rodriguez F, Barrett A, Davies B (2006) Hands-on robotic unicompartmental knee replacement: a prospective, randomised controlled study of the acrobot system. J Bone Joint Surg Br 88(2):188–197 4. Diezi C, Wirth S, Meyer DC, Koch PP (2010) Effect of femoral to tibial varus mismatch on the contact area of unicondylar knee prostheses. Knee 17(5):350–355 5. Dunbar NJ, Roche MW, Park BH, Branch SH, Conditt MA, Banks SA (2012) Accuracy of dynamic tactile-guided unicompartmental knee arthroplasty. J Arthroplast 27(5):803–808 6. Engh GA, Dwyer KA, Hanes CK (1992) Polyethylene wear of metal-backed tibial components in total and unicompartmental knee prostheses. J Bone Joint Surg Br 74(1):9–17 7. Hernigou P, Deschamps G (2004) Posterior slope of the tibial implant and the outcome of unicompartmental knee arthroplasty. J Bone Joint Surg Am 86-A(3):506–511
123
Knee Surg Sports Traumatol Arthrosc 8. Hernigou P, Deschamps G (2004) Alignment influences wear in the knee after medial unicompartmental arthroplasty. Clin Orthop Relat Res 423:161–165 9. Hernigou P, Poignard A, Filippini P, Zilber S (2008) Retrieved unicompartmental implants with full pe tibial components: the effects of knee alignment and polyethylene thickness on creep and wear. Open Orthop J 2:51–56 10. Koch PP, Mu¨ller D, Pisan M, Fucentese SF (2013) Radiographic accuracy in TKA with a CT-based patient-specific cutting block technique. Knee Surg Sports Traumatol Arthrosc 21(10):2200–2205 11. Koeck FX, Beckmann J, Luring C, Rath B, Grifka J, Basad E (2011) Evaluation of implant position and knee alignment after patient-specific unicompartmental knee arthroplasty. Knee 18(5):294–299 12. Lustig S, Scholes CJ, Oussedik SI, Kinzel V, Coolican MR, Parker DA (2013) Unsatisfactory accuracy as determined by computer navigation of VISIONNAIRE patient-specific instrumentation for total knee arthroplasty. J Arthroplast 28(3):469–473 13. Mariani EM, Bourne MH, Jackson RT, Jackson ST, Jones P (2007) Early failure of unicompartmental knee arthroplasty. J Arthroplast 22(6 Suppl 2):81–84 14. Martinez-Carranza N, Weidenhielm L, Crafoord J, Hedstrom M (2012) Deviation between navigated and final 3-dimensional implant position in mini-invasive unicompartmental knee arthroplasty: a pilot study in 13 patients. Acta Orthop 83(6):625–628 15. Noticewala MS, Geller JA, Lee JH, Macaulay W (2012) Unicompartmental knee arthroplasty relieves pain and improves
123
16.
17.
18.
19.
20.
21.
22.
function more than total knee arthroplasty. J Arthroplast 27(8 Suppl):99–105 O’Rourke MR, Gardner JJ, Callaghan JJ, Liu SS, Goetz DD, Vittetoe DA, Sullivan PM, Johnston RC (2005) The John Insall Award: unicompartmental knee replacement. A minimum twenty-one-year follow-up, end-result study. Clin Orthop Relat Res 440:27–37 Servien E, Fary C, Lustig S, Demey G, Saffarini M, Chomel S, Neyret P (2011) Tibial component rotation assessment using CT scan in medial and lateral unicompartmental knee arthroplasty. Orthop Traumatol Surg Res 97(3):272–275 SooHoo NF, Sharifi H, Kominski G, Lieberman JR (2006) Costeffectiveness analysis of unicompartmental knee arthroplasty as an alternative to total knee arthroplasty for unicompartmental osteoarthritis. J Bone Joint Surg Am 88(9):1975–1982 Stronach BM, Pelt CE, Erickson J, Peters CL (2013) Patientspecific total knee arthroplasty required frequent surgeon-directed changes. Clin Orthop Relat Res 471(1):169–174 Swienckowski JJ, Pennington DW (2004) Unicompartmental knee arthroplasty in patients sixty years of age or younger. J Bone Joint Surg Am 86-A Suppl 1(Pt 2):131–142 Valenzuela GA, Jacobson NA, Geist DJ, Valenzuela RG, Teitge RA (2013) Implant and limb alignment outcomes for conventional and navigated unicompartmental knee arthroplasty. J Arthroplast 28(3):463–468 Weber P, Crispin A, Schmidutz F, Utzschneider S, Pietschmann MF, Jansson V, Mu¨ller PE (2013) Improved accuracy in computer-assisted unicondylar knee arthroplasty: a meta-analysis. Knee Surg Sports Traumatol Arthrosc 21(11):2453–2461
