The Journal of Arthroplasty Vol. 27 No.

1 2012

Brief Communication

The Value of Patient-Matched Instrumentation in Total Knee Arthroplasty

John W. Noble Jr., MD,* Chris A. Moore, MS,y and Ning Liu, MSy

Abstract: The purpose of the current prospective, randomized study was to compare the value of a new mechanically aligned patient-matched instrument system for total knee arthroplasty (TKA) (Visionaire; Smith & Nephew, Inc, Memphis, Tenn) (VIS) to that of standard TKA surgical instrumentation (STD). Twenty-nine primary TKA patients were enrolled and completed surgery (15 VIS and 14 STD). Postoperatively, mechanical alignment was significantly closer to neutral zero in the VIS group (1.7 vs 2.8; P = .03). Furthermore, the VIS group demonstrated significant reductions in duration of hospital stay, operative time, incision length, and number of used instrument trays (P b .05). Although additional research is underway to confirm these preliminary results, this evidence suggests that patient-matched instrumentation may support repeatable improvements in surgical accuracy and hospital efficiency. Keywords: patient-matched, custom, alignment, cost, value, function. 2012 Elsevier Inc. All rights reserved.

Total knee arthroplasty (TKA) remains the treatment standard for advanced knee arthritis, demonstrating repeatable pain relief, restoration of basic function, and acceptable clinical longevity. However, there is an increased focus in the orthopedic community toward supporting improved patient satisfaction and postoperative knee function beyond minimal daily requirements [1,2]. In addition, as the annual volume of TKA procedures continues to grow, mounting cost pressure may impose the need for increased efficiency, including reduced operative time, duration of hospital stay, and equipment inventory [1,3]. Patient-matched instrumentation is a relatively new arthroplasty technology that may help the community meet these emerging needs. Here, data from preoperative medical imaging enable the creation of a 3dimensional model of the distal femur and proximal tibia [4]. After visualization, proprietary software is used to virtually map all bone resections and to accurately size
From the *Imperial Calcasieu Medical Group, LLP Lake Charles, Louisiana; and ySmith & Nephew, Inc, Memphis, Tennessee. Submitted May 19, 2011; accepted July 18, 2011. The Conflict of Interest statement associated with this article can be found at doi:10.1016/j.arth.2011.07.006. Reprint requests: John W. Noble, Jr, MD, Center for Orthopaedics, Imperial Calcasieu Medical Group, LLP, 1747 Imperial Boulevard, Lake Charles, LA 70605. 2012 Elsevier Inc. All rights reserved. 0883-5403/2701-0025$36.00/0 doi:10.1016/j.arth.2011.07.006

and position the knee implant. Disposable cutting blocks are then manufactured to fit the patient's unique articular deformity. This methodology is thought to minimize tissue loss and optimize implant positioning while effectively reducing instrument inventory in the operating room [5]. Although all patient-matched systems theoretically share this same purpose, there are considerable differences in their respective design methodologies. One method is to restore the patient's prearthritic anatomical leg alignment by filling all articular defects [1,2]. Although preliminary operative results have been promising [1,2], there has been 1 report citing alignment error in a series of 4 patients [4]. Alternative patientmatched systems are designed to restore a neutral postoperative mechanical axis. However, with the exception of 1 surgeon's feasibility experience [5], there are no published, peer-reviewed clinical data supporting the efficacy of this technology. The purpose of the current prospective, randomized clinical study was to compare the accuracy and efficiency of a new mechanically aligned patient-matched system (Visionaire; Smith & Nephew, Inc, Memphis, Tenn) to that of standard TKA surgical instrumentation.

Materials and Methods

The current study was initiated following institutional review board approval and registration on ClinicalTrials. gov (Identifier: NCT01107769). Thirty-eight primary

153

154 The Journal of Arthroplasty Vol. 27 No. 1 January 2012

TKA patients were randomly assigned to 1 of 2 groups (1:1), using either Visionaire patient-matched cutting blocks (VIS) or standard instrumentation (STD). Nineteen VIS patients completed surgery, whereas 4 of 19 STD patients were preoperatively withdrawn from the study before operation. Causes of withdrawal included patient concern over undergoing anesthesia (1), withdrawn consent (1), failure to schedule operation (1), and myocardial infarction (1). Furthermore, 5 patients were excluded from the study because of protocol deviation (body mass index [BMI], N40 kg/m2). Therefore, the final study sample was 29 patients (15 VIS and 14 STD). In the VIS group, there were 8 men and 7 women aged 65.4 years (range, 57-76 years) with an average BMI of 29.8 (range, 22.1-35). In the STD group, there were 6 men and 8 women aged 68 years (range, 56-80 years) with an average BMI of 31.2 (range, 21-39.8). The Legion Total Knee System (Smith & Nephew, Inc) was implanted in all subjects. A medial parapatellar surgical approach with initial distal femoral cut was used for all cases. All cases had cruciate retaining femoral components. Dished articular inserts were used depending on the perceived status of the posterior cruciate ligament. All procedures were performed at CHRISTUS St Patrick's Hospital (Lake Charles, La). The only treatment difference between groups was the cutting block assignment. Intraoperative alignment and placement of the custom cutting blocks were confirmed using well-described anatomical reference points. Accuracy of these blocks was presumed based on this information as well as fit. Regarding the preoperative procedure for all VIS patients, magnetic resonance imaging (MRI) scans were first performed to construct the requisite 3dimensional model of the knee. Magnetic resonance imaging supports full visualization of the femoral and tibial cartilage and, in contrast to alternative computed tomographic techniques, does not require undue radiation exposure to the patient. Next, full-length anterior/ posterior radiographs were performed to determine each patient's current and planned mechanical axis. Femoral varus/valgus alignment was established by plotting a line between the femoral head and sulcus, then measuring against the femoral shaft axis. Tibial alignment was determined by a perpendicular line transecting the tibial plateau, passing through the medial-distal tibial shaft. Preoperative imaging data were used to preoperatively plan all tissue resections and optimal implant orientation. Furthermore, all 3 rotational axes provided by the MRI image enabled the cutting block to be designed with optimal rotational and mechanical alignment, reducing risk of operative malalignment. Each detailed operative plan was reviewed and approved by the surgeon (JN) before the manufacture of the VIS cutting block. An intraoperative image of the VIS cutting blocks can be found in Fig. 1. Surgical accuracy was assessed by comparing preoperative and postoperative mechanical alignment between groups. Efficiency was measured by capturing duration of hospital stay, operative time, incision length, number of instrument trays used, tray set-up time, and blood loss for each case. All statistical analyses were performed using SAS version 9.1 (SAS Institute Inc, Cary, NC). Pearson 2 test, t test, and Fisher exact test were used to compare baseline characteristics between groups. All outcome measures were assessed using a WilcoxonMann-Whitney U test. The significance level for all tests was 0.05.

Results
There were no significant differences in age, BMI, or preoperative mechanical alignment between groups (P N .05). Regarding surgical accuracy, postoperative

Fig. 1. Intraoperative image of a femoral (A) and a tibial (B) patient-matched cutting block.

Value of Patient-Matched Instrumentation in TKA  Noble et al

155

mechanical alignment in the VIS group was found to be statistically closer to neutral zero (1.7; range, 0-6 vs 2.8; range, 0-5; P = .03). Further statistically significant reductions were observed in the VIS group for duration of hospital stay (59.2 hours; range, 48-72 vs 66.9 hours; range, 48-72; P = .043), operative time (121.4 minutes; range, 105-165 vs 128.1 minutes; range, 115-157; P = .048), incision length (136 mm; range, 120-160 vs 151.8 mm; range, 120-175; P = .014), and the number of used instrument trays (4.3; range, 3-8 vs 7.5; range, 4-10; P b .0001). There was no significant difference between groups for intraoperative blood loss (VIS: 71 mL; range, 40-200 vs STD: 62.5 mL; range, 50-100; P = .395). Regarding blood loss, an outlier value of 200-mL blood loss was identified. This patient had a medical history of coronary artery disease and severe peripheral vascular disease with extensive calcification of the popliteal artery. Because of this, a tourniquet was not used during surgery. One posterior-cruciate ligament release was required in each group, with no incidence of collateral ligament release. At the time of surgery, all patientmatched cutting blocks demonstrated acceptable fit and stability. There were no instrument-related adverse events or complications.

Discussion
Patient-matched cutting blocks are instruments intended to aid the surgeon in correcting mechanical malalignment and to reduce hospital resource use during TKA. However, there are limited, peer-reviewed, published clinical data supporting these purported benefits. The current study is the first prospective, randomized report assessing the value of a mechanically aligned patient-matched system. The primary finding of this study was that the mean VIS mechanical alignment value of 1.7 was statistically closer to neutral zero than that observed for the STD group. These results compare favorably to a previously published feasibility study of an unspecified mechanically aligned system, which reported alignment ranging from 4 to 8 valgus [5]. There has been debate regarding the true effect of alignment on patient outcomes [1,2,4]. However, although TKA failure must be considered a multifactorial phenomenon, mechanical alignment exceeding 3 is still associated with increased failure risk [4,6]. Although postoperative alignment for all patients in the current study did fall in this range, it is possible that aligning the knee closer to neutral zero could support improved long-term outcomes. Additional follow-up is needed to confirm this hypothesis. In addition to improved mechanical alignment, the VIS group demonstrated significant reductions in the

number of instrument trays used, duration of hospital stay, and skin-to-skin time of operation. These findings are of particular interest in respect to the current health economic climate. Total knee arthroplasty has become one of the largest expenses in the Medicare budget and is a primary contributor to evergrowing health care cost pressures [3]. Patient-matched cutting blocks require less surgical instrumentation, which can effectively reduce hospital sterilization costs. Furthermore, any reduction in operative time and the duration of hospital stay may support a cumulative decrease in resource use. Although this study showed a statistically significant reduction in the number of trays used, the number of trays can certainly be reduced in the future to 1 or 2 as experience with this new technique grows. The adoption of new TKA technologies should be considered if repeatable improvements in overall hospital efficiency and return on investment are demonstrated [3]. The results of the current randomized study do support the value of patient-matched cutting blocks. However, although the potential benefits of this technology are certainly encouraging, this study is only representative of 1 surgeon's experience. A statistically powered, multicenter, randomized clinical study is currently underway to confirm these initial observations (ClinicalTrials.gov Identifier: NCT01084772).

Acknowledgment
The authors thank Stacy Leake-Gardner, RN, BSN, CCRP; DeAnne Winey-Ward, BS, CCRP, CCRC; and David Piper, BS, for their valuable contributions during the execution of this study.

References
1. Howell SM, Kuznik K, Hull ML, et al. Results of an initial experience with custom-fit positioning total knee arthroplasty in a series of 48 patients. Orthopedics 2008;31:857. 2. Spencer BA, Mont MA, McGrath MS, et al. Initial experience with custom-fit total knee replacement: intraoperative events and long-leg coronal alignment. Int Orthop 2009;33:1571. 3. Beringer DC, Patel JJ, Bozic KJ. An overview of economic issues in computer-assisted total joint arthroplasty. Clin Orthop Relat Res 2007;463:26. 4. Klatt BA, Goyal N, Austin MS, et al. Custom-fit total knee arthroplasty (OtisKnee) results in malalignment. J Arthroplasty 2008;23:26. 5. Lombardi Jr AV, Berend KR, Adams JB. Patient-specific approach in total knee arthroplasty. Orthopedics 2008;31:927. 6. Jeffery RS, Morris RW, Denham RA. Coronal alignment after total knee replacement. J Bone Joint Surg Br 1991;73:709.