Contribution of a new augmented reality system in total knee arthroplasty implantation. An in vivo, prospective, non-comparative study of 60 cases.

INTRODUCTION: Achieving planned alignment is one of the keys to implant survival and patient satisfaction in total knee arthroplasty (TKA). New technologies have been developed, such as robotics, patient-specific instruments (PSI), and augmented reality (AR). This study evaluated the contribution of AR through the Knee+™ device (PIXEE Medical Company, Besançon, France). This TKA navigation device uses AR to orient the distal femoral and proximal tibial cutting guides in the frontal and sagittal planes. Measurements are independent for the tibia and femur. The device requires no robot, no engineer, and no additional preoperative three-dimensional imaging. There are several papers dedicated to the Knee+™ device; this study is the largest prospective study to date and is independent of the device's promoter, to determine: (1) the accuracy of the Knee+™ device, (2) the existence of a learning effect on operative time and on accuracy, (3) the risk of using the device by reporting the intra-operative events and adverse events occurring within 3 months postoperatively.

HYPOTHESIS: Knee+™ admits an error of less than 1.5 ° between the value announced by the device and post-operative radiographic measurements.

MATERIAL AND METHODS: This was a prospective, monocentric study involving 60 patients undergoing TKA with the device under study; mechanical alignment was planned. The accuracy of the procedure was analyzed during the intraoperative cutting inspection stage and then by measuring the Lateral Distal Femoral Angle (LDFA), Medial Proximal Tibial Angle (MPTA), and tibial slope on postoperative radiographs.

RESULTS: There were no intraoperative adverse events and no revision surgery during the 3-month follow-up period. The mean error of the device was 1.23 ° ± 0.94 ° for the LDFA, 1.22 ° ± 1.05 ° for MPTA, and 2.15 ° ± 1.43 ° for the tibial slope. For the femur, no patient had an angle value with an error >3°; for the tibia, one patient (1.7%) had an angle value with an error >3° in the frontal plane, and twelve patients (21%) had an angle value with an error >3° in the sagittal plane. The learning curve showed no effect on accuracy, but operating time was reduced by 20 min after the tenth prosthesis was fitted with the device.

CONCLUSION: This study showed that the Knee+™ device is a safe, reliable, and reproducible AR tool. It has a rapid learning curve for TKA placement. It can be considered as an alternative or an aid to conventional instrumentation in TKA.

LEVEL OF EVIDENCE: IV; prospective non-comparative study.