Jess H Lonner1,2, Ari D Seidenstein3,4, Michael A Charters5, W Trevor North5, Nathan L Cafferky6, Sridhar M Durbhakula7, Atul F Kamath8. 1. Rothman Orthopaedic Institute, 925 Chestnut St, Philadelphia, PA, 19107, USA. Jess.lonner@rothmanortho.com. 2. Sidney Kimmel Medical College, Thomas Jefferson University, Philadelphia, PA, 19107, USA. Jess.lonner@rothmanortho.com. 3. Rothman Orthopaedics, Montvale, NJ, 07645, USA. 4. Hackensack Meridian School of Medicine, Nutley, NJ, 07110, USA. 5. Henry Ford Health System, Detroit, MI, 48202, USA. 6. Vail Summit Orthopaedic Foundation, Vail, CO, 81657, USA. 7. Washington Joint Institute, Bethesda, MD, 20817, USA. 8. Orthopaedic and Rheumatologic Institute, Cleveland Clinic, Cleveland, OH, 44113, USA.
Abstract
PURPOSE: Alignment errors in medial unicompartmental knee arthroplasty (UKA) predispose to premature implant loosening and polyethylene wear. The purpose of this study was to determine whether a novel CT-free robotic surgical assistant improves the accuracy and reproducibility of bone resections in UKA compared to conventional manual instrumentation. METHODS: Sixty matched cadaveric limbs received medial UKA with either the ROSA® Partial Knee System or conventional instrumentation. Fifteen board-certified orthopaedic surgeons with no prior experience with this robotic application performed the procedures with the same implant system. Bone resection angles in the coronal, sagittal and transverse planes were determined using optical navigation while resection depth was obtained using calliper measurements. Group comparison was performed using Student's t test (mean absolute error), F test (variance) and Fisher's exact test (% within a value), with significance at p < 0.05. RESULTS: Compared to conventional instrumentation, the accuracy of bone resections with CT-free robotic assistance was significantly improved for all bone resection parameters (p < 0.05), other than distal femoral resection depth, which did not differ significantly. Moreover, the variance was significantly lower (i.e. fewer chances of outliers) for five of seven parameters in the robotic group (p < 0.05). All values in the robotic group had a higher percentage of cases within 2° and 3° of the intraoperative plan. No re-cuts of the proximal tibia were required in the robotic group compared with 40% of cases in the conventional group. CONCLUSION: The ROSA® Partial Knee System was significantly more accurate, with fewer outliers, compared to conventional instrumentation. The data reported in our current study are comparable to other semiautonomous robotic devices and support the use of this robotic technology for medial UKA. LEVEL OF EVIDENCE: Cadaveric study, Level V.
PURPOSE: Alignment errors in medial unicompartmental knee arthroplasty (UKA) predispose to premature implant loosening and polyethylene wear. The purpose of this study was to determine whether a novel CT-free robotic surgical assistant improves the accuracy and reproducibility of bone resections in UKA compared to conventional manual instrumentation. METHODS: Sixty matched cadaveric limbs received medial UKA with either the ROSA® Partial Knee System or conventional instrumentation. Fifteen board-certified orthopaedic surgeons with no prior experience with this robotic application performed the procedures with the same implant system. Bone resection angles in the coronal, sagittal and transverse planes were determined using optical navigation while resection depth was obtained using calliper measurements. Group comparison was performed using Student's t test (mean absolute error), F test (variance) and Fisher's exact test (% within a value), with significance at p < 0.05. RESULTS: Compared to conventional instrumentation, the accuracy of bone resections with CT-free robotic assistance was significantly improved for all bone resection parameters (p < 0.05), other than distal femoral resection depth, which did not differ significantly. Moreover, the variance was significantly lower (i.e. fewer chances of outliers) for five of seven parameters in the robotic group (p < 0.05). All values in the robotic group had a higher percentage of cases within 2° and 3° of the intraoperative plan. No re-cuts of the proximal tibia were required in the robotic group compared with 40% of cases in the conventional group. CONCLUSION: The ROSA® Partial Knee System was significantly more accurate, with fewer outliers, compared to conventional instrumentation. The data reported in our current study are comparable to other semiautonomous robotic devices and support the use of this robotic technology for medial UKA. LEVEL OF EVIDENCE: Cadaveric study, Level V.