Unicompartmental arthroplasty of the knee (UKA) is technically challenging because the prosthetic devices must function in concert with a mostly normal joint. Malalignment is common, leading to patient dissatisfaction and early failures. However, UKA remains attractive as a temporizing treatment in early disease. Until now, resurfacing UKAs were performed with free-hand techniques. This study is only the second report investigating the use of a tactile guidance system (TGS—essentially, a robotically assisted surgery) for the performance of UKA.

Methods. The first 20 patients who underwent resurfacing using a Mako Surgical Inc. TGS system by a single surgeon were studied. Surgical goals were to place the components to replicate closely the patient’s native bony architecture. The surgical plan was completed on a workstation, and then executed with the TGS system through a mini-arthrotomy. Stelkast, Inc resurfacing components were implanted with methymethacrylate. Intraoperative measurements of component position were obtained. Pre- and postoperative radiographs were also measured for alignment correction, change in angulation of the joint line relative to the femoral and tibial anatomic axes, femoral component alignment relative to the femoral anatomic axis, and change in tibial slope.

Results. All cases could be completed as planned. None were converted to a full arthrotomy. None required conversion to a different implant. There were no failures of the TGS, associated navigation, or the CAT-scan based preoperative plan.

Intraoperative measurements showed an average femoral component position of 0.89+3.36 degrees of varus relative to the mechanical axis, with 62.5% being varus and 37.5% being valgus. The average femoral component flexion was 11.1+2.11 degrees, with no outliers (less than 5 degrees; greater than 15 degrees). The tibial component position was 4.60+1.76 degrees of varus, with all components in varus as desired. There was an average of 5.00+2.37 degrees of slope, with 25% outliers (less than 3 or greater than 7 degrees).

Postoperative measurements showed an overall limb alignment correction of 4.29+2.60 degrees, femoral joint line change of only 0.43+0.49 degrees, and an overall component alignment relative to the anatomic axis of 4.54+3.77 degrees of valgus. On the tibial side, the joint line varus was corrected by 3.00+2.04 degrees and the slope was changed by 4.29+3.24 degrees, including 19% outliers (less than 3 degrees, more than 7 degrees). However, 33% of the outliers were outliers preoperatively as well. Interestingly, the bone level after resection on the tibial side averaged 5.36+3.00 degrees of varus, suggesting that component placement must be carefully watched.

Discussion. TGS seems to be extremely accurate and precise in recreating individual patient anatomy. This also applies to cases in which the patient anatomy dictates placement of components in so-called “outlier” positions. It is unknown whether these “outlier” positions really translate into poorer outcomes. Impressively, there were no failures to execute the intended surgical plan and no failures of the TGS system. Future research will attempt to correlate component placement in native anatomical positions with functional outcomes and failures, as well as cost-effectiveness of the system.