CORR Insights®: Does Robotic Milling For Stem Implantation in Cementless THA Result in Improved Outcomes Scores or Survivorship Compared with Hand Rasping? Results of a Randomized Trial at 10 Years

Abstract

Where Are We Now? Implant fixation and restoration of a patient’s hip biomechanics are essential for a successful cementless THA. Although these objectives are frequently achieved with THA, complications such as implant loosening and dislocation resulting from component malposition persist namely because of human error or the use of manual instruments that are inconsistent in bone preparation and implant positioning [6]. Robotic surgery, introduced to orthopaedics more than two decades ago, aimed to reduce human error by rendering preoperative planning more precise, improving implant sizing and bone preparation to obtain implant-host bone contact, and more-accurately align the implants so as to better restore the biomechanics of the hip [3]. Enthusiasm for robotic surgery waned following unacceptably high intraoperative complication rates, the lack of clinical superiority, and the substantial capital expenditure [2, 9]. But recently, with the refinement of older technologies and the introduction of new technologies, the interest in the use of robotic devices in THA has re-emerged [11]. The ways in which we use robotics in health care, and specifically in primary THA, will continue to evolve [6]. Various clinical studies have demonstrated that the presently available robotic systems decrease variability and increase precision of component positioning and alignment [2]. In the current study, Nakamura and colleagues [5] reported their minimum 5-year followup of their previously published prospective, randomized study of a second generation, fully active robotic system, called ROBODOC (Integrated Surgical Systems, Davis, CA). The authors compared the results of the robotic system, which is used to mill the femoral canal, to hand rasps for femoral stem implantation during THA. Although they found no difference in the clinical outcomes or the rate of implant osseointegration, the authors did observe that the robotic-milling group had less variance in limb-length inequality and less stress shielding of the proximal femur. Where Do We Need To Go? Although the study by Nakamura and colleagues [5] addressed the mid-term utility of a fully active robotic system to prepare the femur during THA, there are still gaps in our knowledge that need to be understood prior to the routine adoption of robotics in primary THA. These gaps include determining the efficacy of the different types of robotic technologies, the efficacy of robots with various types of cementless implants, the long-term benefits and the cost-effectiveness of the technology, and how robotic technology compares to other technologies used to enhance THA surgery. Presently, there are two main types of robotic technology, fully active and semiactive, or surgeon assisted. The current study only looks at one type of robotic system (fully active) manufactured by a specific company and uses this system to implant one type of cementless femoral stem. Therefore, the results are most likely not generalizable for all robotic systems and/or all implants. In addition, since the fully active robotic system used in this study cannot be used to prepare the acetabulum, this study only looks at the outcomes of robotic surgery on the femoral component and does not take into account the equally important role of the acetabular component on hip biomechanics and outcomes [1]. Likewise, there are also concerns that semiactive systems may not improve combined anteversion, as they only involve acetabular reaming [10]. Further studies are required to clarify of each of these variables. Other technologies, such as imageless navigation have demonstrated the ability to achieve improved accuracy of implant positioning in THA, as well as decrease the number of outliers [4, 8]. In a randomized study by Lass and colleagues [4], the authors did not find any outliers outside of the Lewinnek’s safe zone for inclination compared to 8% in the conventional group, and fewer outliers with respect to anteversion. Ther