Abstract
The success of total knee replacement surgery depends critically on proper limb alignment and implant position. Even with contemporary mechanical alignment instrumentation, errors in limb alignment and implant position do occur. To improve upon the accuracy and biomechanical efficacy of conventional surgical instrumentation while limiting the need for substantial pre-operative planning, a non-image-based computer-aided navigation system was developed for total knee replacement surgery. Clinical studies have demonstrated that use of this system, OrthoPilot® (Aesculap AG, Tuttlingen, Germany), for knee replacement surgery can lead to improved limb alignment and implant position.
In this study we investigated the repeatability and sensitivity of the OrthoPilot® computer-aided navigation system for total knee replacement surgery. To assess repeatability, total knee replacement surgeries were simulated on an idealized test bench using identical input parameters and the variation in output measurements was measured. To assess sensitivity, the effect of moderate movement of position sensors on system-level accuracy was measured. The results indicate that (1) the system functions in a highly repeatable manner if it is supplied with repeatable inputs; and (2) unintentional relative movement of position sensors during surgery can substantially affect accuracy of the system outputs.
Because computer-aided navigation systems are powerful tools for orthopaedic surgery, it is important to recognize that their accuracy and precision are highly dependent on pre-operative and intra-operative registration techniques. Like all instrumentation systems, their use is associated with a learning curve, even in the hands of experienced orthopaedic surgeons. The results of this study demonstrate that the OrthoPilot® in an inherently precise instrument that is sensitive to variations in surgical technique. It is critical that the users of these systems (i.e. surgeons) be aware of system sensitivities and pay careful attention to operative techniques required by the system.
The abstracts were prepared by Nico Verdonschot. Correspondence should be addressed to him at Orthopaedic Research Laboratory, University Medical Centre, PO Box 9101, 6500 HB Nijmegen, The Netherlands.
