CT validation of intraoperative imageless navigation (Naviswiss) for component positioning accuracy in primary total hip arthroplasty in supine patient position: a prospective observational cohort study in a single-surgeon practice.

Abstract

Background: The aim of this study was to report on the validity of the Naviswiss handheld image-free navigation device for accurate intraoperative measurement of THA component positioning, in comparison with the three-dimensional (3D) reconstruction of computed tomography (CT) images as the gold standard.

Methods: A series of patients presenting to a single-surgeon clinic with end-stage hip osteoarthritis received primary hip arthroplasty with the anterolateral muscle-sparing surgical approach in the supine position. Imageless navigation was applied during the procedure with bone-mounted trackers applied to the greater trochanter and ASIS. Patients underwent routine CT scans before and after surgery and these were analyzed by using three-dimensional reconstruction to generate cup orientation, offset and leg length changes, which were compared to the intraoperative measurements provided by the navigation system. Estimates of agreement between the intraoperative and image-derived measurements were assessed with and without correction for bias and declared cases with potential measurement issues.

Results: The mean difference between intraoperative and postoperative CT measurements was within 2° for angular measurements and 2 mm for leg length. Absolute differences for the two indices were between 5° and 4 mm. Mean bias was 1.9°-3.6° underestimation for cup orientation and up to 2 mm overestimation for leg length change, but absolute thresholds of 10° and 10 mm were not exceeded by 95% limits of agreement (LOA), especially after correction for bias. Four cases (12%) were declared intraoperatively for issues with fixation on the greater trochanter. Inclusion of these cases generated acceptable accuracy overall and their omission failed to improve between-case variability in accuracy or LOA for both offset and leg length.

Conclusions: The accuracy of the Naviswiss system applied during primary THA in a supine position and anterolateral surgical approach falls within clinically acceptable recommendations for acetabular cup placement, femoral offset, and length. With refinements to surgical technique to adapt to the navigation hardware, the system could be further improved with regression-based bias correction.

Trial registration: Registered with the Australian New Zealand Clinical Trials Registry (ACTRN12618000317291).