Accuracy and Reliability of Software Navigation for Acetabular Component Placement in THA: An In Vitro Validation Study

Abstract

Background and Objectives: Intraoperative fluoroscopy can be used to increase the accuracy of the acetabular component positioning during total hip arthroplasty. However, given the three-dimensional nature of cup positioning, it can be difficult to accurately assess inclination and anteversion angles based on two-dimensional imaging. The purpose of this study is to validate a novel method for calculating the 3D orientation of the acetabular cup from 2D fluoroscopic imaging. Materials and Methods: An acetabular cup was implanted into a radio-opaque pelvis model in nine positions sequentially, and the inclination and anteversion angles were collected in each position using two methods: (1) a coordinate measurement machine (CMM) was used to establish a digitalized anatomical coordinate frame based on pelvic landmarks of the cadaveric specimen, and the 3D position of the cup was then expressed with respect to the anatomical planes; (2) AP radiographic images were collected, and a mathematical formula was utilized to calculate the 3D inclination and anteversion based on the 2D images. The results of each method were compared, and interrater and intrarater reliably of the 2D method were calculated. Results: Interrater reliability was excellent, with an interclass correlation coefficient (ICC) of 0.988 (95% CI 0.975-0.994) for anteversion and 0.997 (95% CI 0.991-0.999) for inclination, as was intrarater reliability, with an ICC of 0.995 (95% CI 0.985-0.998) for anteversion and 0.998 (95% CI 0.994-0.999) for inclination. Intermethod accuracy was excellent with an ICC of 0.986 (95% CI: 0.972-0.993) for anteversion and 0.993 (95% CI: 0.989-0.995) for inclination. The Bland-Altman limit of agreement, which represents the error between the 2D and 3D methods, was found to range between 2 to 5 degrees. Conclusions: This data validates the proposed methodology to calculate 3D anteversion and inclination angles based on 2D fluoroscopic images to within five degrees. This method can be utilized to improve acetabular component placing intraoperatively and to check component placement postoperatively.

Keywords: arthroplasty; hip; replacement.

Figure 1

Experimental set-up including pelvis model secured with acetabular component in place, fluoroscopy unit, and the coordinate measure machine Romer Arm.

Figure 2

Representation of the 3D data collected using the CMM. The dashed lines represent the anatomic coordinate frame with the anterior axis (red) perpendicular to the plane of the X-ray, the lateral axis (pink) parallel to the inter-teardrop line (full green line connecting the yellow and purple points), and the inferior axis (green) mutually perpendicular to the other two axes. The red circles represent the points collected along the acetabular rim, and the blue ellipse describes the contour of the acetabular rim, calculated using a best fit circle in 3D. The light blue line represents the normal axis to the plane of the acetabular rim, and the red arc between the inferior axis (dashed green line) and the light blue line is the anatomic inclination angle described by Murray et al. The full purple line is the projection of the normal axis onto the transverse plane (plane described by the red dashed line and the pink dashed line). The yellow arc between the lateral axis (dashed pink line) and the purple line is the anatomic anteversion angle described by Murray et al.

Figure 3

Calculation of the anatomic anteversion and inclination based on the ellipse described by the acetabular component rim. The horizontal axis is aligned with the medial/lateral axis of the pelvis using the teardrop line.

Figure 4

Fluoroscopic images used for 3D angle calculations. (A) shows the full-pelvis AP view with the teardrop line passing through the two dark circles created by the previously described steel beads, and the brim line. (B) Shows the acetabular cup view with the brim line and the ellipse described by the acetabular cup rim. The teardrop-brim line angle is used to identify the medial–lateral axis on the acetabular cup view.

Figure 5

Bland–Altman limits of agreement for the interrater analysis for the cup only images.

Figure 6

Bland–Altman limits of agreement for the intrarater analysis for the cup only images.

Figure 7

Bland–Altman limits of agreement for the intermethod analysis comparing the software analysis to the CMM analysis for the cup only images.