Abstract
INTRODUCTION
Poor acetabular cup orientation in total hip arthroplasty (THA) can cause dislocation and impingement, and lead to osteolysis (Little et al., 2009) and inflammatory soft tissue reactions (Haan et al., 2008). While the intrinsic accuracy of cup positioning in navigation is reported as low as 1° (Parratte et al., 2009), a large anterior pelvic tilt may lead to an offset of the same magnitude in the final cup anteversion (Wolf et al., 2005). The objectives of this study are to demonstrate feasibility of a new, non-invasive radiographic tool for accurate preoperative determination of a patient's specific pelvis angle, and intraoperative and postoperative assessment of the acetabular cup orientation with respect to boney landmarks.
METHODS
The methodology stitches multiple radiographic views around the pelvis using a multi-planar radiography setup (Amiri et al., 2011) and reconstructs the reference boney landmarks and the acetabular cup in three dimensions using previously developed algorithms and software (Amiri et al., 2012). To validate the methodology, a Sawbone model of the pelvis and femur was implanted with a standard cementless metal-on-polyethylene THA, and was tracked and digitized by an Optotrak motion tracking system. Five radiographic views were acquired at the pubic tubercle (PT) and anterior-superior iliac spine (ASIS) levels (Views 1 to 5 in Fig 1). Imaging and analysis were repeated 10 times. Custom software (Joint 3D) was used to reconstruct the right and left PT and ASIS by fitting spheres to the corresponding pairs of images (Fig 1). The three-dimensional pose of the acetabular cup was reconstructed in the software by solving a back-projection equation of the elliptical shadow of the cup opening. Accuracies were measured as mean differences from the digitized references. A sample of the reconstructed graphical output for the anterior pelvic plane (APP) and the cup, in comparison to the digitized reference, is shown in Fig 2. Repeatability was estimated as standard deviation of the measures for the reconstructed locations of the boney landmarks and the APP (known as a standard reference plane for cup placement).
RESULTS
Accuracy for the pelvis pose angles was <1.6°, with SD <0.8° (Fig 3). Pelvic tilt was the most accurate with accuracy of 0.1° and SD=0.4°. For the acetabular cup, accuracy was 2.5° or better, with SD <0.2°. Accuracies in the cup operative anteversion and inclination were 2.4° and 0.6°, with SD=0.4° and 0.9°, respectively.
DISCUSSION
The measured accuracies were within an acceptable range, according to previous studies that recommended a 5° cut-off error for acetabular anteversion. The method shows accuracy and radiation dose advantages over current radiographic, fluoroscopic and computed tomography methods. These results suggest that the proposed method is feasible for assessing cup placement with reference to the functional and anatomical references.
CONCLUSION
Use of this technique could improve acetabular cup placement and reduce the incidence of instability, wear and loosening, by providing tools to incorporate the individual's pelvic pose in preoperative planning of the surgery, and by serving as an accurate and reliable tool for intraoperative and postoperative assessment of the acetabular cup position.
