Minimal-Dose Computed Tomography for Measurement of Cup Orientation in Total Hip Arthroplasty

Introduction: Accurate assessment of cup orientation is important for evaluation of outcome after total hip arthroplasty (THA). The orientation is clinically described as a combination of inclination angle and version angle. Measurements of these angles based on plane radiographs are inaccurate, because no solution exists for perfect positioning of the pelvis, as Liaw(1) stated. Computed tomography (CT) provides image data for accurate measurements of inclination angle and version angle of the cup, but requires greater radiation exposure than radiographs. The purpose of this study was to determine a CT scanning protocol with minimum radiation dose while maintaining image quality for measurements of cup orientation. Materials and Methods: Direct measurements on a pelvic model: A spherical acetabular component (Trilogy, Zimmer, Warsaw, USA), a polyethylene liner and a femoral component (Versys, Zimmer) were fixed to the right acetabulum of a human cadaver pelvis using ethyl acetate resin. We measured inclination angle and version angle of the cup in this pelvic model as follows. The plane passing through the bilateral anterior superior iliac spines and pubic tubercles was defined as an anterior pelvic plane (APP). The line passing the bilateral anterior superior iliac spines was defined as a transverse line. We took digital photographs of the model from two directions (anterior to posterior, and caudal to cranial) using the APP and the transverse line as references. The imaging data were preserved in a computer, and inclination angle and anatomical version angle were measured with use of computer-assisted design (CAD) software. Measurements were performed ten times, and the mean values were defined as inclination and version angles in the direct measurements. CT of the phantom: A simulated phantom was constructed fixing the pelvic model in a case made of styrene foam, and filling it with water. The CT scanning of this phantom was performed in six protocols. Collimations (0.5 mm x 64), helical pitch (0.828), gantry rotation (0.5sec), scan length (220 mm), and tube voltage (120 kV) were the same in each protocol. The tube current was varied for each protocol as 300, 100, 50, 30, 20, and 10mA, respectively. The effective doses of these conditions were 11.1, 3.7, 1.9, 1.1, 0.7, 0.4mSv, respectively. DICOM data of the CT scans were imported to a computer, and analyzed with use of template software. We defined the images as good if they showed clearly all of the following landmarks, which are necessary for measurements of cup orientation. The mandatory landmarks were bilateral anterior superior iliac spines, pubic symphysis and edge of the cup. We compared the values between direct measurements and CT. Among the protocols giving good images, that with the minimum radiation dose was defined as the minimal-dose CT. CT of patients: Minimal-dose CT scanning was performed for four patients after total hip arthroplasty in order to confirm the usefulness of this protocol. There were three females, and one male. Mean age was 60 years (range, 53-73). The mean of body mass-index was 22.9 (range, 19.128.0). The same acetabular components (Trilogy) were used in all hips. The study was performed after institutional review board approval and informed patient consent. Results: In direct measurements of the phantom, inclination and version angles were 55.2 and 21.8 degrees, respectively. In the CT of the phantom, protocols with tube currents of 20 and 10mA did not give good images (Table 1). Four protocols with tube currents of 300, 100, 50 and 30mA did give good images. Inclination and version angles were measured on CT data. The differences of measured angles between direct and CT measurements were very small in all of those four protocols (Table 2). The effective dose was smallest in the 30mA protocol among the four good image protocols. Therefore, the protocol with 30mA (1.1mSv) was defined as the minimal-dose CT. Clinically, accurate measurements of inclination and version angles were possible by minimal-dose CT in all of four patients. Discussion: Conventionally, only radiographs of the pelvis had been used for measurements of cup orientation after THA. Tannast (2) described that measurements of cup orientation based on postoperative AP radiographs are highly inaccurate, and that those based on CT used with APP as a reference plane are more accurate. On the other hand, they stated that their method for measuring cup orientation postoperatively is suitable for research purposes but not for routine clinical use because it requires a postoperative CT scan of the pelvis with a radiation dose of 10mSv. This dose is fourteen times that of prevalent pelvis radiographs (0.7mSv). Henckel (4) had looked at ways of minimizing radiation dose while maintaining image quality for pre-operative planning and postoperative assessment in knee replacement. The radiation dose of their CT scan protocol for total knee arthroplasty was between 0.53mSv and 0.84mSv. Our study on the phantom showed that measurements of cup orientation are possible with a CT scan protocol of 1.1mSv, which is less than two times that of a radiograph. One of the limitations of this study was the very small number of patients who had a CT scan. Further studies are needed to adjust the CT scanning protocol according to physical status of each patient in clinical use of the minimal-dose CT. Conclusions Measurement of cup orientation after total hip arthroplasty is possible with a minimal-dose CT with radiation dose of 1.1mSv. References: (1) Liaw et al, Clin. Orthop. Relat. Res. Oct; 451:134-9 2006 (2) Tannast et al, Clin. Orthop. Relat. Res. Jul ; 436:144-5