A total of 344 consecutive THAs were performed for DDH patients between January 2017 and December 2018. All operations were performed by the same group of surgeons, with the posterolateral approach and “acetabular first” technique with the concept of combined anteversion. The procedure has been described in detail previously [4, 18]. Routine intraoperative fluoroscopy was used to verify the size and position of the final femoral broach to achieve “best-fit”.
In total, 211 hips were excluded from this study because of insufficient preoperative or postoperative CT data (103 hips), obvious flexion contracture of the hip (26 hips), Crowe type IV hips and hips with Wagner Cone or S-rom stems (47 hips), and more than 3° malalignment of the stem in the coronal and/or sagittal plane (21 hips). The remaining 133 hips (111 patients) were included in the study, including 87 women and 24 men with a mean age of 59.77 ± 11.22 (range, 29–85) years. There were 76 Crowe type I hips, 27 type II hips, and 30 type III hips. The single-wedge stem was used in 49 hips (Accolade, Stryker, Howmedica, Mahwah, NJ, USA), and the double-wedge stem was used in 84 hips (Secur-fit No. 47, Stryker, Howmedica, Mahwah, NJ, USA; Summit No. 37, DePuy, Warsaw, IN, USA).
CT evaluation was performed as we previously reported [19] by using the Hitachi Radix Turbo (Tokyo, Japan) (120 kVp, 200 mA, 5-mm collimation, 5 mm/s table speed, and 5-mm resolution index) device. By using axial sections passing from the anterior superior iliac spine to the tibial tubercle, patients underwent 1-mm-interval CT in the supine position with the hips and knees fully extended and the lower limbs as horizontal and parallel as possible. Pre- and postoperative CT data was stored in Digital Imaging and Communication in Medicine (DICOM) format. Femoral measurement was performed using the Radiant DICOM Viewer (version 4.6.9, 64-bit, Medixant Company, Poland).
The femoral neck height was measured as the vertical distance between the proximal base of the lesser trochanter and the medial edge of the femoral head-neck junction. The femoral neck osteotomy height was measured as the vertical distance between the proximal base of the lesser trochanter and the medial edge of the femoral neck osteotomy plane.
The anteversion of the anterior cortex at two levels of the lesser trochanter (Levels a and b) and the anteversion of the posterior cortex at five levels of the femoral neck (Levels b, c, d, e, and f) were measured via preoperative two-dimensional CT (Fig. 2). The levels were as follows: Level a, center of the lesser trochanter; Level b, proximal base of the lesser trochanter; Level c, 5 mm above Level b; Level d, 10 mm above Level b; Level e, 15 mm above Level b; and Level f, femoral head-neck junction (just below the head). First, the anteversion of every cortex was measured as the angle formed by the cortical line and the posterior condylar axis. The posterior condylar axis was defined as the line drawn along the largest femoral condyle on the CT slice [10]. Based on the principle of three-point fixation in the sagittal plane (Fig. 1), the predictive anteversion by the new method was then calculated as the average anteversion of the anterior cortex at two levels of the lesser trochanter and posterior cortex at five levels of the femoral neck. Thus, 10 predictive values of anteversion from 10 level combinations (Level ab, Level ac, Level ad, Level ae, Level af, Level bb, Level bc, Level bd, Level be, and Level bf) were obtained.
The native femoral anteversion at the femoral head-neck junction (Level f) was measured by the conventional method of Suh et al. [10], defined as the angle between the midcortical line and the posterior condylar axis, which was often used for the prediction of postoperative stem anteversion in previous studies [11]. Postoperative stem anteversion was measured as the angle formed by the stem neck major axis and the posterior condylar axis [20].
All CT measurements were performed twice over an interval of more than 2 weeks by two orthopedic surgeons.
Statistical analysis
Data were expressed as the mean ± standard deviation. All statistical analyses were performed using SPSS software for Windows (version 18.0; SPSS, Chicago, IL, USA). The intraclass correlation coefficients of interobserver and intraobserver reliabilities were calculated. Normal distribution was tested before any statistical analysis. The difference and correlation between preoperative predictive anteversion and postoperative stem anteversion were analyzed. A P value < 0.05 was considered to represent a significant difference. Correlation (r) was characterized as poor (0.00–0.20), fair (0.21–0.40), moderate (0.41–0.60), good (0.61–0.80), or excellent (0.81–1.00).