As per Altman nomogram, using a standardized difference of 0.5 with a power of 80% and a P-value of 0.05 the sample size required for this study was 128. We measured the patients who needed to undergo total hip replacement in the osteonecrosis and joint reconstruction ward of Xi'an Honghui Hospital from August 2015 to March 2016.
AP radiographic views of the pelvis were obtained in the same radiology unit based on a routine protocol at a standard distance of 1 m and centered 2 cm above the pubic symphysis. The patients were placed in supine position with their feet rotated internally 15° to maintain neutral hip position. The sizes of the lesser trochanters were similar and the anterior and posterior borders of the greater trochanters were superimposed on both sides [5]. The coccyx was in line with the pubic symphysis to prevent pelvic rotation in the transverse plane[6]. The distance between the sacrococcygeal joint and the pubic symphysis was 1–4 cm in men and 4–6 cm in women to exclude abnormal pelvic tilting in the sagittal plane[6, 7]. Exclusion criteria: 1. Unclear teardrops; 2. Pelvic rotation or tilt; 3. Hip flexion contracture before or after surgery, cannot be straightened; 4. Other diseases of the lower limbs lead to shortening of the limbs or knee disease resulting in knee joints varus, valgus or flexion contracture deformity.
The following measures are given to eligible patients after admission:
Use a medical tape to measure the distance from the left and right anterior superior iliac spine(ASIS) to the ipsilateral medial malleolus(MM).
Use the marker on AP radiograph of the bilateral hip joint to mark the following (Figs. 1 and 2).1. Bilateral tip of the greater trochanter(GT); 2. Teardrop inferior edge line (inter-teardrop line, IT); 3. Ischia tuberosity line (bi-ischial line, BI); 4. Vertex of the lesser trochanter(LT).
Made perpendicular lines from GT to IT, and draw lines from LT to IT and BI, and measured the length of each vertical segment. Calculated the difference between the left and right sides, and the data were divided into 4 groups, the difference between GT to IT formed GT-IT group, the difference between ASIS to MM formed ASIS-MM group, the difference between LT to IT formed LT-IT group, and the difference between LT to BI formed LT-BI group. The postoperative radiograph measurement ÷ magnification1 was the actual value, and the preoperative radiograph measurement ÷ magnification 2 was the actual value. The difference in the length of the bilateral vertical line segments can be considered as LLD.
Magnification1= | Measured mediolateral dimension of the acetabular component on radiograph |
True mediolateral dimension of the acetabular component |
Magnification2= | Preoperative distance between teardrops | ⅹ | Magnification1 |
Postoperative distance between teardrops |
Thus, 128 patients were involved in the study, including 69 females and 59 males. The mean age at the time of index arthroplasty was 56.1 ± 11 years (range 26–77 years). All operations were performed by four senior authors through a posterolateral approach.
Institutional review board approval was obtained. The preoperative diagnoses in these patients included femoral neck fractures (15 cases), osteonecrosis of the femoral head (64 cases), developmental hip dysplasia (16 cases), THA revision due to aseptic loosening (five cases), primary osteoarthritis (23 cases), rheumatoid arthritis (four cases), and ankylosing spondylitis (one cases). All implants were cementless: 68 Trilogy IT (Zimmer, USA), 52 Reflection SP3 (Smith & Nephew, USA), six CombiCup PF (LINK, Germany), and two Zuanshi(IRENE, China). A total of 41 cases had ceramic-on-ceramic bearing surfaces, 39 had ceramicon-polyethylene bearing surfaces, 29Oxinium-polyethylene bearing surfaces, and 19 had metal-on-polyethylene bearing surfaces.
To determine intra- and interobserver variability, 20 randomly selected patients were measured initially by Observer I(Lu) and Observer II(Yuan). After 3 weeks the radiographs were measured again by the same authors except physical measurements. An intraclass correlation coeffcient (ICC) > 0.8 was considered excellent agreement, ICC 0.6–0.8 was fair to good agreement, and ICC < 0.6 was poor agreement.
Statistical analysis
Quantitative data are expressed as means ± standard deviation (SD). Statistical analyses were performed using PASW statistics 18 (SPSS Inc., Chicago, IL, USA). The normality assumption of our data was checked using the Kolmogorov-Smirnov test. The One-way ANOVA test and paired t test were used to detect the measurements of LLD of the lower limbs between GT-IT group and the other three groups. The Bland-Altman diagram was used to check the agreement of the four measurements. P < 0.05 was considered statistically significant.