Patients
From August 2013 and June 2017, 29 consecutive patients (7 males and 22 females) presenting DDH underwent derotational femoral shortening osteotomy were included in our hospital. Patients with subluxation of hip or classified as grade I were not involved in this study. Based on different surgical methods, they were divided into the conventional group (n = 14) and navigation template group (n = 15). There was no significant difference between the two groups in terms of age, gender, side, Tönnis classification, and surgery history (Table 1). This study was approved by the Ethics Review Committee of Xiangya Hospital Central South University, and all parents in the present study signed an informed consent to participate.
Table 1 Comparison of demographic data and deformity characteristics between two groups
Characteristics
|
Conventional group
(n = 14)
|
Navigation template group
(n = 15)
|
P value
|
Mean age (range), years
|
4.2 ± 1.6 (3-8)
|
3.7 ± 2.2 (2-8)
|
0.478
|
Gender, n (%)
|
|
|
0.742
|
Male
|
3 (21.4)
|
4 (26.7)
|
|
Female
|
11 (78.6)
|
11 (73.3)
|
|
Side, n (%)
|
|
|
0.597
|
Left
|
8 (57.1)
|
10 (66.7)
|
|
Right
|
6 (42.9)
|
5 (33.3)
|
|
Tönnis classification
|
|
|
0.657
|
II
|
4 (57.1)
|
3 (20.0)
|
|
III
|
7 (54.5)
|
10 (66.7)
|
|
IV
|
3 (18.2)
|
2 (13.3)
|
|
Surgery history
|
|
|
0.573
|
Yes
|
8 (57.1)
|
7 (46.7)
|
|
No
|
6 (42.9)
|
8 (53.3)
|
|
Digital design and navigation template preparation
Preoperatively, all patients underwent X-rays and femoral CT scanning. The obtained slice CT scanning data in navigation template group were saved as DICOM (Digital Imaging and Communications in Medicine) format and then transferred into the Mimics 20.0 software (Materialise, Leuven, Belgium) for three-dimensional reconstruction. 3D models of femur were accurately reconstructed and exported in STL (Standard Tessellation Language) format. Thereafter the deformed 3D models were imported into UG 6.0 software. The osteotomy plane and accurate planning for the derotational degrees were defined based on the femoral parameters (Fig. 1a). And then the personalized corrective osteotomy template model with four guide holes was fitted to the bone surface with reference to the characteristic configuration of the deformed femur and anatomical landmarks (Fig. 1b). The designed corrective osteotomy template was then printed using 3D printing technology with materials of acrylic resin.
3D reconstruction was performed using Amira 3.1 software. By tracking the contours of the prominent bone structure for reconstruction, adjusting the geometric alignment of the overlapping point contours, modeling the surface by meshing into a polygonal point frame, and reconstructing the target contour based on the point data. The result is a fully interactive 3D visualization of the reconstructed structure based on radiology and CT imaging data, allowing full visualization of the area. The 3D models were saved in STL format for stereolithography, which is a universal international file standard for 3D modeling and printing, and then imported into Imageware (version 12.0; EDS, Palo Alto, CA) software for further analysis. The optimal osteotomy plane and accurate planning for the derotational degrees were defined based on the femoral parameters using 3D printing technology. System parameters included the thickness of the processing layer at 0.1 mm, processing speed at 500 mm/s. The entire process of prototype construction required approximately 5-18 h, with an average of 8.9 h.
Surgical procedures
All the proximal femoral corrective osteotomies in the present study were performed by one senior orthopedic surgeon in our department. The navigation template was sterilized and applied intraoperatively to assist derotational femoral shortening osteotomy based on preoperative simulation (Fig. 2). When the proximal femur was exposed, usually via a separate lateral incision, the navigation template was matched to the proximal lateral femur in the best possible manner (Fig. 3a). Four K-wires were inserted through the navigation holes of the template (Fig. 3b), then the bone was osteotomized through a slit on the template. The template was then removed with the K-wires left in place, matching with the shortening osteotomy bone which was required to provide a pressure-free reduction. Rotating the K-wires into the parallel status and this position was maintained with a reduction guide (Fig. 3c). Therefore we were able to perform all osteotomies as preoperatively simulated. Finally, they were managed by internal fixation with Locking Compression Plate (LCP) (Fig. 3d). On the other hand, the derotational degrees of the osteotomy and shortening distance were performed on the basis of preoperative and intraoperative measurements in the conventional group.
Postoperative management
No significant difference in postoperative management procedures after derotational femoral shortening osteotomy between the two groups. A spica cast fixation was used for 8 weeks and double lower limb brace with hip abduction for another 8 weeks. Weight bearing was guided to start after removal of the brace. Moreover, radiographs of the pelvis and femur were taken regularly until the region of osteotomy completely healed and then the internal fixation was removed. The mean follow-up was 4.8 years (3 to 7 years) and patients were reviewed at eight weeks, four and 12 months and then every year until skeletal maturity.
Statistical analysis
Quantitative data in this study were statistically analyzed by the SPSS 25.0 software (SPSS, Inc., Chicago, USA) and manifested as count (percentage) or mean ± standard deviation (SD). Student’s t test, chi-squared test, and Fisher’s exact test were applied to analyze the data in this study. Different parameters measured between two groups were evaluated with independent t test for continuous variables, and chi-square test or Fisher’s exact test for the categorical variables. A P value < 0.05 was considered statistically significant.