Efficacy evaluation of three-dimensional printing assisted osteotomy guide plate in accurate osteotomy of adolescent cubitus varus deformity

A total of 25 patients (15 males and 10 females) with cubitus varus deformity who were admitted to the Jiangxi Provincial People’s Hospital Affiliated with Nanchang University from June 2014 to December 2017 were included in this study. Patients were enrolled into the conventional group (n = 11) and 3D printing group (n = 14) by the random number table method after admission and treated with the different surgical procedures, respectively. The cubitus varus deformity of all patients was caused by the malunion of humeral supracondylar fracture, and there was no bone metabolic disease. This study was approved by the Ethics Review Committee of Jiangxi People’s Hospital Affiliated to Nanchang University, and all the patients signed an informed consent to participate in the study.

All patients were photographed with the X-rays and computed tomography (CT) of bilateral upper limbs, and the carrying angle of the healthy side and the cubitus varus angle of the deformed side were measured, respectively. On the basis of the previous studies by Takeyasu et al. [15], the osteotomy angle was calculated to be the sum of the carrying angle of the healthy side and the cubitus varus angle of the deformed side. In addition, the CT scanning data of patients in 3D printing group were collected by the dual-source 64-slice spiral CT system (Siemens, Germany) in our hospital, and the scanning parameters were voltage 120 KV and pitch 0.625 mm. The initial CT data of deformed upper limbs were stored in Digital Imaging and Communications in Medicine (DICOM) format and input into the Mimics 19.0 software (Materialize, Leuven, Belgium) to generate three-dimensional reconstructed upper limb models. Then, the osteotomy plane was defined parallel to the articular surface 1 cm above the olecranon, and the other osteotomy plane was defined by the calculated osteotomy angle. The two planes intersected at 2 mm from the medial bone cortex. The simulated reduction of bone ends after osteotomy was then performed, and the carrying angle after osteotomy was measured again to check whether the cubitus varus deformity was corrected (Fig. 1). Then, the anatomical data corresponding to the bony surface of distal humerus were extracted and treated with reverse thickening of 5 mm. The basal plate with the same shape was established, and the data of Kirschner wire guide holes were imported at the same time. After Boolean operation, all guide holes were penetrated and the boundary was trimmed to complete the design and fabrication of the guide plate (Fig. 2). Ultimately, the design data were saved in STL format and output to 3D printer (Dongwang Med, Inc., Xian, Shanxi, China), and the osteotomy guide plate was then printed with materials of photosensitive resin.

All operative procedures were performed by senior surgeons in the same treatment team. Under the general anesthesia or brachial plexus block anesthesia, the patient was placed into a supine position and a pneumatic tourniquet was used to block the blood circulation at the root of the upper limb. In 3D printing group, the placement of osteotomy guide plate and the determination of osteotomy lines were directly referred to the results of preoperative simulation. In this process, two Kirschner wires with a diameter of 2.0 mm were drilled into the humeral shaft along the guide plate holes to fix the osteotomy guide plate. After the firm fixation, the wedge-shaped osteotomy was performed along the upper and lower osteotomy planes of the guide plate, and the medial bone cortex was preserved (Fig. 3). However, the determination of osteotomy angle and osteotomy lines in the conventional group were only based on the preoperative measurements and intraoperative attempts. After checking the carrying angle was satisfactory, different specifications of Kirschner wires or reconstruction locking plates (Dabo Med, Inc., Xiamen, Fujian, China) were selected for shaping and fixation, and the incision was routinely closed with 2/0 absorbable sutures.

There was no significant difference in postoperative management between the two groups. Patients in the two groups were treated with long arm plaster support to fix the upper limb. Moreover, postoperative radiographs of the elbow joint were reexamined regularly, and the healing of the incision, range of elbow joint motion, carrying angle, postoperative complications, and the healing of osteotomy ends were observed and recorded. Until the imaging examinations confirmed the formation of continuous callus at the osteotomy end, the plaster support was removed and the patient was guided to start partial weight-bearing exercise and gradually transitioned to complete weight-bearing.

The parameters of operation time, intraoperative blood loss, osteotomy degrees, and osteotomy end union time of patients between two groups were observed and recorded. Moreover, at the last follow-up, the prognosis of cubitus varus deformity was evaluated based on the Bellemore criteria [16]: (1) excellent: the difference of carrying angle between the deformed side and healthy side was less than 5°, and the restriction of elbow joint flexion and extension was less than 10°; (2) good: the difference of carrying angle between the deformed side and healthy side was 6° to 10°, and the restriction of elbow joint flexion and extension was 11° to 15°; (3) fair: the difference of the carrying angle between the deformed side and healthy side was 11° to 15°, and the restriction of elbow joint flexion and extension was 15° to 20°; and (4) poor: the restriction of elbow joint flexion and extension was more than 20°, accompanied with residual varus deformity, or the complications requiring secondary surgery. In addition, parents’ satisfaction with appearance after the deformity correction was also recorded. In the scoring mechanism with a full score of 100 points, a score ≥ 90 was considered as excellent, 75–89 as good, 50–74 as fair, and < 50 as poor.

The overall data in this study were statistically analyzed by the SPSS 23.0 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 regarded as statistically significant.

Table 1 revealed the demographic data and deformity characteristics of patients in two groups. Therein, the carrying angle of the healthy side in the conventional group and 3D printing group was 9.6 ± 1.3° and 10.1 ± 1.4° respectively, and the carrying angle of the deformed side was − 17.8 ± 6.3° and − 18.3 ± 6.7°, respectively. However, the comparison between two groups showed no statistical significance in gender, age, causes of deformity, time from fracture to operation, deformity side, and preoperative carrying angle of healthy and deformed sides (P > 0.05 for all).

Table 2 revealed the clinical data of patients in two groups. There was statistical significance in operation time between the conventional group (73.5 ± 10.3 min) and 3D printing group (48.3 ± 8.9 min, P < 0.001). The intraoperative blood loss in the 3D printing group (35.6 ± 9.7 ml) was significantly less than that of the conventional group (52.1 ± 11.5 ml, P < 0.001). Regarding the osteotomy degrees, there was no statistical significance between the conventional group (24.9 ± 6.6°) and 3D printing group (25.7 ± 7.1°, P = 0.183). As for the osteotomy ends union time, there was also no statistical significance between the conventional group (9.8 ± 2.7 weeks) and 3D printing group (9.3 ± 2.1 weeks, P = 0.417). According to the Bellemore criteria [16], 12 patients in the 3D printing group obtained excellent correction, 1 patient obtained good prognosis, and 1 patient obtained fair prognosis, while 8 patients obtained excellent correction, 2 patients obtained good prognosis, and 1 patient obtained fair prognosis in the conventional group. In addition, it is worth noting that the rate of excellent deformity correction in the 3D printing group (85.8%) was significantly higher than that of the conventional group (72.7%, P = 0.019).

Table 3 revealed the postoperative functional outcomes of patients in two groups. There was no statistical significance in the follow-up time between the conventional group (17.8 ± 3.1 months) and 3D printing group (18.3 ± 2.9 months, P = 0.317). The carrying angle of the deformed side after correction in the conventional group and 3D printing group was 8.3 ± 2.6° and 8.7 ± 2.4°, respectively, which is not statistically different (P = 0.626). Furthermore, the motion of flexion was 128.7 ± 4.9° in the conventional group and for the 3D printing group was 131.6 ± 5.8° (P = 0.451). The motion of rotation was 143.5 ± 7.7° in the conventional group and for the 3D printing group was 146.2 ± 8.1° (P = 0.192). The motion of extension was 2.4 ± 1.1° in the conventional group and for the 3D printing group was 2.6 ± 0.9° (P = 0.523). Besides, no significant difference in the range of elbow joint motions was noted between the two groups at the last follow-up. Regarding the parents’ satisfaction with appearance after deformity correction, nine patients in the conventional group were regarded as excellent, and two patients were regarded as good. However, 13 patients in the 3D printing group were regarded as excellent, and one patient was regarded as good. Compared with the conventional group (81.8%), the 3D printing group (92.9%, P = 0.023) exhibited a higher rate of excellent satisfaction.

Table 4 revealed the postoperative complications of patients in two groups. Therein, one patient in the conventional group developed the ulnar nerve paralysis after operation, and the symptoms disappeared after 3 weeks of treatment with neurotrophic drugs. Besides, two patients in the 3D printing group and one patient in the conventional group have experienced the temporary restriction of elbow joint motion, but returned to normal after the removal of plaster support and the performance of progressive functional exercise. In addition, no other complications such as the incision infection, osteotomy nonunion, and internal fixation loose or broken were observed in the remaining patients. The total complication rate of the conventional group and 3D printing group was 2 out of 11 (18.2%) and 2 out of 14 (14.3%), with no statistical significance (P = 0.371).

A male, 13 years old, who experienced a bicycle fall when he was 10 years old and suffered a humeral supracondylar fracture is a typical case. Due to the lack of attention at that time, the cubitus varus deformity appeared 4 months later (Fig. 4). After careful measurements, the cubitus varus angle of his left elbow joint was 15°, the carrying angle of the right elbow joint was 9°, the flexion motion of the left elbow joint was 100°, and the extension motion of the left elbow joint was 10°. After the HSWO assisted by 3D printing osteotomy guide plate, X-ray showed that the osteotomy end was aligned well and the fixation effect was satisfactory (Fig. 5). Three months after the operation, his osteotomy end achieved bone healing. After 18 months of follow-up, the carrying angle of the left side was 8°, the flexion motion of the left elbow joint was 129°, and the extension motion of the left elbow joint was 0°, which was evaluated as excellent according to the Bellemore criteria [16].