Data from patients with OVCF accompanied by scoliosis who underwent PKP at The First Affiliated Hospital of Soochow University (Suzhou, China) between January 2018 and December 2020 were retrospectively analysed. Since the choice between a unilateral or bilateral approach for treatment of patients with OVCF and scoliosis has always been controversial, prior to the present study, it was unclear which approach was more suitable for these patients. The patients were informed about the advantages and disadvantages of the two surgeries and the bilateral or unilateral puncture approach was selected following the patients' decision. According to the inclusion and exclusion criteria, 52 patients were included in the present study, which was approved by the Ethics Committee of The First Affiliated Hospital of Soochow University. Informed written consent was obtained from all patients. The patients were split into two groups: Unilateral PKP group (n=26; mean age, 71.7±8.1 years) and bilateral PKP group (n=26; mean age, 69.2±9.6 years). All patients were followed up for at least one year with their consent. The preoperative, intraoperative and postoperative information of these patients was gathered via radiographic and medical records and routine outpatient follow-ups.

Inclusion criteria were as follows: i) Patients aged ≥50 years with local tenderness, low back pain and lumbar dysfunction; ii) T-score for bone mineral density (BMD) ≤-2.5, compliant with the diagnostic criteria of osteoporosis (14); iii) radiograph showing a cobb angle >10 degrees in the coronal plane, compliant with the diagnostic criteria of scoliosis without neurological deficits or gross instability (15); iv) MRI showing a single-level vertebral body with a fresh vertebral compression fracture; and v) data integrity of the patients during preoperative, intraoperative and postoperative periods. Exclusion criteria were as follows: i) Pathological fractures ascribed to tuberculosis or tumour; ii) patients with mental illness, such as depression or Alzheimer's disease; iii) presence of coronary heart disease; iv) patients with long-term use of glucocorticoids for treating rheumatic disease and v) abnormal coagulation.

The standard procedure of PKP was performed by senior surgeons as described in a previous study (16). The patient was placed on an operating table in the prone position. In the bilateral group, according to the aforementioned surgical methods, symmetrical incisions were made at 2-2.5 cm proximal to the apical spinous process. A puncture needle was inserted at an abduction angle of 25-35 degrees into both sides of the vertebral body through the pedicle with the guidance of a C-arm. Subsequently, a working channel was established through a guidewire. An appropriate inflatable balloon was inserted into the vertebral body, which was inflated to provide space for cement injection into the compressed vertebral body. OSTEOPAL^®V bone cement was used in this study. The formulation of the bone cement included the powder phase and the liquid phase. The powder phase was composed of Poly (methyl acrylate, methyl methacrylate), Zirconium dioxide, Benzoyl peroxide. The liquid phase was composed of methyl methacrylate (MMA); a peroxide decomposer, N,N-dimethyl-p-toluidine; and hydroquinone. The formulation of the bone cement was most effective when the ratio of powder to liquid was 2:1. The bone cement should be of ‘tooth paste’ consistency prior to injection (17). High-viscosity cement was slowly infused into the inflated vertebral body through the working channel. In the unilateral group, only one incision was made, and the puncture site was selected on the side with a clear pedicle shadow under the C-arm fluoroscopy. Under anteroposterior fluoroscopy, the tunnel was expanded to the anterior part of the vertebral body using a dilator to reach the middle of the vertebral body. The balloon was placed as deep as possible in the middle of the vertebral body. Balloon dilation and bone cement injection were performed in the same way as in the bilateral approach. After surgery, all patients received >1 year anti-osteoporosis treatment with 600 mg of supplemental calcium and 800 IU of vitamin D daily. All patients received preoperative X-ray, MRI and CT to identify the fracture segment and postoperative 1-day X-ray to assess the outcome of treatment (Figs. 1 and 2).

The operation time, bone cement injection volume, intraoperative fluoroscopy frequency, duration of hospital stay and apical vertebral rotation were recorded. Nash-Moe classification ranged from 0 (both sides of the pedicles were symmetrical with no vertebral rotation degree) to 4 (the most severe vertebral rotation degree) and was used to reflect the vertebral rotation degree (18). The vertebral body height and local kyphotic angle (KA) were measured on a standing lateral radiograph pre- and postoperatively. The fractured vertebral body height was measured at the point of most pronounced compression (anterior or middle). The KA was calculated from the angle of intersection of the lines parallel to the upper and lower end plates of the fractured vertebrae (Fig. 3). The compression rate (CR) was calculated as the height of each fractured vertebral body divided by the height of the neighbouring normal vertebral body. Postoperatively, the restoration rate (RR) was calculated as follows: (Restored vertebral body height-original fractured vertebral body height)/adjacent normal vertebral body height (Fig. 4) (19). The visual analogue scale (VAS) and Oswestry disability index (ODI) scores were evaluated preoperatively and postoperatively. VAS ranged from 0 (no pain) to 10 (the worst pain experienced) and was used for measuring back pain. Daily life function was estimated using the ODI score (20). The ODI score system includes 10 sections: Pain intensity, personal care, lifting, walking, sitting, standing, sleeping, sexual life, social life and traveling. Each section consists of six statements with a score of 0-5. The score is calculated as follows: (Total score/5x number of questions answered) x100% (21). Calculation of the VAS and ODI was described in a previous study (22). The occurrence rates of the adjacent vertebral body fractures and bone cement leakage were calculated postoperatively at follow-up.

Statistical analysis was performed using SPSS (version 25.0; IBM Corp.). Independent sample Student's t-test was used to compare the age, BMD T-score, intraoperative data, CR and RR between the two groups. A mixed ANOVA followed by Bonferroni correction was used for comparing KA and average vertebral height between the two groups at different time points and within the same group pre- and postoperatively. The Mann-Whitney U test was used to compare VAS and ODI scores between the two groups, while the Wilcoxon signed-rank test was used to compare the preoperative and postoperative VAS and ODI scores of the same group, and each test was followed by Bonferroni correction. The variables of Nash-Moe grade and adjacent vertebral fracture were analysed by Fisher's exact test, while the remaining variables of sex, fracture location, scoliosis direction and cement leakage rate were analysed using the Chi-square test. The parametric numerical data are presented as mean ± standard deviation, the non-parametric ordinal data are presented as the median (interquartile range), and the count data are shown as n (%). P<0.05 was considered to indicate a statistically significant difference.

The mean BMD T-score was -2.9±0.3 in the unilateral group and -3.0±0.4 in the bilateral group. In the unilateral group, 12 patients had single thoracic fractures whereas 14 patients had single lumbar fractures. In the bilateral group, 10 patients had single thoracic fractures whereas 16 patients had single lumbar fractures. There were no significant differences between the two groups in terms of BMD and fracture location. An apical vertebral rotation was present in 76.9% (20/26) of the patients in the unilateral group and 65.3% (17/26) of the patients in the bilateral group, with no significant difference between the two groups. The direction of scoliosis also showed no significant difference between the two groups (P>0.05) (Table I).

The operation time in the unilateral group was shorter than that in the bilateral group (40.5±9.0 vs. 57.2±11.4 min; P<0.001). In addition, the injected cement volume was significantly lower in the unilateral group than that in the bilateral group (4.7±0.9 vs. 7.7±1.2 ml; P<0.001). Furthermore, the frequency of intraoperative fluoroscopy was lower in the unilateral group than that in the bilateral group (28.3±7.4 vs. 52.6±9.6 times; P<0.001). Regarding the duration of hospital stay, there was no significant difference between the two groups (P>0.05). The VAS and ODI scores at 1-day after surgery and final follow-up improved significantly compared with those before surgery in both groups (P<0.05). However, these scores did not vary significantly between the two groups both before and after surgery (P>0.05; Table II).

The CR of the vertebral body before surgery was not significantly different between the two groups (P>0.05). The average vertebral body height and KA significantly improved after surgery in both groups compared with the preoperative measurements (P<0.05). However, after surgery, RR and KA were not statistically significant between the two groups (P>0.05). During a mean follow-up of 13.5 months (range, 12-16 months) after surgery, the unilateral group included three cases (11.5%) with bone cement leakage, whereas the bilateral group included nine cases (34.6%) (P<0.05). In patients with bone cement leakage, neither clinical nor neurological signs were observed. In the unilateral group, three of 26 (11.5%) patients exhibited adjacent vertebral fractures, the first patient at 1 month, the second at 6 months and the third at 12 months postoperatively. In the bilateral group, two of 26 (7.7%) patients exhibited adjacent vertebral fractures, one patient at 8 months and the other at 14 months postoperatively. The incidence of adjacent vertebral fractures did not vary significantly between the two groups (Table III). Preoperative X-ray, MRI and CT images of the typical cases of the both groups showed a single-level vertebral body with a fresh vertebral compression fracture in a patient with scoliosis. Postoperative 1-day X-ray showed even distribution of bone cement and partial recovery of the vertebral height and kyphotic angle (Figs. 1, 2).