The most significant finding of this study is the design of a novel imaging-based scoring system for OVCFs, and the acquisition of evaluation data of this system is facilitated. In this retrospective study, postoperative pain relief, functional improvement, prevention of progressive kyphosis, and residual back pain were all observed to be inferior in the high-score group compared to the low-score group. Furthermore, through univariate and multivariate regression analyses, the NIBSS score was identified as a high-risk factor for progressive kyphosis and residual back pain. Therefore, for patients with a preoperative NIBSS score exceeding 5, recommendations may include rigid bracing, proactive osteoporosis treatment, and even adjunctive internal fixation.
The NIBSS system is primarily used to evaluate OVCFs from three perspectives: X-ray, CT, and MRI. Preoperative X-ray lateral views are primarily utilized for assessing vertebral height loss and kyphosis angle. Previous literature indicate that increased vertebral height loss is associated with heightened instability of vertebral compression fractures and increased risks of fracture healing deformity and non-union12. Additionally, previous studies have suggested that a large preoperative kyphosis angle is a risk factor for post-kyphoplasty progressive kyphotic deformity. KN Chou reported 14 cases of patients with postoperative progressive kyphosis following PVP, with preoperative kyphotic angles averaging 22.86°21. In this scoring system, preoperative CT was used to evaluate the integrity of the vertebral posterior wall. According to Denis' three-column theory, compression fractures mainly involve the anterior column of the vertebral body, while burst fractures affect the anterior and middle columns, increasing fracture instability11. Furthermore, MRI signals and the intervertebral disc endplate complex were assessed on preoperative MRI T2WI. Tsujio first reported the classification of MRI signals on T2WI of OVCFs and found that diffuse low signal increases the risk of vertebral collapse and progressive kyphosis compared to high signal change16. In this scoring system, their classification was also adopted. A larger area of low signal on T2WI indicates more severe fractures, suggesting more fracture of trabecular bone, whereas high signal represents local edema, indicating relatively mild trabecular bone damage. Furthermore, previous studies have reported that intervertebral disc endplate complex injury is a high-risk factor for post-kyphoplasty progressive kyphotic deformity12. The possible reason is that intervertebral endplate disc complex injury causes accelerated disc degeneration, loss of disc height, and exacerbates kyphotic deformity. The data of three parts of this scoring system are easy to obtain, highly reproducible, and easy to promote and apply clinically.
In terms of preventing vertebral re-collapse and kyphotic deformity, the low-score group significantly superior to the high-score group. In this study, the VHL difference at the last follow-up in the low-score group was 2.556 ± 4.8836, which was markedly lower than that in the high-score group, 5.1356 ± 7.611, with statistically differences. Additionally, the KA difference at the last follow-up in the low-score group was 1.2 (-0.9, 4.125), also significantly lower than that in the high-score group, 10.7 (5.1, 11.9), with a p-value less than 0.001. Furthermore, the low-score group also showed superior postoperative pain relief and functional improvement compared to the high-score group. In this study, both postoperative and last follow-up VAS and ODI scores in the low-score group were significantly lower than those in the high-score group.
In this study, we found that NIBSS score is one of the high-risk factors for progressive kyphotic deformity after PKP surgery. Preoperative evaluation of vertebral height and kyphotic angle is an important component of the NIBSS system, and as height loss and kyphotic angle increase, the score correspondingly increases. Previously, Yu et al. reported that severe preoperative kyphotic deformity is a high-risk factor for vertebral re-collapse after PKP, which is consistent with our study findings22. Additionally, Yu et al. reported that better postoperative vertebral height restoration is a high-risk factor for vertebral re-collapse after PKP22, while Wang et al. also reported that higher postoperative vertebral height is a high-risk factor for vertebral re-collapse23. We also found that a higher postoperative kyphotic angle difference is a high-risk factor for progressive kyphotic deformity, which is consistent with previous studies. Dong and Yu also reported that the fracture location of thoracolumbar is a high-risk factor for post-PKP re-collapse22,24, while in this study, fracture location was not identified as a high-risk factor. Furthermore, previous literature has reported that cement distribution and low cement injection volume are high-risk factors for vertebral re-collapse after PKP, but in this study, these factors were not found to be associated with progressive kyphotic deformity.
Characteristic signals, including homogeneous high signal and black-liner signal, are discernible on MRI STIR images in OVCFs, as initially documented by Omi17. Omi also highlighted that non-homogeneous high signal and black line signal are deemed high-risk factors for nonunion of fractures 17. In our study, the black line signal was recognized as an independent risk factor for PK after PKP (OR = 19.686). This suggests that, in clinical practice, OVCF patients displaying black-liner signal should receive heightened attention, prompting a more proactive approach to treatment.
The NIBSS score is also one of the important risk factors for residual lower back pain after surgery. Yang et al. reported that kyphotic angle greater than 20 degrees and AO type A4 fractures are high-risk factors for residual lower back pain after surgery13. In the NIBSS scoring system, the larger the preoperative and postoperative kyphotic angles, the higher the score. Additionally, in preoperative CT evaluation, burst fractures are assigned 2 points, higher than compression fractures which are assigned 1 point. This finding is somewhat consistent with the results reported by Yang et al. Furthermore, Yu and Li also reported that cement distribution is a risk factor for residual lower back pain after surgery25,26. In our study, we also found that Type D cement form, where the cement does not contact the upper and lower endplates, is a risk factor for residual lower back pain after surgery. Additionally, Yang reported that cement injection volume less than 5ml13, and Yu reported that cement injection volume less than 3ml26 are risk factors for postoperative residual lower back pain, but in our study, we did not find a correlation between cement injection volume and postoperative residual lower back pain. Moreover, we also found correlations between residual back pain and progressive kyphosis, the Spearman's rank correlation coefficient is 0.618. For patients in the high-score group, necessary strategies need to be taken to prevent progressive kyphotic deformity and reduce the occurrence of postoperative residual back pain.
The main innovation of this study is the design of a new scoring system for OVCFs. Importantly, the evaluation scores for various parameters can be easily obtained preoperatively with high reproducibility. Despite the widespread use of the TLICS scoring system for evaluating thoracolumbar fractures, it primarily focuses on traumatic thoracolumbar fractures and does not consider osteoporotic factors10. Similarly, the Genant and Magerl classifications mainly emphasize morphological assessment of fractures, lacking a comprehensive evaluation of osteoporotic compression11. Yang also reported a new scoring system for predicting post-vertebroplasty residual lower back pain, which evaluates age, fracture location, fracture type, kyphotic angle, and cement injection volume among five aspects13. However, this evaluation system is relatively complex with poor reproducibility. Additionally, cement injection volume is a postoperative factor, making preoperative assessment of fractures challenging. In this study, by calculating the score of NIBSS, orthopedic surgeons can easily estimate the risk of PK and RBP after PKP in OVCF patients. For patients assessed with low-score, PKP remains a safe and effective treatment option. However, for those assessed with high-score, recommendations may include rigid bracing, proactive osteoporosis treatment, and even adjunctive internal fixation.
Our study still has some limitations. Firstly, it is a retrospective study, which may introduce biases in case selection and data completeness. Secondly, it is a single-center study, the data originated from the same hospital and were not validated across multiple centers in other regions and countries. This may restrict the applicability of the model in certain hospitals. Therefore, further validation in large-sample multi-center studies is warranted in the future.