Exploring the MRI-based vertebral bone quality score as a novel predictor of spongy cement distribution in osteoporotic vertebral compression fractures following percutaneous vertebral augmentation
Previous studies have demonstrated that a spongy cement distribution pattern is associated with superior clinical and radiological outcomes in patients with osteoporotic vertebral compression fractures (OVCFs). However, the relationship between MRI-based vertebral bone quality (VBQ) score and spongy cement distribution pattern remains unclear. This study aimed to evaluate the predictive value of the VBQ score for achieving a spongy cement distribution following percutaneous vertebral augmentation (PVA) and to identify independent influencing factors. We retrospectively reviewed patients with single-level OVCFs who underwent percutaneous vertebroplasty (PVP) or percutaneous kyphoplasty (PKP) from January 2020 to December 2024. Patients were divided into spongy distribution group (Spongy group, n=94) and non-spongy distribution group (Non-spongy group, n=142) based on postoperative anteroposterior and lateral X-rays. Potential factors, including age, sex, body mass index (BMI), bone mineral density (BMD), hypertension, diabetes, serum uric acid levels, smoking history, glucocorticoid use, fractured vertebral level, injected cement volume, surgical technique (PVP or PKP), VBQ score, intravertebral vacuum cleft (IVC), vertebral compression ratio (VCR), local cobb angle (LCA), and bone cement leakage, were collected and analyzed. Clinical outcomes, including visual analog scale (VAS) and Oswestry Disability Index (ODI) scores, were recorded and compared between groups at multiple postoperative time points. Complications such as adjacent vertebral fracture (AVF) and bone cement displacement (BCD) were recorded. Logistic regression analysis was performed to identify independent factors associated with the spongy cement distribution pattern. The predictive performance of VBQ score was evaluated by receiver operating characteristic (ROC) curve analysis. A total of 236 patients were included. Univariate analysis revealed significant differences between the two groups in age, serum uric acid level, VBQ score, BMD T-score, cement volume, presence of IVC, VCR, LCA, and surgical technique (all P < 0.05). Binary logistic regression demonstrated that VBQ score (OR = 0.179, 95% CI 0.053–0.602, P = 0.005), VCR (OR = 0.770, 95% CI 0.677–0.875, P = 0.001), LCA (OR = 0.750, 95% CI 0.667–0.884, P = 0.001), IVC (OR = 0.171, 95% CI 0.045–0.648, P = 0.009), and cement volume (OR = 0.247, 95% CI 0.142–0.430, P = 0.001) were associated with a lower likelihood of achieving a spongy pattern, whereas higher BMD T-score (OR = 6.101, 95% CI 2.097-17.750, P = 0.001) and the PVP (OR = 4.914, 95% CI 1.374-17.574, P = 0.014) surgical technique were associated with a higher likelihood. The VBQ score demonstrated good predictive ability, with an area under the ROC curve (AUC) of 0.759 (95% CI: 0.697-0.821) and an optimal cut-off value of 3.155 (sensitivity: 72.3%, specificity: 73.2%). Patients with spongy cement distribution showed significantly better pain relief and functional improvement, as well as a lower incidence of complications (all P < 0.05). The preoperative VBQ score is a simple, non-invasive, and reliable predictor of spongy cement distribution pattern following PVA in patients with OVCFs. The VBQ score, LCA, VCR, IVC, and higher cement volume were associated with a lower likelihood of achieving a spongy pattern, whereas higher BMD T-score and the PVP technique were associated with a higher likelihood. These findings may assist surgeons in preoperative risk stratification and surgical planning.