Aly Mohamed M, El-Sharkawi Mohammad, Joaquim Andrei F, Pizones Javier, Santander Espinoza Xavier A, Popescu Eugen C, Bin Shebree N Abdulaziz, Gerdhem Paul, Öner Cumhur F
Department of Neurosurgery, Mansoura University, Mansoura, Egypt.
Department of Neurosurgery, Prince Mohammed Bin Abdulaziz Hospital, Riyadh, Saudi Arabia.
Clin Spine Surg. 2025 Jul 1;38(6):266-277. doi: 10.1097/BSD.0000000000001764. Epub 2025 Mar 24.
To review the historical thoracolumbar burst fractures (TLBFs) classifications and discuss the probable gaps for their clinical validation.
Despite multiple classification schemes, the treatment decisions for TLBFs in neurologically intact patients remain controversial. There are gaps between the current classifications and their predictive validation.
A narrative literature review.
The potential barriers to establishing the predictive value of the current classifications of TLBFs could be connected to validation studies' flaws such as nonvalidated outcome measures and challenges of randomization. It could also be related to limited interobserver reliability in diagnosing A3/A4 fractures. Finally, it might be attributed to the inability to incorporate all prognostic variables, such as computed tomography (CT) parameters, patient-related factors, and traumatic disc injury, may result in failed validation.
AOSpine Patient and Clinical Reported Outcome Spine Trauma (PROST) and a recently proposed natural experiment observational study hold promise for mitigating methodological challenges. A structured approach for distinguishing A3/A4 fractures and standardized CT criteria for PLC injury is critical to improving reliability. Finally, a treatment algorithm incorporating all potential prognostic variables, independent of the morphologic classification, may improve the predictive value of the classification. Machine learning techniques could be helpful in this context.
回顾胸腰椎爆裂骨折（TLBFs）的历史分类，并讨论其临床验证中可能存在的差距。
尽管有多种分类方案，但对于神经功能完整的TLBFs患者的治疗决策仍存在争议。目前的分类与其预测性验证之间存在差距。
进行叙述性文献综述。
目前TLBFs分类的预测价值难以确立，可能与验证研究的缺陷有关，如未经验证的结果测量指标和随机化的挑战。这也可能与诊断A3/A4骨折时观察者间可靠性有限有关。最后，可能归因于无法纳入所有预后变量，如计算机断层扫描（CT）参数、患者相关因素和创伤性椎间盘损伤，可能导致验证失败。
AOSpine患者和临床报告的脊柱创伤结局（PROST）以及最近提出的自然实验观察性研究有望缓解方法学挑战。区分A3/A4骨折的结构化方法和PLC损伤的标准化CT标准对于提高可靠性至关重要。最后，一种独立于形态学分类并纳入所有潜在预后变量的治疗算法可能会提高分类的预测价值。机器学习技术在这方面可能会有所帮助。
