中文摘要：

目的探讨胸腰椎爆裂性骨折椎体后上缘骨折块形成与椎基静脉孔的关系及椎体后上缘骨折块的形成机制。方法2013年6月至2016年6月收治带有椎体后上缘骨折块的胸腰椎爆裂性骨折患者62例，术前行胸腰椎CT扫描及矢状面重建。在横断面图像上测量后上缘骨折块长度（RL）和椎体长度（VL），并计算RUVL值。在矢状面图像上测量后上缘骨折块宽度（RW）和椎体宽度（vw），并计算Rw，vw值；测量后上缘骨折块高度（RH）和椎体高度（VH），并计算RH／VH值，进而确定后上缘骨折块与椎基静脉孔的位置关系。选取8个新鲜冰冻腰椎标本行Micro-CT扫描，将图像分为三层、每层九个区域进行分析（上层SR1-SR9，中层MR1-MR9，下层IR1MR9），比较各个区域骨微结构——骨体积分数、骨密度、连接密度、结构模型指数、骨小梁数量、骨小梁厚度的差异，观察椎基静脉孔能否影响其周围的骨小梁分布。将腰椎标本制作为体外爆裂性骨折模型，观察椎体后上缘骨折块的形态及影像学表现，进一步分析椎体后上缘骨折块与椎基静脉孔的解剖位置关系。结果RL／VL、RH／VH值接近1／2（分别为0．497±0．059、0．485±0．036），RW，VW值接近1／3（0．319±0．025），提示骨折块位于椎基静脉孔所在区域上方。Micro-CT结果显示椎基静脉孔所在区域MR2、MR5骨微结构参数（骨体积分数、结构模型指数、骨小梁数量）明显较同层其他区域薄弱，上层SR5相应参数较同层其他区域薄弱，提示椎基静脉孔可能改变其上方区域的骨小梁分布而使强度降低。体外模拟爆裂骨折模型显示骨折线经过椎基静脉孔的顶角或上缘，椎基静脉孔的上缘为骨折块的底面，后纵韧带损伤的位置主要邻近椎基静脉孔边缘。结论作为骨质缺损区，椎基静脉孔是正常椎体最薄弱的区域，且影响其周边区域的骨小梁分布。?

英文摘要：

Objective To clarify the relationship between the basivertebral foramen （BF） and the retropulsed bone frag- ment （RBF） in thoracolumbar burst fracture （TLBF） and further explain the mechanism of RBF formation. Methods From June 2013 to June 2016, Sixty-two patients suffering from TLBF with RBF were collected. The characteristics of RBF as well as the pa- rameters of vertebral body were studied using CT reconstruction imaging. In the transverse images, the lengths of RBF （RL） and vertebral body （VL） were measured. In median sagittal images, the heights and widths of RBF （RH, RW） and vertebral body （VII, VW） were also obtained. The ratios of different parameters of RBF and vertebral body （RL/VL, RW/VW, RI-I/VH） were calculated, and then defined the location relationship of RBF and BF. Eight frozen cadaveric spine were selected and evaluated by Micro-CT scans. Each vertebral body was divided into three layers （Superior, Middle, Inferior）. Each layer was further divided into 9 regions （R1-R9）, named as SR1-SR9, MR1-MR9, IRI-IR9. Microarchitecture parameters of each region in each layer, including bone volume fraction （BV/TV）, bone mineral density （BMD）, trabecular connectivity （Conn.D）, and trabecular number （Tb.N） and thickness （Tb.Th） were calculated, and their differences were also analyzed to see if the trabecular bone distribution would be affected by BF. In vitro study, burst fractures were simulated on cadaveric spines by using bursting fracture simulator, aiming to observe the RBF morphology and imaging findings to future investigate the relationship between RBF and BF. Results The length and height of RBF were close to half of vertebral body length and height （RL/VL： 0.497±0.059, RH/VH： 0.485±0.036）. The width of RBF was usually one-third of vertebral body width （RW/VW： 0.319±0.025）, which indicated that the fracture block was often located in the posterior of vertebral body above the BF. BV/TV, Tb.N in the MR2 and MR5 regions were