| BackgroundThe retropulsed bone fragment (RBF) is the characteristic of the thoracolumbar burst fracture (TLBF) in axial CT images. It always comes from the posterior superior vertebral body. It is very important to evaluate the neurologic deficit, canal compromise, stability or instability in TLBF, and posterior longitudinal ligament injury. Also, it is one of the indications for the surgery and making a choice for surgical approach in TLBF. The mechanism of the RBF is not absolutely clear. Various studies in the literature have dealt with the issue. Under the axial impact, the endplate in the central portion is failure, allowing the nucleus material to enter the vertebral body, thereby pressurizing it more, squeezing the fat and marrow contents of the vertebral body out of the cancellous bone. When nucleus material enters the vertebral body faster than fat and marrow being expulsed, burst fracture is said to occur. At the same time, the shear force at the facet joints is transmitted through the pedicles to the posterior half of the vertebral body, creating a wedging effect, which dislodges and displaces the bony fragments. Additionally, the biomechanical study of the endplates has demonstrated that comparing with the caudal endplate, the cranial endplate is preferential failure under the same stress. This finding have accounted for the phenomenon that the RBF always originates from the posterior superior vertebral body. However, there is a same defect in these theories, that is an obvious bone default (Basivertebral foramen, BF) in the posterior vertebral body has been ignored. Nowadays, Chinese anatomy textbooks do not even describe this special anatomic structure, and English anatomy textbooks only provided a brief description of the approximate location. The morphological features of the BF and its relevance to the RBF in TLBF are not still clear.Objective1. to observe the morphological features of the BF in T12, L1, L2 and L3,including BF weight (BFW), BF depth (BFD), BF high (BFH), BF relative to the body weight (BFWr), BF relative to the body depth (BFDr), BF relative to the body high (BFHr). the location in the vertebral body, the difference of BF in different levels.2. To certify the relationship between the BF and the RBF in TLBF by clinical and experiment study.Methods1. A total of 36 health adults were underwent multi-slice CT thin slice scans and three-dimensional reconstruction for T12, L1, L2 and L3. In the horizontal and sagittal CT reconstruction images, the BFW, BFD, BFH, BFWr, BFDr and BFHr were measured. The distance between the BF and each side in the horizontal and sagittal CT images were also measured, to investigate the precise location in the body. The correlation between the measured parameters of BF and the gender, age, body mass index (BMI) were analyzed statistically.2.In the midsagittal reconstruction images of 36 health adults (T12, L1, L2 and L3), The BFSL (the distance between the posterior superior corner of BF and posterior corner of cranial endplate) and BFSW (the distance between posterior corner of cranial endplate and perpendicular from the end of BF to the cranial endplate) were measured. The difference between the average of BFSL in upper and lower level and BFSL of middle level was analyzed statistically. The BFSW was analyzed statistically by the same method. A total of 50 TLBF cases,29 cases in L1 and 21 cases in L2. All cases were underwent multi-slice CT thin slice scans and three-dimensional reconstruction. In the midsagittal reconstruction images, the parameters of BF in upper and lower adjacent fracture level were measured, including BFSL and BFSW. In the midsagittal reconstruction images, the parameters of RBF in fracture level were also measured, including RBFL (the length of RBF closed to the canal) and RBFW (the distance between posterior corner of cranial endplate and perpendicular from the anterior inferior corner of RBF to the cranial endplate). The difference between the average of BFSL in upper and lower adjacent fracture level and RBFL was analyzed statistically. Also, the difference between the average of BFSW in upper and lower adjacent fracture level and RBFW was analyzed statistically. By the statistical analysis, the relationship between the BF and RBF was studied.3. A total of 9 frozen samples of thoracolumbar motion segment (T12-L2). An incremental trauma model was applied to reproduce the L1 burst fracture. Before and after fracture, the multi-slice CT thin slice scans and three-dimensional reconstruction was undergone in all samples. The BFSL and BFSW of L1 were measured before fracture, and the RBFL and RBFW of L1 were measured after fracture. The difference between the BFSL and RBFL was analyzed statistically, to investigate the relationship between the BF and RBF. The difference between the BFSW and RBFW was also analyzed statistically for the same purpose. The samples were cut through midsagittal plane to observe the relationship between the BF and RBF by eye.Results1.In the horizontal and sagittal reconstruction images,81% BF was triangular or trapezoid. There was a bone interval within 6% BF. In the same level in 36 cases, the maximum of BFW was approximately 1/3 vertebral body width, which was measured in the same plane as BFW. The maximum of BFD and BFH respectively were also approximately 1/3 vertebral body depth and high. The average BFW, BFD and BFH were approximately 25% vertebral body in the same plane. In T12 and L3, the distance from the BF to right vertebral border was no significant difference than the distance from the BF to left vertebral border (P> 0.05). In the L1 and L2, however, there was statistical difference (P<0.05). In all four levels, the distance from the BF to the cranial endplate was shorter than to the caudal endplate (P<0.05). The mean BFW, BFH. BFWr, BFHr in different levels were no significant difference (P> 0.05). The mean BFD and BFDr in L3, however, were greater than other levels. There was not a correlation between gender and BFW, BFWr, BFD, BFDr, BFH, BFHr. There was a correlation between the age and L1BFHr and L2BFHr,and a correlation between BMI and L1BFH, L2BFW,L3BFH and L3BFHr.2.In all 36 health adults, BFSL and BFSW were measured and statistically analyzed. The mean of BFSL of the upper and lower levels was no significant difference than the BFSL of the middle level (P> 0.05). The mean of BFSW of the upper and lower levels was also no significant difference than the middle level (P> 0.05).In all L1 and L2 TLBF cases, there was no significant difference between the mean of BFSL of upper and lower levels and RBFL in fracture level (P> 0.05). There was also no significant difference between the mean of BFSW of upper and lower levels and RBFW in L1 (P> 0.05). However, the mean of BFSW of upper and lower levels was greater than RBFW in L2 (P<0.05).3. Experimental fracture was reproduced successfully by incremental trauma model in 8 frozen samples. The number of impact was 2-6 times, the average was 3.75 times. By comparing the midsagittal CT images pre- and post-fracture, the RBF was directly related with the BF, and the fracture line of RBF passed through the top border BF. There was no significant difference between the BFSL pre-fracture and the RBFL post-fracture (P>0.05). The BFSW pre-fracture was greater than the RBFW post-fracture (P<0.05). Posterior longitudinal ligament was torn variously in 4 samples. The fracture of cranial endplate was observed in all 8 samples, and only 2 samples showed the fracture in caudal endplate. The samples were cut through midsagittal plane and revealed that the nucleus material or cartilage endplate entered the vertebral body and the fracture line of RBF passed through the top border of BF.Conclusions1. There was a BF in all 144 levels in 36 health adults. In the horizontal and sagittal CT images, the most common appearance of BF was triangular or trapezoid. There was a bone interval within a few BF. The location of BF in T12 and L3 was more central than L1 and L2 in the horizontal CT images, and the BF was closer to the cranial endplate in all four levels in the sagittal CT images. In different individuals, the physical parameters of BF were quite different. The maximum of BFW,BFD and BFH respectively were approximately 1/3 vertebral body width, depth and high. The average BFW, BFD and BFH respectively were approximately 25% vertebral body in the same plane. The mean BFW, BFH, BFWr and BFHr in different levels were no significant difference. The mean BFD and BFDr in L3, however, were greater than other levels. There was not a correlation between gender and BFW, BFWr, BFD, BFDr, BFH, BFHr.2. In health adults, the mean of BFSL of the upper and lower levels was no significant difference than the BFSL of the middle level. The mean of BFSW of the upper and lower levels was no significant difference than the BFSW of the middle level. In 50 TLBF cases, the RBF was closely related to the BF.3. The RBF was closely related to the BF. The incremental trauma model was an effective method to reproduce the experimental fracture, which meet the purpose of this study.
|
PDF Full Text Request