The superior stability of the hybrid technique with MCBT and TT at the L4-L5 segment without fusion technique was previously demonstrated by our team [9]. However, patients with lumbar spinal stenosis were more common in clinical practice and often required thorough decompression and fusion, thus the L1-S1 lumbar spine models were established and TLIF procedures were performed at the L4-L5 segment. The selection of the fixation type is correlated with the occurrence of postoperative complications such as screw loosening and breakage, especially in patients with osteoporosis. As known, the CBT technique increased the pullout load by 30%, and torque by 1.7 times [5]. As previously shown by our team that the MCBT technique provides superior biomechanical stability compared to the pedicle screw technique and the CBT technique [6, 7]. Fujiwara et al. [23] demonstrated that increasing the medio-lateral angle and decreasing the distance to the anterior lateral edge of the upper endplate increases the insertional torque of the screw. The MCBT technique compensates for the deficiencies of the CBT technique. However, we also found that the MCBT technique still has deficiencies, such as limited decompression of the lateral recess and intervertebral foramen, splitting or fracture of the screw insertional point during decompression, and pars fracture are contraindications to the CBT technique [24]. If the BMCS-BMCS was implanted first and then proceed with the decompression, the decompression of the lateral recess might be incomplete. While the decompression proceeded first the screw insertion point might be damaged. It is advisable to use hybrid BMCS-BPS and BPS-BMCS fixation techniques according to the patient's preoperative radiographic data.
In this study, the stability of the BMCS-BMCS group in flexion and lateral bending, and the BPS-BMCS group in extension and rotation were superior to that of the BPS-BPS and BMCS-BPS groups. Perez-Orribo et al. [25] demonstrated that the stability of the CBT technique was superior in flexion and extension, but inferior in lateral bending and rotation and the differences between this study and the currenct study may be due to the greater MCBT screw diameter of 5.0 mm and length of 40 mm compared to the CBT screw with a diameter of 4.5 mm and length ranging from 25 to 35 mm in Perez-Orribo et al. [25] and contact between the MCBT screw and the cortical bone of the lateral edge of the upper endplate and medial wall of the pedicle. BPS-BMCS group offered superior stability in extension and rotation, but in flexion and lateral bending. Huang et al. [26] demonstrated that the inter-screw angle of 15° between the screw at the cranial and caudal level resulted in the lowest L4-L5 segmental ROM in flexion, extension, and rotation. In this study, the inter-screw angle was 8° in the BPS-BMCS group and 33° in the BMCS-BPS group. The morphology of the L5 vertebral arch is unique in that the superior edge of the vertebral body is almost at the same level as the superior edge of the vertebral arch, but the distance between the superior edge of the L4 vertebral arch and the vertebral body is somehow different from that of the L5 vertebral body, which has an "/\" pattern and moves more obliquely to the left and right, making the cranio-caudal angle of the screw in the sagittal plane more constant and milder than that of the L4 vertebral body. Therefore, when using the MCBT technique in the L5 vertebral body, the cranio-caudel angle should be reduced (around 25°) to keep the screw from penetrating the upper endplate of the vertebral body, while the L4 vertebral body has a larger cranio-caudel angle (22°- 35°) [6]. Although the inter-screw angle at a sagittal plane in the BPS-BMCS group was different from that of Huang et al. [26], the variation rule of ROM in extension and rotation was similar. Except for the inter-screw angle at the sagittal plane, which in the axial plane was also different.
The peak von Mises stress of the cage was lower in the BPS-BMCS group and BMCS-BMCS group than in the BPS-BPS group and BMCS-BMCS group in extension and rotation. In flexion and lateral bending, the cage stress of the BMCS-BMCS group was the greatest, The greater the cage stress, the higher the subsidence rate [27]. Numerous scholars have concluded that different types of fixation were essential to maintain the stability of the surgical segment and to reduce cage subsidence [28], but there is still no consensus regarding the ideal instruments. Different fixation methods have been reported previously, but the differences among the BPS-BPS, BMCS-BMCS, BPS-BMCS, and BMCS-BPS have not been reported. The previous study showed that the BMCS-BPS group had the lowest intervertebral disc stress in rotation. The BMCS-BPS group in the current study showed the least cage stress (51.7 ± 23.3 MPa) in rotation, which was comparable to the tendency in the previous model without fusion technique [9]. In the biomechanical analysis of Xu et al. [28], the largest von Mises stress was found in lateral bending, and the peak von mises stress of the single-cage (77.23 MPa) was greater than paired-cage (49.77 MPa) in TLIF model with BPS-BPS technique [29]. In this study, the peak von Mises stress was found in lateral bending in the BMCS-BMCS group (72.66 ± 12.4 MPa), but in the BPS-BPS group, The peak von Mises stress was found in extension (66.27 ± 18.85 MPa), variation rule of BPS-BPS group was inconsistent with Xu et al. [28]. This may be due to the different screw diameters, lengths, and cage sizes. The mean cage stress in the BMCS-BPS group was lowest in extension and rotation and lower in flexion and lateral bending, but did not reach a statistical difference (p > 0.05). In the RCT (randomized control trial) of Lee et al. [29], the fusion rates of CBT and pedicle screw technique were comparable (CBT 94.5% versus PS 91.4%, p > 0.05) at 24 months postoperatively, but the MCBT technique has not been used in the clinic. There was no statistical difference in the stability between the single and paired cage in the TLIF model [11]. Single cage implantation was preferable because of its convenience, and lower cost. As for the hybrid fixation with the BMCS and BPS, the different effects of single and paired cages on the stability of the fixation and fusion rate need further biomechanical study.
Increasing the medio-lateral angle of the pedicle screw reduced the risk of screw loosening and breakage [30]. Peak von mises stress of the screw in the hybrid fixation technique with CBT and pedicle screw, in Su et al. [31], were lower than that of the BPS-BPS technique in flexion, extension, and lateral bending. In this study, As for the BPS-BMCS group, the von Mises stress of the screw was lower than those of the BPS-BPS group in motion. As for the BMCS-BPS group, the von Mises stress of the screw was lower than that in the BPS-BPS group in flexion, extension, and rotation and the stress variation rules were comparable to those of Su et al. [31], BPS-BMCS group and BMCS-BPS group may reduce the incidence of screw breakage. The lowest screw stress was found in the BPS-BMCS (30.12 ± 5.73 MPa) and BMCS-BPS (32.79 ± 4.63 MPa) group in flexion. As for BCS-BCS (CBT at the L3 and L4), the von mises stress of the screw was greater than the counterparts in flexion and rotation [31], However, the opposite result was found in the present study, which may be related to the greater medio-lateral angle of the MCBT technique than that of the CBT technique as mentioned by Newcomb et al. [30]. Huang et al. [26] suggest that the longer rod length may decrease the screw stress. In this study, the BMCS-BPS group had the shortest rod length and the highest screw stress in lateral bending (200.34 ± 91.11 MPa) and rotation (193.35 ± 68.09 MPa), while the BPS-BMCS group had the longest rod length and the lowest von mises stress of the screw stress in flexion (60.81 ± 23.70 MPa) and rotation (120.37 ± 21.98 MPa) (p > 0.05). Results only in Rotation were consistent with Newcomb et al. [30], The reasons were probably due to the different sample sizes, different material properties, and inconsistent screw and rod sizes in the two studies. As shown in Fig. 8, the maximum stress in the screw occurs near the screw hub, that is, the BPS-BMCS group may reduce the risk of screw breakage.
Commonly, the junction of the screw and the connecting rod is the most common site for breakage. In this study, the von Mises stress of the rod in the BMCS-BMCS group was greater in all motions than its counterparts, with the maximum stress occurring in lateral bending (119.35 ± 43.20 MPa) at the lower end of the junction between the screw and the connecting rod (Fig. 8). The rod stress of the BMCS-BPS group was lower than that of the counterparts in extension and rotation, and the BPS-BMCS group also showed lower rod stress than the counterparts in flexion, and approximately similar to that of the BPS-BPS group in lateral bending. It can be deduced that the BPS-BMCS and BMCS-BPS groups may reduce the risk of rod breakage. Wang et al. [32] showed that the CBT-CBT group (CBT screw: 5.0 mm in diameter, 35 mm in length) had the highest rod stress in almost all working conditions while the UP-CBT group (TT screw: 6.5 mm in diameter, 45 mm in length), similar with the BMCS-BPS group, had the lowest rod stress in all motions. Although the length of the MCBT screw in this study was slightly longer and the TT screw was slightly shorter, the variation rules were consistent with the Wang et al. study [32]. Xiao et al. [33] demonstrated that the Dynesys hybrid fixation technique combined the advantages of dynamic stabilization and rigid fusion, using a "soft" rod in the upper segment to reduce the incidence of ASD (adjacent segment degeneration), further manifesting the principle of "overcoming rigidity with flexibility". By analogy, the effect on the adjacent segment of the "rigid" fixed segment using the hybrid fixation techniques discussed in this study and connecting with the "soft" rod needs to be further investigated.
Lumbar musculature plays an important role in spinal stability [34]. A previous study showed that S1 screw loosening was closely related to the degeneration of paraspinal muscles [35]. CBT screw reduces the postoperative serum creatinine phosphokinase concentration [36]. Since the insertion point of the MCBT technique is closer to the midline than the CBT technique [6, 7], the hybrid technique can further reduce the damage to the facet joint [9] and iatrogenic paravertebral muscle injury. Compared with the traditional BPS-BPS and CBT-CBT fixation techniques, the hybrid BPS-BMCS and BMCS-BPS techniques can not only overcome the embarrassing situation of incomplete decompression of the lateral recess after implantation of the screw but also reduce the surgical incision to reflect the principle of "minimally invasive surgery" and avoid the degeneration of the facet joint.
Perez-Orribo et al. [25] demonstrated that cortical screw systems offered greater rigidity in flexion and extension. However, it is worth noting that 20% of CBT screws may cause medial pedicle wall fractures resulting in nerve injury [5]. The medio-lateral angle of the MCBT technique is greater than that of the CBT technique, and it is extraordinarily difficult to complete the high precision screw placement with freehand in high accuracy, therefore, a 3D (3-dimensional) printed template or a robotic technique is required to assist in the nail placement process. Our research team has previously investigated the 3-dimensional printed template of MCBT techniques with in vitro experiments and demonstrated that it improves the accuracy of the MCBT screw placement [37].
There were some limitations in this study. First, the muscle tissues were not reconstructed, which may affect the stability of the lumbar spine. Second, this study did not analyze the effect of the hybrid technique on the adjacent segments in the TLIF model. Third, screw sizes of different diameters and lengths were not discussed in this study. Fourth, the sample size of this study is only three which needs to be increased in further study.