New Techniques and MIS: The Cortical Bone Trajectory Screws—Indications and Limits

Fig. 10.1

Comparison between the traditional trajectory and cortical bone trajectory. The entry point of CBT requires less soft tissue dissection compared with that of TT. (used with permission from Medtronic Inc. (Memphis, TN, USA))

Additionally, screw insertion from a more medial and caudal entry point allows us to minimize the procedure-related morbidity: (1) minimizes paraspinal muscles dissection and retraction [10–12], (2) lessens iatrogenic facet joint injury supradjacent to the fused segment [13], and (3) avoids injury to the posteromedial branch of the nerve root passing near the mammillary process [14]. The direction away from the neural elements leads to a lower risk of neurologic injury, contributing to reduce the surgeon’s stress in clinical practice. Thus, this technique has attracted attention as a new minimally invasive method for lumbar spinal fusion, especially in osteoporotic patients. The purpose of this paper is to describe indications, limits, and pitfalls of the CBT technique from biomechanical and clinical standpoints.

10.2 Indications

10.2.1 Pathology

The CBT technique is indicated for almost all spinal disorders without spondylolysis and severe spinal deformity. It is best suited for short segment fusion as a minimally invasive method but can be effectively adapted for long-segment fusion due to the anchoring ability to achieve rigid spinal fixation.

This technique has a biomechanical advantage as an alternative to the traditional fixation technique, particularly in poorly trabeculated osteoporotic bone. Osteoporosis is characterized by decreased bone strength and is becoming more frequent with the aging of the population. Despite several efforts to enhance the strength of the bone-screw interface in the osteoporotic spine, problems of screw loosening, which may lead to the loss of correction and nonunion, have not been resolved. Santoni et al. first reported the superiority of CBT screws in osteoporotic cadaveric lumbar spines (mean age, 80.8 years) [1]. According to their report, CBT screws (average 4.5-mm diameter and 29-mm length) demonstrated a 30% greater uniaxial pullout strength and equivalent characteristics against toggle loading than TT screws (average 6.5-mm diameter and 51-mm length). Another author also compared the fixation strength of the screw-rod construct using CBT with that using TT under physiological cyclic loading in the osteoporotic lumbar spine [15] and showed increased fixation of the CBT construct in the lower lumbar spine.

One of the noteworthy indications of CBT is as a salvage procedure. Salvaging pedicle screws is necessary in cases of errors in screw placement, poor fixation resulting in screw loosening, and pseudoarthrosis, and increasing the diameter and length of screws and augmenting the screws with cement are conventional methods to improve the integrity of the bone-screw interface. CBT, which follows a different screw path from the traditional anatomical pedicular trajectory, can be another valid option for fixation failure (Fig. 10.2). Supporting this, Calvert et al. conducted a cadaveric study to investigate the fixation of CBT and TT screws when each was used to rescue the other in the setting of a compromised screw track (revision surgery) [16]. They concluded that CBT and TT screws each retain adequate construct stiffness and pullout strength for revision at the same lumbar level.

Fig. 10.2

Comparison of sagittal and axial direction between the traditional trajectory and cortical bone trajectory. CBT is directed 25–30° cranially in the sagittal plane and 10° laterally in the sagittal plane. There is a tendency in the axial plane to gradually increase the difference between medial angle of the traditional trajectory and lateral angle of cortical bone trajectory from L1 to L5. TT traditional trajectory, CBT cortical bone trajectory

Contrary to TT screws which achieve 60–80% of the fixation strength within the pedicle [17], CBT screws utilize the anatomically denser bone of the posterior elements to enhance fixation (Fig. 10.3). This feature indicates the possibility of CBT application for lumbar pathology with compromised vertebral bodies, such as vertebral spondylitis, compression fracture, and combination with vertebroplasty [18].

Fig. 10.3

Finite element models showing distribution of the equivalent stress on the vertebra. Illustration of a 54-year-old woman with loading of the pullout strength on the sagittal plane (upper row) and axial plane (bottom row). The various colors indicate the magnitude of the stresses. Moderate stresses occur in the pedicle for the traditional trajectory. CBT demonstrates higher stress concentrations within the posterior element

10.2.2 Spinal Level

The CBT technique is applicable for the thoracolumbar spine from T9 to S1 and can be used for multilevel fusion without severe deformity requiring correction. Care should be taken when placing screws in the upper lumbar vertebrae. Because of a narrow and medialized pars and small pedicle [2], there is a potential risk of pars and pedicle fracture, leading to fixation failure.

For the lower thoracic vertebrae, the thoracic CBT technique has been reported [19]. The trajectory starts at the 6 o’clock position of the pedicle and is directed straight forward in the transverse plane and ends up at the posterior third of the superior vertebral end plate. A cadaveric biomechanical study revealed that thoracic CBT screws demonstrated a 54% higher insertional torque than the traditional pedicle screws by contacting with the anatomically denser bone regions of the pedicle and vertebral body. Surgeons should note the following fact that, compared with thoracic CBT, lumbar CBT can involve a high level of purchase in the cortical bone with the posterior element and contribute to superior fixation with respect to the anatomical variations in bone density in the vertebrae.

In an attempt to engage with the sacral cortical bone, penetrating the S1 superior end plate screw (PES) technique has been introduced [20]. The trajectory starts at more medial points than the traditional sacral pedicle screw and penetrates the densest sacral bone of the proximal end plate without any risk of neurovascular injury. This technique demonstrated a 141% higher insertional torque than the traditional monocortical technique. One advantage of the PES technique is that the PES anchors are collinear with CBT screws and an S2 alar iliac screw when utilizing the CBT technique for multilevel fusion to the sacrum.

At the thoracolumbar spinal level, CBT is superior to TT with regard to both the biomechanical aspect and less surgical invasiveness, particularly in the lower lumbar spine. From a biomechanical aspect, the lower spine has a larger pedicle than the upper spine and involves difficulty in obtaining optimal filling within the subcortical bone of the pedicle when using TT [15]. In contrast, CBT screws can achieve rigid fixation, regardless of the pedicle size, by increasing contact with the abundant cortical bone between the pars interarticularis and inferior part of the pedicle in the whole lumbar spine [2, 21–23]. In terms of less invasiveness, the anatomical characteristics of the lower spine, such as the deep-seated screw entry point, large amount of paraspinal muscles, and larger medialized pedicle axis, necessitate extensive muscle dissection and traction to implant a pedicle screw in a convergent direction from a more lateral entry point (Fig. 10.2). This tendency is more marked in the case of obese patients with a deep surgical corridor.

10.3 Future Advantage to Prevent Fusion Disease

Several clinical studies have assessed surgical outcomes using CBT. In the short-term results, the CBT technique facilitates similar clinical and radiologic outcomes with low surgical morbidity and blood loss and a short postoperative hospital stay compared with the TT technique [11, 24]. More interestingly, a recent study of 177 patients who underwent posterior lumbar interbody fusion for degenerative lumbar spondylolisthesis (82 controls by the TT technique; 95 by the CBT technique) reported a 3.2% rate of symptomatic adjacent segment disease (ASD) using the CBT technique and a 11% rate using the TT technique during a 3-year postoperative follow-up (p < 0.05) [25]. The development of ASD is one of the major undesirable factors and an inevitable complication after lumbar spinal fusion, which requires further surgical treatment. Spinal fusion alters the biomechanical properties on the nonoperated adjacent segment and shows symptomatic degeneration. However, the essence to achieve spinal fusion is the same even though a diversity of minimally invasive surgical techniques (including percutaneous pedicle screw, CBT, lateral interbody fusion, etc.) are applied in clinical practice. From this point of view, reducing the incidence of ASD is a meaningful advantage of using the CBT technique, probably due to reducing the posterior soft tissue dissection and avoiding iatrogenic injury to the cranial facet joint during placement of the proximal pedicle screws (Fig. 10.4) [13, 26, 27]. In the future, CBT may become a standard procedure and be used as a “reducing ASD fusion technique” regardless of the bone density or generation.

Fig. 10.4

Adjacent cranial facet violation following screw insertion. CT scans show adjacent cranial facet violation (left) and no violation (right). The selection of the optimal entry point sufficiently caudal from the inferior border of the facet joint and leaving the screw proud about 5 mm from the dorsal lamina to maintain safe distance from the facet joint are essential to reduce the risk of violation

10.4 Limitations

10.4.1 Spondylolysis

There is a definite limitation on indicating the CBT technique for spondylolytic vertebrae because the pars interarticularis is a key structure for the fixation of CBT [28]. Spondylolysis is defined as an anatomical defect in the pars interarticularis that allows separation of the vertebral body from the lamina and inferior facet joint, leading to anterior translation of the affected vertebra. The traditional pedicle screw system is the gold standard for the surgical treatment of isthmic spondylolisthesis due to a strong vertebral anchoring capacity, and it has led to favorable clinical outcomes. If CBT is applied to isthmic spondylolisthesis, one could easily insert screws along a craniolaterally directed path because the entry point of CBT can be effectively confirmed under direct visualization after the complete removal of the floating lamina.

From an anatomical aspect, the cortical bone is most concentrated between the pars interarticularis region and inferior part of the pedicle [21–23]; however, a spondylolytic vertebra lacks these regions on which CBT screws rely for most of their stability (Fig. 10.5). A biomechanical study on the fixation strength of pedicle screws in spondylolytic vertebrae revealed that CBT screws provided similar pullout strength to TT screws, but CBT constructs showed a significantly lower vertebral fixation strength compared with TT constructs [28]. The absence of a solid purchase in the cortical bone in the posterior lamina, in spite of penetrating the sclerotic surface at the pars defect, and the divergent trajectory of CBT screws were suggested to be the causes of this drawback. Surgeons should note that (1) the fixation strength of the construct is a critical factor to achieve better bony fixation rather than that of a single screw itself, and (2) the TT technique is superior for spondylolytic vertebrae to the CBT technique, even though the latter can reduce muscle dissection.

Fig. 10.5

Finite element models showing distribution of the equivalent stress on the spondylolytic vertebra. Illustration of a 54-year-old woman with loading of the pullout strength on the sagittal plane (upper row) and axial plane (bottom row). The various colors indicate the magnitude of the stresses. Both TT and CBT screws demonstrate similar moderate stresses in the pedicle. CBT lacks higher stress concentration at the posterior element compared with that in the normal vertebra (Fig. 10.3)

Similarly, relative contraindications include a lack of pars conditions secondary to a wide decompression and an iatrogenic pars fracture [29]. When applying the CBT technique to a pars defect with careful pre- and intraoperative considerations, special attention must be taken to select large-sized screws and to place screws sufficiently deep into the vertebral body for better fixation and effective load sharing [30, 31].

10.4.2 Rotational Spinal Deformity

Some biomechanical studies suggested that CBT screws provided less rigid fixation on lateral bending and axial rotation. A cadaveric biomechanical study showed that the CBT screw-rod construct exhibited almost the same stability as the TT construct; however, when the intervertebral disc was left intact or a transforaminal lumbar interbody fusion implant was used, the TT construct was significantly stiffer than the CBT construct during lateral bending and axial rotation [7]. Another biomechanical study using the finite element method revealed that the CBT paired-screw construct showed significantly higher vertebral fixation on flexion (51%) and extension (35%) and a significantly lower vertebral fixation strength on lateral bending (20%) and axial rotation (37%) compared with the TT construct, although a single CBT screw demonstrated a significantly higher pullout strength and higher resistance to multidirectional loading than a TT screw [3]. The weaknesses on lateral bending and axial rotation using CBT were universally observed regardless of the BMD; therefore, the disadvantage of the CBT construct (the features of a divergent and short lever arm from the medial axis) could be associated with these results.

On the basis of these biomechanical results, two essential points are recommended. One point is enhancing each screw’s fixation within a construct. The selection of the optimal screw path and screw size, which varies widely among surgeons, is important to obtain the best fixation strength. According to previous studies investigating these issues, the optimal screw path should be directed 25–30° cranially along the inferior border of the pedicle to achieve maximum contact with the hard bone of the lamina and end up around the posterior third to posterior half of the superior vertebral end plate to distribute the loads applied to the vertebra effectively (Fig. 10.6) [32]. A finite element study using L4 vertebrae showed that the ideal screw size for CBT is a diameter larger than 5.5 mm and length longer than 35 mm so as to be placed sufficiently deep into the middle column of the vertebra [33]. We always select screws of 5.5 mm in diameter and 35–40 mm in length for the middle or lower lumbar spine.