Biomechanical Evaluation of Cross Trajectory Technique for Pedicle Screw Insertion: Combined Use of Traditional Trajectory and Cortical Bone Trajectory

Abstract

Objective

To introduce a novel double‐screw (cross trajectory) technique that combines use of the traditional trajectory (TT) and cortical bone trajectory (CBT) and to investigate its fixation strength quantitatively by finite element (FE) analysis.

Methods

Three‐dimensional FE models of 30 osteoporotic L₄ vertebrae (patients' mean age: 77.3 ± 7.4 years, 11 men and 19 women) were computationally created. Each vertebral model was implanted with bilateral pedicle screws by TT (using 7.5 mm × 40 mm screws), CBT (using 5.5 mm × 35 mm screws) and cross trajectory (combined use of TT screws of 5.5 mm × 40 mm and CBT screws of 5.5 mm × 35 mm) and compared among three groups. The vertebral fixation strength of a bilateral‐screw construct was examined by applying forces simulating flexion, extension, lateral bending, and axial rotation to the vertebrae by non‐linear FE analyses.

Results

Fixation strength using the cross trajectory was the highest among the three different techniques (P < 0.01). The cross trajectory construct demonstrated 320% higher strength than the TT construct in flexion, 293% higher in extension, 102% higher in lateral bending, and 40% higher in axial rotation (P < 0.01). Similarly, the cross trajectory construct showed 268% higher strength than the CBT construct in flexion, 269% higher in extension, 210% higher in lateral bending, and 178% in axial rotation (P < 0.01).

Conclusions

The cross trajectory technique offered superior fixation strength over the TT and CBT techniques in each plane of motion. This technique may be a valid option for posterior fusion, especially in osteoporotic spine.

Introduction

Various spinal disorders accompanied by osteoporosis are increasing with the aging of the population. Osteoporosis is characterized by decreased bone strength and susceptibility to fracture, and has a great impact on the progress of spinal deformities and potential neurological compromise1. The pedicle screw fixation technique has become the gold standard for spinal fusion, which contributes to rigid fixation, a high fusion rate, and early mobilization. Despite the advantage of superior biomechanical capability using pedicle screws, problems of screw loosening, which may lead to the loss of correction and nonunion, have not been resolved due to the fragile characteristics of the bone2.

Over the past few decades, a considerable number of studies have been performed on the fixation strength of pedicle screws and clarified factors affecting the anchoring ability, such as bone quality3, the design and size of the pedicle screw3, 4, 5, and the insertion technique6, 7, 8. To enhance the integrity of the bone–screw interface within osteoporotic vertebra, augmenting the screw with cement has been explored. Although polymethylmethacrylate (PMMA), which is a frequently used cement, immediately increases the fixation of the screw9, it is an exothermic polymer that may cause thermal bony necrosis and neural injury. Other concerns in using PMMA are the potential risk of embolic events and the difficulty of removing it under unavoidable circumstances10. Regarding factors related to the screw's mechanical properties, previous experimental studies by Zindrick et al. suggested that increasing the diameter and length of the screw improves the performance of the fixation; however, the screw size is limited by the anatomical background and there may be a risk of neurological or vascular injury due to its perforation of the pedicle and penetration of the anterior wall of the vertebra5. Moreover, several studies have found that larger screws had little or no effect on fixation strength in osteoporotic bone because of the thinner cortex of the pedicle11, 12. In terms of the insertional technique, two strategies for enhancing screw fixation can be considered. One involves the preparation techniques of the pilot hole before screw implantation. It has been recognized that a pilot hole size smaller than the core diameter of the screw and undertapping a pilot hole improves pullout strength6. Another technique is altering the transpedicular screw path from the traditional trajectory (TT). Several studies have demonstrated that variations of bone density within the vertebra play a major role in the biomechanical stiffness of the bone–screw interface7, 8. Recently, the cortical bone trajectory (CBT) technique has been developed as an attractive alternative for the lumbar transpedicular path8. The CBT follows a caudal to cephalad and medial to lateral directed path, which takes advantage of maximizing engagement with the cortical bone, providing enhanced screw purchase. Biomechanical studies have revealed equivalent or superior fixation strength using CBT in comparison with the TT8, 13, 14, 15, 16. However, achieving solid implant fixation to severely osteoporotic bone presents a challenge to spinal surgeons even when using these techniques. In this article, we propose a novel double‐screw technique with the combined use of TT and CBT, which we name the cross trajectory technique. The advantage of this technique is the creation of strong constructs by inserting two screws into an elliptical shape of a single pedicle using different transpedicular trajectories. The objectives of the present study were: (i) to introduce a novel double‐screw (cross trajectory) technique; (ii) to conduct a biomechanical evaluation of the cross trajectory technique by the finite element (FE) method; and (iii) to confirm its validity quantitatively.

Materials and Methods

Data Collection

Computed tomography (CT) scans of 30 osteoporotic patients who underwent surgery for degenerative spinal disorders from July 2012 to March 2015 were used. There were 11 men and 19 women, with a mean age of 77.3 ± 7.4 years (range, 67 to 88 years). Patients who were diagnosed with vertebral malformation or metastasis or had previously undergone surgery were excluded. In each case, a CT scan was taken with a slice thickness of 1 mm before the surgery. This study received ethics committee approval at our institution and all patients provided their signed informed consent.

Cross Trajectory Technique

The cross trajectory technique is as follows. Firstly, a CBT screw was placed in the optimal position. According to the previously described technique17, the starting point for CBT was located at the lateral aspect of the pars interarticularis projecting in the 5 o'clock orientation in the left pedicle and the 7 o'clock orientation in the right pedicle. The CBT was directed 10° laterally in the axial plane and 25° cranially in the sagittal plane along the inferior border of the pedicle. Then, the TT was made following Weinstein's technique18; however, its starting point was positioned slightly cranially to utilize the upper residual space in the pedicle along the superior border of the pedicle (Fig. 1).

Figure 1.

Illustration showing the cross trajectory. Upper: The cortical bone trajectory (red arrow) is angulated laterally and cranially along the inferior border of the pedicle (yellow circle). Lower: The entry point of the traditional trajectory (blue arrow) is slightly cranial to follow the superior border of the pedicle.

Finite Element Models

Three‐dimensional FE models of the L₄ vertebra were constructed from the patient's CT data individually using Mechanical Finder software (vers. 6.2, extended edition; Research Center of Computational Mechanics, Tokyo, Japan). Each model was divided into 0.5–1 mm tetrahedral solid elements with 150,000–200,000 nodes and 800,000–1,000,000 solid elements to reflect the smooth surface of the spinal bone using the automatic mesh function. The details of the FE model construction were described previously14. FE models of the pedicle screw (Solera Spinal System; Medtronic, Memphis, TN, USA) were developed separately from high resolution micro‐CT to obtain the detailed geometry, and were also divided into 0.5–1 mm tetrahedral elements. These screws were assumed to have the material properties of cobalt chromium alloy for the screw heads and titanium alloy for the screw shafts. The bone–screw interface was modeled using contact condition and the friction coefficient was determined to be zero, based on previous studies4, 14.

Then, each vertebra was implanted with bilateral pedicle screws. Three different techniques of screw insertion were compared: the TT technique (using screws of 7.5 mm in diameter and 40 mm in length), the CBT technique (using screws of 5.5 mm in diameter and 35 mm in length), and the cross trajectory technique (combined use of TT screws of 5.5 mm in diameter and 40 mm in length and CBT screws of 5.5 mm in diameter and 35 mm in length; Fig. 2). All screws were carefully placed in an appropriate position without any cortical breaching and penetration into the anterior vertebral cortex because these two conditions may potentially influence screw fixation values. The computer solution time per analysis ranged from 18 to 36 hours, and a total of 360 analyses using the L₄ vertebrae of 30 individuals were performed.

Figure 2.

Finite element models of L₄ vertebra using the traditional trajectory (left), the cortical bone trajectory (middle), and the cross trajectory (left; combined use of the traditional trajectory (blue screw) and the cortical bone trajectory (green screw).

Loading and Boundary Conditions

In each model, a non‐linear FE analysis was performed. The bilateral screw heads were rigidly fixed in conditions reported by Chen et al.19 An incremental loading rate of 20 N was gradually applied to the surface of the vertebral body to simulate flexion, extension, lateral bending, and axial rotation according to the previous testing protocols14 (Fig. 3). Displacement was obtained from the average movement of the whole vertebra. Under destructive loading, the ultimate failure loads were defined as the load at the inflexion point of the load‐displacement curve (Fig. 4). Vertebral fixation strength (N/mm) was defined as the slope of the line fitting the load‐displacement curve until the ultimate failure load. The vertebral strength was correlated with regional bone mineral density (BMD) of the center of the pedicle and the vertebral body measured by quantitative CT.

Figure 3.

Illustration of loading conditions for flexion, extension (green arrow), lateral bending, and axial rotation (yellow arrow).

Figure 4.

A typical load‐displacement curve demonstrating the ultimate failure load.

Statistical Analysis

Results are presented as mean ± standard deviation. Data were compared using repeated‐measures anova and Tukey's significant difference multiple comparison tests. Pearson's correlation coefficient was used to analyze how the vertebral fixation stiffness changed with BMD. We used JMP vers. 11 software (SAS Institute, Cary, NC, USA) for all analyses and significance was defined as P < 0.05.

Results

The ultimate failure load using the cross trajectory technique was the highest among the three different insertional techniques in each plane of movement (P < 0.05, Table 1). The cross trajectory construct demonstrated 320% higher strength than the TT construct in flexion (P < 0.01), 293% higher in extension (P < 0.01), 101% higher in lateral bending (P < 0.01), and 40% higher in axial rotation (P < 0.01, Fig. 5). Similarly, the cross trajectory construct showed 268% higher strength than CBT in flexion (P < 0.01), 269% higher in extension (P < 0.01), 210% higher in lateral bending (P < 0.01), and 178% higher in axial rotation (P < 0.01). The superiority of the cross trajectory construct in comparison with TT and CBT was more remarkable in flexion and extension loadings than in lateral bending and axial rotation. There were significant positive linear correlations between the vertebral fixation strength of each moment and BMD parameters among the three techniques (Table 2).

Table 1.

Summary of load to failure data

Index	Flexion			Extension			Lateral bending			Axial rotation
	TT	CBT	Cross	TT	CBT	Cross	TT	CBT	Cross	TT	CBT	Cross
Ultimate failure load (N)	412 ± 43	527 ± 82	818 ± 124	396 ± 47	480 ± 89	783 ± 98	580 ± 89	568 ± 74	761 ± 168	1807 ± 360	1007 ± 97	2105 ± 371
Displacement (mm)	0.69 ± 0.12	0.76 ± 0.11	0.31 ± 0.08	0.66 ± 0.10	0.74 ± 0.08	0.32 ± 0.12	0.33 ± 0.06	0.50 ± 0.09	0.23 ± 0.10	0.52 ± 0.13	0.52 ± 0.07	0.39 ± 0.12
Vertebral fixation strength (N/mm)	619 ± 108	706 ± 179	2602 ± 715	615 ± 121	655 ± 137	2417 ± 708	1802 ± 335	1166 ± 225	3624 ± 972	3979 ± 967	2006 ± 430	5589 ± 1167

All data given as means ± SD. CBT, cortical bone trajectory; cross, cross trajectory; TT, traditional trajectory.

Figure 5.

Comparisons of vertebral fixation strength among the three insertional techniques. Traditional trajectory (TT) construct strength was set to 100% for each type of motion. CBT, cortical bone trajectory; Cross, cross trajectory. (*P < 0.05, **P < 0.01).

Table 2.

Correlation coefficient between bone mineral density and vertebral fixation strength

Location	Flexion			Extension			Lateral bending			Axial rotation
	TT	CBT	Cross	TT	CBT	Cross	TT	CBT	Cross	TT	CBT	Cross
Pedicle	0.51*	0.47*	0.63*	0.52*	0.46*	0.55*	0.79*	0.52*	0.78*	0.47*	0.59*	0.54*
Vertebral body	0.49*	0.53*	0.62*	0.51*	0.55*	0.59*	0.68*	0.66*	0.61*	0.50*	0.66*	0.55*

*P < 0.05. CBT, cortical bone trajectory; cross, cross trajectory; TT, traditional trajectory.

Discussion

FE analysis is a reliable and helpful method to predict mechanical strength and dynamic characteristics by simulating effects acting on a structure, which has been increasingly used in the spinal field4, 14, 19. In cadaveric studies, wide individual variations of bone quality and age, the variability of the level tested, and an insufficient sample size may cause biased study results and lead to misunderstandings. In contrast, the FE method can minimize sample variation for a fair comparison, provide reproducibility, and reduce the expense and time spent on repeated mechanical tests.

Advantages of Cross Trajectory Technique

The most important advantage of the cross trajectory technique is achieving rigid fixation strength by combining two different trajectories to obtain triangulation effects in both sagittal and horizontal planes, which leads to solid arthrodesis and the maintenance of correction. Indications for using the cross trajectory technique are cases necessitating strong fixation, such as severe osteoporosis, unstable spinal fracture, obesity, neuromuscular conditions, and revision surgeries. The cross trajectory technique produced a greater biomechanical effect than the single trajectory technique, particularly in flexion/extension loadings, and seems to be effective in short‐segment spinal fusion for treating instability after an osteoporotic vertebral fracture (Fig. 6). From a morphological point of view, pedicle size is a critical determinant of its ability to accommodate this technique; however, the thoracic pedicle above T₉ is not wide enough to accept two screws20. The cross trajectory would be applicable until T₁₀, at least 6 mm in pedicle width and 14 mm in pedicle height.

Figure 6.

Representative case using the cross trajectory technique. A 78‐year‐old man undergoing L₃–L₅ short segmental spinal fixation with L₄ vertebroplasty for osteoporotic vertebral fracture. A and B: preoperative radiographs demonstrating the L₄ vertebral collapse. C and D: postoperative radiographs show that two screws using different trajectories are inserted into one pedicle and connected with the same rod (using 5.5 mm × 40 mm screws for traditional trajectory and 5.5 mm × 35 mm screws for cortical bone trajectory). E: postoperative computed tomography demonstrates that the screws are placed in the correct positions in L₃ and L₅.

In terms of the double‐screw technique, inserting two pedicle screws into one pedicle, Jiang et al. have performed a biomechanical test using osteoporotic cadaveric spine21. They compared the use of double 5‐mm‐diameter screws with a single 6‐mm‐diameter screw in the thoracolumbar spine, and demonstrated the mechanical superiority of double‐screw fixation. Rodriguez et al. reported a double‐screw technique using CBT for adjacent‐segment lumbar disease previously instrumented with TT screws22. They placed additional CBT screws in the residual transpedicular path of at least 5 mm in diameter under intraoperative CT image‐guided navigation. Similarly, Ueno et al. have reported a corrective fusion technique using double screws of both CBT and TT and achieved solid fixation without the loss of correction in an osteoporotic woman with degenerative lumbar scoliosis23. Good clinical results were demonstrated, but there are several points to be improved. They inserted CBT screws from a higher entry point than the original CBT technique with a lower caudocranial angle. CBT is not only a transpedicular trajectory directed laterally, but an exceptional trajectory engaging with denser cortical bone to the maximum extent8, 24.

Practical Points

What is important in the cross trajectory technique is to choose the ideal screw path for the optimal fixation of each trajectory as well as to insert two screws in one pedicle. A biomechanical study measuring the insertional torque of CBT screws demonstrated that CBT displayed higher fixation ability when inserted from the lower aspect of the pars interarticularis, which possesses dense cortical bone, and maximizing thread contact with the highest concentration cortical bone of the lamina22. In clinical practice, we usually use the CBT as the first screw and the TT as the second screw. A screw path of CBT that follows the inferior border of the pedicle is mandatory for the cross trajectory in terms of enhancing the strength of the bone‐screw interface and providing as large a residual space for TT as possible. Then, the placement of TT screws along the superior border of the pedicle offers several merits for their combined use with CBT screws. Taking the distance between two screw heads of TT and CBT provides an easier rod connection and allows a longer screw path. According to previous reports, regions of higher density bone are found adjacent to the vertebral endplate and screw placement near the superior endplate resists higher displacing forces25.

Another advantage of the cross trajectory technique is that additional screws can be applied to the opposite trajectory for necessary parts and conditions. This may become an augmentation strategy for the instrumentation of the osteoporotic spine to create multiple points of fixation within a vertebra similar to the commonly used technique of laminar hooks and sublaminar wires. When utilizing this option, the screw heads of different trajectories are not collinear in the coronal plane; however, two screws can be connected with the same rod by using offset connectors. On the basis of the fact that screw pullouts are often observed at the top or bottom levels of the spinal construct2, the cross trajectory is a reliable and useful technique to reinforce the fixation strength of these critical regions individually.

Disadvantages

There are some concerns with respect to clinical use of the cross trajectory technique. One is the economic issue, namely, that additional screws increase the total cost of instrumentation. The routine use of this technique should be avoided, considering the issue of cost‐effectiveness and the posterior space available for bone grafting. We recommend this technique for use in patients with a higher risk of screw loosening and fixation failure, which may result in subsequent revision surgery.

Another concern is that while the elliptical shape of thoracolumbar pedicle can accommodate two screws, however, the safe placement of both screws is technically demanding. There are potential risks of neural injury and pedicle fracture inducing loss of fixation. Intraoperative fluoroscopic or image guidance techniques are mandatory to improve accuracy and ensure appropriate bone purchase.

Limitations

There are some limitations in this study to be mentioned. First, this double screws technique demonstrated superior fixation compared with the single traditional screw technique. If there was enough bony space in the pedicle, larger screws could be placed by the single screw technique, resulting in higher strength. Misenheimer et al. suggested that the screw diameter should be less than 80% of the outer pedicle diameter to avoid plastic deformation of the pedicle and consequent loss of screw fixation26. In accordance with this suggestion, we selected a screw diameter of 7.5 mm, which corresponded to 77.2% ± 12.1% fit to the minor diameter of the outer pedicle cortex in the present study. Next, we used FE models of the single vertebral segment. A multiple‐segment and screw–rod construct model provides a better representation of the actual clinical situation; however, the inclusion of intervertebral elements requires the consideration of more material and the geometry of the components and may lead to more complex results. We believe that the present models are useful to evaluate the straightforward holding power of inserted screws. Lastly, the loading conditions do not perfectly replicate the fixation failure in vivo, as FE models have difficulty testing the effects of cyclic loading. Law et al. reported the importance of repeated loading in the mechanism of pedicle screw loosening27; therefore, additional research with long‐term clinical and radiographic results including loss of correction, fusion rate, and junctional disease is necessary to verify the effectiveness of the cross trajectory technique.

Conclusions

The biomechanical advantage of fixation strength using the cross trajectory technique is demonstrated in this study. This technique may be a valid option for posterior fusion, especially in osteoporotic spine.