BIOMECHANICAL STUDY OF PULLOUT RESISTANCE OF CORTICAL SCREWS IN SWINE LUMBAR VERTEBRAE

ABSTRACT

Introduction:

This study aimed to evaluate the mechanical resistance to pullout of the lumbar cortical bone trajectory (CBT) screw with and without transfixation of the second cortical in a biomechanical assay in swine vertebrae.

Methods:

A pullout test was performed in 16 swine vertebrae (L5 and L6) fixed with 32 CBT screws using the same protocol. Of these, 16 were inserted in a monocortical manner according to the original technique and 16 through a modified technique which included the transfixation of the second cortical bone.

Results:

The average maximum load value for pulling out the inserted screws with the modified technique including the transfixation of the second cortical bone was 140.50 Kgf (± 56.56), while in the original technique it was 95.69 Kgf (± 42.19) (p = 0.004).

Conclusion:

The biomechanical tests applied to the studied sample showed that the pullout resistance of the CBT screws inserted with the bicortical technique was, on average, 46% higher than the screws inserted according to the original monocortical technique. Level of Evidence II; Experimental study.

INTRODUCTION

Spinal surgery is increasingly common worldwide, with a considerable growth in the last two decades, especially in the elderly population. ^1,2

Among the various fixation techniques to promote thoraco-lumbar arthrodesis, the most commonly used today is the pedicular fixation, ³ in which the screw is inserted into the cancellous bone inside the vertebral pedicle. This technique is considered the gold standard nowadays.

Spine fixation with pedicle screws shows superior biomechanical performance compared to the old methods, but still with fixation failures, especially in osteoporotic bones. ^4–6

Searching for a new spine fixation system, in 2009, Santoni et al ⁷ described a screw technique for the lumbar spine, the so called cortical bone trajectory (CBT) screw, with mechanical advantages, especially when applied to osteoporotic bones ^7–9

The particular characteristic of this screw which supports its own name - cortical bone trajectory - is the pericortical bone positioning in almost its entire path.

The CBT screw can reach up to four points of contact with the cortical bone, namely: the screw entry point in the pars interarticularis, the inferomedial wall of the pedicle, the anterolateral wall of the pedicle and the lateral cortical of the vertebral body, where the screw ends its path by touching it, but not transfixing it. ¹⁰ Among these, pars interarticularis is the anatomically highest bone density point that the screw has contact with. ¹⁰

This study uses a modification of the Santoni et al ⁷ technique, which consists of the transfixation of the second cortical bone, which would lead it to be considered as a bicortical screw technique. The hypothesis of the study is that this modification of the original technique has better mechanical performance in pullout tests, when compared to screws inserted without cortical transfixation.

This study aimed to compare the mechanical strength of CBT screws when used in a bicortical way compared to those used in a unicortical way, as described by Santoni et al. ⁷

MATERIAL AND METHODS

To carry out the study, the two most distal lumbar vertebrae (L5 and L6) of pigs slaughtered for human consumption were used. These samples were obtained from animals weighing approximately 100 kilograms (Kg), with an average age of 150 days, all male and from the Landrace breed. The total of 10 L5 and 10 L6 vertebrae from 10 recently slaughtered pigs were dissected and kept in a refrigerator at an average temperature of −20° Celsius until the tests were carried out. For the execution of the biomechanical tests, the anatomical pieces were placed at the room temperature until they were completely thawed, in order to have their mechanical properties recovered.

To carry out the tests, the vertebral bodies were adapted and supported on a firm base through a metallic device similar to a hollow cube, which had its lateral walls with a groove to fit the screws. An Instron 5882 universal testing machine with a load cell with a maximum capacity of 100 KiloNewtons was used, coupled with the Bluehill2 software that allowed the registration of the force required to pull the screws out of the vertebrae.

All screws used in the tests were a 4.5 mm-diameter and 45 mm-length titanium alloy polyaxial pedicle screws (Spherx II Nuvasive - Manufacturer: Nuvasive, INC - USA).

The preparation of CBT screws paths, through the pedicles of the pigs lumbar vertebral bodies (L5 and L6), was made according to the technique described by Santoni et al. ^7,11,12 followed by the fixation with the screws, which were performed by a single surgeon, freehand, identical to the technique of the surgical procedures.

The screw entry point was at the junction of a vertical line that passes through the medial face of the lower facet of the adjacent cranial vertebra and a horizontal line, perpendicular to it, which passes approximately 1 millimeter below and parallel to the lower edge of the transverse process. ^7,11,12 (Figure 1)

Figure 1. Entry point of the cortical screw as described by Santoni et al. ⁷ and used in this study.

The initial perforation of the entry point was made with the initial awl, followed by the Lenke probe until it touched the vertebral body's cortical wall, to complete the CBT screw path. The path is 8-10 degrees from medial to lateral in the axial plane and 25-30 degrees from caudal to cephalic in the sagittal plane. ^7,11,12 The same procedure was made through the contralateral pedicle. (Figure 2)

Figure 2. Demonstration of the bicortical bone trajectory screws in the coronal, sagittal, axial and oblique planes.

Once the paths were created bilaterally, following the technique described by Santoni et al. ^7,11,12 and with the cannula having already touched the cortical wall, a simple randomization was carried out, for each independent vertebra, to define on which side the screw would transfix the second cortical bone - using the technique described by the authors.

The pedicles were then fixed in sequence, one side at a time, one side with transfixation of the cortical bone and the other without transfixation and subjected to mechanical tests. Therefore, for each vertebra, there was a conventional Santoni et al. ⁷ fixation and a bicortical fixation proposed by the present authors. (Figure 3 and 4)

Figure 3. Demonstration of the CBT screw with cortical vertebral transfixation.

Figure 4. Demonstration of the pull out force in the same direction as the screw.

The tests were all carried out at the room temperature and with a test speed of 5 mm/min. The force required to pull out the screw inserted in the two types of CBT paths was measured in kilogram-force (Kgf) and was recorded during the application of the mechanical tests. The definition of the force required to pull out the screw was the maximum load required to start displacing the screw.

The D'Agostino-Pearson and Shapiro-Wilk tests were used for exploratory analysis to determine the normality of the data. Given the parametric nature of the data, the evaluation of possible statistical differences was performed using the paired "t" test between the two techniques for CBT screws. In all analyzes performed, the differences obtained were statistically significant when the p value was less than 0.05 (p<0.05). All analyzes were performed using the GraphpadPrism® program, version 5.0 for Windows. The mean and standard deviation data for the maximum load variable required for removing the screws were obtained.

RESULTS

Tests were performed on 20 lumbar vertebrae testing a total of 40 screws. Twenty screws were inserted in a CBT path with transfixation of the second cortical bone and 20 without transfixation. The loss rate was 20%, considering that 4 vertebrae were discarded out of a total of 20, due to broken pedicles during the passage of the screws or during the tests; 16 vertebrae with 32 pedicles for screw insertion made up the final sample of this study.

Based on the pullout force (maximum load), the sample used in the present study obtained a statistical power of 81.99% [95% CI = 66.10 to 23.51; α = 0.05]. The standard deviation of the difference between screws with and without transfixation was ± 39.97 and the average difference between them was 44.81.

The mean value for pulling out the screws inserted with the transfixation of the second cortical bone was 140.50 Kgf (± 56.56), while in the technique without transfixation it was 95.69 Kgf (± 95.69) (p = 0.0004). In the technique with transfixation of the second cortical bone, the force required to pull out the screw was, on average, 46% higher. (Table 1) (Graphic 1)

Table 1. Maximum load values (Kgf) with and without transfixation of the second cortical bone.

Vertebrae studied	Type of vertebrae	Without cortical transfixation (Kgf)	With transfixation of the second cortical bone (Kgf)	p value
V1	L5	68.59	73.65	0.0004 T
V2	L6	30.25	66.60
V3	L5	27.66	42.22
V5	L5	103.30	238.06
V6	L6	96.11	154.30
V7	L5	124.85	125.40
V9	L5	99.34	120.38
V10	L6	140.94	251.54
V11	L5	63.10	135.94
V12	L6	64.30	103.09
V14	L6	46.48	120.10
V15	L5	114.05	166.47
V16	L6	128.58	174.09
V18	L6	162.91	183.72
V19	L5	105.00	154.90
V20	L6	155.62	137.52
MEAN		95.69	140.50
SD		42.19	56.56

Database = 32 tests were performed (total of 16 vertebrae in general - 2 independent tests on each vertebra). 16 screws with transfixation and 16 without cortical transfixation were used. SD = standard deviation. T = paired t-test.

DICUSSION

The idea of using bicortical screws in the lumbar spine has been tempting for years, especially for the population with osteoporotic bones. Several authors ^13–16 described the traditional pedicle insertion technique in a bicortical manner and demonstrated mechanical advantages when compared with the monocortical pedicle screws, but the unfavorable vascular anatomy ^15–16 increases the risk of vascular injuries with the technique of bicortical pedicle screws and therefore it is not routinely used.

Unlike pedicle screws, the CBT screws used with transfixation of the second cortical bone, as reported in this study, would be prominent in the lateral vertebral body cortex, in its cranial third and in its two posterior thirds ^7,10 in a region considered safe, regarding the risks to the great vessels or to the lumbar arteries. ^15,16

Figure 5. Average value of the maximum load presented by CBT screws without transfixing the cortical bone (monocortical) and with transfixation of the second cortical bone (bicortical).

The bars represent the average of each technique and the lines, the standard deviation. A) *** = represents a statistically significant difference between the techniques, according to the paired t test (p <0.05); B) maximum individual load obtained in each tested vertebra.

In order to study the vascular safety, Guimarães at al. ¹⁷ studied, through a retrospective analysis of 50 abdominal angiotomography scans, the anatomical relationship of the probable region corresponding to the exit point of the CBT screw, used in a bicortical manner, on the anterolateral face of the vertebra with the vascular structures adjacent to the spine. The measurements obtained between the prevertebral vessels and the possible exit points of the screw did not demonstrate contact in any of the vertebrae studied. This initial result suggests safety in the use of the cortical screw transfixing the second cortical bone, but further studies are necessary for a better understanding. Knowing the position of the vessels is essential to reduce intra- and postoperative complications related to spinal instrumentation.

Matsukawa et al. ¹⁸ described the CBT screw with transfixation of the second cortical bone in the first sacral vertebrae (S1) in 19 patients. This technical variation was considered safe by the authors. The authors demonstrated an average of 141% higher insertion torque than the traditional monocortical technique.

The belief in the literature is that the inclusion of the second bone cortex increases the mechanical resistance of the implants. ^13–16,18

The pullout of the screws is the main failure mechanism of the fixation systems of the spine and, therefore, this mechanical property, pullout resistance, was chosen for this study. ^4–6 However, the superiority of the transfixation of the second cortical bone, demonstrated in this study, cannot be yet confirmed for other situations of biomechanical.

Spirig et al. ¹⁹ described the biomechanical performance of pericortical-CBT versus bicortical-CBT screws (pCBT versus bCBT) in a randomized cadaveric study submitted to physiological cyclic loading tests, including shear and tension loads, as well as bending moments. They demonstrated a slight decrease in translational and angular displacements of the bCBT screws when compared to pCBT, however, the results were non-significant (p>0.05) and therefore the authors do not recommend the placement of CBT screws bicortically.

In this study, an average 46% (p = 0.0004) increase in pullout resistance was demonstrated when the second cortical bone was transfixed instead of just touching it.

CONCLUSION

The technique of the CBT screw inserted in a bicortical manner showed, in the lumbar column of swine, a pullout resistance 46% higher, on average when compared with the CBT screw described by Santoni et al. Further studies are needed to better understand the biomechanical characteristics and safety of this alternative technique of fixation of the lumbar spine.

DATA AVAILABILITY DECLARATION

The content underlying the research is included in the manuscript.