INTRODUCTION
Posterior fixation using screws and rods is a useful and widely used surgical method for treating various degenerative and traumatic spinal diseases. In this approach, the screw is used as an anchor to the vertebral body. Most screws are made of titanium (Ti) alloys, due to their already established strength, biocompatibility, and cost effectiveness.
However, Ti-alloy screws have two major drawbacks. The first is loosening, which is thought to be due to the micromovement, and the second is the difficulty in postoperative image evaluation caused by the strong Ti-alloy-induced artifacts in computNeurospine ed tomography (CT) and magnetic resonance imaging (MRI). Several solutions have been proposed to address these issues. For example, loosening can be prevented by using cement when inserting the screw; however, this makes screw removal difficult, and does not address the imaging problems [1,2]. In this context, it was proposed that using devices made of carbon fiber reinforced plastic (CFRP) might resolve these issues. Pedicle screws made of CFRP have excellent radiolucency as compared to those made of Ti alloys and are particularly useful for fixation in cases where bone structure visualization and radiation therapy are required [3-15]. In the general manufacturing industry, in terms of screw thread shape, resin and CFRP items are considered less likely to loosen than are metals. However, CFRP pedicle screws manufactured to date have not shown any superiority over Ti alloys in terms of loosening resistance [16]. More importantly, no CFRP pedicle screws produced to date have had strength comparable to those of Ti alloys, mostly because the thread shape of CFRP is technically difficult to produce. However, CFRP screws in which this problem has been addressed by integrating long carbon fibers into the resin, have been introduced in manufacturing in recent years, making it possible to apply this technology to the production of medical pedicle screws.
This study aimed to compare the performance of a long fiber type CFRP pedicle screw, manufactured using this new technology, with that of existing Ti-alloy screws.
MATERIALS AND METHODS
1. Ethics
All animal-related testing was conducted with ethical considerations in accordance with the ISO 10993-2:2022 Biological Evaluation of Medical Devices Part 2: Animal Welfare Requirements.
2. Development and Preparation of CFRP Screws
The strength of CFRP depends on the length and arrangement of the integrated carbon fibers. It is easy to ensure strength in rods or plates by weaving materials to form a mesh-like texture. However, it is difficult to ensure strength in screws and small parts, as it is challenging to adjust the length and arrangement of the fibers. As a result, the CFRP screws developed to date have been molded by cutting carbon fibers and mixing them with resin, in which they have a disordered arrangement (short fiber type) (Fig. 1A), which had significantly inferior strength. Fabrication of a long fiber type (Fig. 1A) screw, in which the fibers are molded by arranging them along the long axis of the screw while maintaining the fiber length and direction, is a major challenge for manufacturing. There have been no successful examples of practical applications of such screws to date. In recent years, this challenge has been overcome, and examples of the development and practical application of long fiber type screws in general manufacturing, such as semiconductor manufacturing equipment, have been reported.
We successfully applied this technology to manufacture pedicle screws (Fig. 1B). The CFRP part was molded into a long fiber type by adjusting the temperature and extrusion pressure of the material during extrusion molding, and using an epoxy resin, which is a thermosetting resin. In addition, a polyaxial-type head and central tube were placed in the center to accommodate minimally invasive surgery, so that surgery could be performed in the same way as with existing pedicle screws. The head parts (head and shank part of the screw body) were made of Ti alloy because it was impossible to mold it, even when using the manufacturing technology for long fiber type CFRP, and the central tube was also made of Ti alloy, to serve as a radiological marker, even though its main purpose was to guide the wire. Because the strength of the screw body structure was built using only long fiber CFRP material, metal coating or high-strength core material for the screw body was not needed.
Thus, we developed a long fiber type CFRP pedicle screw and conducted experiments to evaluate its mechanical properties (shear strength, torsional strength, and resistance to loosening) and artifacts during image evaluation.
3. Mechanical Properties
Screw strength was determined based on shear and torsional breaking strengths. The shear strength was measured by fixing the head of the screw and applying an external force to the neck in a direction perpendicular to the long axis until breakage or irreversible deformation occurred (Fig. 2A). The test materials used were three long fiber type CFRP pedicle screws (diameter, 6.5 mm; shaft length, 45 mm) and three Ti-alloy screws (diameter, 6.5 mm; shaft length, 45 mm) (Reline, Nuvasive Inc., USA) with a similar screw neck shape. The strength was measured using an ElectroPuls E3000 (Instron, USA).
Torsional breaking strength was measured as the maximum strength when a twisting external force was applied in the torque direction of the screw (Fig. 2B). The test was set based on the international standard American Society for Testing and Materials (ASTM) F543-17 “Standard Specification and Test Methods for Metallic Medical Bone Screws.” The screw was fixed with five threads remaining from the head, and a load was applied in the twisting direction (clockwise direction, as viewed from the head side) at three rotations/min. The torque was recorded until the specimen was significantly deformed. Five long fiber type CFRP pedicle screws (diameter, 6.5 mm; shaft length, 45 mm) were used as test materials, and three Ti-alloy screws (diameter, 6.5 mm, shaft length, 45 mm) (Solera, Medtronic Sofamor Danek USA, USA) were used as comparison objects. A 22.5-N· m torsion tester (Instron) was used as a strength measurement device. For reference, 5 short fiber type CFRP screws were fabricated with the same shape as the long fiber type CFRP screws; the torsional breaking strength was measured using the same method.
The maximum torque required to remove the screw from the simulated bone was measured to evaluate resistance to loosening (Fig. 2C). The higher the maximum value, the less likely it is for the screw to loosen. The test was set based on ASTM F543-17 “Standard Specification and Test Methods for Metallic Medical Bone Screws.” The simulated bone (Sawbones, #1522-03, 20 pcf, solid rigid polyurethane foam block) was cut into 40 mm× 40 mm×60-mm pieces to prepare a test block. A pilot hole of 6-mm diameter was drilled into the 40 mm×40-mm surface of the test block. The entire screw was screwed into the test block. From the head to the tip, the screw was unscrewed counterclockwise at a speed of 30 rotations/min, and the torque and angular displacement were recorded until the specimen was removed from the test block. Five long fiber type CFRP pedicle screws (diameter, 6.5 mm; axial length, 45 mm) were used as test materials, and three Ti-alloy screws (diameter 6.5 mm, axial length 45 mm) (Solera, Medtronic Sofamor Danek USA, Inc.) were used as comparison objects. A 22.5-N· m torsion tester (Instron) was used as the test device.
4. Artifacts During Image Evaluation
Pedicle screws were implanted into the right side of a swine vertebral body. The test material was a long fiber type CFRP pedicle screw (diameter, 6.5 mm; axial length, 20 mm), and the comparison object was a Ti-alloy screw (diameter, 6.5 mm; axial length 20 mm) (Solera, Medtronic Sofamor Danek USA Inc.). The Infinix Celeve-i imaging device (Canon Medical Systems, Japan) was used. In addition to qualitative image evaluation, quantitative artifact evaluation was performed using the Gumbel technique [17]. Image size was adjusted to the anterior-posterior distance (ventral body surface to spinous process tip) of the vertebra as 250 pixels, and a 10-row by 20-column region within the spinal canal center was designated as the region of interest (ROI) (Fig. 3A, the red square area). The maximum differ-ence in CT values within each 10-cell column of the ROI was measured across all 20 columns. Variation in CT value brightness (256 levels) was subsequently used as a quantitative parameter for artifact measurement. ImageJ (1.54f, Wayne Rasband and contributors, National Institutes of Health, USA) and Java 1.8.0_322 (64-bit) were used to obtain measurements [18].
To evaluate the interference of the long fiber type CFRP screws with MRI, a test was conducted based on the ASTM F2119-07 “Standard Test Method for Evaluation of MR Image Artifacts from Passive Implants.” The device was placed in a phantom made of 1.32 g/L NaCl and 10 g/L partial sodium salt of polyacrylic acid. MRI was performed using a 3-T MAGNETOM Vida (Siemens Healthcare, Germany), and images were acquired using spin-echo and gradient magnetic field echo sequences. The range of artifacts was measured as the maximum distance (mm) from the boundary of the test piece to the outer edge of the artifact. The test material was a long fiber type CFRP pedicle screw (diameter, 6.5 mm; axial length, 45 mm), and the comparison object was a Ti-alloy screw (diameter, 6.5 mm; axial length 45 mm) (Solera, Medtronic Sofamor Danek USA, Inc.).
5. Statistical Analyses
Stata 18 (StataCorp., USA) was used for statistical analyses to evaluate the results of strength testing. Differences between the features and effects of each screw type were analyzed using Student t-test. Statistical significance was set at p-value<0.05.
6. Biosafety and Other Tests
For medical devices that are to remain implanted in the body for long periods, comprehensive tests need to be conducted to confirm their safety to obtain approval for clinical use. We conducted biological safety tests and magnetic resonance (MR) safety tests necessary to compare long fiber type CFRP rods (diameter, 5.5 mm) and Ti-alloy rods (diameter, 5.5 mm) (Solera, Medtronic Sofamor Danek USA Inc.). A 5.5-mm diameter screw was used because a 6.5-mm diameter screw was too large to be placed inside the swine’s spine. To assess biological safety, tests for cytotoxicity, skin sensitization, intracutaneous reactivity, pyrogenic activity, acute systemic toxicity, reverse mutation, chromosomal aberration, and the implantation itself were performed according to ISO 10993. Because this CFRP screw has exposed the CFRP on its surface and it may react with bone or other biological tissues, we planned a long-term biological safety test as well as the required implant biological interaction conditions. The implantation test was conducted over 24 weeks to confirm the long-term safety, with an evaluation of organs and tissues throughout the body of swine.
In addition to the artifact measurement test mentioned above, MR safety tests were conducted in accordance with ASTM F2052 “Standard Test Method For Measurement Of Magnetically Induced Displacement Force On Medical Devices In The Magnetic Resonance Environment,” F2213 “Standard Test Method for Measurement of Magnetically Induced Torque on Medical Devices in the Magnetic Resonance Environment,” and F2182 “Standard Test Method for Measurement of Radiofrequency Induced Heating On or Near Passive Implants During Magnetic Resonance Imaging.” These tests measure and evaluate the torque and deflection forces exerted on a device placed in a magnetic field, as well as the degree of accompanying heat generation. The MRI equipment used for these tests was a 3T MRI, as used in the artifact measurement tests.
RESULTS
1. Mechanical Properties
The shear strength (mean±standard deviation) of the long fiber type CFRP pedicle screw was 11,377.7±245.1 N, while that of the Ti-alloy pedicle screw was 10,300.3±249.7 N (Fig. 2D). The long fiber type CFRP pedicle screw was thus superior to the Ti-alloy pedicle screw in terms of shear strength (p<0.01).
The torsional strength of the long fiber type CFRP pedicle screw was 4.4±0.5 N·m, while that of the Ti-alloy pedicle screw was 22.4±0.6 N· m (Fig. 2E). The long fiber type CFRP pedicle screw was thus inferior to the Ti-alloy pedicle screw in terms of torsional strength (p<0.01). For reference, the strength of the short fiber type CFRP screw was 1.7±0.0 N· m, which was less than half of the strength of the long fiber type CFRP pedicle screws.
In terms of loosening resistance, the maximum torque value of the long fiber type CFRP pedicle screw was 0.99±0.08 N· m, while that of the Ti-alloy pedicle screw was 0.75±0.05 N ·m. The long fiber type CFRP pedicle screw was significantly less prone to loosening than the Ti-alloy pedicle screw (p<0.01), amounting to a median 32% higher resistance (Fig. 2F).
2. Artifacts During Image Evaluation
In the cone-beam CT images, the long fiber type CFRP pedicle screw showed a clear reduction in artifacts as compared to the Ti-alloy pedicle screw, making it easier to evaluate the area around the spinal canal. Furthermore, the CFRP screw made it possible to evaluate ossification on the screw surface and quantitative evaluation using the Gumbel technique also revealed a significant reduction in artifact interference (Fig. 3A). In the MR images, even though some metal components were used in these screws, the interference range was halved, and the artifact range was clearly reduced (Fig. 3B), as compared to the artifacts associated with the Ti-alloy screws. Although the long fiber type CFRP screw uses Ti-alloy parts in some areas, these parts are inside the implant; thus, interference with the contact area between the screw surface and bone is minimal.
3. Biosafety and Other Tests
Long fiber type CFRP screws demonstrated biological safety equivalent to that of Ti alloys in biological safety tests, with no observed toxicity. Although the resin used in the screw was a new type of epoxy resin, there was no experimental evidence of toxicity to living tissue, including bone (Table 1). Since the reverse mutation, chromosomal aberration, and implantation tests showed no toxicity, it was considered that the compound would not exhibit any carcinogenic toxicity; therefore, no carcinogenicity test was performed. Moreover, the MR safety tests indicated that the torque force, deflection force, and heat generation were comparable to or less than those of the Ti-alloy screws (Table 1).
DISCUSSION
1. Advantages of the Application of the Long Fiber Type CFRP Screw
In the present study, we evaluated the main parameters required for spinal implants, particularly screws, comparing between the newly created long carbon fiber screws and the most widely used for the moment Ti screws.
The shear strength was measured with a load applied perpendicular to the axis, that is, in the direction in which the screws usually brake. In clinical practice, the screw body is embedded within the vertebral body. Therefore, the point where the load is most likely to be applied is its neck, which is close to the connection to the rod. One of the standard tests for determining screw strength is the ASTM F1798 “Standard Test Method for Evaluating the Static and Fatigue Properties of the Interconnection Mechanisms and Subassemblies used in Spinal Arthrodesis Implants.” In this test, a pedicle screw was connected to a rod, set in a T-shape, and both ends of the rod were fixed. A load was applied perpendicular to the axis to one point of the screw body, to assess the durability of the entire system. The screw body often brakes during such test. In these breakage cases, the neck usually breaks, probably because the neck is the part that is most likely to be subjected to the strongest mechanical mo-ment (torque). Therefore, we considered that the neck was the part where the strength of the screw body was utmost importance, and we performed a shear strength measurement test on the neck. We demonstrated that the CFRP screw was significantly stronger than existing Ti-alloy screws. This is because the neck structure is a hybrid of Ti-alloy components and long fiber type CFRP (Fig. 1B). Theoretically, it has been demonstrated that it has almost the same level of strength as the Ti alloy alone. In addition, although the maximum diameter of the shaft was the same (6.5 mm), the minimum diameter at the neck of the long fiber type CFRP pedicle screw was 0.1 mm greater than that of the Ti-alloy screw used for comparison. However, we don’t believe that this difference of neck diameter has contributed to the superiority of shear strength of the long fiber type CFRP pedicle screw.
In terms of torsional strength, the long fiber type CFRP pedicle screw was inferior to the Ti-alloy screw. Ishikawa et al. [19] measured the insertion torque of 190 pedicle screws with diameters of 7.5 mm or 8.0 mm to evaluate the relationship between screw tightening pressure and stability, and found that the torque was 2.38±0.95 N ·m. Other studies also measured the insertion torque in the same direction [20-25] and found values that were approximately half of the strength limit of the long fiber type CFRP pedicle screw, at 4.4±0.5 N· m in the present study [20-25]. The strength limit of the long fiber type CFRP pedicle screw was significantly higher than the maximum torque applied to the screw in clinical practice (p<0.01), and the possibility of failure was considered statistically extremely low (Fig. 4). In addition, the long fiber type CFRP pedicle screw is currently undergoing numerous non-clinical trials, including animal experiments, and hundreds of screws have been manually inserted into simulated and living bones, without any cases of failure to date. Thus, although the torsional strength of the long fiber type CFRP pedicle screw was lower than that of the Ti alloy, it was considered to have achieved an acceptable level for clinical use. However, it is still important to use an appropriate size of tapping device before placing the pedicle screw. A slightly smaller in diameter tapping shaft than that of the screw prevents excessive torque use during surgery, prevents loosening, and reduces postoperative complications [26-28].
The strength of the shirt fiber type CFRP screws, which was measured for reference, was below the range of the highest strength that can be applied clinically; thus, there is a considerable possibility that the screw will break when inserted. It cannot be denied that this inferiority in terms of strength is the reason that the CFRP screws currently used in clinical practice cannot be used for the same indications as metal screws.
The results showed that long fiber type CFRP pedicle screws were significantly less likely to loosen than Ti-alloy pedicle screws, which may be directly related to the resistance to screw loosening during and after surgery in clinical practice, and are expected to lead to a reduction in the rate of reoperation and intraoperative complications. The reason for this is thought to be that CFRP, unlike metals, is flexible and absorbs vibrations [29,30]. In the past, a similar test was conducted with a short fiber type CFRP pedicle screw; however, it did not produce superior results [16], unlike our long fiber type CFRP pedicle screw. Screws with deep threads and large surface area or those with thick threads that penetrate the surrounding matrix are considered to be less likely to loosening. Therefore, there are various shapes of pedicle screws used in clinical practice. When developing the long fiber CFRP pedicle screw, the thread shape was determined by prioritizing the above two factors. However, in deciding the shape, it was necessary to consider the technical issues in molding. The long fiber CFRP pedicle screw needs to have long carbon fiber inside the thread. If the carbon fiber does not sufficiently run inside the thread, the strength of the screw will be decreased. Consequently, we selected a thick thread shape for our screw, which increases the surface area of the thread while ensuring the carbon fiber runs inside the thread, preserving strength and preventing loosening. In this case, compared to the Ti-alloy screws, the long fiber type CFRP screw was found to be more resistant to loosening, implying that it should have high postoperative stability in clinical practice.
In terms of image evaluation, previous short fiber type CFRP devices also showed good results in terms of low interference with medical imaging, as compared to Ti-alloy screws [3-15]. Long fiber type CFRP screws have a similar composition, and a similar low interference was considered appropriate from the perspective of material engineering. The long fiber type CFRP screw, similar to the short fiber type CFRP screw, demonstrated superiority over metal screws in radiation image evaluation, which has benefits for radiation therapy planning. Although it is theoretically possible to achieve complete stealth performance, this is not appropriate in actual clinical practice. Therefore, a Ti-alloy cannula was placed inside the screw to confirm the position of the screw and to allow monitoring for breaks. This cannula also served as a tunnel for passing the guidewire, as a cannulated screw. Because it was technically impossible to form the part that connects to the rod using CFRP alone, Ti-alloy parts were used. The volume of the Ti-alloy parts was 32% of the entire screw body: a 68% reduction in metal volume compared to existing screws. In addition, because the metal was used inside the screw and in the case of a 6.5 mm diameter screw, the thickness of the components near the head portion of body was 2.5 mm for CFRP and 2.5 mm for metal, and for the central tube part it was 4.4 mm for CFRP and 0.4 mm for metal (Fig. 1B), the impact of artifacts on image evaluation in the spinal canal was greatly reduced. The central tube and head were made of Ti alloy, but this is not expected to have a significant effect on the evaluation of intervertebral fusion or nerve tissue in the spinal canal.
Currently, existing short fiber type CFRP screws present problems in terms of strength in general, and none have demonstrated the same strength as that of Ti-alloy screws when directly measured. Consequently, none have been approved for fixation surgery in degenerative diseases, and their application is limited to tumor lesions of the spine. CFRP screws (short fiber type) currently used in clinical practice are not used in the same way as Ti-alloy screws. In this context, the long fiber type CFRP pedicle screws can be used in the same way as Ti-alloy screws, are less likely to loosen, and produce low image interference, making them more clinically useful than Ti-alloy screws. In addition, the manufacturing cost is the same as that of Ti-alloy screws, making it economically feasible. Considering its physical properties and advantages in image evaluation, we believe that these screws could be a significant advance and will change the current trend of spinal posterior fixation devices from Ti alloys to CFRP. We are currently systematizing the test data, including the contents of this research, and preparing for clinical approval and practical use (patent No. PCT/JP2023/025979 and Japanese patent No. 2023-211726).
2. Clinical Applications
This study did not include clinical data and presented mainly non-clinical mechanical and biological safety test data; however, the results of nonclinical trials are a very strong indicator to the usefulness of the device in actual surgery. Considering the following four reasons, especially the property related to resistance to loosening, it is suggested that the long fiber type CFRP screws are likely to demonstrate clinical usefulness equal to or greater than that of Ti-alloy screws and to reduce the risk of postoperative complications.
First, results of mechanical tests are very important for the approval of posterior spinal internal fixation devices for clinical use. Furthermore, the results of public standard tests, including ASTM, are important for approval, and in many cases, the results of clinical trials are not decisive for approval. ASTM tests are conducted by applying very large loads, and thus, it is possible to perform evaluations at levels that cannot be reproduced in actual clinical practice. Therefore, compared to Ti-alloy screws, this long fiber type CFRP screw is less likely to break during actual clinical use.
Second, physical resistance to the loosening of the screw is thought to be directly linked to the prevention of pullout in clinical practice. Past studies have reported that the amount of the applied torque at the time of insertion is directly linked to the prevention of pullout [22,23]. This does not contradict the empirical rule that, in actual surgery, spinal surgeons decide to place the screw while checking the strength of the torque during insertion. Because the long fiber type CFRP screw require strongertorque and have higher resistance to loosening compared to Tialloy screws, it is highly likely that the use of the long fiber type CFRP screw will contribute to reducing the rate of pullout in actual clinical practice.
Third, although no surgical procedures were performed on patients during this study, the implantation test involved placing the screw in swine using the same procedure as supposed to be in the actual surgical practice. Consequently, no defects were noted during the procedure compared to existing Ti-alloy screws. Furthermore, the strength of the torque at the time of insertion of the long fiber type CFRP screw could be felt, and the second issue mentioned above was proven in actual procedures. From these findings, the possibility of unexpected malfunctions occurring during actual surgical use is expected to be low, and that the screw is expected to be less likely to loosen in patients.
Fourth, the superiority of the screw in radiographic evaluation was clear in swine and phantoms, and the evaluation of the area around the spinal canal was unaffected. This has also been proven with existing short fiber type CFRP devices; it is believed that the superiority of this device compared to metal devices in clinical image evaluation is clear.
CONCLUSION
The long fiber type CFRP pedicle screw that we have developed using innovative manufacturing technology has demonstrated sufficient strength for clinical use, is significantly less likely to loosen than are existing Ti-alloy screws, facilitates image evaluation, and is economical to produce. This study established that long fiber type CFRP screws offer superior mechanical performance and reduce imaging artifacts, suggesting that they are a viable alternative to current metal screws in spinal fixation.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Full text via PMC
Download Citation
Print
