Second-generation bone cement-injectable cannulated pedicle screws for osteoporosis: biomechanical and finite element analyses

Congcan Li; Lei Song; Jun Xiao; Wenwen Wu; Yifan Jiang; Rui Zhou; Fei Dai

doi:10.1186/s13018-023-03752-2

. 2023 May 10;18:343. doi: 10.1186/s13018-023-03752-2

Second-generation bone cement-injectable cannulated pedicle screws for osteoporosis: biomechanical and finite element analyses

Congcan Li ^1,^#, Lei Song ^1,^#, Jun Xiao ², Wenwen Wu ³, Yifan Jiang ⁴, Rui Zhou ^1,^✉, Fei Dai ^1,^✉

PMCID: PMC10170841 PMID: 37161530

Abstract

Background

Biomechanical and finite element analyses were performed to investigate the efficacy of second-generation bone cement-injectable cannulated pedicle screws (CICPS) in osteoporosis.

Methods

This study used the biomechanical test module of polyurethane to simulate osteoporotic cancellous bone. Polymethylmethacrylate (PMMA) bone cement was used to anchor the pedicle screws in the module. The specimens were divided into two groups for the mechanical tests: the experimental group (second-generation CICPS) and control group (first-generation CICPS). Safety was evaluated using maximum shear force, static bending, and dynamic bending tests. Biomechanical stability evaluations included the maximum axial pullout force and rotary torque tests. X-ray imaging and computed tomography were used to evaluate the distribution of bone cement 24 h after PMMA injection, and stress distribution at the screw fracture and screw–cement–bone interface was assessed using finite element analysis.

Results

Mechanical testing revealed that the experimental group (349.8 ± 28.6 N) had a higher maximum axial pullout force than the control group (277.3 ± 8.6 N; P < 0.05). The bending moments of the experimental group (128.5 ± 9.08 N) were comparable to those of the control group (113.4 ± 20.9 N; P > 0.05). The screw-in and spin-out torques of the experimental group were higher than those of the control group (spin-in, 0.793 ± 0.015 vs. 0.577 ± 0.062 N, P < 0.01; spin-out, 0.764 ± 0.027 vs. 0.612 ± 0.049 N, P < 0.01). Bone cement was mainly distributed at the front three-fifths of the screw in both groups, but the distribution was more uniform in the experimental group than in the control group. After pullout, the bone cement was closely connected to the screw, without loosening or fragmentation. In the finite element analysis, stress on the second-generation CICPS was concentrated at the proximal screw outlet, whereas stress on the first-generation CICPS was concentrated at the screw neck, and the screw–bone cement–bone interface stress of the experimental group was smaller than that of the control group.

Conclusion

These findings suggest that second-generation CICPS have higher safety and stability than first-generation CICPS and may be a superior choice for the treatment of osteoporosis.

Keywords: Cement-injectable cannulated pedicle screws, Osteoporosis, Safety, Stability

Introduction

An aging population is increasingly becoming a global problem, and spinal degenerative diseases are now frequently diagnosed in older patients. Pain caused by intervertebral disc herniation, spinal stenosis, and spondylolysis impose a heavy burden on the economy, and seriously affect the health and quality of life of patients [1]. Surgery is indicated when conservative treatment fails to achieve spinal recovery.

Pedicle screw fixation has been widely used in the surgical treatment of spinal degenerative diseases [2–5]; however, patients with osteoporosis provide a greater challenge. Osteoporosis seriously affects bonding at the bone–screw interface and screw stability owing to the destruction of bone tissue microstructure and reduction of bone mass; this results in screw loosening, displacement, or protrusion [6, 7].

Polymethylmethacrylate (PMMA)-reinforced screws have been proven to be the most effective system for strengthening pedicle screws and achieving stable fixation [2, 8–11]. PMMA bone cement penetrates the bone trabeculae to form a cement–bone interface in addition to the screw–bone interface, improving the stability of pedicle screws in osteoporotic vertebrae [12–14].

Since the initial application of first-generation cement-injectable cannulated pedicle screws (CICPS) enhanced with PMMA in clinical practice in 2011, the incidence of screw loosening, bone cement leakage, and even pulmonary embolism has reduced [15–18].

In 1975, Liu et al. [19] constructed the first three-dimensional finite element model of the vertebral body for the study of direct shear resistance in the lumbar spine, marking the beginning of finite element analysis in spinal biomechanics. This analysis has been increasingly used with the rapid development of computer technology [20, 21] and involves the creation of a detailed and accurate computer-simulated three-dimensional stereo model [22, 23]. Stress analysis can then be performed using this model, which greatly assists in the correct selection of surgical methods, reducing the rates of surgical failure and complications, and optimizing implant design.

To further optimize screw design and reduce complications, we have improved the design of first-generation CICPS to create second-generation CICPS. This study aimed to evaluate screw safety through biomechanical experiments and analyze stress distribution at the screw fracture and screw–cement–bone interface using finite element analysis.

Methods

Screw design

Second-generation CICPS design features a dual thread with an inner diameter of 1.6 mm, a thread lead of 6 mm, and a pitch of 1.5 mm in the proximal two-fifths and 3 mm in the distal three-fifths. The three side holes in the front two-fifths of the screw are separated from each other by 120° and two threads. From the near to far sides, the holes are round (2 × 2 mm), oval (3 × 2 mm), and U-shaped (4 × 2 mm) and can be designed with different specifications. All pedicle screws used in this study were manufactured by Chongqing FWS Medical Device Co., Ltd. (Chongqing, China) and had a diameter of 6.5 mm and a length of 45 mm (Fig. 1).

Fig. 1 — The proximal, middle, and distal side holes are shown in the left, middle, and right panels, respectively

Specimen preparation

Commercially available polyurethane (PU; catalog no. 1522-507; open cell 7.5 PCF, 18 × 13 × 4 cm; Sawbones, Pacific Research Laboratories, Vashon Island, WA, USA) was used to simulate the mechanical properties of osteoporotic cancellous bone. An open cone was used to drill holes in the side of the specimen, and a nail hole with a depth of 40 mm was prepared perpendicular to the side of the specimen. After installation of the screw placement instrument, the pedicle screw was slowly screwed vertically into the cancellous bone substitute, until the screw head was in contact with the surface of the specimen. During screw placement, the specimen structure was protected from damage by left–right shaking. All specimens were used self-tapping screws to avoid weakening the screw-holding force (Fig. 2A).

Fig. 2 — Screw placement was done according to the standard clinical method (A) and was followed by injection of 1.5 mL PMMA into each screw at room temperature (B)

Bone cement injection

At room temperature, 20 g PMMA bone cement and 10 mL solvent were mixed in a stainless-steel bowl, with a strict powder to liquid ratio of 2:1. When the bone cement became a paste, 10 mL was placed into a syringe, and the air was removed from the connecting device and screw. When the bone cement had developed the consistency of toothpaste, 1.5 mL was injected into each screw (Fig. 2B). After allowing the bone cement to set for 24 h, X-ray imaging and computed tomography (CT) were performed to observe the distribution of the bone cement in the module.

Mechanical testing

Static bending test

This was performed using an electronic universal testing machine (model no. CMT-5105). The long axis of the adjusting screw was perpendicular to the bending direction, and the load was applied at a constant rate of 25 mm/min perpendicular to the long axis of the pedicle screw. A computer automatically recorded the continuity value and drew the load–displacement curve until the load dropped significantly (Fig. 3A).

Fig. 3 — Experimental setups for the static bending test (A) and rotary torque test (B)

Dynamic bending test

This was performed using a fatigue tester (model no. Fd3000-P), with the screw installed in the same manner as in the above test, and the maximum loads set at 65%, 75%, and 80% of the yield load. The loading wave was sinusoidal, R-value was 0.1, and frequency was 10 Hz. Dynamic tests were conducted for each screw in turn, and the load–cycle curves were drawn using a computer analysis of the results until the screw broke or the cycle stopped at 2.5 × 10⁶ cycles.

Rotary torque test

This was performed using a surgical implant electromagnetic dynamic mechanics test system (model no. M-100T). The adjusting screw was perpendicular to the specimen surface and a force of 11.17 N was applied vertically at a constant torsional speed of 3 r/min. The computer automatically recorded the continuity value and drew the torque–angular displacement curve; the recording stopped when all threads had entered the module. Unscrewing was performed in the same manner until the threads had completely exited the module (Fig. 3B).

Maximum axial pullout force test

This was performed using an electronic universal testing machine (model no. CMT-5105). The long axis of the adjusting screw was parallel to the stretching direction. The screw was pulled out along the long axis of the pedicle screw at a constant rate of 5 mm/min. The computer automatically recorded the continuity value and drew the load–displacement curve until the load dropped significantly (Fig. 4A).

Fig. 4 — Experimental setups for the maximum axial pullout force test (A) and maximum shear force test (B)

Maximum shear force test

The test element was a fatigue tester (model no. FD5000-P) with maximum loads of 1000, 1500, and 2000 N. The loading wave was a sine wave with an R-value of 0.1 and a frequency of 5 Hz. Dynamic tests were carried out for each screw in turn, and the load–cycle curves were drawn by computer analysis of the results until the screw broke or the cycles stopped at 2.5 × 10⁶ cycles (Fig. 4B).

Finite element analyses

The CT scan results at 0.625 mm intervals were imported into SolidWorks geometric model processing software version 1.0 (SolidWorks Corp., Waltham, MA, USA), and ANSYS Workbench software version 19.0 (Ansys, Inc., Canonsburg, PA, USA) was used for the finite element calculation. Screw and screw–cement–bone models were established by meshing. A grid convergence analysis was conducted, and the network type was a 10-node tetrahedral element, as shown in Fig. 5A. The results indicated that the maximum stress difference was less than 5% and that the model converged. Considering the calculation accuracy, speed, and stress distribution of the key parts, a 0.04 mm element size was adopted for mesh division in all calculations, and 0.04 mm was also used for all contact surface elements [24, 25].

Screw (Fig. 5B) and screw–bone cement–bone models (Fig. 5C) were established according to the elastic modulus and Poisson’s ratio of each material in the model [26, 27]. The screw model is a homogeneous, continuous, isotropic, and linearly elastic material [17, 18]. Bone is also considered to be homogeneous, isotropic, and linearly elastic [28]. In the screw–bone cement–bone model, friction contact exists between the screw, bone cement, and bone block [18, 29]. The maximum shear force test was used to evaluate the stress distribution at the screw fracture, and the maximum axial pullout test was used to evaluate the stress distribution at the screw–cement–bone interface.

Table 1 lists the mechanical properties of the investigated material [17, 29].

Table 1.

Properties of materials used in this study

Material	E (Young’s modulus)	v (Poisson’s ratio)
Pedicle screw (Ti alloy)	110 GPa	0.3
Polymethylmethacrylate (PMMA)	2.27 GPa	0.46
Cancellous bone model (rigid polyurethane foam)	320 MPa	0.25

Open in a new tab

Statistical analysis

All statistical analyses were performed using SPSS software version 21 (IBM Corp., Armonk, NY, USA); a P value < 0.05 was considered significant. Comparisons of all indices between the two groups were performed using a two-tailed Student’s t-test or analysis of variance.