A new concept of motion preservation surgery of the cervical spine: PEEK rods for the posterior cervical region

PEEK rods provided greater ROM for both models in comparison to the Ti rod for all diameters. The ultimate results of the study are outlined below (an interpretation of the results is shown in Table 1).

4. Discussion

The presented FE study is based on posterior cervical decompression surgery and instrumentation. In order to deal with the postoperative problems of this surgery, a spinal surgeon must be clear about cervical spinal stability, sagittal balance, spinal motion segment and the concepts of cervical laminectomy and laminoplasty.

According to White and Panjabi [10,11], spinal stability can be defined as the ability of the spine to protect the neural elements and its integrity under physiological forces, without causing neurological deficits or intractable pain. Normally, the cervical spine has a lordotic curvature. The lordosis angle is the angle between the lines parallel to the inferior end plate of the second cervical vertebra (C2) and superior end plate of the seventh cervical vertebra (C7). An alternative method to measure the lordosis angle is to find the angle between the lines tangential to the posterior borders of C2 and C7, or the angle between the perpendicular lines to C1 (axis) and the lower end plate of C7 (Fig. 13). In a normal vertebral column, the cervical region possesses a lordosis angle of 9 degrees on average (2–24 degrees), but a wide variety of measurements have been taken in previous studies [12]. In a geriatric cervical spine, normal lordosis decreases, and kyphosis may progress as a consequence of the loss of intervertebral disc height. The ligaments, intervertebral discs and facet joints are important structures that maintain the stability of the cervical vertebra. Failure of these structures with advancing age is associated with a decrease in the intervertebral disc height, osteophytes, ligament calcifications and disc degenerations, leading to the loss of compliance and physiological contour.

OPEN IN VIEWER

Fig. 13.

Two methods to calculate the cervical lordosis angle.

A spinal motion segment is composed of two adjacent vertebrae and an intervertebral disc. ROM of the cervical spine is calculated according to the lateral roentgenograms at flexion and extension, and each is approximately 10° per level [13]. The principal spinal motions are forward-backward and lateral translations, compression and distraction, and rotational motions, namely flexion-extension, lateral flexion and axial rotation. Generally, these motions occur in coupling motions (two or more together) rather than isolated motions, such as lateral flexion with axial rotation, or translation together with anterior flexion, etc.

This FE study is planned on instrumentation after laminectomy. Cervical laminectomy is one of the main surgical procedures preferred in the treatment of degenerative disorders, such as cervical stenotic myelopathy, excision of medullary tumors via a posterior approach, trauma surgery, multilevel degenerated discs and osteophytes with myelopathy, ossified posterior longitudinal ligament, and congenital cervical spinal stenosis. Laminectomy is a destructive procedure because the spinous processes and laminae are excised by rongeur up to the facet-laminae border (Fig. 14). In some cases after extensive decompression (more than 25% of facetectomy during laminectomy), postoperative kyphosis is inevitable and posterior stabilization is necessary [14]. Also, multilevel laminectomy may lead to segmental instability over time [2–4]. Postlaminectomy kyphosis usually leads to severe axial pain, especially a few months after surgery. The destruction of sagittal balance and iatrogenic muscular damage are the cause of this pain. Cervical kyphosis is a self-propagating process because the axial load tends to deform the spine contour over time as a result of longer bending movement, and deformity begets deformity [15].

OPEN IN VIEWER

Fig. 14.

Illustration of posterior cervical laminectomy. The figure shows two levels of extracted lamina.

Stand-alone cervical laminectomy for the aforementioned cervical pathologies has become increasingly rare due to the risk of postlaminectomy kyphosis and should only be used for a selected patient group with relative stiff lordotic cervical spines, taking care not to disrupt facets and C2 and C7 muscle attachments [16–19]. Hence, fixation of the posterior cervical region is necessary after the laminectomy procedure.

The laminoplasty technique via a unilateral approach was described as an alternative to laminectomy and is the unique alternative to laminectomy at the moment [14]. It is thought to decrease postoperative kyphotic deformity incidence [8] because posterior ligamentous and muscular elements, and the cervical anatomy are preserved [13]. In this procedure, the posterior spinal elements are kept open by titanium miniplates (Fig. 15). In the study by Pepa et al., the postoperative kyphotic deformity rate of laminoplasty was lower than that of laminectomy (without instrumentation) (0% versus 12.5%). However, the authors also reported a minimal reduction of approximately 18% in cervical lordotic alignment after the laminoplasty technique. The results of a meta-analysis in 2003 showed that the deformity rate following laminoplasty was comparable to that following posterior laminectomy [8]. Thus, laminoplasty and laminectomy may both cause sagittal plane deformities by destroying or weakening the dorsal tension band, eventually leading to postoperative spinal deformity, and laminoplasty does not completely preserve the lordotic angle in a long-term follow up [4]. Hence it is necessary to fixate the spine dorsally [8].

OPEN IN VIEWER

Fig. 15.

In the laminoplasty technique, unilaterally cut laminae is fixed by titanium miniplates.

To address these complications, the subaxial cervical spine is stabilized posteriorly by a variety of techniques, such as interspinous wiring using bone grafts, laminar clamps, hook plates, lateral mass screw and plates, and transpedicular screw and rod systems. Among these procedures, interspinous wiring was the most common technique before the advent of lateral mass screw fixation for cervical multilevel subaxial stabilization (Fig. 16).

OPEN IN VIEWER

Fig. 16.

Left: interspinous wiring. Right: the cervical lateral mass screw stabilisation system.

However, all the systems mentioned above are rigid stabilization systems. Rigid fixation has been accepted as the gold standard in the treatment of spinal disorders. But unfortunately it decreases the normal physiological ROM of the spine, eventually leading to undesirable complications. It has numerous drawbacks such as increased mechanical stress on the adjacent segment and pseudoarthrosis and fatique of the implants. This in turn causes long-term degenerative changes necessitating additional fusion surgery. For the cervical region, totally, a prevalence rate of adjacent segment disease ranging from 9% to 17% is reported [20]. The annual incidence of adjacent segment disease needing additional surgery is between 1.5% and 4%.

As a result, concerns regarding the long-term viability of the spinal motion segments adjacent to fusions is increasing and alternative spinal implants were developed on unsatisfactory outcomes of rigid implants. These motion preservation systems include anterior and posterior dynamic stabilization systems and are commonly used in the last two decades [20–26].

Our presented system is based on the posterior polyaxial screw and PEEK rod system. PEEK material in posterior dynamic systems can restore stability of the spine without adverse stress-shielding effects that have often been found with ‘rigid’ fusion devices made of ‘rigid’ Ti alloys. The elastic modulus of PEEK is 3.2 GPa and 114 GPa of Ti so it is a flexible material permitting physiologic motion of the spine [27].

Numerous studies in the literature have examined motion preserving PEEK posterior spinal systems [27–31]. But almost all of these are about the lumbar region. Besides, there are more than 23,000 journals in the medical literature about FE. Only a few of them are about the biomechanical efficacies of cervical systems and none of them are about motion preserving the posterior cervical systems [32–35].

To the best of our knowledge, just one semi dynamic system was described by Cusick et al. [36] in 2018. They described a stiff crisscross facet wiring mechanism, which was suggested to be less rigid than the conventional cervical plates for flexion motion. This study is still far more behind encountering the needs of a fully dynamic system as our PEEK rod but it deserves to be mentioned here because it is the only published non-rigid posterior cervical stabilization system.

As a result, the posterior cervical PEEK rod system seems to be a reasonable concept to overcome the problems of diminished ROM after posterior cervical instrumentation. Our FE model is conducted on two levels of laminectomy. It is a pioneer study, so future studies should be planned on more laminectomy levels and concurrent cadaver studies. Even for two levels of laminectomy, PEEK rods allow more ROM than Ti rods. Although these differences cannot be statistically verified (because this is a FE study), there are considerable differences. Differences of motion less than one degree are accepted to be in the error range in clinical settings. We believe that on the basis of computer analysis of FE models these differences are significant when they are considered for only one segment of the spine. We hope that subsequent studies based on this study on more levels will answer the purpose that PEEK rods let more ROM in the physiologic limits in cervical spine.

A limitation of the study is that a FE study examining only one specimen does not provide enough data for a statistical analysis. FE studies are dissent from statistical analysis but they expand the horizon for the supervening biomechanical, cadaver and clinical studies [37–43].

According to the ultimate accounts of this study, it can be clearly observed that the range of motion is higher with the PEEK rod than the Ti rod, especially for the thinner PEEK rods. For all sorts of motion (i.e. flexion, extension, lateral bending, axial rotation), the stiffness of the posterior stabilization system decreases as the diameter of the rod decreases and the difference between the ROM values becomes more striking, especially for the 3 mm PEEK rod. As the max stress measurements imply, 3, 3.4, 3.7 and 4 mm diameter PEEK rods are rational alternatives to posterior Ti rods of 4 mm diameter, which are used widely in spinal surgery. For instance, considering the axial rotation at the C5-6 level, a 3 mm diameter PEEK rod provides 0.959° of motion whereas the Ti rod provides 0.3408°, which verifies that it is almost three times more flexible. At the C4-5 level, the range of motion for axial rotation is almost four times as high for the PEEK rod than for the Ti rod (1.086° vs 0.2881°).

Another limitation is the simplification of the cervical model. The tissue material of different parts of the cervical spine obviously vary with age and gender. The material properties selected for the current model were derived from previous in vitro experiments and were widely referenced because our model was based on healthy men. However, as age increases, cervical geometry and ligament laxity will change. All these factors should be further studied to obtain accurate changes in the corresponding segments after surgery. The interaction of the instrument composed of lateral mass screws and Ti or PEEK rods and the spine tissue will alter depending on the age, bone mineral density and ligament properties and should be analysed in future studies.

Subsequent FE studies on dynamic cervical spinal systems are necessary for more physiologic surgical interventions. Following these studies, we need to perform cadaver and clinical studies in order to assess the long-term effects of posterior cervical PEEK rods.

In posterior stabilization, PEEK rods allow greater ROM for the cervical spine after laminectomy and stabilization, especially for axial rotation. This eventually protects the adjacent intact segment from degeneration, kyphosis or listhesis of the operated levels and other long-term subsequent fusions.

The ROM degrees of the alternative rods may seem to be close to each other in our measurements because this model is based on two levels of laminectomy. But even for two levels, there is a considerable difference between the two rods, especially when AR is considered. Maximum stress measurement results indicate that posterior cervical PEEK rods allow more ROM than the Ti rods, encouraging us to further study this concept. For more levels of laminectomy, particularly thinner PEEK rods should provide more physiological motion and more comfort for the patient in daily physiologic head motions and prevent long-term negative consequences of spinal fusions.