Chapter 21 Biomechanical Testing
Background
In the second half of the 20th century, the research area of biomechanics encompassed a problem that, at one time, was of interest to Leonardo daVinci–the human spine.1 In the 1970s and 1980s, in particular, there was a rapid increase in the biomechanical analysis and quantitative understanding of the anatomy of the spine and clinical issues related to its treatment.2–8 This new insight enabled researchers to design and develop devices that aimed to restore normal physiologic movement.9,10 However, one of the unforeseen consequences of this flurry of scientific activity was a lack of standards.9,11–13 Depending on the laboratory and the application, devices were being evaluated under different conditions, making comparison difficult. Limitations related to the peculiar nature of the spinal anatomy and testing made standardization difficult. Eventually, however, consensus was achieved and standards evolved.9,13 From the economic point of view, spine biomechanics seems to have delivered on its promise. According to Epsicom, a global market research company, the spinal implant market saw a growth of about 8 billion dollars in 2008. It is estimated that by 2012, the worldwide spinal market will realize revenues of 10 billion dollars.
Another avenue of inquiry has been aiming at resolving clinical issues without the use of any “mechanical” devices. This field, tissue engineering, has started to show great promise in the field of spine biomechanics. With U.S. federal funding available for stem cell research, a global market of 4.8 billion dollars, and about 200 companies working on designing newer and better orthopaedic biomaterials, the future is bound to see a growing influence of tissue engineering in spine deformity correction. Although challenges exist,14,15 our understanding of issues related to the regeneration of the nucleus pulposus16,17 and anulus fibrosus18 has increased manifold. As a case in point, much interest is being paid to scaffolding.19,20 Interested readers are advised to review a classic publication by Lanza et al.21
Back- and neck-related issues led to 86 billion dollars in health care expenses in the United States from 1997 to 2006.22 In the same decade, an increase of nearly 50% was found in the number of patients seeking spine-related healthcare expenditure.22,23 In parallel, a 65% increase in health care expenditure in general was measured. Numerous types of surgical procedures are performed on the spine to prevent further deterioration of spinal components or escalation of pain, and various devices are being conceived, designed, tested, and implanted to aid in these treatments. Most of this instrumentation—for example, interlaminar hooks, transpedicular screws, interbody spacers, and cages—is relevant to spinal fusion.24 The goal of such instrumentation is to fuse two or more vertebrae together to eliminate pain and allow the patient to return to normal activities. Alternatives to fusion include the hydrogel-based prosthetic nucleus, the liquid polymer-based nucleus, motion preservation devices, and artificial discs.24
As shown in Figure 21-1, almost everything that is done in biomechanical testing flows from an existing spinal disorder and the perspective of the individual researcher. We do not claim that this algorithm is comprehensive, a case in point being the regulatory part of the process. Other variables include the types of perspectives (e.g., material science), concepts, and tests. Based on his or her perspective and the clinical objective, a researcher may come up with a concept of a solution, for example, a tissue-engineered nucleus for a damaged intervertebral disc. This concept is tested, proving or disproving a predefined hypothesis. The nature of the specific test may be purely mechanical, biomechanical, or based on biocompatibility. In the case of an engineered nucleus, for example, a test could be any of these (except in vivo, which is rare).20 For the nucleus, such a test could involve measurement of motion after surgical implantation in a cadaveric spine model or, perhaps, a purely mechanical study assessing its compressive modulus (i.e., a bench-type test). On the other hand, if the clinical objective is being met from a mechanical perspective, resulting in a mechanical device, the range of tests would include pure mechanical tests such as fatigue and wear tests. Determination of the chemical composition following corrosion and wear testing complements these mechanical tests. At some stage, testing using animal spine models, cadaveric spines, analogue spines, or computer simulations (i.e., in silico), and, eventually, clinical trials on human subjects, will follow.
Although all types of testing modalities are important in the process of concept evaluation and assessment as shown in the algorithm, this chapter focuses on three: bench type; in vitro, or, more appropriately cadaveric; and in silico testing of devices and engineered tissues under the overarching term of biomechanical testing. Moreover, we differentiate between construct testing and implant testing. The terminology, testing procedures, apparatuses, and protocols that have evolved over the years in testing of spinal implants also are reviewed. We also speculate regarding future prospects for biomechanical devices and note the areas that may need more attention from the spine biomechanics community.
Bench-Type Tests for Approval by the U.S. Food and Drug Administration
To evaluate the endurance and strength of orthopaedic implants, various mechanical and materials testing protocols have been proposed by ASTM International (formerly known as the American Society for Testing and Materials) as standards for testing of such devices under different dynamic and static loading profiles (Table 21-1). These protocols allow researchers to estimate the static strength and fatigue limits of an implant assembly and its individual components in a consistent way, thereby enabling a fair comparison of results. Guidelines also are proposed by ASTM and the International Organization for Standardization (ISO) for evaluation of fixation of the parts and loosening effect at the interface of implant components.
Data from these standardized tests are used by the medical device industry to seek approval for commercial distribution of devices from the U.S. Food and Drug Administration (FDA). Medical devices are categorized by the FDA in classes—class I, class II, and class III—based on the degree of regulatory control. Most class II devices, such as the pedicle screw–based instrumentation systems, require submission of a Premarket Notification 510(k), whereas class III devices—devices that pose a significant risk of illness or injury—require premarket approval (PMA; Fig. 21-2). Motion preservation systems, for instance, are categorized as class III devices. The test protocols listed in Table 21-1 pertain to class II devices. Class III devices also may be assessed using these protocols, but approval for commercial distribution of such devices requires submission of clinical data in support of the manufacturers’ claims.
Similar tests sometimes are carried out on ligamentous motion segments. These tests include subsidence tests, pull-out25 or push-out testing of pedicle screw systems26 and cages, respectively, and fatigue tests. Subsidence is a phenomenon in which one or both vertebral end plates adjacent to the implant collapse and allow the implant to move in, increasing the probability of deformity progression and worsening of the fusion.27 Static, quasi-static, or dynamic tests such as pull-out tests also are performed on pedicle screws to measure bone-implant interfascial strength under such forces. New ASTM guidelines are available for the assessment of facet replacement technologies, wear characterization, and motion preservation systems such as artificial discs.
Wear testing is carried out on a wear simulator (Fig. 21-3). One such simulator (MTS Bionix, MTS Systems Corp., Eden Prairie, MN) consists of six active stations (test stations) and one control station.28
Polymeric components in a disc replacement device are soaked in a bath for a week before the test. These are then cleaned and dried in accordance with ASTM F2423-05 (see Table 21-1). Flexion-extension, lateral bending, and rotations are simulated under a constant preload as per ASTM standards. Mass measurements are performed both before and after testing to assess the wear rate. Particulate characterization and element contributions are evaluated using computer- controlled scanning electron microscopy.
Analysis
Elastic zone: The amount of total deformation that offers resistance to the applied load. It is measured by evaluating the tangent to the curve at the load that causes maximum deformation (Fig. 21-4; points 5 and 6).
Elastic zone stiffness: This is the stiffness that characterizes the amount of elastic (or recoverable) deformation of the specimen.
Energy dissipation: To characterize the viscoelasticity or plasticity of the specimen being loaded, the area enclosed by the load-displacement curve is evaluated. This quantity provides a measure of the dissipated energy.
Neutral zone: The amount of unrecovered deformation once the specimen is under no load. In cycle 3 shown in Figure 21-4, NZ is the neutral zone. It also may be defined as the part of the range of motion wherein the specimen offers the least resistance to the applied deformation.
Neutral zone stiffness: The stiffness of the specimen in the neutral zone, determined by the slope of the load-displacement curve at the point of no deformation
Preconditioning: Cycles of load applied to the specimen—intact or otherwise—to mitigate the impact of the viscoelastic nature of the tissues. From Figure 21-4, cycles 1 and 2 are the preconditioning cycles.
Range of motion (ROM): The linear or the angular distance that a specimen (intact or injured or construct) travels in a plane with the application of load in that plane. From the load-displacement curve of Figure 21-4 the ROM can be calculated as (+ROM) − (−ROM).
Relative range of motion (RROM): The relative motion for the entire spine or a segment or even a vertebral body with respect to the static mounting platform
Sigmoidity: A measure of the non-linearity present in the mechanical behavior of the specimen,29 calculated as the ratio of the neutral zone stiffness and elastic zone stiffness.
Stiffness: The mechanical resistance of a specimen to an applied load, measured by the slope of the load-deformation or load-displacement curve along a linear region or regions in a nonlinear curve.
In Vitro Testing
The human spine is a complex structure composed of hard and soft, active and passive tissue. This structure has multiple degrees of freedom at each one of several joints formed by intervertebral discs. Ideally, from a biomechanical and biochemical point of view, the most physiologically relevant model for testing the efficacy of a device, surgical technique, or engineered tissue is the human spine of a live subject. However, this is not a practical option. In vitro testing offers significant advantages, even though factors such as intra-abdominal pressure and muscular forces are hard to replicate.13 In vitro studies have the advantages of the possibility of standardization, ease of estimation of the impact of a surgical procedure, or a simulated injury or stabilization using an implant, because the loads can be varied with relative ease. Such protocols enable researchers to compare different devices designed and developed for the same clinical requirement. Once the device components have been tested using protocols cited in Table 21-1, in vitro testing brings their performance evaluation closer to in vivo use in patients.
Terminology
Some of the terms most commonly used in in vitro studies are defined in this section.
Anatomic planes: To make it possible to specify the locations and angular configurations of the vertebrae, a coordinate system is defined that has three mutually orthogonal planes: the sagittal plane (side view), the frontal or coronal plane (front view), and the transverse plane (top view). Figure 21-5 shows the three anatomic planes along with the terminology for forward/backward, left/right, and up/down directions.30
Center of rotation (COR) or instantaneous axis of rotation (IAR): In a general planar motion, the axis of rotation may move. If this movement is broken down into steps, the instantaneous axis of rotation can be identified at every step of the motion. Such an axis may pass through the rigid body (in the case of a spinning top) or lie outside it (in the case of the flexion or extension of a spinal segment). To specify the IAR completely, one must provide three numbers: two for translations and one for rotation or any combination of these parameters. The IAR is specified only for plane motion, not for 3D motion—that is, there is no IAR for lateral bending or axial rotation because these involve 3D motion, whereas flexion and extension are considered planar motions for all practical purposes.31 However, there is evidence that relatively small coupled motions are present even in flexion and extension.32
Coordinate system: An orthogonal, right-handed, 3D reference system that makes it possible to define the position and motion of vertebral bodies. In Figure 21-6, the x, y, and z axes represent the three orthogonal directions with the origin of the coordinate system located at the base.29 The positive x-axis represents the left lateral direction, whereas the positive y-axis represents the rostral direction and the positive z-axis represents the ventral (anterior) direction. Such a system is known as a global coordinate system. A reference system can be local, however, in the sense that it allows for the position and motion of rigid bodies to be defined with respect to each other. Wilke et al.29 suggest, for most cases, “the mid-point in the frontal plane of the dorsal (posterior) margins of the two adjacent” vertebral endplates as the origin of the local coordinate system.
Degrees of freedom: The number of independent coordinates necessary for complete specification of the position of a particle or a rigid body in space. Under an applied load, a rigid body may move, in total, in six directions: that is, it has 6 degrees of freedom: three translational and three rotational. In comparison, a particle can have only 3 translational degrees of freedom. A general motion by the vertebra may be broken down into six components of these pure motions.
Envelope of the helical axis of motion: The surface generated by various helical axes of motion of a moving rigid body.
Follower load: A compressive load applied to the spinal segment (through strategic points on each vertebral body) that aims at minimizing the coupled flexion-extension changes in motion and shear force in the disc by following the COR of each functional spinal unit of a specimen.33 In cadaveric experiments, a compressive follower load is applied to the specimen to mimic the upper body weight and muscle force application on the lumbar spine. The application of follower load works well only in flexion and extension.31 Bilateral cables are used to apply this load.
Functional spinal unit (FSU) or motion segment: The macrostructural unit of the spine, representing the broad mechanical behavior of two adjacent vertebrae, ligaments, the intervening intervertebral disc, and zygapophyseal (or facet) joints. Studying the biomechanics of an FSU is convenient and relatively straightforward. Figure 21-7 shows with an FSU with an intact intervertebral disc.
Helical axis of motion (HAM) or screw axis motion: As an alternative to x, y, and z coordinates and Euler angles, motion of a rigid body can be decomposed into a translation and rotation about the axis of translation. This axis is known as the screw axis or helical axis of motion. It can be visualized by observing the motion of a screw being driven into a pedicle of a vertebra. As the screw is being tightened, it not only rotates but also translates into the pedicle along an axis running through the screw. Consistent with the 6 degrees of freedom for a freely moving rigid body in the 3D coordinate system, six scalar quantities are required to define 3D motion using HAM: two for the orientation of the axis, two for its position, one for the amount of rotation about the axis, and one for the amount of translation along the axis. The helical axis of motion, although difficult to visualize, particularly for clinicians, may provide quality of motion when compared with an end-point parameter such as range of motion (ROM), which determines simply the quantity of motion. For example, it recently was found that axial rotation causes the helical axes to migrate dorsally, correlating well with high facet joint forces.34
Injured specimen: A spine specimen with existing or simulated clinical pathoanatomy in terms of injury of ligaments, disc(s), and/or bony tissue
Instability: From a purely mechanical perspective, instability of a specimen undergoing in vitro testing may be characterized by a significant change in the range of motion relative to the intact specimen, for example, 3.5 mm of translation will make a specimen unstable.35 Instability may be related to spinal degeneration and pain.
Intact spine specimen: A portion of the fresh-frozen cadaveric spine consisting of one or more contiguous functional spinal units with intact ligaments and disc(s). Fascia, muscles, and fatty tissues are dissected.
Muscle force simulator (MFS) or replicator (MFR): A system that simulates muscle forces on a spinal motion segment. Unfortunately, this experimental setup was found to be so arduous that repeating similar experiments became unrealistic.36,37
Plane motion: Motion characterized by translation(s) and/or rotation(s) in a single plane. For instance, flexing the vertebra (or, in other words, forward bending) is a plane motion occurring in the sagittal plane. In Figure 21-6, flexion will be fully specified in the y-z plane. Flexion of the vertebral body at the top will involve not only rotation but also translation. Furthermore, there may be some varying degree of out-of-plane motion as well.38
Primary and coupled motion: In terms of plane motion, the motion occurring in the same direction as the one in which the load is applied is known as primary motion. The out-of-plane motion is known as coupled motion.
Primary loading directions: In most cases, spinal motion segments are tested in the following directions: flexion-extension, left-right bending, and left-right axial rotation. Pure moments are applied in one of these directions, and motion is measured. In a complex system such as a spine, application of a pure moment results in six motions. So, in an experiment that involves applying 6 pure moments on a segment, 36 load-displacement curves exist. A specimen may be loaded in a number of ways, which may be understood in terms of their orthogonal constituents or components in a global coordinate system. Reaction loads, for example, can be understood as being composed of three forces and three moments acting at a point of interest, such as the base of the specimen shown in Figure 21-6.
Relative motion: The motion of a rigid body with respect to another rigid body, for example, the motion of a vertebral body with respect to an adjacent vertebral body. However, the motion of a vertebral body relative to the static floor is absolute (or global) motion.
Rigid body: A system of particles in which the distance between any two particles remains unchanged regardless of external loads (forces or moments) applied. In other words, a rigid body does not deform. A rigid body is an idealization of a solid body of finite dimensions for the purpose of analysis. In construct testing, vertebrae are considered as rigid bodies.
Rotation: The vertebra in Figure 21-6 can rotate about three orthogonal axes in a clockwise (positive) or counter-clockwise (negative) direction. Curved arrows in Figure 21-6 show these degrees of freedom. In a 3D global coordinate system, Euler or Cardan angles specify the rotation of a rigid body.
Spinal construct: A portion of the spine instrumented with an implant or several implants of interest. Its characteristic motion is different from that of the intact spine. Figure 21-8 shows a spinal construct prepared for testing. The vertebra at the bottom is embedded in a polyester resin or low-melting-point alloy of choice for attachment to the test fixture, and a loading frame is rigidly secured to the superior-most vertebra for the application of loads.
Spine loading simulator: An apparatus to hold spine specimens and test them under different loading scenarios. Several research groups have come up with various designs of a loading simulator, ranging from fully automated to a system of pulleys and dead weights, for manual application of loads.
Three-dimensional motion: The type of motion seen in a rigid body in a global coordinate system that is free to translate or rotate in one or more of its six degrees of freedom
Translation: As shown in Figure 21-6, a vertebra can translate along three axes, that is, positive or negative x, y, and z axes. These are shown by straight arrows.
