Intervertebral disc kinematics in active duty Marines with and without lumbar spine pathology

Abstract Military members are required to carry heavy loads frequently during training and active duty combat. We investigated if operationally relevant axial loads affect lumbar disc kinematics in forty‐one male active duty Marines with no previous clinically diagnosed pathology. Marines were imaged standing upright with and without load. From T2‐weighted magnetic resonance images, intervertebral disc (IVD) health and kinematic changes between loading conditions and across lumbar levels were evaluated using two‐way repeated measures analysis of variance tests. IVD kinematics with loading were compared between individuals with and without signs of degeneration on imaging. Linear regression analyses were performed to determine associations between IVD position and kinematic changes with loading. Fifty‐eight percent (118/205) of IVDs showed evidence of degeneration and 3% (7/205) demonstrated a disc bulge. IVD degeneration was not related to posterior annular position (P > .205). Changes in sagittal intervertebral angle were not associated with changes in posterior annular position between baseline and loaded conditions at any lumbar level (r < 0.267; P = .091‐.746). Intervertebral angles were significantly larger in the lower regions of the spine (P < .001), indicating increased local lordosis when moving in the caudal direction Disc height at the L5/S1 level was significantly smaller (6.3 mm, mean difference = 1.20) than all other levels (P < .001) and baseline posterior disc heights tended to be larger at baseline (7.43 mm ± 1.46) than after loading (7.18 ± 1.57, P = .071). Individuals with a larger baseline posterior annular position demonstrated greater reduction with load at all levels (P < .002), with the largest reductions at L5/S1 level. Overall, while this population demonstrated some signs of disc degeneration, operationally relevant loading did not significantly affect disc kinematics.

kinematic changes between loading conditions and across lumbar levels were evaluated using two-way repeated measures analysis of variance tests. IVD kinematics with loading were compared between individuals with and without signs of degeneration on imaging. Linear regression analyses were performed to determine associations between IVD position and kinematic changes with loading. Fifty-eight percent (118/205) of IVDs showed evidence of degeneration and 3% (7/205) demonstrated a disc bulge. IVD degeneration was not related to posterior annular position (P > .205). Changes in sagittal intervertebral angle were not associated with changes in posterior annular position between baseline and loaded conditions at any lumbar level (r < 0.267; P = .091-.746). Intervertebral angles were significantly larger in the lower regions of the spine (P < .001), indicating increased local lordosis when moving in the caudal direction Disc height at the L5/S1 level was significantly smaller (6.3 mm, mean difference = 1.20) than all other levels (P < .001) and baseline posterior disc heights tended to be larger at baseline (7.43 mm ± 1.46) than after loading (7.18 ± 1.57, P = .071). Individuals with a larger baseline posterior annular position demonstrated greater reduction with load at all levels (P < .002), with the largest reductions at L5/S1 level. Overall, while this population demonstrated some signs of disc degeneration, operationally relevant loading did not significantly affect disc kinematics.

K E Y W O R D S
intervertebral disc bulge, intervertebral disc degeneration, low back pain, lumbar spine, upright MRI

| INTRODUCTION
Military members are required to carry heavy loads frequently during training and combat. During operations, Marines carry a minimum operational load of 11.3 kg in the form of ballistic protection, which can quickly escalate with the addition of necessary equipment to over 45 kg, exceeding the recommended load carriage limit of 33 kg. 1 Higher rates of intervertebral disc (IVD) degeneration has been observed to occur at a higher frequency in military populations compared to similarly-aged civilians. 2 It is thought that load-induced changes in IVD health may play a role in the development of clinical back pathology in this population. However, the association between operational loading, disc degeneration, and clinical spinal pathology (ie, bulge, herniation) has not been explicitly explored.
Heavy axial loads alter natural spinal posture, which may fatigue paraspinal musculature necessary for stabilization. 3 This may increase a Marines vulnerability to IVD injury and an increased rate of IVD degeneration over time. 4 Furthermore, individuals with disc degeneration demonstrate not only decreased whole lumbar range of motion, but also decreased intervertebral range of motion, specifically at the levels with degenerated IVDs. [4][5][6] This decreased range of motion may alter axial distribution of weight, affecting compression and shear forces at intervertebral joints. 7,8 Previous investigations on the effect of load and position on IVD kinematics (IVD height and intervertebral angular changes) in Marines demonstrated that as local lumbar flexion increased, decreased anterior and increased posterior IVD height occurs under operational loading conditions. 4,9 However, these previous investigations did not examine changes in posterior annular position (defined as focal or asymmetric extension of the disk beyond the vertebral border 10 ) with load, or the influence of disc health on kinematic loading responses. Evidence of IVD kinematic changes in response to postural alterations suggests that axial loading may also affect more specific features of disc morphology, such as annular position.
Disc morphology is often used as an indicator of IVD health, and changes in disc morphology are observed with disc degeneration and injury. 11 Changes in disc morphology with degeneration are thought to be a result of decreased proteoglycan concentration within the nucleus pulposus leading to loss of hydration and ultimately a decrease in disc height over time, or destabilization of the disc due to an annular or nucleus pulposus injury. 12 Although IVD herniation is apparent and well defined, the current literature does not provide a clear clinical definition for the term disc bulge, implicating its dependence on individual patient characteristics. Furthermore, clinically relevant changes in kinematics could include, but are not limited to, significant posterior annular protrusion compressing neural elements, loss of IVD height mimicking fusion, and resultant intervertebral angular derangements.
In order to further understand the influences of load on IVD kinematics in active duty Marines, the purpose of our study was to (a) investigate the effect of operationally relevant load on IVD height, intervertebral angle, and posterior annular position in the lumbar spine, and (b) to compare IVD kinematics between Marines with and without disc degeneration. We hypothesized that under increased axial load from tactical equipment, Marines' lumbar IVDs would demonstrate increased posterior displacement of the annulus fibrosus compared to baseline. Additionally, we hypothesized that Marines with IVD degeneration would exhibit decreased disc height and IVD angles compared to those with nondegenerated IVDs.

| Study design
This is a retrospective analysis of lumbar spine imaging data with repeated-measures design. Independent variables were loading condition and disc degeneration on intervertebral angle, posterior annular position, lordosis, and IVD height.

| Volunteers
Utilizing patients from a previous study, a total of 43 male active duty

| Load carriage
Marines were scanned naturally standing without load and standing with body armor (11.3 kg). The 11.3 kg body armor was used because it is minimum protective equipment that Marines are required to wear during military operations/training. The body armor was retrofitted to remove any metallic components to ensure compatibility with MRI.
Marines were not provided instruction on how to assume each position, but were asked to hold each position steady for the duration of the entire MRI acquisition (approximately 3 minutes).

| Image analysis
Postural measurements (IVD height, IVD angle) were generated from upright MRI images in each load configuration using a previously validated algorithm 13 using OsiriX. 14

| Disc grading
All lumbar discs were graded for disc degeneration using the Pfirrmann grading scale. 17 The Marines' discs were separated into degenerated or nondegenerated groups based on Pfirrmann grade; IVDs with a Pfirrmann grade of III or more were assigned to the 3 | RESULTS

| Participant demographics
Complete image datasets were analyzed from 41 male active duty Marines (Table 1)

| Effect of axial load on local lordosis and disc heights
Changes in sagittal intervertebral angle were not associated with changes in posterior annular position between baseline and loaded conditions at any lumbar level (r < 0.267; P = .091-.746). Intervertebral angles were significantly larger in the lower regions of the spine (P < .001), indicating increased local lordosis when moving in the caudal direction. There was also a trend for the main effect of load on intervertebral angles, in that angles were larger (more lordotic) at baseline (7.15 ± 1.63) than with load (6.77 ± 1.85, P = .064).
There was a main effect of level on disc height, in that disc height at the L5/S1 level was significantly smaller (6.3 mm, mean difference = 1.20) than all other levels (P < .001). There was a trend for baseline posterior disc heights to be larger at baseline (7.43 mm ± 1.46) than after loading (7.18 ± 1.57, P = .071). Additionally, there was a significant interaction (P = .006) between axial load and posterior disc height across levels, such that while the L1/2 disc exhibited an increase in disc height with loading, all other levels exhibited a decrease in disc height. However, of these differences, only the L3/4 and L4/5 discs were statistically significant (P < .022).

| Effect of axial load on posterior annular position
Posterior annular position at baseline was found to be different across lumbar levels ( F I G U R E 3 Intervertebral disc (IVD) measurements across lumbar levels in 41 active duty Marines. The Marines' posterior annular position was recorded in the standing unloaded (white) and standing loaded (gray) conditions at all levels. There was a significant main effect for IVD level (P < .035), but no effect of loading (P = .363) demonstrated that disc bulge was not found to increase at lower lumbar segments. Contrarily, the addition of axial load appeared to cause reduction in posterior annular protrusion, indicating decreasing disc bulge at lower lumbar levels. In this study, we limited our analysis of disc bulge to the midsagittal plane. The prevalence of central versus paracentral disc bulge in healthy populations is unclear. In symptomatic cases, if a disc is protruded or extruded, then paracentral location is most commonly observed. 25,26 However, even when a disc bulge is paracentral, it is most often diffuse, and able to be seen to some extent in the midsagittal plane. 26 In asymptomatic individuals or in individuals with mild disc bulges-not protrusions or extrusions-the location of the bulge is more likely to be central. 26 As diagnosed spinal pathology was an exclusion criterion for this study, and the lack of paracentral disc bulges was visually confirmed, the volunteers in this study are within the latter group.
There are three main limitations to the study. To acquire the imaging data for analysis, an elastic band was used to gently secure the coil to the volunteers' low back. While this may influence posture, the band and the coil are relatively light (approximately 1 kg), and the posture of patients did not appear to change when it was attached. Additionally, we were unable to resolve the PLL due to the short T2 relaxation of collagenous tissues. Development of new ultra-short TE pulse sequences may provide insight into the health and function of the PLL under axial load. We used the posterior border of the annulus fibrosis (AF) as a proxy for disc bulge, which may not be the most accurate characterization of IVD movements.
Rather, we may be observing abnormal nucleus pulposus migration with respect to adjacent vertebral bodies. 23 Lastly, Marines are exposed to significant conditioning, as well as physical demand, which may contribute to differences in findings compared to age and sex matched civilians. Such characteristics may impact the external validity of our findings and limit their applicability to the general population. Future work should be directed toward localizing IVD migration in multiple planes to better characterize kinematic responses to axial load. The findings of this analysis warrant further investigation into axial loading and resultant IVD kinematic changes in hopes of elucidating its unique alterations to disc morphology in a highly active population.

ACKNOWLEDGMENTS
The authors thank the Marines from the 1st and 5th Regiments who supported this effort.