Lumbar Spine Behavior and Injury Due to Cyclic Loading

Thumbnail Image
Limited Access
This item is unavailable until:



Journal Title

Journal ISSN

Volume Title


Low back pain (LBP) is one of the most prevalent conditions worldwide, estimated to affect 50-80% of adults at some point in their lifetime. However, up to 90% of patients that suffer from LBP do not receive a specific diagnosis. The etiology of injury that leads to pain in most LBP cases is largely unknown, though there are some known risk factors. Those regularly exposed to repeated lumbar flexion and compression (i.e., helicopter pilots, high-speed boat operators, truck drivers) are more likely to sustain lumbar injury and to experience LBP. There is a critical need to understand low-level injuries contributing to LBP and how they develop. The primary objectives for this project were to determine lumbar spine properties, behavior, and injury risk to elucidate factors contributing to low-level, pain-inducing injuries.Water content within the intervertebral discs (IVDs) plays a crucial role in the mechanical function of the lumbar spine. Common experimental tissue storage practices involve freezing tissue prior to mechanical testing to avoid degradation. However, it has been theorized that this practice may alter the in vivo hydration state of the lumbar spine IVDs. Additionally, porcine lumbar spine is a common animal surrogate for lumbar spine investigations, yet the hydration content in the porcine discs has not been established throughout both the nucleus pulposus (NP) and annulus fibrosus (AF). Intact porcine spines (89 IVDs) were stored in three different conditions (fresh, frozen with a saline-soaked wrap, frozen without saline) to establish the effects of freezing on IVD water content. IVD tissue from the NP, and three progressive AF regions from inner to outer AF were measured for mass percentage of water. Mean hydration values were found to be 88.8 ± 1.7% in the NP, and 79.6 ± 3.8%, 71.9 ± 3.7%, and 62.3 ± 3.3% from inner to outer AF. Hydration in all four disc regions were significantly different from one another and there were no significant differences in storage condition as a main effect. No meaningful trends were observed for the interaction between storage condition and disc region, suggesting freezing porcine lumbar spine does not affect the hydration within the IVDs. The lumbar spine is viscoelastic, so its behavior under load is complex and nonlinear. Under a constant load, a viscoelastic material will exhibit both time-dependent (creep) and stress-dependent (elastic) behavior. There is a critical need to characterize this complex lumbar spine behavior prior to injury to understand how injuries develop. Porcine lumbar spinal units (n=15) were loaded in repeated flexion-compression to characterize the non-injurious primary creep behavior inherent to viscoelastic tissues. Using a quasilinear viscoelasticity (QLV) approach, a computational optimization model determined two creep time constants suitable for all fifteen tests (β1 = 24s, β2 = 580s), then determined the creep and elastic parameters optimal for each individual specimen. Each model had high model accuracy (average R2=0.997) to the experimental data. The overall behavior was found to be 50% non-transient, with the transient contributions to the overall behavior to be 30% due to the long creep time constant (β2) and 20% due to the short creep time constant (β1). Results suggest lumbar spine behavior is highly time-dependent prior to injury and the QLV approach, deemed suitable for individual spinal components in previous work, is also a suitable approach for the complete lumbar functional spinal unit loaded in combined flexion-compression. Studies suggest some LBP could originate from lumbar endplate damage. Cadaveric human (n=16) and porcine (n=20) lumbar functional spinal units were loaded in long durations of repeated flexion-compression to observe endplate failure. Both human and porcine specimens were used to establish interspecies differences in injury between human and a common lumbar animal surrogate. Five human and twelve porcine specimens experienced at least one endplate fracture. Accelerated life analyses were preformed to produce the 50% endplate fracture risk curves based on compressive stress and cycles for both human and pig. The 95% confidence intervals overlapped, suggesting the pig was a suitable model for endplate failure, however, porcine specimens demonstrated an increased tolerance for endplate fracture compared to human. For the same compressive stress, human specimens fractured at approximately 25% of the duration that porcine specimens experienced fracture at 50% risk. The human injury risk curve can be immediately used for injury mitigation efforts in those exposed to this combined loading. The human and porcine injury risk curves provide necessary interspecies scaling information for interpreting human lumbar injury risk using past and future porcine experimental results. Finally, there is a need to capture soft tissue changes in the lumbar spine prior to endplate failure. Preliminary post-test microCT images from specimens loaded in repeated flexion-compression without endplate damage revealed that there may be changes in the IVD tissue prior to endplate failure due to repeated loading. Because the outer AF in healthy IVDs is innervated and degenerated discs have increased nerve density sometimes propagating as far into the disc as the NP, tissue damage in the IVD may be inducing pain. A novel microCT imaging technique was created to resolve individual NP and AF components within porcine IVDs, with resolutions up to 43 voxels/mm. This high-resolution, non-destructive imaging technique was then used to capture how NP and AF components change and potentially fail prior to endplate failure. Porcine lumbar spinal units (n=9) were loaded in repeated flexion-compression at a peak compressive stress of 2.75 MPa for loading durations below the 50% endplate fracture threshold developed previously. For specimens that were loaded the longest (15 minutes, 45 minutes, 60 minutes), post-test images revealed a smaller NP, a less distinct NP-AF boundary, and separation between lamellae compared to pre-test imaging and the control specimen. Small tissue changes captured with the presented high-resolution, high-quality, and non-destructive microCT imaging technique could have substantial implications for understanding low-level injuries contributing to low back pain. The studies presented in the following dissertation determine critical lumbar spine properties and behaviors that precede injury while directly assessing injury risk and injury patterns in the lumbar spine after repeated flexion and compression. These results improve our understanding of lumbar injury initiation and development, with a substantial impact on future prediction modeling, live animal studies, soft tissue imaging, and injury mitigation efforts.





Morino, Concetta (2024). Lumbar Spine Behavior and Injury Due to Cyclic Loading. Dissertation, Duke University. Retrieved from


Dukes student scholarship is made available to the public using a Creative Commons Attribution / Non-commercial / No derivative (CC-BY-NC-ND) license.