How are adjacent spinal levels affected by vertebral fracture, and by vertebroplasty? A biomechanical study on cadaveric spines

Background Context: Spinal injuries and surgery may have important effects at neighbouring spinal levels, but previous investigations of adjacent-level biomechanics have produced conflicting results. We use ‘stress profilometry’ and non-contact strain measurements to investigate thoroughly this long-standing problem. Purpose: To determine how vertebral fracture and vertebroplasty affect compressive load-sharing and vertebral deformations at adjacent spinal levels. Study Design: Mechanical experiments on cadaver spines. Methods: 28 cadaveric spine specimens, comprising three thoracolumbar vertebrae and the intervening discs and ligaments, were dissected from 14 spines aged 67-92 yrs. A needle-mounted pressure transducer was used to measure the distribution of compressive stress across the antero-posterior diameter of both intervertebral discs. ‘Stress profiles’ were analysed to quantify intradiscal pressure (IDP), and concentrations of compressive stress in the anterior and posterior annulus. Summation of stresses over discrete areas yielded the compressive force acting on the anterior and posterior halves of each vertebral body, and the compressive force resisted by the neural arch. Creep deformations of fractured and adjacent vertebral bodies under load were measured using an optical MacReflex system. All measurements were repeated following compressive injury to one of the three vertebrae, and again after the injury had been treated by vertebroplasty. The study was funded by a grant from Action Medical Research, UK ($143,230). Authors of this study have no conflicts of interest to disclose. Results: Injury usually involved endplate fracture, often combined with deformation of the anterior cortex, so that the affected vertebral body developed slight anterior wedging. Injury reduced IDP at the affected level, to an average 47% of pre-fracture values (P<0.001), and transferred compressive load-bearing from nucleus to annulus, and also from disc to neural arch. Similar but reduced effects were seen at adjacent (non-fractured) levels, where mean IDP was reduced to 73% of baseline values (P<0.001). Vertebroplasty partially reversed these changes, increasing mean IDP to 76% and 81% of baseline values at fractured and adjacent levels, respectively. Injury also increased creep deformation of the vertebral body under load, especially in the anterior region where a 14-fold increase was observed at the fractured level and a three-fold increase at the adjacent level. Vertebroplasty reversed these changes also, reducing deformation of the anterior vertebral body (compared to post-fracture values) by 62% at the fractured level, and by 52% at the adjacent level. Conclusions: Vertebral fracture adversely affects compressive load-sharing and increases vertebral deformations at both fractured and adjacent levels. All effects can be partially reversed by vertebroplasty.