Vertebral compression fractures are a common and often debilitating complication of osteoporosis. Percutaneous vertebroplasty is a minimally invasive procedure that provides prompt pain relief with a low complication rate. Vertebral fractures increase significantly the lifetime risk of further fractures, and new vertebral fractures at the adjacent or non-adjacent vertebrae can occur repeatedly after vertebroplasty. The mechanism of the new vertebral fractures is unclear, but from experimental and computational studies, it appears that a change in the mechanical load after augmentation might be responsible [1-8]. If the adjacent vertebrae are already rigid, the dynamic hammer effect can lead to a non-adjacent fracture [9].

In contrast to the growing recognition of the importance of adjacent segment disease after lumbar arthrodesis, relatively little attention has been paid to the relationship between osteoporotic vertebral fractures and instrumented lumbar arthrodesis. Solid fusion alters the biomechanics at the motion segments, resulting in increased mechanical demand. Increased biomechanical forces, mobility and intradiscal pressure in the adjacent and non-adjacent segments after fusion have been suggested to accelerate these pathological changes [10-14]. One of the findings next to a fused segment was vertebral compression fractures [15]. There is limited data available on incidence of new osteoporotic vertebral fractures after lumbar spine fusion [16]. The ability of the vertebral body to carry a load depends on the structural capacity of the vertebrae and on the loading conditions that arise from the activities of daily living or trauma. The vertebral bone in an aging spine fails because it cannot support the internal stresses and strains that result from loads applied to the vertebral body [17]. It is possible to sustain a new vertebral fracture at the osteoporotic vertebra because instrumented lumbar arthrodesis alters the biomechanics of load transfer to the remained vertebrae. Therefore, this study examined the relationship between a new osteoporotic vertebral fracture and instrumented lumbar arthrodesis.

The clinically diagnosed vertebral fractures in the study group were associated with a slip down in 6 cases, back strain in one case and the aggravation of back pain without trauma in 18 cases. In the control group, the patient history showed a slip down in 7 cases, back strain in 2 cases and the aggravation of back pain without trauma in 19 cases. There was a better BMD in the study group than the control group (0.67 ± 0.12 g/cm² vs. 0.60 ± 0.13 g/cm², p = 0.013). The fracture following instrumented arthrodesis was diagnosed after a mean of 47 months (range, 7 to 100 months). Fractures in the study group were found at T11, T12, L1, L2 and L3 in 6, 7, 7, 4, and 1 case, respectively. In the control group, the fracture site was found in T8, T11, T12, L1, L2, L3 and L4 in 1, 1, 5, 12, 6, 2, and 1 case, respectively. There was no difference in the fracture site between both groups.

The VAS score for back pain was 4.6 ± 1.4 before the fracture. The VAS score improved from a mean of 7.5 ± 1.0 at the time of the fracture to a mean of 4.9 ± 2.0 at 1 year after the fracture (p = 0.001). However, 12 patients (48%) complained of an aggravation of back pain 1 year after the fracture compared to the pre-fracture state.

Height loss of the fracture showed a mean of 24.1 ± 12.1% at the time of the fracture and 24.3 ± 12.1%, 24.2 ± 10.1° at 1 year after the fracture. The lumbar lordosis revealed an average of 22.8 ± 10.9° at the time of the fracture and 24.2 ± 10.1° at 1 year after fracture. There was no significant change in radiological height loss of the fracture and lumbar lordosis at 1 year after the fracture. In addition, there was no significant correlation between back pain and the loss of vertebral height (r = - 0.198, p = 0.342), back pain and lumbar lordosis (r = 0.070, p = 0.741) at the last follow up. On the other hand, there was significant negative correlation between the BMD of the femoral neck and back pain at the last follow up (r = - 0.455, p = 0.022) (Fig. 2).

Given the number of spinal fusions performed annually, there has been increasing concern regarding the potential for adjacent segment degeneration, radiographic changes in degeneration at the levels adjacent to the spinal fusion, adjacent segment disease, and the development of new symptoms associated with adjacent segment degeneration. Osteoporotic vertebral fractures may be a sentinel sign of failing health in elderly patients, and are likely to become an increasing health-care concern as the population continues to age. We wondered whether an osteoporotic vertebral fracture develops through an increased biomechanical load on the vertebrae adjacent to the fusion sites.

A decrease in the number, thickness and interconnectivity of vertebral trabeculae in combination with altered load transmission across the degenerated disc predisposes the vertebral body to fracture from minor trauma. Biomechanically, compressive forces in the vertebra are transferred from the intervertebral discs to the vertebral endplates and are then distributed between the trabecular centrum and the thin shell of condensed bone that comprise the vertebral body. Most of the axial force is carried out by the trabecular bone [18-21]. Therefore, the axial forces external to the vertebra impart stresses and strains to the trabecular centrum. The ultimate strength of the trabecular bone is determined by the maximum stress that it can sustain within the overall structure of the vertebral body. The trabecular bone, particularly osteoporotic vertebra, will fail if the strength is not greater than the working stresses in the vertebra under physiological or traumatic conditions. Such local failures or cracks can lead to a fracture of the entire bone. Therefore, strength is an important mechanical property when considering the risk of fracture.

The total flexibility of the lumbar spine was decreased after lumbar fusion. There was less segmental deformation of the juxta-fused segments after either a single-or doublelevel fusion. During lumbar motion, the remaining segments were not deformed beyond their physiological limits after fusion. However, the percentage mechanical load at the un-fused segments was increased due to the reduced number of motion segments for mechanical load-sharing [22]. The thoracolumbar junction connects the relatively rigid, kyphotic thoracic spine, which is stabilized by the rib cage to the more mobile, lorodotic un-fused vertebrae. This transitional zone might experience substantial biomechanical stresses during minor trauma, such as slip down, back strain and the activities of daily living, which make it more susceptible to fractures than adjacent segment disease. Most osteoporotic vertebral fractures affected the motion segments between T11 and L2 in both groups.

In this study, the study group had a better BMD than the control group (0.67 ± 0.12 g/cm² vs. 0.60 ± 0.13 g/cm², p = 0.013). However, a vertebral fracture after instrumented fusion occurred at a mean 47 months after lumbar fusion. The fracture was noted at the thoracolumbar junction in contrast to adjacent segment disease. At the remaining vertebra with a degenerated disc, the loads are transmitted unevenly to the end plates, resulting in possible load concentrations on parts of the end plate [23]. A fracture results when this load overcomes the resistance of the fragile end plate.

It is unclear why the lumbar spine cannot be used to measure the BMD instead of the femoral neck BMD. The BMD is normally measured at the lumbar spine and proximal femur because these sites are affected early in the course of osteoporosis and are the most common sites of osteoporotic fractures [24-26]. Measurements of the lumbar spine BMD for an osteoporotic vertebral fracture could be necessary because the femoral neck BMD cannot predict the lumbar spine BMD precisely [27]. Unfortunately, the BMD could not be measured at the lumbar spine because of lumbar fusion with instrumentation. Black et al. [28] reported that the decreased bone density measurements in the hip, spine, and extremities were equally predictive of the increased fracture risk in older women.

This study is limited by the small number of cases and lack of a comparison of fusion level and fusion methods. To provide more evidence for the influence of instrumented lumbar arthrodesis on the osteoporotic vertebral fracture, it will be necessary to run additional studies such as comparison of fusion level and fusion methods with prospective design.

These results suggest that an osteoporotic vertebral fracture after instrumented arthrodesis can aggravate back pain and affect the final outcome of degenerative lumbar disorders. Therefore, it is important to evaluate the possibility of a new osteoporotic vertebral fracture when new-onset back pain occurs after lumbar instrumented arthrodesis.