Risk of Adjacent Vertebral Body Fractures After Balloon Kyphoplasty: A Biomechanical Study

Introduction: This biomechanical study investigated the incidence, location, morphology, and load required to create subsequent VB fractures adjacent to balloon kyphoplasty.

Methods: Ten fresh human thoracolumbar specimens (9F/1M), mean age 78±8.9yrs, each consisting of 5 adjacent vertebrae (four T10-L2, six T5-T9) were used. BMD was measured using pQCT. VB cortices were instrumented with strain gauges. After cancellous bone disruption in the middle VB, the specimens were compressed under follower load until a fracture was observed with >25% anterior height loss. Fracture reduction was performed by balloon kyphoplasty under a physiologic preload of 250N. After bone cement hardening the specimen was recompressed until an adjacent fracture was observed either on video fluoroscopy or detected as discontinuity in the strain gauge data. The vertebral kyphosis after the initial fracture and after balloon kyphoplasty, the location and morphology of the adjacent fracture, and fracture load were recorded.

Results: The initial VCF increased the vertebral kyphosis (6.2°vs.18°, p<0 .01). Balloon kyphoplasty significantly corrected the VB deformity; however, residual kyphosis remained larger than intact value (6.2°vs.11°, p<0.01). The adjacent fracture occurred above initial VCF in six specimens, and below four. mean load was 698±328N. BMD of fractured smaller un-fractured (99.0 vs. 119 mg/cc, p<0.05). Macroscopic examination showed four specimens with endplate depression cortical wall fractures, three only depression, fractures.

Discussion: Fracture load for VB adjacent to kyphoplasty appears to be much smaller compared with that reported for the first VCF in osteoporotic spines. Low BMD was a strong risk factor for location of subsequent fractures. The residual kyphosis and bone cement augmentation may also contribute to increased stress at adjacent levels, increasing the risk of subsequent fractures.