Creating reproducible thoracolumbar burst fractures in human specimens: An in vitro experiment

Abstract

OBJECT: The treatment of traumatic burst fractures unaccompanied by neurological impairment remains controversial and ranges from conservative management to 360° fusion. Because of the heterogeneity of fracture types, classification systems, and treatment options, comparative biomechanical studies might help to improve our knowledge. The aim of the current study was to create a standardized fracture model to investigate burst fractures in a multisegmental setting. METHODS: A total of 28 thoracolumbar fresh-frozen human cadaveric spines were used. The spines were dissected into segments (T11-L3). The T-11 and L-3 vertebral bodies were embedded in Technovit 3040 (cold-curing resin for surface testing and impressions). To simulate high energy, a metallic drop tower was designed. Stress risers were used to ensure comparable fractures. CT scans were acquired before and after fracture. All fractures were classified using the AO/OTA classification. RESULTS: The preparation and embedding of the spine segments worked well. No repositioning or second embedding of the specimen, even after fracture, was required. It was possible to create single burst fractures at the L-1 level in all 28 spine segments. Among the 28 fractures there were 16 incomplete burst fractures (Type A3.1), 8 burst-split fractures (Type A3.2), and 4 complete burst fractures (Type A3.3). The differences before and after fracture for stiffness and for anterior, posterior, and central heights were all significant (p < 0.05). CONCLUSIONS: The ability to create reproducible burst fractures of a single vertebral body in a thoracolumbar spine segment may serve as a basis for future biomechanical studies that will provide better understanding of mechanical properties or fixation techniques.