Ex Vivo Model for Percutaneous Vertebroplasty

The testing of novel biomaterials for percutaneous vertebroplasty depends on suitable animal models. The aim of this study was to develop ex vivo a reproducible and feasible model of percutaneous vertebroplasty, for ulterior application in vivo. A large animal model was used (Merino sheep), due to its translational properties. Vertebroplasty was performed under tactile and fluoroscopic control, through a bilateral modified parapedicular access in lumbar vertebrae (n=12). Care was taken in order to avoid disruption of the vertebral foramen. The average defect volume was 1234±240 mm3. This mean volume ensures practical defects to test novel injectable biomaterials. 6 vertebrae were injected with a commercial cement (Cerament®, Bone Support, Sweden). Adequate defect filling was observed in all vertebrae. All vertebrae were assessed by microCT, prior to and post defect creation and after biomaterial injection. All vertebrae were mechanical tested. No mechanical failure was observed under loads higher than the physiological. Ultimately, this model is considered suitable for pre-clinical in vivo studies, mimicking clinical application.

Percutaneous vertebroplasty: a new animal model.

Background Context Percutaneous vertebroplasty (PVP) is a minimally invasive surgical procedure and is frequently performed in humans who need surgical treatment of vertebral fractures. PVP involves cement injection into the vertebral body, thereby providing rapid and significant pain relief. Purpose The testing of novel biomaterials depends on suitable animal models. The aim of this study was to develop a reproducible and safe model of PVP in sheep. Study Design This study used ex vivo and in vivo large animal model study (Merino sheep). Methods Ex vivo vertebroplasty was performed through a bilateral modified parapedicular access in 24 ovine lumbar hemivertebrae, divided into four groups (n=6). Cerament (Bone Support, Lund, Sweden) was the control material. In the experimental group, a novel composite was tested—Spine-Ghost—which consisted of an alpha-calcium sulfate matrix enriched with micrometric particles of mesoporous bioactive glass. All vertebrae were assessed by micro-computed tomography (micro-CT) and underwent mechanical testing. For the in vivo study, 16 sheep were randomly allocated into control and experimental groups (n=8), and underwent PVP using the same bone cements. All vertebrae were assessed postmortem by micro-CT, histology, and reverse transcription-polymerase chain reaction (rt-PCR). This work has been supported by the European Commission under the 7th Framework Programme for collaborative projects (600,000–650,000 USD). Results In the ex vivo model, the average defect volume was 1,275.46±219.29 mm3. Adequate defect filling with cement was observed. No mechanical failure was observed under loads which were higher than physiological. In the in vivo study, cardiorespiratory distress was observed in two animals, and one sheep presented mild neurologic deficits in the hind limbs before recovering. Conclusions The model of PVP is considered suitable for preclinical in vivo studies, mimicking clinical application. All sheep recovered and completed a 6-month implantation period. There was no evidence of cement leakage into the vertebral foramen in the postmortem examination.