Retrospective review of procedural parameters and outcomes of percutaneous vertebroplasty in 673 patients.

Percutaneous vertebroplasty (PVP) is a minimally invasive procedure to treat back pain secondary to osteoporotic vertebral compression fractures (VCF). This study aims to review our techniques and outcomes in patients with VCF. Outcomes of all patients who underwent PVP at our institution from 1998 to 2014 were retrospectively collected from medical records and follow-up telephone interviews. 1174 PVP procedures for VCF in 673 patients were identified to have complete follow-up data. Patients with inadequate data were excluded from the analysis. Procedural aspects such as unipedicular or bipedicular access, vertebral region treated, amount of cement injected into vertebrae, number of levels treated at a single session, refracture rates and location, presence of a necrotic cavity, and pain outcomes were examined. Excellent rates of improvement of back pain for both single level and multilevel PVP were achieved in 92% of patients. Unipedicular or bipedicular approach, cement volume, vertebral region treated, cement extravasation, and presence of a necrotic cavity did not affect pain outcomes or refracture rates. Fractures that did develop after PVP were often adjacent and occurred earlier than distant level fractures. Lumbar vertebrae required more cement than thoracic vertebrae. PVP provides excellent rates of pain relief in both single and multilevel procedures. The procedural aspects evaluated did not affect pain outcome or refracture rates. Adjacent refractures tended to occur sooner than distant ones.