RESUMO
UNLABELLED: In our paper, we systemically retrieved the eligible study evaluating whether increased incidence of subsequent vertebral fracture is associated with vertebroplasty. Main effect sizes were vertebral fracture rates reported in terms of hazard ratio (HR) for time-to-event data or relative risk (RR) for dichotomous outcome. Our results do not support the hypothesis that vertebroplasty contributes to increased risk of subsequent vertebral fracture, neither adjacent nor total vertebral fracture. INTRODUCTION: Vertebroplasty has been implicated in significant changes in vertebral strength, vertebral shape, and consequently increased risk for subsequent vertebral fracture, especially the adjacent level. Here, we further tested the hypothesis whether new-onset vertebral fracture is a natural result of osteoporosis or consequence of cement augmentation. METHODS: Relevant literatures were retrieved using PubMed, Web of Knowledge, and Cochrane Central Register of Controlled Trials (CENTRAL), supplemented by a hand-search of the reference lists of selected articles. Eligible studies assessed whether increased morbidity of subsequent vertebral fracture is associated with vertebroplasty. Main effect sizes were vertebral fracture rates reported in terms of hazard ratio (HR) for time-to-event data or relative risk (RR) for dichotomous outcome. Random-effects model was used to account for clinical or methodological heterogeneity across studies. RESULTS: Thirteen studies with a number of 2,551 individuals (1,631 in vertebroplasty group and 920 in control group) were suitable for this meta-analysis. In trials that reported adjacent vertebral fracture as time-to-event data (two trials, n = 328), we found a similar incidence of vertebral fracture in percutaneous vertebroplasty (PVP) group compared to conservative therapy (HR 0.60, 95% confidence interval 0.29 to 1.26; P = 0.18). In trials that reported overall vertebral fracture as time-to-event data (three trials, n = 704), vertebroplasty was associated with a slightly increased but non-significant risk for vertebral fracture (HR 1.14, 95% confidence interval 0.65 to 2.00; P = 0.65). The outcome was further confirmed in the secondary meta-analysis of studies that reported vertebral fracture as dichotomous data. Subgroup analysis according to study design revealed no difference either. CONCLUSIONS: Our results do not support the hypothesis that vertebroplasty contributes to increased risk of subsequent vertebral fracture, neither adjacent nor total vertebral fracture. However, adequately designed randomized controlled trials are warranted to confirm the present findings.