Risk Factors for Cement Leakage After Vertebroplasty or Kyphoplasty: A Meta-Analysis of Published Evidence.

Abstract

Cement leakage is the most common complication of vertebroplasty and kyphoplasty. So far, the reported risk factors remain conflicting because of limited data and lack of uniform measurement and evaluation. Here, we performed a systematic review and meta-analysis of potential risk factors for cement leakage after vertebroplasty or kyphoplasty. Relevant literature was retrieved using PubMed, EMBASE, Cochrane Controlled Trial Register, and MEDLINE with no language restriction, supplemented by a hand search of the reference lists of selected articles. A fixed-effects model was used if homogeneity existed among included studies; otherwise, a random-effects model was used. The results were presented with weighted mean difference for continuous outcomes and odds ratio (OR) for dichotomous outcomes with a 95% confidence interval (CI). Twenty-two studies consisting of 2872 patients with 4187 vertebrae were included in the meta-analysis. The incidences of cement leakage for percutaneous vertebroplasty and percutaneous balloon kyphoplasty were 54.7% and 18.4%, respectively. The significant risk factors for new vertebral compression fractures were intravertebral cleft (OR, 1.40; 95% CI, 1.09-1.78; P < 0.01), cortical disruption (OR, 5.56; 95% CI, 1.84-16.81; P < 0.01), cement viscosity (OR, 3.32; 95% CI, 1.36-8.07; P < 0.01) and injected cement volume (weighted mean difference, 0.59; 95% CI, 0.02-1.17; P < 0.05). Age, sex and fracture type, operation level, and surgical approach were not significant risk factors. The results of this meta-analysis suggest that patients with intravertebral cleft, cortical disruption, low cement viscosity, and high volume of injected cement may be at high risk for cement leakage after vertebroplasty or kyphoplasty. Rigorous patient selection and individual therapeutic strategy irrespective of age, sex and fracture type, operation level, and surgical approach may reduce the occurrence of cement leakage. Given the inherent limitation of the meta-analysis, more large sample-sized randomized controlled trials are needed to further validate the present findings.