Risk Factors Of Augmented Vertebra Recompression After Vertebroplasty

Objective:Percutaneous vertebroplasty(PVP) and percutaneous kyphoplasty(PKP) are main surgical styles to treat painful osteoporotic vertebral compression fractures(OVCFs) nowadays.However,recompression of augmented vertebrae after surgery is a common complication of PVP and PKP and difficult to deal with clinically. This study was undertaken to explore the clinical relative factors associated with recompression of operated vertebrae after PVP/PKP and assess their clinical significance in order to prevent recompression of augmented vertebrae and improve therapeutic effect.Methods:Patients who had undergone PVP or PKP for OVCFs at our hospital between July 2013 and February 2015 were retrospectively reviewed. According to inclusion and exclusion criteria,we collected all patients' clinical and radiographic parameters including age, gender, preoperative bone mineral density T score,height,weight, levels of vertebral fractures, the presence of intervertebral cleft(IVC),the surgical style and approach, the volume of bone cement injected,leakage of cement into adjacent intervertebral disc, non-PMMA-endplate-both-contact(NPEBC) and visual analogue scale(VAS) and measured the anterior body height and kyphotic angle and then calculated the compression rate of the anterior body height(ABHCR) and reduction rates and angles. According to the loss of the anterior body height of the operated vertebral body at last follow-up period, the patients were divided into two groups, recompressed group and well-maintained group. Univariate and multivariate analysis were used to analyse the correlation of recompression with the suspicious factors in order to identify.risk factors that affected recompression of augmented vertebrae after PVP/PKP.Results:In total,151 patients meeting with the inclusion criteria were revirewed. There were 31 men and 120 women. The mean age of all patients was 72 years old. The follow-up time was 12 to 45 months and the average of follow-up time was 20 months. The average of preoperative bone mineral density T score was -3.5. We found 10 patients with the presence of preoperative IVC according to radiographic examination. There were 93 cases with PVP and 58 cases with PKP.16 vertebrae were reinforced through unipedicular approach and the others were reinforced through bipedicular approach.The mean amount of cement injected was 6.3 ml. We found 8 cases with leakage of bone cement into adjacent intervertebral disc and 47 cases with NPEBC after surgry. Among these patients,31 operated vertebrae recollapsed after surgery.The incidence of recompression of treated vertebrae was 20.5%. The occurrence of recompression was 3 months(range 1-8 months).The distribution of fracture regions was as follows:1 case at T10, one case at T11,9 cases at T12,16 cases at L1,1 case at L2,2 cases at L4. The univariate analysis revealed that reduction ABHCR and NPEBC showed highly significant differences (P<0.05). Age, sex, body mass index, preoperative bone mineral density T score, fracture region,IVC,preoperative and postoperative ABHCR, preoperative and postoperative KA,reduction KA,surgery style, approach,the amount of cement injected,leakage of cement into adjacent intervertebral disc showed no significant differences(P>0.05). Multivariant logistic regression demonstrated that NPEBC was the most significant risk factor which resulted in recompression of previously treated vertebrae after surgery.Conclusion:NPEBC would increase the risk of recompression of treated vertebrae postoperatively due to the presence of PMMA-nonsupported area. As a consequence, we suggest that cement injected should be diffused as sufficiently as possible and contact the superior and inferior endplate simultaneously to reinforce the vertebral body integrally in order to avoid the recompression of operated vertebrae after surgery.