OBJECTIVETo explore the reasons of secondary fracture after percutaneous vertebroplasty (PVP) for osteoporotic vertebral compression fractures (OVCFs) and discuss the measure of prevention and cure.

METHODSFrom January 2011 to January 2013, the clinical data of 180 patients with primary OVCFs treated by PVP were retrospectively analyzed. There were 75 males and 105 females, aged from 68 to 95 years old with an average of (79.50 ± 5.45) years. The involved vertebrae were identified according to the clinical symptoms and imaging data. PVP were performed in 362 vertebrae and the patients were followed up with an average of 12 months. Subsequent vertebral fractures were found through the pain's reappearance and MRI or bone scan. The patients were divided into secondary fracture group and no-secondary fracture group according to the subsequent fractures or no. Secondary fracture group was divided into two groups according to gender, and the patients with secondary fracture were also categorized into the original surgical vertebral fractures, adjacent vertebral fracture and remote vertebral fractures. The age, gender, the cement volume, the cement leakage, secondary fracture site, the incidence and type of secondary fracture were observed and compared among different groups.

RESULTSAmong the 362 vertebrae of PVP, there were 109 vertebrae in male and 253 vertebrae in female. And 27 vertebrae (10 in male and 17 in female) of 22 cases (9 males and 13 females) occurred secondary fracture. The second PVP were performed in 13 cases (16 vertebrae) and the third PVP in 2 cases (4 vertebrae); 7 cases (7 vertebrae) were treated with conservative therapy. There was no statistically significant difference on age, gender, cement volume and leakage between secondary fracture group and no-secondary fracture group (P > 0.05). There was no statistically significant difference on the incidence and type of secondary fracture between male and female (P > 0.05). No significant difference was found on the adjacent and remote vertebral fractures (P > 0.05). Most of secondary fracture occurred in 6 months, and whether the single and double side injection, cement leakage had no obvious relation.

CONCLUSIONThere is no significant difference in the subsequent fracture after PVP for the OVCFs different gender and fractured site, and also no significant difference in the adjacent and remote vertebral fractures. The report didn't support the biomechanical viewpoint that vertebral body stiffness increasing after PVP would lead to adjacent vertebral stress increasing and result easily in adjacent vertebral fracture. Secondary fracture occurs always in 6 months after operation, which is the natural course of osteoporosis.