BACKGROUND: Percutaneous vertebroplasty is an efficient procedure to treat painful osteoporotic vertebral compression fractures. However, refracture of cemented vertebrae occurs rarely after percutaneous vertebroplasty. This study was undertaken to investigate the incidence, characteristics, predisposing factors, and mistakes in technique associated with refracture of the same vertebra after percutaneous vertebroplasty. METHODS: From 2001 to 2008, PVP with bone cement (polymethylmethacrylate, PMMA) was carried out in 2,291 patients with 2,581 PVP procedures. The etiologies including pathologic fracture (including metastasis, osteolytic tumor, hemangioma) in 299 patients, infectious spondylitis in 4 patients and osteoporotic compression fractures in 1,988 patients. A total of 1,988 patients with 2,110 VCFs underwent PVP with PMMA cement after failing conservative treatment for at least 3 months. New recollapsed vertebral fractures were diagnosed as recurrent intractable back pain, postoperatively correlated with serial plain radiography and MR image. Clinical parameters such as age, gender, body mass index, and fracture-free interval (from the date of the initial intervention with percutaneous vertebroplasty to the diagnosis of subsequent fractures) were recorded. Parameters related to imaging and technical characteristics, including the amount of bone cement injected per procedure, level, the presence of osteonecrosis in the vertebral body, and the surgical approach (uni- or bipedicles), the restoration of kyphosis angle and height of the anterior border of the collapsed vertebral body, and any leakage of cement into the disk space were also recorded. RESULTS: In a 2-year follow-up, 1,800 patients with 1,820 VCFs were retrospectively reviewed and 10 patients with 10 VCFs developed refracture of the same vertebra after PVP with an incidence rate of 0.56% (10 in 1,800). The mean age of the ten patients (nine females and one male) was 79.6 years, and the mean BMI is 22.3. Levels of refracture after PVP were all located in the thoracolumbar junction (T12-L2): three in T12; four in L1; and three in L2. Osteonecrosis was present in all patients and intradiscal cement leakage was noted in five patients. The mean of the restoration of kyphosis angle was 7.7° and height of the anterior border was 26%. Osteonecrosis, greater anterior vertebral height restoration, lesser kyphosis angle correction and cystic filling pattern were statistically significant. CONCLUSIONS: Our study suggests that larger height restoration and solid lump filling cement are risk factors of refracture of cemented vertebral bodies. Symmetric cement distribution and fluid aspiration would be the potential ways to avoid refracture of cemented vertebral bodies.
BACKGROUND: Percutaneous vertebroplasty is an efficient procedure to treat painful osteoporotic vertebral compression fractures. However, refracture of cemented vertebrae occurs rarely after percutaneous vertebroplasty. This study was undertaken to investigate the incidence, characteristics, predisposing factors, and mistakes in technique associated with refracture of the same vertebra after percutaneous vertebroplasty. METHODS: From 2001 to 2008, PVP with bone cement (polymethylmethacrylate, PMMA) was carried out in 2,291 patients with 2,581 PVP procedures. The etiologies including pathologic fracture (including metastasis, osteolytic tumor, hemangioma) in 299 patients, infectious spondylitis in 4 patients and osteoporotic compression fractures in 1,988 patients. A total of 1,988 patients with 2,110 VCFs underwent PVP with PMMA cement after failing conservative treatment for at least 3 months. New recollapsed vertebral fractures were diagnosed as recurrent intractable back pain, postoperatively correlated with serial plain radiography and MR image. Clinical parameters such as age, gender, body mass index, and fracture-free interval (from the date of the initial intervention with percutaneous vertebroplasty to the diagnosis of subsequent fractures) were recorded. Parameters related to imaging and technical characteristics, including the amount of bone cement injected per procedure, level, the presence of osteonecrosis in the vertebral body, and the surgical approach (uni- or bipedicles), the restoration of kyphosis angle and height of the anterior border of the collapsed vertebral body, and any leakage of cement into the disk space were also recorded. RESULTS: In a 2-year follow-up, 1,800 patients with 1,820 VCFs were retrospectively reviewed and 10 patients with 10 VCFs developed refracture of the same vertebra after PVP with an incidence rate of 0.56% (10 in 1,800). The mean age of the ten patients (nine females and one male) was 79.6 years, and the mean BMI is 22.3. Levels of refracture after PVP were all located in the thoracolumbar junction (T12-L2): three in T12; four in L1; and three in L2. Osteonecrosis was present in all patients and intradiscal cement leakage was noted in five patients. The mean of the restoration of kyphosis angle was 7.7° and height of the anterior border was 26%. Osteonecrosis, greater anterior vertebral height restoration, lesser kyphosis angle correction and cystic filling pattern were statistically significant. CONCLUSIONS: Our study suggests that larger height restoration and solid lump filling cement are risk factors of refracture of cemented vertebral bodies. Symmetric cement distribution and fluid aspiration would be the potential ways to avoid refracture of cemented vertebral bodies.