OBJECTIVETo explore how to determine painful vertebrae treated by kyphoplasty in multiple-level osteoporotic vertebral compression fractures and clinical outcome.

METHODSFrom October 2002 to June 2005, 51 consecutive procedures with kyphoplasty were performed on 35 patients with multiple-level osteoporotic vertebral compression fractures. There were 51 painful vertebrae among 120 vertebral compression fractures. The painful vertebra was determined by the signal intensity changes in MR images, combined with radiography and local percussion pain before operation. Only painful vertebrae were treated by kyphoplasty. Preoperative, postoperative and final follow-up visual analog scale (VAS) and radiographic findings such as vertebral height and Cobb angle were recorded and analyzed.

RESULTSAll patients tolerated the procedure well with immediate relief of their back pain after kyphoplasty and they can walk at 1-3 days after the procedure. There were 3 vertebrae (3/51) occurred asymptomatic extravasation of cement. 31 cases were followed up for mean 16.2 months (range 6-44 months). VAS reduced from preoperative 8.7 to final follow-up 2.1 (P <0. 01). At final follow-up the vertebral height had a recovery rate of 59.17%, and the mean Cobb angle was improved 10.1 degrees. There was a significant improvement between preoperative and final follow-up values (P < 0.01).

CONCLUSIONSThe painful vertebra can be determined by signal intensity changes in MR series images in multiple-level osteoporotic vertebral compression fractures. Selecting painful vertebrae to be treated by kyphoplasty can make patients with multiple-level VCFs gain an excellent result.