Cement directed kyphoplasty reduces cement leakage as compared with vertebroplasty: results of a controlled, randomized trial.

STUDY

DESIGN

A novel randomized, controlled, unblinded clinical trial comparing 2 procedural interventions for painful osteoporotic vertebral compression fractures.

OBJECTIVE

The primary study objective was to evaluate cement leakage for a cement directed kyphoplasty system (CDKS) with anteriorly biased cement flow and vertebroplasty. The secondary study objective was to compare adjacent level fracture rates and vertebral body height for these 2 intervention methods.

SUMMARY OF BACKGROUND DATA: Cement leakage remains a significant clinical problem associated with vertebroplasty and kyphoplasty procedures. Uncontrolled cement flow in the posterior direction can result in leakage into the vertebral veins or spinal canal, leading to potentially serious clinical complications.

METHODS

Seventy-seven patients with painful osteoporotic vertebral compression fractures were enrolled. Patients were randomized 2:1 for treatment with CDKS (49 patients, 65 levels) or vertebroplasty (28 patients, 39 levels). Cement leakage was evaluated from radiographs and computed tomographic scans. Three- and 12-month follow-ups included additional radiographs and computed tomographic scans to assess changes in vertebral body height and the incidence of new fractures.

RESULTS

Treatment with CDKS significantly reduced the number of levels with leaks and the total number of leaks per level, as compared with vertebroplasty (P = 0.0132 and P = 0.0012, respectively). Significantly, fewer lateral cortical and spinal canal leaks (posterior leaks) occurred in the CDKS group (P = 0.0050, P = 0.02260, respectively). Three adjacent level fractures occurred in the vertebroplasty group, as compared with 2 in the CDKS group. Vertebral body height maintenance was equivalent.

CONCLUSION

Cement directed kyphoplasty effectively reduces posterior cement leakage, reducing the risk of leakage related complications.

LEVEL OF EVIDENCE: 2.