Optimal intravertebral cement volume in percutaneous vertebroplasty for painful osteoporotic vertebral compression fractures

STUDY DESIGN.: A prospective follow-up study. OBJECTIVE.: Assessment of the relation between accomplishment of pain relief through percutaneous vertebroplasty (PVP) in painful osteoporotic vertebral compression fractures (OVCFs) and the cemented fraction of the vertebral body and subsequent determination of the optimal intravertebral cement volume. SUMMARY OF BACKGROUND DATA.: The mechanism of pain relief of PVP as a treatment modality for painful OVCFs remains unclear. Generally, benefit of PVP is thought to result from stabilization of micromovements and collapse of the fractured vertebral body. However, studies indicating a relation between intravertebral cement volume and pain relief are lacking and an optimal value of the intravertebral cement volume is unknown. METHODS.: One hunderd six patients who received PVP for 196 painful OVCFs were prospectively followed on back pain (score 0-10) and occurrence of new OVCFs during the first postoperative year. Patients were classified as responders (average postoperative back pain ≤6) and nonresponders (average postoperative back pain >6). The cemented fraction of the vertebral body was determined using volumetric analysis of the postoperative CT scan of the treated levels. Analysis was performed using receiver-operating characteristic (ROC) analysis and multivariable regression techniques. RESULTS.: Twenty-nine patients (27.3%) were found to be nonresponders. Mean intravertebral cement volume in all 196 treated OVCFs was 3.94 mL (SD = 1.89, range 0.13-10.8). The mean cemented vertebral body fraction was significantly lower in nonresponders (0.15 vs. 0.21, P = 0.002). The ROC area-under-curve of the cemented fraction as a predictor of accomplishment of pain relief was 0.67 (95% CI: 0.57-0.78, P = 0.006). In subgroups without specific influential factors (new OVCFs, intravertebral clefts), significantly stronger associations were found. A vertebral body fraction of 24% was identified as the optimal fraction to be cemented. This fraction corresponded to a 93% to 100% specificity for accomplishment of pain relief (i.e., few to no cases without pain relief in the presence of adequate cementing) without being significantly associated with a higher risk of occurrence of cement leakage or new OVCFs. Corresponding values for the recommended (optimal) intravertebral cement volume were provided based on its governing characteristics (fracture level, fracture severity, and patient's sex). CONCLUSION.: An optimal intravertebral cement volume was identified for accomplishment of pain relief through PVP in painful OVCFs. Appropriate thresholds were provided to guide the operator.