The volumetric analysis of cement in vertebroplasty: relationship with clinical outcome and complications.

STUDY
DESIGN: Prospective study.
OBJECTIVE: The aim of this study was to demonstrate the safe range of cement volume during percutaneous vertebroplasty. SUMMARY OF BACKGROUND DATA: A few clinical reports have addressed the relationship between cement volume and clinical outcome. However, the weakness of these studies was that subjects included were not homogeneous. No study in the clinical setting has confirmed results from biomechanical and computational studies.
METHODS: We examined 96 patients with single compression fractures who underwent percutaneous vertebroplasty and postoperative three-dimensional CT scan within a week between June 2006 and April 2009. The volume and fraction were measured by a CT volumetry program. Relationships between predictors and volumetric data, outcome, leakage, intraverterbal vacuum cleft (IVVC), and subsequent fracture were examined. Relationships between volumetric data and outcome, leakage, IVVC, and subsequent fracture were analyzed with stratification by the treated level. Receiver-operator characteristic (ROC) curves were plotted to acquire cut-off values of volumetric data.
RESULTS: Seventy-three patients (76%) were female, and the mean age was 76.3 ± 8.4 years (range 53-97). The mean duration of follow-up was 11 months (range 6-21). Locations were as follows: T4-T10 9, T11-L1-L57, and L2-L4 30. Seventy-eight patients (81%) reported a favorable outcome. Fractured body volume (FBV) and the level treated were associated with fraction, which had an influence on outcome. The fraction of the favorable group was significantly higher. Cut-off values to acquire a favorable outcome were 11.64% (P = 0.026) on the T4-L4 level and 3.35 cm (P = 0.059), 11.65% (P = 0.059) on the T11-L1 level. Group with intradiscal leakage had a smaller volume than nonleakage group on the L2-L4 level (3.86 cm vs. 5.65 cm, P = 0.002). There were no relationships of volumetric data with epidural leakage and pulmonary embolism. The presence of IVVC increased volume on the T4-L4 and L2-L4 level (P < 0.03). Larger volume increased significantly the incidence of adjacent fracture on the L2-L4 level. The significant cut-off volume to avoid adjacent fracture was 4.90 cm on the ROC curve.
CONCLUSION: It is suggested that fraction is superior to volume for predicting outcome on the T11-L1 level and an amount of cement should be determined in terms of FBV and fraction according to the treated level. A lower fraction than required for the restoration of mechanical property was enough to obtain pain relief. Intradiscal leakage on the L2-L4 level may be inevitable to obtain appropriate mechanical properties in the case of severe endplate breakdown connected with the disc space. Smaller volume is needed to avoid an adjacent fracture on the L2-L4 level. Although we did not know the reason why there was a difference among the treated level groups, one thing that is certain is the fact that level-specific approaches may be necessary for good outcome in terms of volume, fraction and FBV.