Vertebroplasty for Osteoporotic Fractures with Spinal Canal Compromise

Patients

In our institution, vertebroplasty was performed in patients who failed conservative treatment and were not candidates for surgery because of age, severe osteoporosis, or debilitation. Exclusion criteria included radicular symptoms, unstable fractures involving the posterior column, or cord compression. We reviewed all 55 patients with 139 osteoporotic compression fractures treated with vertebroplasty and selected all those who had spinal canal compromise due to retropulsion. This study was based on 21 patients (12 women and 9 men; age range, 51–93 years; mean age, 79 years) in whom preoperative imaging showed retropulsion. We analyzed 25 vertebral bodies in these 21 patients. Most of the fractures were around the thoracolumbar junction. The locations and numbers of the treated vertebrae were as follows: T4 (n = 1), T5 (n = 1), T6 (n = 1), T7 (n = 1), T8 (n = 1), T10 (n = 2), T11 (n = 1), T12 (n = 5), L1 (n = 4), L2 (n = 5), L3 (n = 1), and L4 (n = 2).

Vertebroplasty Technique

Vertebroplasty was performed through a transpedicular or parapedicular approach by using 13-gauge bone-biopsy needles (Osteo-Site, Cook, Bloomington, Ind) placed into the anterior one third of the vertebral body. The procedure was performed under biplane fluoroscopic control with use of conscious sedation and local anesthetic on an outpatient basis.

The patient lay prone on the angiographic table. Once the needle or needles were placed in the vertebral body, the liquid and powder polymethylmethacrylate (Cranioplastic; Codman, Raynham, Mass) was mixed with 12 g of barium sulfate to a doughlike consistency. The cement was thicker than toothpaste and relatively hard to inject through the 13-gauge needle by using a 1-mL syringe. However, this was not an “extra-thick” cement formulation. Under biplane fluoroscopic guidance, the cement was injected through the needles. The injection continued until the vertebral body was filled toward the posterior 25% or there was prominent leakage. Special attention was paid to worsening of retropulsion on fluoroscopy. After cement injection, the patient remained prone on the angiographic table for 15–30 minutes and was then transferred to a regular bed. The patient remained in bed until CT had been performed. The patients were evaluated for pain relief and presence of neurologic symptoms. Thereafter, the patients were discharged home as tolerated.

Imaging Technique

Eighteen patients underwent preoperative MR imaging. The other 3 patients underwent preoperative CT. MR imaging was performed with a 1.5T imager. At least sagittal T1-weighted (TR/TE, 500–767/8–22 ms), sagittal T2-weighted (TR/TE, 3800–4550/98–123.6 ms), and sagittal fat-suppressed contrast-enhanced T1-weighted (TR/TE, 567–750/8–23 ms) images were obtained. In some cases, additional sequences were also available. The patients were given an intravenous injection of 0.2 mmol/kg of gadodiamide. Typical imaging parameters were as follows: FOV, 34–36 × 25.5–36 cm; matrix size, 256–512 × 160–256; section thickness, 3 mm; intersection gap, 0.5–1 mm; and echo-train length, 1–4.

CT scans were obtained by using a single-detector or 4–detector row CT scanner with the following parameters: 0.625- to 3-mm collimation, 3.75- to 7.5-mm/s table speed, 16- to 24-cm FOV, 250–280 mAs, and 120 kVp. Reconstructions were performed at 2- to 3-mm section thickness.

The average duration between the preoperative imaging and treatment was 11 days (range, 0–40 days). All CT scans were obtained within 24 hours of completion of procedures.

Imaging Assessment

The raw CT and MR imaging data were transferred to a workstation. CT and MR images were shown with regular clinical window width and level settings. Sagittal reformatted images were created for CT. Vertebral body height was measured in the anterior, central, and posterior portions of the vertebra and wedge angle, in the midsagittal plane as previously described.^3,15 Retropulsion of the bone fragment was also measured in the same plane by using a distance from the line between the posterior margin of the adjacent vertebral bodies and the retropulsed bony fragment (Fig 1). Measurement was performed by using a magnified image to the nearest 0.1 mm and 1°, twice at the workstation, and the mean was recorded. One untreated vertebral body adjacent to the treated vertebral body was also measured as a control in each patient to evaluate the validity of our measurements.

• Download figure
• Open in new tab
• Download powerpoint

Fig 1.

Lines used to measure retropulsion of the bone fragment over sagittal CT scans. Retropulsion is measured by using a distance from the line between the posterior margin of the adjacent vertebral bodies (arrowhead) and the bony fragment (arrow).

Statistical Analysis

Statistical analysis was performed with commercially available software. The statistical significances of changes in retropulsion, vertebral body height, and wedge angle were evaluated with the Wilcoxon signed rank test for treated vertebral bodies. Control vertebral bodies were evaluated with the paired t test. A P value of less than .05 was considered statistically significant.