Comparison of a fluoroscopic 3-dimensional imaging system and conventional CT in detection of pars fractures in the cadaveric lumbar spine.

STUDY DESIGN: Cadaveric Study. OBJECTIVE: To compare a fluoroscopic imaging system with computed tomography (CT) and radiographs in detection of spondylolysis and radiation exposure in a cadaver model. SUMMARY OF BACKGROUND DATA: Lumbar spondylolysis is defined as a defect or fracture of the pars interarticularis and occurs with or without anterior spondylolisthesis. CT scan is the gold standard imaging study for spondylolysis but is limited by the supine position, which may cause reduction of anterolisthesis and by ionizing radiation, which limits the frequency of follow-up scans. METHODS: Thirteen intact cadaveric lumbar spine segments with 26 pars were randomized to be left intact or to undergo simulated fracture using a 1.3 mm oscillating microsurgical saw. Fifteen pars underwent simulated fracture and 11 pars were left intact. Lumbar spine segments were imaged using plain radiographs, multiplanar fluoroscopic imaging, and conventional CT scan. The images were interpreted by 3 observers blinded to the number and location of defects. Radiation exposure and doses were recorded from all imaging units. RESULTS: Average radiation doses were 0.0025 mSv for each radiograph, 0.23 mSv (low dose) and 0.47 mSv (high dose) for fluoroscopic imaging, and 1.5 mSv for conventional CT imaging (pediatric dose setting). Evaluation of radiographs for spondylolysis had sensitivity of 98% and specificity of 97%. Evaluation using low-dose fluoroscopic images, high-dose fluoroscopic images, and CT scan images correctly identified the status of all pars based on multiplanar images; sensitivity and specificity were 100%. Kappa analysis demonstrated a value of 0.89 for radiographic interpretation indicating excellent agreement. Kappa values describing agreement for image interpretation for fluoroscopic imaging and CT scan were equal to 1.0, representing perfect agreement. CONCLUSIONS: Three-dimensional fluoroscopic imaging provides comparable diagnostic imaging with CT scan in an experimental cadaveric model of spondylolysis using up to 85% less radiation than conventional CT scan.