Background and purpose: To explore the value of CT spinal angiography with 64-detector row spiral CT in diagnosing spinal vascular malformations. Methods: Seventeen patients with initial MR and clinical findings suggestive of spinal vascular diseases underwent CT spinal angiography. Among these, 14 patients took DSA examination within 1 week after CT scan, 7 patients underwent surgical treatment, and 6 patients underwent vascular intervention embolotheraphy. CT protocol: TOSHIBA Aquilion 64 Slice CT scanner, 0.5 mm thickness, 0.5 s/r, 120 kV and 350 mA, positioned at the aortic arch level, and applied with "sure start" technique with CT threshold of 180 Hu. Contrast agent Iohexol (370 mg I/ml) was injected at 6 ml/s velocity with total volume of 80 ml. The post-processing procedures included MPR, CPR, MIP, VR, etc. Among the 17 patients, four patients underwent fast dynamic contrast-enhanced 3D MR angiography imaging. CT spinal angiography and three-dimensional contrast-enhanced MR angiography (3D CE-MRA) images were compared and evaluated with DSA and operation results based on disease type, lesion range, feeding arteries, fistulas, draining veins of vascular malformation by three experienced neuroradiologists independently, using double blind method. The data were analyzed using SPSS analytic software with chi(2)-test. We compared the results with DSA and operation results. Results: The statistical analysis of the diagnostic results by the three experienced neuroradiologists had no statistical difference (P > 0.05). All of the 17 patients showed clearly the abnormality of spinal cord vessels and the range of lesions by CT spinal angiography. Among them, one patient was diagnosed as arteriovenous fistulas (AVF) by MRI and CT spinal angiography, which was verified by surgical operation. DSA of the same patient, however, did not visualize the lesion. One case was diagnosed as AVM complicated with AVF by DSA, but CT spinal angiography could only show AVM not AVF. The type differentiations of all the other 16 patients were consistent with DSA results. For 13 cases with positive CT spinal angiography results, DSA displayed 20 feeding vessels, among which 16 vessels were displayed correctly by CT spinal angiography, four could not be visualized, and two turned out to be false-positive. Fistulas were not displayed in six cases by CT spinal angiography. Draining veins were displayed clearly in all cases, which agreed with DSA results. Four cases who took CE-MRA obtained the same type diagnosis as that from CT spinal angiography. Feeding arteries were not displayed in CE-MRA of one case, but could be clearly visualized in other three cases, and the results agreed with CTA and DSA results. Fistulas could be seen in two cases. Draining veins and the disease range could be displayed distinctly by 3D CE-MRA. Conclusion: CT spinal angiography is quite valuable for diagnosing vascular malformation of spinal cord. It can be a screening exam before DSA, and has a guiding effect on DSA, reducing the amount of time required for DSA. (C) 2008 Elsevier Ireland Ltd. All rights reserved.
