Digital subtraction CT angiography for detection of intracranial aneurysms: comparison with three-dimensional digital subtraction angiography

Abstract

Purpose: To evaluate the diagnostic accuracy of digital subtraction computed tomographic (CT) angiography in the detection of intracranial aneurysms compared with three-dimensional (3D) rotational digital subtraction angiography (DSA), as reference standard, in a large cohort in a single center.

Materials and methods: The study was waived by the institutional review board because of its retrospective nature. A total of 513 patients clinically suspected of having or with known intracranial aneurysms and other cerebral vascular diseases underwent both digital subtraction CT angiography with a dual-source CT scanner and 3D DSA, with a median interval of 1 day; 436 patients (84.9%) had acute subarachnoid hemorrhage at presentation. The sensitivity, specificity, and accuracy of digital subtraction CT angiography in depicting aneurysm were analyzed on a per-patient and per-aneurysm basis, with 3D DSA as the reference standard. The sensitivity, specificity, and accuracy of digital subtraction CT angiography in depicting aneurysms of different diameter (ie, <3 mm, 3-5 mm, 5-10 mm, and >10 mm) and of aneurysms at different locations in the anterior and posterior circulation were calculated. Kappa statistics were calculated to quantify inter- and intrareader variability in detecting aneurysms by using digital subtraction CT angiography for 100 patients.

Results: Of 513 patients, 106 (20.7%) had no aneurysms, while 407 patients (79.3%) had 459 aneurysms at 3D DSA. Digital subtraction CT angiography correctly depicted 456 (99.3%) of the 459 aneurysms. By using 3D DSA as the standard of reference, the sensitivity and specificity of depicting intracranial aneurysms were 97.8% (398 of 407) and 88.7% (94 of 106), respectively, on a per-patient basis, and 96.5% (443 of 459) and 87.8% (94 of 107), respectively, on a per-aneurysm basis. Digital subtraction CT angiography had sensitivities of 91.3% (42 of 46), 94.0% (140 of 149), 98.4% (186 of 189), and 100% (75 of 75) in depicting aneurysms of less than 3 mm, between 3 mm but less than 5 mm, between 5 mm but less than 10 mm, and 10 mm or greater, respectively, and of 95.8% (276 of 288) and 97.7% (167 of 171) in depicting anterior circulation and posterior circulation aneurysms, respectively. Excellent inter- and intrareader agreement was found on a per-patient (κ=0.900 and 0.939, both P<.001) and per-aneurysm basis (κ=0.846 and 0.921, both P<.001) for the detection of intracranial aneurysms with digital subtraction CT angiography.

Conclusion: Digital subtraction CT angiography has a high sensitivity and specificity in depicting intracranial aneurysms with different sizes and at different locations, compared with 3D DSA.