OBJECTIVE: Non-invasive imaging methods have become primordial in subarachnoid hemorrhage. The aim of our study was to evaluate the sensitivity and specificity of multislice computed tomographic angiography (MSCTA) for the diagnosis of cerebral aneurysm. METHODS: The 28 included consecutive patients with SAH underwent both MSCTA and digital subtraction angiography (DSA). The MSCTA studies were interpreted by two independent readers (A and B) for the presence, the location and size of the aneurysm comparatively to the DSA as reference examination. RESULTS: In 20 patients, 38 aneurysms were diagnosed and in eight no aneurysm was found. Per patient basis, the diagnostic sensitivity and specificity were excellent. Per aneurysm basis, the diagnostic sensitivity and specificity of MSCTA were, respectively, 97.4 and 100% for reader A, 100 and 100% for reader B. For aneurysms less than 3mm, sensitivity was 100% for both readers. Interobserver agreement was excellent for the detection of aneurysm (kappa=0.98, 95% CI [0.96-1]). Intertechnique and interobserver agreements were excellent for the measurement of aneurysms (slope=0.86, r=0.91 p=3.1x10(-7) and slope=1.04, r=0.99, p<10(-6), respectively). CONCLUSION: MSCTA was an accurate and reproducible non-invasive imaging technique for preoperative diagnosis of ruptured cerebral aneurysm. The MSCTA may be proposed in first intention after the diagnosis of SAH was established, with special care regarding injection procedure and a strict reading method using native images and thin MPR.