Agreement between CT-Angiography and Digital Subtraction Angiography in Predicting Angiographic Vasospasm in Patients with Subarachnoid Hemorrhage

Abstract

Background/Objectives: Digital subtraction angiography (DSA) is the gold standard in the diagnosis of cerebral vasospasm, frequently observed after subarachnoid hemorrhage (SAH). However, less-invasive methods, such as computed tomography angiography (CTA), may be equally accurate. To further clarify comparability, this study evaluated the reliability of CTA in detecting cerebral vasospasm. Methods: This retrospective study included 51 patients with SAH who underwent both CTA and DSA within 24 h. The smallest diameter of the proximal cerebral arterial segments was measured in both modalities at admission and during the vasospasm period. The mean difference in diameter, the intraclass correlation coefficient (ICC) of CTA and DSA, the difference in grade of vasospasm and sensitivity, the specificity and the positive predictive value (PPV) for CTA were calculated. Results: A total of 872 arterial segments were investigated. At time of admission, arterial diameters were significantly smaller on CTA compared to DSA in all segments (−0.26 ± 0.12 mm; p < 0.05). At time of suspected vasospasm (day 9 ± 5), these differences remained significant only for the M1 segment (−0.18 ± 0.37 mm, p = 0.02), the P1 segment (−0.13 ± 0.24 mm, p = 0.04) and the basilar artery (−0.20 ± 0.37 mm, p = 0.0.04). The ICC between CTA and DSA was good (0.5–0.8). The sensitivity of CTA for predicting angiographic vasospasm was 99%, the specificity was 50% and the PPV was 92%. Conclusions: Arterial diameters measured on CTA may underestimate the arterial caliber observed in DSA; however, these absolute differences were minor. Importantly, vessel diameter alone does not fully reflect malperfusion, requiring additional imaging techniques such as CT perfusion.