Evaluation of Noninvasive Follow-up Methods for the Detection of Intracranial In-Stent Restenosis

Abstract

Intra-arterial digital subtraction angiography (IA-DSA), an invasive procedure, is the current reference examination after percutaneous transluminal angioplasty and stenting for the detection of in-stent restenosis (ISR). In this phantom study, we evaluated flat-panel angiographic computed tomography after intravenous contrast agent application (IV-ACT) and multidetector computed tomographic angiography (MDCTA) as potential noninvasive follow-up alternatives after intracranial percutaneous transluminal angioplasty and stenting. We simulated an intracranial vessel using a silicon tube placed inside a human skull. Three different stent systems were deployed inside the silicon tubes, each with diameters of 3 or 4 mm. Three grades of ISR (25%, 50%, and 75%) were simulated. The IA-DSA and IV-ACT examinations were performed on a flat-panel detector angiography system. The MDCTA images were acquired with a 128-slice computed tomographic scanner. The mean stenosis diameters, measured with each technique, were compared using the Bland-Altman plot. The difference between the known stenosis diameter and the measured stenosis diameter was calculated for each examination. Stenosis measurements on the IA-DSA images showed no statistically significant differences compared with the known stenosis diameters (P = 0.19). In the 3-mm stent category, when compared with the known stenosis diameter, mean (SD) differences of 0.01 (0.15) mm, 0.03 (0.24) mm and 0.16 (0.5) mm were calculated for the IA-DSA, IV-ACT, and MDCTA stenosis measurements, respectively. As for the 4-mm stents, IA-DSA and IV-ACT were again very accurate, with mean (SD) differences of -0.03 (0.11) mm and 0.07 (0.19) mm, respectively, compared with the known stenosis diameters, whereas MDCTA overestimated ISR, with a mean (SD) difference of 0.49 (0.53) mm. The Bland-Altman plots show a mean (SD) difference of 0.08 (0.2) mm between IA-DSA and IV-ACT (95% confidence interval, 0.05-0.11) and a mean (SD) difference of 0.34 (0.56) mm between IA-DSA and MDCTA measurements (95% confidence interval, 0.25-0.42). In our phantom study, IA-DSA was the only examination to predict accurately degrees of stenosis compared with the known stenosis diameters. The results of the IV-ACT measurements were comparable with those of IA-DSA. Multidetector computed tomographic angiography was less accurate in the quantification of stenosis, usually overestimating ISR.