Diagnostic accuracy of 64-slice multidetector CT angiography for detection of in-stent restenosis of vertebral artery ostium stents: comparison with conventional angiography. (October 2014)

• Record Type: Journal Article
• Authors: Lee, Youn Joo; Lim, Yeon Soo; Lim, Hyun Wook; Yoo, Won Jong; Choi, Byung Gil; Kim, Bum Soo
• Is Part Of: Acta radiologica. Volume 55:Number 8(2014:Oct.)
• Page Start: 1000
• Page End: 1007
• Publication Date: 2014-10
• Subjects: Vertebral artery ostium -- stent -- in-stent restenosis -- multidetector CT -- CT angiography; Radiology, Medical -- Periodicals; Radiography, Medical -- Periodicals; Radiotherapy -- Periodicals; 616.0757
• DOI: 10.1177/0284185113507922
• Languages: English
• ISSNs: 0284-1851

Abstract

Background: There are very few reports assessing in-stent restenosis (ISR) after vertebral artery ostium (VAO) stents using multidetector computed tomography (MDCT). Purpose: To compare the diagnostic accuracy of computed tomography angiography (CTA) using 64-slice MDCT with digital subtraction angiography (DSA) for detection of significant ISR after VAO stenting. Material and Methods: The study evaluated 57 VAO stents in 57 patients (39 men, 18 women; mean age 64 years [range, 48–90 years]). All stents were scanned with a 64-slice MDCT scanner. Three sets of images were reconstructed with three different convolution kernels. Two observers who were blinded to the results of DSA assessed the diagnostic accuracy of CTA for detecting significant ISR (≥50% diameter narrowing) of VAO stents in comparison with DSA as the reference standard. The sensitivity, specificity, positive and negative predictive values, and accuracy were calculated. Results: Of the 57 stents, 46 (81%) were assessable using CTA, while 11 (19%) were not. No stents with diameters ≤2.75 mm were assessable. DSA revealed 13 cases of significant ISR in all stents. The respective sensitivity, specificity, positive and negative predictive values, and accuracy were 92%, 82%, 60%, 97%, and 84% for all stents. On excluding the 11 non-assessable stents, the respective values were 88%, 95%, 78%, 97%, and 93%. Of the 46 CTA assessable stents, eight significant ISRs were diagnosed on DSA. Seven of eight patients with significant ISR by DSA were diagnosed correctly with CTA. The area under the receiver-operating characteristic curve (AUC) was 0.87 for all stents and 0.91 for assessable stents, indicating good to excellent agreement between CTA and DSA for detecting significant ISR after VAO stenting. Conclusion: Sixty-four-slice MDCT is a promising non-invasive method of assessing stent patency and can exclude significant ISR with high diagnostic values after VAO stenting.