Real World Accuracy of Carotid Artery Stenosis Measurements. (16th November 2020)

• Record Type: Journal Article
• Authors: Farrell, Mary Beth; Sacks, David; Sila, Cathy; Terry, Johnn; Cockroft, Kevin M
• Is Part Of: Neurosurgery. Volume 67(2010)Supplement 1
• Publication Date: 2020-11-16
• Subjects: Nervous system -- Surgery -- Periodicals; 617.48005
• DOI: 10.1093/neuros/nyaa447_250
• Languages: English
• ISSNs: 0148-396X

Abstract

Abstract: INTRODUCTION: Degree of stenosis is an important factor in treatment decision making for patients with cervical carotid stenosis. This is especially true for asymptomatic patients. While participants in clinical trials are typically trained in specific measurement systems, real world assessors may be less rigorous. METHODS: Representative images were selected from random cases submitted to the Intersocietal Accreditation Commission as part of accreditation applications. Operator-reported stenosis (ORS) as documented in the patient's operative report was compared to a consensus-measured stenosis (CMS) determined by three IAC board members using NASCET criteria. Results were categorized into three stenosis levels: 0–39%, 40–69%; and 70–100%. The results were compared using Fleiss' and Cohen's kappa. RESULTS: Angiograms from 65 patients were reviewed. The maximum range of measurement between the three reviewers was 37% with a mean of 12% (SD 8%). Fleiss' kappa showed moderate agreement between the three reviewers (Κ = 0.543; 95%CI:0.539-0.547) with agreement between reviewers 2 and 3 good (Κ = 0.621; 95%CI:0.439-0.803) and moderate between reviewers 1 and 2 (Κ = .476, (95%CI:0.270 to 0.682) and between reviewers 1 and 3 (Κ = .580; 95%CI:0.388-0.772). Agreement between ORS and CMS was poor (Κ = 0.080; 95%CI:0.066-0.166). ORS was significantly higher than the CMS (86% vs 61%, P<. 001). ORS was >70% for 88% of patients (n = 57), while CMS was >70% for only 32% (n = 21). Of the 36 symptomatic patients, ORS was >70% in 28 patients, while CMS was >70% for only 12 patients. For the 29 asymptomatic patients, ORS was >70% for all 29, but CMS found only 9 patients with the same degree of stenosis. CONCLUSION: Real world operators tend to overestimate the degree of cervical carotid artery stenosis. Since decisions regarding carotid revascularization, particularly for asymptomatic patients, are often based at least in part on the percent stenosis, such measuring discrepancies may lead to procedural over utilization.