However, among those patients with symptomatic carotid stenosis, the cumulative incidence of the primary endpoint was 16.8% in the CAS arm and 16.5% in the CEA arm. Based upon this trial, CAS has equivalent one-year outcomes versus CEA in a high-risk population.
click for large version
click for large version
The Carotid and Vertebral Artery Transluminal Angioplasty Study trial was the first large prospective trial comparing CEA and CAS among symptomatic patients with severe carotid stenosis (mean 86.4% stenosis).11 At 30 days, the rate of death or disabling stroke was 6.4% with CAS and 5.9% with CEA, which were not significantly different in this trial of about 500 patients.
The trial was begun in 1994, with a large portion of angioplasty performed without stents or distal embolic protection. There were fewer local complications but higher rates of restenosis in the CAS arm. The authors noted “no substantial difference in the rate of ipsilateral stroke … up to three years after randomization” but cautioned that the confidence intervals were wide.
Two recently published trials of CAS versus CEA in lower-risk populations do not support the overall safety of CAS among symptomatic patients. The Stent-Protected Angioplasty versus Carotid Endarterectomy trial randomized 1,200 average-risk patients with symptomatic carotid stenosis of 50% or greater by angiography or 70% of greater by ultrasound to either CAS or CEA.12
The trial design stipulated that both surgeons and percutaneous interventionalists perform at least 25 procedures prior to inclusion in the study and that independent quality committees review these procedures. The use of distal embolic protection devices was left to the discretion of the operators. The 30-day rate of death or ipsilateral ischemic stroke was 6.34% in the CEA arm and 6.84% in the CAS arm (p=0.09 for noninferiority).
The investigators concluded that CAS is not non-inferior to CEA (i.e., that CAS is inferior). The Endarterectomy versus Angioplasty in Patients with Symptomatic Severe Carotid Stenosis trial randomized 527 patients with symptomatic carotid stenosis of 70% or greater by angiography or magnetic resonance angiography (MRA) to either CAS or CEA within two weeks of the index event.13
This trial design also stipulated that surgeons had performed at least 25 CEAs in the prior year. Percutaneous interventionalists did not have similar numeric procedure requirements, although the investigators provided for tutoring of less experienced operators. The trial was stopped prematurely due to futility (in terms of noninferiority) and harm within the CAS arm.
The 30-day cumulative incidence of death or any stroke was 3.9% in the CEA arm and 9.6% in the CAS arm (p=0.01 for superiority of CEA). The trial was powered to detect only large differences among low- and high-volume operators. Nearly 10% of patients did not have distal embolic protection devices used during their CAS procedures. Ongoing trials will further define the role of CAS versus CEA in the interventional treatment of carotid stenosis.
Accurate Diagnosis
Different trials used different criteria for defining the percent stenosis of the diseased carotid arterial segment. These differences were based primarily on the mode of testing (i.e., conventional angiography versus ultrasound), and on what portion of the carotid artery was used as the reference or baseline segment to calculate the percent stenosis.
A meta-analysis of various non-invasive modes of testing for carotid stenosis concluded that duplex ultrasound had a pooled sensitivity and specificity of 86% and 87%, respectively, to distinguish 70% to 99% stenosis from less than 70% stenosis.14 MRA had a pooled sensitivity and specificity of 95% and 90%, respectively.