Major Finding: Carotid endarterectomy, perioperative stroke, and death rates for symptomatic patients have declined over the last 2 decades.
Data Source: NASCET I and II, ACAS, and CREST trials.
Disclosures: The research was supported by the National Institute of Neurological Disorders and Stroke, with supplemental support from Abbott Vascular Inc. Dr. Lal reported that he has no relevant financial relationships.
BOSTON — Perioperative stroke and death rates associated with carotid endarterectomy have improved for symptomatic patients over the last 20 years, based on a comparison of newly released data.
“The periprocedural stroke and death rate has not changed dramatically in asymptomatic patients, whereas it has almost halved in patients with symptomatic disease,” Dr. Brajesh K. Lal observed at the meeting.
Dr. Lal and his coinvestigators compared periprocedural stroke and death rates associated with carotid endarterectomy (CEA) in CREST (Carotid Endarterectomy vs. Stenting Trial), NASCET I and II (North American Symptomatic Carotid Endarterectomy Trials I and II), and ACAS (Asymptomatic Carotid Atherosclerosis Study).
The perioperative stroke plus death rate for symptomatic patients who underwent CEA was 5.8%, 6.7%, and 3.2% in the NASCET I and II (N. Engl. J. Med. 1991;325:445-53; 1998;339:1415-25) and CREST (N. Engl. J. Med. 2010;363:11-23) trials, respectively. “Operative stroke and death in symptomatic patients [in the CREST trial] were lower than in NASCET and could be related to increased intraoperative shunting, patch angioplasty, antilipid therapy, or antihypertensive therapy,” said Dr. Lal of the division of vascular surgery at the University of Maryland in Baltimore.
Although antiplatelet therapy following CEA was used at similar rates in all of the trials, postprocedure antilipid therapy was used much more frequently in the CREST trial: 76% for both symptomatic and asymptomatic patients in CREST, compared with 14% in NASCET I and 40% in NASCET II. Likewise, postprocedure antihypertensive therapy was used more frequently in CREST than in NASCET I and II: 80%, 54%, and 68%, respectively.
For asymptomatic patients, the perioperative stroke plus death rate was 2.3% and 1.4% in ACAS (JAMA 1995;273:1421-8) and CREST, respectively. However, the ACAS rate included a 1.2% angiography complications rate, Dr. Lal noted. If angiographic complications are excluded from the ACAS results, the perioperative stroke plus death rate for asymptomatic patients was comparable for ACAS and CREST (1.5% for ACAS vs. 1.4% for CREST).
The sesearcherr also compared inclusion criteria, enrollment details, criteria used to credential surgeons, criteria used to define perioperative stroke, surgical technique, and perioperative medical therapy.
NASCET I was open to patients older than 80 years, and ACAS was open to patients aged 40-79 years. NASCET II and CREST were open to patients of all ages.
NASCET I included symptomatic patients with 70%-99% stenosis; NASCET II was open to symptomatic patients with 50%-69% stenosis. ACAS was open to asymptomatic patients with 60% or greater stenosis. CREST was open to asymptomatic patients with 60% or greater stenosis and symptomatic patients with 50% or greater stenosis.
In patients in both NASCETs and roughly half of the patients in ACAS, the percentage of stenosis was determined by means of preoperative angiograms. In CREST, the percentage of stenosis was determined by ultrasound, CT angiography, MR angiography, and angiogram. There were only slight differences among the trials in terms of the definition of symptomatic status.
Both NASCETs as well as ACAS started enrolling patients in 1987 and stopped enrolling patients in the 1990s, whereas CREST enrolled patients during 2000-2008. Both NASCET I and ACAS were conducted in a smaller number of centers (50 and 39, respectively). NASCET II and CREST were conducted in more than 100 centers each (106 and 117, respectively). There were more patients randomized to CEA in the CREST trial (1,240) than in the other three trials (328 for NASCET I, 430 for NASCET II, and 825 for ACAS).
In terms of baseline characteristics, patients in the trials did not differ by age, sex distribution, or race. The rates of smoking, cardiovascular disease, and preprocedure antiplatelet therapy also did not differ among the trials. However, CREST enrolled more patients with hypertension, diabetes, and dyslipidemia. Also, shunting and patch angioplasty were used more frequently in CREST.
“The credentialing criteria for surgeons were fairly consistent across all of the trials,” said Dr. Lal. Both CREST and ACAS required surgeons to perform at least 12 CEAs each year. However, there were no established surgeon volume requirements in the two NASCETs. All four trials included surgical management committees, site evaluations, and reviews of indications, hospital courses, and outcomes in the 50 most recent CEAs.