In a debate on CREST (Carotid revascularization endartarectomy versus stenting trial) held at the VEITHsymposium, three investigators in the trial showed different views on the messages delivered by the study.
William Gray, Columbia University Medical Center, gave the interventionalist perspective. He said that CREST has restored carotid artery stenting to its rightful place as best treatment for most carotid stenoses. Gray told delegates that CREST established both carotid endarterectomy and stenting as safe and effective revascularisation options, with certain clinical and angiographic characteristics driving the choice of therapy offered. “Ultimately these therapies should be complementary, not competitive or exclusive,” he said.
Gray pointed out that carotid artery stenting outcomes in CREST have already been improved on in multiple studies due to operator experience, patient selection, and possibly devices. He said that medical therapy for asymptomatic patients has improved “to a point where equipoise exists” and added that a trial comparing medical therapy and revascularisation is appropriate.
Wesley S Moore, UCLA Medical Center, who gave the surgeon’s perspective, stressed that CREST may be helpful to advocates of carotid stenting but it is not the final answer. He proposed a moratorium on both procedures as a way to accelerate a new clinical trial that could provide clarity for determining best practices for patients with carotid stenosis.
Moore drew attention to the initial delay and difficulties in recruiting patients and the subsequent decision by CREST researchers and the FDA to include asymptomatic patients in order to reach the 2,500 cohort called for in the study’s design. Including both asymptomatic and symptomatic patients, he noted, diluted the study’s power to compare the two procedures because asymptomatic patients tend to have fewer complications than symptomatic patients.
That inherent flaw, Moore said, attenuated the somewhat conflicting findings that have fuelled the controversy over CREST’s outcomes. The primary endpoints of death, stroke and myocardial infarction were 4.5% for endarterectomy and 5.2% for stenting. These differences were not statistically significant. However, when the primary endpoints of death and stroke were analysed alone, the rate was 2.3% for endarterectomy versus 4.4% for stenting and those differences were statistically significant at p=0.005. There were more non-fatal myocardial infarctions with endarterectomy, which resulted in equalising the aggregate of the three primary endpoints between the two procedures.
Moore endorsed the need for a new three-arm trial of asymptomatic patients comparing outcomes of stenting combined with optimal medical treatment, endarterectomy combined with optimal medical treatment, and medical treatment alone. And, to avoid the recruitment delays that plagued CREST, he called for a temporary halt for stenting or endarterectomy treatment of asymptomatic patients with carotid stenosis so that the pool of potential subjects for the new trial would be as large as possible thus ensuring that “accrual of patients in the trial would occur rapidly and the results would be known in short order.”
The neurologist’s perspective was given by Thomas Brott. He showed papers on CREST which have already been published. He said that this year a paper on age, myocardial infarction, sex and also safety will be submitted for publication. The quality of life results, Brott said, have been accepted by the American Heart Association International Stroke Conference and will be presented by David Cohen in February 2012. “There are differences with regard to endpoints and also differences with the entire cohort,” he said.
He added that cost effectiveness analyses will completed this year, and the results will be ready for presentation and publication by June 2011.
Brott spoke on the severity and disability from myocardial infarctions in CREST. Myocardial infarctions are associated with significant increase in later mortality up to four years, he said. “However, we were a little surprised to see that the occurrence of periprocedural strokes was also associated with significant increase in later mortality, up to four years. Whether or not these findings indicate markers for later morbidity, mortality or cause and effect we are trying to tease up.”
William Gray, Columbia University Medical Center, gave the interventionalist perspective. He said that CREST has restored carotid artery stenting to its rightful place as best treatment for most carotid stenoses. Gray told delegates that CREST established both carotid endarterectomy and stenting as safe and effective revascularisation options, with certain clinical and angiographic characteristics driving the choice of therapy offered. “Ultimately these therapies should be complementary, not competitive or exclusive,” he said.
Gray pointed out that carotid artery stenting outcomes in CREST have already been improved on in multiple studies due to operator experience, patient selection, and possibly devices. He said that medical therapy for asymptomatic patients has improved “to a point where equipoise exists” and added that a trial comparing medical therapy and revascularisation is appropriate.
Wesley S Moore, UCLA Medical Center, who gave the surgeon’s perspective, stressed that CREST may be helpful to advocates of carotid stenting but it is not the final answer. He proposed a moratorium on both procedures as a way to accelerate a new clinical trial that could provide clarity for determining best practices for patients with carotid stenosis.
Moore drew attention to the initial delay and difficulties in recruiting patients and the subsequent decision by CREST researchers and the FDA to include asymptomatic patients in order to reach the 2,500 cohort called for in the study’s design. Including both asymptomatic and symptomatic patients, he noted, diluted the study’s power to compare the two procedures because asymptomatic patients tend to have fewer complications than symptomatic patients.
That inherent flaw, Moore said, attenuated the somewhat conflicting findings that have fuelled the controversy over CREST’s outcomes. The primary endpoints of death, stroke and myocardial infarction were 4.5% for endarterectomy and 5.2% for stenting. These differences were not statistically significant. However, when the primary endpoints of death and stroke were analysed alone, the rate was 2.3% for endarterectomy versus 4.4% for stenting and those differences were statistically significant at p=0.005. There were more non-fatal myocardial infarctions with endarterectomy, which resulted in equalising the aggregate of the three primary endpoints between the two procedures.
Moore endorsed the need for a new three-arm trial of asymptomatic patients comparing outcomes of stenting combined with optimal medical treatment, endarterectomy combined with optimal medical treatment, and medical treatment alone. And, to avoid the recruitment delays that plagued CREST, he called for a temporary halt for stenting or endarterectomy treatment of asymptomatic patients with carotid stenosis so that the pool of potential subjects for the new trial would be as large as possible thus ensuring that “accrual of patients in the trial would occur rapidly and the results would be known in short order.”
The neurologist’s perspective was given by Thomas Brott. He showed papers on CREST which have already been published. He said that this year a paper on age, myocardial infarction, sex and also safety will be submitted for publication. The quality of life results, Brott said, have been accepted by the American Heart Association International Stroke Conference and will be presented by David Cohen in February 2012. “There are differences with regard to endpoints and also differences with the entire cohort,” he said.
He added that cost effectiveness analyses will completed this year, and the results will be ready for presentation and publication by June 2011.
Brott spoke on the severity and disability from myocardial infarctions in CREST. Myocardial infarctions are associated with significant increase in later mortality up to four years, he said. “However, we were a little surprised to see that the occurrence of periprocedural strokes was also associated with significant increase in later mortality, up to four years. Whether or not these findings indicate markers for later morbidity, mortality or cause and effect we are trying to tease up.”
