Interim data find ‘no evidence of benefit’ for carotid revascularization additional to optimized medical therapy

Paul Nederkoorn presenting at ESOC 2023

Interim results from the ECST-2 randomized controlled trial (RCT) have shown no evidence that carotid stenosis patients with a low-to-intermediate stroke risk, treated with optimized medical therapy, will benefit from additional carotid revascularization via carotid endarterectomy (CEA) or carotid artery stenting (CAS).

Nevertheless, Paul Nederkoorn (Amsterdam UMC, Amsterdam, The Netherlands)—who presented these late-breaking findings at the European Stroke Organisation Conference (ESOC; 24–26 May, Munich, Germany)—noted that a longer clinical follow-up is now required. The aforementioned interim results were observed at two years, with ECST-2 set to continue following patients for a total of five years.

In an attempt to provide an update to the original European carotid surgery trial (ECST), which was published in the late 1990s and found CEA to be effective in symptomatic patients with carotid stenosis >80%, ECST-2 hypothesized that there would be no benefit from additional carotid revascularization alongside optimized medical therapy in patients who have carotid stenosis ≥50% and a low-to-intermediate risk of stroke. Nederkoorn stated during his presentation that, in the decades following the ECST and NASCET RCTs, medical management—the comparator in both trials—has improved “significantly”.

ECST-2 was designed as a multicenter, prospective RCT with blinded-outcome adjudication, and saw patients randomized to either immediate carotid revascularization plus optimized medical therapy or optimized medical therapy alone. Patients were considered eligible if they had an asymptomatic or symptomatic carotid stenosis ≥50% and an estimated five-year stroke risk <20%, calculated using the Carotid Artery Risk (CAR) score. Recruitment was halted after the inclusion of 428 patients, Nederkoorn reported.

For the composite endpoint defined for these interim results—the two-year rate of any stroke, myocardial infarction (MI) or periprocedural death—Nederkoorn and colleagues found a hazard ratio of 0.96 with optimized medical therapy alone in comparison to optimized medical therapy plus revascularization (95% confidence interval [CI] 0.53–1.76). The speaker noted that there was “a sign of more harm” initially with revascularization, signified by a higher rate of stroke, MI or periprocedural death in the first few months following randomization, but that these rates converged later on and Kaplan-Meier analyses ultimately showed no significant difference between the two groups at 24 months.

While the numbers involved were “very small, precluding valid conclusions,” Nederkoorn reported no differences across any of the predefined subgroups in ECST-2—including between symptomatic and asymptomatic stenoses. There was, however, a slight trend towards optimized medical therapy being associated with a greater treatment effect in patients with an increased risk of ipsilateral stroke (CAR score >10%).

From this interim analysis of the ECST-2 intention-to-treat population, the investigators concluded that there is no evidence patients with stenosis ≥50% and a low-to-intermediate stroke risk will benefit from carotid revascularization in addition to optimized medical therapy.

As well as emphasizing that, despite this finding, further follow-up is both warranted and planned in ECST-2, Nederkoorn stated that a complete, two-year analysis including evaluations of silent infarcts on magnetic resonance imaging (MRI) is now scheduled too. Also, the results of an MR plaque-imaging substudy from ECST-2 will soon be available, which Nederkoorn said is a “crucial tool” within a novel prediction rule that is still to be designed but may elucidate future stroke risks and treatment decisions in this domain of patients.

A subsequent discussion at ESOC 2023 saw Seemant Chaturvedi (University of Maryland School of Medicine, Baltimore, USA) posit that the medical therapy given in the trial’s comparator arm could have been optimized “even further” in accordance with some current guideline recommendations, potentially improving that group’s outcomes to a still-greater extent—a point that Nederkoorn agreed with.


Please enter your comment!
Please enter your name here