ROADSTER 3 analysis explores role of TCAR in asymptomatic carotid disease 

The CREST-2 trial provided important new evidence on carotid revascularization in asymptomatic patients, but it left one question unanswered: Where does transcarotid artery revascularization (TCAR) fit? A new analysis from the ROADSTER 3 trial offers a direct look at that question. 

Researchers analyzed outcomes for 288 asymptomatic, standard surgical risk patients enrolled in ROADSTER 3 whose eligibility criteria closely mirrored those used in CREST-2. TCAR was not included as a treatment arm in CREST-2, a gap this analysis was designed to address. 

“ROADSTER 3 is the first independently adjudicated, prospective study evaluating TCAR in a standard surgical risk population,” said Marc Schermerhorn, MD, chief of the division of Vascular and Endovascular Surgery at Beth Israel Deaconess Medical Center and a professor of surgery at Harvard Medical School. “Our objective with this post hoc analysis was to assess outcomes for standard risk, asymptomatic patients, enrolled with eligibility criteria similar to CREST-2, who underwent TCAR in the ROADSTER 3 trial.” The analysis was presented at the 2026 meeting of the Midwestern Vascular Surgical Society in Chicago, Illinois (Sept. 17-20). 

The primary composite endpoint was periprocedural stroke and death within 30 days, plus any ipsilateral stroke from 31 to 365 days. Among intention-to-treat patients, the 30-day stroke and death rate was 1%. In the per-protocol cohort, it was 0.7%. Notably, no deaths occurred. 

 Through one year, one additional ipsilateral stroke was recorded, bringing the composite to 1.4% for intention-to-treat and 1.2% for per-protocol patients. All events were independently adjudicated and patients underwent independent neurological assessments. 

Those numbers take on additional weight when measured against CREST-2. In that trial, the 30-day stroke and death rate was 1.3% for transfemoral CAS and 1.5% for CEA — both higher than what ROADSTER 3 observed for TCAR in a comparable asymptomatic population. 

“The periprocedural incidence of stroke and death for asymptomatic standard surgical risk patients undergoing TCAR was low relative to the benchmarks established for transfemoral CAS and CEA in CREST-2,” said Schermerhorn. “TCAR should be considered in the shared decision-making process with patients needing carotid revascularization.” 

The patient population in this analysis also included characteristics that would have excluded them from the CAS arm of CREST-2. Roughly 11% had a Type III aortic arch and nearly 33% had arch atheroma — both exclusion criteria for transfemoral CAS in CREST-2 — yet TCAR was still performed with low complication rates. That anatomic distinction matters clinically, as it suggests TCAR may reach patients for whom transfemoral stenting is not a viable option. 

“Several anatomic features that increase risk for CEA or transfemoral CAS procedures do not apply to TCAR,” said Schermerhorn. “TCAR may be a feasible option for patients with these characteristics and this analysis from ROADSTER 3 adds to the evidence supporting safety and effectiveness of TCAR for asymptomatic patients.” 

The analysis stops short of a direct randomized comparison and Schermerhorn was direct about what the field still needs. Observational data comparing TCAR with CEA have largely shown equivalent outcomes and outcomes with transfemoral CAS have generally been inferior, yet formal randomized evidence pitting all three approaches against one another in asymptomatic patients remains absent. Schermerhorn also pointed to the ongoing challenge of patient selection, noting that most asymptomatic patients ultimately do not require revascularization and that high-risk plaque characteristics should drive who is treated. 

“Subgroup analysis of asymptomatic patients undergoing TCAR demonstrates lower periprocedural incidence of stroke or death than the recently published CREST-2 trial of transfemoral CAS and CEA,” said Schermerhorn. “With the enthusiasm of CREST-2 reinforcing the benefit of revascularization in asymptomatic patients, we still need to improve our patient selection for these procedures based on high-risk plaque characteristics, since the majority of patients do not need revascularization and we need to focus our interventions on those at greatest risk.”