← Back
NeuroTrials.ai
Neurology Clinical Trial Database

DIRECT ANGIO

Safety and efficacy of direct versus conventional transfer to angiography suite in patients with severe acute stroke treated with thrombectomy (DIRECT ANGIO) in France: a multicentre, open-label, blinded-endpoint, randomised controlled trial

Share Share Copied! Compare

Year of Publication: 2026

Authors: Gory B, Maïer B, Pop R, ..., on behalf of the DIRECT ANGIO Investigators

Journal: Lancet Neurology

Citation: Lancet Neurol 2026; 25: 346-56

Link: https://clinicaltrials.gov/study/NCT03969511

Clinical Question

Does direct transfer to the angiography suite, bypassing baseline imaging, improve outcomes in patients with suspected large vessel occlusion stroke?

Bottom Line

Direct transfer to angiography suite (DTAS) for patients with suspected large vessel occlusion stroke does NOT improve functional outcomes at 90 days and is associated with a significantly higher risk of symptomatic intracranial haemorrhage. The trial was stopped early for safety reasons. Clinicians should continue to use the conventional pathway with baseline imaging before considering endovascular treatment.

Major Points

  • First multicentre RCT comparing DTAS vs conventional pathway in patients without previous imaging
  • Trial stopped early on Dec 1, 2023 for safety concerns after interim analysis showed increased symptomatic ICH in DTAS group
  • DTAS reduced admission-to-puncture time by 38 minutes and onset-to-reperfusion by 37 minutes
  • Despite faster workflow, DTAS did not improve 90-day functional independence (36% vs 42%)
  • Symptomatic intracranial haemorrhage was significantly increased with DTAS (15% vs 0%, adjusted OR 11.0)
  • All-cause mortality numerically higher with DTAS but not statistically significant (18% vs 11%)
  • Findings raise safety concerns for bypassing baseline imaging in patients primarily admitted to endovascular-capable centres
  • Small sample size due to early termination limits precision of effect estimates

Design

Study Type: Pragmatic, investigator-initiated, multicentre, open-label, randomised controlled trial with masked outcome assessment (PROBE design)

Randomization: 1

Blinding: Open-label with blinded endpoint assessment; neuroimaging data centrally assessed by core laboratory masked to trial groups

Allocation: 1:1 web-based randomisation, stratified by centre and time from symptom onset to admission (<2.5h vs ≥2.5h), using minimisation algorithm with 85% allocation factor

Enrollment Period: July 9, 2020 to April 18, 2023 (stopped early Dec 1, 2023)

Follow-up Duration: 90 days (with 1-year follow-up mentioned)

Centers: 10

Countries: France

Sample Size: 115

Analyzed: 115

Analysis: Intention-to-treat analysis; all randomly assigned patients in their originally assigned treatment groups, irrespective of diagnosis, imaging findings, or treatments received

Power Calculation: Original power calculation not completed due to early termination for safety

Registration: ClinicalTrials.gov NCT03969511

Inclusion Criteria

  • Adult patients aged 18-85 years
  • Acute severe neurological deficit highly suggestive of ischaemic stroke due to large vessel occlusion (ASND-LVO)
  • Unilateral motor deficit with NIHSS motor score ≥5 (facial palsy, arm palsy, or leg palsy)
  • Cortical symptom with NIHSS score ≥1 on language or extinction items
  • Direct admission to endovascular-capable centre within 5 hours from witnessed symptom onset
  • Arrival without any previous neuroimaging

Exclusion Criteria

  • Pre-stroke disability (mRS score >2)
  • Terminal illness with expected survival <90 days
  • Pregnancy or breastfeeding
  • Known severe allergy to contrast agents
  • Unwitnessed onset or last well known timing
  • Secondary transfers with imaging already done at primary stroke centres

Arms

FieldDirect Transfer to Angiography Suite (DTAS)Control
N5758
InterventionDirect transfer to angiography suite where cone-beam CT (CBCT) performed to rule out intracranial haemorrhage, followed by catheter cerebral angiogram to identify large vessel occlusion, then immediate endovascular treatment if LVO confirmed. IV thrombolysis with alteplase administered if eligible.Standard of care: admission at emergency or radiology department with baseline imaging followed by transfer to angiography suite for endovascular treatment if eligible
DurationSingle procedureSingle procedure

Outcomes

OutcomeTypeControlInterventionHR / OR / RRP-value
Functional independence defined as modified Rankin Scale (mRS) score 0 to 2Primary22/53 (42%)20/56 (36%)0.73
Reduction in admission to puncture timeSecondaryMedian reduced by 38 minutes with DTAS
Reduction in onset to reperfusion timeSecondaryMedian reduced by 37 minutes with DTAS
Symptomatic intracranial haemorrhageSafety0/42 (0%)5/34 (15%)Adjusted OR 11.0
All-cause mortality at 90 daysSafety6/53 (11%)10/56 (18%)Adjusted OR 1.65
Symptomatic Intracranial Haemorrhage DTASAdverse5/34 (15%)
Symptomatic Intracranial Haemorrhage ConventionalAdverse0/42 (0%)
Mortality DTASAdverse10/56 (18%)
Mortality ConventionalAdverse6/53 (11%)

Criticisms

  • Trial stopped early for safety reasons resulting in small sample size (115 patients) that limits precision of effect estimates
  • Wide confidence intervals on primary safety outcome (OR 11.0, 95% CI 1.28-1406) reflect imprecision
  • Open-label design with potential for bias in management decisions, although outcome assessment was masked
  • Cost-utility assessment could not be performed due to premature termination
  • Discrepancy in denominators (n=34 vs n=42 for symptomatic ICH analysis vs n=56 vs n=53 for primary outcome) suggests post-randomisation exclusions
  • Generalisability may be limited to high-volume comprehensive stroke centres

Funding

French Ministry of Health (PHRC I 2018) and Medtronic. The trial was funded by the French Ministry of Health with no commercial involvement in design, planning, analysis, or reporting of results.

Based on: DIRECT ANGIO (Lancet Neurology, 2026)

Authors: Gory B, Maïer B, Pop R, ..., on behalf of the DIRECT ANGIO Investigators

Citation: Lancet Neurol 2026; 25: 346-56

Content summarized and formatted by NeuroTrials.ai.