PDF: DO-IT
(2025)Objective
To investigate the safety and efficacy of off-label intravenous thrombolysis (IVT) in acute ischemic stroke patients with recent direct oral anticoagulant (DOAC) intake using target trial emulation methodology, and to compare outcomes between IVT with and without prior DOAC reversal with idarucizumab.
Study Summary
• IVT was associated with significantly better functional outcomes (mRS 0-2 or return to baseline: 62.2% vs 43.7%; adjusted difference +14.4%, 95% CI +7.1% to +21.8%)
• No significant difference in sICH, major bleeding, or efficacy outcomes between IVT with vs without idarucizumab reversal
• Safety findings held even in patients with DOAC intake <12 hours and plasma levels >100 ng/mL (sICH 1.96%)
• DOAC plasma levels and time from last intake were not independent predictors of sICH in exploratory LASSO regression
• In patients not receiving EVT, IVT showed an even larger adjusted benefit in good functional outcome (+19.8%)
Intervention
Intravenous thrombolysis (alteplase) administered off-label to acute ischemic stroke patients with recent DOAC intake (within 48 hours), with or without prior DOAC reversal using idarucizumab
Inclusion Criteria
Adults with acute ischemic stroke and IVT indication; arrival within 12 hours of symptom onset; NIHSS score >=2 with clinically disabling deficit; DOAC intake within 48 hours before anticipated IVT bolus or active DOAC prescription with unverifiable last intake time
Study Design
Arms: Target trial: IVT group (n=342) vs No IVT group (n=1000). Reversal comparison: IVT with idarucizumab reversal (n=289) vs IVT without reversal (n=283)
Patients per Arm: IVT: 342; No IVT: 1000; IVT with reversal: 289; IVT without reversal: 283
Outcome
• Primary efficacy: Good functional outcome (mRS 0-2 or return to baseline at 90 days) — IVT 62.2% vs No IVT 43.7% (adjusted difference +14.4%, 95% CI +7.1% to +21.8%)
• 90-day mortality: IVT 15.7% vs No IVT 22.7% (adjusted difference -3.3%, 95% CI -9.5% to +2.9%)
• mRS shift analysis favored IVT (aOR 0.60, 95% CI 0.45-0.80)
• No significant difference in any radiological ICH or major bleeding between groups
Bottom Line
This target trial emulation confirms the safety of off-label IVT in patients with recent DOAC intake, showing no increase in symptomatic ICH (3.0% vs 5.9% without IVT) and improved functional outcomes (62.2% vs 43.7% good outcome). DOAC reversal with idarucizumab before IVT did not significantly improve safety or efficacy outcomes compared to IVT without reversal, though any ICH was lower with reversal. More data needed for patients with confirmed high DOAC plasma levels.
Major Points
- International multicenter target trial emulation from 28 comprehensive stroke centers (2019-2023)
- 1342 patients on DOACs meeting IVT criteria; 342 (25%) received IVT
- Median age 80 years, median NIHSS 11, 50% female, 88% had atrial fibrillation
- 52% of patients received endovascular therapy
- Primary safety: sICH 3.0% with IVT vs 5.9% without IVT (adjusted difference -2.1%, p=NS)
- Primary efficacy: good functional outcome 62.2% with IVT vs 43.7% without (adjusted difference +14.4%, 95% CI +7.1 to +21.8)
- Among IVT patients: 41% had DOAC intake <12h, 27% within 12-24h
- 67% of IVT patients had DOAC plasma levels measured; median 29 ng/mL
- 25% of IVT patients had plasma levels >50 ng/mL, 10% had >100 ng/mL
- No association between DOAC plasma levels or time from intake with sICH risk
- Reversal comparison: 289 patients with idarucizumab vs 283 without
- sICH numerically lower with reversal (2.1% vs 3.7%) but not statistically significant
- Any ICH significantly lower with reversal (3.8% vs 21.4%, p<0.001)
- No difference in functional outcomes between reversal and non-reversal groups
- Benefits of IVT only tangible in centers with liberal approach (>20% IVT rate)
- Target trial methodology used to minimize observational bias
- LASSO regression identified EVT, ASPECTS, platelet count, glucose as predictors of sICH, not DOAC levels or time from intake
Study Design
- Study Type
- Retrospective, multicenter, observational cohort study using target trial methodology to emulate a randomized controlled trial
- Randomization
- No
- Blinding
- Not applicable (observational study). Target trial framework used to minimize bias by simulating conditions of hypothetical RCT.
- Sample Size
- 1342
- Follow-up
- 90 days (±2 weeks)
- Centers
- 28
- Countries
- Multiple international sites - predominantly European, some New Zealand sites
Primary Outcome
Definition: Safety: Symptomatic intracerebral hemorrhage (sICH) defined by ECASS III criteria (any hemorrhage with neurological deterioration defined as NIHSS increase ≥4 points or leading to death, within 36 hours of treatment). Assessed in 328 IVT patients (96%) and 921 no-IVT patients (92%). Efficacy: Good functional outcome defined as mRS score 0-2 at 90 days or return to baseline mRS.
| Control | Intervention | HR/OR | P-value |
|---|---|---|---|
| sICH: 54/921 (5.9%, 95% CI 4.4-7.6%); Good functional outcome: 405/921 (43.7%, 95% CI 40.5-47.0%) | sICH: 10/328 (3.0%, 95% CI 1.5-5.5%); Good functional outcome: 181/328 (62.2%, 95% CI 56.4-67.8%) | - (sICH adjusted difference: -2.1% (95% CI -5.3 to +1.2); Good outcome adjusted difference: +14.4% (95% CI +7.1 to +21.8)) | Not reported for primary outcomes |
Limitations & Criticisms
- Observational study despite target trial methodology - potential for unmeasured confounding
- Indication bias evident: IVT patients had lower stroke severity, fewer LVOs, fewer comorbidities
- Higher mortality in non-IVT group suggests sicker population not captured by measured variables
- Limited number of patients with DOAC plasma levels >100 ng/mL (n=35 in IVT group)
- Most IVT patients had low/intermediate DOAC levels - limited generalizability to high levels
- Reversal comparison compromised by different populations (New Zealand vs target trial)
- Significant baseline differences in reversal comparison groups
- Higher rate of MRI follow-up in non-reversal group (higher sensitivity for hemorrhage detection)
- Only 67% of IVT patients and 40% of non-IVT patients had DOAC plasma levels measured
- 18.4% of IVT patients had unverifiable last DOAC intake time
- Approaches to IVT in DOAC patients varied widely between centers
- Most centers were large academic EVT-capable centers - limited generalizability
- Limited non-European sites prevented geographic comparison
- No systematic TOAST classification data available
- Treating physicians not blinded (observational study)
- Sample size calculations based on noninferiority framework, but efficacy also analyzed
- Multiple testing without adjustment for Type I error
- Only patients with verified intake within 48h or ongoing prescription included
- Centers with restrictive IVT policies (<20% rate) showed no benefit of IVT
Citation
Stroke. 2025;56:2836–2845