PDF: BAOCHE
(2022)Objective
To assess whether endovascular thrombectomy plus medical therapy improves outcomes compared to medical therapy alone in patients with basilar-artery occlusion stroke presenting 6 to 24 hours after symptom onset.
Study Summary
• Trial was stopped early due to efficacy.
• Symptomatic intracranial hemorrhage and procedural complications were more frequent with thrombectomy.
Intervention
Multicenter, open-label, randomized controlled trial conducted in China. Patients were randomized to receive either thrombectomy plus standard medical therapy or standard therapy alone. Imaging eligibility required no large infarcts (PC-ASPECTS ≥6). Thrombectomy used Solitaire device within 6–24 hours of last known well.
Inclusion Criteria
Adults aged 18–80 with occlusion of the basilar artery or intracranial segments of both vertebral arteries, presenting 6–24 hours from symptom onset. Baseline NIHSS ≥10 initially required; amended to ≥6 after the first 61 patients due to slow recruitment. Required small infarct burden (PC-ASPECTS ≥6, Pons–Midbrain Index ≤2) and good pre-stroke function (mRS 0–1).
Study Design
Arms: Thrombectomy + Medical Therapy vs. Medical Therapy Alone
Patients per Arm: Thrombectomy: 110; Control: 107
Outcome
• Functional independence (mRS 0–2): 39% vs. 14%; adjusted rate ratio 2.75 (95% CI, 1.65–4.56)
• Revascularization: 88% successful (TICI ≥2b)
• Symptomatic ICH: 6% vs. 1%
• Mortality at 90 days: 31% (thrombectomy) vs. 42% (control); adjusted risk ratio 0.75 (95% CI, 0.54–1.04)
Bottom Line
Thrombectomy plus medical therapy resulted in good functional status (mRS 0-3) at 90 days in 46% of patients vs. 24% with medical therapy alone (adjusted rate ratio 1.81; 95% CI, 1.26-2.60; P<0.001; NNT ~5). The trial was stopped early for efficacy at a prespecified interim analysis of the first 212 enrolled patients who had completed 90-day follow-up (218 total underwent randomization; planned enrollment was 318). However, thrombectomy was associated with more procedural complications (11%) and a numerically higher rate of symptomatic intracranial hemorrhage (6% vs. 1%).
Major Points
- Primary outcome positive: mRS 0-3 at 90 days was 46% (thrombectomy) vs. 24% (control), adjusted rate ratio 1.81 (95% CI 1.26-2.60; P<0.001), NNT ~5.
- Original primary outcome (mRS 0-4) was null: 55% vs. 43% (adjusted rate ratio 1.21; 95% CI 0.95-1.54) — primary endpoint was changed from mRS 0-4 to mRS 0-3 mid-trial (before unblinding) based on external data from BEST and BASICS.
- Functional independence (mRS 0-2) nearly tripled: 39% thrombectomy vs. 14% control (adjusted rate ratio 2.75; 95% CI 1.65-4.56).
- Ordinal mRS shift favored thrombectomy: common odds ratio 2.64 (95% CI 1.54-4.50).
- Mortality was not significantly different: 31% thrombectomy vs. 42% control (adjusted risk ratio 0.75; 95% CI 0.54-1.04).
- High reperfusion rate: mTICI 2b/3 achieved in 88% of thrombectomy patients; basilar-artery patency at 24 hours was 92% vs. 19%.
- Extended time window validated: benefit consistent in 6-12h window (adjusted rate ratio 1.89) and >12-24h window (adjusted rate ratio 1.71).
- sICH (SITS-MOST) was 6% vs. 1% (risk ratio 5.18); sICH (ECASS II) was 9% vs. 2% (risk ratio 3.88). NNH for sICH ~20.
- Procedural complications in 11%: vessel dissection (4%), vessel perforation (3%), distal embolization (5%). Rescue angioplasty/stenting needed in 55% of patients, reflecting high atherosclerotic burden.
- Trial stopped early after prespecified interim analysis of the first 212 enrolled patients (who had completed 90-day follow-up) crossed the O'Brien-Fleming boundary (P<0.012 threshold).
Study Design
- Study Type
- Investigator-initiated, multicenter, open-label, randomized, controlled trial with blinded outcome evaluation (PROBE design)
- Randomization
- Yes
- Blinding
- Open-label treatment; blinded outcome assessment by local assessors unaware of treatment plus central assessors via video/audio recordings
- Sample Size
- 218
- Follow-up
- 90 days (primary); also 6 and 12 months for EQ-5D-3L
- Centers
- 30
- Countries
- China
Primary Outcome
Definition: mRS 0-3 at 90 days
| Control | Intervention | HR/OR | P-value |
|---|---|---|---|
| 26/107 (24%) | 51/110 (46%) | - (1.26-2.60) | <0.001 |
Limitations & Criticisms
- Primary outcome changed mid-trial from mRS 0-4 to mRS 0-3 (though before unblinding, based on external BEST/BASICS data). Original primary (mRS 0-4) was null.
- Open-label design, though outcome assessment was blinded.
- Limited to Han Chinese population — generalizability uncertain, especially for embolic vs atherothrombotic etiology.
- 101 of 319 excluded patients underwent endovascular therapy outside the trial — selection bias.
- Low IV thrombolysis rate (14-21%) — patients had to pay for the drug; less than two-thirds arriving within 4.5h received it.
- Early stopping for efficacy after interim analysis of 212 patients (218 randomized of 318 planned) may overestimate treatment effect.
- Original primary outcome (mRS 0-4) had CI crossing 1.0 (adjusted rate ratio 1.21, 0.95-1.54).
- Rescue intracranial angioplasty/stenting was needed in 55%, reflecting predominantly atherosclerotic etiology that may differ from Western populations.
Citation
N Engl J Med 2022;387:1373-84.