PDF: CASSISS
(2022)Objective
Stenting plus medical therapy vs medical therapy alone in patients with symptomatic severe intracranial atherosclerotic stenosis.
Study Summary
Intervention
Medical therapy plus stenting or medical therapy alone. Medical therapy included dual-antiplatelet therapy for 90 days (single antiplatelet therapy thereafter) and stroke risk factor control.
Inclusion Criteria
Patients with transient ischemic attack or nondisabling ischemic stroke (modified Rankin Scale score, 0-2) and severe stenosis (degree of stenosis: 70%-99%) of a major intracranial artery supplying the territory of the ischemic event. beyond a duration of 3 weeks from the latest ischemic symptom onset
Study Design
Arms: Medical therapy plus stenting vs medical therapy alone
Patients per Arm: Stenting: 176, Medical therapy alone: 182 (FAS)
Outcome
Bottom Line
Adding PTAS (Wingspan stent) to medical therapy did not reduce the 1-year risk of stroke or death compared with medical therapy alone (8.0% vs 7.2%; HR 1.10; 95% CI 0.52-2.35; P=0.82). No differences at 2 or 3 years. The 30-day periprocedural stroke/death rate was 5.1% (stenting) vs 2.2% (medical), and 3-year mortality was numerically higher with stenting (4.4% vs 1.3%; HR 3.75; P=0.08).
Major Points
- No benefit of stenting: primary composite (stroke/death ≤30 days or qualifying-territory stroke through 1 year) was 8.0% (stenting) vs 7.2% (medical), HR 1.10 (95% CI 0.52-2.35; P=0.82).
- Higher 30-day periprocedural risk: 5.1% (9/176) stenting vs 2.2% (4/181) medical — more than double.
- sICH only in stenting group: 4 patients (2.3%) within 30 days, 2 were fatal (guidewire perforation). 0% in medical group.
- No difference at 2 or 3 years: territory stroke at 3 years was 11.3% vs 11.2% (HR 1.00; P>.99).
- Numerically higher mortality with stenting: 3-year death 4.4% (7/160) vs 1.3% (2/159), HR 3.75 (95% CI 0.77-18.13; P=0.08).
- Refined selection did NOT help: excluded perforator strokes, required ≥3 weeks from event, credentialed high-volume sites (>30 cases/year, <15% complication rate) — stenting still did not outperform medical therapy.
- Much lower event rates than SAMMPRIS: 1-year primary 8.0%/7.2% vs SAMMPRIS 20%/12.2%, likely due to longer event-to-enrollment interval (median 35 vs 7 days) and exclusion of perforator strokes.
- Predominantly Chinese Han population (98%), mean age 56 years — much younger than SAMMPRIS (60) and VISSIT (62). Generalizability uncertain.
- Only Wingspan stent evaluated; drug-coated balloons and drug-eluting stents not tested.
- Large screened-to-enrolled ratio: 1,152 screened, 380 randomized (33%), 222 excluded for unfavorable anatomy.
Study Design
- Study Type
- Multicenter, open-label, randomized, outcome assessor-blinded trial
- Randomization
- Yes
- Blinding
- Open-label; independent outcome committee and imaging core laboratory blinded to treatment assignment. Simple 1:1 randomization without blocking or stratification via Interactive Voice Response System.
- Sample Size
- 358
- Follow-up
- 3 years (final follow-up November 10, 2019). Regular follow-up at 1 month, 1, 2, and 3 years.
- Centers
- 8
- Countries
- China
Primary Outcome
Definition: Composite of stroke or death within 30 days, or stroke in territory of qualifying artery beyond 30 days through 1 year
| Control | Intervention | HR/OR | P-value |
|---|---|---|---|
| 13/181 (7.2%) | 14/176 (8.0%) | 1.1 (0.52-2.35) | 0.82 |
Limitations & Criticisms
- Only Wingspan stent evaluated; drug-coated balloons, drug-eluting stents, and angioplasty alone not tested.
- Conducted only in Chinese centers — generalizability uncertain. Han ethnicity 98%.
- Enrolled 2014-2016 — management changes since then may limit applicability.
- Medical management may not have been as aggressive as SAMMPRIS (LDL target <100 mg/dL vs SAMMPRIS <70 mg/dL).
- Lower than expected event rates (7.2% medical vs assumed 18%) — study likely underpowered.
- Open-label design (mitigated by blinded adjudication).
- Nonnegligible periprocedural complications: 2.3% sICH, 2 fatal hemorrhages from guidewire perforation.
- Large screened-to-enrolled ratio (33%) with 222 excluded for anatomy — highly selected population.
- Numerically higher 3-year mortality with stenting (4.4% vs 1.3%) is concerning though not significant.
Citation
JAMA. 2022;328(6):534-542.