Scope Meta-Analysis (PFO PASCAL Outcomes)
(2026)Objective
To determine whether the PASCAL (PFO-Associated Stroke Causal Likelihood) classification system identifies patients who will experience net benefit vs net harm from PFO closure
Study Summary
• PASCAL probable/possible groups (85% of patients): PFO closure reduced strokes more than it caused AF — probable: −2.5% strokes vs +1.3% AF; possible: −3.4% strokes vs +1.1% AF
• PASCAL unlikely group (15%): PFO closure did NOT reduce strokes (+0.4%) but substantially increased AF (+4.6%) — net harm from closure
Intervention
Transcatheter PFO closure + antithrombotic therapy vs antithrombotic therapy alone, stratified by PASCAL classification
Inclusion Criteria
Adults aged 18-60 with PFO and otherwise cryptogenic ischemic stroke enrolled in 6 pivotal PFO closure RCTs
Study Design
Arms: PFO Closure (n=1889) vs Medical Therapy (n=1851), analyzed by PASCAL classification (Probable, Possible, Unlikely)
Patients per Arm: PFO Closure: 1,889, Medical Therapy: 1,851
Outcome
• Possible: −3.4% strokes (95% CI −5.4% to −1.3%) vs +1.1% AF — net benefit
• Unlikely: +0.4% strokes vs +4.6% AF (95% CI 0.3% to 8.9%) — net harm
Bottom Line
The PASCAL classification system effectively identifies 4 of 5 patients (probable + possible groups) who benefit from PFO closure and 1 of 5 (unlikely group) who experience net harm, providing a practical clinical tool for patient selection.
Major Points
- Individual participant-level meta-analysis of all 6 pivotal PFO closure RCTs (3,740 patients from SCOPE consortium)
- PASCAL probable group (37%): PFO closure reduced strokes by 2.5% while increasing late AF by only 1.3% — clear net benefit
- PASCAL possible group (48%): Stroke reduction of 3.4% outweighed AF increase of 1.1% — net benefit
- PASCAL unlikely group (15%): No stroke reduction (+0.4%) but substantial AF increase (+4.6%) — net harm from closure
- Provides first evidence-based patient selection tool for PFO closure decisions using a validated classification system
Study Design
- Study Type
- Individual participant-level data meta-analysis of 6 randomized controlled trials
- Randomization
- Yes
- Blinding
- Variable by included trial
- Sample Size
- 3740
- Follow-up
- 5 years
- Countries
- USA, Canada, Europe, Australia, Brazil, South Korea
Primary Outcome
Definition: Recurrent ischemic stroke (efficacy) and first-ever post-periprocedural AF (safety) over 5 years, stratified by PASCAL classification
| Control | Intervention | HR/OR | P-value |
|---|---|---|---|
| - | - | - | - |
Limitations & Criticisms
- Post-hoc secondary analysis of pooled trial data — PASCAL classification was not prespecified in original trials
- PASCAL classification applied retrospectively, not prospectively validated in a clinical decision pathway
- Different devices and antithrombotic regimens across 6 trials may confound results
- Follow-up limited to 5 years; longer-term effects of PFO closure and AF unknown
- Unlikely group (15%) is relatively small, limiting power for subgroup analysis
Citation
Saver JL et al. JAMA Neurol. 2026;83(3):242-249. DOI: 10.1001/jamaneurol.2025.5446