Randomized Evaluation of Recurrent Stroke Comparing PFO Closure to Established Current Standard of Care Treatment - RESPECT


Numerous observational studies have suggested a beneficial effect of patent foramen ovale (PFO) closure on recurrent stroke in patients with cryptogenic stroke and PFO. However, the only randomized controlled trial on this topic, CLOSURE I, failed to show a benefit with PFO closure using the STARFlex device over routine medical management with anticoagulant/antiplatelet agents. The RESPECT trial sought to compare outcomes after PFO closure with the Amplatzer PFO Occluder over medical therapy in patients with cryptogenic stroke and evidence of a PFO.

Contribution to the Literature: The RESPECT trial showed that PFO closure with the Amplatzer PFO Occluder is not superior to medical management in reducing recurrent strokes in patients with presumed cryptogenic stroke and evidence of a PFO.

Study Design

  • Randomized
  • Blinded
  • Parallel

Patient Populations:

  • Patients ages 18-60 years
  • Documented PFO
  • Cyptogenic stroke within 270 days

    Number of enrollees: 980
    Duration of follow-up: 5 years (median)
    Mean patient age: 46 years
    Percentage female: 45%


  • Cerebral, cardiovascular, and systemic conditions that suggest other mechanisms for stroke
  • Contraindications to antiplatelet therapy
  • Anatomical contraindications to device placement
  • Any other reason to expect limited life expectancy, inability to attend follow-up visits, or inability to provide informed consent

Primary Endpoints:

  • Recurrence of a nonfatal ischemic stroke or fatal ischemic stroke or early post-randomization death (defined as all-cause mortality: device group–within 30 days after implant or 45 days after randomization; medical group–within 45 days after randomization)

Secondary Endpoints:

  • Complete closure of the defect demonstrated by transesophageal echocardiography and bubble study at the 6-month follow-up (Device Group)
  • Absence of recurrent symptomatic cryptogenic nonfatal stroke or cardiovascular death
  • Absence of transient ischemic attack

Drug/Procedures Used:

Patients with a documented PFO and evidence of cryptogenic stroke were randomized to either percutaneous closure with the Amplatzer PFO Occluder or medical management (aspirin 46.5%, warfarin 25.2%, clopidogrel 14%, aspirin + dipyridamole 8.1%, aspirin + clopidogrel 6.2%). Patients with PFO closure received aspirin 81-325 mg for 6 months and clopidogrel for 1 month.

Principal Findings:

This was an event-driven trial, and the trial was stopped when the total number of events was met. At this time, a total of 980 patients had been randomized at 69 centers in North America, 499 to PFO closure and 481 to medical management. Baseline characteristics were fairly similar between the two arms. Approximately 7% had diabetes, 13% were current smokers, 3% had evidence of coronary artery disease, and 23% had experienced another cerebrovascular accident event. Procedural success was achieved in 96.1% of the device-treated patients, with effective closure in 93.5% of the patients. Dropout rates were significantly higher in the medical therapy arm (18.7% vs. 9.6%).

Over a median duration of follow-up of 2.1 years, the primary endpoint (all recurrent strokes) was similar between the PFO closure and medical therapy arms (0.66/100 patient-years [PY] vs. 1.38/100 PY; hazard ratio [HR] 0.49, 95% confidence interval [CI] 0.22-1.11, p = 0.08). Event rates were nonsymmetric: 1.3% vs. 1.7% at 1 year, 1.6% vs. 3.0% at 2 years, and 2.2% vs. 6.4% at 5 years. On per-protocol analysis, there was a benefit favoring PFO closure (0.46/100 PY vs. 1.30/100 PY; HR 0.37, 95% CI 0.14-0.96, p = 0.03). On subgroup analysis, there were trends toward a benefit in patients with large shunts (p = 0.07), atrial septal aneurysms (p = 0.10), and those on concomitant antiplatelet medications (p = 0.19).

Atrial fibrillation was noted in 0.6% of patients in both arms. Other procedural complications were rare, including major bleeding (1.6%), major vascular complications (0.8%), and cardiac thrombus formation (0.4%). There were no device embolizations.

Longer-term follow-up: No difference was noted in all-cause strokes between PFO and medical management arms on extended follow-up (mean ~5 years) (p = 0.16). A blinded assessment was performed of the recurrent strokes, and one third of them were noted to be noncryptogenic (i.e., due to other mechanisms such as small-vessel disease, atrial fibrillation, atherosclerosis, etc.). There was a significant benefit in favor of PFO closure among patients with a recurrent cryptogenic stroke (HR 0.46, p = 0.042), in patients aged <60 years (HR 0.48, p = 0.035), and in those with an atrial septal aneurysm (HR 0.25, p = 0.007). All-cause mortality was 0.22% vs. 0.42%/100 PY.


The results of the RESPECT trial indicate that PFO closure with the Amplatzer PFO Occluder is not superior to medical management in reducing recurrent strokes in patients with presumed cryptogenic stroke and evidence of a PFO. There were, however, trends toward benefit on per-protocol analysis and in patients with large shunts and those with atrial septal aneuryms. On extended follow-up, the results were mostly similar. An interesting observation was that nearly one third of strokes thought to be cryptogenic likely had another underlying mechanism, and in patients with true cryptogenic strokes, PFO closure appeared to be superior compared with medical management.

Broadly speaking though, the overall negative results are consistent with those noted in the CLOSURE I trial (with the STARFlex device) and the PC trial (also with the Amplatzer device). These data are hypothesis-generating, but do not support routine PFO closure in patients with presumed cryptogenic stroke and PFO. It is possible that there might be a benefit in select subgroups. Results of ongoing trials including REDUCE and CLOSE are awaited.


Presented by Dr. John Carroll at the Transcatheter Cardiovascular Therapeutics meeting (TCT 2015), San Francisco, CA, October 15, 2015.

Carroll JD, Saver JL, Thaler DE, et al., on behalf of the RESPECT Investigators. Closure of patent foramen ovale versus medical therapy after cryptogenic stroke. N Engl J Med 2013;368:1092-100.

Presented by Dr. John Carroll at the Transcatheter Cardiovascular Therapeutics meeting (TCT 2012), Miami, FL, October 25, 2012.

Clinical Topics: Anticoagulation Management, Congenital Heart Disease and Pediatric Cardiology, Congenital Heart Disease, CHD & Pediatrics and Prevention, CHD & Pediatrics and Quality Improvement, Statins

Keywords: Coronary Artery Disease, Stroke, Follow-Up Studies, Platelet Aggregation Inhibitors, Warfarin, Ticlopidine, Foramen Ovale, Patent, Drug Combinations, Thrombosis, Confidence Intervals, Dipyridamole, Diabetes Mellitus, Septal Occluder Device, Transcatheter Cardiovascular Therapeutics

< Back to Listings