Meta-Analysis Provides Little Closure to PFO-Stroke Debate

 Download this article's Factoid (PDF & PPT for Gold Subscribers

A meta-analysis of randomized trials and observational studies of over 4,000 patients reveals slight benefit from percutaneous closure of patent foramen ovale (PFO) in preventing secondary stroke. However, according to the report published online June 21, 2013, ahead of print in Heart, PFO closure also was associated with an increased incidence of atrial fibrillation (A-fib).

Researchers led by Pascal Meier, MD, of The Heart Hospital, University College London Hospitals (London, United Kingdom) looked at 3 randomized trials and 11 controlled nonrandomized studies comparing percutaneous PFO closure with best medical therapy in 4,335 patients with cryptogenic stroke.

Randomized trials included CLOSURE I, PC, and RESPECT. Closure devices included:

  • Amplatzer PFO Occluder
  • PFO STAR
  • CardioSEAL/STARFlex
  • CardioStar
  • Amplatzer Cribriform Occluder
  • BioSTAR
  • Sideris Buttoned
  • Angel Wing
  • Amplatzer ASD
  • CardioSEAL
  • Rashkind occluder
  • ASDOS

Among the 3 randomized trials, the weighted incidence of recurrent stroke was 1.7% in the closure group and 2.9% in the best medical therapy group. For the nonrandomized trials, the rates were 0.7% and 6.9%, respectively.

Signal for Stroke Reduction

PFO closure showed a trend for stroke reduction in the randomized trials (RR 0.66; 95% CI 0.37-1.19; P = 0.171) and a reduction in stroke risk in the non-randomized trials (RR 0.37; 95% CI 0.20-0.67; P < 0.001).

In addition, a time-to-event analysis that considers the time to a recurrent stroke based on the 3 randomized trials and the 2 non-randomized trials that performed strict multivariate adjustments showed a borderline significant risk reduction with PFO closure (HR 0.58; 95% CI 0.33-0.99; P = 0.047).

There were no differences in risk of TIA, bleeding, or mortality between PFO closure and best medical therapy in the randomized or nonrandomized trials. However, the risk of A-fib was greater after PFO closure (table 1).

Table 1. PFO Closure and Risk of Events

 

RR

95% CI

P Value

TIA
RCTs
Non-RCTs

 
0.77
0.55

 
0.42-1.41
0.19-1.60

 
0.405
0.274

Bleeding
RCTs
Non-RCTs

 
1.17
1.09

 
0.47-2.88
0.07-15.90

 
0.738
0.950

Mortality
RCTs
Non-RCTs

 
0.65
0.76

 
0.23-1.84
0.31-1.88

 
0.458
0.557

A-fib
RCTs
Non-RCTsa

 
3.50
2.90

 
1.47-8.35
0.12-70.50

 
0.005
0.513

a A single observational study reported this outcome.

In a per-protocol analysis that only included patients who adhered to the protocol but excluded those who were, for example, assigned to medical therapy but later underwent PFO closure, there was no difference in stroke risk between the treatment groups (RR 0.66; 95% CI 0.32-1.38; P = 0.270). However, in an analysis excluding the CLOSURE I trial, which used a different closure device (StarFlex), the risk of stroke was significantly reduced (HR 0.44; 95% CI 0.20-0.94; P = 0.034).

The authors note that there was no significant effect of PFO closure on stroke reduction even in the pooled analysis of randomized trials, but nevertheless, signals toward a potential benefit of PFO closure were seen in the non-randomized data as well as in trials that used the Amplatzer closure device.

Regardless, they say, “The crucial question is still not resolved. Is a PFO really a source for stroke?” The data are conflicting, they continue, adding that “Unfortunately, this meta-analysis is not able to contribute to a better understanding of this issue.”

A-fib Defeats Purpose of Procedure

Additionally, the authors stress, “the procedure is associated with some risks such as bleeding, vascular injury, device embolization, thrombus formation on the device, and tamponade; it is also rather costly and, according to this analysis, it significantly increases the risk for atrial fibrillation. The latter is a major problem and defeats the purpose of the procedure in these cases.”

A-fib risk may depend on the type of closure device, Dr. Meier and colleagues observe, adding that A-fib incidence was much higher in CLOSURE I, which used the StarFlex device. “This device is not US Food and Drug Administration approved,” they state. “So far, the FDA has approved the CardioSEAL, Septal Occlusion System, and the Amplatzer PFO Occluder.”

Overall, the findings indicate that “it is too early for a final conclusion” regarding PFO closure and stroke prevention. While percutaneous closure in patients with cryptogenic stroke “does not appear to be superior to medical therapy . . . it [does appear] to be associated with an increased incidence of [A-fib] and has a certain risk for procedural complications,” the study authors note.

Occult A-fib Evaluation Advocated

In an e-mail communication with TCTMD, CLOSURE I principal investigator Anthony J. Furlan, MD, of Case Western Reserve University (Cleveland, Ohio), stressed that “patients should be evaluated for occult atrial fibrillation with 30-day event monitoring if necessary before any device is considered. The Amplatzer device seems less arrythmogenic, although A-fib can still occur.”

A-fib is definitely underestimated in PFO closure cases, he added.

In general, Dr. Furlan observed that cryptogenic stroke with PFO represents a non-homogeneous patient population, of whom only a small fraction have true paradoxical embolism. “The likelihood of paradoxical embolism is greatest in young patients under age 45 with no vascular risk factors, a substantial shunt, and a cortical infarct on baseline MR,” he said. “If your patient does not meet these criteria then medical therapy is preferable.”

Nevertheless, the best medical therapy has not been determined and will likely vary from “aspirin for many, but in others (eg, those with substantial likelihood of paradoxical embolism) antithrombotics may be superior,” Dr. Furlan said, noting that in terms of devices, “Amplatzer performed somewhat better than StarFlex, so device selection may also be a factor, but the major issue will be patient selection.”

Study Details

Roughly two-thirds of patients in the best medical therapy arm were on an antiplatelet therapy. CLOSURE I recommended either warfarin adjusted to INR 2-3 or aspirin 325 mg daily. RESPECT recommended either warfarin adjusted to INR 2-3, aspirin, clopidogrel or combination aspirin with dipyridamole. In the PC trial, antithrombotic treatment was left to the discretion of the treating physician and could have included antiplatelet therapy or oral anticoagulation, given that patients received at least 1 antithrombotic drug.

All 3 randomized trials suffered from slow enrollment. Each was also industry sponsored, “and several of the investigators received consulting fees from these companies,” the authors note, in addition to many of them being “PFO closure enthusiasts.”

 


Source:
Wolfrum M, Froehlich GM, Knapp G, et al. Stroke prevention by percutaneous closure of patent foramen ovale: a systematic review and meta-analysis. Heart. 2013;Epub ahead of print.

 

Disclosures:

  • Drs. Meier and Furlan report no relevant conflicts of interest.

 

Related Stories:

Jason R. Kahn, the former News Editor of TCTMD, worked at CRF for 11 years until his death in 2014…

Read Full Bio

Comments