March 14th, 2012

The $800 Million Gamble: Jumping Aboard or Jumping the Gun?


In the CLOSURE I trial (Evaluation of the STARFlex Septal Closure System in Patients with a Stroke and/or Transient Ischemic Attack Due to a Presumed Paradoxical Embolism through a Patent Foramen Ovale), 909 patients with a PFO who had a stroke or a transient ischemic attack (TIA) of unclear etiology were randomly assigned to device closure of the PFO or to optimal medical therapy. The risks for stroke, TIA, and death (individually or combined) did not differ between the two groups.

2-year outcome

Device Closure

Medical Therapy

P value

Stroke or TIA












Death from neurologic cause





Furthermore, PFO closure increased the risks for a major vascular event (3.2% absolute increase, P<0.001) and for atrial fibrillation (5% absolute increase, P<0.001).

But here’s the really interesting thing.  

Patient enrollment was slow because of frequent off-label use of closure devices. Each study site recruited an average of only two patients per year, primarily because of decisions to close many PFOs outside the trial. During the 9-year study period, approximately 80,000 patients had a PFO closure with the use of a device. At an average cost of $10,000 per procedure, we spent ≈$800 million on a procedure that had no demonstrable benefit (and, as noted, increased risks).

In short, many cardiologists were convinced of the benefit of PFO closure devices before the study was performed.

In the future, should third-party payers withhold reimbursement for unproven device therapy unless such treatment is part of a randomized trial?


5 Responses to “The $800 Million Gamble: Jumping Aboard or Jumping the Gun?”

  1. Steven Greer, MD says:

    I think that the St Jude trial has a better chance or working (meeting endpoints). The TIA is not part of the composite endpoint. As long as the devices show good safety and biological plausibly mechanisms of benefit, and data trend in favor of reducing stroke, look for PFOs to be approved next year.

    We filmed Dr. Johnston of UCSF about is NEJM editorial viewable here

  2. Joel Wolkowicz, MDCM says:

    I think this country can easily afford $800 million for something that is useless.

  3. Steven Greer, MD says:

    The CLOSURE I trial was doomed to fail because it was run by a small company, when in fact this was one of the most challenging trials ever. Ample clinical evidence, case reports, etc all support the notion that PFO’s allow for embolic debris to cause stroke, and that closing the PFO helps. If the St Jude trial shows slight efficacy then the FDA will likely approve PFO closure for stroke prevention as an indication.

    This discussion by Dr. Johnston is quite useful

  4. Bruce Wells, MD says:

    Another example of use of an expensive device with absolutely no proof of efficacy.

    We need well-controlled studies with evidence of usefulness before any insurance company should have to pay for it.

  5. I agree entirely with Bruce Wells’ comments. Procedures should not be approved until there is evidence of benefit. With new technologies RCTs should precede widespread roll-out of technology.