JOHN YANG: Americans rely on the Food and Drug Administration to make sure that medical devices are safe.

Now, reporting by KFF Health News raises questions about how the agency has dealt with some products, including one that treats a heart condition.

Ali Rogin has more.

ALI ROGIN: The device is called MitraClip, and its goal is to fix leaky heart valves, often in patients too sick for surgery, itús implanted by snaking a small clip into the heart through a major vein.

Now, a new investigation by KFF Health News looks into questions surrounding the deviceús development, promotion and use since it was approved by the FDA back in 2013.

David Hilzenrathis the senior and Washington correspondent for KFF, and was the lead author of this story.

David, thank you so much for being here.

What type of patient was this device developed for and why was it such a game changer?

DAVID HILZENRATH, Senior Correspondent, KFF News: MitraClip is used to treat patients with a condition called mitral regurgitation, or MR, in which blood flows backwards through the heartús mitral valve.

MR can lead to heart failure and death.

Iúve seen estimates Ali that millions of people in the United States experience MR including 10 percent of people over 75.

MitraClip system includes a tiny clip that is implanted in the mitral valve to fasten the two flaps of the mitral valve together and enable them to achieve a tighter seal.

ALI ROGIN: And what was then the reality of what you found about how what this product says it does matches up with the reality?

DAVID HILZENRATH: For many patients, it may be a game changer.

We found that the story of MitraClip is in many ways, a cautionary tale, however, about the science, the business and the regulation of medical devices.

We found that since the FDA approved MitraClip in 2013, MitraClip has been named in thousands of reports to the FDA about malfunctions or cases in which it might have caused or contributed to a patientús injury or death.

We found 17,000 reports, of which more than 1,100 involved deaths.

There has long been debate about research surrounding MitraClip.

Before the FDA approved the device in 2013, it issued a paper criticizing data marshaled in support of MitraClip, and one outside advisor to the agency compared that data to poop, more specifically, horse poop and dog poop.

ALI ROGIN: Now in terms of the deaths that you mentioned, certainly the people associated with MitraClip would point out that this was not a physically well population to begin with, so some deaths are to be expected.

But is the issue here that the claims are that these devices actually cause the deaths.

DAVID HILZENRATH: The reports that I described do not prove causality.

Itús one of the flaws in the FDA post market surveillance system that these reports are not definitive.

The FDA relies largely on reports submitted by the manufacturers themselves, such as, in this case, Abbott.

ALI ROGIN: And as you mentioned, Abbott denies wrongdoing.

They say in part that, quote, it is a safe and effective treatment option with a strong foundation of clinical evidence for more than 20 clinical trials.

I do want to ask you, though about the clinical trial aspect of this.

It does seem like a Catch 22 that the very companies that are charged with proving to the FDA that their products are safe and effective, largely bankroll a lot of these clinical trials because theyúre extremely expensive and theyúre the only ones that can do so.

Is that an inherent conflict of interest in just how the system is run in the United States?

DAVID HILZENRATH: Itús built into the system.

What you describe Ali is absolutely true.

The FDA relies primarily on clinical trials conducted by the very companies that manufacture these products.

Much of the research that Abbott alluded to in that statement was funded by Abbott.

Many of the doctors involved themselves had financial relationships with Abbott, and the research is not always clear cut.

It can be debated.

ALI ROGIN: This investigation also, though, reveals the amount of uncertainty that is allowed for in the FDA approval process.

Is that also something that is pretty standard when FDA approval is sought?

DAVID HILZENRATH: Well at times, and weúve seen this with other products.

The FDA has approved medical products over the objections of its own staff professionals.

In the case of MitraClip, the FDA initially argued that the clinical research and other data the manufacturer had produced to argue for approval was fundamentally flawed.

To cite just one example, the allegation was that the MitraClip patients in that testing were treated by surgeons who were highly experienced, while those in the control group were treated by doctors who performed surgery much less frequently.

In that testing, MitraClip was compared to surgery, and the FDA convened this committee of outside advisers.

The committee was quite split on what to recommend to the FDA.

The committee voted against the proposition that the device was proven effective.

However, it voted that the benefits outweighed the risks.

Ultimately, the FDA approved it for a somewhat narrower universe of patients than originally proposed.

ALI ROGIN: I should note that the FDA has a statement that says that the agency believes that the overall benefits of this device continue to outweigh the risks in these patients, and that Abbottús recall strategy was appropriate and adequate.

Much left to discuss, but we have to leave it there.

David Hilzenrath with KFF, thank you so much for coming in.

DAVID HILZENRATH: Thank you, Ali.

Great to be here.