CardioMEMS HF Implant Narrowly Passes FDA Advisory Hurdle

October 10, 2013

GAITHERSBURG, MD (updated) — A Food and Drug Administration (FDA) advisory committee narrowly voted in favor of a novel device that measures pulmonary-artery (PA) pressures in patients with heart failure. Despite the favorable vote, panel members weren't entirely convinced the PA-pressure data were responsible for reducing heart-failure hospitalizations but still gave the device a close thumbs-up on the overall risk/benefit question.

Overall, more members of the Circulatory System Devices advisory panel believed the benefits of the Champion HF Monitoring System PA-pressure monitor (CardioMEMS, Atlanta, GA), which would be the first permanent diagnostic cardiovascular implant, outweighed the risks when implanted in NYHA class 3 patients with a history of hospitalizations.

The panel was unanimously convinced the implant was safe but was not convinced of the effectiveness of the device. In fact, despite the 6-4 favorable vote on the question of risk vs benefit, a majority of panel members felt the company and clinical investigators did not provide sufficient evidence of effectiveness.

Asked whether there is "reasonable assurance that the CardioMEMS HF Pressure Management System is effective" in patients meeting criteria for use, seven panel members said no compared with just four members who said yes.

The vote was based largely on an open-label study conducted with patients enrolled in the CardioMEMS Heart Sensor Allows Monitoring of Pressure to Improve Outcomes in NYHA Class III Heart Failure Patients (CHAMPION) trial. In the original CHAMPION study, previously reported by heartwire , the PA-pressure sensor implanted in NYHA class 3 heart-failure patients with a history of decompensation reduced hospitalizations 30% compared with standard care.

In December 2011, an FDA advisory panel collectively decided CardioMEMS device wasn't quite ready for prime time. Although the panel's vote was again mixed—in fact, seven panel members said the device was effective compared with three who voted against—the vote was six to four against on the overall question of whether the benefits outweigh its risks.

Majority Feels Benefits Outweigh Risks

One of the concerns of the panelists in 2011 was a number of treatment recommendations sent by nurses, employed by the sponsor, to the treating physicians. The nurses would receive the PA pressures from patients in the treatment arm and would make recommendations based on these data. As a result, the impact of the nurse communications was believed to limit the interpretability of the data in terms of its effectiveness. Instead of conducting a new trial, the sponsor proposed additional follow-up of the CHAMPION patients to evaluate the effectiveness of the CardioMEMS device in the absence of nurse communications.

It was these ancillary analyses, particularly how to interpret the results of the open-access period (also known as part 2), that left some of the advisory committee members a little flat. Specifically, they were concerned about the potential for bias, given that the ancillary analyses were not defined a priori and that more than one-third of patients in CHAMPION were not included in the part 2 open-access study.

The open-access study showed that treatment with the CardioMEMS device could avoid 20 to 32 heart-failure hospitalizations per 100 subjects per year when it was used to guide medical therapy; therefore, the number needed-to-treat to prevent one rehospitalization was approximately three to five patients.

Dr Brent Blumenstein (Trial Architecture Consulting, Washington, DC), the statistician on the FDA panel, called the open-access data level 3 evidence and said that he simply did not know what to make of it. Similarly, when the FDA had its own members analyze the data, they too were concerned about the potential for bias given the nonrandomized nature in which patients left the trial.

In contrast, Dr Magnus Ohman (Duke University Medical Center, Durham, NC) said that while he didn't want to argue with a statistician, he felt the quality of evidence was better than level 3. He called the ancillary studies an "imperfect experiment" but told the panel that he believed there is an effect regardless of the way one cuts the data.

Dr Jeffrey Borer (State University of New York) said that while limitations of the ancillary analyses are valid, the data are largely consistent with the overall CHAMPION study. He said the limitations do not negate the validity of the data.

"Reducing heart-failure–related hospitalizations by 20 to 32 per 100 subjects per year, and I'll assume that number is correct, it might be off by a little bit, but that is phenomenal," said Borer. "That's really fantastic. That would be a tremendous benefit, and that benefit is cumulative. It's not just this year, it's again next year, and the year after that, and it's against a risk that is a one-time-only risk. I just don't see an issue here."

While Borer believes the clinical significance is high, other panel members were alarmed by the enormous number of alerts generated in the study and whether these were responsible for the improvement in clinical outcomes. Dr Valluvan Jeevanandam (University of Chicago, IL) said the device was effective in monitoring PA pressure but questioned whether it was effective in improving outcomes beyond hospital admissions. He would have liked to see an effect on heart function, such reducing the need for left-ventricular-assist devices or the need for transplant, or even improvements in six-minute-walk tests.

During the daylong discussion, there were also some concerns raised about the data with regard to women, especially whether or not the treatment effect existed in this subgroup. Borer even went as far to call for an additional study to address the safety and efficacy in this subgroup, given the apparent lack of decrease in heart-failure hospitalizations in females.

Still, despite the frequent back-and-forth about effectiveness and the merits of the ancillary analysis, the majority of panel members echoed the sentiment that they believed the CardioMEMS implanted device was an effective monitoring tool for physicians and that this additional signal could help reduce heart-failure hospitalizations.

Speaking during the panel meeting, electrophysiologist Dr Kristin Patton (University of Washington, Seattle) noted that it is very difficult to get patients to wear devices required in electrophysiology studies and wondered about patient motivation to take daily pressure and heart-rate readings. Dr Jay Yadav (Piedmont Heart Institute, Atlanta, GA), the chief executive officer and founder of CardioMEMS, said the patients were highly motivated and happy to be engaged in taking steps needed for their heart health.

Sponsor's Proposed Indications for Use

The CardioMEMS Champion HF Monitoring System is indicated for wirelessly measuring and monitoring PA pressure and heart rate in NYHA class 3 heart-failure patients who have been hospitalized for heart failure in the previous year. The hemodynamic data are used by physicians for heart-failure management and to reduce heart-failure hospitalizations.

The CardioMEMS Champion HF Monitoring System is used by the physician in the hospital or office setting to obtain and review PA-pressure measurements. The CardioMEMS Champion HF Monitoring System is used by the patient in the home or other remote location to wirelessly obtain and send hemodynamic and PA-pressure measurements to a secure database for review and evaluation by the patient's physician.

CardioMEMS sponsored the CHAMPION trial.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: