GAITHERSBURG, MD. – An implantable wireless device that provides measurements of pulmonary arterial pressure and heart rate in patients with heart failure was supported for approval by a majority of a Food and Drug Administration advisory panel, almost 2 years after it was rejected for approval.
The Champion Heart Failure Monitoring System, a permanent implantable pressure-measurement system, includes a wireless sensor that is implanted in a small branch of the pulmonary artery during a right-heart cardiac catheterization, providing wireless PA measurements.
At a meeting of the FDA’s Circulatory System Devices Panel, the panel voted 6-4, with 1 abstention, that the benefits of the device outweighed its risks for monitoring patients who meet the criteria in the proposed indication, which is for wirelessly measuring and monitoring PA pressures and heart rate in patients with New York Heart Association (NYHA) class III heart failure who have been hospitalized for HF in the previous year.
The proposed indication includes the statement that the hemodynamic data "are used by physicians for heart failure management and to reduce heart failure hospitalizations" in hospital or office settings, and that it can also be used by patients at home or another location "to wirelessly obtain and send hemodynamic and PA pressure measurements" to a database to be reviewed and evaluated by their physicians. The manufacturer is CardioMEMS.
The panel was divided regarding the strength of the data supporting the effect the device had on reducing HF hospitalizations, the primary endpoint of studies, and this was reflected in the vote on the device’s effectiveness. The panel voted 7-4 that there was not reasonable assurance that the device was effective for use for the proposed indication. But they unanimously voted that there was reasonable assurance it was safe, based on the finding that most complications occurred in the 30 days after implantation.
However, no benefit was seen on HF hospitalizations in women, which panelists thought may have been due to the low number of women enrolled in the studies, and panelists recommended that the device needed to be studied further in women. Another recommendation was to drop the specific HF class in the indication because patients with HF tend to move from one class to the other.
The original study enrolled 550 patients with NYHA class III HF and a recent hospitalization for HF. All patients had the device implanted, but physicians had access to daily PA measurements for those randomized to the treatment group, and were able to adjust HF medications based on the values provided.
At 6 months, there were 84 HF hospitalizations in the treatment group (0.32 per patient), compared with 120 (0.44 per patient) in the control group, a highly significant difference that represented a 28% reduction in the rate of HF hospitalizations. In addition, 5.6% of the patients in the treatment group died, versus 7.1% in the control group, a nonsignificant 23% reduction in the relative risk of death.
In December 2011, the FDA’s Circulatory System Devices Panel met for the first time to review the device, and voted 6-4 that the risks outweighed the benefits. Like the FDA reviewers, the panel raised concerns that the nurse communications in this study may have influenced treatment decisions and biased the results, and the FDA did not approve the device.
In response, the company did not conduct another trial but conducted various analyses of HF hospitalizations and deaths during the open-access period of the study, which eliminated nurse input. Patients in the control and treatment groups in the randomized study were included in the study’s open-access period, where the PA pressures on all patients were provided to physicians.
Among the results was that HF hospitalization rates among those patients who were controls in the first part of the study dropped to a rate comparable to the rate among those who were in the original treatment group, when access to the PA measurements of the original controls were available. A third-party analysis showed that the possible effect the nurses’ input had on the outcomes in the randomized study was minimal, according to the company.
On the basis of these analyses, the company concluded that the effect on HF hospitalizations was a result of physicians knowing the PA measurements provided by the device, and was not related to the communications from nurses during the randomized segment of the study or during the open-access period.
Panelist Dr. Jeffrey Borer, chief of the division of cardiovascular medicine at the State University of New York Downstate Medical Center, Brooklyn, who voted yes on all three questions, said he found the ancillary studies very helpful in clarifying the utility of the device seen in the original study, and that this was "an approvable and probably very useful device." He, like other panelists, remained concerned about the use in women, noting that there was no benefit on HF hospitalizations in women who had the device implanted, but this may have been due to the low number of women enrolled in the study.