Success in Heart Failure: Keeping Patients Free of Hospitals

Some implantable cardiac devices capture data that may indicate worsening health in patients with heart failure. Cardiologists are beginning to use that information to identify at-risk patients and intervene before their conditions deteriorate and require hospitalization.

Proactive patient care

More than a million people are hospitalized for heart failure in the U.S. annually at a cost of almost $31 billion, according to the American Heart Association (AHA). Readmissions are high, too: Among Medicare beneficiaries alone, 134,500 patients were readmitted within 30 days for congestive heart failure in 2011, for a bill of $1.7 billion, the Agency for Healthcare Research and Quality calculated. The AHA projected that the cost of caring for patients with heart failure could reach $69.7 billion by 2030.

All these dollar signs prompted the Centers for Medicare & Medicaid Services to make heart failure one of three initial conditions that it would target in a program that now withholds up to 3 percent of Medicare reimbursement to hospitals with higher than expected 30-day readmission rates. That gives hospitals an even greater incentive to monitor heart failure patients for early signs of trouble. The simplest method asks patients to weigh themselves daily to detect signs of fluid retention and report back to physicians.

“If their weight goes up, they call their doctor and take more diuretics,” says E. Kevin Heist, MD, PhD, a cardiac electrophysiologist at Massachusetts General Hospital in Boston. But the approach is often hit or miss. “It is a sensitivity/specificity issue. Either you get so many signals that you stop listening or it is not sensitive enough and lots of people get sick and come in before they have this signal.”

Rather than rely on such blunt instruments, physicians now have an opportunity to get to the heart of the matter, literally. Using implantable devices, they hope to spot signs of early decompensation remotely and intervene before the patient’s condition worsens and he or she needs hospitalization. Two different options have emerged as contenders: standalone devices that are prognostic but not therapeutic and therapeutic devices that could do double duty to predict risk.

“Features embedded in cardiac resynchronization devices [CRT-Ds] and implantable cardioverter-defibrillators [ICDs] have proven to be of some value, particularly in predicting risk,” says William T. Abraham, MD, of the Ohio State University Heart and Vascular Center in Columbus. In one recent study, Abraham and other researchers showed they could stratify patients at risk of heart failure hospitalization using diagnostic notifications in CRT-Ds with defibrillation capability (Heart Lung 2015;44[2]:129-136). “They kind of tell us which patients to worry about more and less but they don’t provide data that are actionable on a day-to-day or week-to-week basis.”

Heist also explored the use of implanted cardiac devices to detect pulmonary congestion in the pivotal study DEFEAT-PE (Detect Fluid Early from Intrathoracic Impedance Monitoring). The study used an algorithm in CRT-Ds and ICDs by St. Jude Medical to monitor changes in intrathoracic impedance, which could be a way to detect pulmonary congestion. “The idea is that as the lung, which is mostly filled with air, begins to fill with fluid, the impedance will drop,” Heist says. That change possibly could be a predictor of heart failure hospitalization.

They concluded that while the algorithm predicted pulmonary congestion events, it failed to find the sweet spot: a low false positive rate along with high sensitivity (Am J Cardiol 2014;114:1249-1256). “It was considered not sufficiently predictive for it to create a red flag that would alert the physician and lead to a change in therapy,” Heist says.

Raising red flags early

After an earlier rejection, the FDA approved CardioMEMS (St. Jude Medical), a miniaturized implantable device for managing heart failure patients, in 2014 based on additional analyses of the CHAMPION trial (CardioMEMS Heart Sensor Allows Monitoring of Pressure to Improve Outcomes in NYHA Class III Heart Failure Patients). The 550 trial participants all had been rehospitalized for heart failure in the previous 12 months and after enrollment implanted with the CardioMEMS pressure sensor. They then were randomized to be remotely monitored and receive the standard of care or receive standard of care alone.

Physicians in the CardioMEMS group followed a medical treatment strategy based on information relayed through the pressure system. At six months, the rate of hospitalization was 32 percent with CardioMEMS vs. 44 percent with standard of care alone—a 37 percent reduction in heart failure readmissions (Lancet online Feb. 10, 2011).

“We now get a direct measurement of pulmonary artery pressure on a day-to-day basis in the ambulatory setting,” says Abraham, CHAMPION’s principal investigator. “We have actionable data. We can use that information as often as needed, on a day-to-day or week-to-week or month-to-month basis, whenever as things change, to make adjustments in medications, that, based on the CHAMPION trial, have been shown to reduce the risk of hospitalization.”

One of the advantages of the pressure sensor approach, according to Abraham, is that the device can be implanted in a broad patient population. The trial enrolled patients with New York Heart Association Class III heart failure, with no restrictions on ejection fraction or whether they had an existing CRT-D or ICD. In a recently published prespecified subanalysis of CHAMPION, Abraham and his team dug into data on patients with left ventricular ejection fractions of 40 percent or greater (preserved ejection fraction), a cutoff that is consistent with guidelines, as well as patients with ejection fractions of 50 percent and greater (Circ Heart Fail 2014;7:935-944).

The rehospitalization rate for patients with a preserved ejection fraction managed with CardioMEMS was 50 percent lower than the control group of patients who received standard of care at 17.6-month follow-up. They reported similar results for patients with ejection fractions of 50 percent and greater at 18 months. Abraham points out that today’s heart failure guidelines for systolic heart failure are evidence-based while they offer consensus recommendations for diastolic heart failure. Their results, which are statistically significant and from a randomized, controlled clinical trial, meet the higher standard. Abraham hopes this new data will be incorporated into the next set of heart failure guidelines.

“To date, every subgroup we looked at in the CHAMPION trial benefitted from the use of the [CardioMEMS] heart failure system,” he says. “Systolic heart failure, diastolic heart failure, chronic kidney disease, no chronic kidney disease, pulmonary hypertension and no pulmonary hypertension, and so on. It really looks like the information CardioMEMS provides is applicable across a broad spectrum of patients who fit that FDA labeled indication.”

Heist agrees that the pressure sensor approach offers more predictive power than thoracic impedance but in patients who already have an implanted device it also requires another implantation. Any procedure, even a minimally invasive one, holds some potential risk, although Abraham points out that all-cause as well as heart-failure-related readmissions were lower with CardioMEMS. “What that tells you is there is no trade-off,” Abraham says.

ICDs and CRT-Ds alone may one day meet the required predictive standards, but it likely will take more than a thoracic impedance algorithm to get there, Heist predicts. Instead, thoracic impedance may be one of several features that can be bundled together in a device to provide both therapy and provide information physicians can use to keep patients out of the hospital.

Given the high burden heart failure places on patients and healthcare systems, any process that helps cardiologists detect early warning signs and act on them is welcome. It is “the Holy Grail, so to speak,” Heist says.

Making Atrial Fibrillation Less Elusive

There is no shortage of devices for monitoring atrial fibrillation but their effectiveness hinges in part on the type of disease—paroxysmal, persistent or permanent—and the duration of monitoring. Here is a look at a few of the outpatient options.

Holter monitors
A portable electrocardiogram monitor, the Holter monitor is a mainstay for remotely monitoring patients for atrial fibrillation. But its diagnostic yield is variable and some patients find the monitors cumbersome. 

Wireless telemetry devices
These devices include sensors and remote communication capabilities to capture data on arrhythmia and relay it to physicians.

Insertable cardiac monitors
These devices detect and record atrial fibrillation while companion networks remotely transmit the data. They allow long-term monitoring. 

Patches
The Zio patch, from iRhythm Technologies, is a lighter-weight alternative to a Holter monitor that can be worn for 14 days.  

Apps
The AliveCor app can be downloaded to a smart phone and used to detect the presence of atrial fibrillation. Certain apps can record, store and transfer electrocardiogram rhythms.