Halting the Revolving Door: Could Remote Monitoring Reduce Heart Failure Rehospitalizations?

Remote monitoring has emerged as a possible strategy for stemming the costly pattern of repeat hospitalization common among heart failure patients.

Revolving hospital door

Nearly one million hospitalizations for heart failure occur in the United States each year. Unplanned readmissions contribute to these hospitalizations and to the country’s approximately $37.2 billion price tag paid annually on heart failure care. Heart failure rehospitalization rates have continued to rise, approaching 30 percent within 30–90 days post-discharge (J Am Coll Cardiol 2013;61[4]:391-403).

The problem with heart failure is not a lack of best practices guidelines. In an editorial on preventing heart failure hospitalizations, Liviu Klein, MD, MS, of the University of California, San Francisco, noted that the heart failure rehospitalization rate is close to 50 percent over six months despite clinicians’ efforts to manage patients in heart failure clinics, follow-up early after discharge, collect daily weights and stay in frequent contact with patients (JACC Heart Fail 2016;4[5]:345-7).

Neither is there a shortage of incentives for clinicians to halt heart failure’s revolving hospital door. The Hospital Readmissions Reduction Program (HRRP), which was launched in 2012 under the Affordable Care Act, levied penalties totaling $517 million in 2013 and 2014 for above-national-average all-cause readmissions within 30 days of discharge following treatment for heart failure, myocardial infarction or pneumonia, according to a Kaiser Family Foundation analysis.

A major challenge in treating heart failure has been that clinicians often don’t learn that a patient’s condition has worsened until symptoms that require urgent treatment in the hospital—such as shortness of breath or congestion causing the ankles, feet, legs and abdomen to swell—have returned. If cardiologists were able to catch congestion earlier, then they might be able to make medication changes and stave off both the dangerous symptoms and the rehospitalizations.

This is where remote monitoring could play a role, experts say, even though most of the studies reported so far have yielded disappointing results. 

Try, try again 

“In three large trials, telemonitoring did not lead to reductions in hospitalization,” says Gregg Fonarow, MD, director of the Ahmanson-UCLA Cardiomyopathy Center at the University of California-Los Angeles. The trials he’s referencing: Tele-HF (Telemonitoring to Improve Heart Failure Outcomes), which in 2010 failed to improve readmissions or reduce all-cause mortality in heart failure patients despite having monitored their general health and heart failure symptoms (N Engl J Med 2010;363[24]:2301-9); TIM-HF (Telemedical Interventional Monitoring in Heart Failure), which tracked daily weights and vital signs but  also failed to decrease all-cause mortality (Circulation 2011;123:1873-80); and BEAT-HF (Better Effectiveness After Transition-Heart Failure), where daily collection of blood pressure, heart rate, weight and symptom data did not improve 180-day readmissions (JAMA Intern Med 2016;174[3]:310-8).

A theme of these trials is that most of the data captured by remotely monitoring heart failure patients has not been “highly actionable,” says Fonarow, who was a BEAT-HF investigator. For example, he explains, “changes in daily weight are neither sensitive nor very specific for [detecting] true congestion that would lead to hospitalization.”

Arrival of a champion?

The ability to deliver information that clinicians can use to take preemptive action to halt the heart failure cycle of hospital admission–discharge–symptomatic congestion–readmission is what differentiated CHAMPION (CardioMEMS [St. Jude Medical] Heart Sensor Allows Monitoring of Pressure to Improve Outcomes in Class III Heart Failure) from the other heart failure monitoring trials. In CHAMPION, participants who had been rehospitalized for heart failure in the previous 12 months were implanted with a CardioMEMS pressure sensor, which collects data on pulmonary artery pressure and heart rate. The patients were then randomized to be remotely monitored and receive the standard of care or to receive only standard care. At six months, there was a 28 percent decrease in heart failure hospitalization rates for the patients whose treatment was guided by the data obtained with the monitoring device compared to the control patients (Lancet 2011;377[9766]:658-66). The monitored patients also had their medications changed twice as many times as the control group (JACC Heart Fail 2016;4[5]:333-44).

The intervention worked because it captured PA pressure, says Klein, who was a CHAMPION investigator and is conducting a follow-up registry study requested by the FDA to determine whether the benefits seen in CHAMPION can be achieved in real-world settings. “The [CHAMPION] study linked decreases in elevated pulmonary artery pressures to decreased rates of heart failure hospitalization,” he says. “This was done through more regular adjustment of medications like diuretics and vasodilators.”

Some cardiologists have questioned the cost-effectiveness of the CardioMEMS sensor, noting that the $17,000 device must be implanted during a right heart catheterization, while others remain unconvinced of its effectiveness. “Important uncertainties remain about the ability of the CardioMEMS device to reduce [heart failure]-related hospitalizations, improve quality of life, and decrease mortality,” wrote Sanket S. Dhruva, MD, and Harlan M. Krumholz, MD, SM, of the Yale School of Medicine in New Haven, Conn., in an editorial published this spring in JACC Heart Failure (2016;4[5]:376-9).

Pros & cons of proxy data

Some clinicians are on the lookout for a monitoring option that could deliver early signals of congestion without the risks or costs of an invasive strategy.

It’s possible that measuring lung impedance could serve as a proxy for clinical congestion, says J. Thomas Heywood, MD, a cardiologist at Scripps Clinic in La Jolla, Calif., who was asked to comment on IMPEDANCE-HF (Non-invasive Lung IMPEDANCE-Guided Preemptive Treatment in Chronic Heart Failure Patients), which was presented at the 2016 American College of Cardiology scientific sessions. A small, proof-of-concept trial, IMPEDANCE-HF found that measuring lung impedance-guided treatment was superior to routine management in reducing heart failure hospitalizations at one year (J Card Fail, online Apr 4, 2016).

“Doctors have not embraced [lung impedance] because there is perceived to be a lot of false positives,” Heywood explains, emphasizing that for surrogate measurements to work in remote monitoring, clinicians “have to be smarter in responding. You cannot respond as reflexively as you can when you are dealing with something that is a direct relationship to the patient’s disease,” such as PA pressure.

The value of such noninvasive remote monitoring approaches and their surrogate markers may be in their negative predictive value, Heywood adds. “If you did not see things like change in impedance or change in heart rate variability, then you would not need to see those patients back in hospital,” he says.

[[{"fid":"22837","view_mode":"media_original","type":"media","attributes":{"height":467,"width":424,"style":"font-size: 13.008px; width: 424px; height: 467px; float: left; margin: 5px;","alt":" - readmissions-glance","class":"media-element file-media-original"}}]]

Recipe for data overload?

Another challenge with remotely monitoring heart failure patients is incorporating the task into the daily workflow without overwhelming the clinical team.

“If you increase the frequency of data collection, you increase the amount of data that the clinical team needs to review,” says Peter Eckman MD, heart failure section head at the Minneapolis Heart Institute and lead investigator for the SMILE (Sensible Medical Innovations Lung Fluid Status Monitor Allows Reducing Readmission Rate of Heart Failure Patients) study, which is testing an experimental technology that uses a low-power electromagnetic energy to measure fluid in the lungs.

Ekcman says the sweet spot—where data are captured in a timely fashion without inundating the team—is still elusive, but it might be enough to record data once every day. That frequency of data capture would provide the clinical team with sufficient time to determine whether a patient is decompensating and make treatment modifications without creating information overload.

Biykem Bozkurt MD, PhD, director of the Winters Center for Heart Failure Research at Baylor College of Medicine in Houston, notes that once-daily data collection proved effective in the CHAMPION trial, whereas investigators in the IMPEDANCE-HF measured lung impedance at each monthly outpatient visit. Questions remain, she says, about who would monitor CardioMEMS transmissions and whether an expanded workforce would be necessary to scale the technology into community settings.

Integrating remote monitoring and responding to the data generated will ultimately be the responsibility of nurses and nurse practitioners, predicts Lisa Rathman, MSN, CRNP, CHFN, CCRN, president-elect of the American Association of Heart Failure Nurses and a heart failure nurse practitioner at Lancaster General Hospital in Pennsylvania.

“It is a workflow issue in many places,” says Rathman. The questions to focus on are “how we organize to look at the data and how often should we look at it? That is not standardized.”