No Dent in Readmissions, Mortality Seen With Telemonitoring-Guided HF Care

November 17, 2010

November 17, 2010 (Chicago, Illinois) — In an age when seemingly everyone is electronically connected to everyone else, providing patients with heart failure a similar kind of connectivity with their providers seems like a natural. Regular monitoring for signs of clinical deterioration should allow clinicians to intervene sooner and more effectively, possibly preventing hospitalization or death. But sometimes great ideas that ought to work are hard to prove or simply don't pan out.

Two randomized trials of intervention in chronic heart failure guided by remote monitoring, presented here today at the American Heart Association 2010 Scientific Sessions, failed to show that the strategy improved care in ways that prolonged survival or kept the patients out of the hospital.

Dr Clyde W Yancy

"There was no benefit seen by either of these well-designed clinical trials on outcomes that are important to patients with heart failure," according to Dr Clyde W Yancy (Baylor University Medical Center, Dallas, TX), who spoke as the discussant after presentation of one of the trials.

"I think this is an important moment in our understanding of the contribution of this novel intervention in the overall management of heart failure," he said. "There is no need to parse the data any further." The weight of evidence now shows that telemonitoring doesn't improve on standard modern heart-failure disease management, even though the strategy had shown promise in meta-analyses and observational studies.

"The benefit of telemedicine that was presumed to be present has always been demonstrated to be less in the few randomized controlled trials than the cohort studies, and we've allowed hyperbole and excitement to guide our judgment rather than evidence."

Yancy continued, "We should hold all treatments to the same threshold of evidence, be they drugs, devices, or processes of care, and avoid an overexuberant rush to enticing but unproven technology."

Dr Lynne W Stevenson

Also as discussant for one of the presentations, Dr Lynne W Stevenson (Brigham and Women's Hospital, Boston, MA) contended that broad acceptance and implementation of telemonitoring systems, as they depend on providers to act on signaled changes in patients' clinical status, would demand so many trained professionals across the healthcare system as to be impractical. "We talk a lot about the number of patients needed to treat," she said. "I would propose that we need to talk, in addition, about the number needed to treat patients."

The alternative approach that she outlined in her talk, although not enough is probably yet known to carry it out, she noted, would actually diminish the need for clinician intervention.

Telemonitoring After HF Hospitalization

In one of the studies, the multicenter Telemonitoring to Improve Heart Failure Outcomes (Tele-HF) trial [1], published online November 16, 2010 in the New England Journal of Medicine to coincide with presentation at the meeting, patients with an HF hospitalization within the previous month provided answers to questions about their general health and weight and symptom status over the phone daily to an automated, interactive system (Tel-Assurance, Pharos Innovations). Every 30 days, they were queried about any symptoms of depression.

Dr Sarwat I Chaudhry

The reports were regularly reviewed for signs of clinical deterioration and possible need for intervention by nurses, who either responded independently or "obtained physician input as needed," reported lead author Dr Sarwat I Chaudhry (Yale University, New Haven, CT) here at the meeting. Medications were adjusted as needed.

Patients who didn't use the telemonitoring system for two consecutive days received an automated phone call to remind them, which if needed was followed up by direct encouragement from the program's staff, she said.

But six months of this kind of telemonitoring-guided management, compared with usual care, had no significant effect on readmission or death from any cause, all-cause mortality, days in the hospital, or other important end points, according to Chaudhry.

Clinical Outcomes at Six Months, Hazard Ratio (HR, 95% CI) for Telemonitoring-Guided HF Care vs Usual Care in Tele-HF

End point Telemonitoring, n=826 (%) Usual care, n=827 (%)x HR (95%CI) p
All-cause readmission/death* 52.3 51.5 1.04 (0.91–1.19) 0.75
All-cause readmission 49.3 47.4 1.06 (0.93–1.22) 0.45
All-cause mortality 11.1 11.4 0.97 (0.73–1.30) 0.88

*Primary end point

The study provided evidence that telemonitoring-based HF care can be hobbled by noncompliance issues, much like drug therapy itself. Of those assigned to the strategy, 14% never used it, Chaudhry noted.

Adherence (at least three calls per week) was best at the beginning, and it reached 90% the first week. But only 55% of patients were using it at least three times per week by the last week of the trial.

Adherence in Tele-HF most likely represents a best-case scenario for telemonitoring-based disease management, according to Chaudhry, as "we carefully selected practice sites and patients to maximize engagement."

Suggestions of Subgroup Benefits

In the other trial, conducted in Germany and called Telemedicine Interventional Monitoring in Heart Failure (TIM-HF) [2], a similar remote-monitoring strategy designed to sharpen conventional care of patients with NYHA class 2-3 heart failure (who had either an HF hospitalization within the previous two years or an LVEF <25%) had no effect on the primary end point of death from any cause over 24 months. Nor did it significantly influence secondary end points such as the composite of CV death or HF hospitalization, reported Dr Stefan Anker (Charité Medical School, Berlin, Germany) here at the meeting.

Hazard Ratio (95% CI) for 24-Month Outcomes, Telemonitoring-Guided Management vs Usual Care in TIM-HF

End point HR (95% CI) p
All-cause mortality* 0.97 (0.67–1.41) 0.87
CV death/HF hospitalization 0.89 (0.67–1.19) 0.44

*Primary end point

Dr Stefan Anker

The 354 patients in the telemonitoring group showed a significant (p<0.05) improvement in SF-36 physical-functioning score and a trend to less depression (p<0.06) compared with the 356 who received usual care, suggesting that the remote-monitoring strategy might well improve patients' quality of life, according to Anker.

Marginal benefits or trends suggesting possible survival gains from telemonitoring-guided management were observed in some patient subgroups, including those with prior decompensation without clinical depression or with an LVEF of at least 25%.

"Remote telemedical management," concluded Anker, retaining some optimism for the strategy, isn't for all patients with heart failure but could be used in clinically unstable patients as a "bridge to stability" and should be explored in further trials for effectiveness in specific patient subgroups.

Tightening the Circle of Management

An editorial accompanying the published Tele-HF report [3] notes that existing approaches for remote-monitoring–guided heart-failure management are based on the idea that close surveillance of markers of clinical deterioration facilitates timely intervention, with the goal of improving clinical outcomes.

"An effective home-monitoring strategy must therefore contain all of the elements necessary to complete the circle of heart-failure management," write lead author Dr Akshay S Desai (Brigham and Women's Hospital) and Stevenson, the senior author. Those elements include: measuring markers of early decompensation "while it can still be averted," timely and efficient transmission of the information, reception of the data by persons qualified to recommend appropriate intervention, and correct implementation of the intervention by the patient.

As discussant for the Tele-HF presentation, Stevenson observed that "it's not clear in this trial how long it took for the alerts to be received and how long it took for a decision to be made and communicated back to the patient."

Also important, if telemonitoring were to be mandated as the standard of care in heart failure, would be the likely monumental demand for such "persons qualified to recommend appropriate intervention," according to Stevenson.

She estimated that there would be 26 000 calls for a management response every year for each provider with an average caseload of 200 patients, based on their frequency in Tele-HF and assuming 90% patient compliance with the system.

"This would be about 100 phone calls every working day, for telemonitoring alone, for a person carrying an average caseload. Yikes. This doesn't even include time for that person to provide education, medication refills, [and handle] the expected unexpected and, of course, the unexpected unexpected."

More efficient, she said, would be to have patients self-manage their medications guided by some kind of physiologic marker. "This could clearly be achieved if we could get the right measurement. It has certainly been achieved for diabetic management. We never thought it would be possible, but most of our patients several times a day actually adjust their medications according to their own glucose. We probably don't yet have the right measurements and have not designed the right algorithms to allow this type of information to be translated for helping us maintain fluid balance."

Yancy reports that he has had no consultancies or speakerships or has received relevant honoraria since June 2008 at the latest. Chaudhry said she has no disclosures. Disclosures for the other Tele-HF authors and Desai and Stevenson are provided online in the journal; Stevenson herself reports that she consulted for Medtronic and Cardiomems prior to December 2009. With regard to TIM-HF, Anker reported that "the technology development as well as the clinical trial was funded in a public-private partnership through a research grant of the German Federal Ministry of Economics & Technology" and by Robert-Bosch Healthcare, InterComponentWare, and Aipermon; he has consulted for Robert Bosch Healthcare, Thermo Fisher Scientific, and St Jude Medical and received honoraria for speaking from Thermo Fisher Scientific and St Jude Medical.