This article is more than 5 years old. Information may no longer be current.
MultiSENSE: Diagnostic service predicts future events in patients with HF
Click Here to Manage Email Alerts
NEW ORLEANS — A novel diagnostic tool was successful at predicting decompensation events in patients with HF, according to findings from the MultiSENSE study presented at the American Heart Association Scientific Sessions.
In the study, 900 patients with HF (mean age, 66 years; 72% men) had the diagnostic tool, consisting of sensors that collect physiological data (HeartLogic Heart Failure Diagnostic Service, Boston Scientific), activated in their existing cardiac resynchronization therapy defibrillator, John P. Boehmer, MD, from Penn State Hershey Medical Center in Hershey, Pennsylvania, said at a press conference.
The system contains an algorithm to detect signs of worsening HF, based on indicators such as heart sounds, respiratory rate, thoracic impedance, heart rate and physical activity, researcher Jagmeet P. Singh, MD, PhD, associate chief, cardiology division, Massachusetts General Hospital and Harvard Medical School, told Cardiology Today.
“The propensity of patients with HF to be readmitted is almost eight- to ninefold higher,” said Singh, a member of the Cardiology Today Editorial Board. “The [financial] drain will be potentially as much as $100 billion by 2030. The study centered on developing a composite risk strategy from within the [CRT-D] and then testing it.”
All patients were aged at least 18 years, were NYHA class II to IV within the prior year and previously implanted with a specific CRT-D system (Cognis, Boston Scientific).
After the diagnostic system was downloaded to the CRT-D, each patient was evaluated with it for 12 months, with data retrieved during a visit every 6 to 8 weeks or transmitted remotely every 3 weeks. At 12 months, the diagnostic software was disabled and the device was turned back into a conventional CRT-D.
Each patient’s data were used either for the development set (n = 500), which developed the algorithm, or the test set (n = 400), which validated alert algorithm performance. Assignment was sequential, not randomized.
“The composite score integrated information from all the variables to communicate whether the patient was at a high risk or a low risk for an event,” Singh told Cardiology Today. “In the test cohort, it was tested whether this was sensitive enough to predict HF events, whether there were unexpected alerts ... and how much lead time did [an alert] give you before the patient got rehospitalized for HF.”
An independent clinical events committee adjudicated HF events, true positive alerts, HF-related alerts and unexplained alerts.
The endpoints of interest were sensitivity for detecting usable HF events > 40% and an unexplained alert rate of less than 2 per patient per year.
Sensitivity for detecting usable HF events was 70% (95% CI, 55.4-82.1) and the unexplained event rate was 1.47 per patient per year (95% CI, 1.32-1.65), so both primary endpoints were met, Boehmer said.
Median time from alert onset to HF event was 34 days, according to the researchers.
“The study gives us an idea that we can use a composite score from within the device using measured variables to risk-stratify the patients and potentially in the future prevent HF readmissions,” Singh said in an interview. “What we haven’t done is test this prospectively to see if we can prevent HF hospitalizations, and [determine] how it may impact the use of various therapeutic interventions. That will probably be the next step.”
Unlike an existing technology to predict HF events through pulmonary artery pressure monitoring (CardioMEMS, St. Jude Medical), the novel diagnostic tool does not require implantation of a new device, according to Singh. “It can pick up these signals without any additional hardware,” he said.
Another factor that distinguishes the new technology is that it is based on a composite measure instead of a single measure. “As a result of this, the sensitivity is much higher than other strategies ... and the false-positive rate is very low,” Singh said. “This will give us the opportunity to risk-stratify patients more thoughtfully, and to more thoughtfully intervene and prevent HF readmissions. We will be more proactive after they’ve developed an HF event; 34 days is enough time to intervene or monitor.”
In a discussant presentation, Mary Norine Walsh, MD, FACC, medical director of heart failure and cardiac transplantation at St. Vincent Heart Center in Indianapolis, said a limitation of the technology is that “HF [with preserved ejection fraction] is increasing in outpatient and inpatient prevalence, so many of our patients, in the hospital and out, have HFpEF and are not CRT-D candidates.” Many patients with HF with reduced ejection fraction are also not CRT-D candidates, she said.
“Although the signal may be interesting from the standpoint of preventing HF events more aggressively, the number of patients who may be eligible for this type of monitoring is small,” said Walsh, president-elect of the American College of Cardiology and a member of the Cardiology Today Editorial Board.
The diagnostic service is not currently approved for use in the United States or elsewhere.
– by Erik Swain
Reference:
Boehmer JP, et al. LBCT.04 – Guiding the Momentum to Effect HF Outcomes – Ironing Out the Wrinkles. Presented at: American Heart Association Scientific Sessions; Nov. 12-16, 2016; New Orleans.
Disclosure: The study was funded by Boston Scientific. Boehmer reports consulting for Boston Scientific. Singh reports consulting for Biotronik, Boston Scientific, Impulse Dynamics, Medtronic, Respicardia and St. Jude Medical. Walsh reports no relevant financial disclosures.
Perspective
Back to Top
Christopher B. Granger, MD
This is an interesting trial, but it was not randomized. It represents one of the latest attempts for us to be better at monitoring with new technologies. The study showed that a device can be effective at monitoring parameters that seem to be related to worsening of HF. The key is the next step: Can this be used to improve outcomes and prevent hospitalizations for worsening HF?
At this point, the CardioMEMS system has better data in terms of having at least some evidence for reducing hospitalization and saving resources. We’re not there yet with this technology.
Click Here to Manage Email Alerts