Multisensor Chronic Evaluation in Ambulatory Heart Failure Patients (MultiSENSE Study)

Status: Completed in 2015

Heart failure (HF) involves costly hospitalizations with adverse impact on patient outcomes.  The aim of this study was to develop and validate a device-based diagnostic algorithm and alert (HeartLogic™) to predict heart failure events (HFE).


  • Sensitivity of 70% (95% confidence interval: 55.4% to 82.1%) 
  • Unexplained alert rate of 1.47 per patient-year (95% confidence interval: 1.32 to 1.65)
  • The median lead time prior to HFE was 34.0 days (interquartile range: 19.0 to 66.3 days)


  • International, multicenter, nonrandomized
  • 900+ patients were followed for up to 1 year 
  • Heart Failure Events (HFE) were independently adjudicated to include events where primary cause was worsening HF and patient admission included an overnight stay or intravenous medications.
  • Sensor data included heart rate, accelerometer-based heart sounds, respiration rate, relative tidal volume, activity, and intrathoracic impedance.
  • Algorithm development combined key feature trends into a composite index and associated alert using HFE and sensor data from 500 patients in Development Set.
  • Performance was evaluated against HFE and sensor data in sequestered Test Set patients.  Two co-primary endpoints included:
    • Sensitivity to detect HFE >40%
    • Unexplained alert rate <2.0 per patient-year

Patient Population

  • Age 18 or above
  • Currently implanted with a COGNIS CRT-D system
  • NYHA Class II, III, IV within the last 6 months


HeartLogic index and alert mechanisms are investigational and not available for sale.