Pramodsingh Hirasingh Thakur

Haemodynamic monitoring of cardiac status using heart sounds from an implanted cardiac device

The aim of this study was to evaluate the haemodynamic correlates of heart sound (HS) parameters such as third HS (S3), first HS (S1), and HS‐based systolic time intervals (HSTIs) from an implantable cardiac device.

1 A Multi-Sensor Algorithm Predicts Heart Failure Events in Patients with Implanted Devices: Results from the MultiSENSE Study

Purpose: Toevaluate toperformanceof analgorithmdevelopedusing diagnostic sensor data from implanted cardiac resynchronization therapy defibrillators (CRT-D) to detect impending heart failure decompensation events. Background: Heart Failure (HF), a growing health-care challenge globally, involves costly hospitalizations with adverse impact on patient outcomes. Reliable monitoring for early signs of worsening HF is needed to enable proactive interventions for prevention of acute decompensations. We hypothesize that an algorithm combining information from a diverse set of implanted device based sensors judiciously chosen to target different aspects of HF pathophysiology can effectively detect worsening HF. Methods: MultiSENSE enrolled patients with HF and reduced EF (HFrEF) implanted with CRT-D, converted into an investigational device to enable chronic ambulatory data collection. HF events (HFEs) were defined as HF admissions or unscheduled visits with augmented intravenous HF treatment, and were independently adjudicated. Patients were assigned to Development or Test set cohorts in chronological order of enrollment. The development set was used to construct a composite index and alert algorithm (HeartLogic) combining Heart Sounds, Respiration, Thoracic Impedance, Heart Rate and Activity; whereas the test set was sequestered for its subsequent independent validation. Sensitivity was defined as the proportion of usable HFEs detected by HeartLogic alerts. Unexplained alert rate (UAR)was defined as the ratio of alerts not explained by HF to the total usable follow-up duration. The two co-primary endpoints: 1. Sensitivity performance goal (PG) of> 40%; 2. UAR PG of< 2 alerts per patient year were testedwith a 2-sided 95% confidence interval (CI). Results: Overall, 900 (Development 1⁄4 500, Test 1⁄4 400) patients had sensor data collection enabled and followed for up to a year. Primary endpoints were evaluated using the 320 patient years of follow-up data and 50 adjudicated usable HFEs in the Test Set cohort (72% male; age 66.8 10.3 years; NYHA Class at enrollment I/II/III/IV/unknown: 5%/69%/25%/1%/1%; LVEF 30.0 11.4%). With an observed sensitivity of 70% (Lower 2-sided 95% CI: 55.4%) and UAR of 1.47 (Upper 2-sided 95% CI: 1.65), both endpoints were significantly exceeded. Conclusion: The HeartLogic multi-sensor HF diagnostic algorithm significantly exceeded its pre-specified endpoints demonstrating compelling performance for worsening HF detection.