Activation of the HeartLogic algorithm on top of heart failure care: a multicenter propensity-matched cohort analysis

Abstract

Abstract Background Hospitalization for decompensated heart failure may be prevented by early detection of fluid retention. The multisensory HeartLogic™ algorithm, incorporated in a cardiac implantable electronic device (CIED) aims to detect impending fluid retention and thereby enables timely adjustment of medical therapy. However, it is to be investigated whether HeartLogic™ provides clinical benefit compared to heart failure care with conventional telemonitoring. This analysis investigates the effects of activating HeartLogic™ on top of heart failure care with telemonitoring on the number of episodes with fluid retention and heart failure related hospitalizations. Methods Heart failure patients with a CIED were recruited from the outpatient clinics of four European cardiology departments. All patients were included from January 2017 until December 2020, and followed-up for 365 days. Patients with a CIED and an activated HeartLogic™ algorithm were compared to a 1:1 propensity score-based matched control group consisting of patients with CIED on routine telemonitoring. Data of all episodes of (impending) fluid retention with ≥2 signs and symptoms of congestion were included for analyses. Results Data of 127 patients with an activated HeartLogic™ algorithm were adequately matched with 127 heart failure patients with a CIED on routine telemonitoring. Median age was 68 [59–75], majority of patients were male (80%), 46% had an ischemic etiology of heart failure. Total follow-up consisted of 254 patient years. During follow up, 77 (61%) individual patients with HeartLogic™ experienced an episode of fluid retention, compared to 85 (67%) induvial patients on routine telemonitoring. Patients with an activated HeartLogic™ algorithm had 1.62±1.78 events of fluid retention per patient year (PPY) compared to 2.61±3.71 events PPY in patients on routine telemonitoring, p<0.01 (Figure 1). Hospitalization for fluid retention occurred in 7 (6%) HeartLogic™ patients (0.06±0.27 hospitalizations PPY) compared to 13 (10%) patients on routine telemonitoring (0.15±0.45 PPY), p=0.07 (Figure 2A). Mean length of hospitalization in days PPY was 0.29±1.46 in patients with HeartLogic™ and 1.59±6.30 in patients on routine telemonitoring, p=0.02 (Figure 2B). Conclusion In a real-world multicenter heart failure population, activation of the HeartLogic™ algorithm was associated with a lower number of episodes of fluid retention per patient and a shorter duration of hospitalization for congestive heart failure. Funding Acknowledgement Type of funding sources: None.