Implementation of a Cardiovascular Implantable Electronic Device Heart Failure Prediction Tool-Guided Management Pathway

Abstract

Introduction and ObjectivesCardiovascular implantable electronic devices (CIEDs) monitor physiologic variables that could identify subacute heart failure (HF) decompensation and impending HF hospitalization (HFH). One such algorithm uses measurements from the previous 30 days of CIED remote monitoring data to predict low-, medium-, or high-probability of HFH in the next 30 days. We sought to understand how to prospectively implement use of such algorithms in routine HF care. MethodsFrom 1/18/24 to 4/19/24, HF risk categories were predicted from scheduled remote transmissions every 30 days and from unscheduled transmissions for all patients at two distinct cardiology clinics. Clinicians contacted and assessed patients at high-risk regarding symptoms and then provided an empiric 3-day diuretic intervention (initiation or dose augmentation), adjusted guideline-directed medical therapy (GDMT), or performed other clinical action as appropriate. ResultsAmong 358 patients with 1140 remote transmissions, 72 (20%) had ≥1 transmission categorized as high-risk. Mean patient age was 72.8 years, 346 (97%) were male, and 221 (62%) had a pre-existing diagnosis of HF. Of these 72 patients, 67 (93%) were successfully contacted; 34 (51%) had no HF symptoms, 24 (36%) had mild to moderate symptoms, and 2 (3%) had severe symptoms. Forty-six (69%) patients had clinical action taken, including 28 (42%) with a diuretic intervention and 12 (18%) with GDMT augmented. ConclusionsIn this implementation study, clinicians contacted and assessed nearly all patients at high-risk for HF decompensation based on CIED remote monitoring data and intervened on more than two-thirds. A randomized clinical trial is needed to determine whether this algorithm and subsequent intervention improves clinical outcomes.