Real-World Diagnostic Performance for Heart Failure Events of TriageHF® Algorithm.

BACKGROUND AND

OBJECTIVES

Heart failure (HF) remote monitoring programs collect HF-related parameters from cardiac implantable electronic devices, enabling early interventions and reducing healthcare burdens. This study aimed to characterize HF patients, alert burden, and the association between TriageHF® alerts and HF decompensation.

METHODS

This retrospective observational study, 134 patients enrolled in an outpatient remote monitoring HF program with the TriageHF® algorithm between February 2022 and July 2023. Data on alert frequency, data driving the alerts, and alert duration were collected. Clinical outcomes included unplanned emergency room visits, HF hospitalizations, all-cause death, and a composite of these events.

RESULTS

Over a median follow-up of 1.7 years, 77 individual high-risk alerts (0.44 alerts per patient-year) occurred. HF patients who eventually had an alert were similar to those without any alerts, except for the increased prevalence of chronic kidney disease, stroke history and use of furosemide. The event rate was 1.1% per month, and the alert state was 3.2% of the total follow-up duration. Six events occurred during high-risk days (event rate of 7.9% per month vs. 0.8% per month during standard-risk days). TriageHF® alerts demonstrated a sensitivity of 35%, a specificity of 71%, a positive predictive value of 25%, and a negative predictive value of 80% for predicting a composite of HF hospitalization, unplanned ER visits, and death.

CONCLUSIONS

The TriageHF® algorithm identified periods with a higher incidence of HF-related adverse events, supporting efficient resource allocation. The low alert burden also supports feasibility in resource-limited settings.