Predicting all-cause mortality by means of a multisensor implantable defibrillator algorithm for heart failure monitoring

Background

The HeartLogic algorithm (Boston Scientific) has proved to be a sensitive and timely predictor of impending heart failure (HF) decompensation.

Objective

The purpose of this study was to determine whether remotely monitored data from this algorithm could be used to identify patients at high risk for mortality.

Methods

The algorithm combines implantable cardioverter-defibrillator (ICD)–measured accelerometer-based heart sounds, intrathoracic impedance, respiration rate, ratio of respiration rate to tidal volume, night heart rate, and patient activity into a single index. An alert is issued when the index crosses a programmable threshold. The feature was activated in 568 ICD patients from 26 centers.

Results

During median follow-up of 26 months [25th–75th percentile 16–37], 1200 alerts were recorded in 370 patients (65%). Overall, the time IN-alert state was 13% of the total observation period (151/1159 years) and 20% of the follow-up period of the 370 patients with alerts. During follow-up, 55 patients died (46 in the group with alerts). The rate of death was 0.25 per patient-year (95% confidence interval [CI] 0.17–0.34) IN-alert state and 0.02 per patient-year (95% CI 0.01–0.03) OUT of the alert state, with an incidence rate ratio of 13.72 (95% CI 7.62–25.60; P <.001). After multivariate correction for baseline confounders (age, ischemic cardiomyopathy, kidney disease, atrial fibrillation), the IN-alert state remained significantly associated with the occurrence of death (hazard ratio 9.18; 95% CI 5.27–15.99; P <.001).

Conclusion

The HeartLogic algorithm provides an index that can be used to identify patients at higher risk for all-cause mortality. The index state identifies periods of significantly increased risk of death.

Introduction

Implantable cardioverter-defibrillators (ICDs) and defibrillators for resynchronization therapy (cardiac resynchronization therapy defibrillator [CRT-D]) are widely adopted for the treatment of chronic heart failure (HF).¹ Some modern ICDs are equipped with automated algorithms that provide detailed information on the patient’s HF condition on a daily basis. Many studies have reported combining ICD diagnostics in order to better stratify and manage patients at risk for HF events.2, 3, 4 In the MultiSENSE (Multisensor Chronic Evaluation in Ambulatory Heart Failure Patients) study,⁵ a novel algorithm for HF monitoring was implemented: the HeartLogic^TM (Boston Scientific, St. Paul, MN) index, which combines physiological data from multiple ICD-based sensors. The index enabled dynamic assessment of HF, identifying periods when patients were at significantly increased risk for worsening HF.⁶ However, no study has explored whether the index predicts all-cause death. In the present study, we sought to determine whether remotely monitored data from this algorithm could be used to identify patients at high risk for mortality, and whether its predictive ability was independent of the patient’s demographic and clinical variables.