PRecision Event Monitoring for PatienTs with Heart Failure using HeartLogic (PREEMPT‐HF) study design and enrolment

Abstract

Aims

The HeartLogic multisensor index has been found to be a sensitive predictor of worsening heart failure (HF). However, there is limited data on this index's association and its constituent sensors with HF readmissions.

Methods and results

The PREEMPT‐HF study is a global, multicentre, prospective, observational, single‐arm, post‐market study. HF patients with an implantable defibrillator device or cardiac resynchronization therapy with defibrillator with HeartLogic capabilities were eligible if sensor data collection was turned on and the HeartLogic feature was not enabled. Thus, the HeartLogic Index/alert and heart sounds sensor trends were unavailable via the LATITUDE remote monitoring system to clinicians (blinded). Evaluation of subject medical records at 6 months and a final in‐clinic visit at 12 months was required for collection of all‐cause hospitalizations and HF outpatient visits. The purpose of this study is exploratory, no formal hypothesis tests are planned, and no adjustment for multiple testing will be performed. A total of 2183 patients were enrolled at 103 sites between June 2018 and June 2020. A significant proportion of the patients were implanted with implantable defibrillator devices (39%) versus cardiac resynchronization therapy with defibrillator (61%); were female (27%); over 65 (61%); New York Heart Association class I (13%), II (53%), and III (33%); ejection fraction < 25% (21%); ischaemic (50%); and with a history of renal dysfunction (23%).

Conclusions

The PREEMPT study will provide clinical data and blinded sensor trends for the characterization of sensor changes with HF readmission, tachyarrhythmias, and event subgroups. These data may help to refine the clinical use of HeartLogic and to improve patient outcomes.

Introduction

Heart failure (HF) is a complex clinical syndrome with high morbidity, mortality, and economic burden.^1,2 Chronic HF is persistent, gradually progressive, and punctuated by episodes of acute worsening, leading to hospitalizations.³ The recognition and interpretation of worsening HF signs and symptoms, such as oedema, dyspnoea, and fatigue are difficult, and failure can lead to delayed intervention.⁴ Therefore, there remains an unmet clinical need for better tools to help monitor the progression of HF and prevent hospitalizations. Physiological changes may occur early and changes in individual sensor trends, such as thoracic impedance, respiration, heart rate, heart sounds, and activity have been associated with HF events.⁵ Recently, an algorithm in implantable defibrillators, called HeartLogic™, was developed to combine the information from multiple sensors into a single index value and an alert for worsening HF. This algorithm incorporates data from the first and third heart sounds, respiration rate and volume, night-time heart rate, thoracic impedance, and time spent active, and was shown to be useful in detecting worsening HF with a median early notification of 34 days, a sensitivity of 70%, and 1.46 alerts per patient per year that were not associated with worsening HF (unexplained alert rate)⁶

Since commercialization, several studies have confirmed the performance of HeartLogic in clinical practice. An independent retrospective evaluation confirmed HeartLogic's performance in predicting HF events with 100% sensitivity and 58% positive predictive value, when clinicians were blinded to HeartLogic.⁷ While another study, in Italy, highlighted a low unexplained alert rate of 0.37 alerts/patient-year and that most of the alerts were clinically meaningful and provided information previously unknown to clinicians (60% positive predictive value), thus allowing proactive opportunity for intervention.⁸ A 2021 publication by Calò et al. also found that patients in HeartLogic alert were 31-fold more likely to have an HF event than those out of alert.⁹ However, there is limited data on this index's association with the risk of readmissions, tachyarrhythmias, or for phenotyping the broad spectrum of HF patients.

To address these clinical gaps, a larger and more diverse cohort of patients with HeartLogic blinded to clinicians is needed to evaluate better the dynamics of various HF sensors leading up to events. The PREcision Event Monitoring of PatienTs with Heart Failure using HeartLogic (PREEMPT-HF) trial of implantable defibrillator device (ICD)/cardiac resynchronization therapy with defibrillator (CRT-D) patients was designed to assess the association of sensor data with 30 day readmission following an index hospitalization for HF, characterize sensor data for association with risk of tachyarrhythmias and device therapy, and characterize sensor data for phenomapping of HF events. Additionally, analyses will be performed to characterize the association of sensor data with non-HF hospitalizations and investigate the change in patient sleep incline leading up to and following clinical events. Finally, clinical and device data will also be linked with real-world data sources for an evaluation of health economics and to develop new patient management and therapy algorithms and applications.

Study design

PREEMPT-HF is a multinational, multi-centre, post-market, prospective, non-randomized study of HF patients with an ICD or CRT-D device with multi-sensor capability (HeartLogic™, Boston Scientific Corp., Marlborough, MA). It is a single-arm trial being conducted at 103 sites in the United States, Canada, Europe, and Pacific Asia with an upper enrolment cap of 3750 patients. The trial is registered at ClinicalTrials.gov, identifier NCT03579641, and is being conducted in accordance with the Declaration of Helsinki and the International Conference on Harmonization Good Clinical Practice guidelines. The institutional review board, ethics committee, or relevant national competent authority of each participating centre approved the study before any enrolments. Subjects who met eligibility criteria and provided informed consent were considered enrolled in the trial. The study was designed and funded by Boston Scientific Corporation.

Eligibility

All enrolled patients met the inclusion and exclusion study eligibility criteria shown in Table 1.

Table 1 Study inclusion and exclusion criteria

Inclusion criteria

Boston Scientific CRT-D or ICD device implant that had HeartLogic, with Heart Failure Sensors turned ON, Respiratory Sensor turned ON, and Sleep Incline Sensor turned ON

Age 18 or above, or of legal age to give informed consent specific to each country and national laws

Documented diagnosis of heart failure

Active bipolar RV lead implant (required for sensor data)

Enrolled in the LATITUDE remote monitoring

Willing to be remotely monitored from the baseline visit for approximately 12 months with HeartLogic disabled

Exclusion criteria

Had received or were scheduled to receive a heart transplant or ventricular assist device

Life expectancy of less than 12 months

History of non-compliance to medical care

Known inability to comply with requirements of the clinical study protocol

Devices

The implanted portion of the study system included the CRT-D or ICD pulse generator with HeartLogic capabilities, along with its associated commercially available leads. For the PREEMPT-HF study, Boston Scientific Corporation CRT-D and ICDs could be associated with leads from any manufacturer.

The external portion of the system included the commercially available LATITUDE NXT 5.0 patient management system and the Programmer/Recorder/Monitor. LATITUDE is a remote monitoring system that gathers data from the pulse generator and transmits it to a web server. The primary means of device data collection for the PREEMPT-HF study was through the LATITUDE patient management system. Therefore, LATITUDE enrolment was required for subject participation in the study, and successful data transmission via LATITUDE was confirmed at the baseline visit.

Visits

Subject informed consent and enrolment occurred during the first visit (Figure 1). A baseline visit followed at least 7 days after implant and within 30 days of enrolment, when demographic, medical, and physical assessments were obtained. The protocol instructed investigators to calibrate the Sleep Incline Sensor in the seated and supine positions. The device was interrogated with a Programmer/Recorder/Monitor at enrolment and baseline visits to confirm the appropriate programming of the device (Heart Failure Sensors ON, Sleep Incline Sensor ON, and Respiratory Sensor ON) and to ensure that remote patient monitoring follow-ups were programmed to occur at a frequency no less often than once every 3 months. Importantly, the HeartLogic feature was also disabled (i.e. not reported to the clinician via LATITUDE). In this state, the device continues to collect sensor data. Still, the HeartLogic Index and Heart Sounds sensor trends are not displayed on the LATITUDE system, and clinicians do not receive HeartLogic alerts. An interim review of subject medical records to assess for study-related and reportable adverse events was required at 6 months. Study-related clinical events were defined as follows:

• hospitalization (all-cause): the subject is admitted to inpatient hospital care and discharged on a different calendar date.
• HF hospitalization: the subject is admitted with signs/symptoms of congestive heart failure and receives unscheduled augmented HF therapy with oral or intravenous medications, ultrafiltration therapy, or other parenteral therapy.
• HF readmission (30 day): the subject is admitted for an unplanned hospitalization for any cause within 30 days post discharge from an HF hospitalization.
• HF outpatient visit: the subject has signs/symptoms of congestive HF and receives unscheduled intravenous decongestive therapy (e.g., IV diuretics, IV inotropes, IV vasoactive drugs, and ultrafiltration), in a setting that does not involve a hospitalization (e.g. emergency room, HF clinic, and primary care clinic).

[IMAGE OMITTED. SEE PDF]

A final clinic visit occurred at 12 months, followed by study exit. During the follow-up period, patients received standard of care at physician's discretion. During the final visit, after confirmation of the download of device sensor data into LATITUDE, the HF sensors were programmed on or off per the physician discretion.

Data linkage

Patients in the PREEMPT-HF study provided consent for permission to link study data to third-party data sources. Data linkage will enable research to improve clinical applications for patient management and therapy (both device and non-device), develop algorithms for patient risk stratification and prognosis, evaluate health economics, and optimize clinical trial methodologies. The third-party data sources include payer administrative claims and health analytics (e.g. Centers for Medicare and Medicaid Services, Truven, and Optum), provider electronic health records, pharmacy benefit management (or other third-party administration of prescription drug programs), clinical registries (used for post-market surveillance or quality improvement), biobanks, and government databases (such as census data or social security death index).

Primary study objective

Individual sensor trends have been found to change leading up to, during, and following HF hospitalizations.⁵ These changes include increases in heart rate, decreases in thoracic impedance, increases in the third heart sound, increases in respiration rate and rapid shallow breathing index, and decreases in activity prior to HF hospitalizations.⁵ The primary objective of PREEMPT is to evaluate if the pattern of sensor changes is different when an index hospitalization for HF is followed by a readmission (all cause and/or HF) within 30 days versus when it is not. This analysis is exploratory, and no formal tests of the hypothesis are planned.

Additional study objectives

The association of sensor trends before a VT/VF event will be evaluated using an odds ratio. Phenomapping of HF events will be evaluated using cluster analysis. The association of sensor changes before a non-HF hospitalization will be evaluated using a paired t-test for each sensor.

Statistics

The purpose of this study is exploratory and to characterize the association of sensor data with various clinical events. No formal tests of the hypothesis are planned, no claims of safety or efficacy are intended, and no adjustment for multiple testing will be performed.

The primary objective will evaluate the difference in sensor trends pre-admission and post-discharge in hospitalized subjects for worsening HF. Comparisons will be made between subjects with and without 30 day readmission using an independent two-sample t-test with an alpha of 0.05 for sensor measurement. All subjects with usable sensor and episode data will be included in the analysis of characterizing sensors for the association with risk for device VT/VF therapy. The analysis will be assessed using an odds ratio. Cluster analysis will be used to evaluate all subjects with an HF event and usable sensor data for HF phenomapping. All subjects with a non-HF hospitalization and usable sensor data will be included in the analysis of characterizing sensors for association with non-HF hospitalizations with an assessment using a paired t-test.

Sample size

An index HF hospitalization is the first HF hospitalization during the study follow-up period. In estimating sample size, an HF hospitalization is deemed usable if sensor data are available during the event, and sufficient follow-up time is recorded to determine if a readmission occurs. Due to the different ranges and variability of the sensor metrics observed in the MultiSENSE study, a standardized difference was used to calculate power which would apply to any of the sensor parameters. Assuming a 30 day readmission rate of 20%, a total of 215 usable index HF hospitalizations are required to have at least 80% power to detect a standardized difference of 0.5 in the mean sensor change (admission to discharge) between the groups with and without a 30 day HF readmission using a two-side type I error (alpha) of 0.05 and 3:1 ratio of group sizes (no readmission group and a 30 day readmission group). Figure 2 shows the estimated numbers of events. Assuming a 20% loss of usable HF events due to proximity to study start and exit, a total of 270 index HF hospitalizations are required to obtain 215 usable HF hospitalizations. A maximum of 3750 subject enrolments was allowed per protocol to obtain 270 HF hospitalizations based on the assumptions of a 9% index HF hospitalization rate, 20% death or withdrawal rate, and an average of 12 months of sensor data per patient. Index HF hospitalization and readmission rates were estimated from previous studies.^6,10 The investigators conservatively estimated attrition rates considering the nature of the study, while the enrolment numbers and follow-up duration were derived from the sample size calculations.

[IMAGE OMITTED. SEE PDF]

Subgroup analyses

Analyses may be performed to evaluate the study objectives in various subgroups. The list of baseline characteristics and their corresponding subgroups to be analysed include, but are not limited to, the following:

• Device: CRT-D and ICD
• Sex: male and female
• Age: <65 and ≥65 years
• New York Heart Association class: Class I, II, III, IV
• Left ventricular ejection fraction: <25% and ≥25%
• Aetiology: ischaemic and non-ischaemic
• Renal dysfunction: yes and no

Baseline enrolment characteristics

A total of 2183 patients were enrolled from June 2018 to June 2020. Enrolment was truncated short of the upper limit after interim evaluation of event rates confirmed that a smaller sample size could fulfil the study objectives. Eighteen (18) per cent were enrolled after the US national emergency declaration concerning coronavirus disease 2019 (COVID-19) pandemic (13 March 2020). A significant proportion of the patients were implanted with ICDs (39%) versus CRT-D (61%); were female (27%); over 65 (61%); New York Heart Association class I (13%), II (53%), and III (33%); ejection fraction < 25% (21%); ischaemic (50%); and with a history of renal dysfunction (23%). Baseline characteristic of the patients are shown in Table 2.

Table 2 Baseline characteristics of patients enrolled in the PREEMPT-HF study

Characteristic	PREEMPT-HF (N = 2155)
Age, mean ± SD	67.3 ± 11.5
Male, N (%)	1569 (73%)
Race/ethnicity, N (%)
Hispanic or Latino	51 (2%)
American Indian or Alaska native	9 (0.4%)
Asian	50 (2%)
Black or African American	243 (11%)
Native Hawaiian or other Pacific Islander	2 (0.1%)
White	1639 (76%)
Other race	7 (0.3%)
CRT-D, N (%)	1323 (61%)
NYHA, N (%)
Class I	271 (13%)
Class II	1136 (53%)
Class III	707 (33%)
Class IV	26 (1%)
HF hospitalization previous 12 months, N (%)	679 (32%)
Ischaemic heart disease, N (%)	1069 (50%)
Dilated cardiomyopathy, N (%)	947 (44%)
Valvular disease, N (%)	532 (25%)
Valvular surgery, N (%)	187 (9%)
Thoracic surgery, N (%)	220 (10%)
Previous myocardial infarction, N (%)	772 (36%)
Coronary artery bypass grafting, N (%)	458 (21%)
Chronic obstructive lung disease, N (%)	359 (17%)
Pulmonary hypertension, N (%)	132 (6%)
Peripheral vascular disease, N (%)	188 (9%)
Cerebrovascular disease, N (%)	205 (10%)
Renal dysfunction, N (%)	489 (23%)
Hypertension, N (%)	1572 (73%)
Diabetes, N (%)	765 (35%)
Hyperlipidaemia, N (%)	1383 (64%)
Sleep apnoea, N (%)	399 (19%)
Depression, N (%)	271 (13%)
Hepatic disease, N (%)	81 (4%)
Anaemia, N (%)	263 (12%)
Smoking history, N (%)
Current	227 (11%)
Never	915 (42%)
Previous	1013 (47%)
NT-proBNP measurement past 90 days, N (%)	386 (18%)
NT-proBNP, median (Q1, Q3)	1482 (564, 3724)
ACE/ARBs/ARNI, N (%)	1734 (80%)
Beta-blockers, N (%)	1921 (89%)
Diuretics, N (%)	1381 (64%)
Anticoagulants, N (%)	1752 (81%)
Antiarrhythmics, N (%)	477 (22%)
Cardiac glycosides, N (%)	155 (7%)
Vasoactive drugs, N (%)	53 (2%)
Aldosterone antagonists, N (%)	835 (39%)
Calcium channel blockers, N (%)	160 (7%)

Discussion

The PREEMPT-HF study represents the largest and most geographically diverse study to date evaluating HeartLogic sensors in relationship to HF hospitalization and death. It is also the first large study to include ICD and CRT-D patients. There are limited observational studies to date evaluating HeartLogic in relationship to patient status. The studies that have evaluated HeartLogic continue to demonstrate a strong association between HeartLogic index and worsening heart.^7,11,12 However, these studies were not blinded to HeartLogic and did not have a sample size to sufficiently evaluate the objectives of PREEMPT-HF. Importantly, the present study has limited inclusion and exclusion criteria, which helped ensure the patient population broadly represents patients indicated for an ICD or CRT-D. The results will inform the real-world performance of HeartLogic for several key outcomes that have not been evaluated previously.

Heart failure hospitalization readmissions

Hospitalizations due to worsening HF are associated with high post-acute re-hospitalization and mortality, which has led payers to impose financial penalties to reduce readmissions while improving care quality.¹³ Data from fee-for-service Medicare beneficiaries suggest that these incentives have reduced the readmission rate. Still, the magnitude of readmission reduction has been modest at best, the rate of decline has slowed, and mortality has not improved.^14,15 In fact, from 2010 to 2017, an analysis of the National Readmission Database showed an increase in all-cause and HF-specific 30 and 90 day readmissions¹⁶ Readmission risk stratification algorithms developed to date have been focused on clinical and patient data.^17,18 The ability to further reduce readmissions will require new approaches. Additional HF sensors, such as pulmonary artery pressure sensors, have been proposed to risk-stratify and prevent HF readmissions.¹⁹ Sensors in implantable devices such as heart rate,²⁰ thoracic impedance,²¹ and activity²² have been shown to have predictive value for HF readmission. The addition of unique sensors may improve the predictive value, particularly when combined with a risk index which has been shown to have strong risk prediction abilities.⁶ Given the large sample size, PREEMPT-HF is anticipated to have sufficient events to evaluate sensor changes leading up to hospital readmission. These data may lead to a better understanding of discharge readiness and an improved understanding of sensor changes during the 30 day period following hospital discharge.

Arrhythmias

It is well known that arrhythmias are more common as HF progresses. Acute HF exacerbations are presumed to trigger arrhythmias through multiple mechanisms, including increased cardiac filling pressures and neurohormonal activation.^23–26 Current patient risk assessment for malignant arrhythmias is subjective, and device sensor-based algorithms for risk stratification using traditional sensors have had only modest accuracy.^27–30 The management of ventricular arrhythmias is largely reactive, with further evaluation and treatment occurring after the arrhythmia is detected and often following ICD therapy. New or worsening atrial³¹ or ventricular tachyarrhythmias³² may also trigger worsening HF. Given the known associations of arrhythmias with worsening HF, a better understanding of the association of HF-related sensor data with the risk of tachyarrhythmias and device therapy is warranted.

Phenomapping

Clinically, HF is composed of multiple heterogeneous entities including HF with reduced, preserved, and mildly-reduced ejection fraction.³³ Similarly, acute worsening HF has been classified by haemodynamic profiles³⁴ incorporating congestion (wet/dry) and perfusion (warm/cold). However, physiological measurements underlying clinical classifications exhibit significant variation. For example, even though the majority (>80%) of patients admitted for worsening HF show clinical signs or symptoms of volume overload, there is a wide distribution of body fluid and its composition.³⁵ Data-driven approaches such as cluster analysis have been applied to phenotype HF with reduced ejection fraction and HF with preserved ejection fraction,³⁶ but continuous data collected by implanted devices have not been utilized. Thus, PREEMPT-HF offers a large data set for using phenomapping techniques³⁷ to identify subgroups of HF patients.

Night-time sleep angle

Orthopnoea and paroxysmal nocturnal dyspnoea are hallmark symptoms of worsening³⁸ HF. Implantable devices can be equipped with a multi-dimensional accelerometer that can measure a subject's night-time sleep angle. However, the cross-sectional association of night-time sleep angle with nocturnal symptoms and the longitudinal changes in night-time sleep angle in the vicinity of HF hospitalization events are unknown.

Real-world data

PREEMPT-HF trial patients consented to allow for linkage to other data sources. The capability to link broad data sets to device sensor data being collected in large groups of patients and to validate the findings from sensor data with multiple sources using data-driven approaches may lead to more cost-effective and more robust device diagnostics in the future. If such claims data linked to remote monitoring data provide valid and accurate assessments of clinical events, future device diagnostic development may be done without the need for expensive clinical trials.

Coronavirus disease 2019

The use of remote monitoring and telemedicine was forever changed by the COVID-19 pandemic and are now more firmly entrenched in healthcare delivery than ever before. Multiple publications have also shown how the HeartLogic sensors are directly impacted in patients that contracted COVID-19.^39–44 As of 6 October 2020, six subjects from PREEMPT were reported to have presented to the hospital with COVID-19 and were included in an analysis comparing sensor changes with COVID-19 to those presenting with decompensated HF or pneumonia.⁴⁵ COVID-19 was distinguishable from worsening HF by an extreme and fast rise in respiratory rate and no changes in S3. Further insights may be gained from PREEMPT-HF because the study had a significant portion of additional follow-up during the pandemic's subsequent years.

Conclusions

The PREEMPT-HF trial will provide blinded sensor trend data from a large population of HF patients with ICD and CRT-D devices. Corresponding patient medical history and clinical event data will enable an exploration of relationships between physiological monitoring and index HF events in patient subgroups, 30 day hospital readmissions, and ventricular arrhythmias. These data may help refine the clinical use of HL, aiming to improve patient outcomes.

Conflict of interests

John Boehmer has received research funding from Abbott industries and compensation for consulting from Boston Scientific, Medtronic, Nanowear, and Zoll Medical Corporation. Roy Gardner and Andrew J. Sauer have received research funding and compensation for speaking and advising for Boston Scientific. Craig Stolen, Brian Kwan, Ramesh Wariar, and Stephen Ruble are Employees of Boston Scientific.

Funding

This study was funded by Boston Scientific Corporation, St Paul, MN, USA.

References

Correction: heart disease and stroke statistics‐2017 update: a report from the American Heart Association (circulation (2017) 135 (e146‐e603) DOI: 10.1161/CIR.0000000000000485). Circulation. 2017: 136.

Ambrosy AP, Fonarow GC, Butler J, Chioncel O, Greene SJ, Vaduganathan M, Lam CSP, Sato N, Shah AN, Gheorghiade M. The global health and economic burden of hospitalizations for heart failure. J Am Coll Cardiol. 2014; 63: 1123–1133.

Gheorghiade M, Pang PS. Acute heart failure syndromes. J Am Coll Cardiol. 2009; 53: 557–573.

Gravely‐Witte S, Jurgens CY, Tamim H, Grace SL. Length of delay in seeking medical care by patients with heart failure symptoms and the role of symptom‐related factors: a narrative review. Eur J Heart Fail. 2010; 12: 1122–1129.

Gardner RS, Thakur P, Hammill EF, Nair DG, Eldadah Z, Stančák B, Ferrick K, Sriratanasathavorn C, Duray GZ, Wariar R, Zhang Y, An Q, Averina V, Boehmer JP. Multiparameter diagnostic sensor measurements during clinically stable periods and worsening heart failure in ambulatory patients. ESC Heart Fail. 2021; 8: 1571–1581.

Boehmer JP, Hariharan R, Devecchi FG, Smith AL, Molon G, Capucci A, An Q, Averina V, Stolen CM, Thakur PH, Thompson JA, Wariar R, Zhang Y, Singh JP. A multisensor algorithm predicts heart failure events in patients with implanted devices: results from the MultiSENSE study. JACC Heart Fail. 2017; 5: 216–225.

Capucci A, Santini L, Favale S, Pecora D, Petracci B, Calò L, Molon G, Cipolletta L, Bianchi V, Schirripa V, Santobuono VE, la Greca C, Campari M, Valsecchi S, Ammirati F, D'Onofrio A. Preliminary experience with the multisensor HeartLogic algorithm for heart failure monitoring: a retrospective case series report. ESC Heart Fail. 2019; 6: 308–318.

Santini L, D'Onofrio A, Dello Russo A, Calò L, Pecora D, Favale S, Petracci B, Molon G, Bianchi V, de Ruvo E, Ammirati F, la Greca C, Campari M, Valsecchi S, Capucci A. Prospective evaluation of the multisensor HeartLogic algorithm for heart failure monitoring. Clin Cardiol. 2020; 43: 691–697.

Calò L, Bianchi V, Ferraioli D, Santini L, Dello Russo A, Carriere C, Santobuono VE, Andreoli C, la Greca C, Arena G, Talarico A, Pisanò E, Santoro A, Giammaria M, Ziacchi M, Viscusi M, de Ruvo E, Campari M, Valsecchi S, D’Onofrio A, Minati M, Tota C, Martino A, Tavoletta V, Manzo M, Ammirati F, Mahfouz K, Colaiaco C, Guerra F, Zorzin Fantasia A, Amato V, Savarese G, Pellegrini D, Pimpinicchio L, Pecora D, Bartoli C, Borrello VM, Ratti M, de Rosa F, Quirino F, Tomaselli C, Marino E, Baiocchi C, de Vivo O, Baccani B, Amellone C, Lucciola MT, Angeletti A, Frisoni J, Brignoli M, Costa A, Pangallo A, Benedetto F, Pepi P, Nicolis D, Petracci B, Giubilato G, Carbonardi L, Porcelli D, Romani B, Zuccaro LM. Multiparametric implantable cardioverter‐defibrillator algorithm for heart failure risk stratification and management: an analysis in clinical practice. Circ Heart Fail. 2021; 14: [eLocator: E008134].

Gillis AM, Kerr CR, Philippon F, Newton G, Talajic M, Froeschl M, Froeschl S, Swiggum E, Yetisir E, Wells GA, Tang AS. Impact of cardiac resynchronization therapy on hospitalizations in the resynchronization‐defibrillation for ambulatory heart failure trial. Circulation. 2014; 129: 2021–2030.

Calò L, Bianchi V, Ferraioli D, Santini L, dello Russo A, Carriere C, Santobuono VE, Andreoli C, la Greca C, Arena G, Talarico A, Pisanò E, Santoro A, Giammaria M, Ziacchi M, Viscusi M, de Ruvo E, Campari M, Valsecchi S, D’Onofrio A, Minati M, Tota C, Martino A, Tavoletta V, Manzo M, Ammirati F, Mahfouz K, Colaiaco C, Guerra F, Zorzin Fantasia A, Amato V, Savarese G, Pellegrini D, Pimpinicchio L, Pecora D, Bartoli C, Borrello VM, Ratti M, de Rosa F, Quirino F, Tomaselli C, Marino E, Baiocchi C, de Vivo O, Baccani B, Amellone C, Lucciola MT, Angeletti A, Frisoni J, Brignoli M, Costa A, Pangallo A, Benedetto F, Pepi P, Nicolis D, Petracci B, Giubilato G, Carbonardi L, Porcelli D, Romani B, Zuccaro LM. Multiparametric implantable cardioverter‐defibrillator algorithm for heart failure risk stratification and management: an analysis in clinical practice. Circ Heart Fail. 2021; 14: [eLocator: e008134].

Santini L, D'Onofrio A, dello Russo A, Calò L, Pecora D, Favale S, Molon G, Bianchi V, de Ruvo E, Ammirati F, la Greca C, Campari M, Valsecchi S, Capucci A. Prospective evaluation of the multisensor HeartLogic algorithm for heart failure monitoring. Clin Cardiol. 2020; 43: 691–697.

Desai AS. The three‐phase terrain of heart failure readmissions. Circ Heart Fail. 2012; 5: 398–400.

Dharmarajan K, Wang Y, Lin Z, Normand SLT, Ross JS, Horwitz LI, Suter LG, Drye EE, Bernheim SM, Krumholz HM. Association of changing hospital readmission rates with mortality rates after hospital discharge. JAMA. 2017; 318: 270–278.

Gupta A, Fonarow GC. The hospital readmissions reduction program—learning from failure of a healthcare policy. Eur J Heart Fail. 2018; 20: 1169–1174.

Khan MS, Sreenivasan J, Lateef N, Abougergi MS, Greene SJ, Ahmad T, Anker SD, Fonarow GC, Butler J. Trends in 30‐ and 90‐day readmission rates for heart failure. Circ Heart Fail. 2021; 14: [eLocator: e008335].

Eapen ZJ, Liang L, Fonarow GC, Heidenreich PA, Curtis LH, Peterson ED, Hernandez AF. Validated, electronic health record deployable prediction models for assessing patient risk of 30‐day rehospitalization and mortality in older heart failure patients. JACC Heart Fail. 2013; 1: 245–251.

Ross JS, Mulvey GK, Stauffer B, Patlolla V, Bernheim SM, Keenan PS, Krumholz HM. Statistical models and patient predictors of readmission for heart failure: a systematic review. Arch Intern Med. 2008; 168: 1371–1386.

Heywood JT, Jermyn R, Shavelle D, Abraham WT, Bhimaraj A, Bhatt K, Sheikh F, Eichorn E, Lamba S, Bharmi R, Agarwal R, Kumar C, Stevenson LW. Impact of practice‐based management of pulmonary artery pressures in 2000 patients implanted with the CardioMEMS sensor. Circulation. 2017; 135: 1509–1517.

Gopinathannair R, Sharma A, Jones P, English C, Furmanek S, Olshansky B. Heart rate score and outcomes in ICD patients: insights from the prospective randomized INTRINSIC RV trial. J Interv Card Electrophysiol. 2022; 64: 87–93.

Conraads VM, Tavazzi L, Santini M, Oliva F, Gerritse B, Yu CM, Cowie MR. Sensitivity and positive predictive value of implantable intrathoracic impedance monitoring as a predictor of heart failure hospitalizations: the SENSE‐HF trial. Eur Heart J. 2011; 32: 2266–2273.

Conraads VM, Spruit MA, Braunschweig F, Cowie MR, Tavazzi L, Borggrefe M, Hill MR, Jacobs S, Gerritse B, van Veldhuisen D. Physical activity measured with implanted devices predicts patient outcome in chronic heart failure. Circ Heart Fail. 2014; 7: 279–287.

Ip JE, Cheung JW, Park D, Hellawell JL, Stein KM, Iwai S, Liu CF, Lerman BB, Markowitz SM. Temporal associations between thoracic volume overload and malignant ventricular arrhythmias: a study of intrathoracic impedance. J Cardiovasc Electrophysiol. 2011; 22: 293–299.

Sullivan RM, Olshansky B. OptiVol fluid index: does an association with ventricular arrhythmias hold water? J Cardiovasc Electrophysiol. 2011; 22: 300–301.

Popjies E, Boehmer J. Therapeutic implications of implantable device‐based monitoring of patients with heart failure. Curr Treat Options Cardiovasc Med. 2008; 10: 371–379.

Abi‐Saleh B, Malozzi C, Charaf E, Omar B, Massey C, Khoury M, Verma S. Worsening thoracic impedance as a ventricular tachyarrhythmia risk. Rev Cardiovasc Med. 2014; 15: 226–231.

Jędrzejczyk‐Patej E, Kowalski O, Średniawa B, Pruszkowska P, Sokal A, Szulik M, Mazurek M, Kowalczyk J, Kalarus Z, Lenarczyk R. Trying to predict the unpredictable: variations in device‐based daily monitored diagnostic parameters can predict malignant arrhythmic events in patients undergoing cardiac resynchronization therapy. Cardiol J. 2014; 21: 405–412.

Hariharan R, Cao M, Sweeney R, Zhang Y, Averina V, An Q. Heart failure diagnostic sensor measurements change during atrial arrhythmias. Heart Rhythm. 2017; 14: [eLocator: S404].

Pubbi D, Hudson M, Molon G, Capucci A, Slotwiner D, Cao M, Thakur PH, Averina V, Zhang Y, Wariar R, Thompson J. AF in HF: the chicken or the egg? Role of Crt device based sensor data in identifying a causal relationship between AF and worsening HF. J Card Fail. 2016; 22: [eLocator: S65].

Capucci A, Merkely B, Sweeney R, Zhang Y, Thakur P, Averina V. Arrhythmia burden increases prior to worsening heart failure events. Heart Rhythm. 2017; 14: [eLocator: S216].

Ponikowski P, Alemayehu W, Oto A, Bahit MC, Noori E, Patel MJ, Butler J, Ezekowitz JA, Hernandez AF, Lam CSP, O'Connor CM, Pieske B, Roessig L, Voors AA, Westerhout C, Armstrong PW, VICTORIA Study Group. Vericiguat in patients with atrial fibrillation and heart failure with reduced ejection fraction: insights from the VICTORIA trial. Eur J Heart Fail. 2021; 23: 1300–1312.

Poole JE, Johnson GW, Hellkamp AS, Anderson J, Callans DJ, Raitt MH, Reddy RK, Marchlinski FE, Yee R, Guarnieri T, Talajic M, Wilber DJ, Fishbein DP, Packer DL, Mark DB, Lee KL, Bardy GH. Prognostic importance of defibrillator shocks in patients with heart failure. N Engl J Med. 2008; 359: 1009–1017.

McDonagh TA, Metra M, Adamo M, Gardner RS, Baumbach A, Böhm M, Burri H, Butler J, Čelutkienė J, Chioncel O, Cleland JGF, Coats AJS, Crespo‐Leiro MG, Farmakis D, Gilard M, Heymans S, Hoes AW, Jaarsma T, Jankowska EA, Lainscak M, Lam CSP, Lyon AR, McMurray JJV, Mebazaa A, Mindham R, Muneretto C, Francesco Piepoli M, Price S, Rosano GMC, Ruschitzka F, Kathrine Skibelund A, ESC Scientific Document Group, de Boer RA, Christian Schulze P, Abdelhamid M, Aboyans V, Adamopoulos S, Anker SD, Arbelo E, Asteggiano R, Bauersachs J, Bayes‐Genis A, Borger MA, Budts W, Cikes M, Damman K, Delgado V, Dendale P, Dilaveris P, Drexel H, Ezekowitz J, Falk V, Fauchier L, Filippatos G, Fraser A, Frey N, Gale CP, Gustafsson F, Harris J, Iung B, Janssens S, Jessup M, Konradi A, Kotecha D, Lambrinou E, Lancellotti P, Landmesser U, Leclercq C, Lewis BS, Leyva F, Linhart A, Løchen ML, Lund LH, Mancini D, Masip J, Milicic D, Mueller C, Nef H, Nielsen JC, Neubeck L, Noutsias M, Petersen SE, Sonia Petronio A, Ponikowski P, Prescott E, Rakisheva A, Richter DJ, Schlyakhto E, Seferovic P, Senni M, Sitges M, Sousa‐Uva M, Tocchetti CG, Touyz RM, Tschoepe C, Waltenberger J, Adamo M, Baumbach A, Böhm M, Burri H, Čelutkienė J, Chioncel O, Cleland JGF, Coats AJS, Crespo‐Leiro MG, Farmakis D, Gardner RS, Gilard M, Heymans S, Hoes AW, Jaarsma T, Jankowska EA, Lainscak M, Lam CSP, Lyon AR, McMurray JJV, Mebazaa A, Mindham R, Muneretto C, Piepoli MF, Price S, Rosano GMC, Ruschitzka F, Skibelund AK. 2021 ESC guidelines for the diagnosis and treatment of acute and chronic heart failure. Eur Heart J. 2021; 42: 3599–3726.

Chioncel O, Mebazaa A, Maggioni AP, Harjola VP, Rosano G, Laroche C, Piepoli MF, Crespo‐Leiro MG, Lainscak M, Ponikowski P, Filippatos G, Ruschitzka F, Seferovic P, Coats AJS, Lund LH, ESC‐EORP‐HFA Heart Failure Long‐Term Registry Investigators. Acute heart failure congestion and perfusion status – impact of the clinical classification on in‐hospital and long‐term outcomes; insights from the ESC‐EORP‐HFA heart failure long‐term registry. Eur J Heart Fail. 2019; 21: 1338–1352.

Miller WL, Mullan BP. Understanding the heterogeneity in volume overload and fluid distribution in decompensated heart failure is key to optimal volume management: role for blood volume quantitation. JACC Heart Fail. 2014; 2: 298–305.

Cogliati C, Ceriani E, Gambassi G, de Matteis G, Perlini S, Perrone T, Muiesan ML, Salvetti M, Leidi F, Ferrara F, Sabbà C, Suppressa P, Fracanzani A, Montano N, Fiorelli E, Tripepi G, Gori M, Pitino A, Pietrangelo A. Phenotyping congestion in patients with acutely decompensated heart failure with preserved and reduced ejection fraction: the Decongestion duRing therapY for acute decOmpensated heart failure in HFpEF vs HFrEF‐ DRY‐OFF study. Eur J Intern Med. 2022; 97: 69–77.

Shah SJ, Katz DH, Selvaraj S, Burke MA, Yancy CW, Gheorghiade M, Bonow RO, Huang CC, Deo RC. Phenomapping for novel classification of heart failure with preserved ejection fraction. Circulation. 2015; 131: 269–279.

Wang CS, FitzGerald JM, Schulzer M, Mak E, Ayas NT. Does this dyspneic patient in the emergency department have congestive heart failure? JAMA. 2005; 294: [eLocator: 1944].

Yapejian AR, Fudim M. Novel findings of respiratory rate increases using the multisensor HeartLogic heart failure monitoring algorithm in COVID‐19‐positive patients: a case series. Eur Heart J Case Rep. 2021; 5: [eLocator: ytab067].

Mitter SS, Alvarez‐Garcia J, Miller MA, Moss N, Lala A. Insights from HeartLogic multisensor monitoring during the COVID‐19 pandemic in new York City. JACC Heart Fail. 2020; 8: 1053–1055.

Egolum UO, Parikh K, Lekavich C, Wosik J, Frazier‐Mills C, Fudim M. Applications of the multisensor HeartLogic heart failure monitoring algorithm during the COVID‐19 global pandemic. JACC Case Rep. 2020; 2: 2265–2269.

Ebrille E, Amellone C, Lucciola MT, Suppo M, Antonacci G, Gotta F, Birolo M, Favro E, Orlando F, Giammaria M. HeartLogic algorithm for heart failure monitoring in the era of CoViD‐19 pandemic and remote monitoring/telemedicine. Eur Heart J. 2021; 42: 42.

Heggermont W, Nguyen PAH, Lau C‐W, Tournoy K. A steep increase in the HeartLogic index predicts COVID‐19 disease in an advanced heart failure patient. Case Rep Cardiol. 2020; 2020: [eLocator: 8896152].

Shumway JL, Stolen CM, Ahmed R, Plumer M, Capodilupo RC. Case reports of implantable cardiac device physiologic sensor changes in subjects with coronavirus disease‐2019 infection. J Card Fail. 2021; 27: 373–378.

Gardner RS, Capodilupo RC, Ahmed R, Stolen CM, An Q, Averina V, Hernandez AF, Boehmer JP. Multiparameter diagnostic sensor measurements in heart failure patients presenting with SARS‐CoV‐2 infection. ESC Heart Fail. 2021; 8: 4026–4036.