Validation of a peak endocardial acceleration-based algorithm to optimize cardiac resynchronization: early clinical results

Left Ventricular Pressure Estimation Using Machine Learning-Based Heart Sound Classification

Objective

A method to estimate absolute left ventricular (LV) pressure and its maximum rate of rise (LV dP/dtmax) from epicardial accelerometer data and machine learning is proposed.

Methods

Five acute experiments were performed on pigs. Custom-made accelerometers were sutured epicardially onto the right ventricle, LV, and right atrium. Different pacing configurations and contractility modulations, using isoflurane and dobutamine infusions, were performed to create a wide variety of hemodynamic conditions. Automated beat-by-beat analysis was performed on the acceleration signals to evaluate amplitude, time, and energy-based features. For each sensing location, bootstrap aggregated classification tree ensembles were trained to estimate absolute maximum LV pressure (LVPmax) and LV dP/dtmax using amplitude, time, and energy-based features. After extraction of acceleration and pressure-based features, location specific, bootstrap aggregated classification ensembles were trained to estimate absolute values of LVPmax and its maximum rate of rise (LV dP/dtmax) from acceleration data.

Results

With a dataset of over 6,000 beats, the algorithm narrowed the selection of 17 predefined features to the most suitable 3 for each sensor location. Validation tests showed the minimal estimation accuracies to be 93% and 86% for LVPmax at estimation intervals of 20 and 10 mmHg, respectively. Models estimating LV dP/dtmax achieved an accuracy of minimal 93 and 87% at estimation intervals of 100 and 200 mmHg/s, respectively. Accuracies were similar for all sensor locations used.

Conclusion

Under pre-clinical conditions, the developed estimation method, employing epicardial accelerometers in conjunction with machine learning, can reliably estimate absolute LV pressure and its first derivative.

Analysis of Cardiac Vibration Signals Acquired From a Novel Implant Placed on the Gastric Fundus

The analysis of cardiac vibration signals has been shown as an interesting tool for the follow-up of chronic pathologies involving the cardiovascular system, such as heart failure (HF). However, methods to obtain high-quality, real-world and longitudinal data, that do not require the involvement of the patient to correctly and regularly acquire these signals, remain to be developed. Implantable systems may be a solution to this observability challenge. In this paper, we evaluate the feasibility of acquiring useful electrocardiographic (ECG) and accelerometry (ACC) data from an innovative implant located in the gastric fundus. In a first phase, we compare data acquired from the gastric fundus with gold standard data acquired from surface sensors on 2 pigs. A second phase investigates the feasibility of deriving useful hemodynamic markers from these gastric signals using data from 4 healthy pigs and 3 pigs with induced HF with longitudinal recordings. The following data processing chain was applied to the recordings: (1) ECG and ACC data denoising, (2) noise-robust real-time QRS detection from ECG signals and cardiac cycle segmentation, (3) Correlation analysis of the cardiac cycles and computation of coherent mean from aligned ECG and ACC, (4) cardiac vibration components segmentation (S1 and S2) from the coherent mean ACC data, and (5) estimation of signal context and a signal-to-noise ratio (SNR) on both signals. Results show a high correlation between the markers acquired from the gastric and thoracic sites, as well as pre-clinical evidence on the feasibility of chronic cardiovascular monitoring from an implantable cardiac device located at the gastric fundus, the main challenge remains on the optimization of the signal-to-noise ratio, in particular for the handling of some sources of noise that are specific to the gastric acquisition site.

Sensors for rate-adaptive pacing: How they work, strengths, and limitations

Chronotropic incompetence is the inability of the sinus node to increase heart rate commensurate with increased metabolic demand. Cardiac pacing alone may be insufficient to address exercise intolerance, fatigue, dyspnea on exertion, and other symptoms of chronotropic incompetence. Rate-responsive (adaptive) pacing employs sensors to detect physical or physiological indices and mimic the response of the normal sinus node. This review describes the development, strengths, and limitations of a variety of sensors that have been employed to address chronotropic incompetence. A mini-tutorial on programming rate-adaptive parameters is included along with emphasis that patients' lifestyles and underlying medical conditions require careful consideration. In addition, special sensor applications used to respond prophylactically to physiologic signals are detailed and an in-depth discussion of sensors as a potential aid in heart failure management is provided.

Change in indication for cardiac resynchronization therapy?

Cardiac resynchronization therapy (CRT) has rapidly evolved as a standard therapy for heart failure (HF) patients with ventricular conduction delay. Although in early trials, only patients with sinus rhythm and advanced stages of HF have been candidates for CRT, more recent data have expanded the indications to patients with mild-to-moderate HF and atrial fibrillation and patients in need of antibradycardia pacing with reduced left ventricular function. On the other hand, it is now well recognized that patients with a wide QRS (>150 ms) and left bundle branch block morphology benefit most from CRT, whereas in patients with a more narrow QRS complex (<130 ms) CRT may actually be harmful despite the evidence of ventricular dyssynchrony by echocardiography. There is no prospective randomized study showing mortality benefit from a combined CRT defibrillating device over a CRT pacer alone. This is especially important because recent data indicate that older patients with non-ischaemic cardiomyopathy may not benefit from the implantable cardioverter-defibrillator as much as previously thought. Thus, the decision for a CRT pacer versus CRT defibrillating should be tailored to the therapeutic goal (improvement in prognosis versus symptomatic relief), patient age, underlying cardiac disease and comorbidities. This article gives an overview over the current indications for CRT according to published literature and the European guidelines for pacing and HF.

Multi-Sense CardioPatch: A Wearable Patch for Remote Monitoring of Electro-Mechanical Cardiac Activity

This study describes the conceptual design and the first prototype implementation of the Multi-Sense CardioPatch, a wearable multi-sensor patch for remote heart monitoring aimed at providing a more detailed and comprehensive heart status diagnostics. The system integrates multiple sensors in a single patch for detection of both electrical (electrocardiogram, ECG) and mechanical (heart sounds, HS) cardiac activity, in addition to physical activity (PA). The prototypal system also comprises a microcontroller board with a radio communication unit and it is powered by a Li-Ion rechargeable battery. Results from preliminary evaluations on healthy subjects have shown that the prototype can successfully measure electro-mechanical cardiac activity, providing useful cardiac indexes. The system has potential to improve remote monitoring of cardiac function in chronically diseased patients undergoing home-based cardiac rehabilitation programs.

Assessment of Myocardial Contractility by SonR Sensor in Patients Undergoing Cardiac Resynchronization Therapy

BACKGROUND

SonR sensor signal correlates well with myocardial contractility expressed in terms of left ventricular (LV) dP/dt max. The aim of our study was to evaluate the changes in myocardial contractility during isometric effort in heart failure patients undergoing cardiac resynchronization therapy (CRT) with right atrial SonR sensor.

METHODS

Thirty-one patients (19 men, 65 ± 7 years, LV ejection fraction [LVEF] 28% ± 5%, in sinus rhythm) were implanted with a CRT-defibrillator (CRT-D) device equipped with SonR sensor, which was programmed in VVI mode at 40 beats/min. Twenty-four hours after implantation, each patient underwent a noninvasive hemodynamic evaluation at rest and during isometric effort, including: (1) measurement of beat-to-beat endocavitary SonR signal; (2) echocardiographic assessment; and (3) continuous measurement of blood pressure with Nexfin method (BMEYE, Amsterdam, the Netherlands). The following contractility parameters were considered: (1) mean value of beat-to-beat SonR signal; (2) mean value of LV dP/dt by Nexfin system; and (3) fractional shortening (FS) by echocardiography.

RESULTS

At the third minute of the isometric effort, mean value of SonR signal significantly increased from baseline (P < 0.001). Similarly, mean value of both LV dP/dt by Nexfin and FS significantly increased compared to the resting condition (P < 0.001; P < 0.001). While in 27 (88%) patients SonR signal increased at the third minute of the isometric effort, in four (12%) patients SonR signal decreased. In these patients, both LV dP/dt by Nexfin and FS consensually decreased.

CONCLUSIONS

In CRT patients, SonR sensor is able to detect changes in myocardial contractility in a consensual way like noninvasive methods such as Nexfin system and echocardiography.

Automatic optimization of cardiac resynchronization therapy using SonR-rationale and design of the clinical trial of the SonRtip lead and automatic AV-VV optimization algorithm in the paradym RF SonR CRT-D (RESPOND CRT) trial

Although cardiac resynchronization therapy (CRT) is effective in most patients with heart failure (HF) and ventricular dyssynchrony, a significant minority of patients (approximately 30%) are non-responders. Optimal atrioventricular and interventricular delays often change over time and reprogramming these intervals might increase CRT effectiveness. The SonR algorithm automatically optimizes atrioventricular and interventricular intervals each week using an accelerometer to measure change in the SonR signal, which was shown previously to correlate with hemodynamic improvement (left ventricular [LV] dP/dtmax). The RESPOND CRT trial will evaluate the effectiveness and safety of the SonR optimization system in patients with HF New York Heart Association class III or ambulatory IV eligible for a CRT-D device. Enrolled patients will be randomized in a 2:1 ratio to either SonR CRT optimization or to a control arm employing echocardiographic optimization. All patients will be followed for at least 24 months in a double-blinded fashion. The primary effectiveness end point will be evaluated for non-inferiority, with a nested test of superiority, based on the proportion of responders (defined as alive, free from HF-related events, with improvements in New York Heart Association class or improvement in Kansas City Cardiomyopathy Questionnaire quality of life score) at 12 months. The required sample size is 876 patients. The two primary safety end points are acute and chronic SonR lead-related complication rates, respectively. Secondary end points include proportion of patients free from death or HF hospitalization, proportion of patients worsened, and lead electrical performance, assessed at 12 months. The RESPOND CRT trial will also examine associated reverse remodeling at 1 year.

Heart failure monitoring with a cardiac resynchronization therapy device-based cardiac contractility sensor: a case series

Introduction

The SonR signal has been shown to reflect cardiac contractility. It is recorded with an atrial lead connected to a cardiac resynchronization therapy defibrillator. For the first time, clinical evidence on the use of the SonR signal in the monitoring of the clinical status of heart failure patients implanted with cardiac resynchronization therapy defibrillator are presented through three clinical cases.

Case presentation

In the two first patients (non-Hispanic/Latino white), the SonR amplitude increases concomitantly to clinical status improvement subsequent to cardiac resynchronization therapy defibrillator implantation. In the third patient (non-Hispanic/Latino white), a decrease in SonR amplitude is observed concomitantly to atrial fibrillation and clinical status deterioration.

Conclusions

This case series reports the association between SonR signal amplitude changes and patients’ clinical status. Combined with remote monitoring, early SonR signal amplitude remote monitoring could be a promising tool for heart failure patients’ management.

Association between frequent cardiac resynchronization therapy optimization and long-term clinical response: a post hoc analysis of the Clinical Evaluation on Advanced Resynchronization (CLEAR) pilot study

Aims

The long-term clinical value of the optimization of atrioventricular (AVD) and interventricular (VVD) delays in cardiac resynchronization therapy (CRT) remains controversial. We studied retrospectively the association between the frequency of AVD and VVD optimization and 1-year clinical outcomes in the 199 CRT patients who completed the Clinical Evaluation on Advanced Resynchronization study.

Methods and results

From the 199 patients assigned to CRT-pacemaker (CRT-P) (New York Heart Association, NYHA, class III/IV, left ventricular ejection fraction <35%), two groups were retrospectively composed a posteriori on the basis of the frequency of their AVD and VVD optimization: Group 1 (n = 66) was composed of patients ‘systematically’ optimized at implant, at 3 and 6 months; Group 2 (n = 133) was composed of all other patients optimized ‘non-systematically’ (less than three times) during the 1 year study. The primary endpoint was a composite of all-cause mortality, heart failure-related hospitalization, NYHA functional class, and Quality of Life score, at 1 year. Systematic CRT optimization was associated with a higher percentage of improved patients based on the composite endpoint (85% in Group 1 vs. 61% in Group 2, P < 0.001), with fewer deaths (3% in Group 1 vs. 14% in Group 2, P = 0.014) and fewer hospitalizations (8% in Group 1 vs. 23% in Group 2, P = 0.007), at 1 year.

Conclusion

These results further suggest that AVD and VVD frequent optimization (at implant, at 3 and 6 months) is associated with improved long-term clinical response in CRT-P patients.

A randomized pilot study of optimization of cardiac resynchronization therapy in sinus rhythm patients using a peak endocardial acceleration sensor vs. standard methods

AIMS

Non-response rate to cardiac resynchronization therapy (CRT) might be decreased by optimizing device programming. The Clinical Evaluation on Advanced Resynchronization (CLEAR) study aimed to assess the effects of CRT with automatically optimized atrioventricular (AV) and interventricular (VV) delays, based on a Peak Endocardial Acceleration (PEA) signal system.

METHODS AND RESULTS

This multicentre, single-blind study randomized patients in a 1 : 1 ratio to CRT optimized either automatically by the PEA-based system, or according to centres' usual practices, mostly by echocardiography. Patients had heart failure (HF) New York Heart Association (NYHA) functional class III/IV, left ventricular ejection fraction (LVEF) <35%, QRS duration >150 or >120 ms with mechanical dyssynchrony. Follow-up was 1 year. The primary endpoint was the proportion of patients who improved their condition at 1 year, based on a composite of all-cause death, HF hospitalizations, NYHA class, and quality of life. In all, 268 patients in sinus rhythm (63% men; mean age: 73.1 ± 9.9 years; mean NYHA: 3.0 ± 0.3; mean LVEF: 27.1 ± 8.1%; and mean QRS duration: 160.1 ± 22.0 ms) were included and 238 patients were randomized, 123 to PEA and 115 to the control group. At 1 year, 76% of patients assigned to PEA were classified as improved, vs. 62% in the control group (P= 0.0285). The percentage of patients with improved NYHA class was significantly (P= 0.0020) higher in the PEA group than in controls. Fatal and non-fatal adverse events were evenly distributed between the groups.

CONCLUSION

PEA-based optimization of CRT in HF patients significantly increased the proportion of patients who improved with therapy, mainly through improved NYHA class, after 1 year of follow-up.

Clinical validation of a real-time data processing system for cardiac output and arterial pressure measurement during intraoperative biventricular pacing optimization

Biventricular pacing (BiVP) improves cardiac output (CO) and mean arterial pressure (MAP) after cardiopulmonary bypass (CPB) in selected patients at risk for acute left heart failure after cardiac surgery. Optimization of atrioventricular delay (AVD) and interventricular delay (VVD) to maximize the hemodynamic effect of pacing requires rapid and accurate data processing. Conventional post hoc data processing (PP) is accurate but time-consuming, and infeasible in the intraoperative setting. We created a customized, real-time data processing (RTP) system to improve data processing efficiency, while maintaining accuracy. Biventricular pacing optimization was performed within 1 hour of the conclusion of CPB in 10 patients enrolled in the Biventricular Pacing After Cardiac Surgery trial. Cardiac output, measured by an electromagnetic flow meter, and arterial pressure were recorded as AVD was randomly varied across seven settings and VVD across nine settings. Post hoc data processing values calculated by two observers were compared to RTP-generated outputs for CO and MAP. Interexaminer reliability coefficients were generated to access the dependability of RTP. Interexaminer reliability coefficient values ranged from 0.997 to 0.999, indicating RTP is as reliable as PP for optimization. Real-time data processing is instantaneous and therefore is more practical in a clinical setting than the PP method. Real-time data processing is useful for guiding intraoperative BiVP optimization and merits further development.

Acute animal and human study of tensiometric pacing lead sensor based on triboelectricity

Cardiac contractions bend the implanted cardiac lead body, extend and compress the lead conductors, their insulation and the inserted stylet. Magnitude of lead deflection depends on cardiac muscle contraction forces. The purpose of study was to measure the charge generated due to triboelectric effect between one of the lead conductors and the inserted stylet. The charge was measured by differential charge amplifier being connected to isolation amplifier and power supply. Sensor signal, ECG and intracardiac electrograms were acquired. Three models of custom designed leads were implanted in 8 sheep. Measurements were done in 18 patients undergoing pacemaker implantation and replacement procedures. Atrial and ventricular tensiometric signals were recorded in dual chamber and in single-lead VDD patients. Recordings in sinus rhythm at various AV intervals and in supraventricular tachycardia were done. In average, charge variation between 1 and 600 pC was measured. Tensiometric stylet could be feasible hemodynamic sensor for myocardial contraction detection. Its main advantage is that it is easily exchangeable and universal for all leads.

Usefulness of hemodynamic sensors for physiologic cardiac pacing in heart failure patients

The rate adaptive sensors applied to cardiac pacing should respond as promptly as the normal sinus node with an highly specific and sensitive detection of the need of increasing heart rate. Sensors operating alone may not provide optimal heart responsiveness: central venous pH sensing, variations in the oxygen content of mixed venous blood, QT interval, breathing rate and pulmonary minute ventilation monitored by thoracic impedance variations, activity sensors. Using sensors that have different attributes but that work in a complementary manners offers distinct advantages. However, complicated sensors interactions may occur. Hemodynamic sensors detect changes in the hemodynamic performances of the heart, which partially depends on the autonomic nervous system-induced inotropic regulation of myocardial fibers. Specific hemodynamic sensors have been designed to measure different expression of the cardiac contraction strength: Peak Endocardial Acceleration (PEA), Closed Loop Stimulation (CLS) and TransValvular Impedance (TVI), guided by intraventricular impedance variations. Rate-responsive pacing is just one of the potential applications of hemodynamic sensors in implantable pacemakers. Other issues discussed in the paper include: hemodynamic monitoring for the optimal programmation and follow up of patients with cardiac resynchronization therapy; hemodynamic deterioration impact of tachyarrhythmias; hemodynamic upper rate limit control; monitoring and prevention of vasovagal malignant syncopes.

Validation of automated monitoring of cardiac output for biventricular pacing optimization

Biventricular pacing (BiVP) can increase cardiac output (CO) during acute failure of the left ventricle (LV) after cardiac surgery. This CO benefit is maximized by adjustment of atrioventricular (AVD) and interventricular (VVD) pacing delays. Real-time CO calculation could facilitate this optimization. Accordingly, we compared real-time automated analysis (AA) of CO with manual analysis (MA) in an animal model of pressure overload of the right ventricle (RV). In six anesthetized pigs, pacing leads were placed on the right atrium, RV, and LV. Complete heart block was induced with ethanol injection, and RV systolic pressure was doubled with a pulmonary artery snare. Atrioventricular pacing delay was varied over seven common values and VVD over nine, in random sequence. Two LV pacing sites (LVPS) were also tested. Aortic flow velocity, measured by ultrasonic flow probe, was integrated by AA and MA to calculate CO. Interexaminer Reliability Coefficient (IRC) was determined by Analysis of Variance (ANOVA) for two 10-second runs in each animal. Cardiac output-AVD and CO-VVD relations were similar for AA and MA. Interexaminer Reliability Coefficients were 0.997 and 0.994 for MA vs. AA. Automated analysis was available in real-time. Manual analysis was delayed at 2 hours or more. Automated analysis merits development for real-time optimization of intraoperative BiVP.

Optimization techniques in cardiac resynchronization therapy

Cardiac resynchronization therapy improves symptoms and cardiac function, as well as reduces mortality in patients with progressive congestive heart failure, reduced left ventricular ejection fraction and a left bundle branch block on the surface electrocardiogram. As many as 30% of patients fail to have an adequate response. The interplay between the atrioventricular delay and the contribution of a properly timed atrial contraction to ventricular filling along with a properly timed sequence of activation of the right and left ventricular is crucial to maximizing the benefits of cardiac resynchronization therapy devices.