One-year follow-up of automatic adaptation of the rate response algorithm of the QT sensing, rate adaptive pacemaker

Machine learning-based atrial fibrillation detection and onset prediction using QT-dynamicity

Objective. This study examines the value of ventricular repolarization using QT dynamicity for two different types of atrial fibrillation (AF) prediction.Approach. We studied the importance of QT-dynamicity (1) in the detection and (2) the onset prediction (i.e. forecasting) of paroxysmal AF episodes using gradient-boosted decision trees (GBDT), an interpretable machine learning technique. We labeled 176 paroxysmal AF onsets from 88 patients in our unselected Holter recordings database containing paroxysmal AF episodes. Raw ECG signals were delineated using a wavelet-based signal processing technique. A total of 44 ECG features related to interval and wave durations and amplitude were selected and the GBDT model was trained with a Bayesian hyperparameters selection for various windows. The dataset was split into two parts at the patient level, meaning that the recordings from each patient were only present in either the train or test set, but not both. We used 80% on the database for the training and the remaining 20% for the test of the trained model. The model was evaluated using 5-fold cross-validation.Main results.The mean age of the patients was 75.9 ± 11.9 (range 50-99), the number of episodes per patient was 2.3 ± 2.2 (range 1-11), and CHA2DS2-VASc score was 2.9 ± 1.7 (range 1-9). For the detection of AF, we obtained an area under the receiver operating curve (AUROC) of 0.99 (CI 95% 0.98-0.99) and an accuracy of 95% using a 30 s window. Features related to RR intervals were the most influential, followed by those on QT intervals. For the AF onset forecast, we obtained an AUROC of 0.739 (0.712-0.766) and an accuracy of 74% using a 120s window. R wave amplitude and QT dynamicity as assessed by Spearman's correlation of the QT-RR slope were the best predictors.Significance. The QT dynamicity can be used to accurately predict the onset of AF episodes. Ventricular repolarization, as assessed by QT dynamicity, adds information that allows for better short time prediction of AF onset, compared to relying only on RR intervals and heart rate variability. Communication between the ventricles and atria is mediated by the autonomic nervous system (ANS). The variations in intraventricular conduction and ventricular repolarization changes resulting from the influence of the ANS play a role in the initiation of AF.

Rate-Responsive Cardiac Pacing: Technological Solutions and Their Applications

Modern cardiac pacemakers are equipped with a function that allows the heart rate to adapt to the current needs of the patient in situations of increased demand related to exercise and stress ("rate-response" function). This function may be based on a variety of mechanisms, such as a built-in accelerometer responding to increased chest movement or algorithms sensing metabolic demand for oxygen, analysis of intrathoracic impedance, and analysis of the heart rhythm (Q-T interval). The latest technologies in the field of rate-response functionality relate to the use of an accelerometer in leadless endocavitary pacemakers; in these devices, the accelerometer enables mapping of the mechanical wave of the heart's work cycle, enabling the pacemaker to correctly sense native impulses and stimulate the ventricles in synchrony with the cycles of atria and heart valves. Another modern system for synchronizing pacing rate with the patient's real-time needs requires a closed-loop system that continuously monitors changes in the dynamics of heart contractions. This article discusses the technical details of various solutions for detecting and responding to situations related to increased oxygen demand (e.g., exercise or stress) in implantable pacemakers, and reviews the results of clinical trials regarding the use of these algorithms.

Contemporary pacemakers: what the primary care physician needs to know

Pacemaker therapy is most commonly initiated because of symptomatic bradycardia, usually resulting from sinus node disease. Randomized multicenter trials assessing the relative benefits of different pacing modes have made possible an evidence-based approach to the treatment of bradyarrhythmias. During the past several decades, major advances in technology and in our understanding of cardiac pathophysiology have led to the development of new pacing techniques for the treatment of heart failure in the absence of bradycardia. Left ventricular or biventricular pacing may improve symptoms of heart failure and objective measurements of left ventricular systolic dysfunction by resynchronizing cardiac contraction. However, emerging clinical data suggest that long-term right ventricular apical pacing may have harmful effects. As the complexity of cardiac pacing devices continues to grow, physicians need to have a basic understanding of device indications, device function, and common problems encountered by patients with devices in the medical and home environment.

Developments in sensor-driven pacing

This article reviews the recent major developments in the field of rate adaptive pacing. Including, the improved instrumentation of existing sensors, the use of multiple sensors to enhance sensor specificity or sensitivity, and the automation of sensor calibration. The physiologic benefits and programming of rate adaptive pacing are reviewed.

Relationship between atrioventricular delay and oxygen consumption in patients with sick sinus syndrome: relevance to rate responsive pacing

To develop a dromotropic-controlled rate adaptive algorithm for patients with sick sinus syndrome (SSS) and intact AV conduction, 14 pace-maker patients with SSS underwent cardiopulmonary exercise testing (CPX). During exercise, the pace-maker was programmed in an AAT mode without rate adaptation, whereby 3 patients developed supraventricular arrhythmia and 11 patients kept sinus rhythm. Chronotropic incompetence (CI) at heart rate (HR) < 95 beats/min at the anaerobic threshold (AT) was found in five patients. In patients with chronotropic competence (CC), the HR increase was significantly greater than in CI patients (rest: 73.2 +/- 12.6 vs. 64.2 +/- 4.0 beats/min;AT:101.2 +/- 6.2 vs. 82.0 +/- 5.1 beats/min;peak: 135.2 +/- 10.7 vs. 103.2 +/- 10.9 beats/min). There was no significant difference in the AVD between CC and CI patients (rest: 167.7 +/- 38.6 vs. 170.8 +/- 22.5 ms, AT: 156.2 +/- 30.7 vs. 163.6 +/- 21.6 ms, peak: 144.7 +/- 29.0 vs. 152.4 +/- 15.0 ms). The correlation coefficient between HR increase and VO2 was +1.0 and between AVD decrease and VO2 - 1.0 in both groups. An increase in pacing rate from 75 beats/min to 120 beats/min without exercise (overpacing) led to a prolongation of the AV interval of about 30.6 +/- 14.2 ms. Based on this closed loop control with negative feedback, a dromotropic rate adaptive algorithm for patients with SSS and intact AV conduction could be developed.

Initial experience with a new single chamber, dual sensor rate responsive pacemaker. The Topaz Study Group

In August 1991, a new single chamber pacemaker became available that utilizes information from two sensors, activity and stimulus-to-T wave (QT) interval. We are reporting on the first 90 implants in 21 centers. T wave sensing was adequate at implantation in 88/90 patients, with a safety margin of > 100% in 86/90. Activity sensing was adequate in all patients. The contribution of each sensor (sensor blending) is programmable for each patient. Of 75 patients assessed at 1 month after implant, three have been programmed to "Activity-Only" mode, and 72 to dual sensor mode. Of these, 18 have been programmed to "QT < Activity," 48 to "QT = Activity," and 6 to "QT > Activity." Forty-five patients underwent exercise testing in dual sensor mode and a subgroup of 15 also underwent exercise testing in Activity-Only mode. The dual sensor mode produced a more gradual increase in pacing rate. Sensor Cross Checking satisfactorily prevented a sustained high pacing rate in tests of false-positive activity sensing (tapping, vibrating pacemaker, or static pressure). The maximum pacing rate on walking downstairs (94.2 +/- 7.2 ppm) was similar to that produced by walking upstairs (91.6 +/- 5.9 ppm). We conclude that initial assessment of this dual sensor, single chamber, rate responsive pacemaker confirms that the algorithm for combining data from two sensors functions satisfactorily. Dual sensor rate responsive pacing may offer significant advantages over single sensor devices, and further studies of this novel device are indicated.

New and combined sensors for adaptive-rate pacing

A new potential indication for cardiac pacing is chronotropic incompetence, that is, an inadequate cardiac rate response to exercise and other metabolic demands. Many patients who have been paced for indications such as complete heart block or sick sinus syndrome also have chronotropic incompetence. Such patients are not adequately treated when fitted with a constant rate pacemaker. Adaptive-rate pacemakers increase the pacing rate in proportion to signals derived from a biosensor which is sensitive to exertion and possibly to other metabolic requirements. These pacemakers have proven valuable for patients with overt chronotropic incompetence. However, no single sensor/algorithm is ideal and improvement has been sought by introducing new sensors, adjusting the algorithms by which biosensor signals are converted to the most appropriate pacing rate, or by combining sensors in such a way that a composite biosensor signal is derived which bears a close linear relationship with the appropriate heart rate. An example of a new sensor is the accelerometer, which is sensitive to a fuller range of movements than the piezo crystal. A successful new algorithm is the rate augmentation algorithm for use with minute ventilation, which provides a better initial pacing rate response. A combination of minute ventilation sensed by impedance changes and movement sensed with piezo crystals maintains the rapid response from the piezo crystal and overcomes its lack of proportionality. Another successful new combination of sensors is QT sensing from the evoked ventricular potential and motion sensing with a piezo crystal. As yet, these innovations have not been exhaustively tested and shown to confer clinical benefit but the improvements are such that an advantage can be expected.

Clinical results of automatic slope adaptation in a dual sensor VVIR pacemaker

Manual slope programming in rate adaptive pacemakers can be time consuming. This may become worse with dual sensor devices. The remedy is to let the pacemaker automatically learn the slopes. Fast learning replaces initial manual slope programming. Daily learning is a continuous process to determine and optimize slopes during daily life. Both methods are known for a QT sensing pacemaker. Fast learning is known for other single sensor devices. The aim of this study was to follow daily learning in a QT and activity dual sensor pacemaker, starting with factory slope settings. Six patients were studied for about 8 weeks. The daily learning algorithm appeared to be effective, showing the desired regulation processes. It took 2-5 weeks to reach full rate response.