Electrogram morphology discriminators in implantable cardioverter defibrillators: A comparative evaluation

Real-world data of patients affected by advanced heart failure treated with implantable cardioverter defibrillator and left ventricular assist device: Results of a multicenter observational study

BACKGROUND

Left ventricular assist device (L-VAD) implantation is increasingly used in patients with heart failure (HF) and most patients also have an implantable cardioverter defibrillator (ICD). Limited data are available on the incidence of ICD therapies and complications in this special setting. The aim of this study was to analyze the real-world incidence and predictors of ICD therapies, complications and interactions between ICD and L-VAD.

METHODS

We conducted a multicenter retrospective observational study in patients with advanced HF implanted with ICD and a continuous-flow L-VAD, followed-up in five advanced HF centers in Northern Italy.

RESULTS

A total of 234 patients (89.7% male, median age 59, 48.3% with ischemic etiology) were enrolled. After a median follow-up of 21 months, 66 patients (28.2%) experienced an appropriate ICD therapy, 22 patients (9.4%) an inappropriate ICD therapy, and 17 patients (7.3%) suffered from an interaction between ICD and L-VAD. The composite outcome of all ICD-related complications was reported in 41 patients (17.5%), and 121 (51.7%) experienced an L-VAD-related complication. At multivariable analysis, an active ventricular tachycardia (VT) zone and a prior ICD generator replacement were independent predictors of ICD therapies and of total ICD-related complications, respectively.

CONCLUSIONS

Real-world patients with both L-VAD and ICD experience a high rate of ICD therapies and complications. Our findings suggest the importance of tailoring device programming in order to minimize the incidence of unnecessary ICD therapies, thus sparing the need for ICD generator replacement, a procedure associated to a high risk of complications.

Artificial intelligence for detection of ventricular oversensing: Machine learning approaches for noise detection within nonsustained ventricular tachycardia episodes remotely transmitted by pacemakers and implantable cardioverter-defibrillators

BACKGROUND

Pacemakers (PMs) and implantable cardioverter-defibrillators (ICDs) increasingly automatically record and remotely transmit nonsustained ventricular tachycardia (NSVT) episodes, which may reveal ventricular oversensing.

OBJECTIVES

We aimed to develop and validate a machine learning algorithm that accurately classifies NSVT episodes transmitted by PMs and ICDs in order to lighten health care workload burden and improve patient safety.

METHODS

PMs or ICDs (Boston Scientific, St Paul, MN) from 4 French hospitals with ≥1 transmitted NSVT episode were split into 3 subgroups: training set, validation set, and test set. Each NSVT episode was labeled as either physiological or nonphysiological. Four machine learning algorithms-2DTF-CNN, 2D-DenseNet, 2DTF-VGG, and 1D-AgResNet-were developed using training and validation data sets. Accuracies of the classifiers were compared with an analysis of the remote monitoring team of the Bordeaux University Hospital using F2 scores (favoring sensitivity over predictive positive value) using an independent test set.

RESULTS

A total of 807 devices transmitted 10,471 NSVT recordings (82% ICD; 18% PM), of which 87 devices (10.8%) transmitted 544 NSVT recordings with nonphysiological signals. The classification by the remote monitoring team resulted in an F2 score of 0.932 (sensitivity 95%; specificity 99%) The 4 machine learning algorithms showed high and comparable F2 scores (2DTF-CNN: 0.914; 2D-DenseNet: 0.906; 2DTF-VGG: 0.863; 1D-AgResNet: 0.791), and only 1D-AgResNet had significantly different labeling from that of the remote monitoring team.

CONCLUSION

Machine learning algorithms were accurate in detecting nonphysiological signals within electrograms transmitted by PMs and ICDs. An artificial intelligence approach may render remote monitoring less resourceful and improve patient safety.

Progressive implantable cardioverter-defibrillator therapies for ventricular tachycardia: The efficacy and safety of multiple bursts, ramps, and low-energy shocks

BACKGROUND

The Heart Rhythm Society, the European Heart Rhythm Association, the Asia Pacific Heart Rhythm Society, the Latin American Heart Rhythm Society expert consensus statement on optimal implantable cardioverter-defibrillator programming recommends burst antitachycardia pacing (ATP) for the treatment of ventricular tachycardia (VT) up to high rates. The number of bursts is not specified, and treatment by ramps or low-energy shocks is not recommended.

OBJECTIVES

We investigated the efficacy and safety of progressive therapies for VTs between 150 and 200 beats/min. After 3 failed bursts, we compared 3 ramps vs 3 bursts followed by a low-energy shock vs high-energy shock.

METHODS

Using remote monitoring, we included monomorphic VT episodes treated with ≥1 burst.

RESULTS

A total of 1126 VT episodes were included. A single burst was as likely to terminate VT between 150 and 200 beats/min as VT between 200 and 230 beats/min (63% vs 64%; P=.41), but was more likely to accelerate the latter (3.2% vs 0.25%; P<.01). For VT <200 beats/min, the likelihood of ATP success increased progressively (73% with 2 bursts, 78% with 3 bursts). Three additional bursts further increased VT termination to 89%, similar to the success rate with 3 additional ramps (88%; P=.17). Programming 6 bursts is associated with the probability of acceleration requiring shock of 6.6%. A low-energy first shock was less successful than a high-energy shock (66% vs 86%; P<.01) and more likely to accelerate VT (17% vs 0%; P<.01).

CONCLUSION

Programming up to 6 burst ATP therapies for VTs 150-200 beats/min can avoid implantable cardioverter-defibrillator shocks in most patients. Ramp ATP after failed bursts were similarly effective. Low-energy shocks are less effective and more arrhythmogenic than high-energy shocks.