Interpreting device diagnostics for lead failure

Navigating the Landscape of Medical Device Advisories: A Special Report From the Canadian Heart Rhythm Society Device Advisory Committee

Cardiac implantable electronic devices (CIEDs) are often important for regulating cardiac rate and rhythm. Pacemakers and defibrillators are among the top 10 most implanted medical devices, with > 1.5 million devices implanted annually. Although millions of patients have benefited with improved quality of life and survival, CIED systems are becoming increasingly complex and do not always perform according to expectations. Advisory notices communicate important information about the safety and performance of a medical device to health care providers and patients. Medical device recalls are common, with > 35 unique device recalls in the past 5 years. From an ethical standpoint, CIED recalls highlight a range of considerations including the consent process, duty to report, how best to promote autonomous decision-making, trust in the health care system, as well as disproportionate effects of these considerations on equity-deserving groups. The purpose of the current article is to review and advise regarding the process around medical device advisory and recall, with a specific focus on clinicians caring for patients affected by these devices. We have sought the input of a lawyer, a patient advocacy group, and an ethicist to guide the clinical management of, and communications regarding, device recalls and advisories. Diligent surveillance and a clear, transparent patient consent process regarding these small but potentially serious device anomalies is paramount in ensuring patients believe they are safe and informed. Meaningful patient engagement helps to ensure optimal communication and disclosure mechanisms before implantation and throughout follow-up, accessibility of information in the initial implant and recall action process, and trust in health care systems and providers.

Sensing of concealed ventricular extrasystoles by pacemakers. Fifty years later

Whether a pacemaker can sense concealed ventricular extrasystoles still remains debatable since its occurrence was first proposed in 1972. It must remain a diagnosis of exclusion if it really exists. Isoelectric complexes and all the causes of oversensing especially discrete false signals generated by a defective pacemaker lead must be excluded before concealed ventricular extrasystoles can be postulated.

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.

Time course of oversensing and impedance changes in developing implantable cardioverter-defibrillator lead fracture

Background

Pace-sense conductors comprise a pacing coil to the tip electrode and cable to the ring-electrode. Implantable cardioverter-defibrillator (ICD) lead-monitoring diagnostics include pacing impedance (direct current resistance [DCR]) and measures of oversensing. How they change as fractures progress is unknown.

Objectives

To characterize the relationship between oversensing, impedance, and structural changes in ICD leads developing pace-sense conductor fractures.

Methods

We performed bending tests on 39 leads connected to ICD generators in an electrolyte bath with simulated electrograms. DCR was recorded every 3 minutes; electrograms were telemetered continuously. Twenty-two leads were tested to develop partial or complete fracture criteria confirmed by imaging, using DCR or DCR variability measured by standard deviation (σ_DCR). Results are reported for 17 other test leads.

Results

Initial oversensing occurred with partial pacing coil fracture vs complete ring cable fracture and correlated with bending-induced DCR peaks. These peaks were too small to be detected by clinical impedance measurements and were characterized by small increases in σ_DCR (≥0.5 Ω). Impedance threshold alerts occurred at complete pacing coil fracture but only later for ring cable fractures. The oversensing alert triggered before device-detected ventricular fibrillation more frequently than impedance alerts (94% vs 17%; P = .00002).

Conclusions

In conductor fracture, early oversensing corresponds to partial pacing coil fracture or complete ring cable fracture and correlates with transient bending-induced impedance increases, which are detected by impedance variability but too small to trigger clinical impedance alerts. This explains why clinical oversensing alerts provide more warning for device-detected ventricular fibrillation than impedance alerts and suggests how to improve impedance diagnostics based on short-term variability.