No abstract

The morphological analysis of implantable cardioverter defibrillator (ICD) stored electrograms (EGM) using a multilayer perceptron (MLP) has been proposed for discrimination between supraventricular and ventricular arrhythmias. However, a reliable estimation of the accuracy of MLP methods is lacking. The aim of the study was to compare the morphology and spectrum-based MLP with more conventional morphology-based algorithms in a large series of ICD-stored episodes of arrhythmia. One set of ICD-stored electrograms was used for control and training purposes and a second one, consisting of spontaneous episodes in patients with dual chamber ICDs, for validation of the MLP performance. The correlation waveform analysis (CWA) and the EGM width criterion were compared with MLP methods. Bootstrap resampling techniques were used to extract the relevant information in the MLP training. The morphology-based MLP achieved better discrimination than any other method, with areas under the receiver operating characteristic (ROC) curve (tolerance intervals): 0.96 (0.81, 0.96) for MLP, 0.91 (0.77, 0.94) for CWA, and 0.68 (0.49, 0.78) for EGM width in the validation set. A specificity of 73.0% was obtained at 95% sensitivity, compared with 38.1% and 55.1% using CWA and EGM width criteria, respectively. In contrast, the generalization capabilities of spectral-based MLP methods are poor, showing a lower area under the ROC curve in the validation set. Time-domain MLP techniques may be useful for the morphological analysis of the intracardiac EGM signal stored by ICD devices. When properly trained and validated, these methods perform better than other commonly used morphological criteria for discrimination between supraventricular and ventricular arrhythmias.

No abstract

No abstract

Funding Acknowledgements Type of funding sources: None. Background Left ventricular ejection fraction (LVEF) is at present the main parameter used to guide the implantation of cardioverter-defibrillators (ICD) in patients with ischaemic cardiomyopathy. However, LVEF alone is a poor predictor of ventricular arrhythmias (VA) and sudden cardiac death (SCD). The majority of ICDs implanted for the primary prevention of SCD never deliver appropriate therapies, while being an expensive therapy and being potentially iatrogenic. Therefore, insertion of ICDs based solely on LVEF is not efficient and calls for an improvement in the risk-stratification of post-myocardial infarction patients. Inducibility of ventricular arrhythmias with programmed ventricular stimulation (PVS) has been used in the past to assess the risk of malignant arrhythmias. However, its current role in clinical practice is unknown. Purpose The aim of this meta-analysis was to study the role of PVS to predict future arrhythmic events in patients with previous myocardial infarction. Methods We performed a systematic review and a random-effect meta-analysis of studies reporting VAs –defined as SCD, sudden cardiac arrest, sustained monomorphic ventricular tachycardia (SMVT) or appropriate ICD therapies– after PVS in patients with ischaemic heart disease. Whenever possible, we used inducibility of SMVT as the positive response to PVS. Otherwise, we used the definition pre-specified in the corresponding study. The primary outcome was to estimate the positive and negative predictive value (PPV and NPV, respectively) of PVS. Subgroup analyses were performed according to the PVS timing (<1 or ≥1 month after infarction), definition of a positive PVS, use of ICD, or study era (before or after 2000). We searched in MEDLINE, EMBASE and Cochrane Library electronic databases up to November 2021. Results Thirty studies were included with a total of 8253 patients (mean age 58.5 years, mean LVEF 38.8%). A total of 909 arrhythmic events occurred after a mean follow-up of 24.5 months. Overall, 23% (95% confidence interval [CI] 18-29%) of subjects had inducible arrhythmias with PVS. Inducibility was associated with a higher risk of subsequent VAs (odds ratio 6.26, 95% CI 3.97-9.89, p<0.001). The pooled NPV was 95%, (95% CI 93-96%, p<0.001) and the PPV was 25% (95% CI 20-31%, p<0.001). There was significant heterogeneity between studies (I2 78% for the odds ratio, 96% for the NPV and 68% for the PPV). There were no statistically significant differences in any of the values according to the pre-specified subgroups. Meta-regression showed a reduction of the NPV with longer follow-up (B coefficient –0.033; p=0.012). Conclusion PVS has a high NPV (95%) and could be useful to rule out subjects at low risk of arrhythmic events. However, it decreases with longer follow-ups and, thus, the arrhythmic risk may need to be reassessed.