Abstract 20131: Prediction of Ventricular Tachyarrhythmias in Patients With Implantable-Cardioverter- Defibrillator
Introduction: Implantable-cardioverter defibrillator (ICD) therapy has been shown to be effective in reducing mortality in patients with or at risk of sudden cardiac death. However, the burden of ICD therapies has recently been observed to correlate with increase in morbidity and mortality.
Hypothesis: Ventricular tachyarrhythmias (VTVF events) could be predicted days in advance with a weighted score based on device-derived parameters.
Methods: Twenty-five ICD-indicated patients were prospectively enrolled. After implantation of ICD, each patient underwent weekly data collections by remote monitoring (Carelink, Medtronic) for 6 months and appropriate VTVF events were determined. A VTVF event could be a single discrete ventricular tachycardia (VT) or ventricular fibrillation (VF) episode or a VTVF storm with inter-episode interval less than 24 hours. An algorithm was developed to create a weighted score based on the assessment of directional changes in multiple device-derived parameters in Carelink Cardiac Compass over a 15-day moving window. The algorithm for VTVF prediction was then retrospectively tested based on the comparison between the weighted score one day before an event vs. the score calculated for immediately prior 15-day window. The efficiency of the prediction in one day before an event was determined with the sensitivity and the specificity.
Results: The patients enrolled in this study had LVEF 45.0±15.4% and NYHA class 1.4±0.6. Eleven (44%) had coronary heart disease, 14 (56%) had ischemic or non-ischemic cardiomyopathy, 7 (28%) had hypertension, and 20 (80%) had a history of sustained VT or VF. Of 25 patients, 10 (40%) developed a total of 123 appropriate VTVF episodes (12.4±13.2, median 5 per patient) that were terminated by ICD therapies. Of these 123 episodes, 100 were classified as VT while remaining 23 episodes were VF. When the weighted-score algorithm was tested in all 25 patients, the prediction of VTVF events one day before the occurrence had a 100% sensitivity and 74% specificity.
Conclusions: VTVF events in ICD patients can be predicted one day in advance by a weighted score algorithm. This may provide a time window for executing appropriate measures to prevent the occurrence of VTVF and especially VTVF storms.
- Arrhythmias, treatment of
- Implantable cardiovertor defibrillator
- Sudden cardiac death
- Ventricular tachycardia
- Ventricular fibrillation
Author Disclosures: N. Chan: Other Research Support; Modest; Statistical support. L. Tam: None. X. Zhou: Employment; Significant; Medtronic employee. T. Ling: Employment; Significant; Medtronic employee. H. Lu: Employment; Significant; Medtronic employee.
- © 2016 by American Heart Association, Inc.