TL;DR: The proposed method is advantageous because it uses only the RR-interval signal for arrhythmia beat and episode classification and the results compare well with more complex methods.
TL;DR: Systematic continuous ECG monitoring is made compulsory to improve the overall results of subarachnoid hemorrhage, irrespective of early or delayed surgical treatment.
Abstract: The frequency and severity of cardiac arrhythmias were studied in 70 patients with spontaneous subarachnoid hemorrhage investigated prospectively with 24-hour Holter monitoring. Patients were less than 70 years old and without clinical and/or ECG signs of previous heart disease; Holter monitoring was initiated within 48 hours of subarachnoid hemorrhage. Arrhythmias were detected in 64 of the 70 patients (91%). Twenty-nine of the 70 patients (41%) showed serious cardiac arrhythmias; malignant ventricular arrhythmias, i.e., torsade de pointe and ventricular flutter or fibrillation, occurred in 3 cases. Serious ventricular arrhythmias were associated with QTc prolongation and hypokalemia. No correlation was found between the frequency and severity of cardiac arrhythmias and the neurologic condition, the site and extent of intracranial blood on computed tomography scan, or the location of ruptured malformation. The extremely high incidence of cardiac arrhythmias, sometimes serious, in the acute period after subarachnoid hemorrhage and the absence of clinical and radiologic predictors make systematic continuous ECG monitoring compulsory to improve the overall results of subarachnoid hemorrhage, irrespective of early or delayed surgical treatment.
TL;DR: Subjects with short APERPs and multiple pathways are at higher risk of developing life-threatening arrhythmic events and are the best candidates for prophylactic ablation.
TL;DR: Three patients developed severe ventricular arrhythmias while taking flecainide for atrial fibrillation and indicate the limitations of classifying patients as "high-risk" or "low- risk" for proarrhythmic complications of anti-arrHythmic therapy.
Abstract: Flecainide acetate has a recognized proarrhythmic effect in patients treated for ventricular tachycardia. Three patients developed severe ventricular arrhythmias while taking flecainide for atrial fibrillation. Patient 1 had normal ventricular function and idiopathic atrial fibrillation. Treadmill exercise tests during digoxin therapy showed no ventricular arrhythmia; however, during flecainide therapy the patient developed ventricular flutter at his peak exercise level that required cardioversion. Patient 2 had normal ventricular function and a prosthetic mitral valve. During therapy with flecainide, 150 mg twice daily, he had an episode of sustained ventricular tachycardia, also at his peak exercise level. Patient 3 had paroxysmal atrial fibrillation and hypertrophic cardiomyopathy but no previous ventricular arrhythmia. She died suddenly within 10 days of starting flecainide therapy. Judged from previous findings none of these patients was considered at high risk for proarrhythmia. These cases suggest a possible relation between vigorous exercise, atrial fibrillation, and the proarrhythmic properties of flecainide and indicate the limitations of classifying patients as "high-risk" or "low-risk" for proarrhythmic complications of anti-arrhythmic therapy.
TL;DR: It is demonstrated that accurate methods of computing the time-frequency domain should be found for ECG signals and only then should future work be done to design discriminatory features and classifiers for arrhythmias.
Abstract: The short time Fourier transform (STFT), smoothed pseudo Wigner Ville distribution (SPWVD), and cone-shaped kernel distribution (CKD) have been used to compare the time-frequency distribution of normal sinus rhythm, ventricular tachycardia, ventricular flutter, and ventricular fibrillation signals. This work is a pilot study to illustrate that the CKD and SPWVD have better time and frequency resolution than the STFT. It demonstrates that accurate methods of computing the time-frequency domain should be found for ECG signals. Only then should future work be done to design discriminatory features and classifiers for arrhythmias. >