Abstract

This study did investigate how well a moving dipole approximation of the electrical activity of the heart can contribute in differentiating strict left bundle branch block (sLBBB) from other, unspecified depolarization abnormalities. 561 ECGs with given sLBBB diagnoses from the LBBB initiative of the ISCE 2018 meeting were used for applying a moving dipole model to the individual heart beats, providing the 3-dimensional time courses of the dipole's position and momentum. The dipole position and momentum traces plus sex and QRS width were used as input for a logistic regression model. The classification performances of the various input dataset were compared by calculating the area under the receiver operating characteristic curve (AUC) and through a twofold cross validation approach. Best classification performance was observed using 10 ms time segments including 9 dipole position and 12 dipole momentum parameters with average cross validation AUC value of 0.920 (95% CI [0.896; 0.945]). The moving dipole analysis based on the standard 12lead ECG provides new and physiologically relevant information about the location of the electrical activity during depolarization, which may help in improving the assessment of depolarization abnormalities.

Automated detection of T-end in high precision is required for ECG safety assessment of new chemical entities (NCEs

16We present a new family TrX of ECG biomarkers based on the T vector velocity (TVV) for assessing 17 drug effects on ventricular repolarization. Assuming a link between the TVV and the instantaneous 18 change of the cellular action potentials, drugs accelerating repolarization by blocking inward 19 (depolarizing) ion currents cause a relative increase of the TVV, while drugs delaying repolarization by 20 blocking outward ion currents cause a relative decrease of the TVV. 21 Evaluating the published data from two FDA funded studies, the TrX effect profiles indicate 22 increasingly delayed electrical activity over the entire repolarization process for drugs solely reducing 23 outward potassium current (dofetilide, moxifloxacin). For drugs eliciting block of the inward sodium or 24 calcium currents (mexiletine, lidocaine), the TrX effect profiles were consistent with accelerated 25 electrical activity in the initial repolarization phase. For multichannel blocking drugs (ranolazine) or 26 drug combinations blocking multiple ion currents (dofetilide + mexiletine, dofetilide + lidocaine), the 27 overall TrX effect profiles indicate a superposition of the individual TrX effect profiles. 28 The parameter Tr40c allows separating pure potassium channel blocking drugs from multichannel 29 blocking drugs with an area under the ROC curve (AUC) value of 0.90, CI = [0.88 to 0.92]. This is 30 significantly larger than the performance of J-T peak c (0.81, CI = [0.78 to 0.84]) using the published data 31 from the second FDA study. Further performance improvement was achieved by combining the ten 32 parameters Tr10c to Tr100c in a logistic regression model, resulting in an AUC value of 0.94. 33 The TVV based approach substantially improves assessment of drug effects on cardiac repolarization, 34 providing a plausible and improved mechanistic link between drug effects on ionic currents and overall 35 ventricular repolarization reflected in the body surface ECG. TVV may contribute to a better 36 assessment of the proarrhythmic risk of drugs beyond QTc prolongation and JT peak c. 37 3 38

Background We present a new family of ECG biomarkers for assessing drug effects on ventricular repolarization. We show that drugs blocking inward (depolarizing) ion currents cause a relative increase of the T vector velocity (TVV) and accelerate repolarization, while drugs blocking outward ion currents cause a relative decrease of the TVV and delay repolarization. The results suggest a link between the TVV and the instantaneous change of the cellular action potentials that may contribute to bridge the gap between the surface ECG and myocardial cellular processes. Methods We measure TVV as the time required to reach X % of the total Tr ajectory length of the T vector loop, denoted as TrX. Applied to data from two FDA funded studies (22+22 subjects, 5232+4208 ECGs) which target ECG effects of various ion-channel blocking drugs, the Tr X effect profiles indicate increasingly delayed electrical activity over the entire repolarization process for drugs solely reducing outward potassium current (dofetilide, moxifloxacin). For drugs eliciting block of the inward sodium or calcium currents (mexiletine, lidocaine), the Tr X effect profiles were consistent with accelerated electrical activity in the initial repolarization phase. For multichannel blocking drugs (ranolazine) or drug combinations blocking multiple ion currents (dofetilide + mexiletine, dofetilide + lidocaine), the overall Tr X effect profiles indicate a superposition of the individual Tr X effect profiles. Results The parameter Tr40c differentiates pure potassium channel blocking drugs from multichannel blocking drugs with an area under the ROC curve (AUC) of 0.90, CI = [0.88 to 0.92]. This is significantly better than the performance of J-T peak c (0.81, CI = [0.78 to 0.84]) identified as the best parameter in the second FDA study. Combining the ten parameters Tr10c to Tr100c in a logistic regression model further improved the AUC to 0.94, CI = [0.92 to 0.96]. Conclusions TVV analysis substantially improves assessment of drug effects on cardiac repolarization, providing a plausible and improved mechanistic link between drug effects on ionic currents and overall ventricular repolarization reflected in the body surface ECG. TVV contributes to an enhanced appraisal of the proarrhythmic risk of drugs beyond QTc prolongation and J-T peak c.