Maria P Bonomini1, Daniel F Ortega2, Luis D Barja2, Emilio Logarzo2, Nicolás Mangani2, Analía Paolucci2. 1. Instituto de Ingeniería Biomédica, Facultad de Ingeniería, Universidad de Buenos Aires, Argentina; Instituto Argentino de Matemática, 'Alberto P. Calderón' CONICET, Buenos Aires, Argentina. Electronic address: paula.bonomini@conicet.gov.ar. 2. Clínica San Camilo, Buenos Aires, Argentina; Hospital Universitario Austral, Buenos Aires, Argentina.
Abstract
AIMS: Left ventricular (LV) dyssynchrony lengthens the left ventricular electrical delay (LVED), measured from QRS onset to the first peak of the LV electrogram. We constructed an ECG model to predict LVED noninvasively. METHODS: Intrapatient LVED was measured during a baseline vs nonselective His bundle pacing (nHBP) protocol. This setup provided paired synchronic/non-synchronic LVEDs, allowing intrapatient comparisons. Crosscorrelation of leads II and V6 was accomplished and extracted features together with age and gender fed a linear mixed effects model to predict LVED. RESULTS: Hemodynamic increments were consistent with LVED advances under nHBP in a subset of 17 patients (dP/dtmax, baseline: 938.82 ± 241.95 mm Hg/s vs nHBP: 1034.94 ± 253.63 mm Hg/s, p = 6.24e-4). The inclusion of the area under V6 (AV 6) and the time shift of R-peaks obtained from the crosscorrelation signal (CorS) grouped by patient significantly improved LVED estimation with respect to the model based only on QRS duration, age and gender (p = 1.7e-5). CONCLUSIONS: Interlead ECG changes explained LVED, providing clues about the electrical impulse conduction within the left ventricle noninvasively.
AIMS: Left ventricular (LV) dyssynchrony lengthens the left ventricular electrical delay (LVED), measured from QRS onset to the first peak of the LV electrogram. We constructed an ECG model to predict LVED noninvasively. METHODS: Intrapatient LVED was measured during a baseline vs nonselective His bundle pacing (nHBP) protocol. This setup provided paired synchronic/non-synchronic LVEDs, allowing intrapatient comparisons. Crosscorrelation of leads II and V6 was accomplished and extracted features together with age and gender fed a linear mixed effects model to predict LVED. RESULTS: Hemodynamic increments were consistent with LVED advances under nHBP in a subset of 17 patients (dP/dtmax, baseline: 938.82 ± 241.95 mm Hg/s vs nHBP: 1034.94 ± 253.63 mm Hg/s, p = 6.24e-4). The inclusion of the area under V6 (AV 6) and the time shift of R-peaks obtained from the crosscorrelation signal (CorS) grouped by patient significantly improved LVED estimation with respect to the model based only on QRS duration, age and gender (p = 1.7e-5). CONCLUSIONS: Interlead ECG changes explained LVED, providing clues about the electrical impulse conduction within the left ventricle noninvasively.
Authors: Andres Di Leoni Ferrari; Guilherme Ferreira Gazzoni; Luis Manuel Ley Domingues; Jessica Caroline Feltrin Willes; Gustavo Chiari Cabral; Flavio Vinicius Costa Ferreira; Laura Orlandini Lodi; Gustavo Reis Journal: Arq Bras Cardiol Date: 2022-02 Impact factor: 2.000