Martín Yebra, Alba Pilar
Martínez Cortés, Juan Pablo (dir.) ; Caiani, Enrico (dir.)
Universidad de Zaragoza,
2017
Resumen: Cardiovascular diseases (CVDs) represents the leading cause of mortality worldwide [1,2]. These pathological conditions are mainly characterized by a structurally abnormal heart, that is, a vulnerable substrate, prone to the abnormal generation and/or propagation of the electrical impulse, determining the onset of ventricular arrhythmias, which can result in sudden cardiac death (SCD) [3]. In this context, the assessment of ventricular repolarization from the electrocardiogram (ECG) signal has been shown to provide with valuable information for risk stratification and several electrocardiographic indices have been proposed in the literature [4].
The main objective of this thesis is to propose methodological advances for the assessment of ventricular repolarization instability in pathological and abnormal conditions. These contributions are aimed at improving the prediction of ventricular arrhythmias and, consequently, better identifying SCD risk. In particular, we have addressed this objective by developing robust methodologies for the assessment of T-wave alternans (TWA) and ventricular repolarization instability, in invasive and non-invasive cardiac signals, that have been evaluated in both experimental and clinical conditions.
In the first part of the thesis, TWA was simultaneously characterized (prevalence, magnitude, time-course, and alternans waveform) in body-surface ECG and intracardiac electrograms (EGMs) signals during coronary artery occlusion. Signals from both body surface ECG and intracardiac EGMs recorded from 4 different anatomical heart locations (coronary sinus, epicardial space and left and right ventricles) were analyzed following a multilead strategy. Leads were linearly combined using the periodic component analysis (πCA) [5], which maximizes the 2-beat periodicity (TWA periodicity) content present on the available leads. Then the Laplacian Likelihood Ratio method (LLRM) [6] was applied for TWA detection and estimation. A sensitivity study for TWA detection from the 5 different locations of leads was performed, revealing that it is the combination of the ECG leads that better performs. In addition, this multilead approach allowed us to find the optimal combination of intracardiac leads usable for in-vivo monitorization of TWA directly from an implantable device, with a sensitivity comparable to the ECG analysis. These results encourage further research to determine the feasibility of predicting imminent VT/VF episodes by TWA analysis implemented in implantable cardioverter defibrillator’s (ICD) technology.
Then, we have studied the potential changes induced by a prolonged exposure to simulated microgravity on ventricular repolarization in structurally normal hearts. It is well known that this environmental condition affects the control of autonomic and cardiovascular systems [7], with a potential increase on cardiac electrical instability. The effects of short- (5 days), mid- (21 days) and long- (60 days) exposure to simulated microgravity on TWA using the head-down bed-rest (HDBR) model [8] were assessed. TWA was evaluated before (PRE), during and after (POST) the immobilization period, by the long-term averaging technique in ambulatory ECG Holter recordings [9]. Additionally, we proposed an adapted short-term averaging approach for shorter, non-stationary ECG signals obtained during two stress manoeuvres (head-up tilt-table and bicycle exercise tests). Both approaches are based on the multilead analysis used in the previous study. The absence of significant changes between PRE and POST-HDBR on TWA indices suggests that a long-term exposure to simulated microgravity is not enough to induce alterations in healthy myocardial substrate up to the point of reflecting electrical instability in terms of TWA on the ECG.
Finally, methodological advances were proposed for the assessment of ventricular repolarization instability from the ECG signal in the presence of sporadic (ventricular premature contractions, VPCs) and sustained (atrial fibrillation) rhythm disturbances.
On the one hand, a methodological improvement for the estimation of TWA amplitude in ambulatory ECG recordings was proposed, which deals with the possible phase reversal on the alternans sequence induced by the presence of VPCs [10]. The performance of the algorithm was first evaluated using synthetic signals. Then, the effect of the proposed method in the prognostic value of TWA amplitude was assessed in real ambulatory ECG recordings from patients with chronic heart failure (CHF). Finally, circadian TWA changes were evaluated as well as the prognostic value of TWA at different times of the day. A clinical study demonstrated the enhancement in the predictive value of the index of average alternans (IAA) [9] for SCD stratification. In addition, results suggested that alternans activity is modulated by the circadian pattern, preserving its prognostic information when computed just during the morning, which is also the day interval with the highest reported SCD incidence. Thus, suggesting that time of the day should be considered for SCD risk prediction.
On the other hand, the high irregularity of the ventricular response in atrial fibrillation (AF) limits the use of the most common ECG-derived markers of repolarization heterogeneity, including TWA, under this clinical condition [11]. A new method for assessing ventricular repolarization changes based on a selective averaging technique was developed and new non-invasive indices of repolarization variation were proposed. The positive impact in the prognostic value of the computed indices was demonstrated in a clinical study, by analyzing ECG Holter recordings from CHF patients with AF. To the best of our knowledge, this is the first study that attempts a non-invasive SCD stratification of patients under AF rhythm by assessing ventricular repolarization instability from the ECG signal.
To conclude, the research presented in this thesis sheds some light in the identification of pro-arrhythmic factors, which plays an important role in adopting efficient therapeutic strategies. In particular, the optimal configuration for real-time monitoring of repolarization alternans from intracardiac EGMs, together with the prognostic value of the proposed non-invasive indices of alternans activity and ventricular instability variations in case of AF rhythms demonstrated in two clinical studies, would increase the effectiveness of (ICD) therapy. Finally, the analysis of ECG signals recorded during HDBR experiments in structurally healthy hearts, also provides interesting information on cardiovascular alterations produced in immobilized or bedridden patients.
Resumen (otro idioma):
The main objective of this thesis is to propose methodological advances for the assessment of ventricular repolarization instability in pathological and abnormal conditions. These contributions are aimed at improving the prediction of ventricular arrhythmias and, consequently, better identifying SCD risk. In particular, we have addressed this objective by developing robust methodologies for the assessment of T-wave alternans (TWA) and ventricular repolarization instability, in invasive and non-invasive cardiac signals, that have been evaluated in both experimental and clinical conditions.
In the first part of the thesis, TWA was simultaneously characterized (prevalence, magnitude, time-course, and alternans waveform) in body-surface ECG and intracardiac electrograms (EGMs) signals during coronary artery occlusion. Signals from both body surface ECG and intracardiac EGMs recorded from 4 different anatomical heart locations (coronary sinus, epicardial space and left and right ventricles) were analyzed following a multilead strategy. Leads were linearly combined using the periodic component analysis (πCA) [5], which maximizes the 2-beat periodicity (TWA periodicity) content present on the available leads. Then the Laplacian Likelihood Ratio method (LLRM) [6] was applied for TWA detection and estimation. A sensitivity study for TWA detection from the 5 different locations of leads was performed, revealing that it is the combination of the ECG leads that better performs. In addition, this multilead approach allowed us to find the optimal combination of intracardiac leads usable for in-vivo monitorization of TWA directly from an implantable device, with a sensitivity comparable to the ECG analysis. These results encourage further research to determine the feasibility of predicting imminent VT/VF episodes by TWA analysis implemented in implantable cardioverter defibrillator’s (ICD) technology.
Then, we have studied the potential changes induced by a prolonged exposure to simulated microgravity on ventricular repolarization in structurally normal hearts. It is well known that this environmental condition affects the control of autonomic and cardiovascular systems [7], with a potential increase on cardiac electrical instability. The effects of short- (5 days), mid- (21 days) and long- (60 days) exposure to simulated microgravity on TWA using the head-down bed-rest (HDBR) model [8] were assessed. TWA was evaluated before (PRE), during and after (POST) the immobilization period, by the long-term averaging technique in ambulatory ECG Holter recordings [9]. Additionally, we proposed an adapted short-term averaging approach for shorter, non-stationary ECG signals obtained during two stress manoeuvres (head-up tilt-table and bicycle exercise tests). Both approaches are based on the multilead analysis used in the previous study. The absence of significant changes between PRE and POST-HDBR on TWA indices suggests that a long-term exposure to simulated microgravity is not enough to induce alterations in healthy myocardial substrate up to the point of reflecting electrical instability in terms of TWA on the ECG.
Finally, methodological advances were proposed for the assessment of ventricular repolarization instability from the ECG signal in the presence of sporadic (ventricular premature contractions, VPCs) and sustained (atrial fibrillation) rhythm disturbances.
On the one hand, a methodological improvement for the estimation of TWA amplitude in ambulatory ECG recordings was proposed, which deals with the possible phase reversal on the alternans sequence induced by the presence of VPCs [10]. The performance of the algorithm was first evaluated using synthetic signals. Then, the effect of the proposed method in the prognostic value of TWA amplitude was assessed in real ambulatory ECG recordings from patients with chronic heart failure (CHF). Finally, circadian TWA changes were evaluated as well as the prognostic value of TWA at different times of the day. A clinical study demonstrated the enhancement in the predictive value of the index of average alternans (IAA) [9] for SCD stratification. In addition, results suggested that alternans activity is modulated by the circadian pattern, preserving its prognostic information when computed just during the morning, which is also the day interval with the highest reported SCD incidence. Thus, suggesting that time of the day should be considered for SCD risk prediction.
On the other hand, the high irregularity of the ventricular response in atrial fibrillation (AF) limits the use of the most common ECG-derived markers of repolarization heterogeneity, including TWA, under this clinical condition [11]. A new method for assessing ventricular repolarization changes based on a selective averaging technique was developed and new non-invasive indices of repolarization variation were proposed. The positive impact in the prognostic value of the computed indices was demonstrated in a clinical study, by analyzing ECG Holter recordings from CHF patients with AF. To the best of our knowledge, this is the first study that attempts a non-invasive SCD stratification of patients under AF rhythm by assessing ventricular repolarization instability from the ECG signal.
To conclude, the research presented in this thesis sheds some light in the identification of pro-arrhythmic factors, which plays an important role in adopting efficient therapeutic strategies. In particular, the optimal configuration for real-time monitoring of repolarization alternans from intracardiac EGMs, together with the prognostic value of the proposed non-invasive indices of alternans activity and ventricular instability variations in case of AF rhythms demonstrated in two clinical studies, would increase the effectiveness of (ICD) therapy. Finally, the analysis of ECG signals recorded during HDBR experiments in structurally healthy hearts, also provides interesting information on cardiovascular alterations produced in immobilized or bedridden patients.
Resumen (otro idioma):
Titulación: Programa de Doctorado en Ingeniería Biomédica
Plan(es): Plan 510
Departamento: Instituto de Investigación en Ingeniería de Aragón (I3A)
Nota: Presentado: 17 11 2017
Nota: Tesis-Univ. Zaragoza, Instituto de Investigación en Ingeniería de Aragón (I3A), 2017
Enlace permanente:
Registro creado el 2018-02-28, última modificación el 2021-05-20
