Julia Ramírez

Automatic SVM classification of sudden cardiac death and pump failure death from autonomic and repolarization ECG markers.

BACKGROUND Considering the rates of sudden cardiac death (SCD) and pump failure death (PFD) in chronic heart failure (CHF) patients and the cost-effectiveness of their preventing treatments, identification of CHF patients at risk is an important challenge. In this work, we studied the prognostic performance of the combination of an index potentially related to dispersion of repolarization restitution (Δα), an index quantifying T-wave alternans (IAA) and the slope of heart rate turbulence (TS) for classification of SCD and PFD. METHODS Holter ECG recordings of 597 CHF patients with sinus rhythm enrolled in the MUSIC study were analyzed and Δα, IAA and TS were obtained. A strategy was implemented using support vector machines (SVM) to classify patients in three groups: SCD victims, PFD victims and other patients (the latter including survivors and victims of non-cardiac causes). Cross-validation was used to evaluate the performance of the implemented classifier. RESULTS Δα and IAA, dichotomized at 0.035 (dimensionless) and 3.73 μV, respectively, were the ECG markers most strongly associated with SCD, while TS, dichotomized at 2.5 ms/RR, was the index most strongly related to PFD. When separating SCD victims from the rest of patients, the individual marker with best performance was Δα≥0.035, which, for a fixed specificity (Sp) of 90%, showed a sensitivity (Se) value of 10%, while the combination of Δα and IAA increased Se to 18%. For separation of PFD victims from the rest of patients, the best individual marker was TS ≤ 2.5 ms/RR, which, for Sp=90%, showed a Se of 26%, this value being lower than Se=34%, produced by the combination of Δα and TS. Furthermore, when performing SVM classification into the three reported groups, the optimal combination of risk markers led to a maximum Sp of 79% (Se=18%) for SCD and Sp of 81% (Se=14%) for PFD. CONCLUSIONS The results shown in this work suggest that it is possible to efficiently discriminate SCD and PFD in a population of CHF patients using ECG-derived risk markers like Δα, TS and IAA.

T‐Wave Morphology Restitution Predicts Sudden Cardiac Death in Patients With Chronic Heart Failure

Background Patients with chronic heart failure are at high risk of sudden cardiac death (SCD). Increased dispersion of repolarization restitution has been associated with SCD, and we hypothesize that this should be reflected in the morphology of the T‐wave and its variations with heart rate. The aim of this study is to propose an electrocardiogram (ECG)‐based index characterizing T‐wave morphology restitution (TMR), and to assess its association with SCD risk in a population of chronic heart failure patients. Methods and Results Holter ECGs from 651 ambulatory patients with chronic heart failure from the MUSIC (MUerte Súbita en Insuficiencia Cardiaca) study were available for the analysis. TMR was quantified by measuring the morphological variation of the T‐wave per RR increment using time‐warping metrics, and its predictive power was compared to that of clinical variables such as the left ventricular ejection fraction and other ECG‐derived indices, such as T‐wave alternans and heart rate variability. TMR was significantly higher in SCD victims than in the rest of patients (median 0.046 versus 0.039, P<0.001). When TMR was dichotomized at TMR=0.040, the SCD rate was significantly higher in the TMR≥0.040 group (P<0.001). Cox analysis revealed that TMR≥0.040 was strongly associated with SCD, with a hazard ratio of 3.27 (P<0.001), independently of clinical and ECG‐derived variables. No association was found between TMR and pump failure death. Conclusions This study shows that TMR is specifically associated with SCD in a population of chronic heart failure patients, and it is a better predictor than clinical and ECG‐derived variables.

QT/RR and T-peak-to-end/RR curvatures and slopes in chronic heart failure: Relation to sudden cardiac death ☆

BACKGROUND Previous studies investigated the QT/RR relationship by linear regressions of QT and RR intervals. However, the pattern of the QT/RR relationship is not necessarily linear. This study investigated the QT/RR and T-peak-to-end (Tpe)/RR curvatures and corresponding slopes in chronic heart failure (CHF) patients, and studied their differences between sudden cardiac death (SCD) victims and others. METHODS Holter ECG recordings of 650 CHF patients were analyzed. RR, QT and Tpe series were obtained and for each patient, the data of each subject were fitted with a non-linear regression function of the form: QT=χ+ϕ(1-RR(γ)), where γ is the QT/RR curvature. The same regression formula was applied to the Tpe interval series. The slopes (dimensionless units) were calculated at the averaged RR intervals and at RR of 1 second. RESULTS The median (difference between 75th and 25th percentile) of the curvature parameter was 0.226 (2.39) for QT/RR and -0.002 (3.64) for Tpe/RR in the overall sample. For the QT/RR slope, these values were 0.170 (0.12) and 0.190 (0.10) when evaluated at RR=1 and at the averaged RR, respectively, while for the Tpe/RR slope the values were 0.016 (0.04) and 0.020 (0.04), respectively. The Tpe/RR slope showed high statistical significance for separation of SCD victims and others, particularly when evaluated at the averaged RR (median values of 0.040 vs 0.020, p=0.002), but also when evaluated at RR=1 second (0.026 vs 0.015, p=0.023). Patients with values of Tpe/RR slope above 0.042 had double incidence of SCD, for the case of the slope being evaluated at RR=1 second, and triple incidence for the case of the slope being evaluated at the averaged RR. The QT/RR slope and curvature, as well as the Tpe/RR curvature, were not different in SCD victims and in others. CONCLUSIONS Non-linear regression models based on curvature and slope characteristics, individually obtained for each patient, were used to characterize the QT/RR and Tpe/RR relationships. Steeper Tpe/RR slopes, obtained after adjusting for the curvature parameter, were associated with higher incidence of SCD. The curvature parameter itself did not show SCD predictive value.

Sudden cardiac death and pump failure death prediction in chronic heart failure by combining ECG and clinical markers in an integrated risk model

Background Sudden cardiac death (SCD) and pump failure death (PFD) are common endpoints in chronic heart failure (CHF) patients, but prevention strategies are different. Currently used tools to specifically predict these endpoints are limited. We developed risk models to specifically assess SCD and PFD risk in CHF by combining ECG markers and clinical variables. Methods The relation of clinical and ECG markers with SCD and PFD risk was assessed in 597 patients enrolled in the MUSIC (MUerte Súbita en Insuficiencia Cardiaca) study. ECG indices included: turbulence slope (TS), reflecting autonomic dysfunction; T-wave alternans (TWA), reflecting ventricular repolarization instability; and T-peak-to-end restitution (ΔαTpe) and T-wave morphology restitution (TMR), both reflecting changes in dispersion of repolarization due to heart rate changes. Standard clinical indices were also included. Results The indices with the greatest SCD prognostic impact were gender, New York Heart Association (NYHA) class, left ventricular ejection fraction, TWA, ΔαTpe and TMR. For PFD, the indices were diabetes, NYHA class, ΔαTpe and TS. Using a model with only clinical variables, the hazard ratios (HRs) for SCD and PFD for patients in the high-risk group (fifth quintile of risk score) with respect to patients in the low-risk group (first and second quintiles of risk score) were both greater than 4. HRs for SCD and PFD increased to 9 and 11 when using a model including only ECG markers, and to 14 and 13, when combining clinical and ECG markers. Conclusion The inclusion of ECG markers capturing complementary pro-arrhythmic and pump failure mechanisms into risk models based only on standard clinical variables substantially improves prediction of SCD and PFD in CHF patients.

Quantification of t-wave morphological variability using time-warping methods

The aim of this study is to quantify the variation of the T-wave morphology during a 24-hour electrocardiogram (ECG) recording. Two ECG-derived markers are presented to quantify T-wave morphological variability in the temporal, \(d_w\), and amplitude, \(d_a\), domains. Two additional markers, \(d_w^{NL}\) and \(d_a^{NL}\), that only capture the non-linear component of \(d_w\) and \(d_a\) are also proposed. The proposed markers are used to quantify T-wave time and amplitude variations in 500 24-hour ECG recordings from chronic heart failure patients. Additionally, two mean warped T-waves, used in the calculation of those markers, are proposed to compensate for the rate dependence of the T-wave morphology. Statistical analysis is used to evaluate the correlation between \(d_w\), \(d_w^{NL}\), \(d_a\) and \(d_a^{NL}\) and the maximum intra-subject RR range, \(\Delta \)RR. Results show that the mean warped T-wave is able to compensate for the morphological differences due to RR dynamics. Moreover, the metrics \(d_w\) and \(d_w^{NL}\) are correlated with \(\Delta \)RR, but \(d_a\) and \(d_a^{NL}\) are not. The proposed \(d_w\) and \(d_w^{NL}\) quantify variations in the temporal domain of the T-wave that are correlated with the RR range and, thus, could possibly reflect the variations of dispersion of repolarization due to changes in heart rate.