Publications

27

Journals

M Calvo, P Gomis, D Romero, V Le Rolle, N Behar, P Mabo, A Hernandez. Heart rate complexity analysis in Brugada syndrome during physical stress testing. Physiological Measurement 2017; (IF 2015 = 1.576, Q2); doi: 10.1088/1361-6579/aa513c

Symptoms such as ventricular arrhythmias in Brugada syndrome (BS) typically occur at rest, especially during sleep, suggesting that the autonomic nervous system (ANS) function may be relevant in the arrhythmogenesis of the disease. The aim of this work was to assess the ANS response captured by nonlinear heart rate variability (HRV) measures in 69 patients diagnosed with BS, who underwent a standardized physical stress test. Heart rate complexity (HRC) was evaluated by the power-law scaling analysis (β slope) during rest, exercise, recovery and rest post-recovery, in order to discriminate between symptomatic and asymptomatic BS patients. Symptomatic patients showed a significant reduction in HRC in comparison to asymptomatic subjects, after exertion (p = 0.015); during the whole recovery period (p = 0.023), and in particular within the passive recovery phase (p = 0.025), as well as during rest post-recovery (p = 0.022). Based on these results, symptoms could be associated with a lower ANS complexity during the stress test stages where parasympathetic activity is predominant. Therefore, the proposed HRV indicators could be of help in the risk stratification of asymptomatic patients.

F Bisbal, F Gómez-Pulido, P Cabanas-Grandío, N Akoum, M Calvo, D Andreu, S Prat-González, R J Perea, R Villuendas, A Berruezo, M sitges, A Bayés-Genís, J Brugada, N F Marrouche, L Mont. Left Atrial Geometry Improves Risk Prediction of Thromboembolic Events in Patients With Atrial Fibrillation. Journal of Cardiovascular Electrophysiology 2016; (IF 2014 = 3.156, Q2); doi: 10.1111/jce.12978

Background: Left atrial (LA) sphericity (LASP) is a new remodeling parameter based on LA shape analysis, with independent predictive value for recurrence after atrial fibrillation (AF) ablation.
Objectives: To evaluate the association between LASP and thromboembolic events (TE) in patients with AF.
Methods: Twenty-nine AF patients and prior TE and 29 age- and gender-matched controls were included. LASP was calculated using a 3D-LA reconstruction. The LA appendage (LAA) volume and morphology were assessed. ROC curve analysis was performed for LASP, LA volume, LAA volume, and CHAD/CHA2D–VASc scores (Stroke2—the grouping variable—was excluded).
Results: Mean age of the study population was 61 ± 11 years (79.3% males, 53.4% hypertension, 8.6% diabetes). Patients with prior TE had higher LASP than those without (82.5 ± 3.3% vs. 80.2 ± 3.1%, P = 0.008); there were no differences in CHAD or CHA2D–VASc scores, LA volume, LAA volume, or LAA morphology. The C-statistic was higher for LASP (0.71) than for other tested variables (CHAD score = 0.58, CHA2D–VASc score = 0.59, LA volume = 0.50, LAA volume = 0.46; P < 0.01 for all vs. LASP). The best cutoff value for LASP was 83.6% (sensitivity 0.52, specificity 0.90). Logistic regression analysis showed predictive value for LASP (OR 1.26 per each 1% increase [1.85–52.20], P = 0.013), but not for clinical risk scores. The addition of LASP to the CHAD and CHA2D–VASc scores increased the predictive value over the risk scores alone (P = 0.004), and reclassified 45.5% of patients with CHAD = 0 (no anticoagulation indicated) to moderate-risk (anticoagulation indicated). Conclusion: LA sphericity is associated with prior TE in AF patients and improves the performance of the CHAD and CHA2D–VASc scores alone.

D Andreu*, F Gomez-Pulido*, M Calvo, A Carlosena-Remírez, F Bisbal, R Borràs, E Benito, E Guasch, S Prat-Gonzalez, R J Perea, J Brugada, A Berruezo, L Mont. Contact Force Threshold for Permanent Lesion Formation in Atrial Fibrillation Ablation: A Cardiac Magnetic Resonance-based Study to Detect Ablation Gaps. Heart Rhythm 2015 (in press, accepted manuscript); (IF 2014 = 5.076, Q1); doi: http://dx.doi.org/10.1016/j.hrthm.2015.08.010

Background: Catheter contact force (CF) has a strong correlation with lesion formation during radiofrequency ablation. Delayed-enhancement cardiac magnetic resonance (DE-CMR) provides lesion information in patients with a prior atrial fibrillation (AF) ablation.

Objective: The aim of this study was to determine the CF threshold to create permanent lesions detected by DE-CMR.

Methods: A total of 36 patients referred for AF ablation were included. A CF catheter was used during the ablation procedure and DE-CMR was performed at three months post-ablation. Eighteen pulmonary vein (PV) segments were defined and 3D reconstructions of the left atrium (LA) derived from the DE-CMR images were obtained. One observer evaluated the presence of any discontinuity of previous ablation lesions (gap) in the 3D LA reconstructions and another observer (blinded to the gap findings) determined the minimum CF value in each PV segment.

Results: The PV segments where a gap was observed had a lower maximal CF than the segments without gap in the 3D LA reconstructions (6.7±4.4 g vs 12.2±4.7 g; p<0.001). In ROC analysis, a CF threshold >8 g provided 73% sensitivity and 81% specificity in the prediction of a complete PV lesion (Positive Predictive Value [PPV] =84%). A CF threshold >12 g had a specificity of 94% and increased the PPV to 91% in creating a complete lesion in the LA wall (area under the curve=0.834).

Conclusion: A CF threshold >12 g predicts a complete lesion with high specificity and PPV when a dragging ablation strategy is used in AF ablation.

Bisbal F, Calvo M, Trucco E, Arbelo E. Left Atrial Tachycardia after atrial fibrillation: can MRI assist the ablation? Canadian Journal of Cardiology 2014; (IF 2013 = 3.940, Q1); doi: 10.1016/j.cjca.2014.10.002

We present a case of a MRI-assisted ablation of an atrial tachycardia in a patient with prior AF ablation. A 3D reconstruction of the delayed-enhanced CMR (CMR model) was created to identify prior ablation lesions and gaps. Multiple gaps around right-sided PVs were observed. The activation map identified a dual-loop reentry around both right-sided PVs, confirming that the substrate identified by DE-CMR was critical to sustain the tachycardia. Ablation at this site converted to sinus rhythm. The present case shows the usefulness of DE-CMR substrate characterization to complement activation mapping of the tachycardia and more accurately define the anatomical circuit.

URL: www.onlinecjc.ca

Alcaine A, Soto-Iglesias D, Calvo M, Guiu E, Andreu D, Fernández-Armenta J, Berruezo A, Laguna P, Camara O, Martínez J P. A Wavelet-Based Electrogram Onset Delineator for Automatic Ventricular Activation Mapping. IEEE Transactions on Biomedical Engineering 2014, 61 (12), 2830 - 2839; (IF 2013 = 2.233, Q2); doi: 10.1109/TBME.2014.2330847

Electroanatomical mapping (EAM) systems are commonly used in clinical practice for guiding catheter ablation treatments of common arrhythmias. In focal tachycardias, the ablation target is defined by locating the earliest activation area determined by the joint analysis of electrogram (EGM) signals at different sites. However, this is currently a manual timeconsuming and experience-dependent task performed during the intervention and thus prone to stress-related errors. In this paper we present an automatic delineation strategy that combines electrocardiogram (ECG) information with the wavelet decomposition of the EGM signal envelope to identify the onset of each EGM signal for activation mapping. Fourteen electroanatomical maps corresponding to 10 patients suffering from non-tolerated premature ventricular contraction (PVC) beats and admitted for ablation procedure were used for evaluation. We compared the results obtained automatically with two types of manual annotations: one during the intervention by an expert technician (on-procedure) and other after the intervention (offprocedure), free from time and procedural constraints, by two other technicians. The automatic annotations show a significant correlation (0.95, p < 0.01) with the evaluation reference (offprocedure annotation sets combination) and has an error of 2.1 ± 10.9 ms, around the order of magnitude of the onprocedure annotations error (-2.6 ± 6.8 ms). The results suggest that the proposed methodology could be incorporated into EAM systems to considerably reduce processing time during ablation interventions. URL: ieeexplore.org

Calvo M, Subirats L, Ceccaroni L, Maroto J M, de Pablo C, Miralles F. Automatic assessment of socioeconomic impact on cardiac rehabilitation. International Journal of Environmental Research and Public Health 2013, 10, 5266-5283; (IF 2013 = 1.993, Q2); doi:10.3390/ijerph10115266

Disability-Adjusted Life Years (DALYs) and Quality-Adjusted Life Years (QALYs), which capture life expectancy and quality of the remaining life-years, are applied in a new method to measure socioeconomic impacts related to health. A 7-step methodology estimating the impact of health interventions based on DALYs, QALYs and functioning changes is presented. It relates the latter (1) to the EQ-5D-5L questionnaire (2) to automatically calculate the health status before and after the intervention (3). This change of status is represented as a change in quality of life when calculating QALYs gained due to the intervention (4). In order to make an economic assessment, QALYs gained are converted to DALYs averted (5). Then, by inferring the cost/DALY from the cost associated to the disability in terms of DALYs lost (6) and taking into account the cost of the action, cost savings due to the intervention are calculated (7) as an objective measure of socioeconomic impact. The methodology is implemented in Java. Cases within the framework of cardiac rehabilitation processes are analyzed and the calculations are based on 200 patients who underwent different cardiac-rehabilitation processes. Results show that these interventions result, on average, in a gain in QALYs of 0.6 and a cost savings of 8,000 €.

URL: www.mdpi.com/1660-4601/10/11/5266

Book chapters

M Calvo, V Le Rolle, D Romero, N Béhar, P Gomis, P Mabo, A Hernandez. Chapter 7b: Gender differences in the autonomic response to exercise testing in Brugada syndrome. Sex-specific analysis of the cardiovascular function. Springer Publishing 2017 (accepted)

Introduction: Cardiac events in Brugada syndrome (BS) typically occur at rest and mainly during sleep, suggesting that changes in the autonomic modulation play an important role in
the arrhythmogenesis of the disease. Moreover, gender differences in clinical manifestations of BS have been reported, identifying male patients with worse prognosis. The aim of our work was to assess and compare, according to gender, the autonomic response to exercise in a clinical series including 105 BS patients.

Method: The standard 12-lead ECG recordings were collected during a physical stress test divided in four phases: warm-up, incremental exercise, active and passive recovery. Spectral
non-stationary heart rate variability (HRV) indicators were extracted by means of a Smoothed Pseudo Wigner Ville distribution (SPWVD) approach that adapts frequency bands to respiratory information. These indicators were then averaged in non-overlapped windows of 1 minute for each patient, in order to compare groups at each minute of the physical stress test.

Results: From the last minute of warm-up and until the third minute of incremental exercise, asymptomatic male patients presented significantly higher LF values than asymptomatic females. Conversely, asymptomatic women showed an increased vagal modulation during the first minutes of incremental exercise. However, no significant differences were observed between symptomatic male
and female patients.

Conclusion: As previously reported on healthy subjects, enhanced parasympathetic and decreased sympathetic tones appear to be not only greater in women, but also defensive
during cardiac stress. Based on the results, asymptomatic patients presented the same gender tendencies. However, we observed that symptomatic males developed a more female-like autonomic modulation, probably related to a more protective autonomic response to exercise. These results could be a step forward towards the understanding of the autonomic function in BS, with a potential impact on risk stratification.

International conferences

Calvo M, Le Rolle V, Romero D, Béhar N, Gomis P, Mabo P, Hernández A. Time-frequency analysis of the autonomic response to head-up tilt testing in Brugada syndrome. September 24th - 27th, Rennes, France 2017. Computing in Cardiology Conference (CinC).

Cardiac events in Brugada syndrome (BS) usually occur at rest and mainly at nighttime, commonly relating changes in autonomic modulation to arrhythmogenesis. A major challenge today is to differentiate symptomatic and asymptomatic patients in order to optimize their therapy. Since autonomic assessment can be improved by the application of standard maneuveurs, this study analyzes and compares the autonomic response to head-up tilt (HUT) testing on 20 symptomatic and 45 asymptomatic BS patients. Their standard 12-lead ECG recordings were collected during a HUT test composed of the following three phases: (1)
supine rest for 10 minutes, (2) tilting with an inclination of 60º for 45 minutes and (3) supine rest for 10 minutes. Spectral non-stationary heart rate variability indicators were extracted by means of a Smoothed Pseudo Wigner Ville distribution approach that adapts frequency bands to respiratory information. Symptomatic patients showed an increased parasympathetic tone and a reduced sympathovagal balance with respect to asymptomatic patients during tilting, and particularly before the 15th minute of the
test. The results provide new insights into the autonomic function in BS, with a potential impact on the use of timereduced
HUT testing for risk stratification.

Calvo M, Le Rolle V, Romero D, Béhar N, Gomis P, Mabo P, Hernández A. Recursive model identification for the evaluation of the autonomic response to exercise in Brugada syndrome. September 6th - 8th, Compiègne, France 2017. International Conference on Computational Bioengineering (ICCB).

Brugada syndrome (BS) is a genetic disorder characterized by a distinctive electrocardiographic pattern, associated with an increased risk for sudden cardiac death (SCD) due to malignant ventricular arrhythmias. The autonomic nervous system (ANS) plays a relevant role in the pathophysiology, arrhythmogenesis and prognosis of the disease. Indeed, ventricular arrhythmias in this population tend to occur at rest and mainly at nighttime, thus being commonly related to an augmented parasympathetic tone. Furthermore, many studies have revealed an anomalous sympathetic function in BS. However, despite this connection between BS prognosis with autonomic imbalance, it remains to determine which autonomic tests and indicators are the most appropriate so as to identify those BS patients at high risk for SCD.

The cardiovascular response to exertion has been extensively studied through clinical trials and computational models. Exercise causes a sympathetic activity increase and a parasympathetic inhibition, leading to higher heart rates (HR). Conversely, post-exercise cardiodeceleration is regulated by an increase in parasympathetic activity, as well as a gradual sympathetic withdrawal. Since classical HR analysis at rest frequently fails to identify patients at risk, we hypothesize that the evaluation of HR modulation during physical stress testing may provide additional information for risk stratification in this population. Moreover, although classical markers are widely used in clinical practice to estimate sympathetic and parasympathetic levels, since computational models directly represent interactions between the ANS and the cardiovascular system (CVS), a model-based approach could provide useful knowledge to support autonomic response interpretation.

Therefore, a recursive parameter identification method is proposed and applied to a closed-loop mathematical model of the baroreflex and cardiovascular systems, in order to estimate the evolution of sympathetic and parasympathetic contributions to HR modulation during exercise. The model was evaluated with data from a BS patient, acquired during a physical stress test. The results show a close match between experimental and simulated signals (mean error = 0.85%) and the estimations of sympathetic and parasympathetic components were consistent with physiological knowledge.

In this work, we propose an original method to capture ANS dynamics in response to exertion. It is based on a recursively identified closed-loop model of the baroreflex and the cardiovascular systems, introducing: i) a subject-specific model parameter identification and ii) an estimation of the time-varying sympathetic and parasympathetic activities, by the application of a recursive evolutionary algorithm. Results demonstrate the feasibility of the model to reproduce realistic autonomic responses to non-stationary physiological conditions, such as exercise. This approach unmasks indicators capturing cardiovascular and autonomic dynamics never before studied in BS, in order to better understand the underlying mechanisms of the ANS in response to exercise that can be useful for risk stratification in this population. Current developments are focused on the application of the proposed method to a large population of BS patients with different levels of risk for SCD.

Calvo M, Hernandez J, Vidorreta S, Brugada J, Gomis P, Arbelo E. Automatic Brugada pattern detection on continuous electrocardiographic monitoring. June 18th - 21st, Vienna, Austria 2017. Cardiostim 2017 - EHRA Europace.

Background: As described in the literature, Brugada syndrome (BrS) patients presenting a spontaneous type 1 electrocardiographic (ECG) pattern are related to a greater arrhythmic risk compared to those with a flecainide-induced type 1 ECG pattern. The 24-hour continuous ECG provides relevant cardiac information. However, it is usually challenging and time-consuming to analyze.

Purpose: This work presents an automatic algorithm able to detect significant ECG parameters in the disease, such as Brugada-like pattern burden on ECG monitoring.

Methods: The algorithm integrates a previously developed and validated ECG delineator in order to locate the beginning, peak and ending of cardiac waves. Then, it classifies each beat between normal or Brugada-like, based on the morphology of the detected waves (see table below). Likewise, the algorithm includes the detection of other relevant ECG features: ST-segment elevation, PR, QRS and QTc intervals duration, heart rate variability and number of premature ventricular contractions, among others. Validation of the Brugada-pattern detector was done by assessment of the beat-to-beat correlation of the 24-hour 12-lead ECG signals in 2 BrS patients with spontaneous type 1 ECG and 1 healthy individual, 2 blinded arrhythmia specialists.

Results: From a random sample of 212 beats, type 1 ECG pattern was correctly detected in the 98.11% of beats in V1, 84.91% in V2, 92.45% in aVR and 100% in aVF. Erroneous detections mainly resulted from incorrect T-wave delineations in noisy segments. Regarding type 2 ECG detections, all analyzed beats in V1 and the 92.45% in V2 were properly classified.

Conclusion: This study shows the feasibility of the algorithm to automatically detect Brugada-like beats on a 24-hour continuous ECG monitoring. Future work will be focused on testing the performed detector in an extensive sample of patients diagnosed with BrS that, based on their cardiac events, will allow to determine the association of this ECG parameter burden with the individual risk of suffering sudden cardiac death.

Romero D, Calvo M, Béhar N, Mabo P, Hernández A. Ensemble classifier based on linear discriminant analysis for distinguishing Brugada syndrome patients according to symptomatology. September 11th - 14th, Vancouver, Canada 2016. Computing in Cardiology Conference (CinC).

Identifying high-risk patients requiring an ICD among asymptomatic Brugada patients is nowadays a bit challenging. In this study, 62 patients suffering from Brugada syndrome (14 symptomatic) were studied by analyzing the 12-lead ECG recording acquired during a physical exercise test. For each patient, conventional HRV indices from time-frequency analysis and heart rate recovery (HRV features), as well as several morphological depolarization indices (QRS features), were evaluated at relevant periods of the test. Most discriminant features from both the HRV and QRS sets were selected using a two-stage feature selection algorithm and used for model classification building. For the detection step, an ensemble classifier using stacking approach plus a fixed combiner was designed, using linear discriminant analysis as the base classification algorithm. Best features from each model were then used for building the final individual and combined classification models. Detection performance using the symptomatic group as the target class, was as follows: HRV-based model: Se=1, Sp=0.67, AUC=0.87; QRS-based model: Se=75, Sp=0.67 AUC=0.73. When joining best features of both models (HRV-QRS-based model), the performance increased up to Se=1, Sp=0.83, AUC=0.90. The study showed that by combining both HRV and depolarization analysis, a better risk stratification can be performed. This could be useful for the identification of Brugada patients with previous symptoms, and it may help to the decision making process of asymptomatic patients needing an ICD.

Calvo M, Le Rolle V, Romero D, Béhar N, Gomis P, Mabo P, Hernández A. Analysis of a Cardiovascular Model for the Study of the Autonomic Response of Brugada syndrome patients . August 16th - 20th, Orlando, USA 2016. 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

This paper describes the integration of mathematical models of the cardiac electrical activity, the cardiovascular system and the baroreceptor reflex control of the autonomic nervous system, with a model representing a head-up tilt test. Sensitivity analyses are performed in order to determine those model parameters producing the greatest significant effects on heart rate and blood pressure. An optimization step is then applied to the most influential parameters in order to find the best model fit to real cardiac data obtained from a patient suffering from Brugada syndrome and a healthy subject, in supine and upright postures during a tilt test. Results show the feasibility of the model to generate realistic autonomic responses to head-up tilt testing in health and disease.

Calvo M, Gomis P, Hernández A, Andreu D, Arbelo E, Caminal P. Automatic detection of Brugada-­like pattern on continuous electrocardiographic monitoring. September 14th-16th, Barcelona, Spain 2015. International Conference on Computational Bioengineering (ICCB).

Brugada syndrome is an inherited disease characterized by a specific electrocardiographic pattern as well as an increased risk of cardiac arrhythmias and sudden death. There are currently no objective indicators to determine individual risk, so the eligibility for the implantation of implantable cardioverter defibrillators to prevent sudden death events still generates controversy.

Since the electrocardiogram (ECG) provides with noninvasively heart information, this work presents an automatic algorithm able to detect significant electrocardiographic parameters in the disease, such as the Brugada-like pattern burden on 24-hour continuous ECG monitoring. The algorithm integrates a previously developed and validated ECG delineator in order to locate the beginning, peak and ending of cardiac waves. Then, it classifies each beat between normal or Brugada, based on the morphology of the detected waves. Likewise, the algorithm includes the detection of other relevant ECG features: ST-segment elevation, PR, QRS and QTc intervals duration, heart rate variability and premature ventricular contractions load, among others.

The detector validation was based on the assessment of 24-hour 12-lead ECG signals of a high-risk Brugada patient from the Hospital Clínic de Barcelona, in Spain. The subject carried the Holter monitor SEER 12 (GE Healthcare Inc., Milwaukee, WI, USA) for a whole day, placing the electrodes in a regular Mason-Likar 12-lead ECG system configuration and recording at a sampling frequency of 128 Hz. The obtained sensitivity and specificity of a random sample of 600 beats across the 12 leads (50 beats/lead) were of 93.7% and 98.69%, respectively. Erroneous detections resulted from incorrect T-wave delineations in noisy segments.

Future work will be focused on testing the performed algorithm in an extensive sample of patients diagnosed with Brugada syndrome that, based on their cardiac events, will automatically determine the association of these ECG parameters with the individual risk of suffering sudden cardiac death.

Calvo M, Le Rolle V, Romero D, Béhar N, Gomis P, Mabo P, Hernández A. Comparison of methods to measure baroreflex sensitivity in Brugada syndrome. September 6th-9th, Nice, France 2015. Computing in Cardiology Conference (CinC).

Baroreflex sensitivity (BRS) is a valuable index to assess cardiovascular autonomic control and therefore to provide prognostic evaluation in many cardiac diseases. Although several methods have been developed to noninvasively capture spontaneous BRS, they are difficult to compare and often provide conflicting results. In this study we analyzed different BRS estimates obtained from a clinical series of thirty-two patients diagnosed with Brugada syndrome. They took part in a standardized head-up tilt test in order to quantify the level of agreement between the following measures: (1) BRS+/+ and (2) BRS-/- from sequence analysis, (3) BRS-LF, (4) BRS-HF, (5) BRS-LHF, (6) BRS-TF and (7) BRS-TFLF from cross-spectral analysis and (8) BRS-SD. Measures resulting from the sequence method failed to provide results in many recordings and showed low agreement with the remaining methods based on intraclass correlation coefficient results. Assuming an age-associated reduction in BRS, which has been previously reported, the lowest Pearson’s correlation coefficients came from sequence analysis results. Thus, among the analyzed BRS estimates, those derived from sequence analysis showed a lower reliability when capturing baroreflex function in Brugada syndrome patients.

Andreu D, Gomez-Pulido F, Calvo M, Benito E, Carlosena-Remirez A, Arbelo E, Tolosana JM, Berruezo A, Brugada J, Mont L. Usefulness of contact force maps to predict ablation gaps in catheter ablation of atrial fibrillation. Europace 2015, 17 (suppl3):iii105. 21st - 24th June, Milan, Italy 2015. Cardiostim.

Introduction: Catheter contact force (CF) is strongly correlated with lesion formation during radiofrequency ablation. Late-gadolinium enhancement cardiac magnetic resonance (LGE-CMR) provides lesion information in patients with a prior atrial fibrillation (AF) ablation. The aim of this study is to determine if CF maps can predict the development of lesion gaps after an ablation procedure.

Methods: A total of 36 consecutive patients referred for AF ablation were included. A CF catheter was used in the procedure and a dragging ablation strategy was performed. Three months postablation, LGE-CMR was performed. Eighteen pulmonary vein (PV) segments were defined for both left and right PVs, and 3D reconstructions of the left atrium (LA) derived from the LGE-CMR images were obtained. One observer evaluated the presence of gaps in the 3D LA reconstructions and a blinded second observer determined the minimum CF in each PV segment of the CF map.

Results: A total of 638 PV segments were analyzed. The PV segments where a gap was observed in the 3D LA reconstructions had a lower CF than the segments without gaps (6.7±4.4g vs 12.2±4.7g; p<0.001). The ROC curve analysis showed that a CF threshold >8 g provided a sensitivity of 72.9% and a specificity of 80.4% in the prediction of a complete PV lesion. A CF threshold >12 g had a probability of 94.0% to create a complete lesion in the LA wall (area under the curve=0.834).

Conclusion: A CF >12 g had a high specificity of predicting a complete lesion when using a dragging ablation strategy in AF ablation.

Alcaine A, Soto-Iglesias D, Calvo M, Guiu E, Andreu D, Fernández-Armenta J, Laguna P, Cámara O, Martínez J P, Berruezo A. Evaluation of an automatic delineation algorithm for activation mapping of focal ventricular tachycardias. August 30th – September 3rd, Barcelona, Spain 2014. European Society of Cardiology Congress.

Purpose: Activation mapping (AM) based on electrogram (EGM) analysis is used to guide ablation treatments of focal tachycardias. However, AM based on EGM activation onset is a manual, observer-dependent and time-consuming task within the EP lab. We propose and evaluate an automatic EGM onset detection algorithm for AM.

Methods: Clinical data correspond to 8 patients (9 electroanatomical maps, a total of 1673 mapping points) admitted for ablation procedure due to non-tolerated premature ventricular contraction (PVC) beats. During the intervention, for each mapping point, a 2.5 s EGM segment was recorded including a PVC beat and the bipolar EGM activation onset was manually marked (on-procedure annotation). After the intervention, an automatic EGM activation onset algorithm was applied and electroanatomical maps were reconstructed using the automatically measured local activation times (LATs). Additionally, two different experts manually blind-annotated those maps after the intervention (off-procedure annotations).

Results: Considering the average value of the two off-procedure annotation sets as the reference for evaluation, we measured the error (mean ± standard devia- tion) of LATs computation committed by the automatic algorithm. For comparison purposes, the error is also computed for the on-procedure annotations. The errors were, for automatic method: 2.1±10.9 ms, and for on-procedure annotations: -2.6±6.8 ms. The off-procedure inter-expert difference is -0.8±4.5 ms. The figure shows a representative example in RAO view obtained with the 4 sets of annotations.

Conclusions: Automatic EGM activation onset detection during clinical routine allows an observer-independent, accurate estimate of LATs which could be ex- tended to AM in high-density catheter configurations.

Bisbal F, Gómez F, Akoum N, Calvo M, Cabanas P, Vidal B, Brugada J, Marrouche N, Mont L. Left Atrial Sphericity improves CHADS2 score stroke prediction in patients with atrial fibrillation. Europace 2014, 16 (suppl2):ii89. 18th - 21st June, Nice, France 2014. Cardiostim.

Introduction: Left atrial sphericity (LASP) is a new remodeling parameter that has independent predictive value for atrial fibrillation (AF) recurrence after AF ablation. We sought to evaluate whether LASP adds prognostic information to CHADS2 score and LA appendage (LAA) characteristics.

Methods: Twenty-nine patients with history of prior stroke and 29 age- and gender-matched controls were included. All patients underwent cardiac MRI prior to the AF ablation procedure. LASP was calculated using a 3D left atrial (LA) reconstruction that excluded pulmonary veins and the LAA. Manual LAA segmentation was used to calculate the volume. LAA morphology was classified as previously reported: chicken wing, cauliflower, windsock, and cactus. Area under the ROC curve (AUC) was calculated for LASP, LA volume, LAA volume and CHAD score (Stroke2 excluded). A cut-off value was determined for optimal stroke prediction.

Results: Mean age of the study population was 61±11 years, 79.3% were male, 53.4% had hypertension, and 8.6% had diabetes. Compared to controls, patients with history of prior stroke had significantly higher LASP (80.2±3.1 vs 82.5±3.3, p=0.008); there were no differences in CHAD score (0.66±0.76 vs 0.90±0.86, p=0.26), LA volume (95±31 vs 94±25ml, p=0.97), LAA volume (5.9±2.9 vs 5.5±2.9, p=0.66), or LAA morphology (p=0.514). LASP had the only significant ROC curve for stroke prediction (AUC 0.706 [0.571-0.842], p=0.007). The CHAD score had an AUC of 0.58 (0.432-0.728, p=0.297), LA volume of 0.497 (0.345-0.649, p=969), and LAA volume of 0.463 (0.312-0.613, p=0.624). The best cutoff value for LASP was 83.6% (52% sensitivity, 90% specificity). A significantly greater proportion of patients with high LA sphericity (>83.6%) had prior stroke (83.3% vs 35.0% below the cut-off; P=0.001). Logistic regression showed predictive value for LASP (OR 1.26 per each 1% increase [1.85-52.20], P=0.013), but not for CHAD score (OR 1.457 [0.755-2.81]; p=0.262). The combination of both parameters (CHAD-LASP) increased the predictive value over CHAD score alone (AUC of 0.719 [0.585-0.852], p=0.004).

Conclusion: LA sphericity is associated with history of prior stroke in patients undergoing AF ablation and increases the predictive value over CHAD score alone.

Cabanas P, Bisbal F, Gómez-Pulido F, Guíu E, Calvo M, Berruezo A, Prat S, Perea RJ, Brugada J, Mont L. MRI characterization of cryoballoon and radiofrequency ablation lesions after pulmonary vein isolation. Europace 2014, 16 (suppl2):ii141. 18th - 21st June, Nice, France 2014. Cardiostim.

Introduction: Cryoballoon (CA) and radiofrequency (RF) ablation are effective techniques to treat patients with paroxysmal atrial fibrillation (AF). Delayed-enhanced cardiac magnetic resonance (De-CMR) allows the identification of ablation lesions (scar) and gaps. We aimed to compare the DE-CMR characteristics of the ablation lesions and gaps between both CA and RF ablation in patients with paroxysmal AF, and evaluate the clinical implications (recurrence rate).

Methods: We prospectively included 23 patients with paroxysmal AF undergoing a first ablation procedure (9 RF, 14 CA). A DE-CMR was performed 6±3 months postablation. The endocardium and epicardium of the left atrium (LA) were manually segmented on axial plane slices. A 3D volume-rendered LA reconstruction was created using the segmented DE-CMR data. A pixel signal intensity map was projected on the 3D reconstruction and color-coded to display healthy areas and scar. The gap size was defined as the proportion (%) between the gap length and the total perimeter of the PV antrum. To evaluate the gap location, each group of ipsilateral PVs was divided into 4 quadrants (anterior, posterior, carina and superior or inferior).

Results: We analyzed a total of 92 PVs (36 RF, 56 CA). There were gaps in 80.6% and 91.1% of PVs in RF and CA group, respectively. The gap size was 31% and 41% in the RF and CA group, respectively (p=0.217). The most common location of the gaps differed between groups: the inferior quadrant of the right inferior PV in the RF group (78% vs. 86% in CA group, p=0.624) and the superior quadrant of the left superior PV in the CA group (93% vs. 44% in the RF group, p= 0.01). After 8 (6-12) months follow-up, 33% and 64% in the RF and CA group had recurrences, respectively (p=0.214).

Conclusion: Anatomic gaps on DE-CMR were common after RF and CA ablation. There was a trend toward a higher recurrence rate in the CA group compared to RF group; this finding might be related to a bigger gap size in the CA group or the differences in the location of the gaps between techniques.

Cabanas P, Bisbal F, Gómez-Pulido F, Guiu E, Calvo M, Berruezo A, Prat S, Perea RJ, Brugada J, Mont L. MRI characterization of cryoballoon ablation lesions: predicting recurrences after pulmonary vein isolation. Europace 2014, 16 (suppl2):ii111. 18th - 21st June, Nice, France 2014. Cardiostim.

Introduction: Cryoablation is a safe and effective technique for pulmonary vein isolation (PVI). Delayed-enhanced cardiac magnetic resonance (DE-CMR) can detect scar and gaps after ablation. The aim was to characterize the scar and gaps of prior ablation lesions on DE-CMR after cryoablation, as well as evaluate the differences between patients with and without AF recurrence.

Methods: Sixteen patients (54±10 years, 13 paroxysmal AF, 9 with AF recurrence) referred for PVI with cryoballoon (Artic Front and Artic Front Advance 28 mm, Medtronic®) were included. A post-procedural DE-CMR was performed after 7.2 ± 4.1 months. The endocardium and epicardium of the left atrium (LA) were manually segmented on axial plane slices. A 3D volume-rendered LA reconstruction was created using the segmented DE-CMR data. A pixel signal intensity map was projected on the 3D reconstruction and color-coded to display healthy areas and scar. The gap size was defined as the proportion (%) between the gap length and the total perimeter of the PV antrum. A ROC curve analysis was performed to determine the optimal cut-off value of gap size to predict recurrences.

Results: A total of 64 PV were analyzed. There were no differences in the number of cryoballoon applications or occlusion grade between patients with and without recurrences. All patients had gaps ≥1PV and 12 patients had gaps in all PVs. Compared to patients without recurrences, the patients with recurrences had gaps in all PVs more often (43% vs. 100%, p= 0.019) and of bigger size (29% vs. 51%, p= 0.03). The cut-off of gap size to predict recurrences was 45% (Se 67%, Sp 86%; AUC 0.93).

Conclusion: Anatomic gaps around PVs detected by DE-CMR were common after AF cryoablation. Patients with AF recurrence had more gaps and of bigger size than patients without recurrence.

Cabanas P, Bisbal F, Gomez Pulido F, Guiu E, Calvo M, Berruezo A, Prat S, Perea R J, Brugada J, Mont L. MRI characterization of cryoballoon ablation lesions predicting recurrences after pulmonary vein isolation. Heart Rhythm, 11(5), 7th - 10th May, San Francisco, USA 2014. Heart Rhythm Society Congress.

Introduction: Cryoablation is a safe and effective technique for pulmonary vein isolation (PVI). Delayed-enhanced cardiac magnetic resonance (DE-CMR) can detect scar and gaps after ablation. The aim was to characterize the scar and gaps of prior ablation lesions on DE-CMR after cryoablation, as well as evaluate the differences between patients with and without AF recurrence.

Methods: Sixteen patients (54±10 years, 13 paroxysmal AF, 9 with AF recurrence) referred for PVI with cryoballoon (Artic Front and Artic Front Advance 28 mm, Medtronic®) were included. A post-procedural DE-CMR was performed after 7.2 ± 4.1 months. The endocardium and epicardium of the left atrium (LA) were manually segmented on axial plane slices. A 3D volume-rendered LA reconstruction was created using the segmented DE-CMR data. A pixel signal intensity map was projected on the 3D reconstruction and color-coded to display healthy areas and scar. The gap size was defined as the proportion (%) between the gap length and the total perimeter of the PV antrum. A ROC curve analysis was performed to determine the optimal cut-off value of gap size to predict recurrences.

Results: A total of 64 PV were analyzed. There were no differences in the number of cryoballoon applications or occlusion grade between patients with and without recurrences. All patients had gaps ≥1PV and 12 patients had gaps in all PVs. Compared to patients without recurrences, the patients with recurrences had gaps in all PVs more often (43% vs. 100%, p= 0.019) and of bigger size (29% vs. 51%, p= 0.03). The cut-off of gap size to predict recurrences was 45% (Se 67%, Sp 86%; AUC 0.93).

Conclusion: Anatomic gaps around PVs detected by DE-CMR were common after AF cryoablation. Patients with AF recurrence had more gaps and of bigger size than patients without recurrence.

National conferences

Calvo M, Le Rolle V, Romero D, Béhar N, Gomis P, Mabo P, Hernández A. Model-based approach for the study of the autonomic response in Brugada syndrome patients. June 26th - 29th, Poitiers, France 2017. Colloque de la Société Francophone de Biologie Théorique (SFBT)

Brugada syndrome (BS) is a genetic disorder characterized by a distinctive electrocardiographic pattern, associated with an elevated risk for sudden cardiac death due to malignant ventricular arrythmias in patients having structurally normal hearts. The complexity and multifactorial nature of its prognosis turns risk stratification into a major challenge nowadays. However, since autonomic nervous system (ANS) modulation plays a significant role in the disease, the application of specific maneuvers stimulating the ANS, such as the head-up tilt (HUT) test, could improve autonomic assessment and, thus, risk stratification in BS. Although several temporal and spectral indices are widely used in clinical practice to estimate sympathetic and parasympathetic contributions to the ANS, a model-based approach could provide valuable information to better understand the underlying mechanisms regulating interactions between ANS and CVS. However, for model-based analysis, subject-specific instances of the model with personalized parameters must be designed.

Therefore, we propose a coupled computational model integrating the cardiac electrical activity, the CVS and the baroreceptor reflex control of the ANS, along with a submodel representing the response to HUT testing. The CVS was composed by a representation of the heart-level interaction and both pulmonary and systemic circulations, dividing the latter in three parallel branches in order to account for regulatory differences on each branch, based on their distance from the heart level. Moreover, we performed sensitivity analyses based on the screening method of Morris, in order to assess the infuence of each model parameter on the regulation of heart rate (HR) and systolic blood pressure (SBP). Then, by using a Covariance Matrix Adaptation Evolution Strategy (CMA-ES) approach, we estimated the most significant model parameters, leading to promising subject-specific model fits to real cardiac data obtained from a patient suffering from BS and a healthy subject, in supine and upright postures during a HUT test.

Results show the feasibility of the model to reproduce real autonomic responses to HUT testing in health and disease, making cardiovascular parameters never before analyzed in BS available. Although similar results were obtained for healthy and pathological subjects, lower sympathetic and higher parasympathetic gains were observed in the latter, following the tendencies reported in previous studies where a higher vagal tone and a lower sympathetic activity during HUT testing has been related to worse prognosis in BS. Nevertheless, the identified parameters should be estimated in larger clinical series including BS patients to extract conclusions of clinical relevance. The interpretation of underlying autonomic mechanisms regulating CVS in response to HUT testing could provide new insights in the understanding of BS, with a potential impact on risk stratification.

Calvo M, Gomis P, Andreu D, Caminal P, Arbelo E. Análisis del ECG para la detección automática del patrón característico del síndrome de Brugada. November 26 - 28th, Barcelona, Spain 2014. Congreso Anual de la Sociedad Española de Ingeniería Biomédica (CASEIB).

El síndrome de Brugada es una enfermedad hereditaria caracterizada por un patrón electrocardiográfico característico, así como por un aumento del riesgo de sufrir arritmias cardíacas y muerte súbita. No existen actualmente indicadores objetivos que permitan determinar el riesgo individual, por lo que la elegibilidad para la implantación de desfibriladores automáticos implantables con el fin de prevenir eventos de muerte súbita genera todavía gran controversia. Dado que el electrocardiograma (ECG) proporciona información cardíaca de forma no invasiva, se presenta un algoritmo automático capaz de detectar parámetros electrocardiográficos relevantes en la enfermedad. De este modo, futuros estudios que incluyan grandes series de pacientes permitirán determinar el valor pronóstico de las diferentes variables detectadas. El algoritmo integra funciones disponibles previamente desarrolladas para la lectura y delineación de señales ECG y propone una expansión centrada en la detección de parámetros como la carga de latidos con patrón de Brugada, elevación del segmento ST, duración del segmento ST y de los intervalos PR, QRS y QTc, la variabilidad del ritmo cardíaco (VRC), la alternancia de la onda T o la carga de extrasístoles ventriculares, entre otros. El código se testeó en registros ECG continuos de 12 derivaciones y 24 horas de duración de dos pacientes con diferentes grados de severidad y un caso control, obteniendo tendencias relevantes en la VRC y la carga de latidos con patrón de Brugada. Asimismo, se analizaron diferentes colocaciones de los electrodos con el objetivo de concluir la configuración más sensible para la detección del patrón característico.

Benito E, Gómez F, Bisbal F, Calvo M, Prat S, Vidal B, Berruezo A, Mont J L. Reablación de fibrilación auricular guiada por resonancia magnética: análisis de la eficacia a medio plazo. October 30th -­ November 1st, Santiago de Compostela, Spain 2014. Congreso de la Sociedad Española de Cardiología.

Introducción: La presencia de gaps en las líneas de ablación es la principal causa de recurrencia de fibrilación auricular (FA) tras el aislamiento de las venas pulmonares (AVP). La resonancia magnética de realce tardío (RMN-RT) permite identificar los gaps y guiar el procedimiento de reablación reduciendo el tiempo de radiofrecuencia (RF). El objetivo de este estudio fue evaluar la eficacia de la estrategia de reablación guiada por RMN-RT a medio plazo (grupo 1) y compararla con un grupo control (técnica convencional; grupo 2).

Métodos: Para el grupo 1 se incluyeron pacientes consecutivos con recurrencia arrítmica tras un procedimiento de AVP a los que se realizó una RMN-RT. En el grupo 2 se compuso de una cohorte histórica pareada con una ratio 1:2 (edad, sexo, tipo FA). El seguimiento clínico consistió en registros holter de 48 horas a los 3, 6 y 12 meses tras la ablación, visita clínica y ECG en cada visita. Analizamos la recurrencia (definida como > 30 segundos de fa) a los 12 meses.

Resultados: Se incluyeron un total de 25 pacientes en el grupo 1 y 50 en el grupo 2 (edad media 53,1 ± 7,6, 84% varones, 60% fa paroxística). El tiempo de RF fue de 1.236 ± 560 seg en el grupo 1 y de 2.454 ± 1.236 seg en el grupo 2 (p = 0,001), sin diferencias significativas en el tiempo de procedimiento o de escopia. Solo se reportó una complicación menor en cada grupo que consistió en pericarditis en el grupo 1 y fiebre en el grupo 2. A los 12 meses la tasa de recurrencia de FA fue del 33,3% en el grupo 1 y del 46,8% en el grupo 2 (p = 0,16).

Conclusiones: La reablación guiada por RMN-RT supone un menor de tiempo de RF para el paciente presentando además una tendencia a una menor tasa de recurrencia a medio plazo comparada con el abordaje convencional. El tamaño muestral podría justificar la falta de significación estadística.

Bisbal F, Gómez-Pulido F, Akoum N, Calvo M, Cabanas P, Vidal B, Marrouche N F, Mont L. La esfericidad auricular incrementa el valor predictivo para ictus de la escala CHADS2 en pacientes con fibrilación auricular. April 23rd – 25th, Girona, Spain 2014. Sociedad Española de Cardiología, Sección de Electrofisiología y Arritmias.

La Esfericidad Auricular (EA) es un nuevo parámetro de remodelado auricular que evalúa la geometría auricular izquierda (AI) cuantificando la similitud con la esfera que mejor se adapta a dicha AI. Un estudio previo demostró que la EA tiene valor predictivo independiente para recurrencia tras ablación de fibrilación auricular (FA). El objetivo de este estudio fue evaluar si la EA puede incrementar la capacidad predictiva de ictus de la escala CHADS2 y de las características de la orejuela.

Thesis

PhD Thesis - Signal processing & telecommunications; Biomedical Engineering

Analysis of the cardiovascular response to autonomic nervous system modulation in Brugada syndrome patients

Brugada syndrome (BS) is a genetic arrhythmogenic disease characterized by a distinctive electrocardiographic pattern, associated with a high risk for sudden cardiac death. Its complex and multifactorial nature turns risk stratification into a major challenge. Although variations in autonomic modulation are commonly related to arrhythmic events in this population, novel markers with higher predictive values are still needed so as to identify those patients at high risk. Since the autonomic function can be better characterized through the application of standardized maneuvers stimulating the autonomic nervous system (ANS), the main objective of this thesis is to evaluate and compare the cardiovascular response to ANS modulations overnight, as well as in response to exercise and HUT testing, on a series of BS patients with different levels of risk (symptomatic and asymptomatic subjects). In a first part of this work, we apply previously described methods for the analysis of heart rate complexity, baroreflex sensitivity, and non-stationary heart rate variability, never before studied in the context of BS patients. In a second part, in order to address the multifactorial nature of the disease, a multivariate approach based on a step-based machine learning method is introduced. By employing markers extracted at signal-processing analysis, robust classifiers capable of distinguishing patients at different levels of risk are proposed. The third part of this work has been focused on the proposal of novel mathematical models and the associated model analysis methods, so as to study the autonomic and hemodynamic responses to exercise and HUT testing. Finally, a prospective application of a multivariate approach integrating parameters extracted at the model-based stage is also presented. Overall, the obtained results provide new insights into the underlying autonomic mechanisms regulating the cardiovascular system in BS, improving physiopathology and prognosis interpretation. The proposed approach may be used as an instrument for the identification of those asymptomatic patients at high risk who may benefit from a cardioverter defibrillator implantation.

Master’s Thesis - Biomedical Engineering

ECG analysis for the automatic detection of Brugada pattern

Brugada syndrome is characterized by a distinctive electrocardiographic pattern, as well as an increased risk of cardiac arrhythmias and sudden death. Since there are currently no indicators to determine the individual risk, the goal of the project was to design an automatic algorithm for the detection of relevant electrocardiographic parameters in order to determine their prognostic value in future studies. The algorithm was tested in two patients with different degrees of severity and a control case, obtaining relevant trends in heart rate variability and in the burden of beats presenting type-1 Brugada pattern.

Master’s Thesis - Telecommunication Engineering

Channelized Hotelling Observer Optimization for Medical Image Quality Assessment in Lesion Detection Tasks

An alternative increasingly used in the evaluation of medical image quality are the Numerical Observers, mathematical methods that estimate the human behavior in the detection of lesions in medical images. In this manuscript the Channelized Hotelling Observer was studied, using different channels and internal noise models, optimizing its estimation of the human decisions.

URL: upcommons.upc.edu