Machine Learning of 12-Lead QRS Waveforms to Identify Cardiac Resynchronization Therapy Patients With Differential Outcomes.

BACKGROUND: Cardiac resynchronization therapy (CRT) improves heart failure outcomes but has significant nonresponse rates, highlighting limitations in ECG selection criteria: QRS duration (QRSd) ≥150 ms and subjective labeling of left bundle branch block (LBBB). We explored unsupervised machine learning of ECG waveforms to identify CRT subgroups that may differentiate outcomes beyond QRSd and LBBB. METHODS: We retrospectively analyzed 946 CRT patients with conduction delay. Principal component analysis (PCA) dimensionality reduction obtained a 2-dimensional representation of preCRT 12-lead QRS waveforms. k-means clustering of the 2-dimensional PCA representation of 12-lead QRS waveforms identified 2 patient subgroups (QRS PCA groups). Vectorcardiographic QRS area was also calculated. We examined following 2 primary outcomes: (1) composite end point of death, left ventricular assist device, or heart transplant, and (2) degree of echocardiographic left ventricular ejection fraction (LVEF) change after CRT. RESULTS: Compared with QRS PCA Group 2 (n=425), Group 1 (n=521) had lower risk for reaching the composite end point (HR, 0.44 [95% CI, 0.38-0.53]; P<0.001) and experienced greater mean LVEF improvement (11.1±11.7% versus 4.8±9.7%; P<0.001), even among patients with LBBB with QRSd ≥150 ms (HR, 0.42 [95% CI, 0.30-0.57]; P<0.001; mean LVEF change 12.5±11.8% versus 7.3±8.1%; P=0.001). QRS area also stratified outcomes but had significant differences from QRS PCA groups. A stratification scheme combining QRS area and QRS PCA group identified patients with LBBB with similar outcomes to non-LBBB patients (HR, 1.32 [95% CI, 0.93-1.62]; difference in mean LVEF change: 0.8% [95% CI, -2.1% to 3.7%]). The stratification scheme also identified patients with LBBB with QRSd <150 ms with comparable outcomes to patients with LBBB with QRSd ≥150 ms (HR, 0.93 [95% CI, 0.67-1.29]; difference in mean LVEF change: -0.2% [95% CI, -2.7% to 3.0%]). CONCLUSIONS: Unsupervised machine learning of ECG waveforms identified CRT subgroups with relevance beyond LBBB and QRSd. This method may assist in objective classification of bundle branch block morphology in CRT.

Lifestyle and Risk Factor Modification for Reduction of Atrial Fibrillation: A Scientific Statement From the American Heart Association.

Atrial fibrillation (AF), the most common sustained cardiac arrhythmia, is associated with substantial morbidity, mortality, and healthcare use. Great strides have been made in stroke prevention and rhythm control strategies, yet reducing the incidence of AF has been slowed by the increasing incidence and prevalence of AF risk factors, including obesity, physical inactivity, sleep apnea, diabetes mellitus, hypertension, and other modifiable lifestyle-related factors. Fortunately, many of these AF drivers are potentially reversible, and emerging evidence supports that addressing these modifiable risks may be effective for primary and secondary AF prevention. A structured, protocol-driven multidisciplinary approach to integrate lifestyle and risk factor management as an integral part of AF management may help in the prevention and treatment of AF. However, this aspect of AF management is currently underrecognized, underused, and understudied. The purpose of this American Heart Association scientific statement is to review the association of modifiable risk factors with AF and the effects of risk factor intervention. Implementation strategies, care pathways, and educational links for achieving impactful weight reduction, increased physical activity, and risk factor modification are included. Implications for clinical practice, gaps in knowledge, and future directions for the research community are highlighted.

Effectiveness of catheter ablation of atrial fibrillation according to heart failure etiology.

BACKGROUND: Catheter ablation is an important rhythm control therapy in patients with atrial fibrillation (AF) with concomitant heart failure (HF). The objective of this study was to assess the comparative efficacy of AF ablation patients with ischemic vs nonischemic heart failure. METHODS: We conducted a retrospective, observational cohort study of patients with HF who underwent AF ablation. Outcomes were compared based on HF etiology and included in-hospital events, symptoms (Mayo AF Symptom Inventory [MAFSI]), and functional status (New York Heart Association class) and freedom from atrial arrhythmias at 12 months. RESULTS: Among 242 patients (n = 70 [29%] ischemic, n = 172 [71%] nonischemic), patients with nonischemic cardiomyopathy were younger (mean age 64 ± 11.5 vs 69 ± 9.1, P = .002), more often female (36% vs 17%, P = .004), and had higher mean left-ventricular ejection fraction (47% vs 42%, P = .0007). There were no significant differences in periprocedural characteristics, including mean procedure time (243 ± 74.2 vs 259 ± 81.8 minutes, P = .1) and nonleft atrial ablation (17% vs 20%, P = .6). All-cause adverse events were similar in each group (15% vs 17%, P = .7). NYHA and MAFSI scores improved significantly at follow-up and did not differ according to HF etiology (P = .5; P = .10-1.00 after Bonferroni correction). There were no significant differences in freedom from recurrent atrial arrhythmia at 12-months between ischemic (74%) and nonischemic patients (78%): adjusted RR 0.63, 95% confidence interval 0.33-1.19. CONCLUSIONS: Catheter ablation in patients with AF and concomitant heart failure leads to significant improvements in functional and symptom status without significant differences between patients with ischemic vs nonischemic HF etiology.

Predictors of long-term outcomes greater than 10 years after cardiac resynchronization therapy implantation.

BACKGROUND: Cardiac resynchronization therapy (CRT) is indicated in patients with medically refractory heart failure and wide QRS duration. While much is known about predictors of left ventricular (LV) remodeling after CRT implantation and short-term mortality, limited data exist on long-term outcomes after CRT placement. METHODS: We retrospectively reviewed all patients undergoing CRT implantation at our center between 2003 and 2008 and examined mortality using institutional electronic records, social security death index, and online obituary search. We included only patients with preimplant echoes with LV ejection fraction (LVEF) 35% or below. Variable selection was performed using stepwise regression and models were compared using goodness-of-fit criteria. A final model was validated with the bootstrap regression method. RESULTS: Out of the 877 CRT patients undergoing implantation during this time, 287 (32.7%) survived longer than 10 years. Significant (P < .05) predictors of survival in our multivariate model were age, left ventricular diastolic diameter, sex, presence of nonischemic vs ischemic cardiomyopathy, QRS duration, atrial fibrillation, BNP levels, and creatinine levels at the time of CRT implantation. A model using the odds ratios from these variables had a receiver operating curve with an area under the curve score of 0.816 (standard error, 0.019) at predicting survival or freedom from LVAD or heart transplant for longer than 10 years after CRT implantation. The specificity for factors 3 or above and 5 or above was 68% and 77%, respectively. CONCLUSION: A large proportion of patients are still alive 10 years after CRT implantation. Variables at the time of CRT implant can help provide prognostic information to patients and electrophysiologists to determine the long-term benefit and survival of patients after CRT implantation.

Scar burden is an independent and incremental predictor of cardiac resynchronisation therapy response.

Objective: Determine the prognostic impact of scar quantification (scar %) by cardiac magnetic resonance (CMR) in predicting heart failure admission, death and left ventricular (LV) function improvement following cardiac resynchronisation therapy (CRT), after controlling for the presence of left bundle branch block (LBBB), QRS duration (QRSd) and LV lead tip location and polarity. Methods: Consecutive patients who underwent CMR between 2002 and 2014 followed by CRT were included. The primary endpoint was death or heart failure admission. The secondary endpoint was change in ejection fraction (EF) ≥3 months after CRT. Cox proportional hazards, linear regression models and change in the area under the receiver operating characteristic curve (AUC) were used. Results: A total of 84 patients were included (63% male, 51% with ischaemic cardiomyopathy). After adjusting for clinical factors, presence of LBBB and QRSd and LV lead tip location and polarity, greater scar % remained associated with a higher risk for clinical events (HR=1.06; 95% CI 1.02 to 1.10; p<0.001) and a smaller improvement in EF (slope: -0.61%; 95% CI -0.93% to 0.29%; p<0.001). When adding scar % to QRSd and LBBB, there was significant improvement in predicting clinical events at 3 years (AUC increased to 0.831 from 0.638; p=0.027) and EF increase ≥10% (AUC 0.869 from 0.662; p=0.007). Conclusion: Scar quantification by CMR has an incremental value in predicting response to CRT, in terms of heart failure admission, death and EF improvement, independent of the presence of LBBB, QRSd, LV lead tip location and polarity.

Machine Learning Prediction of Response to Cardiac Resynchronization Therapy: Improvement Versus Current Guidelines.

BACKGROUND: Cardiac resynchronization therapy (CRT) has significant nonresponse rates. We assessed whether machine learning (ML) could predict CRT response beyond current guidelines. METHODS: We analyzed CRT patients from Cleveland Clinic and Johns Hopkins. A training cohort was created from all Johns Hopkins patients and an equal number of randomly sampled Cleveland Clinic patients. All remaining patients comprised the testing cohort. Response was defined as ≥10% increase in left ventricular ejection fraction. ML models were developed to predict CRT response using different combinations of classification algorithms and clinical variable sets on the training cohort. The model with the highest area under the curve was evaluated on the testing cohort. Probability of response was used to predict survival free from a composite end point of death, heart transplant, or placement of left ventricular assist device. Predictions were compared with current guidelines. RESULTS: Nine hundred twenty-five patients were included. On the training cohort (n=470: 235, Johns Hopkins; 235, Cleveland Clinic), the best ML model was a naive Bayes classifier including 9 variables (QRS morphology, QRS duration, New York Heart Association classification, left ventricular ejection fraction and end-diastolic diameter, sex, ischemic cardiomyopathy, atrial fibrillation, and epicardial left ventricular lead). On the testing cohort (n=455, Cleveland Clinic), ML demonstrated better response prediction than guidelines (area under the curve, 0.70 versus 0.65; P=0.012) and greater discrimination of event-free survival (concordance index, 0.61 versus 0.56; P<0.001). The fourth quartile of the ML model had the greatest risk of reaching the composite end point, whereas the first quartile had the least (hazard ratio, 0.34; P<0.001). CONCLUSIONS: ML with 9 variables incrementally improved prediction of echocardiographic CRT response and survival beyond guidelines. Performance was not improved by incorporating more variables. The model offers potential for improved shared decision-making in CRT (online calculator: http://riskcalc.org:3838/CRTResponseScore ). Significant remaining limitations confirm the need to identify better variables to predict CRT response.

Should fast pathway ablation be reconsidered in typical atrioventricular nodal re-entrant tachycardia?

INTRODUCTION: Atrioventricular nodal re-entry tachycardia (AVNRT) is the most common, regular narrow-complex tachycardia. The established treatment is catheter ablation of the AV nodal slow pathway (SP). However, in a select group of patients with long PR intervals in sinus rhythm, SP ablation can lead to AV block due to the absence of robust anterograde conduction through the fast pathway (FP). This report aims to demonstrate that AV nodal FP ablation is a reasonable approach in patients with AVNRT and poor or absent anterograde FP conduction. METHODS AND RESULTS: Standard electrophysiology study techniques were used in the electrophysiology laboratory. Catheter ablations were performed using radiofrequency energy. Mapping of intracardiac activation was performed with electroanatomical mapping systems. Outcomes were assessed acutely during the procedure and during routine clinical follow-up. Six patients with first-degree AV block and recurrent AVNRT who underwent ablation of their tachycardia at our institution are presented. One patient underwent ablation of AV nodal SP resulting in high-degree AV block necessitating pacemaker implantation. The remaining five patients underwent ablation of the AV nodal FP guided by electroanatomical mapping of the earliest atrial activation in tachycardia. These five had successful treatment of the tachycardia with preservation of anterograde AV nodal conduction. Mapping and ablation approach to eliminate retrograde FP conduction are described. CONCLUSION: In select patients with AVNRT and poor anterograde FP conduction, retrograde FP ablation is reasonable and is less likely to result in AV block and pacemaker dependency.

Management and outcomes in mitral valve prolapse with ventricular arrhythmias undergoing ablation and/or implantation of ICDs.

BACKGROUND: While there is an association between isolated mitral valve prolapse (MVP) and sudden cardiac arrest (SCA), the baseline characteristics and outcomes of patients with isolated MVP who experience ventricular arrhythmias (VAs) and then subsequently undergo catheter ablation and/or implantable cardioverter defibrillator (ICD) implantation are unknown. METHODS: We performed a retrospective review of all patients at the Cleveland Clinic with isolated MVP between 1997 and 2016 who underwent VA catheter ablation or secondary prevention ICD implantation. RESULTS: Of 617 screened patients, we identified 43 patients with isolated MVP and significant VA who underwent ICD placement (n = 13, 30%) or catheter ablation (n = 30, 70%). Both leaflets were most commonly involved (n = 22, 52%) with posterior MVP being next most common (n = 15, 36%). The most common foci of VA origin was the left ventricular papillary muscle (n = 9, 27%). Ablation was successful in the majority of cases (n = 20, 65%). At a mean follow-up of 2.5 years, 11 patients (26%) had recurrent VT. CONCLUSIONS: Patients with isolated MVP and VA were more likely to have bileaflet prolapse and at least moderate mitral regurgitation. VA originated more commonly from left-sided foci. While ablation was acutely successful in the majority of cases, there was still a moderate rate of VA recurrence. There is still more study needed on factors that will predict malignant VAs and management of these VAs in the MVP population.

Effect of Cardiac Resynchronization Therapy on Left Ventricular Remodeling in Patients With Cardiac Sarcoidosis.

Cardiac resynchronization therapy (CRT) has been shown to be beneficial in patients with medically refractory heart failure. Although it has been found to be effective in a wide range of etiologies for nonischemic cardiomyopathy, its role in improving remodeling and survival of patients with cardiac sarcoidosis (CS) remains undefined. We performed a retrospective review of all patients at our institution with CS who underwent implantation of a CRT device from 2007 to 2017. The outcomes of this population were compared with the outcomes of a cohort of patients with nonischemic cardiomyopathy with an etiology other than sarcoidosis. Nineteen patients in our institution with CS underwent CRT implantation during the time period. This group was compared with 311 consecutive patients with other etiologies of nonischemic cardiomyopathy who underwent CRT implantation. CRT improved left ventricular ejection fraction (LVEF) from 28.8% to 35.9% (p <0.05) in CS, whereas it improved LVEF from 25% to 36.6% (p <0.01) in non-CS group (difference in means of 0.13). CRT significantly improved diastolic and systolic LV diameters, mitral regurgitation, and right ventricular systolic function in non-CS patients but failed to improve same parameters in CS patients. In conclusion, CRT significantly improved LVEF in patients with CS. There is no significant evidence that survival outcomes of CRT patients with CS are significantly worse than other etiologies of nonischemic cardiomyopathy.

Rhythm control in heart failure patients with atrial fibrillation: contemporary challenges including the role of ablation.

Because nonpharmacological interventions likely alter the risks and benefits associated with rhythm control, this paper reviews the role of current rhythm control strategies in atrial fibrillation. This report also focuses on the specific limitations of pharmacological interventions and the utility of percutaneous ablation in this growing population of patients with concomitant atrial fibrillation and heart failure.

17ß-Estradiol attenuates hypoxic pulmonary hypertension via estrogen receptor-mediated effects.

RATIONALE: 17ß-Estradiol (E2) attenuates hypoxic pulmonary vasoconstriction and hypoxic pulmonary hypertension (HPH) through an unknown mechanism that may involve estrogen receptors (ER) or E2 conversion to catecholestradiols and methoxyestradiols with previously unrecognized effects on cardiopulmonary vascular remodeling. OBJECTIVES: To determine the mechanism by which E2 exerts protective effects in HPH. METHODS: Male rats were exposed to hypobaric hypoxia while treated with E2 (75 µg/kg/d) or vehicle. Subgroups were cotreated with pharmacologic ER-antagonist or with inhibitors of E2-metabolite conversion. Complementary studies were performed in rats cotreated with selective ERα- or ERß-antagonist. Hemodynamic and pulmonary artery (PA) and right ventricular (RV) remodeling parameters, including cell proliferation, cell cycle, and autophagy, were measured in vivo and in cultured primary rat PA endothelial cells. MEASUREMENTS AND MAIN RESULTS: E2 significantly attenuated HPH endpoints. Hypoxia increased ERß but not ERα lung vascular expression. Co-treatment with nonselective ER inhibitor or ERα-specific antagonist rendered hypoxic animals resistant to the beneficial effects of E2 on cardiopulmonary hemodynamics, whereas ERα- and ERß-specific antagonists opposed the remodeling effects of E2. In contrast, inhibition of E2-metabolite conversion did not abolish E2 protection. E2-treated hypoxic animals exhibited reduced ERK1/2 activation and increased expression of cell-cycle inhibitor p27(Kip1) in lungs and RV, with up-regulation of lung autophagy. E2-induced signaling was recapitulated in hypoxic but not normoxic endothelial cells, and was associated with decreased vascular endothelial growth factor secretion and cell proliferation. CONCLUSIONS: E2 attenuates hemodynamic and remodeling parameters in HPH in an ER-dependent manner, through direct antiproliferative mechanisms on vascular cells, which may provide novel nonhormonal therapeutic targets for HPH.