A Comprehensive Approach to Cardiac Resynchronization Therapy

Abstract

This chapter aims to comprehensively examine and offer guidance on the contemporary indications for Cardiac Resynchronization Therapy (CRT) in individuals with pacemaker indications or heart failure. Herein we examine the contemporary understanding of CRT responders by analyzing the latest evidence. We explore the impact of CRT on mortality rates, heart failure hospitalizations, clinical parameters of heart failure, stabilization of ventricular function, and its role in preventing the progression of heart failure. We delve into the latest advancements in physiological pacing, encompassing anatomical and physiological characteristics, while critically evaluating the associated advantages and disadvantages. Additionally, the chapter explores future prospects and directions in the field, providing a well-rounded overview of the evolving landscape of CRT.

1. Introduction

Cardiac Physiologic Pacing (CPP) is at the forefront of device management in Heart Failure with reduced Ejection Fraction (HFrEF). The 2023 HRS/APHRS/LAHRS guideline on Cardiac Physiologic Pacing for the avoidance and mitigation of heart failure defined CPP as “cardiac pacing intended to restore or preserve ventricular synchrony”. CPP includes cardiac resynchronization therapy (CRT) and Conduction System Pacing (CSP). CRT uses left ventricular (LV) stimulation and Biventricular (BiV) pacing by utilizing a lead placed in a coronary sinus branch in the LV epicardium. CSP includes His bundle pacing (HBP), or left bundle branch area pacing (LBBAP). The evidence supporting CRT use in HFrEF is more robust than that of Conduction System Pacing (CSP) due to the longer duration of its widespread application [1]. The primary focus of this chapter is CRT: pathophysiology, indications, patient selection, procedural aspect, current evidence of CRT, and future CRT directions and innovation.

2. Pathophysiology of cardiac dyssynchrony

In the normal physiologic depolarization sequence, the initial electrical activation originates in the LV and Right Ventricular (RV) endocardium. This progression unfolds with the depolarization of the septum from left to right, followed by the swift activation of the remaining LV myocardium, including the lateral wall, through specialized conduction tissue [2].

Conversely, with left bundle branch block (LBBB), depolarization initiates in the right ventricular endocardium. Subsequently, septal activation occurs through impulses transmitted from the right bundle branch, moving in a right-to-left direction. This activation then traverses to the left ventricular endocardium, reaching and activating the remainder of the left ventricular myocardium. Notably, this latter route partially bypasses specialized conduction tissue, leading to a delayed activation of the lateral wall creating electromechanical dyssynchrony [3].

Electromechanical dyssynchrony caused by a LBBB carries significant hemodynamic implications, potentially resulting in diminished LV contraction and compromised relaxation. This can culminate in adverse remodeling over an extended period. Consequently, a subset of individuals experiencing prolonged LBBB may be susceptible to the development of dyssynchrony-induced cardiomyopathy, characterized by a reduction in left ventricular ejection fraction (LVEF) and the onset of heart failure (HF). Similarly, patients with high percentages of RV only pacing may develop dyssynchrony from a similar mechanism [4].

Acknowledging the reversible nature of electromechanical dyssynchrony-induced cardiomyopathy underscores the importance of routine assessments of ventricular function in the specific context of LBBB or high percentage RV pacing. The integration of tailored therapeutic approaches such as CRT can play a pivotal role in mitigating the impact of electrodyssynchrony [5].

3. Evidence

Prior to the MUSTIC (Multisite Stimulation in Cardiomyopathies) trial in 2001 and PATH-CHF (The Pacing Therapies for Congestive Heart Failure) trial in 2002, observational studies had highlighted the negative impact of RV pacing on heart failure patients and the positive hemodynamic changes in the pulmonary capillary wedge pressure and cardiac output [6]. Despite their limitations in sample size and study design, the MUSTIC and PATH-HF trials established the efficacy of CRT pacing in selected chronic systolic heart failure patients with intraventricular conduction delay and showcased positive outcomes such as improved walking distance and reduced HF hospitalizations (Figure 1) [8, 9].

Figure 1.

A timeline of the core CRT clinical trials [7].

These pivotal trials paved the way for the MIRACLE (Multicenter InSync Randomized Clinical Evaluation) study, the first double-blinded randomized clinical trial involving 453 patients with moderate to severe HF and QRS duration ≥130 ms. This trial validated the efficacy of CRT-P + Optimal Medical Therapy (OPT) compared to OPT alone. CRT-P + OPT combination reduced NYHA functional class and HF hospitalization in addition to improving quality of life and LVEF [10]. This was followed by the MIRACLE ICD study involving 369 patients with moderate to severe HF and QRS duration ≥130 ms. This trial demonstrated the safety and efficacy of combined CRT and ICD therapy (CRT-D) compared to ICD therapy alone. CRT-D improves quality of life and functional class [11].

In 2004, the COMPANION (Comparison of Medical Therapy, Pacing and Defibrillation in Heart Failure) trial was the first randomized clinical trial to compare CRT-P and CRT-D + OPT to OPT alone. The primary composite endpoint, encompassing time to death from any cause or hospitalization, showed a decreased risk with both CRT-P (HR 0.81) and CRT-D (HR 0.80) compared to OPT alone [12]. Subsequently, in 2005, CARE-HF (Cardiac Resynchronization-Heart Failure) trial compared CRT + OPT to OPT alone in 813 patients. The findings were consistent with previous clinical trials: the primary endpoint, time to death from any cause or unplanned hospitalization for a major cardiovascular event, was lower in CRT group (39%) vs. OPT group (55%). The secondary endpoint of death from any cause was also lower in the CRT group (20% vs. 30%) [13].

The COMPANION and the CARE-HF landmark trials laid the foundation for using CRT-P and CRT-D as part of the Guideline Directed Medical Therapy (GDMT) in HFrEF (EF ≤35%) with NYHA class III-IV and wide QRS complex (≥120 ms).

The benefit of CRT in mild symptomatic HF was not fully explored until the publication of REVERSE (Resynchronization Reverses Remodeling in Systolic Left Ventricular Dysfunction) and MADIT-CRT (Multicenter Automatic Defibrillator Implantation Trial with Cardiac Resynchronization Therapy) trials in 2008 and 2009. The REVERSE randomized 610 patients with NYHA Class I-II, EF < 40% and wide QRS (≥120 ms) to active CRT vs. control. Active CRT in combination with medical therapy reduces HF hospitalization and improves the LV structure and function [14]. The MADIT-CRT randomized 1820 patients with NYHA Class I-II, EF < 30% and a wide QRS (≥130 ms) to CRT-D vs. ICD alone. CRT-D had HR 0.66, P = 0.001 in the primary endpoint (death or nonfatal heart-failure events) compared to ICD alone. CRT-D arm had 41% reduction in HF hospitalization especially with QRS duration >150 ms. CRT-D was also associated with improved LVEF and volumes [15].

The BLOCK HF (Biventricular versus Right Ventricular Pacing in Heart Failure Patients with Atrioventricular Block) study demonstrated the superiority of the BiV pacing over RV pacing only in patients with atrioventricular block, LVEF ≤50%, and NYHA Class I-II. The primary outcome was comprising death from any cause, an urgent care visit for heart failure requiring intravenous therapy, or a 15% or more increase in the left ventricular end-systolic volume index. The HR for the primary outcome was 0.74 suggesting the CRT role in mitigating adverse LV remodeling and improving overall outcomes compared to RV pacing only (Table 1) [16].

Trial name	Date	Sample Size	EF	QRS duration	NYHA class	Primary outcomes	Results
MUSTIC [8]	2001	48	≤35%	≥150 ms	III	Distance walked in six minutes, quality of life, peak oxygen consumption, hospitalizations related to heart failure, and mortality rate.	CRT-P improved walking distance and quality of life, reduced HF hospitalization
PATH-HF [9]	2002	41	NA	≥120 ms	III-IV	Oxygen uptake during bicycle exercise and the 6-minute walk distance	CRT-P improved walking distance and peak VO2 after 12 months
MIRACLE [10]	2002	571	≤35%	≥130 ms	III-IV	NYHA class, quality of life, and the distance walked in six minutes.	CRT-P improves six minutes walking distance, NYHA functional class, quality of life, time on the treadmill during exercise testing, and EF
MIRACLE ICD [11]	2003	369	≤35%	≥130 ms	III-IV	NYHA class, quality of life, and the distance walked in six minutes.	CRT-D improves quality of life, functional status, and exercise capacity
COMPANION [12]	2004	1520	≤35%	≥120 ms	III-IV	Death from any cause or hospitalization for any cause	CRT-P and CRT-D reduces all cause death and hospitalization
CARE-HF [13]	2005	813	≤35%	≥120 ms	III-IV	Death from any cause, cardiovascular hospitalizations	CRT reduces mortality and heart failure hospitalizations
REVERSE [14]	2008	610	≤40%	≥120 ms	I-II	HF clinical composite response	CRT-P delays time-to-first HF hospitalization and improves in LV end-systolic volume index
MADIT-CRT [15]	2009	1820	≤30%	≥130 ms	I-II	Death from any cause or non-fatal heart failure events	CRT-D reduces heart failure events and all-cause mortality
RAFT [17]	2010	1798	≤30%	≥120 ms	II-III	Death from any cause, heart failure hospitalizations	CRT-D reduces mortality and heart failure hospitalizations
BLOCK-HF [16]	2013	691	≤50%	≥150 ms	I-II	Death, heart failure-related urgent care visits	CRT-P improves clinical outcomes in patients with AV block

Table 1.

A summary of core clinical trials that shaped the current guidelines and selection criteria of CRT [7].

4. Indications

This cumulative evidence supporting CRT use in patients with moderate to severe HF and interventricular conduction delay shaped the current guidelines and recommendations. The 2023 HRS/APHRS/LAHRS guideline on CPP support the following (Table 2) [1].

Recommendation class	Recommendation
Class IA	Patients with LVEF ≤35%, sinus rhythm, LBBB with QRS duration ≥150 ms and NYHA class II-IV on Guideline Directed Medical Therapy (GDMT), CRT with BiV pacing improves symptoms and reduce morbidity and mortality.
Class IIA	Patiens with LVEF ≤35%, sinus rhythm, LBBB with QRS duration ≥150 ms, and NYHA class I symptoms on GDMT, CPP with HBP with LBBB correction or LBBAP is reasonable if effective CRT cannot be achieved with BiV pacing based on anatomical or functional.
Class IIB	In patients with LVEF ≤30%, sinus rhythm, LBBB, QRS duration ≥150 ms, and NYHA class I symptoms on GDMT, CRT with BiV pacing may be considered to reduce the risk of worsening HF and potentially improve LV remodeling.
Class IIB	In patients with LVEF 36–50%, sinus rhythm, LBBB with QRS duration ≥150 ms, and NYHA class II-IV symptoms on GDMT, CPP maybe considered to maintain or improve LVEF
Class IIB	In patients with LVEF <35%, sinus rhythm, LBBB with QRS duration ≥150 ms, and NYHA class II-IV symptoms on GDMT, CSP with HBP or LBBAP may be considered as an alternative to CRT with BIV pacing.

Table 2.

A summary of the 2023 HRS/APHRS/LAHRS guideline on CPP guidelines [1].

5. Patient selection and CRT response

These core CRT clinical trials assessed response on several clinical and echocardiographic parameters. Clinical responses evaluated included NYHA class, improvement in quality of life, increase in peak oxygen consumption, and reduced HF hospitalization and mortality. Echocardiographic responses included >5% absolute increase in LVEF or the absence of worsening in LVEF, reduction in LV size, increase in LV stroke volume, and decrease in mitral regurgitation. Based on these selection criteria, 60–70% of CRT recipients will be responders [1].

Reverse remodeling following CRT has been linked to LBBB, QRS duration (>150 ms), nonischemic etiology, and female sex [18]. Below we discuss the most important predictive factors of CRT response:

5.1 QRS duration and morphology

CRT is effective in patients with QRS duration of >120 to 130 ms. Conversely, individuals with a QRS duration <120 ms are not likely to benefit from CRT based on randomized studies which showed no improvement in peak oxygen consumption or reverse remodeling on the echocardiogram, therefore, CRT is not indicated for this population [19, 20, 21]. A comprehensive meta-analysis utilized patient-level data from key CRT trials, including MIRACLE, MIRACLE-ICD, MIRACLE-ICD II, REVERSE, RAFT, BLOCK-HF, COMPANION, and MADIT-CRT. This meta-analysis aimed to evaluate the benefits of CRT based on QRS morphology LBBB (n = 4549); RBBB, (n = 691); and nonspecific intraventricular conduction delay NSIVCD, (n = 1024) and duration (with a 150-ms partition). The primary endpoint focused on the time to heart failure hospitalization (HFH) or death, while a secondary endpoint considered the time to all-cause death. This meta-analysis showed no advantage of CRT in patients with RBBB or NSIVCD [22].

5.2 Cardiomyopathy type (ischemic versus non-ischemic)

In the Cardiac Resynchronization – Heart Failure (CARE-HF) study, CRT resulted in comparable reductions in all-cause mortality for individuals with both ischemic and non-ischemic cardiomyopathy [23]. However, several studies showed that CRT is more effective in patients with non-ischemic cardiomyopathy compared to ischemic cardiomyopathy. CRT in non-ischemic cardiomyopathy resulted in a more substantial increase in LVEF and a greater reduction in NYHA functional class compared to those with ischemic cardiomyopathy [24]. Furthermore, upon conducting sub-analyses of various prospective randomized studies, including MIRACLE (Multicenter InSync Randomized Clinical Evaluation), CARE-HF and, REVERSE (Resynchronization reVErses Remodeling in Systolic left ventricular dysfunction) [20], and MADIT-CRT it was consistently affirmed that more favorable reverse remodeling occurred in cases of non-ischemic cardiomyopathy compared to ischemic cardiomyopathy [25].

5.3 Atrial fibrillation

Frequent atrial arrythmias including atrial fibrillation (AF), and frequent premature ventricular complexes diminish BiV pacing percentage and therefore the CRT effectiveness. Conversely, achieving a high percentage of BiV pacing in observational studies has shown a potential link between CRT and a decreased burden of AF [26]. Regardless, for CRT to be useful, adequate rate control must be present. Achieving a high percentage of BiV pacing proved unsuccessful in approximately two-thirds of the 8686 patients diagnosed with persistent or permanent AF due to medically refractory rapid ventricular rates [27]. Notably, the subgroup with less well controlled rates experienced an elevated risk of mortality. It is important to consider a more aggressive approach to rate control and increased effective BiV pacing in individuals with AF to optimize the advantages of CRT. In fact, in patients with permanent AF in whom primary AF ablation is not possible, many will benefit from AV node ablation and CRT. Specifically, those with heart failure [28]. In the PAVE trial, a group of 184 individuals diagnosed with permanent atrial AF, of which 83% had NYHA class II or III HF, underwent AV node ablation as a treatment for medically refractory rapid ventricular rates. These participants were then randomly allocated to either a standard RV pacing system or CRT pacing system. After a six-month follow-up period, CRT led to significantly greater enhancements in the six-minute walking distance above baseline (31% vs. 24%, P = 0.04), peak oxygen consumption during exercise, and exercise duration when compared to standard RV pacing [29].

5.4 LV SCAR

LV scar is detected in as many as 40% of individuals eligible for CRT, and its presence is indicative of an anticipated suboptimal response. Absence of scar in LV and RV pacing regions is associated with 81% CRT response rate compared to when the scar occupies RV pacing region (55% CRT response rate), LV pacing region (25% CRT response rate), and both pacing regions (0% CRT response rate) [30].

6. Pacing considerations in patients with preserved LVEF

Selecting appropriate pacemaker type in patients with bradyarrhythmia and preserved LVEF needs careful consideration. Available choices are RV pacing and CPP (CSP or CRT). Several factors should be considered such as AF, AV conduction and patient’s age and comorbidities. Taking these factors into account, both RV pacing and CPP are reasonable options in carefully selected patients.

In PACE, an older study, 177 patients with bradycardia and EF ≥ 45% at baseline were examined for effect if CRT vs. RV only pacing. The EF was better preserved in CRT as compared to those with RV only pacing. In the CRT group, 20.2% dropped their EF ≥5% at 2 year as compared to 62.5% of those in the RV group (p < 0.001^)29. However, no differences were seen in secondary endpoints such as 6-minute walk. Mortality was low in both groups. Thus, mortality benefit was not proven. It is notable that the baseline EF of 45% would by more current standards be regarded as mid-range for heart failure [31]. Similarly, the PREVENT-HF trial, a randomized study involving 108 patients with an initial mean normal LVEF (55% in the RV pacing group and 57% in the CRT group), did not demonstrate a clear advantage of biventricular (BiV) pacing compared to right ventricular pacing (RVP) based on primary and secondary echo parameters of outcome (primary-LV diastolic volume, secondary LV systolic volume, EF and MR). Importantly, while no significant harm was observed with BiV pacing, it is noteworthy that the introduction of an additional left ventricular (LV) lead during the procedure was linked to an extended duration of the procedure and an increased occurrence of procedure-related complications [32].

Based on these data, it is reasonable to perform RV only pacing in patients with preserved EF, particularly in those in whom a low percentage of RV pacing is anticipated. In addition, patients who develop depressed EF because of RV only pacing should be considered for upgrade to CRT [33, 34]. However, it is a class IIb in the HRS/APHRS/LAHRS guidelines to select CPP to reduce the risk of pacemaker-induced cardiomyopathy [1].

7. Procedural considerations and follow up

7.1 Preprocedural evaluation

Preprocedural evaluation includes but is not limited to history and physical exam to evaluate candidacy and severity of HF symptoms, EKG to assess QRS duration and morphology, echocardiogram to evaluate LVEF [35, 36, 37]. Rarely, Cardiac MRI (CMR) and nuclear imaging can be used to evaluate LVEF, ischemia and scar.

7.2 CRT implant

Several factors are taken into consideration that can be barriers when performing a CRT implant: engaging the CS, finding optimal branches, and determining the best pacing strategies to maximize CRT. Current guidelines recommend “a quadripolar LV lead to assist with lead stability, lower capture thresholds, avoid phrenic nerve pacing and decrease need for lead repositioning (Class 1)”, making lead positioning one of the most important factors to consider when performing CRT device implantation. Compared to bipolar leads, quadripolar leads require less fluoroscopy, allow for better distal vein positioning, and facilitate lower pacing thresholds and impedances [38]. Given that pacing vectors can be switched between the different poles, phrenic nerve stimulation can be avoided. Thus, the best hemodynamic response is usually achieved by placing the electrode around the area with the latest LV activation that provides adequate threshold without phrenic nerve stimulation.

Moreover, there are several means of optimization of CRT therapy. In a small study [39], the best hemodynamic response was achieved with the narrowest QRS duration by optimization of the interventricular delay. Additional improvement in reverse remodeling can be achieved by optimizing the AV delay and finding the best fusion-optimized QRS duration during LV pacing [40]. Currently, device manufacturers have different algorithms to enhance QRS duration automatically to provide an individual approach to each patient. For example, certain Medtronic CRT device models use AdaptivCRT™ algorithm which enhances physiological and dynamic pacing by assessing intrinsic conduction to determine appropriate AV interval duration, based on the AV interval, the device decides on Adaptive LV vs. BiV pacing. As a result, the device will continuously optimize the AV and VV delay and optimize pacing configuration [41]. Several studies have shown non-inferiority of this method compared to echocardiographic optimization which is time consuming and rarely performed clinically.

Anatomic lead positioning is also fundamental and plays an important role in the success of a CRT. Apical LV pacing has shown lower event-free survival and LV reverse remodeling compared to basal and midventricular LV lead positions [42]. Additionally, placement of LV leads in areas of electrical delay can confer a greater benefit [43]. QLV (time from the onset of QRS on the ECG to local activation at the site of the LV lead) > 95 ms or > 50% total QRS duration favor optimal response with CRT.

When CS lead placement is unattainable, surgical epicardial LV lead pacing is a reasonable alternative.

7.3 Complications

The risks of complications should be weighed when deciding to implant a CRT device. In those with life expectancy <1, shared decision-making is required to consider the potential improvement in quality of life compared to the risk of procedural complications. CRT-D is not indicated for these patients, whereas CRT-P may be considered.

The complications related to a CRT are those inherent to a pacemaker and ICD with the addition of those related to an LV lead placement. These complications include coronary vein dissection or perforation. Additionally, other complications include the risk of pneumothorax, lead dislodgement, infection and/or hematoma, perforation, tamponade, cardiac arrest, and sustained ventricular tachyarrhythmia.

7.4 Follow up

A multidisciplinary team is required for follow up of patients with implanted CRT devices. This includes advanced heart failure and electrophysiologists, both to assess and optimize GDMT and assess periodically LV lead capture. This is best accomplished in a specialized device clinic [44]. Patients who do not appear to have benefited from CRT need careful evaluation to correct potentially reversible factors, such as inadequate BiV pacing percentage (due to a high burden of ectopy or AF), or suboptimal lead placement position. A chest X-ray can be useful to assess LV lead position. A 12-lead ECG is useful to confirm LV capture and facilitate optimization of LV pacing configurations. An echocardiogram should be repeated within 3–12 months after implant. Ablation or pharmacological suppression of PVCs or atrial fibrillation might be required to promote enhanced BiV pacing.

7.5 Alternatives to CRT

LV lead placement is not possible in up to 10% of patients planned for CRT with BiV pacing due to anatomical/technical reasons, functional issues (high thresholds, diaphragmatic stimulation), and intrinsic ECG considerations. Criteria for optimal lead placement continue to evolve and failure of lead placement at initial implantation has not been standardized. Hence, the decision to abandon the initial approach and crossover to another conduction system pacing (HBP or LBBAP) varies among the operators. Current guidelines recommend crossover from CRT with BiV pacing via CS to CSP with HBP or LBBAP when the CS LV lead placement is unsuccessful or suboptimal (class 2a). Crossover to epicardial LV lead placement is currently a class 2b indication.

We are not fully addressing CSP in this chapter, but briefly note that there is compelling data to support the use of CSP, either as a “fallback” when CRT is not feasible. Or in some cases as a preferred first line means to achieve resynchronization.

7.6 Response to CRT

The response to CRT in heart failure patients is variable. Some may experience improvement in objective (LVEF, NYHA class) and/or subjective parameters and in some others, CRT might manifest as a slowing of the natural progression of HF [45]. In all these patients, continuation of CRT should be sought at the time of battery replacement as some studies have shown worse outcomes when CRT was deactivated in people with improved LVEF [46]. Those who do not respond to CRT as expected are labeled as “nonresponders”. However, this definition does not consider the natural history of the disease and the fact that CRT can stabilize HF progression. Nevertheless, these patients should have medication optimization, evaluation of lead position, device reprogramming to optimize and look for a favorable response.

8. Future directions and innovation

8.1 Leadless CRT

Leadless pacemaker implantation is growing due to the high success rate and low complication rate [47]. Single and dual chamber leadless pacemaker eliminates leads and generator-related complications such as lead fracture or dislodgement, and pocket infection or procedural complications such as pneumothorax and coronary sinus perforation [48]. Leadless CRT efficacy and safety remains under investigation, however, similar to single and dual chamber leadless pacemakers, leadless CRT will eliminate lead and some procedural complications compared to conventional CRT.

The SOLVE-CRT study, presented at Heart Rhythm 2023 scientific meeting, introduced the WiSE® CRT System, a potential endocardial LV pacing as an alternative to conventional CRT. The study included CRT non-responders, untreated cases with lead failures, and those undergoing high-risk upgrades. Results showed 80.9% freedom from Type I complications (device and procedure-related) and a significant 16.4% improvement in Left Ventricular End Systolic Volume [49]. Although a large sample randomized double blinded clinical trials are needed to determine the efficacy and safety of leadless CRT, these positive findings will help customize CRT in HF patients.

8.2 Use of advanced cardiac imaging to optimize response and patient selection

QRS duration as the sole measurement of ventricular dyssyncrhony is not ideal. This is evidenced by the 30% CRT nonresponse rate. Nonresponders may include a subset with wide QRS but no actual mechanical dyssyncrhony. This inspired researchers to evaluate other methods to predict CRT responsiveness and optimize patient selection. Initially, echocardiogram offered a non-electrical assessment of mechanical dyssnchrony using several parameters such as M-mode, color tissue Doppler (TD) M-mode, longitudinal TD velocity, pulsed TD, additional measures such as strain and strain rate imaging and 3-D echocardiography [50]. However, the multi-centric Predictors of Response to CRT (PROSPECT) study showed variable ability of echocardiographic parameters to predict CRT response, sensitivity (ranging from 6–74%) and specificity (ranging from 35–91%) [51]. Similarly, echocardiographic characterization of dyssynchrony in patients with QRS <130 ms in the Cardiac-Resynchronization Therapy in Heart Failure with a Narrow QRS Complex trial failed to improve outcomes in the study population prompting halting the study due to safety concerns [52]. As a result, failure of echocardiographic assessment of mechanical dyssyncrhony opened the avenue for alternative imaging approach.

Various MRI-based methods and indices have been established to improve CRT patient selection and optimize pacing algorithm. CMR may be useful to predict the CRT responsiveness and improve patient selection and predict long term outcomes. Techniques such as cine myocardial tagging, harmonic phase analysis, and strain-encoded MRI offer a more comprehensive evaluation of ventricular dyssyncrhony [53, 54]. Taylor AJ et al. demonstrated the utility of multisequential CMR in predicting CRT response. Mechanical dyssyncrhony identified as ≥65 ms delay between septal and posterolateral wall contraction on cine imaging in combination with lack of transmural scarring of the anteroseptal or posterolateral wall on delayed contrast-enhanced imaging were labeled as CRT-responders. The sensitivity of CMR-predicted CRT clinical response was 90%, with a specificity of 59%. Additionally, transplant-free survival post-CRT was observed in 88% in CMR-predicted CRT responders vs. 58% in CMR-predicted non-responders [55].

Myocardial tagging is a CMR technique that uses temporary tags applied to the myocardium to monitor myocardial deformation and motion throughout the cardiac cycle. Bilchick KC et al. using a circumferential mechanical dyssynchrony index (Circumferential Uniformity Ratio Estimate or CURE), a value derived from myocardial tagging, was able to predict clinical response in CRT HF group with 90% accurate rate. The addition of scar imaging improved the accuracy rate to 95%. This study demonstrates the use of myocardial tagging-CURE combined with scar imaging predicts clinical responsiveness after CRT with 95% accuracy rate [56]. Chalil S et al. evaluated the Cardiovascular Magnetic Resonance-Tissue Synchronization Index (CMR-TSI) ability to independently predict major cardiovascular events in CRT patients. CMR-TSI ≥110 ms was associated with worse outcomes including death or unplanned HF hospitalization [57].

8.3 Artificial intelligence

Artificial Intelligence (AI) and Machine Learning (ML) is at the forefront of cardiovascular medicine. CRT-respondent patient selection remains challenging. AI and ML may be able to provide better classification and characterization of heart failure patients with ventricular dyssyncrhony. Cikes M et al. applied unsupervised machine learning algorithm on 1106 HF patients from the MADIT-CRT dataset utilizing echocardiographic and clinical data to phenotype heart failure patients to identify CRT respondents. This algorithm, groups patients with similarities in clinical parameters, LV volume, and deformation traces at baseline. Four phenogroups were identified, two of these groups had higher clinical characteristics predictive of CRT response [58]. Similarly, Feeny AK et al. used a ML algorithm incorporating 9 variables including QRS morphology, QRS duration, New York Heart Association classification, left ventricular ejection fraction, end-diastolic diameter, sex, ischemic cardiomyopathy, atrial fibrillation, and epicardial left ventricular lead. This model was superior to guidelines in predicting CRT responsiveness and survival [59]. AI and ML serve as an essential building block in predicting CRT responsiveness and outcomes.

9. Conclusion

Decades of research has led to impressive gains in knowledge of the interplay between cardiac conduction abnormalities, symptoms, morbidity and mortality. Impressive strides have been made at correcting electromechanical dyssynchrony. The evidence supporting CRT benefits in selected patients with symptomatic HFrEF and dyssyncrhony is overwhelming. CRT therapy improves quality of life and reduces mortality. When CRT is not feasible, CSP provides a good alternative. Future directions at better patient selection and improved technology hold great promise.

Conflict of interest

The authors declare no conflict of interest.