Residency · Residency · Cardiology

Cardiac Resynchronization Therapy

Pathophysiology of Dyssynchrony

Electrical Dyssynchrony

Left bundle branch block causes delayed activation of the left ventricular lateral wall by 40 to 80 milliseconds after septal activation, creating both interventricular dyssynchrony where the right ventricle contracts before the left ventricle and intraventricular dyssynchrony characterized by a septal flash with lateral wall delay. The left bundle branch block pattern produces paradoxical septal motion, wherein early septal contraction against the still-relaxed lateral wall generates wasted mechanical work, followed by delayed lateral wall contraction and prolonged isovolumic contraction and relaxation times. Non-LBBB patterns including right bundle branch block and intraventricular conduction delay have less clear evidence for dyssynchrony response, as right ventricular-dominant delay does not produce the same degree of left ventricular mechanical inefficiency.

Mechanical Dyssynchrony

Dyssynchronous contraction reduces stroke volume by 20 to 30% compared to synchronous activation at the same level of contractility. This inefficiency increases left ventricular end-systolic volume due to impaired ejection, exacerbates mitral regurgitation through dyssynchronous papillary muscle contraction, and elevates filling pressures. Cardiac resynchronization therapy corrects this dyssynchrony by simultaneously activating the septum through native conduction or the right ventricular lead and the lateral wall through the left ventricular lead positioned in the coronary sinus, thereby restoring coordinated contraction.

Indications

Class I (Strongest Evidence)

The strongest indication for CRT encompasses patients with a left ventricular ejection fraction of 35% or below, left bundle branch block morphology, QRS duration of 150 milliseconds or greater, NYHA Class II through ambulatory Class IV symptoms, optimized guideline-directed medical therapy, and sinus rhythm. This population has the most robust evidence for reductions in both mortality and heart failure hospitalization. The COMPANION trial demonstrated that CRT-D reduced all-cause mortality by 36% compared with medical therapy. MADIT-CRT showed that CRT-D reduced heart failure events by 34% in NYHA Class I to II patients with ejection fraction of 30% or below and QRS of 130 milliseconds or greater, with the left bundle branch block subgroup driving all benefit. The RAFT trial demonstrated a 25% reduction in death or heart failure hospitalization with CRT-D compared to ICD alone in NYHA Class II to III patients with QRS of 130 milliseconds or greater.

Class IIa (Reasonable)

CRT is considered reasonable for patients with an ejection fraction of 35% or below, left bundle branch block, and QRS duration of 120 to 149 milliseconds in NYHA Class II to IV, though the benefit is less pronounced than with QRS of 150 milliseconds or greater and requires individual assessment. Patients with an ejection fraction of 35% or below, non-LBBB morphology, and QRS of 150 milliseconds or greater in NYHA Class II to IV have limited evidence and are less likely to respond than those with LBBB, warranting case-by-case consideration. Patients with AV block who require ventricular pacing with an anticipated pacing burden of 40% or greater should receive CRT rather than isolated right ventricular pacing to avoid pacing-induced cardiomyopathy, as supported by the BLOCK HF trial. Patients with existing pacemakers or ICDs who have right ventricular pacing exceeding 40% and worsening heart failure are candidates for CRT upgrade.

Class III (Not Recommended)

CRT is not recommended for patients with QRS duration less than 120 milliseconds, as demonstrated definitively by the EchoCRT trial which was terminated early for futility and a signal of harm. This finding established that mechanical dyssynchrony on echocardiography does not predict CRT response in patients with narrow QRS. CRT is also not recommended for asymptomatic NYHA Class I patients with non-LBBB morphology or for patients with expected survival of less than one year from non-cardiac conditions.

Special Populations

The benefit of CRT in atrial fibrillation is less well established than in sinus rhythm. AV nodal ablation to ensure biventricular pacing exceeding 99% improves outcomes in this population, as demonstrated by the CERTIFY and APAF-CRT trials, while rate control alone is often insufficient to achieve an adequate biventricular pacing percentage. For patients with right ventricular pacing-induced cardiomyopathy, upgrading to CRT is supported by the BLOCK HF trial, which demonstrated CRT superiority over right ventricular pacing in patients with AV block and ejection fraction of 35 to 50%.

Device and Lead Placement

Coronary Sinus Lead (LV Lead)

The standard approach for left ventricular lead placement involves subclavian or axillary venous access, advancement of a guide catheter into the coronary sinus ostium, venography to map the coronary sinus tributaries, and positioning of the lead in the target vein. The optimal target position is the lateral or posterolateral left ventricular wall, corresponding to the latest activated segment in left bundle branch block. Anterior vein placement, which positions the lead near the septum, and apical positions provide less benefit and should be avoided.

Scar-guided placement positions the lead away from myocardial scar, identified by late gadolinium enhancement on cardiac MRI or voltage mapping, because pacing scar does not generate effective contraction. The TARGET and STARTER trials supported this imaging-guided approach to left ventricular lead placement. Quadripolar left ventricular leads with four pacing electrodes allow selection of multiple pacing vectors without lead repositioning, avoid phrenic nerve stimulation that causes diaphragmatic pacing, and reduce the need for left ventricular lead revision. Lead dislodgement rates have improved from 6 to 10% with older leads to less than 3% with quadripolar leads. Challenges during implantation include difficulty cannulating the coronary sinus, limited target veins, phrenic nerve stimulation, high pacing thresholds, and lead dislodgement.

Conduction System Pacing (CSP) -- Emerging Alternative

His bundle pacing involves pacing directly at the His bundle to recruit the native conduction system, achieving physiologic ventricular activation that corrects left bundle branch block in approximately 70% of cases. Advantages include achievement of a narrow QRS through corrected or native conduction without the need for a coronary sinus lead. Limitations include higher pacing thresholds of 2 to 4 volts, lower sensing amplitudes, potential for loss of capture over time, and technical difficulty.

Left bundle branch area pacing represents an important evolution, with the lead positioned on the right ventricular septum and the deep septal screw penetrating to the left bundle branch region to capture left bundle branch fibers. Compared to His bundle pacing, left bundle branch area pacing offers lower and more stable thresholds, better sensing, greater technical ease, high R-wave amplitudes, and more consistent capture. Criteria for confirming left bundle branch capture include recording of a left bundle branch potential, a V6 R-wave peak time of less than 75 milliseconds, and V1 morphology transition from left bundle branch block to right bundle branch block pattern with increasing output. LOT-CRT, or left bundle optimized CRT, combines a left bundle branch area pacing lead with a coronary sinus lead for optimal resynchronization and is an emerging concept.

<image> A detailed illustration of CRT lead positioning showing a posterior view of the heart with the coronary venous system. The coronary sinus is shown opening into the right atrium, with its tributaries clearly labeled: great cardiac vein (anterior interventricular groove), lateral marginal vein, posterolateral vein (ideal target for LV lead, highlighted in green with a star), middle cardiac vein (posterior interventricular groove). A quadripolar LV lead is shown positioned in the posterolateral vein with four electrode positions labeled (D1, M2, M3, P4 from distal to proximal). A CS venogram image is shown as an inset, displaying the contrast-filled venous anatomy with branches labeled. The RV lead is shown at the RV septum, and the RA lead is in the RA appendage. Include a separate small inset showing the conduction system pacing alternative: a close-up cross-section of the interventricular septum with the LBBAP lead screw penetrating from the RV side through the septum to reach the left bundle branch fibers, with the His bundle, left bundle branch, and Purkinje fibers labeled. Use anatomic coloring with veins in blue/purple, pacing leads in distinct colors (RA lead: blue, RV lead: red, LV lead: green). </image>

Optimization and Response Assessment

CRT Response Definition

Approximately 65 to 70% of patients are classified as responders, defined by symptom improvement of at least one NYHA class, an increase in ejection fraction of at least 5 to 10% absolute, or a reduction in left ventricular end-diastolic volume of at least 15%, indicating reverse remodeling. Super-responders, comprising approximately 15 to 20% of patients, achieve normalization of ejection fraction to 50% or greater with dramatic symptom improvement, and discussion of ICD deactivation may be appropriate if the normalized ejection fraction is sustained. Non-responders, representing approximately 30 to 35% of patients, show no significant clinical or echocardiographic improvement.

Predictors of Response

The strongest predictors of CRT response are left bundle branch block morphology, QRS duration of 150 milliseconds or greater, female sex, and non-ischemic etiology. Moderate predictors include the absence of extensive scar as assessed by cardiac MRI, viable myocardium at the left ventricular lead site, and greater baseline left ventricular dyssynchrony. Factors predicting poor response include non-LBBB morphology, QRS less than 150 milliseconds, ischemic cardiomyopathy with extensive scar, right ventricular failure, severe mitral regurgitation that may require mitral valve intervention, and atrial fibrillation without AV nodal ablation.

AV and VV Optimization

AV delay optimization adjusts the AV interval to maximize diastolic filling by finding the longest AV delay that does not truncate the atrial contribution on mitral inflow. This iterative method uses echo-guided assessment, with typical optimal AV delays of 100 to 140 milliseconds for sensed events and 130 to 170 milliseconds for paced events. VV delay optimization evaluates sequential left ventricle-first, simultaneous, or right ventricle-first pacing and adjusts timing to maximize stroke volume or narrow QRS width, with left ventricle-first pacing by 20 to 40 milliseconds often being optimal. Automated algorithms including AdaptivCRT from Medtronic and SyncAV from Abbott adjust AV and VV timing based on intrinsic conduction and electrograms, reducing the need for manual optimization.

Ensuring Adequate Biventricular Pacing

The target biventricular pacing percentage is at least 98% for maximal benefit, with percentages below 92% associated with worse outcomes. Common causes of reduced biventricular pacing include atrial fibrillation with rapid ventricular response, which is the most common cause, frequent premature ventricular complexes, loss of left ventricular lead capture, and atrial pacing modes not tracking properly. Atrial fibrillation management should target aggressive rate control with a heart rate below 70 to 80 beats per minute to allow biventricular pacing, and AV nodal ablation should be performed if the biventricular pacing percentage remains below 95% despite rate control, as supported by the CERTIFY and APAF-CRT trials. When premature ventricular complexes exceed 10% of total beats and reduce biventricular pacing, antiarrhythmic therapy or PVC ablation should be considered.

Outcomes and Landmark Trials

Major CRT Trials Summary

The COMPANION trial in 2004 demonstrated that both CRT-P and CRT-D reduced heart failure hospitalization, with CRT-D achieving a 36% reduction in mortality. CARE-HF in 2005 provided the strongest evidence for CRT-P mortality benefit, showing a 36% mortality reduction at 3 years with CRT-P alone without an ICD component. MADIT-CRT in 2009 demonstrated a 34% reduction in heart failure events with CRT-D in NYHA Class I to II patients with ejection fraction of 30% or below and QRS of 130 milliseconds or greater, with the left bundle branch block subgroup driving all benefit. RAFT in 2010 showed a 25% reduction in death or heart failure hospitalization with CRT-D in NYHA Class II to III patients, including a significant mortality reduction. REVERSE in 2008 demonstrated that CRT slowed disease progression in mildly symptomatic patients with NYHA Class I to II HFrEF. BLOCK HF in 2013 established CRT superiority over right ventricular pacing in patients with AV block and ejection fraction of 35 to 50%, supporting CRT for pacing-dependent patients with reduced ejection fraction.

CRT-P vs. CRT-D

CRT-D provides defibrillation backup and is standard for most patients who meet ICD criteria. CRT-P alone is reasonable for patients who do not meet primary prevention ICD indications, elderly patients with limited life expectancy, and patients who decline ICD therapy. COMPANION showed that CRT-D achieved a significant mortality benefit over medical therapy, while CRT-P trended toward benefit but did not reach statistical significance due to being underpowered. The DANISH subgroup analysis in non-ischemic cardiomyopathy suggests that CRT-P may be sufficient given the less established benefit of the ICD component, with the decision best approached through shared decision-making.

<image> A before-and-after echocardiographic comparison showing CRT response. Two columns labeled "Before CRT" and "After CRT (6 months)." Top row: Apical 4-chamber view with LV volume tracings. Before: severely dilated LV with LVEDV 280 mL, LVESV 210 mL, EF 25%, severe functional MR (large color jet). After: reduced LV volumes, LVEDV 200 mL, LVESV 100 mL, EF 50%, trivial MR. Middle row: M-mode through LV showing septal motion. Before: paradoxical septal motion (septal flash) with delayed lateral wall contraction, interventricular mechanical delay marked. After: normalized synchronous septal and lateral wall motion, no septal flash. Bottom row: GLS bullseye strain maps. Before: heterogeneous strain with severely reduced basal/mid segments (red/yellow), preserved apical strain, and temporal dispersion marked. After: uniform strain improvement across all segments (blue tones), reduced mechanical dispersion. Include numerical values for LVESV, LVEDV, EF, and MR grade in each panel. Use consistent color scales. </image>

Troubleshooting CRT Non-Response

Systematic Approach

A systematic evaluation of CRT non-response begins with verification of the biventricular pacing percentage at 98% or greater on device interrogation. The next step is confirming adequate left ventricular lead position, ensuring it is not in an anterior vein or positioned over scar, and considering lead revision or an alternative pacing strategy if placement is suboptimal. AV and VV optimization should be performed under echocardiographic guidance. Ongoing atrial fibrillation must be addressed through rate control or AV nodal ablation. Guideline-directed medical therapy should be reviewed to ensure all four HFrEF pillars are maximized. Other causes of heart failure progression should be evaluated, including significant mitral regurgitation that may benefit from transcatheter edge-to-edge repair, progressive valvular disease, arrhythmias, and ischemia. Conduction system pacing with His bundle pacing or left bundle branch area pacing should be considered as a rescue strategy for patients who have not responded to coronary sinus lead-based CRT.

Emerging Strategies

Multisite left ventricular pacing using two left ventricular leads or multipoint pacing from multiple electrodes on a quadripolar lead may improve synchrony. Endocardial left ventricular pacing with the WiSE-CRT wireless endocardial system uses ultrasound-based leadless left ventricular pacing to overcome the limitations of coronary sinus lead placement. Left bundle branch area pacing as a CRT alternative has growing evidence supporting similar or superior outcomes compared to traditional CRT, with the advantage of avoiding the challenges of coronary sinus cannulation.

Key Clinical Pearls

• The triad of LBBB + QRS >= 150 ms + non-ischemic etiology predicts the highest likelihood of CRT super-response -- these patients should be urgently referred for CRT
• BiV pacing percentage must be >= 98% for optimal CRT benefit -- even 5% loss of BiV pacing (from AF, PVCs, or undersensing) significantly attenuates the benefit; AV nodal ablation should be strongly considered in AF patients with CRT when BiV% cannot be maintained
• EchoCRT definitively showed that CRT does NOT help patients with QRS < 120 ms regardless of echocardiographic dyssynchrony -- do not implant CRT based on echo dyssynchrony alone
• LV lead position matters: lateral/posterolateral wall (away from scar) is optimal; quadripolar leads offer multiple pacing vectors and dramatically reduce the need for lead revision
• LBBAP is emerging as a viable alternative or rescue strategy for traditional CRT, with advantages of lower thresholds, stable capture, and physiologic activation -- fellowship training should include exposure to conduction system pacing techniques
• CRT super-responders (EF normalization) pose a unique clinical dilemma regarding ICD generator replacement at end-of-battery life -- discuss ongoing need for defibrillation backup vs. switching to CRT-P

References

• Glikson M, et al. 2021 ESC Guidelines on Cardiac Pacing and CRT. Eur Heart J. 2021;42:3427-3520.
• Bristow MR, et al. Cardiac Resynchronization Therapy with or without an Implantable Defibrillator in Advanced Chronic Heart Failure (COMPANION). NEJM. 2004;350:2140-2150.
• Moss AJ, et al. Cardiac Resynchronization Therapy for the Prevention of Heart-Failure Events (MADIT-CRT). NEJM. 2009;361:1329-1338.
• Cleland JGF, et al. The Effect of Cardiac Resynchronization on Morbidity and Mortality in Heart Failure (CARE-HF). NEJM. 2005;352:1539-1549.
• Vijayaraman P, et al. Left Bundle Branch Area Pacing for Cardiac Resynchronization Therapy: Results from the International LBBAP Collaborative Study Group. JACC Clin Electrophysiol. 2021;7:135-147.

Read this lecture as Markdown