CRT (Cardiac Resynchronization Therapy) Devices: How They Work, What the Evidence Shows

2. What It Is

Cardiac resynchronization therapy (CRT) is a specialized form of cardiac pacing designed to restore synchronous ventricular contraction in patients with heart failure and ventricular dyssynchrony, usually evidenced by a wide QRS complex with left bundle branch block morphology.

A CRT system consists of:

The CRT pulse generator: Larger than a standard pacemaker generator, housing a three-lead interface (right atrium, right ventricle, left ventricle), advanced programming capabilities, and in the case of CRT-D, a full high-voltage defibrillation capacitor and shock coil circuitry.

Right atrial lead: Standard pacing/sensing lead in the right atrial appendage. Senses native atrial activity to enable AV synchrony.

Right ventricular lead: Standard pacing lead at the RV apex or septum. Paces and senses the right ventricle.

Left ventricular lead: The defining element of CRT. This lead is advanced through the right heart, through the coronary sinus (the venous drainage of the left ventricle that runs in the posterior AV groove), and into a lateral or posterolateral branch of the coronary venous system. This positions the pacing tip on the lateral or posterolateral left ventricular wall, which is typically the last segment to activate in LBBB. Pacing this site simultaneously with (or slightly before) the RV pacing stimulus resynchronizes biventricular activation.

Two device variants:

• CRT-P (CRT pacemaker): Resynchronization pacing function only. No defibrillation capability. Smaller generator. Battery life approximately 7-10 years.
• CRT-D (CRT defibrillator): Resynchronization pacing plus high-energy defibrillation capability (like a full ICD). Indicated when the patient has both a CRT indication and an ICD indication (which is most patients with LVEF below 35% and heart failure). Battery life approximately 5-7 years.

Major manufacturers: Medtronic (Claria, Compia families), Abbott (Quadra Assura, Resonate families), Boston Scientific (Resonate, Autogen CRT-D families). All provide FDA PMA-approved devices with MRI-conditional labeling available on current-generation systems.

---

3. The Mechanism

3.1 Ventricular Dyssynchrony in Left Bundle Branch Block

The normal His-Purkinje system activates both ventricles simultaneously through the left and right bundle branches. The right bundle activates the right ventricle; the left bundle, via its anterior and posterior fascicles, activates the left ventricle in a coordinated apical-to-basal, septal-to-lateral sequence.

When the left bundle branch fails (LBBB), the left ventricle is activated exclusively through slow cell-to-cell conduction spreading from the right ventricle across the interventricular septum. This is 3-5 times slower than His-Purkinje conduction. The result:

• Septal wall: Activates early (receiving activation from the RV side)
• Lateral wall: Activates last (activation reaches it after slow cell-to-cell spread from septum to lateral wall)
• Dyssynchrony: The septum contracts (and may paradoxically move inward, compressing the LV cavity, before the lateral wall is activated) while the lateral wall is still relaxed
• Net effect: Wasted energy, reduced stroke volume, reduced cardiac efficiency

The QRS duration reflects this conduction delay: a QRS above 150 ms with LBBB morphology in a patient with heart failure represents the most extreme dyssynchrony subgroup with the clearest benefit from CRT.

3.2 How CRT Corrects Dyssynchrony

The LV lead paces the lateral wall at the same moment the RV lead paces (or senses) the right ventricle. By activating the lateral wall from the outside (epicardial pacing via the coronary venous system) at the same time the septal wall is activated from the inside, both walls contract simultaneously.

The programmed interventricular delay (VV delay) allows fine-tuning: the LV can be pre-excited (paced slightly before the RV) to compensate for the additional activation time needed to reach the late-activating lateral wall. Target VV delay programming improves hemodynamic response.

The mechanical consequence of resynchronization:

• Both ventricular walls contract together
• No wasted pre-systolic septal contraction against a relaxed lateral wall
• Improved stroke volume
• Reduced mitral regurgitation (secondary MR from annular dilation and dyssynchrony improves as the LV contracts more symmetrically)
• Over weeks to months: reverse remodeling (LV dilatation partially reverses, LVEF increases)

3.3 LV Lead Positioning: The Coronary Sinus Challenge

The coronary sinus is the main cardiac vein, running in the posterior AV groove and receiving tributaries from the left ventricular free wall. Accessing it during CRT implantation requires cannulating it from the right side of the heart via a specially shaped guide catheter.

Ideal LV lead target: a lateral or posterolateral branch of the coronary venous system, positioned at the site of latest mechanical activation (typically the lateral wall in LBBB). Imaging techniques (speckle-tracking echocardiography, cardiac MRI mechanical mapping) can identify the latest-activating segment and guide lead placement.

Technical challenges:

• Coronary venous anatomy is highly variable; a suitable lateral branch may not exist
• The target branch may be too small, too tortuous, or too acute-angled for lead passage
• The LV lead can dislodge from the coronary venous branch (lead dislodgement is the most common complication of CRT implantation, occurring in approximately 2-5% of cases)
• Phrenic nerve stimulation (the LV lead pacing tip may stimulate the left phrenic nerve if positioned too anteriorly, causing diaphragmatic contraction with each LV pacing stimulus)

When coronary venous anatomy does not allow target LV lead placement: epicardial LV lead placement (surgical approach through a lateral mini-thoracotomy) or LV endocardial pacing via transseptal approach are alternatives, less commonly performed.

3.4 CRT-D: The Dual Function Device

Most patients who meet CRT criteria also meet ICD criteria (LVEF below 35%, NYHA Class II-III, on maximally tolerated GDMT for at least 3 months). CRT-D combines both functions in a single device:

• CRT pacing: biventricular resynchronization therapy
• ICD therapy: detection of VF and VT with high-energy shock therapy and antitachycardia pacing

The CRT-D generator is larger than CRT-P (approximately 40-80 cc vs 20-40 cc for CRT-P) due to the high-voltage capacitors required for defibrillation. The implantation procedure is identical except for the shock coil on the RV lead.

---

4. How It Is Used

4.1 Indications for CRT

The 2022 ACC/AHA/HFSA Heart Failure Guideline and the 2021 ESC Heart Failure Guidelines provide the current evidence-based indication framework.

Class I (indicated):

• LVEF ≤35%, LBBB with QRS ≥150 ms, NYHA II-III-IV, on maximally tolerated GDMT for ≥3 months, sinus rhythm

This is the highest-evidence indication. The landmark trials (COMPANION, CARE-HF, MADIT-CRT, RAFT) all demonstrated benefit in this population.

Class IIa (reasonable):

• LVEF ≤35%, LBBB with QRS 120-149 ms, NYHA II-III-IV, on maximally tolerated GDMT, sinus rhythm
• LVEF ≤35%, non-LBBB (RBBB or nonspecific IVCD) with QRS ≥150 ms, NYHA III-IV (evidence is weaker than LBBB)
• LVEF ≤35%, NYHA II-IV, with anticipated high-burden RV pacing (pacing >40% of beats) where RV pacing alone would worsen LV function

Class III (not indicated):

• NYHA Class IV refractory heart failure (expected 1-year survival <1 year)
• QRS duration <120 ms without anticipated high-burden RV pacing
• Non-LBBB morphology with QRS <150 ms

The key clinical takeaway: the benefit of CRT is strongly concentrated in patients with LBBB (not RBBB), QRS ≥150 ms (not 120-150 ms), and LVEF ≤35%. The ECHO CRT trial showed that CRT in QRS <130 ms actually increased hospitalization risk (potential harm). 5 / Solid

4.2 The CRT-P vs CRT-D Decision

Most patients who meet CRT criteria also have an ICD indication (LVEF ≤35%). CRT-D is the standard in this population.

CRT-P without defibrillation capability is appropriate when:

• The patient has declined ICD therapy after informed discussion
• The patient has advanced age (≥80 years) with significant competing comorbidities where survival benefit from ICD is unclear
• The patient has NYHA Class IV symptoms not expected to improve with CRT (in whom sudden death prevention from ICD is less certain)
• The patient is already being considered for LVAD or transplant (bridging strategy)

The CRT-P vs CRT-D decision is individualized. A 72-year-old with LVEF 28%, LBBB, QRS 170 ms, and no other significant comorbidities benefits from CRT-D. A 79-year-old with LVEF 25%, LBBB, QRS 160 ms, chronic kidney disease stage 4, and dementia may be better served by CRT-P to relieve heart failure symptoms without the complexity of ICD management.

4.3 The Non-Responder Problem

Approximately 30% of patients who receive CRT by guideline-based criteria do not respond (defined as no measurable improvement in symptoms, LV dimensions, or LVEF at 6 months). This is not a device failure; it is a patient-selection and physiological complexity problem.

Known predictors of CRT non-response:

• Non-LBBB morphology (RBBB, nonspecific IVCD): lower response rates than LBBB
• QRS 120-149 ms (vs ≥150 ms): lower response rate
• AF vs sinus rhythm: AF patients have lower response rates; AV node ablation to ensure reliable biventricular capture can improve response in select AF patients
• Below-target LV lead position (not at latest-activating segment)
• Extensive LV scar (by cardiac MRI or PET): fibrotic scar at the LV pacing site prevents the paced stimulus from producing effective mechanical activation

For patients who do not respond to conventional CRT: reassess lead positioning, improve programming, consider His-bundle or left bundle branch pacing as alternatives or adjuncts. If a scar is present at the conventional LV lead site, repositioning to a scar-free segment may convert a non-responder to a responder. 4 / Promising

4.4 Geographic Access

CRT implantation requires electrophysiology expertise and cardiac surgery backup. Not all cardiology programs in Illinois perform CRT. Major centers in Illinois with CRT programs include:

• University of Illinois Chicago (UIC)
• Northwestern Memorial Hospital (Chicago)
• Rush University Medical Center (Chicago)
• University of Chicago Medicine
• Carle Foundation Hospital (Urbana)
• OSF HealthCare Saint Francis Medical Center (Peoria)
• Memorial Medical Center / SIU Medicine (Springfield)

For patients in rural Illinois, referral to the nearest academic medical center is appropriate for CRT evaluation. CRT follow-up (device interrogations, programming improvement) can often be performed remotely or at a regional center after the initial implant at an academic center.

---

5. The Evidence

5.1 COMPANION Trial (Bristow 2004, NEJM)

This is the first large RCT demonstrating CRT mortality benefit.

Design: 1,520 patients with NYHA Class III-IV HF, LVEF ≤35%, QRS ≥120 ms. Randomized 1:2:2 to medical therapy alone (OPT), CRT-P, or CRT-D.

Primary outcome: Death or all-cause hospitalization.

Results:

• CRT-P vs OPT: HR 0.81 (95% CI 0.69-0.96), p=0.015
• CRT-D vs OPT: HR 0.80 (95% CI 0.68-0.95), p=0.010
• All-cause mortality (secondary endpoint): CRT-D vs OPT: HR 0.64 (95% CI 0.48-0.86), p=0.003

CRT-D significantly reduced all-cause mortality by 36% relative to medical therapy.

5 / Solid

5.2 CARE-HF Trial (Cleland 2005, NEJM)

The definitive European mortality benefit trial for CRT.

Design: 813 patients with NYHA Class III-IV HF, LVEF ≤35%, QRS ≥120 ms (QRS 120-149 ms required echocardiographic dyssynchrony criteria). Randomized to CRT-P vs medical therapy.

Results: Primary outcome (all-cause mortality or hospitalization for major cardiovascular event): HR 0.63 (95% CI 0.51-0.77). All-cause mortality: HR 0.64 (95% CI 0.48-0.85). Mean LVEF improvement: +3.7 percentage points (CRT) vs no change (medical therapy). NYHA class improvement: significant.

5 / Solid

CARE-HF established that CRT-P (without the ICD component) reduces mortality, not merely from sudden death prevention, but through reverse cardiac remodeling and improvement in heart failure progression.

5.3 MADIT-CRT Trial (Moss 2009, NEJM)

The pivotal trial for CRT in NYHA Class I-II patients.

Design: 1,820 patients with LVEF ≤30%, QRS ≥130 ms (LBBB or non-LBBB), NYHA Class I or II. Randomized 3:2 to CRT-D vs ICD alone.

Results: Primary outcome (death or non-fatal heart failure event): CRT-D vs ICD: HR 0.66 (95% CI 0.52-0.84). Reduction driven largely by heart failure events, not mortality.

LBBB subgroup: HR 0.47 (95% CI 0.34-0.65) for LBBB. Benefit was substantially concentrated in LBBB patients.

Non-LBBB subgroup: HR 1.11 (95% CI 0.71-1.75). No significant benefit in non-LBBB.

5 / Solid

MADIT-CRT established CRT-D benefit in earlier (Class I-II) heart failure with LBBB and expanded the indication beyond Class III-IV.

5.4 RAFT Trial (Tang 2010, NEJM)

Design: 1,798 patients with NYHA Class II-III HF, LVEF ≤30%, QRS ≥120 ms (intrinsic) or ≥200 ms (paced). Randomized to CRT-D vs ICD alone.

Results: Death or hospitalization for heart failure: HR 0.75 (95% CI 0.64-0.87). All-cause mortality: HR 0.75 (95% CI 0.62-0.91).

5 / Solid

RAFT was the first large RCT to demonstrate CRT-D mortality superiority over ICD alone in Class II-III HF, confirming the MADIT-CRT signal.

5.5 ECHO-CRT: When CRT Causes Harm

The ECHO-CRT trial is the necessary counterweight to the above positive trials.

Design: 809 patients with LVEF ≤35%, NYHA Class III-IV, QRS <130 ms but echocardiographic dyssynchrony. Randomized to CRT-D vs ICD alone.

Results: Trial stopped early for futility and possible harm. Death or first hospitalization for worsening HF: 28.7% CRT-D vs 25.5% ICD (HR 1.20, 95% CI 0.92-1.57). All-cause mortality: CRT-D 11.1% vs ICD 6.4% (excess mortality in CRT-D arm, p=0.02 at interim analysis, basis for stopping).

5 / Solid

Interpretation: QRS duration, not echocardiographic dyssynchrony assessment, is the reliable patient selection criterion. Narrow-QRS patients (<130 ms) should not receive CRT regardless of echocardiographic dyssynchrony findings. The ECHO-CRT result was practice-changing: it ended the decade-long effort to expand CRT to patients with narrow QRS and echocardiographic dyssynchrony.

5.6 CRT-D Non-Responder Quantification

In pooled analyses of CRT trials, approximately 30-35% of patients do not respond (defined as no improvement in NYHA class, no reduction in LV end-systolic volume, no improvement in exercise capacity). The rate of super-responders (LVEF normalization, often ≥50%) is approximately 20-25% in target-selection populations.

The non-responder rate has not been eliminated by device improvements alone; patient selection, LV lead position improvement, and programming remain the primary modifiable factors. 5 / Solid

5.7 Evidence Summary Table

---

6. The Patient Experience

6.1 The Implantation Procedure

CRT implantation is more complex than standard pacemaker implantation. Procedure duration: 2-3 hours on average. The right atrial and right ventricular leads are placed via subclavian or axillary vein access as for a standard pacemaker. The coronary sinus lead placement requires coronary sinus cannulation via a specialized guide catheter and fluoroscopic advancement of the LV lead through the coronary venous system.

The patient typically receives moderate IV sedation (not general anesthesia). Local anesthetic is administered at the pocket site. A defibrillator is available in the room throughout the procedure.

Post-procedure: one to two overnight stays for monitoring, chest X-ray to confirm lead position and exclude pneumothorax, device interrogation to confirm adequate pacing and sensing thresholds, discharge with activity restrictions.

The LV lead is the most technically demanding element of the implant. In some patients with unfavorable coronary venous anatomy, achieving adequate LV lead position requires significant fluoroscopy time and may occasionally require a hybrid surgical approach.

6.2 The Post-Implant Course

For CRT responders, the expected timeline of response:

• 1-3 months: Patient notices improved exercise tolerance. May not be measurable yet on echocardiogram.
• 3-6 months: Echocardiogram typically shows LVEF improvement (mean response in trials: 5-10 percentage points; super-responders: 20+ point improvement). LV dimensions reduce (reverse remodeling). Mitral regurgitation grade often improves.
• 6-12 months: NYHA class improvement documented in responders. Some patients can be considered for GDMT titration adjustment (reduction of diuretics, consideration of beta-blocker uptitration).

For non-responders, the absence of clinical or echocardiographic improvement at 6 months triggers a reassessment: Is the biventricular pacing actually occurring (is the patient capturing on all three leads)? Is the LV lead position target? Is the AV/VV delay programming improved? Is AF preventing reliable CRT capture?

6.3 Device Alerts and Remote Monitoring

CRT devices transmit daily or nightly via home communicators. The cardiologist receives alerts for:

• Low biventricular pacing percentage (suggesting lead issues or AF with irregular RR preventing reliable LV pacing)
• VT/VF detection or therapy delivered (in CRT-D)
• AF detection (important because AF reduces CRT benefit and may trigger rate-control or rhythm-control changes)
• Battery status decline
• Lead impedance changes (suggesting lead fracture or dislodgement)

The target for biventricular pacing in CRT patients is >98%. The data show that CRT benefit diminishes rapidly below 98% biventricular pacing: patients with <90% biventricular pacing (often from AF with irregular RR intervals) may lose clinical CRT benefit. 5 / Solid

6.4 The ICD Component in CRT-D

The ICD function of a CRT-D device operates exactly as a standalone ICD. VT detection zones are programmed (typically with antitachycardia pacing (ATP) for slower VT and shock therapy for fast VT or VF). An ICD shock in a CRT-D patient is uncomfortable but rarely dangerous; the device works as programmed.

Patients with CRT-D who receive an ICD shock:

• Call their cardiologist or go to the emergency department for same-day evaluation
• Bring the CRT-D identification card to the ED
• Expect a remote or in-person interrogation to review the stored electrogram and determine whether the shock was appropriate (for VT/VF) or inappropriate (T-wave oversensing, SVT with rapid rate)

Multiple ICD shocks in a short period (“electrical storm”): this requires urgent hospitalization. CRT does not prevent electrical storm, and CRT-D programming adjustments alone may not control it.

---

7. Decisions and Trade-Offs

7.1 When to Implant: The 3-Month Rule and Its Exceptions

The Class I indication requires at least 3 months of maximally tolerated GDMT before CRT implantation. This window allows medical therapy to produce its maximal LVEF recovery: many patients whose LVEF is 25% at diagnosis will improve to 35-40% with GDMT alone, eliminating the CRT indication.

The 3-month rule has exceptions:

• Newly diagnosed cardiomyopathy with LVEF 20% and hemodynamic instability: hospitalization management takes priority over waiting
• Patient with LVEF 20%, LBBB QRS 170 ms, and rapid clinical deterioration: earlier CRT may be appropriate
• Patient who has had an ICD placed first (prior to LVEF recovery): ICD upgrade to CRT-D at time of first generator change or earlier if LV dysfunction persists and CRT criteria are met

7.2 CRT in AF: The AV Node Ablation Decision

In patients with persistent AF and CRT indication, irregular RR intervals prevent reliable biventricular pacing. The standard approach: improve rate control to minimize native conduction intrusion. If ventricular rate during AF is irregularly irregular, native beats inhibit LV pacing, reducing biventricular pacing percentage.

AV node ablation: deliberate ablation of the AV node (via catheter radiofrequency energy) renders the patient pacemaker-dependent but enables reliable 100% biventricular pacing. Retrospective data show that AV node ablation in AF patients receiving CRT improves outcomes (functional status and survival) compared to rate-control alone. 4 / Promising This is a major clinical commitment (pacemaker dependence is permanent); it is considered in AF patients who fail rate control and have poor CRT response due to low biventricular pacing percentage.

7.3 His-Bundle and Left Bundle Branch Pacing as CRT Alternatives

Conduction system pacing is being investigated as an alternative or adjunct to conventional CRT, particularly for patients who fail to respond to conventional CRT or in whom the coronary venous anatomy does not allow target LV lead placement.

His-bundle pacing in LBBB: HBP can correct LBBB by pacing the His bundle above the block, engaging the normal conduction system below. It produces narrow QRS complexes and physiological activation without the need for a coronary sinus lead. The technical challenge: LBBB correction via HBP requires capture below the level of the block and is not always achievable.

Left bundle branch pacing (LBBP): Pacing the left bundle branch directly or via septal LBB capture produces near-native left ventricular activation. Multiple case series show QRS narrowing and LVEF improvement comparable to conventional CRT with fewer LV lead dislodgement events. 4 / Promising

Conduction system pacing as a CRT replacement strategy is not yet supported by large RCTs. It remains investigational in most settings but is increasingly performed at centers with strong EP programs. LBBP-RESYNC trial and other ongoing trials will provide more definitive evidence.

---

Clinical Synthesis

Robert’s response to CRT is the best outcome the device can produce: a patient on maximum medical therapy with documented dyssynchrony who improves from LVEF 22% to 38%, from NYHA III to NYHA II, from housebound to active.

The CRT device is, in the language of cardiac mechanics, a correction of a wiring defect. The heart muscle in Robert’s case is not irreversibly damaged. It is poorly coordinated. It contracts out of sequence, wasting much of its energy. CRT does not fix the underlying cardiomyopathy. It does not reverse the fibrosis or normalize the neurohormonal milieu. What it does: it restores mechanical synchrony, allowing the same heart muscle to work more efficiently and produce a higher cardiac output with the same metabolic cost.

This clinical framework identifies volume matters: VO2max, LVEF, cardiac reserve. For patients with HFrEF and dyssynchrony, CRT can increase LVEF by 5-15 percentage points. In a patient at LVEF 25%, a 12-point improvement to LVEF 37% represents a meaningful reduction in the mortality curve. The COMPANION and CARE-HF trials quantified this: 36% relative mortality reduction compared to medical therapy. In absolute terms, roughly 6-8 deaths prevented per 100 patients treated over 2-3 years.

Those are not abstractions. They are lives.

The 30% non-responder rate is the honest counterweight. CRT is not universally effective. The patient who receives a CRT-D, waits 6 months, and finds their LVEF unchanged, their symptoms unchanged, and their exercise capacity unchanged has received a surgical procedure with real risks (lead dislodgement, pocket infection, the ongoing obligation of device management) and obtained no benefit. Improving the identification of CRT responders before implantation, through advanced imaging (cardiac MRI scar burden, speckle tracking mechanical dyssynchrony), conduction system anatomy assessment, and ECG criteria refinement, is one of the active frontiers of heart failure electrophysiology.

For patients in structured remote monitoring with known heart failure and reduced ejection fraction who are on maximally tolerated GDMT and continue to have LVEF ≤35% with LBBB: CRT evaluation is a standard, evidence-based component of their management pathway. This program does not perform CRT implantations; it ensures that patients who meet criteria are referred to programs with the expertise and volume to improve outcomes.

---

Sex Differences in CRT Response and LBBB Diagnosis

9.1 Women Respond Better to CRT: The Evidence

One of the most consistent and clinically important findings in the CRT literature is that women have superior outcomes compared to men with CRT implantation. This is not a subtle difference. Across MADIT-CRT, RAFT, and observational registries, women with CRT have higher rates of reverse remodeling (EF improvement), lower rates of appropriate ICD therapy, and lower mortality than men at equivalent QRS duration thresholds 5 / Solid .

The magnitude: in a sex-stratified meta-analysis of MADIT-CRT, RAFT, and REVERSE, women with QRS 130-150 ms derived mortality benefit from CRT while men at the same QRS duration did not. The current guideline threshold for CRT (LBBB with QRS greater than or equal to 150 ms, Class I) was derived from data including men and women combined; the data show that a lower QRS threshold (possibly 130 ms) is appropriate specifically for women.

The mechanism for female CRT response superiority is not established with certainty. Proposed explanations include:

• Women’s hearts are smaller, so a 150 ms QRS duration represents a larger fraction of total left ventricular activation time (greater relative dyssynchrony relative to chamber size)
• Women with LBBB have more complete left bundle branch anatomy (true left bundle branch block) compared to men, in whom the wide QRS may reflect a more diffuse conduction delay rather than true LBBB
• Women have fewer ischemic cardiomyopathy-associated fibrotic substrate abnormalities that would reduce CRT response by fragmenting lead-to-myocardium coupling

9.2 The LBBB Diagnosis Problem in Women

The ECG definition of LBBB requires QRS duration of 120 ms or more with specific morphological criteria (absence of septal Q in lateral leads, broad notched R in V5-V6, RS or rS in V1). However, in women, QRS duration in health is slightly shorter than in men (average 88 ms in women vs 97 ms in men in population samples). This means that a woman with a QRS of 130 ms is proportionally more abnormal than a man with the same QRS duration.

The 2012 AHA/ACC/HRS diagnostic criteria for LBBB require QRS greater than or equal to 120 ms. This threshold was validated primarily in men. Some electrophysiology experts advocate for a sex-adjusted LBBB threshold (greater than or equal to 116-118 ms in women) to capture the same degree of conduction abnormality that 120 ms represents in men. No major society guideline has adopted sex-specific LBBB thresholds as of 2026, but the debate is ongoing and clinically relevant for CRT candidacy decisions 3 / Early .

9.3 CRT in Women Who Were Excluded from Original Trials

The COMPANION trial (2004) enrolled 30% women; CARE-HF enrolled 27% women; RAFT enrolled 22% women. The consistent pattern of female underrepresentation means that the most influential CRT trials were powered primarily on male populations. The strong subsequent sex-stratified data showing female benefit emerged from post-hoc analyses and pooled individual patient data meta-analyses, not from prospectively sex-stratified designs. This is the standard limitation of post-hoc sex analysis and should be noted with appropriate Honesty Scale tagging: the evidence for superior female CRT response is Promising to Solid in the pooled data but has not been confirmed in a prospective sex-stratified trial.

---

Technical Notes on CRT Implantation and the Coronary Sinus

10.1 The Coronary Sinus and LV Lead Placement

The left ventricular lead in a CRT device is delivered via the coronary sinus (CS), the venous drainage structure that runs in the left atrioventricular groove. The CS ostium opens into the posterior wall of the right atrium, and the main CS leads posteriorly and leftward before branching into lateral, posterior, and anterolateral tributaries. The ideal LV lead position for CRT is the lateral or posterolateral wall, where the LV activation delay is typically maximal in LBBB.

The procedure requires CS angiography to map the venous anatomy, followed by lead advancement through a guiding catheter into the target branch vein. Complications include:

• Coronary sinus dissection (1-2%; typically managed conservatively; pericardial tamponade is rare but can occur)
• LV lead dislodgement (5-8% in older systems; less than 3% with modern active-fixation LV leads)
• Phrenic nerve stimulation (5-20%): the lateral CS tributaries run in proximity to the left phrenic nerve; LV pacing at the target lead position may cause diaphragmatic stimulation requiring lead repositioning or output reduction

Approximately 5-8% of patients have coronary sinus anatomy that does not permit conventional LV lead placement (inadequate branches, venous stenosis, excessive tortuosity). These patients may be candidates for epicardial LV lead placement (surgical) or for the newer left bundle branch area pacing approach as an alternative to lateral wall CS pacing.

10.2 Biventricular Pacing vs Conduction System Pacing for CRT

Left bundle branch area pacing (LBBAP) is an emerging alternative to traditional biventricular CRT that delivers electrical stimulation to the left bundle branch in the interventricular septum, capturing the native left conduction system distal to the AV node. In theory, LBBAP provides LV synchrony by capturing the native Purkinje network rather than pacing the lateral LV epicardial surface through a CS lead.

Early single-center and multicenter registries suggest that LBBAP achieves ECG QRS narrowing and echocardiographic LV reverse remodeling comparable to traditional CRT in patients with LBBB and reduced EF 3 / Early . No large-scale RCT comparing LBBAP to traditional CRT has been completed as of mid-2026. The LBBAP vs CRT debate will likely be resolved by the LBBP-RESYNC trial and similar prospective comparative trials expected to complete enrollment in 2025-2026. Current AHA/ACC guidelines give LBBAP a Class IIb indication in patients who cannot receive a conventional CS LV lead.

10.3 Device Programming: AV and VV improvement

The therapeutic benefit of biventricular pacing depends partly on improving two programmable intervals:

AV delay (the interval between atrial pacing/sensing and ventricular pacing): An AV delay that is too short truncates LV filling; too long allows PR prolongation and delayed LV activation with below-target diastolic filling. Target AV delay maximizes stroke volume and is individualized using echocardiographic assessment of mitral inflow E and A wave separation or automated device-based improvement algorithms.

VV interval (the interval between right and left ventricular pacing pulses): Pre-exciting the LV (LV first) by 20-40 ms relative to RV pacing is target in most patients, reflecting the LBBB physiology where LV activation is delayed. In some patients with RBBB or conduction abnormalities, simultaneous or even RV-first programming may be better. VV improvement is typically performed echocardiographically using the approach validated in the SMART-AV trial paradigm.

Empiric device programming at nominal settings is adequate for many CRT patients and produces a response rate (defined as EF improvement greater than 5% at 6 months) of approximately 60-70%. Systematic AV and VV improvement increases the response rate to 70-75% in centers with dedicated CRT improvement programs 4 / Promising . However, formal improvement requires echocardiography expertise and time, and it is not universally available or reimbursed.

---

The 30% Non-Responder Problem: What Causes CRT Failure

11.1 Defining Non-Response

CRT “non-response” is typically defined as less than 5-10% improvement in LVEF at 6 months, absence of clinical improvement (no improvement in NYHA class or 6-minute walk test), or worsening heart failure requiring hospitalization. By this definition, approximately 25-35% of patients who meet guideline criteria for CRT do not respond 5 / Solid .

The non-response problem is one of the central unresolved issues in CRT therapy. Understanding it has driven both improved patient selection (excluding patients most likely to not respond) and improved implant technique (placing the LV lead at the site of latest mechanical activation rather than simply in the most accessible vein).

11.2 Causes of Non-Response

Below-target LV lead position: The most modifiable cause. Placing the LV lead in the anterior or anteroseptal position (often because the lateral and posterolateral veins are inaccessible) produces minimal resynchronization because the lead is not stimulating the wall of latest activation. Speckle-tracking echocardiography and cardiac MRI can identify the site of latest mechanical activation pre-implant and guide LV lead targeting 4 / Promising .

Scar at the LV lead site: Myocardial scar (from previous MI or any cause) at the site of LV pacing reduces the mechanical response to electrical stimulation. The LV pacing stimulus depolarizes scar tissue that does not contract, producing electrical but not mechanical resynchronization. Scar burden assessed by cardiac MRI LGE is an independent predictor of CRT non-response; patients with transmural scar in the LV lateral wall are unlikely to respond to CRT even if the lead position is anatomically correct 5 / Solid .

True LBBB vs non-specific IVCD: Patients with non-specific intraventricular conduction delay (NICD; wide QRS that does not meet full LBBB criteria) respond significantly less well to CRT than patients with true LBBB 5 / Solid . The ECHO-CRT result (harm in QRS less than 130 ms) represents an extreme version of this: patients with narrow QRS have no dyssynchrony to correct, and CRT pacing introduces artificial dyssynchrony.

Below-target device programming: AV and VV intervals at factory-default settings may be inappropriately programmed for individual patient physiology. Systematic programming improvement can convert non-responders to responders in some cases.

Non-cardiac comorbidities: Advanced chronic kidney disease, severe pulmonary disease, and concurrent active malignancy limit the cardiac benefit that CRT can produce even when resynchronization is achieved. A patient who is functionally limited by advanced emphysema will not demonstrate the expected 6-minute walk test improvement after CRT, not because the device failed to resynchronize but because the primary limitation is now pulmonary, not cardiac.

11.3 What to Do When CRT Fails

When a patient who meets CRT criteria does not respond at 6 months, the clinical algorithm includes:

1. Lead position verification: Review chest X-ray and device fluoroscopy to confirm LV lead position is lateral or posterolateral, not anterior or apical.
2. Pacing threshold check: Confirm the LV output is achieving capture.
3. Programming improvement: Systematic AV and VV interval improvement with echocardiographic guidance.
4. Scar assessment: Cardiac MRI or PET imaging to assess LV lateral wall scar burden.
5. LV lead repositioning: If the lead is in a below-target position and can be safely repositioned, a second procedure to place the LV lead at the site of latest mechanical activation.
6. Alternative pacing approaches: LBBAP as an alternative to lateral CS pacing in anatomically challenging cases.
7. Clinical reassessment: Has the diagnosis changed? Has the patient developed another cardiac condition (new MI, severe valve disease, pericardial constriction) that explains the non-response? Has a non-cardiac cause of symptoms emerged?

---

Illinois-Specific Practice

12.1 CRT Access in Central Illinois

CRT implantation requires electrophysiology subspecialty expertise, a dedicated cardiac EP lab, and substantial echocardiographic support. In Illinois, CRT programs are concentrated in Chicago (Northwestern Medicine, U of Chicago, Rush, UIC), Peoria (OSF HealthCare), and Urbana (Carle Foundation Hospital). For patients in rural Illinois communities more than 60 miles from these centers, the implantation procedure requires travel to a regional center.

Remote monitoring following CRT implantation, as with pacemakers, is the standard for routine surveillance. The Medtronic Carelink, Abbott Merlin.net, Boston Scientific LATITUDE, and Biotronik Home Monitoring networks all provide daily remote data review with automatic alert transmission for clinically significant findings. Rural patients who cannot easily travel for in-person device interrogation are well served by these remote monitoring systems once the implantation procedure has been completed at a regional center.

12.2 The ECHO-CRT Lesson for Clinical Cardiologists in Community Practice

The ECHO-CRT trial enrolled patients with narrow QRS (less than 130 ms) who had echocardiographic evidence of mechanical dyssynchrony by speckle-tracking, hypothesizing that echo-guided CRT would overcome the ECG QRS limitation. The result: harm. CRT in narrow QRS patients increased mortality and hospitalization. This is a cautionary result that has important implications for community cardiology practice.

A cardiologist who sees echocardiographic dyssynchrony in a heart failure patient with a QRS of 115 ms may be tempted to recommend CRT based on the echo finding. ECHO-CRT says not to. The ECG criteria (specifically LBBB with QRS greater than or equal to 150 ms) are not arbitrary bureaucratic thresholds; they predict which patients benefit and which are harmed. Deviation from these criteria in either direction (treating patients below the threshold, or withholding CRT from patients who meet the threshold) produces worse outcomes than strict adherence.

CRT Follow-up

Patients in a structured post-care program with reduced EF heart failure who are on maximally tolerated GDMT are evaluated for CRT candidacy at each annual review. The CRT candidacy checklist:

1. EF at or below 35% confirmed on improved GDMT (minimum 3 months of stable GDMT before EF reassessment)
2. QRS duration at or above 150 ms (preferably at or above 130 ms in women)
3. LBBB morphology confirmed by trained cardiologist review (not algorithm interpretation alone)
4. NYHA Class II-IV symptoms despite improved GDMT
5. Non-ischemic cardiomyopathy: no scar burden assessment required before ILR (but considered if planning CRTD in ischemic patients)
6. Life expectancy at least 1 year with good functional status expected with therapy

Patients who meet all six criteria and who have not received CRT are flagged for urgent cardiology consultation and expedited EP referral. The evidence that CRT reduces mortality in this population is among the strongest in all of interventional cardiology (COMPANION, CARE-HF, MADIT-CRT, RAFT: four large, well-powered RCTs all converging on the same conclusion). Failure to offer CRT to an eligible patient represents a preventable death.

---

The Signal Check is fifteen questions mapping the cardiovascular risk pattern across the physiological domains most commonly missed in standard screenings. It produces a specific starting point for your next clinical conversation.