Heart Failure Therapy

Optimizing Cardiac Resynchronization Therapy: The Electrophysiologist’s Role in Treating Heart Failure

James Kneller, MD, PhD, FHRS, CCDS

Regional Health System,

Yakima, Washington

James Kneller, MD, PhD, FHRS, CCDS

Regional Health System,

Yakima, Washington

Cardiac resynchronization therapy (CRT) represents the electrophysiologist’s unique contribution to heart failure therapy, and is one of the most rewarding aspects of electrophysiology practice. Eligible patients frequently feel much better, demonstrate recovery of LV function, and live longer with CRT. While some studies have found greater benefit in patients with non-ischemic cardiomyopathy, ischemic patients may also experience similar benefit.1,2 Given their level of suffering, patients with heart failure are generally eager to embrace device therapy, particularly after a disappointing response to medication. CRT represents the last device-based therapy for heart failure prior to progressing to LVAD or heart transplant.3 As such, electrophysiologists remain the best hope for many heart failure patients.

Patients most likely to respond favorably to CRT have true left bundle branch block (LBBB)4 and QRS duration >150 msec.1,2 However, we also see markedly positive responses in patients with QRS duration 120-130 msec, an observation backed by the demonstration of improved peak VO2 in this subset of patients.5 In these cases, a final determination of QRS duration may be best made in the EP lab (200 mm/sec ECG). CRT may also be appropriate for RBBB, with consideration that RBBB may mask LBBB, and is overall most beneficial for QRS duration >150 msec.6 Some CRT patients express feeling significantly improved as early as during pocket closure at the time of implant or on post-op day 1. Others may remain uncertain of symptom benefit, even after demonstrating clear improvement in LVEF with CRT. We encourage those who are seemingly true non-responders in their first year to remain optimistic, as >40% of these patients may have a delayed positive response.7 For all patients, a trial of ECG-guided CRT optimization is strongly recommended to ensure an early and sustained response to CRT, and should be performed for non-responders to ensure the paced QRS morphology predicts a positive response while waiting for evidence of benefit.

Beyond patient selection, ECG-guided optimization of CRT pacing parameters (LV offset, AV delay) is vital to maximize the clinical response to CRT and minimize the non-responder rate. In our program, this is done in the EP lab immediately post implant and again at 3-month follow-up. At implant, we select an initial LV offset demonstrating greatest recruitment of LV myocardium without excessive QRS widening. This is done by first programming a lower rate limit approximately 20 bpm faster than the intrinsic HR and short paced AV delay (to avoid any fusion due to intrinsic AV conduction), and comparing QRS morphologies as different LV offsets are programmed (0, -20, -40, -60 msec). QRS morphology (leads I and V1) and duration are compared for each LV offset. The goal is to find the LV offset achieving the greatest negativity in lead I (QS pattern) and positivity in V1 (R wave), beyond which only QRS widening occurs without further morphology benefit. Achieving an optimal paced morphology is facilitated by pacing from the basal-lateral LV, with the most basal-lateral pacing poles on the LV lead determined fluoroscopically. Final programming with nominal AV delays or those suggested by device algorithms (i.e., SMART-AV trial) are acceptable provided no obvious QRS fusion is occurring due to intrinsic AV conduction.8 In the office, our staff will obtain 12-lead ECGs in rhythm strip format for multiple programmed LV offsets in advance of the physician visit. Serial optimization in the outpatient setting is particularly important when the early response to CRT is underwhelming, or when an early responder plateaus but with incomplete recovery of LVEF. For suboptimal responses despite these efforts, echo-guided AV optimization should be attempted (goal to achieve maximal mitral inflow),8 which is frequently very rewarding at this stage of the optimization process. We do not use the Medtronic AdaptivCRT® algorithm as initial programming, given several equivocal early responses that were much improved with an ECG-guided fixed offset as described here. For those meeting criteria for AdaptivCRT® (LBBB, PR<200 msec), this programming option is used as backup to an ECG-guided approach.

Effective CRT begins with a high-quality implant, with LV lead position enabling maximal recruitment of LV myocardium (Figure 1). A sufficiently medial position for the pulse generator should be established, with anchoring to the pectoral fascia to prevent potentially painful device interactions with the patient’s ipsilateral arm and impingement into the axilla. This is facilitated by first identifying the lateral border of the costal margin in AP fluoroscopy (Figure 1A), with the goal to achieve a final device position overlaying the lung field on post-op chest X-ray (Figure 1B). This is especially important in heart failure patients who may develop worsening of cardiac cachexia with lateral migration of the pulse generator. From the base of the pocket, three independent sites of venous access (Figures 1C and 1D) reduce lead interactions during implant. Venous access obtained using a micropuncture kit reduces the risk of a clinically relevant pneumothorax or bleed from inadvertent arterial puncture. Guidewires advanced below the level of the diaphragm (Figure 1E) confirm venous access, which can sometimes be challenging to distinguish from arterial puncture in patients with distorted anatomy. Deflectable electrophysiology catheters are sometimes necessary to locate and cannulate the CS (Figure 1F). We routinely establish an independent sterile prep of the right femoral vein, so femoral venous access may be obtained during cases, with CS cannulation performed from an inferior approach (up arrows). Sometimes a second electrophysiology catheter is also needed from a superior approach to finally advance an outer guide sheath into the CS (down arrows). LV lead placement may be straightforward and highly successful, even after unusually challenging CS access.

The development of quadripolar LV leads represents a tremendous advance in CRT. Quadripolar leads are presently available in the U.S. from St. Jude Medical and Medtronic, whereas Boston Scientific devices are able to accommodate quadripolar leads from other manufacturers. Presently, only St. Jude Medical offers a CRT pacemaker (CRT-P) system with quadripolar technology. Advantages from quadripolar leads include the ability to implant for maximum lead stability and subsequently select from multiple pacing poles for basal-lateral pacing, best thresholds, and to avoid phrenic nerve capture without necessarily moving the lead. With quadripolar leads, it is particularly important to consider the orientation of the lead within the CS when programming to deliver optimal CRT. In Figure 2, occlusive CS venography shows the main body of the CS wrapping around the mitral valve annulus in LAO (Figure 2A), evidenced by the mitral valve ring in situ. A target vein for the LV lead is seen projecting along the lateral LV wall (* * *), which represents an ideal orientation for utilizing quad technology. Lateral LV pacing would be possible from any poles of a quadripolar lead in this position, and various pacing poles may be selected based on the site of latest local activation, best threshold, or optimal paced QRS morphology. In contrast, potential target veins may project to the lateral LV wall, but not along the lateral LV (* * *, Figure 2B). In this case, pacing from the distal poles of a quadripolar lead reaching the lateral LV would provide the greatest recruitment of LV myocardium with CRT pacing, and should be utilized (even with higher thresholds) for the greatest early response to CRT. It is invaluable to document the orientation of the LV lead and position of available pacing poles to guide subsequent optimization attempts in the outpatient setting, at which time such information is otherwise unavailable.

During the implant procedure, the unpaced ECG (Figure 2C) shows LBBB (V1) with activation toward the lateral LV (positivity in lead I). When testing the LV lead with LV-only pacing (Figure 2F), the ECG reflects activation arising from lateral LV, with negativity in lead I and positivity in V1. Even though the QRS is wider, this is an excellent indicator of effective LV lead placement, and that subsequent ECG-guided CRT optimization will be rewarding. The possibility of dislodgement or migration of the LV lead is minimized by placing adequate but not excessive slack on the LV lead prior to anchoring to the pectoral fascia. Following LV lead implant and removal of the outer guide sheath, slack is added until the lead follows the contour of the lateral RA (LAO fluoroscopic view, Figure 2D). In RAO, the lead should be seen coursing along the annulus of the tricuspid valve from the lateral RA towards the ostium of the coronary sinus (Figure 2E), without prolapsing beyond the valve plane.

Figure 3 shows the case of an 88-year-old female with history of remote bypass surgery, LVEF 20-25%, LBBB with QRS >160 msec, who presented after a fall resulting in facial trauma, with pauses >6 seconds observed on hospital telemetry. The patient did not wish for ICD protection, and was referred for CRT-P implant (Figure 3A). The process of ECG-guided CRT optimization is shown (Figure 3B), comparing baseline and biventricular paced ECGs with programmed LV offsets 0, -20, -40, and -50 msec. A favorable paced QRS morphology (negativity in lead I, positivity in V1) was first evident with -20 msec offset. Favorable features were more pronounced at -40 msec without QRS widening. At -50 msec, only additional QRS widening occurred. Based on this analysis, an LV offset of -40 msec was selected for initial programming. The -20 msec was also favorable, and may be attempted during outpatient follow-up for further clinical improvement.

ECG-guided CRT optimization should be attempted in patients previously labeled as non-responders. The case shown in Figure 4 illustrates the potential benefit of performing CRT optimization in a previously implanted system. The patient is an 80-year-old female with history of ischemic cardiomyopathy and CRT system implanted 8 years prior, who underwent epicardial LV lead placed after 3 years due to failure of the endocardial LV lead (Figure 4A). She was first referred to our clinic when her device reached ERI, now with evidence of a failing RV lead. Her LVEF was <30% despite 8 years of CRT, with no symptom improvement. Chronic programming included LV offset 0 msec, with biventricular paced QRS isoelectric in lead I with ongoing LBBB pattern in V1 (Figure 4C), which was at least narrower than her unpaced QRS. Due to proximal venous obstruction, a new RV lead was placed at the time of generator change using the right internal jugular for venous access with tunneling to the device pocket (Figure 4B). ECG-guided CRT optimization was performed, with LV offset -20 msec achieving an ideal biventricular paced QRS, with good negativity in lead I and fully upright QRS in V1 (Figure 4D). The patient felt immediately improved with this change, noting overnight that she felt more energetic with better color in her cheeks, and markedly improved sleep. This patient may have been considered an early responder if ECG-guided optimization had been performed following addition of the epicardial lead 5 years prior.

LV lead revision should be considered for non-responders when attempted ECG-guided CRT optimization fails to appreciably alter the paced QRS morphology. Figure 5 illustrates the case of a 78-year-old male with ischemic cardiomyopathy, LVEF 17%, first referred to our clinic for management of a CRT system implanted 5 years prior. Chronic programming included LV offset -20 msec. ECG showed an unfavorable biventricular paced QRS morphology (Figure 5C) unchanged with different LV offsets, indicating a negligible contribution of LV pacing to ventricular activation. LV lead revision was performed, with placement of a new lead in a more lateral target vein (Figures 5A and 5B). With ECG-guided CRT optimization, LV offset -20 msec was found to be most favorable for the revised system (Figure 5D). The patient felt immediately improved with effective CRT.


CRT represents an essential component of heart failure therapy, and ECG-guided CRT optimization may dramatically improve patient response to CRT. The LV lead should be positioned along the basal-lateral LV whenever possible. The goal of ECG-guided CRT optimization is to achieve a biventricular paced QRS complex with greatest negativity in lead I and positivity in V1, without excessive QRS prolongation. ECG-guided CRT optimization should be performed at the time of implant, in the outpatient setting when early clinical benefit is not evident, for responders who reach a plateau but with incomplete recovery of LVEF, and for those labeled non-responders to CRT. It is essential to know the orientation of the LV lead and available pacing poles when attempting optimization in the outpatient setting. An ineffective LV lead is identified when varying LV offset fails to influence QRS morphology.  When it is certain this is not due to fusion over the AV node, LV lead revision should be strongly considered.

Please also see Dr. Kneller’s presentation on youtube.com: “Cardiac Resynchronization Therapy (CRT): Indications, Implantation Techniques, and Optimal Programming”

Disclosure: The author has no conflicts of interest to report regarding the content herein. Outside the submitted work, Dr. Kneller reports consulting/honoraria from BIOTRONIK, Boston Scientific, Medtronic, and St. Jude Medical.


  1. Daubert C, Gold MR, Abraham WT, et al. Prevention of disease progression by cardiac resynchronization therapy in patients with asymptomatic or mildly symptomatic left ventricular dysfunction: insights from the European cohort of the REVERSE (Resynchronization Reverses Remodeling in Systolic Left Ventricular Dysfunction) trial. J Am Coll Cardiol. 2009;54:1837-1846.
  2. Moss AJ, Hall WJ, Cannom DS, et al. Cardiac-resynchronization therapy for the prevention of heart-failure events. N Engl J Med. 2009;361:1329-1338.
  3. Nauffal V, Tanawuttiwat T, Zhang Y, et al. Predictors of mortality, LVAD implant, or heart transplant in primary prevention cardiac resynchronization therapy recipients: the HF-CRT score. Heart Rhythm. 2015;12(12):2387-2394.
  4. Eschalier R, Pioux S, Ritter P, Haissaguerre M, Ellenbogen KA, Bordachar P. Nonspecific intraventricular conduction delay: Definitions, prognosis, and implications for cardiac resynchronization therapy. Heart Rhythm. 2015;12(5):1071-1079.
  5. Beshai JF, Grimm RA, Nagueh SF, et al. Cardiac-resynchronization therapy in heart failure with narrow QRS complexes. N Engl J Med. 2007;357(24):2461-2471.
  6. Auricchio A, Lumens J, Prinzen FW. Does cardiac resynchronization therapy benefit patients with right bundle branch block. Circ Arrhythm Electrophysiol. 2014;7:532-542.
  7. Burns KV, Gage RM, Curtin AE, Bank AJ. Long-Term Echocardiographic Response to Cardiac Resynchronization Therapy in Initial Nonresponders. JACC Heart Fail. 2015 Oct 31. doi:10.1016/j.jchf.2015.09.006.
  8. Ellenbogen KA, Gold MR, Meyer TE, et al. Primary results from the SmartDelay determined AV optimization: a comparison to other AV delay methods used in cardiac resynchronization therapy (SMART-AV) trial: a randomized trial comparing empirical, echocardiography-guided, and algorithmic atrioventricular delay programming in cardiac resynchronization therapy. Circulation. 2010;122:2660-2668.