Leads for cardiac resynchronization therapy: where and how many?

This editorial refers to ‘A randomized study of haemodynamic effects and left ventricular dyssynchrony in right ventricular apical vs. high posterior septal pacing in cardiac resynchronization therapy’, by H.M. Kristiansen et al., published in this issue on pages 506–516 and ‘A randomized double-blind crossover trial of triventricular versus biventricular pacing in heart failure’, by D.P.S. Rogers et al., published in this issue on pages 495–505.

Cardiac resynchronization therapy (CRT) has now become a well-established therapy for heart failure in patients with evidence of electromechanical dyssynchrony.1 Currently, left ventricular (LV) reverse remodelling is often used as a response indicator and has proven to be an independent predictor for long-term clinical outcome and long-term survival after CRT.2,3 Furthermore CRT indication has been extended to patients with less severe heart failure symptoms,1,4 and a retrospective analysis of the PROSPECT trial suggests that patients with preserved ejection fraction could also benefit from CRT.4,5 Even though implantation rates keep increasing6 and will increase even more if patients in better functional classes and with better systolic functions are included,7 we are still fighting the problem of non-response. Non-response rates remain ~30–40% and, using cardiac remodelling as the primary outcome for response, it will be even closer to 50%.8 Therefore, subanalyses of large studies have focused on the retrospective search for factors that could be used as a predictor for better outcome. Several prognostic predictors of non-response have been identified, e.g. an ischaemic cause of heart failure, a QRS width <150 ms, the absence of left bundle branch block, and male sex.9–11 Lead localization, primarily of the LV lead, is considered to play a substantial role in clinical improvement after CRT. Retrospective analyses suggest that locating the latest activated region of the left ventricle by echocardiographic techniques can aid in optimal lead positioning. As coronary sinus (CS) approaches are limited by their anatomy, and response rates appear to have stabilized, there is an emerging search for new approaches.

In this issue of the journal two prospective randomized studies focus on optimal positioning of the right ventricular (RV) lead and the comparison of triple-site pacing with standard biventricular stimulation.12,13

These approaches are aimed at improving LV reverse remodelling. There is little evidence on the optimal positioning of the RV lead in CRT if an RV lead is necessary at all. Kristiansen et al.12 showed similar effects in both RV outflow tract and apical lead positions. Even though this is a small study, the books on RV lead position can hopefully now be closed and we can focus our emphasis on LV lead location. Rogers et al.13 examined whether a third ventricular lead can result in better electromechanical synchrony. The lead was targeted in an early situated part of the CS or, if technically impossible, it was positioned on the high RV septum. This multisite approach has previously been suggested by Leanrczyk et al.14 and was further investigated in the Triple Resynchronization In paced Heart Failure Patients (TRIP-HF) study group which established a feasible and safe outcome with twoLV leads.15

A larger increase in LV ejection fraction and decrease in left ventricular end-systolic volume (LVESV) was seen in responders using triventricular leads compared with biventricular leads.15 This consistency in ejection fraction and systolic volume is a remarkable outcome as defining evident responders is difficult partly due to discrepancy between ejection fraction, systolic volume, or clinical outcome. However in the TRIP-HF study, no significant differences was found in clinical outcome such as quality of life after adding a third ventricular lead. Rogers et al. corrected this issue by measuring the 6 min walk test and a heart failure questionnaire as parameters for clinical outcome, and also demonstrated further improvement in clinical outcome.13 Improvement in LV reverse remodelling seems to be similar with different RV lead positions, as shown by Kristiansen et al.12 A second RV lead in the high septum in the case of inability to reach a second position in the CS, however, seemed beneficial. This demonstrates that the larger the area that can be resynchronized the better the outcome will be.

Although studies with triple-site pacing like that of Rogers et al. showed some reduction in the non-responder rate, we would like to add a cautionary note. First, CRT and in particular placing the CS lead are associated with a learning curve. This curve is further prolonged when therapy is extended with a fourth lead. Dislocation, intractable phrenic nerve stimulation, lead fracture, infection, and prolonged hospital stay may increase with each additional lead. These issues have not been addressed in the present study and other studies on multisite pacing. There are even ideas extending CRT by placing a second atrial lead for patients who have interatrial conduction delay in order to diminish interatrial delays.16 There are also data emphasizing that less can be more. The GREATER-EARTH trial, for example, demonstrated a similar reduction in LV systolic volumes by using just one LV lead.17 This univentricular pacing was previously proposed by Blanc et al. who demonstrated improvements in ejection fraction, peak VO₂, and New York Heart Association (NYHA) class to the same extent as biventricular pacing.18 Although the GREATER-EARTH investigators showed no fully comparable responder rate, a safe and feasible option was suggested.

Adding leads in several locations without accurate positioning provides a potentially complicated operation and the actual issue will be avoided. Determining the best possible LV lead position and developing the right technique to reach that position would be preferable. Also device costs will increase due to reduced device durability and increased complexity. Dividing leads on two targets on the LV or even in further biatrial pacing will give subsequently lower impedance levels and eventually higher power consumption. Also when lead failure occurs, reoperation will be a complex task, as other leads can be affected by relocating dysfunctional leads. Future studies should consider alternative approaches, e.g. epicardial or transapical leads, which give comparable outcome and bypass the inaccessibility of CS branches. Furthermore, these approaches have better possibilities for optimal lead placement.

Conflict of interest: A.H.M. has received lecture fees from Medtronic, Biotronik, Boston Scientific, and Sorin.