A Pilot Study of Interventricular Septal Puncture for Cardiac Resynchronization Therapy to Treat Heart Failure

CRT as a treatment for heart failure in selected patients has been shown to decrease both hospitalizations and mortality as well as improving symptoms and quality of life, and is hence supported by guidelines including NICE (United Kingdom National Institute for clinical excellence) . It is achieved by placing a lead to pace the epicardial left ventricle (LV) via the coronary sinus and a coronary vein. Unfortunately, in 5-8% of patients the LV lead cannot be delivered and in 15-20% the lead is placed in a suboptimal position due to the presence of left ventricular scarring, phrenic nerve stimulation or adverse coronary vein anatomy. After lead delivery, 20-45% of patients undergoing CRT fail to respond to their implant. Suboptimal lead position is thought to be one of the most significant causes of non-response.

In the case of failed lead placement there are no well-established alternatives. The most common solution is open-chest surgery using a thoracotomy to place a lead on the epicardial surface of the heart, a procedure with considerable morbidity and a risk of death. Given the co-morbidity profile of patients suitable for CRT, and their severe heart failure, many patients are not fit enough for surgical lead placement. An alternative approach is a percutaneous procedure passing the lead through a puncture in the inter-atrial septum and across the mitral valve, in order to pace the endocardial surface of the left ventricle. This is achieved in a complex and technically-challenging procedure using both femoral and subclavian venous access. This approach exposes a long segment of lead to the systemic circulation, particularly in the low-pressure, low-flow left atrium, raising the thromboembolic risk. It can also worsen mitral regurgitation and exposes the mitral valve to a risk of damage and incompetence should lead extraction be required.

Recent studies have suggested that LV endocardial pacing offers a number of physiological advantages over the conventional epicardial approach. The lead can be positioned at any site on the endocardial surface, away from scar tissue and the phrenic nerve and at the site of maximal haemodynamic improvement, rather than being limited by venous anatomy, thus improving response to CRT. There is also evidence that endocardial pacing may of itself improve haemodynamics to a greater extent than epicardial pacing, in both animal models and in acute haemodynamic studies in humans, and provide a more physiological mechanism of activation, which may be less arrhythmogenic. Consistent with this, response to CRT with endocardial pacing in patients who had not responded to epicardial CRT has been reported, and response rates in previous reports of LV endocardial pacing have been high at approximately 90%.

Targeting of CRT lead placement is an important factor and multiple techniques have been investigated. Targeting of lead placement guided by the site of latest mechanical activation assessed with speckle-tracking echocardiography has been shown to increase response and improve long-term clinical outcomes. Anatomy limits epicardial lead positioning and the best outcomes were obtained in the 60% of patients with a lead able to be placed in the optimal site4. Cardiac output monitoring used to guide lead placement maximizes cardiac output achieved and increases the frequency of long term improvement in cardiac function. Electrical mapping can identify the site of latest electrical activation, although there is not yet evidence as to whether guiding lead position with this improves clinical outcomes. No comparisons between these techniques have been performed.

There is little published so far on how to optimize lead positioning for endocardial LV leads. Groups have so far used invasive cardiac output monitoring to assess the best position for endocardial leads, and have shown that this optimizes cardiac output at the time of implant. Despite the best available evidence for guiding conventional LV lead positioning supporting speckle-tracking echocardiography, this has not yet been used for guiding endocardial lead positioning.

The investigators propose to investigate a novel LV endocardial pacing technique using a puncture through the inter-ventricular septum. This technique has been developed and used in Oxford for a small number of patients, with an initial case series soon to be submitted for publication. Potential advantages include that the inter-ventricular septum is punctured from the same subclavian vein access site through which the other leads are inserted, and that the dilated left ventricular cavity provides a large target, reducing complications. Furthermore, lead placement and extraction would not risk damage to the mitral valve.

All patients who have a left ventricular endocardial pacing lead implanted should be anticoagulated to minimize the risk of stroke from systemic thromboembolism. It has previously been shown that most lead thrombi arise on (right) atrial leads. The investigators' study technique would allow LV endocardial pacing with no lead in the left atrium, likely reducing thromboembolic stroke risk. It is likely that a significant proportion of patients eligible for this study will already be on warfarin due to prior atrial fibrillation or other indications.

Over the last 12 months the investigators have used this novel technique on 6 patients, all of whom had an LV lead safely delivered without complication. Having refined the technique, this Phase 2 study aims to assess the feasibility and safety of the procedure in a larger cohort of patients as well as documenting echocardiographic, biochemical, and functional response.

The population under study would be two groups: firstly, patients with conventional indications for CRT who are not able to have lead implantation via the normal coronary sinus route due to adverse anatomy revealed during a failed prior attempt; and secondly, patients with a successful conventional CRT lead placement who have not responded to the therapy (non-responders). Patients with a failed lead implantation would previously either have been considered for surgical epicardial lead implantation or left with no CRT device. Non-responders are also assessed for surgical lead implantation if they are felt to be likely to respond.

No single technique is known to define optimal lead positioning, and accordingly the investigators intend to use multiple techniques to try and optimize lead positioning, both to try and find the optimal site for each patient, and to allow for comparisons between them. This would include speckle-tracking echocardiography before the procedure, as well as electrical mapping of the LV activation pattern and cardiac output monitoring during the procedure.