A Pilot Study: Preventing Adverse Remodelling Following Pacemaker Implantation

All patients booked for pacemaker clinic 6w following a new implant for heart block will receive an information sheet with their appointment letter. Potential participants will attend the National Institute of Health Research Clinical Research Facility at Leeds General Infirmary. Each participant will sign a consent form and undergo a baseline echocardiogram, a quality of life assessment (EQ5D-5L), and blood tests including B-type natriuretic peptide measurement. The investigators will record demographic and clinical variables including co-morbidities and current medication.

Patients will be allocated in a 1:2 allocation stratified for atrial rhythm (atrial fibrillation or sinus rhythm) to either standard care (sham programming) or intervention (protocolised pacing-avoidance programming. Patients will be called at 1w to ensure short term acceptability and safety and then reviewed at 3m after the implant. At that visit, a pacing check will establish the amount of RV pacing required. Those pacing >40% despite optimised programming will be asked to undergo a cardiac magnetic resonance (CMR) scan and will be randomly allocated to Lisinopril 5mg (ACEi) or no treatment. They will take one tablet daily and the dose will be increased 2 weekly to the maximally tolerated dose or 20mg daily. This will be supervised by the HF nurse. All patients will be reviewed at a further 6m (9m after their pacemaker procedure) with a repeat echocardiogram, blood test for BNP, QoL assessment, and those in phase two (lisinopril or not) will undergo a repeat CMR scan.

Choice of endpoints Left ventricular remodelling is a widely accepted surrogate endpoint in clinical studies, and therapy-related changes in LV structure and function are related morbidity/mortality outcomes. A 5% increase in mean LVEF was associated with an odds ratio (OR) of 0.86 [95%CI 0.77-0.96] for one-year mortality. LV end diastolic volume (LVEDV) also demonstrates a reliable link to mortality outcomes. A decrease of 10ml is associated with an OR of 0.95 [0.94-0.97] for one-year mortality, whilst a decrease in LV end systolic volume (LVESV) of 10ml is associated with an OR of 0.96 [0.93-0.98] for mortality at 1yr. The investigators will also measure change in indexed volumes such as LVESVi which is frequently used in pacemaker studies. A reduction in LVESVi of ≥15% is associated with better outcome in recipients of a CRT device.

Safety and patient tolerability The protocol of minimising unnecessary RV pacing was well tolerated. ACEi are well-tolerated with fewer than 15% of HF patients unable to tolerate any dose of short-acting enalapril. We expect a lower rate of discontinuation since patients will not have HF and lisinopril is longer-acting. In the Assessment of Treatment with Lisinopril and Survival study (ATLAS), there was adverse effect rate of 4.6% and symptomatic hypotension in 1%. Higher doses were associated with a greater effect, and the discontinuation rate was not related to dose. The investigators conservatively estimate that 10% patients will not tolerate ACEi.

Statistical considerations In estimating sample size, the investigators have followed the rules of thumb for pilot studies proposed by Browne to use at least 30 subjects, and the suggestions of Julious to include at least 12 subjects per treatment arm, and Teare et al for 70 subjects in total. The investigators have assumed that 20% will not have complete heart block at their 6w check, such that if they consent 75 patients, 60 will go through to randomisation, of whom 20 will be allocated standard care (group 1) and 40 will be randomly allocated to optimised pacing (group 2). Those in group 2 still pacing >40% in the RV, will be randomly allocated lisinopril or not in a 1:1 ratio. In addition to exploring the effects of an ACEi on pacing-induced remodelling using CMR in patients with no programming options, they will also be able to compare 6m remodelling in three groups using echocardiography: pacing <40% (n=25), pacing ≥40% (n=35), and pacing ≥40% but taking an ACEi (n=15).

Feasibility In Leeds, 35% of 450 patients receiving an RV pacemaker have heart block (n=160). Around 20% will be excluded due to an ejection fraction <45% (n=32) and a further 20% (n=26) will already be taking an ACE inhibitor. This leaves a pool of 100 people annually who might be eligible. An anonymised study log of people receiving a pacemaker for heart block but excluded for other reasons will be maintained until recruitment is completed.

Blinding The programming will be blinded but ACEi allocation will be open-label. Echocardiography and CMR scans will be performed/administered/reported by a blinded cardiac physiologist or CMR technician and physician. An unblinded cardiac physiologist will be available at each visit for programming questions.

Data analysis:

Statistical analysis plan The primary endpoint will be change in LVEF at 9m post-implant. Secondary outcomes will include other remodelling variables (LVEDV, LVESV), quality of life (EQ5D-5L), and estimated battery longevity at 9m. Patients will be analysed according to their randomised allocation and the difference between groups estimated using a general linear model adjusting for baseline LVEF. Analysis of other endpoints will use similar methods.