Peri-infarct pacing to prevent left reverse remodelling: an unvalidated concept?

The management of acute ST-segment elevation myocardial infarction (STEMI) has evolved considerably in recent decades. Modern treatment includes STEMI network activation, antithrombotic drugs, and rapid early revascularization (mainly by mechanical reperfusion). Cardioprotective interventions aimed at reducing the extent of myocardial necrosis have been evaluated in basic and clinical research with negative or minimal effects. After myocardial injury, left ventricular (LV) remodelling of the infarct zone and the circumferential residual viable myocardium leads to thinning and dilatation of the affected myocardium and hypertrophy of the viable myocardium. This process of LV remodelling impairs LV function, enlarges the left ventricle, and has a major effect on the patient’s outcome in terms of quality of life, heart failure, and survival. Therapies that may attenuate or reverse the process of LV remodelling might have a favourable effect on the prognosis of patients with STEMI. In this issue of the journal, Stone et al. report the results of a clinical trial that assessed a new concept for preventing or attenuating LV remodelling after a myocardial infarction, namely peri-infarct pacing. Chronic cardiac pacing is not only indicated for patients with bradycardia but, as in most patients implanted with a cardiac resynchronization device, also for haemodynamic purposes. Previous studies in normal hearts using magnetic resonance imaging have shown that electrical pre-excitation reduces local loading characteristics while increasing them in remote and late activated regions. The hypothesis of the potential efficacy of peri-infarct pacing to prevent LV remodelling after myocardial infarction was first evaluated in animal studies. In ischaemic and surrounding tissue, an increase in wall stress is observed, which could be reduced by pacing. Shuros et al. showed, in a myocardial infarction model in swine, that ventricular pre-excitation with 8 weeks of pacing in the border regions significantly reduced regional strain and decreased heart weight and LV and left atrial dimensions compared with no pacing. Although not significant, cardiomyocyte apoptosis was lower in the pacing group. In a rabbit model of myocardial infarction that compared different pacing modalities (no pacing, right ventricular pacing, and biventricular pacing) with a sham-operated group without myocardial infarction, only biventricular pacing (with the LV lead positioned away from the apical infarction) prevented systolic and diastolic LV dilatation and the reduction in fractional shortening. The feasibility of biventricular pacing in patients with recent myocardial infarction was reported in a small pilot dual-centre randomized study. Eighteen patients with myocardial infarction within the previous 30–45 days with an LV ejection fraction (LVEF) ≤30% and QRS duration ≤120 ms and conventional medical therapy were randomized to receive a biventricular intracardiac cardioverter defibrillator (ICD) (treatment group) or a dual ICD (control group). In the biventricular group, the LV lead was positioned, if possible, in a lateral vein close to the lateral edge of the infarction. The goals of the study were to assess the safety of delivering biventricular pacing in this population and the effect of biventricular pacing on the LV volume at 1 year. Most of the patients (15/17 who were implanted) had a myocardial infarction. Out of eight successful biventricular implantations, the lead was positioned near the myocardial infarction in seven. LV end-diastolic and end-systolic volumes did not change in the biventricular group, but increased in the control group. However, the difference was only significant for LV enddiastolic volume and not for LV end-systolic volume, which is widely accepted as the best criterion to evaluate LV reverse remodelling in cardiac resynchronization therapy (CRT) trials. Importantly, no differences were observed for LVEF or various clinical endpoints [New York Heart Association (NYHA) class, quality of life, and 6-min walk test]. With a sample size of only 17 patients, the study needs to be considered as hypothesis-generating at best, and its