DISCUSSION
This study shows, for the first time in a prospective study, the effects of early activation of MPP on reverse remodelling, cardiac function and clinical response in patients with heart failure treated with CRT. The main findings of this study are (i) MPP was associated to a 6-month response rate of 64.6% that did not meet at a 2.5% significance level to be considered as significantly superior to current published CRT-D literature (calculated at 57%), (ii) similarly to previous studies with MPP, it was observed a numerically higher CRT response rate at 6 months in subjects programmed using widest pacing cathodes (72%) compared to other programming (61.1%), (iii) early activation of MPP was associated with an important 6-month LV reverse remodelling and improvement in cardiac function (i.e., LVEF) that was especially significant in patients programmed using anatomical approach, (iv) clinical response to early activation of MPP was high with only 12.2% of patients remaining in class III at 6 months with a percentage of negative responders to CRT according to the clinical composite score of 5%, and, (v) the incidence of mortality and or all cause hospitalizations was low and significantly lower in comparison with the observed in an recent cohort of patients treated with CRT using quadripolar leads and conventional biventricular pacing.
CRT by biventricular pacing is the only heart failure therapy that improves cardiac function, functional capacity and survival while decreasing hospitalizations14. Nevertheless, the response to BiV pacing is variable, ranging from complete normalization of cardiac function to lack of benefit. Suboptimal LV lead position, with less possibilities to pre-excite late activated left ventricular segments, and persistence of mechanical dyssynchrony despite biventricular pacing have been some of the suggested reasons to explain the absence of response to CRT4. The limited ability of conventional pacing to reduce dyssynchrony in some patients could be related to the important variability in the ventricular activation pattern, even in patients with LBBB15,16. In addition, the presence of diseased tissue and or lines of functional conduction block in the LV can induce slow myocardial impulse propagation or left ventricle latency, limiting the ability of a lateral LV lead to reduce the mechanical dyssynchrony. Consequently, intraventricular and interventricular dyssynchrony could persist in up to 30% of patients during conventional biventricular pacing4. MPP has been postulated to improve response by depolarizing large segments of the LV simultaneously and therefore activating early the area of latest electrical activation in left ventricle. Consequently, MPP could reduce LV activation time improving contractility and clinical outcomes.
Initial acute haemodynamic and echocardiographic studies of MPP showed a significant increase in LV dp/dtmax, LV stroke volume and a higher ability to reduce LV dyssynchrony in comparison to conventional biventricular pacing5-8. Nevertheless, the results observed in clinical trials have been somehow disappointing. In the MPP IDE study, patients received biventricular pacing for 3 months and were then randomized to MPP or biventricular pacing for 6 months17. MPP met the primary efficacy endpoint (non-inferiority of response rate based on Clinical Composite Score) in this study. However, MPP was not superior to conventional biventricular pacing reducing the percentage of patients with no response to CRT. In the More CRT MPP study18, patients received conventional biventricular pacing and those not responding at 6 months were randomized to activate MPP or continue with biventricular pacing for another 6 months. In this study, MPP was not superior to biventricular pacing in the rate of conversion to echocardiographic response to CRT (31.8% vs 33.8%, P = 0.72). Similarly, we did not observe a significant benefit of MPP on the echocardiographic response rate to CRT in comparison to the estimated value previously published with conventional biventricular pacing.
Despite these neutral effects of MPP over the response rate to CRT, the MPP IDE study showed that patients randomized to MPP and programmed to pace from anatomically distant poles had a higher response rate to CRT in comparison to pace from close MPP poles. The rate of clinical responders at 9 months was 87% in the group of patients with the MPP programmed with the anatomic separation in comparison to 65% observed in patients with MPP and other programming or 70% in patients with conventional biventricular pacing17. The MORE CRT MPP study also observed a different non-responder to responder conversion rate according to the selection of poles for MPP, 46% in patient with the wide anatomical separation versus 26% in MPP and other programming and 34% in conventional biventricular pacing18. In our study, we also observed important differences in the effects of MPP according to the selection of poles for MPP. The patients programmed with an anatomical approach for MPP had a higher CRT response rate at 6 months (76%) compared to other programming (63%). Moreover, reverse remodelling and improvement in cardiac function was especially important in patients programmed using anatomical approach with a mean absolute increase in LVEF of 14% in comparison to 8% in other MPP programming group. Finally, the percentage of super-responders was 48% in patients with MPP programmed using anatomical approach vs. 28% in other MPP programming. The observed percentage of super responders in this group is strikingly high in comparison to previous published data with conventional biventricular pacing. Interestingly, the QUARTO II study conducted by the same group using conventional biventricular pacing with quadripolar leads reported a super-responder rate of 38%12.
These results are not surprising when considering the initial results of MPP evaluating acute hemodynamic and immediate dyssynchrony reduction in comparison to conventional biventricular pacing. These studies also reported a superiority of an empiric method of selecting MPP vectors based on maximizing anatomical spacing between pacing cathodes. These studies also reported that pacing with the minimum delay between pacing poles produced the best response in hemodynamics and dyssynchrony reduction5-8. The MPP IDE and the MORE CRT MPP trials also reported the highest benefits of MPP when pacing with the minimum delay between MPP electrodes.
All of these observations may be in accordance with the suggested benefits of MPP that entails the depolarization simultaneously of large segments of the left ventricle, reducing its activation time and resulting in a more efficient resynchronization. In this sense, wide separation of the 2 pacing sites seems to be crucial to obtain a benefit from MPP. When pacing sites are close, the area of initial myocardium activation is smaller than the area activated from 2 widely separated sites of pacing. Additionally, pacing from anatomically distant poles increases the probability of stimulate early the of latest activated area within the left ventricle and to avoid stimulation of overlying myocardial scar from at least one of 2 cathodes during MPP. On the other side, MPP by close pacing poles may be similar to conventional bipolar CRT. Indeed, it has been shown with conventional bipolar LV leads unwitting anodal capture in at least half of all cases19. This cathodal-anodal pacing of 2 adjacent LV lead poles may not be different from MPP from adjacent poles and would explain the absence of benefit of MPP over conventional CRT when pacing from narrow-spaced electrodes.
These results suggest the selection of widest-space electrodes with simultaneous pacing when considering MPP activation to increase CRT response. Otherwise, the probability of obtain a significant benefit could be null with a negative impact over device battery longevity.
Clinical outcomes of MPP have been less studied. In a registry, Forleo et al.20 showed that MPP was associated with a better clinical outcome based on Clinical Composite Score, increased LVEF and reduction in QRS duration compared to biventricular pacing. In MORE CRT MPP study18, patients randomized to the MPP arm showed a 21.8% reduction in heart failure events per 100 patient-years compared to before randomization and the biventricular arm showed a 9.1% reduction compared to before randomization (P=NS). In our study, we observed an important clinical response to MPP with only 12% of patients remaining in class III at 6 months of follow-up and a non-responder rate of 5% according to composite clinical score. Moreover, at 6 months mortality was only 1.9% and 11.4% of patients were admitted to the hospital for any reasons. Aiming to compare the clinical outcomes of MPP to conventional biventricular CRT the QUARTO III and QUARTO II cohorts were compared. Despite that groups were not completely equivalent; QUARTO II and QUARTO III cohorts were consecutives and were included, in most cases, by the same hospitals and investigators what entails a homogeneous plan of treatment and monitoring for the patients with heart failure involved in both studies. We observed better clinical outcomes in the group of patients treated with MPP with a significant reduction in the incidence of of mortality and or all cause hospitalizations in comparison to conventional biventricular pacing after adjusting for possible confounders variables.
Presently, MPP is available in devices from most of the companies. Despite the absence of an undoubtedly evidence of benefit, clinicians may have the opportunity of activating MPP in patients treated with CRT and a capable device. The published evidence may indicate a potential benefit from MPP only when it is possible to program wide separated pacing poles. The results of this study are in line with this observation and suggest a potential benefit in terms of clinical outcomes.