Multipoint pacing for cardiac resynchronisation therapy in patients with heart failure: A systematic review and meta‐analysis

Multipoint pacing (MPP) has been proposed as an effective way to improve cardiac resynchronisation therapy (CRT) response. We performed a systematic review and meta‐analysis evaluating the efficacy of CRT delivered via MPP compared to conventional CRT.


| INTRODUCTION
Cardiac resynchronisation therapy (CRT) is an effective treatment for patients with heart failure (HF) and electrical dyssynchrony characterised by left bundle branch block (LBBB), however, between 30% and 50% fail to respond. CRT nonresponse is multifactorial, however, placement of the left ventricle (LV) pacing lead away from scar may play a significant role in determining response. 1 Multipoint pacing (MPP) is achieved when multiple pacing stimuli are delivered via a single quadripolar lead placed in a branch of the coronary sinus to achieve CRT. 2 Early feasibility and single-centre studies demonstrated positive results with improvements in acute and short-term measures of dyssynchrony and haemodynamic response. [3][4][5] Determining efficacy of MPP is essential as this technology has been demonstrated to reduce battery longevity. 6 Small randomised studies, 7  p < .001). 11 The authors concluded that MPP was more effective than standard CRT, however the meta-analysis included studies which did not only employ MPP, but rather the effects of single site stimulation from a quadripolar lead. Since this prior metanalysis, two further major randomised trials evaluating MPP have been published 9,10 and a meta-analysis incorporating the data from these randomised trials is needed to evaluate the utility of this treatment.

| Literature search
The systematic review and meta-analysis were conducted in accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines for all stages of the design and implementation (see Table S1). 12 We systematically reviewed the relevant literature, comparing MPP to conventional CRT, by searching EMBASE, CENTRAL, and MEDLINE databases from inception to January 2021 without language restriction. The Quality of Reporting of Meta-Analyses statement 13 and the empiric study by McAuley 14 indicate the exclusion of unpublished studies produces a systematic positive bias, and therefore "grey literature" in the form of poster presentations, unpublished data from Cochrane reviews or other meta-analyses, conference abstracts and preprints were included. In addition, references of review articles, meeting abstracts, pre-prints, letters, editorials, and previous meta-analyses were searched. The robustness of the decision to include "grey literature" in the final analysis will be tested by a sensitivity analysis. The following keywords were used for search: "multipoint pacing" OR "multipoint left ventricular pacing" OR "Multi-point pacing" OR "multi-point left ventricular pacing" OR "MPP pacing."

| Selection criteria
We included all eligible studies that met the following inclusion criteria: (i) CRT-Pacemaker/Defibrillator (CRT-P/D), (ii) MPP (defined as two or more pacing sites from a single quadripolar lead) versus control group (i.e., conventional CRT), (iii) human studies only, (iv) minimum of 6 months mean follow-up. We excluded studies that only assessed acute haemodynamic and/or electrical metrics. For each included study, data of the following efficacy endpoints were used

| Data extraction
All data from included studies were independently extracted and assessed for further analysis by two reviewers (CAR and VSM). Any discrepancies were resolved through a third reviewer (MKE   tool and presented graphically. 17 As less than 10 studies were included in the final analysis, no test for funnel plot asymmetry was performed. 18

| RESULTS
A total of 203 unique records were identified through the database and bibliographic searches. Of these, 151 were excluded based on title and abstract content. After screening the full texts of the remaining 52 articles, 7 met inclusion criteria ( Figure S1). A total of 5 randomised controlled trials (3 multicentre) were included, of which 3 had an active run-in period of conventional CRT for a period of at least 3 months, and 2 did not. A total of 2 were observational registry analyses, of which 1 was prospective, and 1 was retrospective. A total of 5 were fully published results, 7-10, 19 1 was an abstract, 20 Table 2. Graphical risk of bias assessment is reported ( Figure S2).

| Primary efficacy endpoint and study design sub-analysis
Overall, MPP was found to be more effective than conventional CRT      Figure 5).

| Sensitivity analysis
In the sensitivity analysis, excluding the abstracts by Al Mussad and Ferreira, resulted in no differences in the odds ratios and significance of results with regards to the prespecified endpoints 20,22 (Table 4).

| DISCUSSION
The Notably, the benefit of MPP is not significant when including randomised studies only. This is an important finding as MPP activation will adversely affect battery life and therefore, there needs to be a robust clinical reason to consider its activation. 6

| Clinical perspective
Since their initial introduction, quadripolar leads have revolutionised CRT with initial studies demonstrating reduced PNS and reintervention. 30 Subsequent studies have shown cost effectiveness 31 and mortality benefit, 32 and as such they have become standard of care.
The ability to program vectors from multiple poles of an LV lead has potential advantages in terms of improving CRT response, however involve trials with a small number of participants with softer outcomes. [37][38][39] It is hoped that further large, randomised studies will help clarify in which patients this treatment may be beneficial and how it should be programmed in those patients.

| LIMITATIONS
The current meta-analysis has important limitations. Several the studies were nonrandomised, and observational cohort studies or registries, and such studies have inherent confounders and biases as highlighted in the risk of bias assessment that we performed. In addition, there were two studies which were not fully peer reviewed. As stated in the methodology, the inclusion of so-called "grey literature" is encouraged to avoid selection and publication bias, these should be approached with caution. To mitigate this and justify their inclusion a sensitivity analysis was performed which justified inclusion of grey literature. Notably, the weighting assigned to studies that are fully peer reviewed and are considered high quality were given greater weighting in the final ROB assessment ( figure S2). There were multiple different measures of MPP outcome and follow-up duration was not standardised across the studies. This may have affected the outcome, as the marginal benefits of MPP over conventional CRT may require longer therapy duration than the follow-up periods allowed in the current studies. In several studies MPP was assessed only in nonresponders to CRT and in fact, this was the case in the largest major published randomised trial by Leclercq et al. 10 It could be argued that MPP in this group of patients who have already failed to respond to conventional CRT may not be of benefit, as response is unlikely due to underlying unfavourable substrate in this patient group which is not amenable to CRT.
Furthermore, MPP programming varied across studies, and was often at the discretion of the implanter making it difficult to systematically assess the electrical benefit of MPP and evaluate an individual effect of each programming type. We mitigated this risk by assuming high levels of heterogeneity, and so a random effects metaanalytical approach was applied to all analyses. In the future, as further studies are performed with more detailed information on programming type, a network meta-analysis may be an appropriate way to evaluate types of MPP programming better. Delivery of MPP was via two different models of CRT devices and not consistently personalised to the patient's cardiac substrate and it is possible that such measures may optimise electrical resynchronisation and improve response. Specifically, ensuring that both pacing sites are outside of scar, may improve response.

| CONCLUSION
The current meta-analysis represents the largest to date on the use of MPP. There appears to be a signal for benefit for MPP mainly derived from nonrandomised trials, whilst more recent larger randomised studies have failed to show a clear benefit. The results of ongoing large scale randomised studies will be required to better assess the potential benefit of MPP. At present it is difficult to justify widespread use of MPP in the CRT population.

ACKNOWLEDGMENT
The study was supported by the Wellcome/EPSRC Centre for Medical Engineering (WT203148/Z/16/Z).

CONFLICT OF INTERESTS
B.S. Sidhu is funded by a project grant from NIHR and receives speaker fees from EBR systems, outside of the submitted work.