Abstract
Background:
For cardiac resynchronization therapy (CRT), image-guided approaches targeting left ventricular (LV) lead placement at the site of latest mechanical activation had inconsistent outcomes. We examined evidence for improved CRT outcomes when LV lead placement concordant with latest mechanical activation occurred.
Methods:
A review of EMBASE and PubMed was performed for randomized controlled trials or prospective observational studies from October 2008 through October 2020 comparing outcomes with concordant versus discordant LV lead placement. Meta-analyses were performed to assess the association between concordance and death, death or heart failure (HF) hospitalization, ≥ 15% reduction in LV end systolic volume (LVESV), and changes in LVESV or ejection fraction (LVEF).
Results:
From 5897 citations, nine publications (eight studies) with 1355 patients were selected; 975 with a concordant LV lead and 380 with a discordant lead. Mean age was 66–68 years, 82% were male, and 64% had ischemic cardiomyopathy. Meta-analyses demonstrated a statistically significant reduction in death/HF hospitalization at 2 years (OR 0.38; 95% CI 0.16, 0.92) and LVESV at 6 months (mean difference [MD] −13.4%; 95% CI −6.7%, −20.0%), and an increase in LVEF (MD 4.03; 95% CI 0.77, 7.30) with the concordant LV lead. There were trends toward decreased death at 2 years (OR 0.49; 95% CI 0.19, 1.23) and ≥ 15% reduction in LVESV at 6 months (OR 3.81; 95% CI 0.24, 61.24) with concordant LV lead placement.
Conclusion:
A concordant LV lead was associated with better CRT outcomes. Further study of feasible methods to achieve LV lead concordance is needed.
Keywords: cardiac resynchronization therapy, heart failure, left ventricular lead placement, left ventricular mechanical dyssynchrony, meta-analysis
1 |. INTRODUCTION
The location of the LV lead for cardiac resynchronization therapy (CRT) is an important predictor of CRT outcomes.1,2 Built upon the premise that coordinated contraction of a dyssynchronous heart with CRT may improve LV function, LV lead placement in the posterolateral or lateral segments of the LV is suggested because of the likelihood of latest mechanical activation occurring in those segments.3 However, subsequent studies determined that the segment of latest mechanical activation was not always the posterolateral or lateral segments, suggesting the need for a more individualized approach.4,5 Early studies using radial strain analysis to identify the site of latest mechanical activation and assess outcomes in patients with LV placement at the identified site versus outside of identified site found better echocardiographic response and prognosis with the concordant LV lead as compared with a discordant LV lead.6 Several randomized controlled trials then compared a strategy of personalized, image-guided targeting of the LV lead to the site of latest mechanical activation versus a nontargeted approach.7–11 Results were mixed, however the personalized strategy was not always successful at achieving a concordant LV lead, and the nontargeted approach unexpectantly achieved a relatively high rate of concordant LV lead placement. Therefore, while an image-guided strategy that attempts to achieve a concordant LV lead placement may not consistently be associated with improved CRT outcomes, achievement of a concordant LV lead may still be optimal.12,13 Given that a randomized controlled trial comparing concordant versus discordant LV lead placement would not be feasible, we conducted a systematic literature review and a meta-analysis to assess evidence for improved CRT outcomes with a concordant LV lead placement as compared with a discordant LV lead placement.
2 |. METHODS
As part of a National Heart, Lung, and Blood Institute funded project to synthesize evidence related to CRT and identify and prioritize clinical and policy evidence gaps, one of the identified evidence gaps was whether the location or extent of left ventricular mechanical dyssynchrony predicted outcomes in CRT patients.14 To address one component of this evidence gap, we conducted a systematic literature review for studies comparing CRT outcomes in patients with placement of the left ventricular (LV) lead within the LV segment with latest mechanical activation as identified by any cardiac imaging method versus patients whose LV lead was not placed within the identified LV segment with latest mechanical activation.
2.1 |. Literature search strategy
A literature search was conducted using PubMed and Embase for RCTs and prospective observational studies of CRT. The Mesh and Emtree terms “cardiac resynchronization therapy” and “cardiac resynchronization therapy device” were used in addition to title and abstract searches for “cardiac resynchronization therapy,” “cardiac resynchronization therapy,” “atrio biventricular pacing,” “atrio-biventricular pacing,” “biventricular pacing,” or “biventricular pacemaker.” Results were limited to English language and human studies with publication dates between October 2008 through October 2020.
2.2 |. Study selection, abstraction, and bias assessment
Titles and abstracts were independently screened by two reviewers. Publications, identified by either reviewer, that met the selection criteria underwent full text review. Selection criteria included RCT or prospective observational study in humans that were published in English, use of CRT in eligible inpatient or outpatient heart failure patients, an assessment of the location and/or extent of LV mechanical dyssynchrony by any imaging method, and CRT outcomes of interest. CRT outcomes of interest included all-cause mortality, heart failure mortality, heart failure hospitalization, ventricular arrhythmias, change in LV dyssynchrony, change in echocardiographic parameters (LV ejection fraction, end-systolic volume or diameter, end-diastolic volume, or diameter), NYHA class improvement, 6MWT, ICD shocks, quality of life, device-related adverse events, and any composite of the outcomes. The full text of each selected publication was independently reviewed by two reviewers to confirm that the study met the inclusion criteria. Finally, one investigator reviewed full text of the selected studies to identify those that included a comparison of CRT outcomes in patients with the LV lead placed within versus outside of the identified LV segment of latest mechanical activation.
One investigator abstracted study characteristics, patient characteristics, and results and a second investigator confirmed study applicability and the accuracy of abstracted data. Abstracted study characteristics included study design, number of study sites, location of study site (s), publication year, study arms, number of subjects, method of identifying site of latest mechanical activation, method of identifying LV lead location, follow-up period, and outcomes. The study arms of interest for our review were concordant and discordant LV lead placement. Concordant LV lead placement was defined as placement of the LV lead within the LV segment with latest mechanical activation. If the location of transmural scar was ascertained in a study, a concordant LV lead was defined as placement of the LV lead within the nonscarred LV segment with latest mechanical activation. An adjacent LV lead placement was placement of the LV lead in a segment next to the segment with latest mechanical activation. Discordant LV lead placement was an LV lead placed in a segment that was neither concordant nor adjacent. Some studies categorized an adjacent lead placement as concordant while others categorized it as discordant for analyses. For study’s presenting results for three categories (concordant, adjacent, or discordant), the concordant and adjacent groups were combined to create the concordant group for our analyses. Patient characteristics included age, sex, ischemic cardiomyopathy, LBBB, QRS duration, sinus rhythm, LV ejection fraction, NYHA class, and location of LV lead placement. The abstracting investigator and the over-reading investigator independently assessed risk-of-bias using the Cochrane risk-of-bias in nonrandomized studies of interventions (ROBINS-I) tool.15 Discrepancies were resolved through discussion.
2.3 |. Statistical analysis
Categorical outcomes were reported as frequencies with percentages. Continuous values were reported as means with standard deviations (SD). For LVEF, the mean (SD) of the absolute change from baseline to follow-up was reported. In two studies, absolute change was estimated from the mean baseline and follow-up values, and the SD was based upon the correlation between baseline and follow-up SD in one study16 and upon a conservative estimate of 0.5 for the other study.6 In a third study, SD for absolute change in LVEF was not provided and thus the SD at baseline and follow-up were assumed to be the same and the correlation was conservatively estimated at 0.5.17 For LVESV, the mean (SD) of the relative change, expressed as a percentage, from baseline was reported. For two studies, the mean relative change in LVESV was estimated using the reported means at baseline and follow-up and the SD was calculated assuming a conservative correlation between baseline and follow-up of 0.5.6,17
Meta-analyses were conducted for all outcomes of interest that were reported in three or more included studies. Meta-analyses were preformed using a DerSimonian-Laird random-effects model and conservatively used the Knapp–Hartung approach to adjust the standard errors of the estimated model coefficients.18,19 Categorical outcomes were pooled and presented as an odds ratio (OR) with 95% confidence intervals (CI). Continuous outcomes were pooled and presented as the mean difference (MD) with 95% CI. Heterogeneity was assessed by Cochran’s Q and I2 index, and the corresponding p-value. I2 values greater than 75% indicate large heterogeneity.
p-value ≤ .05 was considered statistically significant. Analyses were performed using R (version 4.0.2), including the R package “metafor” (version 2.4–0).
3 |. RESULTS
Of 3084 unique citations identified, a total of nine publications from eight unique studies met the selection criteria for this analysis (Figure 1).6–8,11,13,16,17,20,21 Two publications from the Targeted Left Ventricular Lead Placement to Guide Cardiac Resynchronization (TARGET) study were included, but considered as only one study for analyses.7,13 Of the nine publications, four were secondary analyses of data from RCTs, four were prospective observational studies, and one combined patients from a RCT and a prospective observational study. Only three of the included publications were primarily focused on comparing concordant versus discordant LV leads.6,13,17 Of the remaining six publications, three were RCTs comparing image-guided LV lead placement to nonimage guided LV lead placement,7,8,11 one was a RCT comparing two different RV lead placements,16 one compared a prospectively enrolled cohort with image-guided LV lead placement to a historical control group,21 and the last assessed factors associated with CRT outcomes.20 Speckle tracking echocardiography (STE) was the primary imaging method used in all but one study to identify the LV segment of latest mechanical activation.6–8,11,16,20,21 Four papers categorized an adjacent LV lead as concordant and three categorized it as discordant in analyses. Two papers provided outcomes of interest for patients with concordant, adjacent, and discordant LV leads; therefore, for our study, concordant and adjacent LV lead groups were combined.7,21 All studies were single site studies except the TARGET study, which included 2 sites.7,13 Study characteristics are summarized in Table 1. Risk of bias assessments are presented in Table 2.
Figure 1.
Literature flow diagram. The number of articles screened and excluded along with reasons for exclusions are presented. CRT = cardiac resynchronization therapy; LV = left ventricular; LVMD = left ventricular mechanical dyssynchrony; RCT = randomized controlled trial
Table 1.
Summary of study characteristics
| First Author, Year Study Name | Study design # Studysites Location Overall bias | Method for determining site of latest activation Method for confirming LV lead placement | Study arms (N) | All reported outcomes of Interesta | Follow-up |
|---|---|---|---|---|---|
| Ypenburg, 20086 | Prospective Obs Single Center Netherlands Low |
2D Echo/STE - radial strain analysis (6 segment model) Chest x-ray (frontal and lateral view; apical leads excluded from analysis) |
Concordant (153) Adjacent/Discordant (91) |
ACM, heart transplant or HF hosp.(Primary); ACM; ACM or HF hosp.; HF hosp.; Change in LVEDV; Change in LVESV; % Change in LVEF; |
Mean 32 months (SD 16 months) for hospitalizations and mortality 6 months for Echo changes |
| Delgado, 201120 | Prospective Obs Single Center Netherlands Low |
2D Echo/STE - radial strain analysis (6 segment model) Chest x-ray (lateral and frontal views) |
Concordant (271) Adjacent/Discordant (126) |
ACM or HF hosp.; ACM | Median 21 months (range 1 – 90 months) |
| Kristiansen, 201216 | Secondary analysis of RCT Single center Norway Low |
2D Echo/STE - radial strain analysis (6 segment model) Fluoroscopy (single plane) |
Concordant (61) Adjacent/Discordant (24) |
Mean LVEF; % Change in LVEF; Mean LVEDV; % Change in LVEDV; Mean LVESV; % Change in LVESV; |
6 months |
| Saba, 20138 STARTER |
Secondary analysis of RCT Single Center United States Low |
2D Echo/STE - radial strain analysis (8 segment model) Fluoroscopy (biplane) |
Concordant/Adjacent (128) Discordant (37) |
ACM or HF hosp.; ACM, heart transplant, or LVAD; Event free ≥ 5% increase in LVEF; Event free ≥ 15% decrease in LVESV; % Change in LVEF; % Change in LVESV; |
Mean 1.8 years (SD 1.3 years) for hospitalizations and mortality 6–12 months for Echo changes |
| Khan, 20127 TARGET |
Secondary analysis of RCT 2 sites United Kingdom Low |
2D Echo/STE - radial strain analysis (12 non apical segment model) Chest x-ray/Fluoroscopy (biplane) |
Concordant (124) Adjacent (58) Discordant (29) |
ACM or HF hosp.; ACM; |
2 years for hospitalizations and mortality 6 months for Echo Changes |
| Kydd, 201413 TARGET Substudy |
Secondary analysis of RCT + Prospective Obs 2 sites United Kingdom Low |
2D Echo/STE - radial strain analysis (12 nonapical segment model) Chest x-ray/Fluoroscopy (biplane) |
Concordant/Adjacent (202) Discordant (48) |
ACM; % Change in LVESV; ≥ 15% decrease in LVESV; Absolute change in LVEF; Change in NYHA class; Change in MLWHF; |
Median 39 months (range < 1–61 months) for mortality 6 months for Echo Changes |
| Poller, 201417 Charite CRT Registry |
Prospective Obs Single Center Germany Moderate |
2D Echo/TDI - aorticvalve opening to peak systolic velocity (6 segment model) Fluoroscopy (only left anterior oblique view used for concordance determination) |
Concordant/Adjacent (63) Discordant (28) |
Change in NYHA class; Absolute Change in LVEF; Change in LVESV; |
6 months |
| Bertini, 201621 | Prospective Obs Single center Italy Moderate |
2D Echo/STE - longitudinal strain analysis and CMR (16 segment model) Fluoroscopy (biplane) |
Concordant/Adjacent (18) Discordant (3) |
≥ 15% decrease in LVESV | 6 months |
| Borgquist, 202011 | Secondary analysis of RCT Single center Sweden Moderate |
2D Echo/STE - radial strain analysis + CT/CMR(12 nonapical segment model) Chest x-ray (4 projections) |
Concordant/Adjacent (79) Discordant (23) |
ACM or HF hosp.; ACM; HF hosp.; ≥ 15% decrease in LVESV; ≥ 1 increase in NYHA class; |
Mean 47 months (SD 21 months) for hospitalizations and mortality 6 months for Echo Changes |
Outcomes in bold were included in meta-analyses.
Abbreviations: ACM = All-cause mortality; CMR = cardiac magnetic resonance; Echo = echocardiogram; HF = heart failure; Hosp. = hospitalization; LVEDV = left ventricular end diastolic volume; LVEF = left ventricular ejection fraction; LVESV = left ventricular end systolic volume; LVAD; left ventricular assist device; NYHA = New York Heart Association class; N = sample size; Obs = observational; SD = standard deviation; STE = speckle tracking echocardiogram; TDI = tissue doppler imaging.
Table 2.
Risk of bias assessments
| Categories of assessed bias | ||||||||
|---|---|---|---|---|---|---|---|---|
| Confounding | Selection of participants | Classification of intervention | Deviation from intervention | Missing data | Measurement of outcome | Selection of reported results | Overall | |
| Ypenburg et al. | Moderate | Moderate | Moderate | Low | Low | Low | Low | Low |
| Delgado et al. | Moderate | Low | Moderate | Low | Low | Low | Low | Low |
| Kristiansen et al. | Moderate | Low | Low | Low | Low | Low | Low | Low |
| Saba et al. | Moderate | Low | Moderate | Low | Low | Low | Low | Low |
| Khan et al. | Low | Low | Moderate | Low | Low | Low | Moderate | Low |
| Kydd et al. | Moderate | Moderate | Moderate | Low | Low | Low | Low | Low |
| Poller et al. | Serious | Low | Moderate | Low | Low | Moderate | Low | Moderate |
| Bertini et al. | Moderate | Moderate | Critical | Low | Low | Low | Low | Serious |
| Borgquist et al. | Serious | Low | Moderate | Moderate | Moderate | Moderate | Low | Moderate |
Scale for bias ratings: NI (no information), Low, Moderate, Serious, Critical.
A total of 1355 patients were included in the selected studies, and 975 (72%) had a concordant LV lead, while 380 (28%) had a discordant LV lead. Patient characteristics are presented in Table 3. Mean patient age was greater than 65 years in all studies and most patients were male (range of 73% to 90%). The proportion of patients with ischemic cardiomyopathy in each study ranged from 30% in the concordant arm of the Charite CRT registry study to 100% in both arms of the study by Delgado et al.17,20 Mean or median QRS duration was over 150 msec in all studies except the discordant arm of the study by Delgado et.al. which permitted patients with QRS ≥ 120 ms.
Table 3.
Summary of patient characteristics
| Ypenburg et al. | Delgado et al. | Kristiansen et al. | Saba et al. | Kydd et al.a | Poller et al. | Bertini et al. | Borgquist et al. | |
|---|---|---|---|---|---|---|---|---|
| Age | ||||||||
| Concordant | 67 (10) | 67 (10) | 66 (12) | NR | Median (IQR) 71 (63, 77) | 69 (9.5) | 67.3 (9.7) | 68 (8) |
| Discordant | 65 (10) | 67 (10) | 67 (11) | NR | 74 (69, 79) | 67 (9.3) | ||
| Male | ||||||||
| Concordant | 78% | 85% | 85% | 73% | 75% | 90% | 74% | 74% |
| Discordant | 87% | 86% | 92% | 85% | 86% | |||
| Ischemic Cardiomyopathy | ||||||||
| Concordant | 45% | 100% | 52% | 62% | 53% | 30% | 46% | 46% |
| Discordant | 74% | 100% | 47% | 58% | 39% | |||
| Sinus Rhythm | ||||||||
| Concordant | 79% | 81% | 80% | 72% | NR | 73% | NR | 90% |
| Discordant | 96% | 79% | 75% | NR | 79% | NR | ||
| QRS duration (ms) | ||||||||
| Concordant | 164 (31) | 152 (33) | 169 (25) | NR | Median (IQR) 160 (146,176) | 165 (23) | 156 (24) | 170 (19) |
| Discordant | 154 (34) | 149 (34) | 168 (25) | NR | 155 (136,171) | 156 (15) | ||
| LBBB | ||||||||
| Concordant | 75% | NR | 87% | NR | NR | NR | 54% | 74% |
| Discordant | 73% | NR | 83% | NR | NR | NR | ||
| NYHA Class | ||||||||
| Concordant | All II 13% III 72% IV 22% |
All II 58% III 34% IV 8% |
All II 23% III 63% IV 12% |
|||||
| II | 0 | NR | 0% | 0% | NR | |||
| III | 92% | 77% | 89% | |||||
| IV | 13% | 23% | 11% | |||||
| Discordant | ||||||||
| II | 0 | NR | 0% | 0% | NR | |||
| III | 89% | 75% | 83% | |||||
| IV | 11% | 25% | 17% | |||||
| LVEF (%) | ||||||||
| Concordant | 23 (7) | 25 (7) | 24 (4) | NR | Median (IQR) 23 (19, 29) | 25 (6) | 29 (6) | 23 (11) |
| Discordant | 24 (7) | 25 (7) | 24 (4) | NR | 23 (18, 29) | 27 (6) | ||
| LV Lead placement | ||||||||
| All patients | ||||||||
| Anterior | 5% | 9% | 2% | 0% | 7% | 2% | 2% | 22% |
| Anterolateral | 0% | 0% | 29% | 16% | 0% | 5% | 14% | 0% |
| Lateral | 45% | 46% | 44% | 42% | 48% | 47% | 72% | 44% |
| Posterolateral | 0% | 0% | 21% | 34% | 0% | 40% | 12% | 4% |
| Posterior | 49% | 45% | 4% | 8% | 39% | 6% | 0% | 26% |
| Inferior | 0% | 0% | 0% | 0% | 6% | 0% | 0% | 4% |
Values are mean (standard deviation) unless otherwise specified.
IQR = Interquartile range; LBBB = left bundle branch block; LV = left ventricle; LVEF = left ventricular ejection fraction; ms = milliseconds; NYHA = New York Heart Association; NR = not reported.
All relevant patients from the study by Khan et.al. were eligible for inclusion in the substudy by Kydd et al. Therefore, only the patient characteristics from Kydd et al. are provided in this table.
All outcomes of interest reported in each study are listed in Table 1; however, only five outcomes of interest were reported in three or more studies and analyzed in this study. Four studies assessed all-cause mortality 2 years after CRT placement and five studies assessed all-cause mortality or heart failure hospitalizations at 2 years. The meta-analyses showed statistically significant lower odds of death or heart failure hospitalization (pooled OR 0.38, 95% CI 0.16, to 0.92; I2 =65.0%, p=.022), but only a trend toward lower odds of death (pooled OR 0.49, 95% CI 0.19 to 1.23; I2 = 68.5%, p = .023) for those with a concordant versus discordant LV lead (Figures 2 and 3). There was a trend toward a higher likelihood of achieving a ≥ 15% reduction in LVESV at 6 months (pooled OR 3.81, 95% CI 0.24 to 61.24; I2 = 79.4%, p = .008), but a statistically significant relative change in LVESV at 6 months (MD −14.89, 95% CI −22.68 to −7.10; I2 = 39.7%, p = .16) (Figures 4 and 5). In addition, there was a statistically significant increase in absolute LVEF at 6 months (MD 4.03, 95% CI 0.77 to 7.30; I2 = 74.6%, p = .003) with concordant LV lead placement (Figure 6).
Figure 2.
All-cause mortality at 2 years. Point estimates and 95% CI for effect in each study are presented along with the pooled point estimate and 95% CI. CI = confidence interval, OR = odd ratio
Figure 3.
All-cause mortality or heart failure hospitalization at 2 years. Point estimates and 95% CI for effect in each study are presented along with the pooled point estimate and 95% CI.CI = confidence interval,OR = odd tatio
Figure 4.
At least 15% reduction in LVESV at 6 months. Point estimates and 95% CI for effect in each study are presented along with the pooled point estimate and 95% CI. CI = confidence interval, OR = odd ratio
Figure 5.
Mean relative change in LVESV at 6 months. Point estimates and 95% CI for mean difference in LVESV presented as percentages in each study are presented along with the pooled mean difference and 95% CI. CI = confidence interval, MD = means difference
Figure 6.
Absolute mean change in LVEF at 6 months. Point estimates and 95% CI for mean difference in absolute LVEF in each study are presented along with the pooled mean difference and 95% CI CI = confidence interval, MD = means difference
4 |. DISCUSSION
Given the devasting consequences of heart failure, finding a way to optimize CRT so that more patients can realize its full benefit is important. While enhancements in patient selection and optimizing programming are also critical, we focused on the question of optimizing LV lead placement. For ethical and feasibility reasons, a randomized controlled trial of concordant versus discordant LV lead placement is not possible, therefore, evidence is limited to secondary analyses of randomized controlled trials and prospective observational studies. Most of the included studies used radial strain analysis to identify the site of latest mechanical activation, but not all studies considered potential issues from myocardial scarring. Even though results from image-guided LV targeting studies were not universally favorable for the image-guided strategy, we found evidence supporting the concept of improved CRT outcomes with an LV lead placement concordant with the latest mechanical activation. Our pooled analyses demonstrated a statistically significant 64% lower likelihood of death or heart failure hospitalization, 14.9% reduction LVESV, and an increase in LVEF by 4 points with concordant LV lead placement. There was a 51% lower likelihood of death that did not reach statistical significance and an over three-fold greater likelihood of a ≥ 15% reduction in LVESV that also did not reach statistical significance. The challenge remains how to achieve concordant LV lead placement feasibly and efficiently in clinical practice.
In our study, we initially threw a wide net to capture all studies for which concordant and discordant LV lead placements were presented along with CRT outcomes. Relevant data for our research question were anticipated to be found mostly in subanalyses or subset analyses; however, we limited our review to studies for which the data were prospectively collected. Of the nine publications included, two were from the TARGET study.7,13 The main TARGET study randomized patients to either an image-guided approach for LV lead placement or standard of care, but presented results on those with concordant versus discordant LV lead placement from the combined dataset. The TARGET sub study by Kydd et al. combined patients from TARGET with an earlier prospective, observational, feasibility study to focus on the comparison between concordant and discordant LV lead placement.13 Only two other included publications were primarily focused on comparing concordant versus discordant LV leads.6,17 In the remaining six publications, analyses of CRT outcomes with concordant versus discordant LV lead placement were not the primary focus of the studies.7,8,11,16,20,21 Therefore, the included studies were diverse.
To determine concordant LV lead placement, one must first identify the optimal site and then confirm placement of the LV lead at that site. All studies used either chest x- ray, fluoroscopy, or a combination to identify LV lead placement. In addition, all studies considered the LV segment with latest mechanical activation as the optimal site for LV lead placement. However, the methodology to identify the site with latest mechanical activation differed across studies. In six of the eight unique studies, radial strain analysis from STE was the method used.6,8,11,13,16,20 One study used longitudinal strain analysis with STE21 and one study used the longest interval between aortic valve opening and peak systolic velocity as measured by tissue doppler imaging (TDI).17 The study using longitudinal strain analysis was only included in our analysis for one outcome (≥ 15% reduction in LVESV). The relatively small sample size and number of events from this study yielded a very wide confidence interval around the point estimate for the study and contributed to the wide confidence interval around the pooled point estimate. The study using TDI was included in our analyses of LVESV and LVEF and weighted at 11% and 21%, respectively. The results from Poller et al. were equivocal for both LVESV and LVEF, and this was the only study in which the results did not favor the concordant LV lead position. Despite this, the pooled analyses for both outcomes statistically significantly favored the concordant LV lead. Lastly, three studies used echocardiography and/or other imaging methods to identify and exclude areas of scar.11,13,21 This is an important consideration for LV lead placement and additional research is needed to determine the most appropriate method for identifying the site of latest mechanical activation when global scar burden is high.
The definition of a concordant and discordant LV lead was consistent across studies; however, for analyses, an adjacent LV lead placement, that is placement of a LV lead in the segment adjacent to the segment with latest mechanical activation, was categorized as a discordant LV lead in three studies6,16,20 and as a concordant LV lead in four studies.8,11,13,17 In the remaining two studies, results were presented separately for concordant, adjacent, and discordant LV lead placements. Because most studies grouped concordant and adjacent LV lead placement into one group, we combined the concordant and adjacent groups into one group in our analysis for those studies that reported results separately. This resulted in 472 patients within the concordant group of our analyses having either a concordant or adjacent LV lead, 18 patients in the concordant group having an adjacent LV lead, and 241 patients in the discordant group having either a discordant or adjacent LV lead. This should be taken into consideration in interpretation of our study results as it not known whether an adjacent LV lead placement is as good as a truly concordant LV lead placement Future research is needed to understand differences in CRT outcomes between concordant, adjacent, and discordant LV lead positions as an individual’s coronary anatomy or area of scar could make concordant LV lead placement impossible with current technology.
There are several limitations to this study. It is possible that the systematic literature review missed relevant study results given the complexity of the research question and the large number of articles screened. However, reference reviews did not identify any missing studies. Because of the nature of the research question, there were no randomized controlled trials that addressed the study question. Instead, the evidence was limited to secondary analyses of randomized controlled trials and prospective observational studies that have inherent limitations. In addition, two studies were rated as having moderate bias and one with serious bias, which may impact interpretation of our findings. Several of the included studies were small and/or had very small number of events of interest for analysis. Publication bias was not assessed due to the small number of studies in each meta-analysis. LV lead location was determined by chest x-ray and/or fluoroscopy, but may not have produced consistent granularity across studies and may not have been as precise as other more contemporary modalities. Lastly, significant statistical heterogeneity was identified for most of the analyses.
In conclusion, placement of the LV lead for CRT at an echocardiographic identified site of latest mechanical activation, considered a concordant LV lead placement, was associated with better clinical and echocardiographic CRT outcomes in this meta-analysis. Further study of feasible and efficient methods to achieve LV lead concordance is needed to further improve CRT outcomes in clinical practice.
ACKNOWLEDGEMENTS
The authors thank Kathryn Lallinger and Kamaria Kaalund for project coordination support. They also thank Brystana Kaufman, PhD, Marat Fudim, MD, Vanesa Blumer, MD, and Sarah Goldstein, MD, for their help with initial literature screening.
This study was funded by an NHLBI grant (1R01HL131754-01A1) on the Power of Bayesian Methods, RCTs, and Decision Models to Inform CRT Uncertainties.
ABBREVIATIONS:
- 6MWT
six-minute walk test
- CI
confidence interval
- CMR
cardiac magnetic resonance
- CRT
cardiac resynchronization therapy
- CT
computed tomography
- ICD
implantable cardiac defibrillator
- LV
left ventricular
- LVEF
left ventricular ejection fraction
- LVESV
left ventricular end systolic volume
- MD
mean difference
- Msec
millisecond
- NYHA
New York Heart Association
- OR
odds ratio
- RCT
randomized controlled trial
- SD
standard deviation
Footnotes
Conflicts of Interest
Drs. Allen LaPointe, Ali-Ahmed, Dalgaard, Kosinski, and Sanders Schmidler have no disclosures. Dr. Al-Khatib receives research funding from Medtronic, Abbott, and Boston Scientific.
REFERENCES
- 1.Singh JP, Klein HU, Huang DT, et al. Left ventricular lead position and clinical outcome in the multicenter automatic defibrillator implantation trial-cardiac resynchronization therapy (MADIT-CRT) trial. Circulation. 2011;123:1159–1166. [DOI] [PubMed] [Google Scholar]
- 2.Wilton SB, Exner DV, Healey JS, et al. Left Ventricular Lead Position and Outcomes in the Resynchronization-Defibrillation for Ambulatory Heart Failure Trial (RAFT). Canadian Journal of Cardiology. 2014;30:413–419. [DOI] [PubMed] [Google Scholar]
- 3.Butter C, Auricchio A, Stellbrink C, et al. , and Pacing Therapy for Chronic Heart Failure IISG. Effect of resynchronization therapy stimulation site on the systolic function of heart failure patients. Circulation. 2001;104:3026–3029. [DOI] [PubMed] [Google Scholar]
- 4.Ansalone G, Giannantoni P, Ricci R, Trambaiolo P, Fedele F, Santini M. Doppler myocardial imaging to evaluate the effectiveness of pacing sites in patients receiving biventricular pacing. J Am Coll Cardiol. 2002;39:489–499. [DOI] [PubMed] [Google Scholar]
- 5.Becker M, Kramann R, Franke A, et al. Impact of left ventricular lead position in cardiac resynchronization therapy on left ventricular remodelling. A circumferential strain analysis based on 2D echocardiography. Eur Heart J. 2007;28:1211–1220. [DOI] [PubMed] [Google Scholar]
- 6.Ypenburg C, van Bommel RJ, Delgado V, et al. Optimal left ventricular lead position predicts reverse remodeling and survival after cardiac resynchronization therapy. J Am Coll Cardiol. 2008;52:1402–1409. [DOI] [PubMed] [Google Scholar]
- 7.Khan FZ, Virdee MS, Palmer CR, et al. Targeted left ventricular lead placement to guide cardiac resynchronization therapy: the TARGET study: a randomized, controlled trial. J Am Coll Cardiol. 2012;59:1509–1518. [DOI] [PubMed] [Google Scholar]
- 8.Saba S, Marek J, Schwartzman D, et al. Echocardiography-guided left ventricular lead placement for cardiac resynchronization therapy: results of the Speckle Tracking Assisted Resynchronization Therapy for Electrode Region trial. Circ Heart Fail. 2013;6:427–434. [DOI] [PubMed] [Google Scholar]
- 9.Sommer A, Kronborg MB, Norgaard BL, et al. Multimodality imagingguided left ventricular lead placement in cardiac resynchronization therapy: a randomized controlled trial. Eur J Heart Fail. 2016;18:1365–1374. [DOI] [PubMed] [Google Scholar]
- 10.Stephansen C, Sommer A, Kronborg MB, et al. Electrically vs. imagingguided left ventricular lead placement in cardiac resynchronization therapy: a randomized controlled trial. Europace. 2019;21:1369–1377. [DOI] [PubMed] [Google Scholar]
- 11.Borgquist R, Carlsson M, Markstad H, et al. Cardiac Resynchronization Therapy Guided by Echocardiography, MRI, and CT Imaging: a Randomized Controlled Study. JACC Clin Electrophysiol. 2020;6:1300–1309. [DOI] [PubMed] [Google Scholar]
- 12.Sade LE, Saba S, Marek JJ, et al. The association of left ventricular lead position related to regional scar by speckle-tracking echocardiography with clinical outcomes in patients receiving cardiac resynchronization therapy. J Am Soc Echocardiogr. 2014;27:648–656. [DOI] [PubMed] [Google Scholar]
- 13.Kydd AC, Khan FZ, Watson WD, Pugh PJ, Virdee MS, Dutka DP. Prognostic benefit of optimum left ventricular lead position in cardiac resynchronization therapy: follow-up of the TARGET Study Cohort (Targeted Left Ventricular Lead Placement to guide Cardiac Resynchronization Therapy). JACC Heart Fail. 2014;2:205–212. [DOI] [PubMed] [Google Scholar]
- 14.Fudim M, Dalgaard F, Al-Khatib SM, et al. Future research prioritization in cardiac resynchronization therapy. Am Heart J. 2020;223:48–58. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Sterne JA, Hernán MA, Reeves BC, et al. ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions. BMJ. 2016;355:i4919. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Kristiansen HM, Vollan G, Hovstad T, Keilegavlen H, Faerestrand S. A randomized study of haemodynamic effects and left ventricular dyssynchrony in right ventricular apical vs. high posterior septal pacing in cardiac resynchronization therapy. Eur J Heart Fail. 2012;14:506–516. [DOI] [PubMed] [Google Scholar]
- 17.Poller WC, Dreger H, Schwerg M, Bondke H, Melzer C. Not left ventricular lead position, but the extent of immediate asynchrony reduction predicts long-term response to cardiac resynchronization therapy. Clin Res Cardiol. 2014;103:457–466. [DOI] [PubMed] [Google Scholar]
- 18.DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials. 1986;7:177–188. [DOI] [PubMed] [Google Scholar]
- 19.Knapp G, Hartung J. Improved tests for a random effects meta-regression with a single covariate. Stat Med. 2003;22:2693–2710. [DOI] [PubMed] [Google Scholar]
- 20.Delgado V, van Bommel RJ, Bertini M, et al. Relative merits of left ventricular dyssynchrony, left ventricular lead position, and myocardial scar to predict long-term survival of ischemic heart failure patients undergoing cardiac resynchronization therapy. Circulation 2011;123:70–78. [DOI] [PubMed] [Google Scholar]
- 21.Bertini M, Mele D, Malagu M, et al. Cardiac resynchronization therapy guided by multimodality cardiac imaging. Eur J Heart Fail. 2016;18:1375–1382. [DOI] [PubMed] [Google Scholar]