Precordial Electrocardiography-Guided Left Ventricular Lead Positioning Associated with Improved Cardiac Resynchronization Therapy

DOI: 10.19102/icrm.2014.051004

^1,2ALLON RAFAEL, MD, ³JOSHUA MOZES, MD, ¹ORRIN B. MYERS, PhD and ^1,2MICHAEL B. WEST, MD

¹University of New Mexico, Health Sciences Center, Albuquerque, NM

²New Mexico VA Health Care System, Albuquerque, NM
³ St. Vincent Medical Group, Indianapolis, IN

Download PDF

Follow Us >> Tweet

ABSTRACT. Background: Twelve-lead surface electrocardiography (ECG) is a simple tool that can illustrate characteristics of left ventricular (LV) activation. The use of precordial ECG patterns at the time of implantation to guide LV lead positioning and predict clinical outcomes in cardiac resynchronization therapy (CRT) has not previously been investigated. Objectives: The aim of this study is to analyze clinical outcome data in CRT based on targeted LV lead position guided by concurrent ECG data. Methods: Records of 328 consecutive patients from two institutions who received biventricular (BIV) devices for CRT were reviewed. All patients were connected to continuous twelve-lead surface ECG monitoring during device implantation. LV lead position and the earliest post-implant ECG were analyzed. Study endpoints included combined outcomes of all-cause mortality or hospitalization due to heart failure or arrhythmia. Results: 316 patients were analyzed (median follow-up 517 days). Positioning the LV lead at the junction of the mid and basal left ventricle in the right anterior oblique projection and the anterolateral “2 o’clock” position in the left anterior oblique projection was associated with less mortality and hospitalization (OR 0.59; 95% CI 0.35–0.95). Patients with optimal lead grade position and optimal ECG grades had fewer primary outcomes; absolute risk reduction 14%; OR 0.49 (95% CI 0.29–0.83; p = 0.008). After adjusting for lead position the odds ratio for optimal ECG grade was 0.44 (95% CI 0.26–0.75; p = 0.002). Conclusion: The current study provides evidence that simple ECG guidance can be instrumental in identifying a coronary vein that is associated with a higher likelihood of CRT responsiveness.

KEYWORDS. biventricular pacing, cardiac resynchronization therapy, electrocardiogram, left ventricular lead.

The authors report no conflicts of interest for the published content.
Manuscript received June 14, 2014, final version accepted July 16, 2014.

Address correspondence to: Allon Rafael, MD, Cardiac Arrhythmia Service, Cardiology Division, Massachusetts General Hospital, 55 Fruit St, Boston, MA 02114.
E-mail: arafael@mgh.harvard.edu

Introduction

Cardiac resynchronization therapy (CRT) has become a landmark therapeutic intervention in heart failure management reducing mortality and heart failure hospitalization by greater than 35% in patients with left ventricular (LV) systolic dysfunction and wide QRS.^1–5 Despite careful, guideline-directed patient selection, there remains a significant number of CRT non-responders; with rates ranging from 30% to 45% in some reports.^1,4,6,7 The proper sequence of epicardial to endocardial activation of LV tissue followed by endocardial activation of right ventricular (RV) tissue belies the mechanics of effective CRT. Twelve-lead surface electrocardiography (ECG) is a simple tool that can indicate directionality and timing of LV activation.⁸ Early LV activation guided by surface ECG has not previously been studied as a marker for ideal LV lead position and a predictor of effective CRT.

Methods

Records of 328 consecutive patients from two institutions who received biventricular (BIV) devices for CRT from June 2010 to January 2014 were reviewed. According to the routine practice in the laboratories participating in this study, all patients were connected to continuous 12-lead surface ECG monitoring during BIV device implantation with careful attention to correct precordial chest lead positioning. Influenced in part by prior quality improvement data⁹ and the simplistic concept that increased “tip-to-tip” separation will shorten combined RV and LV activation times promoting prompt and complete recruitment of ventricular myocardium, an anterolateral coronary vein was routinely targeted (Figure 2). LV-only pacing from that site was then performed to establish the presence of a right bundle branch block (RBBB) in V1, indicating LV free-wall activation and ensuring adequate distance from the ventricular septum (close septal proximity otherwise indicated by a left bundle branch block in V1). Bipolar capture thresholds below 3 volts (to avoid premature battery depletion) and absence of phrenic nerve stimulation (tested at 10 volts) were then verified. If these criteria were not met, the LV lead was repositioned as needed, typically moving to a slightly more posterior vein where this algorithm was repeated until the LV lead was ultimately deployed at the most appropriate site. Subsequently, “V-to-V” activation timing was adjusted in the laboratory to “grow an R wave” in V1 during BIV pacing. Guided by the precordial ECG, “pre-excitation” of the LV epicardium was made progressively more premature until R>S in V1 with precordial R-wave transition by V3 implying simultaneous ventricular activation (Figure 1).

For the data collection purposes of this study, the earliest post-implant 12-lead surface ECGs were analyzed to assess precordial R-wave patterns. This included intraprocedural surface recordings when possible; otherwise, earliest post-procedure recordings were used. ECG R-wave morphology received optimal, intermediate and suboptimal grading based on the following criteria: presence of an R wave (R>S) in lead V1 with early transition (R<S) by V3 was graded as optimal, reflective of ideal “V-to-V” timing (Figure 1); R>S in V1 with late or no precordial transition received an intermediate grade; and the absence of R>S in V1 was suboptimal.

Intraprocedural venography and fluoroscopic images, as well as earliest post-implant chest X-rays (CXR) were used to analyze lead position. The approach to grading LV lead position was similar to that used in the MADIT-CRT trial. Positioning the pacing electrode near the junction of the mid and basal LV segments in the 30° right anterior oblique projection (RAO) and in the anterolateral position in the 30° left anterior oblique projection (LAO) (referred to as “2 o’clock”) was considered optimal LV lead placement (Figure 2). Suboptimal placement included apical RAO or anterior or posterior LAO positioning. Positioning at the junction of the mid and basal LV segments, RAO, and posterolateral, LAO, as well as mid to mid-apical, RAO, and anterolateral, LAO, received intermediate “A” grading because the lead was considered optimally positioned in one view. Mid to mid-apical RAO and posterolateral LAO were graded intermediate “B”.

Follow-up data were retrospectively reviewed from the time of implant until the time of first end-point, which included combined outcomes of all-cause mortality or hospitalization due to heart failure or arrhythmia, or until June, 2014 (the time of study approval by the Institutional Committee on Human Research); whichever was earlier. The methods of research herein conform to the guiding principles of the Declaration of Helsinki.

Statistical analysis

All statistical analyses were conducted using SAS v9.3 (SAS Institute Inc., Cary, NC). The primary outcome was assessed using univariate and multivariable logistic regression; odds ratios were used with 95% CIs to summarize the strength of association. Analysis models were developed using a purposeful forward selection approach. Association between categorical predictors and the outcome was also explored using Fisher exact tests. Univariate tests of continuous variables among groups were Wilcoxon two-sample tests or Kruskal–Wallis tests. Results were considered to be statistically significant if p<0.05; p-values were not adjusted for multiple comparisons.

Results

A total of 316 patients out of 328 had complete ECG and LV lead position data and were analyzed. The median follow-up time was 517 days (2 days to 45.4 months). The median age was 69 years (40–91 years) and 94% were male. A total of 150 patients (48%) had optimal lead position by imaging, 85 patients (27%) had intermediate “A” lead position, 55 patients (17%) had intermediate “B” position, and 26 patients (8%) had suboptimal lead position. Optimal ECG criteria were met in 210 patients (67%), intermediate in 70 patients (22%), and suboptimal in 36 patients (11%) (Tables 1 and 2).

Outcomes occurred in 101 patients including 39 deaths and 62 hospitalizations for heart failure or arrhythmia. Death or hospitalization occurred in 26.0% of patients (n = 39) with optimal lead position compared to 40% in the intermediate A group, 34.5% in the intermediate B group, and 34.6% in the suboptimal lead position groups. The primary end-point occurred in 25.2% of patients (n = 53) with optimal ECG grades compared to 48.6% and 38.9% of patients with an intermediate or suboptimal ECG, respectively. Patients with primary outcomes were slightly older (72 versus 68 years of age, p = 0.04) and there was no significant difference between men and women (p = 0.067) (Tables 1 and 2).

Patients with combined optimal ECG and optimal imaged lead grades (n = 115) had fewer primary outcomes, 22.6% compared with 37.3% in all other categories or permutations of ECG and lead position grading (n = 201) with an OR of 0.49 (95% CI 0.29–0.83, p = 0.008). Logistic regression analysis adjusting for age and sex yielded an odds ratio of 0.47 (95% CI 0.29–0.83, p = 0.006) (Table 3). After adjusting for ECG grading the odds ratio for optimal lead position grade versus combined intermediate (A+B) and suboptimal lead position was 0.70 (95% CI 0.40–1.16; p = 0.175). Conversely, after adjusting for lead position the odds ratio for optimal ECG grade versus combined intermediate and suboptimal ECG grades was 0.44 (95% CI 0.26–0.75; p = 0.002) (Table 4).

Discussion

After univariate logistic regression analysis, placing the LV lead at the junction of the mid- and basal left ventricle in the RAO projection and the anterolateral “2 o’clock” position in the LAO projection was associated with less mortality and hospitalization compared with all other positions (OR 0.59; 95% CI 0.35–0.95). Despite a sustained trend, the loss of statistical significance when comparing optimal versus intermediate B or suboptimal lead positions is a result of fewer patients in the intermediate B and suboptimal groups and likely explains why statistical significance is recovered when intermediate and suboptimal positions are pooled against optimal lead position.

A similar pattern is seen in univariate analysis of ECG grading as patients with optimal ECGs had fewer outcomes compared with the intermediate or combined intermediate and suboptimal ECG groups (OR 0.36 and 0.41, respectively). A small number of patients in the suboptimal ECG group in this study similarly explains the loss of statistical significance to support the trend favoring optimal versus suboptimal ECG grades.

Analysis of combined ECG grade with imaged lead grade revealed an absolute risk reduction for death or hospitalization of 14% when patients had optimal grades in both ECG and lead position compared with all other groups. Interestingly, when correcting for ECG grade the trend towards benefit of optimal lead position loses statistical significance; however, the benefit of optimal ECG grade versus intermediate and suboptimal ECG grading maintains statistical significance after adjusting for lead position.

Myriad factors have been studied in an attempt to elucidate and mitigate risk factors for non-responsiveness including a variety of pre-implant patient characteristics as well as LV lead position.^2,3 The authors of the COMPANION trial have suggested that posterior LV lead placement is not beneficial, while MADIT-CRT data have shown the apical LV lead position to be harmful. Early recruitment of the LV myocardium is an important concept in CRT and the notion of right ventricular (RV) and LV lead tip-to-tip separation is inherently logical and may explain poor outcomes in apical LV lead positioning.

Several markers, surrogates, and interventions at the time of BIV CRT device implantation or during the post-implant period have been considered. Many including echocardiography-guided measurements of dyssynchrony and timing optimization are cumbersome and are often not reproducible. The current study represents a unique analysis of adjunctive surface 12-lead ECG utility in association with targeted LV lead position as predictors of clinical outcomes in CRT.

Conceptualization of initial electrical depolarization from an LV epicardial pacing site, in turn leading to LV endocardial activation and subsequent initial RV activation from an endocardial site is paramount to effective resynchronization. In this regard LV pacing can be likened to an epicardial focus for a premature ventricular complex (PVC). In the same fashion in which the precordial ECG is used to localize the PVC source, the presence of an R wave in V1 during pacing after BIV implantation indicates early LV epicardial depolarization and myocardial recruitment. An assumption is made that LV stimulation from the epicardium will require a longer period of time to engage the His–Purkinje network, therefore BIV pacing is routinely programmed to LV before RV in patients treated at these study laboratories. The elegance and simplicity of using the surface 12-lead ECG as an indicator of early activation of the LV free wall during BIV pacing have been recognized and recently described by van Deursen and colleagues,⁸ who have verified that R>S morphology in V1 is a surrogate for early activation of the LV free wall. However, the use of precordial ECG patterns at the time of implantation to guide LV lead positioning and appropriate V-V timing to predict clinical outcomes in CRT has not previously been investigated.

Study limitations

This study represents a retrospective analysis of non-randomized data. The data were generated by a single operator at two centers. The predominantly male population in this study recognizably limits application to the general population. The gender discrepancy in this series is due to the high prevalence of patients from the Veterans Affairs Health Care system. Next, while a moderate number of patients were included in this study, after distribution based on lead position and ECG grades, the intermediate and suboptimal cohorts were undersized to support some of the trends; this likely reflects operator bias towards presumed optimal lead position and ideal ECG morphology.

Conclusion

These data support the usefulness of adjunctive surface ECG guidance at the time of BIV device implantation for optimal LV lead positioning. This study suggests that the ideal LV pacing site is in the junction between the mid and basal LV segments in the RAO projection and the anterolateral “2 o’clock” position in the LAO projection. Furthermore, in conjunction with such targeted LV lead placement, the manifestation of an R wave in V1 with early transition to R<S by V3 is associated with a significant reduction in mortality or hospitalization for heart failure or arrhythmia.

Although the importance of positioning the LV lead in a target coronary vein in attempts to optimize CRT is well understood, anatomical complexity and other technical challenges may pose barriers to ideal LV lead placement, often resulting in compromised CRT. The value of persistence in the face of such obstacles has been emphasized¹⁰ and myriad approaches to safely overcome them including subselective venography, venoplasty, and retrograde snare pull-through techniques have been described.¹¹ The current study provides evidence that simple ECG guidance can be instrumental in identifying and confirming the location of a coronary vein that is associated with a higher likelihood of CRT responsiveness. The non-responder rate of 32% for all-comers in this study is consistent with data reported in the CRT literature. The reduction to 25.2% in patients with an optimal ECG grade is supportive of improved outcomes and better responsiveness with ECG guided targeted lead placement; however, this remains an unacceptably high rate. Continued efforts to increase effectiveness of CRT are still needed.

Acknowledgments

This project was partially supported by the National Center for Research Resources and the National Center for Advancing Translational Sciences of the National Institutes of Health through Grant Number UL1 TR000041. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.

References

1. Cleland JG, Daubert JC, Erdmann E, et al. The effect of cardiac resynchronization on morbidity and mortality in heart failure. N Engl J Med 2005; 352:1539–1549. [CrossRef] [PubMed]
2. Moss AJ, Hall WJ, Cannom DS, et al. Cardiac-resynchronization therapy for the prevention of heart failure events. N Engl J Med 2009; 361:1329–1338. [CrossRef] [PubMed]
3. Bristow MR, Saxon LA, Boehmer J, et al. Cardiac resynchronization therapy with or without an implantable defibrillator in advanced chronic heart failure. N Engl J Med 2004; 350:2140–2150. [CrossRef] [PubMed]
4. Abraham WT, Fisher WG, Smith AL, et al. MIRACLE Study group. Cardiac resynchronization in chronic heart failure. N Engl J Med 2002; 346:1845–1853. [CrossRef] [PubMed]
5. McAlister FA, Ezekowitz J, Hooton N, et al. Cardiac resynchronization therapy for patients with left ventricular systolic dysfunction: A systematic review. JAMA 2007; 297:2502–2514. [CrossRef] [PubMed]
6. Van Brommel RJ, Bax JJ, Abraham WT, et al. Characteristics of heart failure patients associated with good and poor response to cardiac resynchronization therapy: a PROSPECT (Predictors of Response to CRT) sub-analysis. Eur Heart J 2009; 30: 2470–2477. [CrossRef] [PubMed]
7. Chung ES, Leon AR, Tavazzi L, et al. Results of the predictors of response to CRT (PROSPECT) trial. Circulation 2008; 117:2608–2616. [CrossRef] [PubMed]
8. Van Deursen CJ, Blaauw Y, Witjens MI, et al. The value of the 12-lead ECG for evaluation and optimization of cardiac resynchronization therapy in daily clinical practice. J Electrocardiol 2014; 47:202–211. [CrossRef] [PubMed]
9. Mozes J. Localizing LV lead placement in the anterolateral segment (LAO) between the basal and mid segments (RAO) followed by adjustment Of V-V timing using intra-operative precordial ECG is associated with improved clinical effectiveness of CRT. Poster presented at: Heart Rhythm Society. 34^th Annual Scientific Sessions; 2013 May 8–11; Denver, CO.
10. Singh JP, Heist EK, Ruskin JN, et al. “Dialing-in” cardiac resynchronization therapy: Overcoming constraints of the coronary venous anatomy. J Interv Card Electrophysiol 2006; 17:51–58. [CrossRef] [PubMed]
11. Rafael A, West MB. Coronary venous floss: a novel technique for left ventricular lead positioning in cardiac resynchronization therapy. J Innov Card Rhythm Management 2012; 3:988–991. [CrossRef]