Preferred left ventricular lead position for upgrade from right ventricular pacing to cardiac resynchronization therapy

Cardiac resynchronization therapy (CRT) is well‐established for treating symptomatic heart failure with electrical dyssynchrony. The left ventricular (LV) lead position is recommended at LV posterolateral to lateral sites in patients with left bundle branch block; however, its preferred region remains unclear in patients being upgraded from right ventricular (RV) apical pacing to CRT. This study aimed to identify the preferred LV lead position for upgrading conventional RV apical pacing to CRT.


| INTRODUCTION
Cardiac synchronization therapy (CRT) is a well-established treatment modality for patients with symptomatic heart failure (HF) and electrical dyssynchrony.In previous large clinical trials with CRT, the left ventricular (LV) lead position was instructed to be at the posterolateral to lateral branch of the coronary sinus (CS). 1,2Therefore, the position from the posterolateral to the lateral branch of the CS is presumably ideal for LV lead placement.However, there were differences in the underlying electrical dyssynchrony between individual patients 3 and the optimal LV pacing site should be individually determined.
Several studies showed that a predictor of symptomatic and structural response to CRT is the most electrically delayed site in the LV that is assessed by the electrical delay from the beginning of the native QRS complex to the local LV electrogram (time interval between the beginning of the native QRS complex and the local LV electrogram [Q-LV]) or by the Q-LV ratio, defined as Q-LV over the QRS duration (QRSd). 4,5This finding is consistent with the concept that CRT eliminates electrical dyssynchrony.Right ventricular (RV) apical pacing leads to nonphysiological electrical myocardial conduction and mechanical LV dyssynchrony. 6Some patients may have LV systolic dysfunction, resulting in HF, 7,8 and they are recommended to be upgraded to CRT by adding a new LV lead. 9,10However, the electrically optimal LV lead site remains unclear in patients planned to be upgraded from conventional RV apical pacing to CRT.Thus, we aimed to identify the preferred LV lead position for upgrading conventional RV apical pacing to CRT.

| Electrophysiological study and acute hemodynamic measurements with temporary pacing
Antiarrhythmic drugs were discontinued for at least five half-lives before catheter ablation.Pulmonary vein isolation was performed in all the patients using radiofrequency currents or cryoballoons under general anesthesia.Additional superior vena cava isolation or cavotricuspid isthmus ablation was performed at the discretion of the operator.If sinus rhythm could not be obtained after pulmonary vein isolation, intracardiac defibrillation was performed to restore sinus rhythm.
After catheter ablation for atrial fibrillation, electrophysiological studies and acute hemodynamic measurements were performed with temporary pacing under continuous general anesthesia.Angiography of the CS using a balloon-occlusion catheter (Wedge pressure catheter, Teleflex) to delineate the CS branches was eligible for LV lead deployment.A pigtail-shaped catheter (Heartcass; Terumo) was introduced into the LV via a transseptal approach for hemodynamic measurements.The electrode catheter was placed in the high right atrium and the RV apex.The CS was cannulated with an 8.5-Fr preshaped SL2 sheath (Abbott Medical).A 3.5 Fr over-the-wire-type pacing catheter (InterNova Monorail Catheter; InterNova) with a 0.014 inch guidewire was placed in each CS branch.
The baseline maximum LV pressure rate (LV− dP/dt max ) was measured during pacing in the DDI mode performed from the RV apex.
Atrial pacing was programmed at 10 beats/min above the intrinsic rate.
The atrioventricular (AV) delay was set at 100 ms below the baseline AV delay, to ensure consistent ventricular capture.The electrical delay to each CS branch during RV pacing was measured as the Q-LV interval and the ratio of Q-LV to QRSd was calculated as the QRS ratio.Simultaneous RV pacing and LV pacing were continuously performed, and changes in QRSd and LV dP/dt max were measured.Each measurement was compared with the immediately preceding baseline measurement and LV dP/dt max was calculated electrically from every heartbeat for at least 10 s under steady-state conditions.

| Location of the CS branches
The CS branches were identified based on CS angiography.CS angiography was performed in two orthogonal views (right anterior oblique [RAO], 30°; left anterior oblique [LAO], 45°).In the RAO view, which represented the longitudinal axis of the heart, the areas were classified into three segments: basal, midventricular, and apical.In the LAO view, which represented the short axis of the heart, a line between the anterior interventricular vein and the middle cardiac vein was drawn.

| Statistical analyses
Continuous data were expressed as the mean and standard deviation and were compared using the Mann-Whitney U test or Wilcoxon signed-rank test, as appropriate.Meanwhile, categorical variables were expressed as absolute numbers and percentages and were compared using Fisher's exact test.Assuming a two-sided α of .05, a statistical power of 80%, and an effect size of 0.5, the target sample size was calculated to be 34.To account for a potential dropout rate of 5%, we calculated that the trial would need to include 37 patients.All statistical analyses were performed using the Statistical Package for Social Science (version 24.0; IBM) and R statistical software version 4.0.2(R Foundation for Statistical Computing).
Statistical significance was defined as a two-tailed p < .05.

| Patient background
Among the 103 patients enrolled to undergo catheter ablation for atrial fibrillation, 47 were excluded from this study because of age ≤64 years (n = 32), history of myocardial infarction or surgery for cardiovascular diseases (n = 8), bundle branch block (n = 5), and postpacemaker implantation state (n = 2).The remaining 56 patients underwent CS angiography, among whom 37 patients (66.1%) showed anterolateral and posterolateral branches in the CS and were eligible for LV lead implantation (Figure 2).The patient characteristics are presented in Table 1.Then, we evaluated the Q-LV interval in the short-axis and LAO view.The Q-LV interval and the Q-LV ratio at the latest activation site were longer and higher at the LV anterolateral site than at the LV posterolateral site (144.2± 14.4 ms vs. 114.3± 18.3 ms, p < .01 and 0.91 ± 0.06 vs. 0.72 ± 0.711, p < .01,respectively).LV stimulation was attempted at the longest Q-LV interval.However, the pacing threshold was too high to stimulate the LV in nine (24.3%) and seven (18.9%) patients at the LV anterolateral and posterolateral branches, respectively (Table 2).In these 16 patients, LV myocardial capture was performed after the pacing site was moved to a more apical site.There was no significant difference in the Q-LV time between the myocardial capture site and the latest activation site within the region between the left anterolateral and posterolateral sites (7.3 ± 3.6 and 7.9 ± 6.1 ms, p = .85,respectively).The Q-LV interval and Q-LV ratio at the pacing LV myocardial capture site were longer and higher at the LV anterolateral site than at the LV posterolateral site (142.5 ± 15.2 ms vs. 112.6 ± 8.7 ms, p < .01 and 0.90 ± 0.06 vs. 0.71 ± 0.11, p < .01,respectively).

| Electrophysiological study and acute hemodynamic measurements with temporary pacing
Simultaneous RV apex and LV pacing was performed at each LV anterolateral or LV posterolateral site.The QRSd was shorter at the LV anterolateral site than at the LV posterolateral site (114 ± 11 ms vs. 127 ± 12 ms, p < .01,respectively).The decreasing ratio of QRSd in biventricular pacing compared with that in RV apical pacing was greater at the LV anterolateral site than at the LV posterolateral site (45.7 ± 18.0% vs. 32.0 ± 17.6%, p < .001).The LV dP/dt max was higher at the LV anterolateral site than at the LV posterolateral site (1188.8± 291.3 mmHg/s vs. 1092.9± 261.2 mmHg/s, p < .01,respectively).The increasing ratio of LV dP/dt max from the RV apical to biventricular pacing was also higher at the LV anterolateral site than at the LV posterolateral site (12.7 ± 8.9% vs. 3.7 ± 8.2%, p < .01,respectively), Figure 3.A representative case is shown in Figure 4A,B.

| DISCUSSION
The findings of the present study can be summarized as follows.First, the latest activation site in the LV during RV apical pacing was not the LV posterolateral site but the LV anterolateral site.Second, almost all patients showed that the latest activation site was the basal site in the RAO view.Finally, simultaneous RV apical and LV anterolateral pacing showed shorter QRSd and better hemodynamic response than simultaneous RV apical and LV posterolateral pacing.
Given that the LV lead position was recommended in the posterolateral to lateral site in a large-scale randomized clinical trial that proved CRT efficacy, 1,2 many cardiologists generally place the LV lead at the LV posterolateral site.However, the appropriate LV lead position to restore electrical synchrony depends on the underlying conduction disturbances. 3A recent study showed that in patients with left bundle branch block (LBBB) and left-axis deviation, the latest activation site was not the LV posterolateral site but the anterolateral site. 11Therefore, the CRT response was weaker in patients with LBBB and left-axis deviation than in patients with LBBB without left-axis deviation. 12,13This observation implies that the empirical LV lead positioning at the LV posterolateral site was inadequate and that patients with LBBB and left-axis deviation might have more advanced cardiac damage such that the marked conduction delay in the LV is apparent, 14 leading to a poor CRT response.
The electric LV lead position, assessed by Q-LV or the Q-LV ratio, was a predictor of symptomatic and structural responses to CRT in several studies. 4,5Our study showed that the Q-LV and Q-LV ratios were greater at the LV anterolateral site than at the LV posterolateral site during RV apical pacing.Positioning the LV lead at the latest activation site is crucial for improving CRT efficacy.A case of left anterior hemiblock that showed a CRT response was recently reported. 15According to the Q-LV measurement, the LV lead position, in this case, was anterolateral, not posterolateral.
There are many reports on the efficacy of conduction system pacing, such as His-bundle pacing 16 and left bundle branch pacing. 17wever, the RV apical site is still used as the RV pacing site because it is easy for operators to deploy the RV lead and for long-term durability. 18The defibrillator lead is generally placed at the RV apex because of its good defibrillation threshold.When it stimulates the RV apical site, a left-axis deviation always appears.Our study demonstrated that the latest activation site in patients with RV apical pacing, as well as in patients with LBBB and left-axis deviation, was the LV anterolateral site.Our study confirmed that simultaneous RV apical and LV anterolateral pacing showed a narrower QRSd and better LV dP/dt max than RV apical and LV posterolateral pacing.
Accordingly, the LV anterior site is highly recommended for LV lead placement in patients who require an upgrade to CRT.
Previous studies that compared biventricular pacing to RV apical pacing have discrepancies.Although the superiority of biventricular pacing was confirmed in the BLOCK-HF trial, 19 the results were inconclusive in the BioPace study. 20Further, there was no statistically significant difference between implantable cardioverter defibrillators and CRT defibrillators among patients who required pacing at baseline in the sub-analysis of the RAFT trial. 21In these three trials, the LV lead position was first recommended in the posterolateral to lateral site whenever possible and LV lead positioning was modified based on the patient's CS anatomy.Although the final LV lead position distribution has not been reported, it is assumed that the failure to prioritize the LV anterolateral site for LV lead placement may have caused discordant results.
T A B L E 2 Difference between the latest activation site and the actual available pacing site.F I G U R E 3 The electrical and hemodynamic differences with biventricular pacing from the anterolateral compared with the posterolateral branch of the coronary sinus.This study demonstrates that the left ventricular (LV) anterolateral site is the preferred position for LV lead placement in patients planned to be upgraded to cardiac resynchronization therapy.The Q-LV ratio during right ventricular (RV) apical pacing is larger at the anterolateral LV site than at the posterolateral LV site.The QRS decreasing ratio and LV dP/dt max increasing ratio during biventricular pacing are higher when the LV anterolateral site is used than when the LV posterolateral site is used.

| Clinical implications
As mentioned above, the LV anterolateral site is preferable for LV lead placement in patients planned to be upgraded to CRT.Gold et al. 5 reported an LV lead placement at which a Q-LV over 95 ms was recommended to achieve a good CRT response.In our study, the Q-LV interval at the LV anterolateral site was over 95 ms in all patients; in contrast, the Q-LV interval at the LV posterolateral site was over 95 ms in 30 (81.1%) patients (p = .01).Duckett et al. 22 reported a significant relationship between an LV dP/dt max increase Recently, fusion-optimized AV interval programming using residual intrinsic conduction with ventricular pacing has been reported to show a narrower QRSd and better CRT response. 23wever, this method does not apply to patients with a long PR interval or an advanced AV block.Patients not eligible for fusionoptimized interval programming should be categorized into the ventricular pacing-dependent group.In this group, the LV anterolateral site is also the preferred LV lead position when the RV is stimulated from the RV apex.Moreover, if conduction disturbances worsen in patients using a fusion-optimized AV interval and CRT becomes ineffective, LV lead repositioning from the LV posterolateral to the anterolateral site might be considered one of the options to improve the CRT effect.

| Limitations
This study has some limitations.First, the registered patients did not have indications for pacemaker implantation, and patients with underlying conditions such as ischemic heart disease or peripheral artery disease were excluded.However, pacemakers are primarily indicated due to age, 24,25 and our registered patients were limited to those aged older than 65 years (average age, 75 ± 8 years).It is difficult to predict which patients will require an upgrade to CRT after pacemaker implantation, but elderly patients who require pacemaker implantation due to RV pacing-induced cardiac dyssynchrony may be potential candidates for CRT upgrade.Although there may be differences in the patient backgrounds between those who meet the criteria for CRT indication and our study population, we believe that our study results should be considered when deciding on the placement of the LV lead during the upgrade procedure.
Second, nearly 30% of the patients in this study did not have LV anterolateral or posterolateral branches eligible for LV lead implantation.When there is only one large branch in the LV lateral region, it is inevitable to place the LV lead in the same site.However, this study highlights a new perspective that attention should be paid to the LV anterolateral site.Even if there is only one branch in the LV lateral region, inserting the tip of the LV lead into further branching towards the LV anterolateral direction may be useful.Third, the threshold was too high to stimulate the LV at the latest activation site in 16 patients.
In these patients, the pacing site was then moved to a more apical site to achieve sufficient myocardial capture.However, there was no significant difference in Q-LV between the LV anterolateral and posterolateral sites (p = .85).Fourth, conduction system pacing is an alternative to RV apical pacing.However, conduction system pacing is complex and usually requires the injection of a contrast medium.Some reports have described complications associated with conduction system pacing. 26,27Therefore, RV apical pacing may be a useful option in clinical practice.
Finally, our study was limited by its single-center design and small sample size.Further multicenter randomized studies with a larger number of patients are warranted to confirm these findings.

| CONCLUSION
The LV anterolateral site is the preferred LV lead position rather than the conventional LV posterolateral site in patients who are planned to be upgraded to CRT from conventional RV apical pacing.

2. 1 |
Patients This study included consecutive patients older than 65 years, who were scheduled to undergo catheter ablation for atrial fibrillation between March 2022 and December 2022.All patients underwent a 12-lead electrocardiogram, laboratory tests, and echocardiography.The exclusion criteria were as follows: (1) bundle branch block, (2) postpacemaker implantation state, (3) history of myocardial infarction or surgery for cardiovascular diseases, and (4) severe renal function damage, defined as an estimated glomerular filtration rate of <30 mL/min/1.73m 2 .
All 37 patients showed regular sinus rhythm after catheter ablation for atrial fibrillation.The mean baseline heart rate was 65 ± 12 beats/min.During pacing in the AAI mode (10 beats/min above the intrinsic rate), the average QRSd and LV dP/dt max were F I G U R E 1 Coronary sinus venography to classify possible left ventricular lead positions.(A) Right anterior oblique (RAO) view representative of the long axis of the heart.These areas are divided into three segments: basal, midventricular, and apical.(B) Left anterior oblique (LAO) view representative of the short axis of the heart.After drawing a line between the anterior interventricular vein (AIV) and the middle cardiac vein (MCV), the areas were divided into four segments based on the angle from the center: anterior (0°-30°), anterolateral (30°-90°), posterolateral (90°-150°), and posterior (150°-180°).F I G U R E 2 Patient selection flow chart.CS, coronary sinus.significantly longer (97 ± 6 ms vs. 159 ± 13 ms, p < .01),whereas LV dP/dt max was significantly shorter (1192.2± 312.5 mmHg/s vs. 1071.0± 293.5 mmHg/s, p < .01)during pacing in the DDI mode from the RV apical site compared with those during pacing in the AAI mode.The Q-LV interval as electrical delay to each CS branch during RV pacing was evaluated at the LV anterolateral and posterolateral branches via CS branches.First, we assessed the longest Q-LV interval site from the perspective of the LV longitudinal view: the RAO view.At the LV anterolateral branch, 36 patients (97.3%) and 1 patient (2.7%) showed the longest Q-LV interval site at the base and middle ventricles, respectively.At the LV posterolateral CS branch, 31 (83.8%) and 6 (16.2%) patients showed the longest Q-LV interval site at the base and middle ventricles, respectively.Notably, no patient showed the longest Q-LV interval site at the apex.

4
Example of electrophysiological studies and acute hemodynamic measurements.(A) Coronary sinus angiography was performed in an 80-year-old male.The anterolateral and posterolateral branches are depicted (upper panel).An electrode catheter is inserted into each branch (white circles).During right ventricular (RV) apical stimulation (white stars), the Q-LV was measured at each branch (lower panel).(B) The Q-LV interval was longer at the left ventricular (LV) anterolateral site (144 ms) than at the LV posterolateral site (100 ms) (left panel).Simultaneous biventricular pacing showed that the QRS duration was shorter at the LV anterolateral site (94 ms) than at the LV posterolateral site (144 ms) (middle panel).The LV dP/dt max and left ventricle pressure (LVP) were higher at the LV anterolateral site (1489 mmHg/s and 132/ 8 mmHg) than at the LV posterolateral site (1252 mmHg/s and 124/8 mmHg) (right panel).LAO, left anterior oblique; LV dP/dt max , maximum LV pressure rate; Q-LV, time interval between the beginning of the native QRS complex and the local LV electrogram; RAO, right anterior oblique. of >10% in acute hemodynamic studies and LV reverse remodeling 6 months after CRT.In our study, 21 (56.8%)patients showed an LV dP/dt max increase of >10% during simultaneous RV and LV anterolateral pacing, whereas only 10 (27.0%) patients showed an LV dP/dt max increase of >10% during simultaneous RV and LV posterolateral pacing (p = .02).
T A B L E 1 Baseline patient characteristics.