Short-Term Hemodynamic and Electrophysiological Effects of Cardiac Resynchronization by Left Ventricular Septal Pacing

JOURNAL OF THE AMERICAN COLLEGE OF CARDIOLOGY

VOL. 75, NO. 4, 2020

ª 2020 BY THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION PUBLISHED BY ELSEVIER

ORIGINAL INVESTIGATIONS

Short-Term Hemodynamic and Electrophysiological Effects of Cardiac Resynchronization by Left Ventricular Septal Pacing Floor C.W.M. Salden, MD,a Justin G.L.M. Luermans, MD, PHD,b,c Sjoerd W. Westra, MD,b,c Bob Weijs, MD, PHD,b Elien B. Engels, MSC, PHD,a,d Luuk I.B. Heckman, MD,a Léon J.M. Lamerichs, RN,b Michel H.G. Janssen, BSC,b Kristof J.H. Clerx, RN,b Richard Cornelussen, PHD,a,e Subham Ghosh, PHD,f Frits W. Prinzen, PHD,a Kevin Vernooy, MD, PHDb,c

ABSTRACT BACKGROUND Cardiac resynchronization therapy (CRT) is usually performed by biventricular (BiV) pacing. Previously, feasibility of transvenous implantation of a lead at the left ventricular (LV) endocardial side of the interventricular septum, referred to as LV septal (LVs) pacing, was demonstrated. OBJECTIVES The authors sought to compare the acute electrophysiological and hemodynamic effects of LVs with BiV and His bundle (HB) pacing in CRT patients. METHODS Temporary LVs pacing (transaortic approach) alone or in combination with right ventricular (RV) (LVsþRV), BiV, and HB pacing was performed in 27 patients undergoing CRT implantation. Electrophysiological changes were assessed using electrocardiography (QRS duration), vectorcardiography (QRS area), and multielectrode body surface mapping (standard deviation of activation times [SDAT]). Hemodynamic changes were assessed as the first derivative of LV pressure (LVdP/dtmax). RESULTS As compared with baseline, LVs pacing resulted in a larger reduction in QRS area (to 73  22 mVs) and SDAT (to 26  7 ms) than BiV (to 93  26 mVs and 31  7 ms; both p < 0.05) and LVsþRV pacing (to 108  37 mVs; p < 0.05; and 29  8 ms; p ¼ 0.05). The increase in LVdP/dtmax was similar during LVs and BiV pacing (17  10% vs. 17  9%, respectively) and larger than during LVsþRV pacing (11  9%; p < 0.05). There were no significant differences between basal, mid-, or apical LVs levels in LVdP/dtmax and SDAT. In a subgroup of 16 patients, changes in QRS area, SDAT, and LVdP/dtmax were comparable between LVs and HB pacing. CONCLUSIONS LVs pacing provides short-term hemodynamic improvement and electrical resynchronization that is at least as good as during BiV and possibly HB pacing. These results indicate that LVs pacing may serve as a valuable alternative for CRT. (J Am Coll Cardiol 2020;75:347–59) © 2020 by the American College of Cardiology Foundation.

Listen to this manuscript’s audio summary by Editor-in-Chief Dr. Valentin Fuster on JACC.org.

From the aDepartment of Physiology, Cardiovascular Research Institute Maastricht (CARIM), Maastricht University, Maastricht, the Netherlands; bDepartment of Cardiology, Cardiovascular Research Institute Maastricht (CARIM), Maastricht University Medical Center (MUMC+), Maastricht, the Netherlands; cDepartment of Cardiology, Radboud University Medical Centre (Radboudumc), Nijmegen, the Netherlands; e

d

Department of Medicine, University of Western Ontario, London, Ontario, Canada;

Medtronic PLC, Bakken Research Center, Maastricht, the Netherlands; and fMedtronic PLC, Cardiac Rhythm and Heart Failure

(CRHF), Mounds View, Minnesota. A part of this study was financially supported by Medtronic (Minneapolis, Minnesota). Dr. Luermans has a consultancy agreement with Medtronic; and has received an educational grant from Biotronik. Dr. Engels has received a post-doctoral research fellowship from the Cardiac Arrhythmia Network of Canada (CANet). Dr. Cornelussen is an employee of Medtronic. Dr. Ghosh is an employee of and stock holder in Medtronic. Dr. Prinzen has received research grants

ISSN 0735-1097/$36.00

https://doi.org/10.1016/j.jacc.2019.11.040

348

Salden et al.

JACC VOL. 75, NO. 4, 2020

LV Septal Pacing for CRT

FEBRUARY 4, 2020:347–59

D

ABBREVIATIONS AND ACRONYMS AV = atrioventricular

uring the last 2 decades, cardiac

with endocardial CRT [7,8]). Another advantage of

resynchronization therapy (CRT)

LVs pacing may be that lead positioning is less critical

has developed as an important

than

treatment option for symptomatic patients

BiV = conventional biventricular

CRT = cardiac resynchronization therapy

ECG = electrocardiogram EP = electrophysiological HB = His bundle LBB = left bundle branch LBBB = left bundle branch block

LV = left ventricular

during

His

bundle

(HB)

pacing,

another

approach that is gaining interest in CRT (9–13).

with left ventricular (LV) systolic dysfunc-

To further explore the potential of LVs pacing in

tion and dyssynchronous ventricular activa-

CRT, the present study aimed to answer the following

tion, especially due to left bundle branch

questions: how do the electrophysiological and he-

(LBB) block (LBBB). Randomized clinical tri-

modynamic effects of LVs in CRT candidates compare

als showed that CRT creates a more coordi-

with those of BiV and HB pacing; and how important

nated and efficient ventricular contraction,

is the location within the LVs? These questions were

improves LV systolic function and functional

investigated in a prospective dual-center interven-

capacity, reverses structural remodeling, and

tional cohort study by measuring short-term electro-

decreases heart failure hospitalization and

physiological and hemodynamic effects in heart

mortality (1,2).

failure patients with an indication for CRT. SEE PAGE 360

LVdP/dtmax = first derivative

METHODS

of left ventricular pressure

CRT is usually applied by stimulation of

LVs = left ventricular septal/

both the right ventricle (RV) and the LV

STUDY

lateral wall (biventricular [BiV] pacing). Our

patients, fulfilling the criteria for CRT implantation

group has investigated a single-lead alter-

according to the current guidelines (14), were pro-

septum

native approach for ventricular resynchro-

spectively enrolled. Patients were included in the

SDAT = standard deviation of

nization: pacing the LV endocardial side of

Maastricht University Medical Centerþ (n ¼ 23) and

activation times

the interventricular septum (LV septal [LVs]

the Radboud University Medical Center (n ¼ 4) from

VCG = vectorcardiogram

pacing). LVs pacing is based on the idea that

November 2017 to April 2019. These patients were

in the normal heart, the earliest exit of electrical im-

enrolled on the basis of the inclusion criteria of the

pulses out of the Purkinje system occurs at the LVs

presence of sinus rhythm, New York Heart Associa-

and that LVs pacing would provide a near physio-

tion functional class II to III, optimal heart failure

logical activation. This idea has been supported by

medication, LV ejection fraction <35%, and QRS

short- and long-term animal studies (3,4). Impor-

duration $130 ms in the presence of LBBB or QRS

tantly, long-term LVs pacing maintained regional

duration >150 ms in the absence of typical LBBB.

cardiac mechanics, contractility, relaxation, and effi-

Patients were excluded when they had moderate-to-

ciency near native levels, whereas RV apical or RV

severe aortic stenosis, frequent premature ven-

septal (RVs) pacing diminished these variables (4). In

tricular complexes ($2 complexes on a standard

relation to CRT, Rademakers et al. (5) showed that in

electrocardiogram [ECG]), significant peripheral vascular

canine LBBB hearts, short-term electrical and hemo-

disease, or an age below 18 years.

septum

RV = right ventricular RVs = right ventricular septal/

POPULATION. Twenty-seven

consecutive

dynamic benefits of LVs pacing were comparable to

This study was performed according to the princi-

those during BiV pacing. Moreover, in a small group

ples of the Declaration of Helsinki, and the study

of patients with an indication for CRT, short-term

protocol was approved by the ethics committee of the

hemodynamic benefits of LVs pacing and BiV pacing

Maastricht University Medical Centerþ (registration

were also comparable.

number NL58809.068.16/METC 161117). All patients

The successful permanent implantation in dog

gave written informed consent before investigation,

hearts was followed by a small study in 10 patients

and the study was monitored by the Clinical Trial

with sinus node disease in which a modified version

Center Maastricht.

of the Medtronic 3830 lead (Medtronic, Dublin,

STUDY PROTOCOL. All participants underwent CRT

Ireland) was implanted permanently using a trans-

implantation according to routine clinical practice.

venous

interventricular

The atrial lead was positioned in the right atrial

septum (6). This approach was feasible and avoids

appendage, the RV lead in the RV apical septum,

access of the LV lead into the LV cavity (as is the case

and a quadripolar LV lead in a suitable vein on the

route

and

crossing

the

from Medtronic, Abbott, Microport CRM, Biosense Webster, MDS, and Biotronik; and has served as an advisor to Oracle Health. Dr. Vernooy has received research grants from Medtronic, Abbott, and Biotronik; and has a consultancy agreement with Medtronic and Abbott. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. Manuscript received July 24, 2019; revised manuscript received October 31, 2019, accepted November 8, 2019.

JACC VOL. 75, NO. 4, 2020

Salden et al.

FEBRUARY 4, 2020:347–59

LV Septal Pacing for CRT

F I G U R E 1 Fluoroscopic Images of the Implantation Procedure

LVs mid level

LVs mid level

LVs basal level

LVs apical level

Right anterior oblique (RAO) (upper left) and left anterior oblique (LAO) (upper right) view of the heart with right atrial (RA), right ventricular (RV), and left ventricular (LV) leads and the pressure wire. Upper panels show position of the electrophysiological catheter during mid-level LV septal (LVs) pacing, whereas the lower panels show the catheter position during LVs basal and apical level pacing.

posterolateral LV wall. Thereafter, a 0.014-inch

by a small atrial and large ventricular electrogram

PressureWire X (Abbott, St. Paul, Minnesota) was

during electrical mapping from the catheter tip was

introduced via a femoral arterial puncture using a

targeted. HB capture was noted according to previ-

4-F multipurpose catheter, into the LV cavity

ously defined criteria with correction of the bundle

(Figure 1). Subsequently, a 6-F steerable electro-

branch block (stimulus to QRS end # His to QRS end

physiological (EP) catheter was advanced via the

and paced QRS narrowing) (12).

same femoral arterial puncture into the LV for

The pacing protocol was performed at >10 beats/min above the intrinsic sinus rate to exclude spontaneous

temporary pacing. Hemodynamic and electrophysiological measure-

changes in heart rate. To ensure full ventricular

during

capture, ventricular pacing was performed during

different pacing configurations. Bipolar BiV was per-

atrioventricular (AV) sequential pacing (DDD mode)

formed at 2 levels of the LV lead, using the 2 distal

with a paced AV delay 60 ms shorter than the

(1 and 2) and 2 proximal (3 and 4) electrodes of the

intrinsic PR interval or a paced AV delay of 150 ms in

quadripolar LV lead. To perform LVs pacing, the EP

case of prolonged PR interval (>210 ms). The inter-

catheter was temporarily positioned at different LVs

ventricular pacing delay during biventricular pacing

levels (basal, mid-, and apical). The position at the

was set to 0 ms. Baseline measurements were per-

ments

were

simultaneously

performed

fluoroscopic

images

formed during atrial pacing (AAI mode) at the same

(Figure 1) and the 12-lead ECG. The LVs was paced

pacing rate, both before and after each mode of

alone as well as in combination with the RV lead

ventricular pacing.

LVs

was

determined

from

(LVsþRV). Subsequently, the EP catheter was posi-

Once the pacing protocol was completed, the EP

tioned on the RV interventricular septum via the right

catheter and pressure wire were withdrawn, and the

femoral vein to perform mid-RVs pacing and on the

permanent leads were connected to the CRT device.

HB for HB pacing. A stable HB potential accompanied

The implantation procedure was completed, which

349

350

Salden et al.

JACC VOL. 75, NO. 4, 2020

LV Septal Pacing for CRT

FEBRUARY 4, 2020:347–59

F I G U R E 2 Example of the VCG Construction

X

QRSx = 88 µVs

Y

QRSy = –10 µVs

Z

Tz = –123 µVs

Ty = 9 µVs

Tx = –64 µVs QRSarea =

QRSz = 167 µVs

QRS2area,x + QRS2area,y + QRS2area,z = 190 µ Vs QRS area

T-wave area

Example of a 3-dimensional vectorcardiogram (VCG) constructed from a standard 12-lead electrocardiogram.

automatically terminated the study participation of

principle directions (X, Y, and Z) and represents the

the patient.

extent

M EA SU R EM EN TS. E l e c t r o c a r d i o g r a p h y . Standard

ventricular activation (Figure 2) (16,17).

of

unopposed

electrical

forces

during

12-lead ECGs were recorded throughout the procedure

The customized VCG software was also used to

using an EP recording system (Bard, Boston Scientific,

measure the QRS duration manually from the 12-lead

Massachusetts, or Claris, Abbott). After the procedure,

ECGs. QRS duration was defined as the maximum

3-dimensional

were

time between the beginning of the QRS complex and

synthesized from the recorded 12-lead ECGs using

the end of the QRS complex in all 12 leads. During

the Kors matrix (15). Ectopic ventricular beats

ventricular pacing, the beginning of the QRS complex

were excluded from the analysis. VCG parameters,

was defined in 2 ways: as the first visible deviation

including QRS area, were calculated using customized

after the pacing artifact and as time of the pacing

software programmed in MATLAB R2010b software

artifact.

(MathWorks, Natick, Massachusetts). QRS area is the

M u l t i e l e c t r o d e b o d y s u r f a c e m a p p i n g . A single-

combined area under the QRS complex of the 3

use disposable ECG belt (Heartscape Technologies,

vectorcardiograms

(VCGs)

JACC VOL. 75, NO. 4, 2020

Salden et al.

FEBRUARY 4, 2020:347–59

LV Septal Pacing for CRT

derived from the pressure wire and analyzed using

T A B L E 1 Baseline Characteristics (N ¼ 27)

Age at implantation, yrs

67  9

Men

20 (74) 26  4

2

Body mass index, kg/m

NYHA functional class II/III

22 (81)/5 (19)

ICM LVEF, % QRS duration, ms

PhysioMon software, version 2.02 (Abbott). LVdP/ dtmax is the maximum rate of LV pressure rise and a marker of LV contractility. During each hemodynamic measurement, LVdP/

12 (44)

dtmax was measured for each heartbeat for 30 s

26  5

during baseline (AAI pacing), 60 s during the ven-

151  14

tricular pacing protocol, and again 30 s during base-

QRS morphology

line. To account for spontaneous changes in baseline

LBBB

24 (89)

IVCD

3 (11)

LVdP/dtmax, each ventricular pacing response was compared with the corresponding baseline as relative

Medical history

percentage change in LVdP/dtmax. The last 10 beats

Myocardial infarction/PCI

12 (44)

Hypertension

7 (26)

Prior AF

6 (22)

for analysis, so that for each pacing mode, a “CRT-on”

Diabetes mellitus

4 (15)

and “CRT-off” effect could be determined (20). The

CABG

4 (15)

results of these 2 steps were averaged. Ectopic ven-

Ischemic stroke

3 (11)

tricular beats and the 2 subsequent heartbeats were

COPD

3 (11)

before and first 10 beats after every switch were used

excluded from the analysis.

Medication 27 (100)

STATISTICAL ANALYSIS. Continuous data are pre-

ACE inhibitors/ARB

23 (85)

sented as mean  SD, and discrete variables as counts

Diuretics

22 (82)

and percentages. Data showed a normal distribution

Aldosterone antagonist

2 (7)

according to the Kolmogorov–Smirnoff and Shapiro–

Angiotensin receptor-neprilysin inhibitor

2 (7)

Wilk tests. For ventricular sites with multiple

Beta-adrenergic blocking agents

measurements (BiV with distal and proximal LV

Values are mean  SD or n (%). ACE ¼ angiotensin-converting enzyme; AF ¼ atrial fibrillation; ARB ¼ angiotensin II receptor blocker; CABG ¼ coronary artery bypass grafting; COPD ¼ chronic obstructive pulmonary disease; ICM ¼ ischemic cardiomyopathy; IVCD ¼ intraventricular conduction delay; LBBB ¼ left bundle branch block; LVEF ¼ left ventricular ejection fraction; NYHA ¼ New York Heart Association; PCI ¼ percutaneous coronary intervention.

electrodes, LVsþRV pacing at different LVs levels, and LVs pacing alone at different levels), the 2 best results were averaged. Differences between the pacing settings were statistically evaluated using repeated-measures

analysis

of

variance

with

Bonferroni post hoc analysis, using SPSS software, Verathon, Seattle, Washington) (with modifications

version 25.0 (SPSS, Chicago, Illinois). A p value <0.05

in recording software by Medtronic) with 17 anterior

was considered statistically significant.

and 36 posterior unipolar ECG electrodes was

RESULTS

applied to the patient before the implantation procedure. with

A

multichannel

customized

amplifier

software

and

27 patients represent a typical CRT population. The patient cohort was predominantly male (74%), nearly

were

times

one-half had an ischemic etiology, and LV ejection

and color-coded isochronal maps per setting with

fraction was 26  5%. Cardiac magnetic resonance

custom-made software written in MATLAB. The

imaging or echocardiography showed septal ischemia

activation time was defined as the time to the

in 4 patients (15%) and akinesia or hypokinesia of the

steepest negative slope, and the standard deviation

septum in 4 other patients (15%). Most patients (89%)

of activation times (SDAT) from all electrodes was

were scheduled for a de novo device implantation,

calculated.

and 11% for a CRT upgrade from an implantable car-

SDAT

is

a

into

measure

for

PATIENT CHARACTERISTICS. Table 1 shows that the

acquisition. The data of each unipolar electrogram offline

used

monitor data

transformed

were

activation

of

the

spatial

synchronicity of ventricular activation, and $10%

dioverter-defibrillator.

SDAT reduction by CRT from intrinsic rhythm has

Twenty-six patients successfully completed the

previously been shown to be a predictor of LV

ECG analysis and body surface mapping measure-

reverse remodeling (18,19).

ments, whereas hemodynamic measurements were

HEMODYNAMIC

MEASUREMENTS. The

short-term

successfully performed in all 27 patients. HB pacing

hemodynamic effect of pacing the different sites

with correction of the bundle branch block was per-

was evaluated by invasive measurement of the

formed in 16 patients, with, in most cases, nonselec-

first derivative of LV pressure (LVdP/dtmax) as

tive HB capture (88%). In 1 patient, RV apical pacing

351

Salden et al.

JACC VOL. 75, NO. 4, 2020

LV Septal Pacing for CRT

FEBRUARY 4, 2020:347–59

F I G U R E 3 Examples of EP Effects

A

Baseline

BiV

His bundle

LV septal + RV

LV septal

137 ms / 96 PVs

141 ms / 72 PVs

QRS duration / QRS area = 147 ms / 124 PVs

135 ms / 70 PVs

B

85 ms / 79 PVs

Anterior Right

Posterior Left

Right

110

Baseline SDAT = 45 ms

88 BiV SDAT = 34 ms

66 His bundle SDAT = 27 ms

44 LV septal + RV SDAT = 34 ms

Activation time (ms)

352

22

LV septal 0

SDAT = 17 ms

Example of ECGs during baseline, conventional biventricular (BiV), His bundle, LV septal in combination with RV, and LV septal pacing alone with corresponding QRS duration and QRS area (A) and examples of isochronal maps with corresponding standard deviation of activation times (SDAT) (B). EP ¼ electrophysiological; other abbreviations as in Figures 1 and 2.

(VVI mode) was used as baseline because of the

ELECTROPHYSIOLOGICAL EFFECTS. An example of

presence of atrial fibrillation during the procedure

ECG recordings and isochronal maps based on body

(after inclusion). No periprocedural complications

surface maps is shown in Figure 3. Baseline QRS

were observed.

duration was 151  14 ms, with 24 patients (89%)

JACC VOL. 75, NO. 4, 2020

Salden et al.

FEBRUARY 4, 2020:347–59

LV Septal Pacing for CRT

F I G U R E 4 Electrophysiological Effects

Subgroup (n = 16)

QRS Duration (ms)

200

150

*

*

*

* *

100

50

0 200

†

§

QRS Area (µ µVs)

‡ 150 *

*

100

*

*

50

0 50

†

SDAT (ms)

40

*

*

*

*

*

LVs

HB

LVs

30 20 10 0

BL

BiV

LVs+RV

QRS duration as time between QRS beginning and end (closed bars) and as time from pacing stimulus to QRS end (open bars) (upper panel), QRS area (middle panel), and SDAT (lower panel) (n ¼ 26) during baseline (BL), conventional BiV, LVs in combination with RV, and LVs pacing alone and in a subgroup (n ¼ 16) during His bundle (HB) and LVs pacing. Results are presented as mean  SD. *p < 0.05 versus BL; †p < 0.05 BiV versus LVs; ‡p < 0.05 LVsþRV versus LVs; §p < 0.05 HB versus LVs. Abbreviations as in Figures 1 and 3.

showing a LBBB morphology of the QRS complex ac-

comparable for 3 CRT conditions (BiV 136  15 ms,

cording to the European Society of Cardiology defi-

LVsþRV 134  21 ms, and LVs pacing 135  23 ms) (all

nition. QRS duration, measured as the time between

p < 0.05 compared with baseline). HB pacing (n ¼ 16)

earliest

resulted in a QRS duration of 110  17 ms (p < 0.05

deviation

and

end

QRS

complex,

was

353

Salden et al.

JACC VOL. 75, NO. 4, 2020

LV Septal Pacing for CRT

FEBRUARY 4, 2020:347–59

F I G U R E 5 Hemodynamic Effects

Subgroup (n = 16)

†

30

*

*

*

LVs

HB

LVs

* Increase in LVdP/dtmax (%) Compared to Baseline (AAI)

354

*

20

10

0

BiV

LVs+RV

Relative change in first derivative of left ventricular pressure (LVdP/dtmax) compared with baseline (n ¼ 27) during BiV, LVs in combination with RV, and LVs pacing alone and in a subgroup (n ¼ 16) during HB and LVs pacing. Results are presented as mean  SD. *p < 0.05 versus BL; †p < 0.05 LVsþRV versus LVs. Abbreviations as in Figures 1, 3, and 4.

compared with baseline). The time from pacing

significantly reduced the increase in LVdP/dtmax as

stimulus to QRS end was comparable for all 4 CRT

compared with LVs pacing alone to 11  9% from

conditions (Figure 4).

baseline. In the subgroup analysis (n ¼ 16), with HB

Baseline QRS area was 116  44 m Vs. BiV pacing

pacing, the increase in LVdP/dtmax was not signifi-

significantly reduced QRS area compared to baseline

cantly different between LVs and HB pacing (20  8%

to 93  26 m Vs, whereas LVs further reduced QRS area

vs. 18  10%, respectively).

to 73  22 m Vs (p < 0.05 compared with baseline and

LVs PACING SITE. No significant differences were

BiV pacing). LVsþRV pacing did not significantly

found in changes in LVdP/dtmax and SDAT between

reduce the QRS area. The reduction in QRS area by

the basal, mid-, and apical LVs levels. QRS area during

LVs was similar to that by HB pacing (72  17 mVs vs. to 74  36 mVs, respectively) (Figure 4). For the entire cohort, baseline SDAT was 39  6 ms. BiV pacing reduced SDAT significantly to 31  7 ms, whereas LVs pacing further decreased SDAT to 26 

basal and mid-LVs pacing was significantly smaller than during apical LVs pacing (Figure 6). Of note, LVs pacing, taken into account all performed LVs pacing measurements (basal, mid-, and apical septum) in this study (n ¼ 80) that resulted in at least 10%

7 ms (p < 0.05 compared with baseline and BiV pac-

reduction in SDAT from baseline had a significantly

ing). HB pacing resulted in a significant SDAT reduc-

larger hemodynamic benefit than LVs pacing, which

tion compared with baseline to 27  7 ms, which was comparable to the reduction by LVs pacing in this subgroup (27  6 ms) (Figure 4).

reduced SDAT by <10% (LVdP/dtmax 20  9% vs. 9  8%, respectively; p < 0.05). The same was found for LVs pacing with $10% versus <10% reduction in QRS

ACUTE HEMODYNAMIC EFFECTS. Baseline LVdP/dtmax

area (LVdP/dtmax 19  8% vs. 9  10%, respec-

was 823  225 mm Hg/s. BiV and LVs pacing resulted

tively; p < 0.05).

in significant increases in LVdP/dtmax, relative to

Changes in QRS area, SDAT, and LVdP/dtmax by

baseline (17  9% vs. 17  10%, respectively)

mid-LVs pacing were significantly larger than those

(Figure 5). Interestingly, combined LVsþRV pacing

by mid-RVs pacing (71  23 m Vs vs. 121  43 m Vs,

JACC VOL. 75, NO. 4, 2020

Salden et al.

FEBRUARY 4, 2020:347–59

LV Septal Pacing for CRT

29  10 ms vs. 40  7 ms, and 1  9% vs. 16  10%, respectively).

F I G U R E 6 Results per LVs Pacing Site

BiV pacing using LV proximal (3 and 4) or distal (1

†

and 2) pacing electrodes resulted in similar SDAT (31 

200

7 ms vs. 32  8 ms, respectively) and LVdP/dtmax (17 

†

with the proximal electrodes resulted in a significantly larger reduction in QRS area from baseline (from 115  45 mVs to 86  28 m Vs) than pacing with the distal electrodes (from 119  48 m Vs to 103  29 m Vs).

DISCUSSION

QRS Area (μVs)

8% vs. 15  10%, respectively). However, BiV pacing

150

100

hemodynamic improvement that is at least as good as ond, basal, mid-, and apical LVs pacing positions

40

resulted in similar results, indicating that within the

pacing creates better electrical resynchronization than BiV pacing, as evidenced by larger reductions in from previous animal studies, showing that pacing cial endocardial fibers, not necessarily Purkinje fibers,

which takes 30 to 70 ms in LBBB (23,24). Other factors may be that the LVs is the location where electrical impulses break out of the rapid conduction system (25) and that activation occurs from endocardium to epicardium. The latter is expected to generate more simultaneous LV activation. Previous studies indicate that SDAT is a more accurate predictor of CRT response than QRS duration (18,19). Interestingly, SDAT was similar during

20

n = 23

n = 25

n = 21

*

*

*

n = 23

n = 26

n = 22

Basal

Mid

Apical

30 Increase in LVdP/dtmax (%) Compared to Baseline (AAI)

slow conduction across the interventricular septum,

30

0

the LV at any endocardial position captures superfi-

tion (4,21,22). Moreover, pacing the LVs bypasses

*

10

QRS area and SDAT. Potential explanations come

providing a physiological way of LV resynchroniza-

*

n = 21

* SDAT (ms)

important finding of the present study is that LVs

n = 25

50

during BiV and HB pacing (Central Illustration). Sec-

ELECTROPHYSIOLOGICAL EFFECTS OF LVs PACING. An

n = 22 0

results in short-term electrical resynchronization and

LVs, the position of the pacing electrode is not critical.

*

50

The present proof-of-principle study demonstrates that in patients with an indication for CRT, LVs pacing

*

20

10

0

HB and LVs pacing in the present study, indicating comparable synchronicity of ventricular activation

QRS area (upper panel), SDAT (middle panel), and relative increase in LVdP/

by HB and LVs pacing. A potential explanation for

dtmax (lower panel) during pacing at 3 levels of the LVs. Results are pre-

this finding comes from previous animal experi-

sented as mean  SD. *p < 0.05 versus BL; †p < 0.05 between LVs levels.

ments, which showed a sinus-rhythm–like activation

Abbreviations as in Figures 1, 3, 4, and 5.

pattern when pacing into septal tissue below the point at which HB penetrates the central fibrous body (26). Conceptually, LVs pacing would not be effective if

these patients is quite proximal in the His-Purkinje system, as recently described by Upadhyay et al. (27).

most of the conduction delay in CRT patients were

Furthermore, the present study corroborates data

peripheral. The observation that HB pacing was able

from previous studies that there is a considerable

to reduce QRS duration to an average of 110 ms in our

difference in electrophysiological effects between

study indicates that most of the conduction delay in

RVs and LVs pacing (4,28). Between RVs pacing and

355

356

Salden et al.

JACC VOL. 75, NO. 4, 2020

LV Septal Pacing for CRT

FEBRUARY 4, 2020:347–59

C E NT R AL IL L U STR AT IO N Left Ventricular Septal Pacing for Cardiac Resynchronization Therapy

Salden, F.C.W.M. et al. J Am Coll Cardiol. 2020;75(4):347–59.

The present study compared acute electrophysiological and hemodynamic effects of left ventricular (LV) septal, biventricular, and His bundle pacing in patients receiving cardiac resynchronization therapy (upper left). Electrophysiological changes were assessed using multi-electrode body-surface mapping (upper right) as standard deviation of activation times. Hemodynamic changes were assessed as LVdP/dtmax. The results indicate that LV septal pacing provides acute hemodynamic improvement (lower left) and electrical resynchronization (lower right), that is at least as good as during biventricular and possibly His bundle pacing.

LVs endocardium activation is an approximately

conventional CRT or because of inability to access the

40-ms delay, and it takes another 50 ms to conduct

coronary sinus. The present study, however, was

the impulse around the LV endocardium. During LVs

prospective, and none of the patients had undergone

pacing, the transseptal conduction occurs (in the

prior attempts for CRT implantation, thereby better

rightward direction) simultaneous with the circum-

reflecting regular clinical practice.

ferential LV endocardial conduction, thus reducing

An unexpected and interesting finding in this

total LV activation time significantly (4).

study is that combined LVsþRV pacing provided a

HEMODYNAMIC

PACING. The

significantly smaller hemodynamic response than LVs

beneficial electrophysiological consequences of LVs

pacing alone. The explanation for this finding is not

pacing appear to translate to beneficial hemodynamic

clear and requires further investigation.

EFFECTS

OF

LVs

effects, especially because LVs pacing with $10%

The similarity of hemodynamic effects of HB

reduction in SDAT or QRS area was associated with a

and LVs pacing is in agreement with the results in

significantly larger hemodynamic response. The

a previous study in patients with sinus node disease.

similarity of hemodynamic effects of LVs and BiV

In these patients, LVs pacing did not induce a sig-

pacing is in line with previous animal studies as well

nificant reduction in LVdP/dtmax as compared with

as a preliminary patient study (5). In this small study,

intrinsic sinus rhythm, despite that LVs pacing

patients (n ¼ 12) underwent investigation of alterna-

increased QRS duration in these patients with a

tive pacing locations because of nonresponse to

narrow QRS complex (6).

JACC VOL. 75, NO. 4, 2020

Salden et al.

FEBRUARY 4, 2020:347–59

LV Septal Pacing for CRT

The finding that HB and BiV pacing provided

the finding that the reduction in SDAT and the in-

similar increases in LVdP/dtmax compares also with

crease in LVdP/dtmax were virtually independent of

previous studies showing no significant differences in

the LVs pacing location. However, Hou et al. (28)

echocardiographic response between HB and BiV

indicate that LBB capture may provide superior

pacing (9,29). By contrast, the recent study by Arnold

benefit compared with myocardial capture, albeit in a

et al. (11) showed that HB pacing provides a larger

population of patients with bradyarrhythmias and

increase in systolic arterial blood pressure, albeit in a

LVEF

pre-selected patient cohort. CLINICAL IMPLICATIONS. Although the present study

compare the electrophysiological and hemodynamic

>55%.

Additional

research

is

needed

to

effects of LV septal pacing with LBB area pacing.

used temporary LVs pacing, several practical benefits

This wide range in LVs pacing location possibility

of single-lead LVs pacing may be envisioned for future

would make LVs lead implantation relatively easy,

clinical application. With a similar hemodynamic

because lead positioning is less critical than during

benefit of LVs, BiV, and HB pacing, LVs may have

HB and LBBB area pacing (12). Further disadvantages

several advantages once a market-released pacing lead

of HB pacing are various sensing issues, high pacing

is available for this purpose. A custom, investigational

thresholds,

lead has been successfully employed in animal studies

implantation leading to more lead revisions and/or

(4,30) and in 10 patients with sinus node dysfunction,

faster battery drainage (12,13). In the experience with

using a transvenous approach by positioning the lead

long-term LVs pacing in patients with sinus node

deep into the interventricular septum (6). Although

disease, LVs pacing was performed from the tip of the

and

further

rising

thresholds

post-

these studies were performed with a modified version

lead with capture thresholds of 0.5  0.2 V. No lead-

of the Medtronic 3830 lead (extended helix), Huang

related complications were observed, and leads

et al. (31) were able to perform deep septal (LBB area)

remained electrically and mechanically stable during

pacing, using the same approach, by implantation of

the 6-month follow-up (6) (and are still so after 3

the standard Medtronic 3830 lead.

years) (Vernooy, unpublished data, April 2019).

As compared with BiV pacing, LVs pacing has the advantages that it avoids potential problems with lead placement in a coronary venous tributary and phrenic nerve stimulation (32). Moreover, the essentially epicardial stimulation, generated by pacing in a coronary vein, creates a less physiological, epicardialto-endocardial, activation sequence, which has been considered as a cause of some proarrhythmic effects of CRT (33). More physiological activation is also obtained using endocardial CRT. However, endocardial CRT is still in its childhood, associated with a lack of proper tools and a relatively high percentage of adverse events (7,8). Recently, pioneering studies showed the feasibility of pacing in or near the LBB (34,35). LBB area pacing is

STUDY LIMITATIONS. The present proof-of-principle

study investigated the short-term response of temporary pacing at the LVs. For long-term LVs pacing, the aforementioned transvenous transventricular septal approach can be used. This may lead to slightly different activation patterns because the penetrating lead may not entirely reach the endocardium. HB pacing was also performed using an EP catheter, which resulted in nonselective HB capture in most cases. Therefore, QRS durations and SDAT may be larger compared with selective HB capture. Finally, there are controversial data as to whether the shortterm increase in LVdP/dtmax translates to a longterm

benefit

of

CRT

(36,37),

especially

when

comparing to short-term changes in stroke work (38).

supposed to capture the rapid conduction system. This LBB area pacing is performed by a similar

CONCLUSIONS

transventricular septal approach as in long-term LVs pacing, although across the most basal part of the

LVs pacing results in short-term hemodynamic

septum (31,34,35). In our LVs pacing approach, pacing

improvement and electrical resynchronization that is

locations were not chosen on the basis of a “His-

at least as good as BiV and possibly HB pacing. These

Purkinje signal” on the local electrogram. However,

results suggest that LVs pacing with a single ven-

the QRS morphology that we present in Figure 3A has

tricular lead may serve as a novel, valuable, and

some similarities to the LBB area–paced QRS mor-

feasible alternative way to apply CRT.

phologies described by Huang et al. (31). These similarities may be explained by accidental capture of the

ADDRESS FOR CORRESPONDENCE: Dr. Floor C.W.M.

His-Purkinje system during our LVs approach. Alter-

Salden, Maastricht University, Department of Physi-

natively, pacing the rapid conduction along non-

ology,

Purkinje endocardial fibers may achieve the same

Netherlands. E-mail: [email protected].

benefit as pacing the LBB. The latter is supported by

Twitter: @FloorSalden.

P.O.

Box

616,

6200

MD

Maastricht,

the

357

358

Salden et al.

JACC VOL. 75, NO. 4, 2020

LV Septal Pacing for CRT

FEBRUARY 4, 2020:347–59

PERSPECTIVES COMPETENCY IN PATIENT CARE AND

TRANSLATIONAL OUTLOOK: Long-term clinical

PROCEDURAL SKILLS: Pacing the cardiac conduction

studies are needed to assess the utility of a left ventric-

system along Purkinje fibers and non-Purkinje left

ular septal pacing approach to cardiac resynchronization

ventricular endocardial fibers may provide an alternative

therapy.

to biventricular, His bundle, and left bundle branch stimulation for cardiac resynchronization therapy.

REFERENCES 1. Cleland JGF, Daubert J-C, Erdmann E, et al. The effect of cardiac resynchronization on morbidity and mortality in heart failure. N Engl J Med 2005; 352:1539–49. 2. McAlister FA, Ezekowitz JA, Wiebe N, et al. Systematic review: cardiac resynchronization in patients with symptomatic heart failure. Ann Intern Med 2004;141:381. 3. Peschar M, De Swart H, Michels KJ, et al. Left ventricular septal and apex pacing for optimal pump function in canine hearts. J Am Coll Cardiol 2003;41:1218–26.

patients with heart failure and left bundle branch block. J Am Coll Cardiol 2018;72:3112–22.

cardiac resynchronization therapy. PACE 2011;34: 217–25.

12. Vijayaraman P, Dandamudi G, Zanon F, et al. Permanent His bundle pacing: recommendations from a Multicenter His Bundle Pacing Collaborative Working Group for standardization of definitions, implant measurements, and follow-up. Heart Rhythm 2018;15:460–8.

21. van Deursen C, van Geldorp IE, Rademakers LM, et al. Left ventricular endocardial pacing improves resynchronization therapy in canine left bundle-branch hearts. Circ Arrhythmia Electrophysiol 2009;2:580–7.

13. Zanon F, Ellenbogen KA, Dandamudi G, et al. Permanent His-bundle pacing: a systematic literature review and meta-analysis. Europace 2018;

Castellanos A, Morales A, Bassett A. The role of canine superficial ventricular muscle fibers in endocardial impulse distribution. Circ Res 1978;42: 27–35.

20:1819–26.

4. Mills RW, Cornelussen RN, Mulligan LJ, et al. Left ventricular septal and left ventricular apical pacing chronically maintain cardiac contractile coordination, pump function and efficiency. Circ Arrhythm Electrophysiol 2009;2:571–9.

14. Brignole M, Auricchio A, Baron-Esquivias G, et al. 2013 ESC guidelines on cardiac pacing and

5. Rademakers LM, van Hunnik A, Kuiper M, et al. A possible role for pacing the left ventricular septum in cardiac resynchronization therapy. J Am

Reconstruction of the Frank vectorcardiogram from standard electrocardiographic leads: diagnostic comparison of different methods. Eur Heart J 1990;11:1083–92.

Coll Cardiol EP 2016;2:413–22. 6. Mafi-Rad M, Luermans JGLM, Blaauw Y, et al. Feasibility and acute hemodynamic effect of left ventricular septal pacing by transvenous approach through the interventricular septum. Circ Arrhythmia Electrophysiol 2016;9:e003344. 7. Morgan JM, Biffi M, Geller L, et al. ALternate Site Cardiac ResYNChronization (ALSYNC): a prospective and multicentre study of left ventricular endocardial pacing for cardiac resynchronization therapy. Eur Heart J 2016;37:2118–27. 8. Gamble JHP, Herring N, Ginks MR, Rajappan K, Bashir Y, Betts TR. Endocardial left ventricular pacing across the interventricular septum for cardiac resynchronization therapy: clinical results of a pilot study. Heart Rhythm 2018;15:1017–22. 9. Upadhyay GA, Vijayaraman P, Nayak HM, et al. On-treatment comparison between corrective His bundle pacing and biventricular pacing for cardiac resynchronization: a secondary analysis of HisSYNC. Heart Rhythm 2019;16:1797–807. 10. Narula OS. Longitudinal dissociation in the His bundle. Bundle branch block due to asynchronous conduction within the His bundle in man. Circulation 1977;56:996–1006. 11. Arnold A, Shun-Shin M, Keene D, et al. His resynchronization versus biventricular pacing in

cardiac resynchronization therapy. Eur Heart J 2013;34:2281–329. 15. Kors J, van Herpen G, Sittig A, van Bemmel J.

16. Engels EB, Mafi-Rad M, van Stipdonk AMW, Vernooy K, Prinzen FW. Why QRS duration should be replaced by better measures of electrical activation to improve patient selection for cardiac resynchronization therapy. J Cardiovasc Transl Res 2016;9:257–65. 17. van Stipdonk AMW, ter Horst IAH, Kloosterman M, et al. QRS area is a strong determinant of outcome in cardiac resynchronization therapy. Circ Arrhythmia Electrophysiol 2018;11: e006497. 18. Johnson W, Vatterott PJ, Peterson MA, et al.

22. Myerburg

R,

Gelband

H,

Nilsson

K,

23. Strik M, van Deursen CJM, van Middendorp LB, et al. Transseptal conduction as an important determinant for cardiac resynchronization therapy, as revealed by extensive electrical mapping in the dyssynchronous canine heart. Circ Arrhythm Electrophysiol 2013;6:682–9. 24. Prinzen FW, Auricchio A. Is echocardiographic assessment of dyssynchrony useful to select candidates for cardiac resynchronization therapy? Echocardiography is not useful before cardiac resynchronization therapy if QRS duration is available. Circ Cardiovasc Imaging 2008;1:70–8. 25. Durrer D, van Dam RT, Freud GE, Janse MJ, Meijler FL, Arzbaecher RC. Total excitation of the isolated human heart. Circulation 1970;41: 899–912. 26. Laske TG, Skadsberg ND, Hill AJ, Klein GJ, Iaizzo PA. Excitation of the intrinsic conduction system through His and interventricular septal pacing. Pacing Clin Electrophysiol 2006;29: 397–405.

Body surface mapping using an ECG belt to characterize electrical heterogeneity for different left ventricular pacing sites during cardiac resynchronization: relationship with acute hemodynamic improvement. Heart Rhythm 2017;14:385–91.

27. Upadhyay GA, Cherian T, Shatz DY, et al. Intracardiac delineation of septal conduction in

19. Gage RM, Curtin AE, Burns KV, Ghosh S, Gillberg JM, Bank AJ. Changes in electrical dyssynchrony by body surface mapping predict left ventricular remodeling in patients with cardiac

28. Hou cardiac branch septum.

resynchronization therapy. Heart Rhythm 2017;14: 392–9.

left bundle branch block patterns: mechanistic evidence of left intra-Hisian block circumvented by His pacing. Circulation 2019;139:1876–88. X, Qian Z, Wang Y, et al. Feasibility and synchrony of permanent left bundle pacing through the interventricular Europace 2019;21:1694–702.

20. Whinnett ZI, Nott G, Davies JER, et al. Maxi-

29. Lustgarten DL, Crespo EM, ArkhipovaJenkins I, et al. His-bundle pacing versus biventricular pacing in cardiac resynchronization ther-

mizing efficiency of alternation algorithms for hemodynamic optimization of the AV delay of

apy patients: a crossover design comparison. Heart Rhythm 2015;12:1548–57.

JACC VOL. 75, NO. 4, 2020

Salden et al.

FEBRUARY 4, 2020:347–59

LV Septal Pacing for CRT

30. Grosfeld MJW, Res JCJ, Vos DHS, de Boer TJM, Bos HJ. Testing a new mechanism for left interventricular septal pacing: the transseptal route. Europace 2002;4:439–44.

34. Vijayaraman P, Subzposh FA, Naperkowski A, et al. Prospective evaluation of feasibility, electrophysiologic and echocardiographic characteristics of left bundle branch area pacing. Heart Rhythm 2019;16:1774–82.

31. Huang W, Su L, Wu S, Xu L, Xiao F. A novel pacing strategy with low and stable output: pacing the left bundle branch immediately beyond the

35. Li X, Li H, Ma W, et al. Permanent left bundle branch area pacing for atrioventricular block:

conduction block. Can J Cardiol 2017;33. 1736.e1– 3.

feasibility, safety and acute effect. Heart Rhythm 2019;16:1766–73.

32. Sieniewicz BJ, Gould J, Porter B, et al. Understanding non-response to cardiac resynchronisation therapy: common problems and potential solutions. Heart Fail Rev 2019;24:41–54.

36. de Roest G, Knaapen P, Gotte M, et al. Stroke work or systolic dP/dtmax to evaluate acute response to cardiac resynchronization therapy: are they interchangeable? Eur J Heart Fail 2009;11: 706–8.

33. Fish JM, Brugada J, Antzelevitch C. Potential proarrhythmic effects of biventricular pacing. J Am Coll Cardiol 2005;46:2340–7.

37. Duckett SG, Ginks M, Shetty AK, et al. Invasive acute hemodynamic response to guide

left ventricular lead implantation predicts chronic remodeling in patients undergoing cardiac resynchronization therapy. J Am Coll Cardiol 2011; 58:1128–36. 38. Zweerink A, Salden OAE, van Everdingen WM, et al. Hemodynamic optimization in cardiac resynchronization therapy: should we aim for dP/ dtmax or stroke work? J Am Coll Cardiol EP 2019; 5:1013–25.

KEY WORDS body surface mapping, cardiac resynchronization therapy, heart failure, hemodynamics, His bundle pacing, ventricular septum

359