Evaluation of cardiac energetics in left bundle branch area pacing using noninvasive pressure–volume loops

Abstract

Background

Conduction system pacing, particularly left bundle branch area pacing (LBBAP), is emerging as the preferred strategy for bradyarrhythmias and cardiac resynchronization therapy because of its clinical benefits. However, the differences in the impact of LBBAP vs right ventricular pacing (RVP) on cardiac energetics remain unclear.

Objective

The purpose of this study was to compare the effects of LBBAP and RVP on cardiac energetics using a novel noninvasive pressure–volume (PV) loop algorithm derived from cardiac magnetic resonance (CMR).

Methods

We included patients who underwent permanent pacemaker implantation with LBBAP or RVP between January 2021 and October 2024, followed by CMR imaging. Noninvasive PV loops were calculated using volumes derived from CMR and a time-varying elastance model. The parameters obtained from the PV loop analysis were compared between the LBBAP and RVP groups. Sensitivity analyses were performed by varying the left ventricular end-diastolic pressure (3–40 mm Hg) and excluding anodal pacing cases.

Results

Of the 235 patients, 15 were analyzed (LBBAP group, n = 7; RVP group, n = 8). The end-systolic PV relationship was significantly higher in the LBBAP group (2.25 mm Hg/mL [1.85–2.55]) compared to the RVP group (1.39 mm Hg/mL [1.33–1.69]; P <.0001). The potential energy and PV area were significantly lower in the LBBAP group (P = .0006 and P = .006, respectively), whereas the ventricular efficiency was significantly higher (63.1% vs 57.4%; P = .006).

Conclusion

LBBAP reduces myocardial energy consumption and improves ventricular efficiency compared with RVP, suggesting its advantages in pacing-associated cardiac energetics.

Graphical abstract

Key Findings.

• ▪Left bundle branch area pacing reduced pressure–volume area and improved ventricular efficiency.
• ▪Left bundle branch area pacing has potential to suppress myocardial energy during pacing.
• ▪Left bundle branch area pacing has potential to suppress oxygen consumption during pacing.
• ▪Left bundle branch area pacing is a beneficial pacing modality from the perspective of cardiac energetics.

Introduction

Conduction system pacing (CSP), particularly left bundle branch area pacing (LBBAP), has become increasingly prevalent as a therapeutic strategy for bradyarrhythmias and cardiac resynchronization therapy (CRT).^1,2 Historically, right ventricular pacing (RVP) posed challenges due to wide QRS complexes and the resulting left ventricular (LV) dyssynchrony and systolic dysfunction.³ In response, the utility of LBBAP in addressing these issues has been widely reported.^4,5 Furthermore, LBBAP has demonstrated clinical advantages over RVP and compared to traditional CRT (biventricular CRT).^5,6 As a result, the position of CSP in the guidelines also was revised.⁷

Despite these advancements, the differences in the impact of RVP and LBBAP on cardiac energetics remain unclear. The gold standard for evaluating cardiac mechanics, including LV stroke work (SW) and energy consumption, is the pressure–volume (PV) loop.⁸ Previous studies utilizing intraventricular pressure catheters have demonstrated that CRT improves energy efficiency.⁹ However, obtaining PV loops traditionally required invasive LV conductance catheterization, thus limiting its routine clinical use. Recently, myocardial work derived from echocardiography and brachial blood pressure (GE Healthcare, Pewaukee, WI) has enabled PV looplike assessments and facilitated studies comparing the efficiency of LBBAP and RVP.¹⁰ Although one study reported significantly improved myocardial work efficiency with LBBAP,¹⁰ these results are specific to myocardial work metrics and differ from traditional PV loop parameters. Moreover, the correlation between myocardial work-derived metrics and the invasively obtained PV loops remains unclear. Technical limitations, such as imaging quality and frame rate, also persist.¹¹

Recently, a novel noninvasive approach using cardiac magnetic resonance (CMR) and brachial blood pressure to derive PV loops has been proposed.¹² CMR offers highly accurate measurements of cardiac volumes. Studies have demonstrated strong correlations between the PV loops derived from this method and those obtained invasively via LV conductance catheters.¹³ This algorithm provides a low-invasive yet reliable means of evaluating cardiac mechanics.

This study aimed to clarify the differences in the impact of LBBAP and RVP on cardiac energetics using a novel noninvasive PV loop algorithm based on CMR. In failing hearts, mechanical workload (pressure–volume area [PVA]) increases, leading to greater myocardial oxygen consumption and worsened prognosis.¹⁴ Demonstrating the superiority of LBBAP over RVP in cardiac energetics would further solidify the benefits of LBBAP. Thus, this study has significant clinical relevance.

Methods

Study design, participants, and measurements

This retrospective study included patients who underwent permanent pacemaker (PM) implantation with LBBAP or RVP to treat bradyarrhythmias between January 2021 and October 2024. We aimed to evaluate the impact of different pacing modalities (LBBAP or RVP) used for the treatment of bradyarrhythmias on myocardial work efficiency using PV loop analysis. To achieve this objective, we excluded patients with ischemic heart disease, severe valvular disease, amyloidosis, sarcoidosis, or unexplained LV dysfunction (left ventricular ejection fraction [LVEF] <50%), as these conditions could confound the interpretation of PV loop results. Additionally, to safely perform CMR in patients with PMs, the final analysis included only those who did not fall under the following exclusion criteria: LVEF <50% on echocardiography; PM implantation for tachybrady syndrome; presence of severe valvular heart disease; chronic atrial fibrillation; chronic kidney disease with estimated glomerular filtration rate <30 mL/min/1.73 m² or on maintenance dialysis; or inability to undergo CMR after PM implantation.

This study was conducted in accordance with the Helsinki Declaration and current ethical guidelines and was approved by the Institutional Ethics Committee (Approval No.: 1287240807). Written informed consent was obtained from all participants for PM implantation, CMR examination, and use of their data for research purposes.

Device implantation

The procedure and success criteria for LBBAP implantation were based on those of previous studies.15, 16, 17, 18, 19, 20 In summary, LBBAP was considered successful when both the following mandatory criteria were met (Figure 1A): (1) V₁ lead QRS morphology demonstrating right ventricular conduction delay (Qr or qR pattern); and (2) paced R-wave activation time in V₆ <75 ms for non–left bundle branch block morphology or <80 ms for left bundle branch block morphology, consistently stable across pacing outputs

Figure 1.

Electrocardiographic (ECG) and cardiac magnetic resonance (CMR) analysis in left bundle branch area pacing (LBBAP) and right ventricular pacing (RVP). A: LBBAP ECG (right: intracardiac electrogram measurements). B: RVP ECG. C: CMR image analysis (green: external ventricular volume; red: internal ventricular volume).

In additionally, the procedure was considered successful if at least one of the following supplementary criteria was met: (1) detection of the left bundle potential; (2) V₆–V₁ interpeak interval >33 ms; or (3) transition of QRS morphology from nonselective left bundle branch pacing to LV septal or selective left bundle branch pacing.²¹

For RVP, a preshaped stylet was used to guide lead placement. Under fluoroscopic guidance in the left anterior oblique and right anterior oblique views, the lead was secured to the right ventricular septum or apex (Figure 1B).²²

CMR and image analysis

CMR examination was performed at least 6 weeks after PM implantation to ensure safety. The pacing mode was set to VOO-90 bpm in asynchronous mode during the examination, and the PM parameters were confirmed to show no abnormal changes before and after the scan.

CMR scans were performed using a Discovery MR750w 3.0T system (GE Healthcare). Imaging was performed using retrospective electrocardiographic gating during end-expiratory breath-holding. The typical spatial resolution was 2.2 × 1.8 × 8 mm, reconstructed into 25 frames. Standard imaging parameters included repetition time of 3.6 ms, echo time of 1.5 ms, and flip angle of 50°.¹²

Image analysis was performed using a Segment 4.0 R11026 (Medviso, Lund, Sweden). LV volumes were calculated across the cardiac cycle using time-resolved delineation of the LV endocardial border from the short-axis stack images (Figure 1C).²³

PV loop computation

Computation of the PV loop in this study utilized the time-varying elastance model expressed by the double-Hill equation.^12,13 In summary, left ventricular pressure (LVP) was estimated using noninvasive blood pressure measurements during CMR. Systolic blood pressure was used to approximate peak LVP, and left ventricular end-diastolic pressure was set at 7.5 mm Hg based on previous studies. The PV relationship was modeled using a time-varying elastance function, where ventricular pressure was calculated as P(t) = E(t) × (V(t) – V₀), with E(t) representing elastance and V₀ is the volume at zero pressure. To enhance accuracy, CMR-derived volume data were interpolated to 100 points per cardiac cycle, ensuring a smooth PV loop calculation. Detailed methods are provided in the Supplemental Material.

Definitions of PV loop parameters

The definitions of PV loop parameters were based on previous studies.¹³ In summary, SW was defined as the area enclosed by the PV loop. The end-systolic pressure–volume relationship (ESPVR) was the slope between volume 0 (V₀, set to zero) and Emax (the point where the time-varying elastance function reaches its maximum value). Arterial elastance (Ea) was calculated as the slope from Emax to the point where the end-diastolic volume (EDV) intersected with a pressure of 0 mm Hg. Ventricular–arterial (VA) coupling was defined as Ea/Emax.²³ Mechanical potential energy (PE) was the triangular area with a base between V₀ and the end-systolic volume (ESV) and a height corresponding to ESPVR at the ESV.

The following equations were used to calculate the additional parameters (Figure 2):

Figure 2.

Scheme of the pressure–volume loop using the time-varying elastance model. Ea = arterial elastance; EDV = end-diastolic volume; Emax = point where the time-varying elastance function reaches its maximum value; EPEV = energy per ejected volume; ESPVR = end-systolic pressure–volume relationship; ESV = end-systolic volume; MEP = mean external power; PE = potential energy; PVA = pressure–volume area; SV = stroke volume; SW = stroke work; VE = ventricular efficiency.

Stroke volume (SV) = EDV – ESV.

Total mechanical energy: PVA = SW + PE.

Mean external power (MEP) = SW ∗ HR/60.

Ventricular efficiency (VE) = SW/PVA.

Energy per ejected volume (EPEV) = PVA/SV.

To account for the influence of body size, EDV, ESV, and SV were indexed by body surface area to obtain body surface area–normalized values.

Sensitivity analysis based on differences in end-diastolic pressure

To evaluate changes in PV loop parameters with varying end-diastolic pressures (EDPs), sensitivity analysis was performed by setting EDP to 3, 7.5, 15, 25, and 40 mm Hg while keeping other values constant. Changes in EDP were achieved by setting the specified EDP for LVP during diastole, in accordance with Emin as defined by the double-Hill equation. Subsequently, the PV loop and its respective parameters were recalculated using the corresponding E(t) values, adjusted to reflect these changes. Intergroup comparisons between the LBBAP and RVP groups were performed for each EDP value.

Statistical analysis

Categorical variables are given as number (percentage). Continuous variables are given as median [interquartile range]. Intergroup comparisons between the LBBAP and RVP groups were performed using nonparametric tests for categorical and continuous variables. Repeated measures analysis of variance was used to compare PV loop parameters between the 2 groups. Cohen d was calculated to evaluate the effect size between the 2 groups. P <.05 was considered significant. All statistical analyses were performed using Stata MP 18.0 (StataCorp 2023, Stata Statistical Software Release 18, StataCorp LLC, College Station, TX).

Results

Patient selection and baseline characteristics

The final analysis included 15 patients (7 in the LBBAP group and 8 in the RVP group) (Figure 3). All patients had atrioventricular block as the underlying type of bradyarrhythmia. Baseline characteristics of the study population are given in Table 1. Median age of the LBBAP group was 79 years, which was slightly higher than that of the RVP group (72 years), although the difference was not statistically significant (P = .27). Serum creatinine levels were significantly lower in the LBBAP group (P = .017). No significant differences were observed in other laboratory parameters or medical history.

Figure 3.

Flowchart of patient selection. CKD = chronic kidney disease; CMR = cardiac magnetic resonance; HD = heart disease; LVEF = left ventricular ejection fraction; PM = permanent pacemaker; PV = pressure–volume; other abbreviations as in Figure 1.

Table 1.

Comparison of patient characteristics

	LBBAP group (n = 7)	RVP group (n = 8)	P value
Age, y	79 [75−83]	72 [67−77]	.27
Sex, male	1 (14.3)	5 (62.5)	.12
BMI, kg/m2	23.0 [20.7−25.1]	24.9 [22.9−28.3]	.46
BSA, m2	1.45 [1.38−1.51]	1.85 [1.62−1.92]	.054
Laboratory data
Creatinine, mg/dL	0.80 [0.71−0.82]	0.94 [0.87−1.05]	.017
eGFR, mL/min/1.73 m2	59.4 [53.3−62.5]	53.0 [46.3−66.4]	.78
HbA1c, %	5.7 [5.7−6.9]	6.0 [5.6−6.3]	.95
BNP, pg/mL	43.6 [31.2−50.9]	30.2 [19.1−97.1]	.78
Medical history
Hypertension	6 (85.7)	3 (37.5)	.12
Diabetes mellitus	2 (28.6)	1 (12.5)	.57
Dyslipidemia	3 (42.9)	(0.0)	.077
Current smoking	0 (0.0)	1 (12.5)	>.99
Coronary artery disease	1 (14.3)	1 (12.5)	>.99
Treatment at baseline
Beta-blockers	1 (14.3)	1 (12.5)	>.99
RAAS inhibitors	4 (57.1)	2 (25.0)	.32
SGLT2 inhibitor	1 (14.3)	1 (12.5)	>.99
Echocardiogram before PM implantation
EDV, mL	82.0 [73.0−90.0]	91.5 [86.3−94.5]	.27
ESV, mL	34.0 [28.5−38.5]	34.5 [30.5−41.5]	.82
LVEF, %	61.0 [55.0−67.5]	60.5 [57.5−67.0]	.82
AS mild or moderate	0 (0.0)	0 (0.0)	>.99
AR mild or moderate	4 (57.1)	4 (50.0)	>.99
MR mild or moderate	7 (100.0)	7 (87.5)	>.99
TR mild or moderate	6 (85.7)	8 (100.0)	.47
Electrocardiogram before PM implantation
LBBB	2 (28.6)	2 (25.0)	>.99
RBBB	2 (28.6)	4 (50.0)	.61
At pacing
QRS duration, ms	110.0 [109.5−114.5]	143.5 [136.8−147.0]	.001
V6 RWAT, ms	61 [59.5−66.5]	NA
V6–V1, ms	42.0 [38.5−45.5]	NA
PM parameters
Atrial pacing rate	33.9 [16.8−59.0]	20.7 [14.0−42.0]	.69
Ventricular pacing rate	95.2 [11.2−98.0]	72.0 [10.0−99.8]	.73
Period from PM implantation to CMR, d	362 [43.0−550]	226 [45.3−455]	.90

Data are given as median [interquartile range] for continuous variables, and n (number of patients) and (%).

AR = aortic regurgitation; AS = aortic stenosis; BMI = body mass index; BNP = brain natriuretic protein; BSA = body surface area; CMR = cardiac magnetic resonance; EDV = end-diastolic volume; eGFR = estimated glomerular filtration rate; ESV = end-systolic volume; HbA_1c = hemoglobin A_1c; LBBAP = left bundle branch area pacing; LBBB = left bundle branch block; LVEF = left ventricular ejection fraction; MR = mitral regurgitation; PM = pacemaker; RAAS = renin-angiotensin-aldosterone system; RBBB = right bundle branch block; RVP = right ventricular pacing; RWAT = R-wave activation time; SGLT2 = sodium-glucose cotransporter 2; TR = tricuspid regurgitation.

Echocardiographic findings revealed no significant differences between the groups in terms of LVEF or frequency of valvular disease. During pacing, QRS duration in the LBBAP group was 110.0 [109.5–114.5] ms, significantly shorter than the 143.5 [136.8–147.0] ms in the RVP group (P = .001).

The period from PM implantation to CMR examination was 362 days [43–550 days], with no significant difference between the 2 groups (P = .90). The changes in EDV between baseline and the time of CMR in the 2 groups are given in Supplemental Figure S1.

Comparison of CMR and PV loop parameters (EDP = 7.5 mm Hg)

The results of CMR analysis, blood pressure measurements, and PV loop parameters at EDP = 7.5 mm Hg were compared between the LBBAP and RVP groups (Table 2 and Figure 4). All the analyzed PV loops are shown in Supplemental Figure S2.

Table 2.

CMR analysis and PV loop parameters (EDP = 7.5 mm Hg)

	LBBAP group (n = 7)	RVP group (n = 8)	P value
Systolic BP, mm Hg	126 [113−151]	125 [114−141]	.91
Diastolic BP, mm Hg	88 [70−99]	81 [76−93]	.64
Estimated LVP systole, mm Hg	113 [97−135]	109 [101−124]	.60
EDV, mL	67.8 [62.0−71.8]	90.9 [87.0−98.1]	<.0001
ESV, mL	33.8 [30.8−37.0]	58.4 [50.1−62.3]	<.0001
EDV index, mL/m2	47.9 [45.1−48.9]	52.2 [49.1−61.4]	.029
ESV index, mL/m2	24.4 [21.3−26.1]	34.1 [28.4−35.9]	.006
Stroke volume, mL	32.0 [29.9−35.1]	35.9 [33.1−39.4]	.33
Stroke volume index, mL/m2	22.6 [21.6−24.1]	21.4 [19.3−24.5]	.69
LVEF, %	58.8 [56.4−61.7]	54.2 [52.6−58.8]	.19
ESPVR, mm Hg/mL	2.25 [1.85−2.55]	1.39 [1.33−1.69]	<.0001
Ea, mm Hg/mL	2.19 [2.14−2.60]	2.32 [2.00−2.69]	.87
Ventricular-arterial coupling	1.16 [1.06−1.23]	1.47 [1.39−1.61]	.006
Stroke work, J	0.26 [0.25−0.39]	0.38 [0.32−0.48]	.12
Potential energy, J	0.19 [0.15−0.20]	0.32 [0.28−0.35]	.0006
PVA, J	0.44 [0.40−0.61]	0.74 [0.64−0.82]	.006
Mean external power, W	0.38 [0.38−0.58]	0.57 [0.48−0.72]	.12
Ventricular efficiency, %	63.1 [61.5−63.6]	57.4 [54.0−59.2]	.006
EPEV, J/L	15.2 [14.1−17.6]	18.0 [17.2−22.5]	.054

Data are given as median [interquartile range] for continuous variables.

BP = blood pressure; Ea = arterial elastance; EDP = end-diastolic pressure; EPEV = energy per ejected volume; ESPVR = end-systolic pressure-volume relationship; LVP systole = left ventricular peak pressure; PV = pressure–volume; PVA = pressure–volume area; other abbreviations as in Table 1.

Figure 4.

Comparison of pressure-volume loop parameters between LBBAP and RVP using box plots. Abbreviations as in Figures 1 and 2.

Systolic blood pressure was 126 [113–151] mm Hg and 125 [114–141] mm Hg in the LBBAP and RVP groups, respectively, with no significant difference (P = .91). Similarly, no significant differences were observed in diastolic blood pressure or estimated LVP during systole (P = .64 and P = .60, respectively).

Conversely, EDV was significantly lower in the LBBAP group (67.8 [62.0–71.8] mL) than in the RVP group (90.9 [87.0–98.1] mL; P <.0001). Similarly, ESV was significantly lower in the LBBAP group (33.8 [30.8–37.0] mL) than in the RVP group (58.4 [50.1–62.3] mL; P <.0001).

There was no significant difference in LVEF between the groups (LBBAP: 58.8% [56.4%–61.7%] vs RVP: 54.2% [52.6%–58.8%]; P = .19). Similarly, SV showed no significant difference (LBBAP: 32.0 [29.9–35.1] mL vs RVP: 35.9 [33.1–39.4] mL; P = .33).

PV loop analysis revealed that the ESPVR was significantly higher in the LBBAP group (2.25 [1.85–2.55] mm Hg/mL) compared to the RVP group (1.39 [1.33–1.69] mm Hg/mL; P <.0001). VA coupling was closer to 1 in the LBBAP group than in the RVP group (P = .006).

Whereas no significant difference was observed in SW (P = .12), PE was significantly lower in the LBBAP group (0.19 [0.15–0.20] J) compared to the RVP group (0.32 [0.28–0.35] J; P = .0006). Consequently, PVA (SW + PE) also was significantly lower in the LBBAP group (0.44 [0.40–0.61] J) than in the RVP group (0.74 [0.64–0.82] J; P = .006). VE was significantly higher in the LBBAP group (63.1% [61.5%–63.6%]) than in the RVP group (57.4% [54.0%–59.2%]; P = .006).

Effect size analysis using Cohen d revealed large effects for the comparison of ESPVR (d = 1.42, 95% confidence interval [CI] 0.17–2.67) and VE (d = 1.29, 95% CI 0.06–2.52), with higher values observed in the LBBAP group. In contrast, PE (d = –2.32, 95% CI –3.76 to –0.87) and PVA (d = –1.66, 95% CI –2.96 to –0.37) showed large negative effects, indicating higher values in the RVP group.

Additionally, evaluation of the correlation between VA coupling and VE showed a very strong and significant negative correlation (correlation coefficient = –0.97, 95% CI –0.99 to –0.92, P <.0001) (Supplemental Figure S3).

Sensitivity analysis of PV loop parameters with changes in EDP (3, 7.5, 15, 25, and 40 mm Hg)

The effects of changes in EDP (3, 7.5, 15, 25, and 40 mm Hg) and pacing modality (RVP and LBBAP) on SW, PVA, MEP, VE, and EPEV were analyzed using repeated-measures analysis of variance (Figure 5).

Figure 5.

Repeated measures analysis of variance analysis of pressure–volume loop parameters with varying end-diastolic pressure (EDP). MEP = mean external power; other abbreviations as in Figure 1, Figure 2.

The main effect of the changes in EDP on SW was significant (P <.0001, F = 330.52). Whereas the main effect of pacing modality was not statistically significant (P = .14, F = 2.45), the interaction between EDP and pacing modality was significant (P <.0001, F = 8.32), suggesting that the effect of EDP changes differed depending on pacing modality.

MEP: Main effect of EDP changes: P <.0001, F = 330.52; main effect of pacing modality: P = .14, F = 2.45; interaction between EDP and pacing modality: P <.0001, F = 8.32.

EPEV: Main effect of EDP changes: P <.0001, F = 867.86; main effect of pacing modality: P = .099, F = 3.15; interaction between EDP and pacing modality: P 0.0003, F=6.52.

For PVA, the main effect of EDP changes was significant (P <.001, F = 312.68). Additionally, a significant difference was observed between the LBBAP and RVP groups (P = .0075, F = 9.98). The interaction between EDP and pacing modality also was significant (P <.0001, F = 8.86). For VE, a trend similar to that observed for PVA was observed. The main effect of EDP change was significant (P <.0001, F = 283.76), and there was a significant difference between the LBBAP and RVP groups (P = .026, F = 6.28). However, the interaction between EDP and pacing modality was not significant (P = .60, F = 0.69).

Sensitivity analysis of PV loop parameters considering the impact of anodal pacing

During CMR imaging, PM settings were adjusted to VOO mode at 5 V to ensure safety. In the LBBAP group, anodal capture was confirmed in 4 cases, with 3 of these cases demonstrating an anodal threshold of 2.5–4V. Consequently, the evaluations in these three cases were performed under anodal pacing conditions. To account for the potential impact of anodal pacing, sensitivity analysis was performed by excluding cases with anodal pacing (Table 3). The results indicated that the statistical difference in ESPVR between the LBBAP and RVP groups was not observed. However, findings for other PV loop parameters remained consistent across groups, suggesting a minimal overall impact of anodal pacing on these parameters.

Table 3.

Sensitivity analysis excluding anodal pacing (EDP = 7.5 mm Hg)

	LBBAP group (n = 4)	RVP group (n = 8)	P value
Systolic BP, mm Hg	120 [111−134]	125 [114−141]	.73
Diastolic BP, mm Hg	88 [71−91]	81 [76−93]	.50
Estimated LVP systole, mm Hg	106 [97−120]	109 [101−124]	.35
EDV, mL	68.7 [66.6−72.3]	90.9 [87.0−98.1]	.008
ESV, mL	37.0 [36.1−38.6]	58.4 [50.1−62.3]	.004
Stroke volume, mL	31.7 [30.3−34.0]	35.9 [33.1−39.4]	.46
LVEF, %	59.1 [57.0−60.6]	54.2 [52.6−58.8]	.34
ESPVR, mm Hg/mL	1.85 [1.78−2.03]	1.39 [1.33−1.69]	.11
Ea, mm Hg/mL	2.27 [2.11−2.50]	2.32 [2.00−2.69]	>.99
Ventricular-arterial coupling	1.17 [1.16−1.21]	1.47 [1.39−1.61]	.049
Stroke work, J	0.31 [0.26−0.37]	0.38 [0.32−0.48]	.28
Potential energy, J	0.20 [0.18−0.22]	0.32 [0.28−0.35]	.008
PVA, J	0.51 [0.43−0.59]	0.74 [0.64−0.82]	.028
Mean external power, W	0.46 [0.38−0.56]	0.57 [0.48−0.72]	.28
Ventricular efficiency, %	63.1 [61.6−63.5]	57.4 [54.0−59.2]	.049
EPEV, J/L	14.8 [14.3−16.3]	18.0 [17.2−22.5]	.072

Data are given as median [interquartile range] for continuous variables.

Abbreviations as in Table 1, Table 2.

Discussion

This study was the first to compare LBBAP and RVP in terms of cardiac energetics using noninvasive PV loop analysis derived from CMR. The analysis suggested that LBBAP may suppress myocardial energy and oxygen consumption compared with RVP.

The evaluation of implanted devices using PV loops has been extensively reported, particularly in CRT research.^24,25 These studies demonstrate that optimizing pacing sites or using multipoint pacing consistently improves SW and LVEF and acute hemodynamic indices, such as dP/dtmax. However, to our knowledge, no studies have evaluated these indices using PV loop analysis for LBBAP.

In the present study, LBBAP showed a leftward shift in the PV loop, resulting in a steeper ESPVR slope and lower PVA compared with RVP. PVA was known to correlate with myocardial oxygen consumption.²⁶ This suggests that LBBAP provides pacing that reduces myocardial oxygen consumption.

From an energetic perspective, the results for VA coupling indicate smoother energy transfer between the ventricle and arteries with LBBAP. The strong correlation observed with VE further underscores the high energy efficiency of LBBAP, supporting its benefits and offering a new rationale for choosing LBBAP. When RVP is used to treat bradyarrhythmias, it has been reported to cause electrical and mechanical dyssynchrony, resulting in cardiac remodeling and a decline in LVEF.³ In this context, LBBAP, with its characteristics of improving energy efficiency and reducing oxygen consumption, is anticipated to become a standard pacing modality for bradyarrhythmias in the future. However, the long-term clinical benefits of LBBAP in this patient population have not yet been fully established, highlighting the need for further large-scale and long-term studies. We speculated that the underlying mechanism for these findings may be long-term pacing-induced changes in LV morphology.

In the baseline characteristics of the present study, no significant differences were observed in preimplantation LVEF, ventricular pacing percentage, period between PM implantation and CMR, or CMR-derived LVEF and SV. However, significant differences were noted in EDV and ESV assessed using CMR. Chronic RVP has been shown to exert multifaceted adverse effects on the LV.²⁷ Conversely, reverse remodeling has been reported with LBBAP.²⁸ Integrating the findings from previous studies and the current investigation, LBBAP was an effective pacing modality electrically, mechanically, and in terms of energy efficiency. Thus, it is anticipated to become the first-line treatment for bradycardia.

However, the potential effects of anodal capture specific to LBBAP warrant further investigation. Bipolar LBBAP may induce anodal capture, which previous studies have reported to reduce hemodynamic benefits.²⁹ Those studies used changes in dP/dtmax as hemodynamic indices, whereas in the present study, sensitivity analysis excluding cases of anodal pacing showed a tendency toward smaller statistical differences between the LBBAP and RVP groups. Whether this was due to the reduced sample size in the LBBAP group or differences in the hemodynamic indices evaluated by the PV loops remains unclear.

Another study involving 21 patients reported that although anodal capture activates the right ventricle, it does not contribute to acute hemodynamic improvements.³⁰ Further large-scale studies are required to clarify these points.

Study limitations

First, because this was a single-center study with a relatively small sample size, it is possible that some factors failed to reach statistical significance. To mitigate this, we established clear inclusion and exclusion criteria to minimize selection bias as much as possible. However, the generalizability of the conclusions in this study may be limited due to the reduced sample size resulting from the restriction of the study population. Additionally, in the sensitivity analysis excluding anodal pacing cases, no statistically significant difference was observed in ESPVR. However, considering the potential impact of the small sample size, we performed an effect size analysis using Hedges g. The analysis yielded Hedges g = 1.22 (95% CI –0.14 to 2.58), indicating that although the result was not statistically significant, a clinically meaningful difference between the 2 groups still may be present. Second, PV loop analysis in the present study did not involve invasive direct measurement of LV pressure but instead was based on an established model from previous research. As such, the PV loop derived from CMR and fixed parameters cannot capture dynamic changes, particularly those related to preload and afterload, which involve multiple interdependent factors. Future studies are required to validate whether the results of this analysis accurately reflect the actual cardiac energetics.

Conclusion

Compared with RVP, LBBAP reduced PVA and improved ventricular efficiency, suggesting its potential to suppress myocardial energy and oxygen consumption during pacing. These characteristics were consistent, even with variations in EDP, indicating that LBBAP is a beneficial pacing modality from the perspective of cardiac energetics.