Positional Preload Changes and Hemodynamic Tolerance in LBBAP

Part one: the mechanistic answer

Short answer: yes, positional preload shifts do transiently change filling pressures in an LBBAP-paced ventricle, and they can produce discomfort — but LBBAP itself is usually the protective variable in that equation rather than the culprit. The dominant determinants are device programming and the patient's diastolic and autonomic reserve.

Why LBBAP generally improves tolerance to preload swings

Conduction system capture preserves the physiologic left ventricular activation sequence, which shortens isovolumic contraction, lengthens effective diastolic filling time, and avoids the dyssynchrony-driven functional mitral regurgitation and left atrial pressure elevation seen with right ventricular apical pacing. That translates into more headroom when standing drops central volume by several hundred milliliters: stroke volume falls, but from a higher baseline and with better diastolic recoil. Conversely, on recumbency the left ventricle handles the acute rise in end-diastolic volume with a more normal pressure–volume relationship, so the pulmonary capillary wedge pressure excursion is blunted relative to a dyssynchronously paced ventricle.

Where the LBBAP-specific vulnerabilities actually sit

• Right ventricular activation delay. LBBAP produces an RBBB-like activation pattern — the right ventricle is depolarized late, via transseptal spread. In most patients this is hemodynamically trivial, but in someone with pre-existing RV dysfunction, pulmonary hypertension, or significant tricuspid regurgitation, an abrupt rise in RV preload (supine, leg elevation, bending forward) meets a ventricle contracting non-uniformly. That can amplify right atrial pressure, TR severity, and the sensation of neck pulsation or right upper quadrant fullness. This is one of the arguments for LOT-CRT or HOT-CRT in selected patients.
• Position-dependent capture behavior. Deep septal leads can show variation in threshold or intermittent loss of selective conduction system capture, with beat-to-beat QRS and hemodynamic change. Worth a supine-versus-standing 12-lead with V6 R-wave peak time if the history is positional.
• Diaphragmatic or intercostal stimulation. Genuinely position-dependent, because the anatomic relationship of a septal lead to the diaphragm changes with posture. This is a very common explanation for "positional discomfort" that gets attributed to hemodynamics.

The programming variables that usually dominate

Chronotropic reserve is the single biggest one. Orthostatic tolerance depends on the baroreflex tachycardia; a patient in VVI, with a poorly configured accelerometer or minute-ventilation sensor, or bumping into upper-rate behavior, cannot compensate for the stroke volume drop regardless of how good the activation sequence is.

AV coupling is second. The atrial contribution to left ventricular filling matters more at reduced preload. Loss of effective AV synchrony — atrial systole against a closed mitral valve — produces exactly the symptom cluster described: neck pulsation, chest fullness, malaise, dyspnea, often worse recumbent because higher venous return amplifies the regurgitant atrial waves.

On the discomfort question specifically

The plausible hemodynamic mechanisms are: supine augmentation of central blood volume raising wedge pressure in a ventricle with limited diastolic reserve (the orthopnea physiology); positional accentuation of tricuspid regurgitation with right atrial and hepatic congestion; and orthostatic hypotension, which in this population is frequently pharmacologic — beta-blockade, ARNI, diuretics, SGLT2 inhibitors — rather than device-related. Non-hemodynamic causes are at least as common and much easier to confirm.

Distinguishing them

Interrogation first: mode, rate-response configuration, paced and sensed AV delays, upper-rate behavior, threshold trends, atrial fibrillation burden, and rate histograms correlated to symptom times. Then orthostatic vitals with simultaneous rhythm capture, positional 12-leads to check for loss of conduction system capture, and echocardiography with AV optimization plus tricuspid regurgitation assessment.

Part two: what the literature actually shows

The specific data you would want still does not exist, but two findings sharpen the picture — and one of them corrects something stated above.

The correction first

Part one suggested that posture shifts intrinsic AV conduction enough to make a rest-optimized AV delay wrong when upright. The direct measurement does not support that. In a multicenter study of patients with cardiac resynchronization devices assessed supine, sitting, and standing, AV intervals changed only minimally with posture at rest, while atrial pacing immediately lengthened the AV interval relative to sinus rhythm, with atrial-to-right-ventricular slopes of roughly 8 ms per beat per minute across all three positions [2].

So the AV timing problem is rate-driven, not posture-driven. Posture matters mainly through the heart rate it provokes — which puts chronotropic reserve back at the center rather than posture-specific interatrial conduction.

The mechanism that does hold up

A CircAdapt virtual-patient study published in Europace in 2025 is the closest thing to a direct answer, because it varies loading and rate systematically [1]. Three findings bear on the question.

First, non-selective left bundle branch pacing lowered mean left atrial pressure more than biventricular pacing in every simulated phenotype, while stroke volume differences between the two were minimal. LBBAP's advantage is disproportionately a filling-pressure advantage — exactly the axis in question.

Second, the optimal AV delay shifted to shorter values as exercise intensity rose, and the parabolic curves relating mean left atrial pressure to AV delay steepened, meaning filling pressure became progressively more sensitive to AV delay at higher heart rates. Holding the rest-optimized AV delay fixed through exercise raised filling pressure, whereas stage-specific adjustment lowered it, most markedly in patients with elevated baseline filling pressures. In the stiff-myocardium virtual patient, that gap reached several mmHg at high workload.

Third — and this is the clinically pointed part — patients with increased chamber stiffness showed the greatest exercise filling-pressure benefit from AV optimization despite showing little stroke-volume gain at rest. The resting acute response in stroke volume or dP/dt_max does not reflect the potential benefit under load. That is precisely the patient who looks fine on the resting optimization echo and still feels awful when loading changes.

The two failure modes map onto the discomfort question directly. A long AV delay produced diastolic mitral regurgitation with a faster pressure rise during early atrial filling; a short one produced a prominent cannon A wave in the left atrial pressure signal, from atrial contraction against a closing mitral valve. Both are especially problematic where filling pressure is already high, since a suboptimal setting pushes it higher still and raises the risk of pulmonary congestion [1].

The right-ventricular concern should be downgraded

Part one overweighted it. In patients with right bundle branch block and RV dyssynchrony manifest as reverse septal flash or basal bulge, LBBAP corrected these abnormalities and rarely produced de novo RV dyssynchrony; despite the RBBB-like paced pattern, the net effect on RV function and synchrony was protective, with limited tricuspid regurgitation deterioration on long-term follow-up [3]. Meta-analysis found RV fractional area change and TAPSE both significantly improved versus intrinsic conduction, and the RBBB pattern did not appear to meaningfully impair RV systolic function [4].

Tricuspid regurgitation is the exception, and it is time-dependent.

The modifiable variable is anatomic: the distance from the lead entry site to the tricuspid annulus is the predictor meta-analysis has focused on for post-LBBAP TR deterioration [5] — though a more distal position, around 21.5 mm from the annulus, has not proven inferior for interventricular synchrony [6]. For someone with positional right-sided congestion more than a year after implant, TR progression is a more plausible substrate than RV activation delay itself.

One programmable lever is worth adding: left bundle branch pacing with anodal capture advanced right ventricular activation compared with pacing without it in all patients studied, with earliest RV activation at the base from septal myocardial capture [7]. If interventricular delay is genuinely suspected, electrode configuration is a cheap thing to test before invoking LOT-CRT [8].

On positional loss of capture

Part one called this documented; that was too strong. What is actually characterized in the literature is microdislodgement and progressive threshold rise, not a postural entity. QRS transition during threshold testing is the reference standard for confirming direct conduction system capture, but its detectability over follow-up is inconsistent across series — one found selective transition in only about 31% of patients at three months, while another reported transition preserved in 88% at one year among those who had it intraoperatively [9]. And a case of progressive threshold rise over the year after implant, ultimately rendering the left bundle branch area lead unusable with mid-septal fibrosis on cardiac MRI, illustrates that loss of conduction system capture during follow-up does occur [10].

A positional 12-lead is still cheap and reasonable — but frame it as ruling out lead-related change rather than looking for a described postural phenomenon.

The gap, now with a name

FOCUS-Right (NCT06601322, Virginia Commonwealth University) is the trial built for this: simultaneous exercise stress cardiac MRI and cardiopulmonary exercise testing in patients with existing LBBAP devices, measuring right ventricular function, left ventricular function and exercise capacity across different pacing settings at rest and during low-intensity exercise — explicitly because the impact of LBBAP on the right ventricle has not been addressed [11]. Still feasibility-stage, and exercise rather than posture, but it is the first design that would detect what is described here.

Practical upshot

Two things change relative to the mechanistic answer in part one. Rate-adaptive AV delay behavior belongs higher on the interrogation list than posture-specific AV assessment. And a normal resting optimization echo deserves less weight as reassurance — particularly in a stiff, high-filling-pressure ventricle, where the resting numbers are the least informative about what happens under load.

References