The core hemodynamic case
The atrial "kick" supplies roughly 10-30% of LV filling normally, but in a stiff, non-compliant ventricle (exactly the HFpEF substrate) that contribution can rise to ~40%, because passive early filling is already curtailed. So AF is doubly damaging in HFpEF: it abolishes the booster contribution the stiff ventricle most depends on, and it shortens diastolic filling time. This is why AF is tolerated far worse in HFpEF than in HFrEF (where the dilated ventricle fills passively and the dominant AF harm is rate-related). AF carries a worse prognosis in HFpEF than HFrEF, and baseline AF prevalence in HFpEF is ~42%.
The strongest direct evidence that restoring the atrial contribution helps comes from the small dedicated ablation RCTs in AF-HFpEF (Aldaas, JACC Heart Failure 2023; the STALL program). Restoring sinus rhythm cut peak exercise PCWP (~30→25 mmHg), raised resting and peak cardiac output, improved peak VO2, and in half of ablated patients eliminated the exercise-hemodynamic HFpEF signature. These are the cleanest invasive demonstrations that returning organized atrial mechanics improves output.
-
Restoring sinus rhythm is the only established way to recover native atrial contraction, and the recovery has a characteristic arc:
Immediately post-cardioversion there is atrial stunning (electromechanical dissociation, the basis for post-cardioversion stroke risk), deeper and slower to recover the longer AF persisted.
Booster-pump function and strain rate begin recovering within ~4 weeks; LA volume reduction and reservoir-strain reverse remodeling progress over 3-12 months in those who stay in sinus rhythm.
This reverse remodeling also reduces atrial functional MR and TR (Soulat-Dufour, JACC 2022 showed serial 3D-echo reduction in both valves; ~80% of atrial functional MR became ≤mild after rhythm control in older series), via recovery of annular dynamics, independent of LA volume change.
On outcomes, the supporting evidence is convergent but not yet definitive for HFpEF specifically: the EAST-AFNET 4 HF subgroup (56% HFpEF) showed early rhythm control reduced CV death/stroke/HF hospitalization, benefit consistent across HF phenotypes, and a mediation analysis attributed most of the benefit to attaining sinus rhythm itself rather than the specific tool. The CABANA HFpEF post-hoc found ablation benefit was actually larger in high-likelihood HFpEF (HR 0.82 for CV hospitalization/death; mortality signal in the echo-defined subset). Meta-analyses favor ablation/rhythm control. Counterpoints: BENEFIT-HFpEF (persistent AF) was neutral on peak VO2, and an editorial flagged that H2FPEF-based classification is confounded in AF. The definitive dedicated trial, CABA-HFPEF-DZHK27 (~1,548 patients), reports primary completion around July 2026; CASTLE-HFpEF and STABLE-SR IV are also running.
-
This is the most important nuance for your specific framing: no approved or clinical-stage device exists whose explicit purpose is to restore native atrial contractile force. The adjacent strategies break down as:
Accelerated/physiologic atrial pacing — myPACE (RCT, n=107) improved QoL, NT-proBNP, and reduced device-detected AF with a personalized higher rate; but RAPID-HF was negative for exercise output. These work by shortening diastole / lowering filling pressure, not by augmenting atrial force, and the results conflict. PACE-HFpEF is ongoing.
Atrial resynchronization — Bachmann bundle pacing (including a first-in-human leadless device in 2025) and the older left-atrial-pacing-for-atrial-dyssynchrony pilot aim to restore timing/coordination of atrial contraction; early/experimental.
Ablate-and-pace + conduction system pacing (APAF-CRT, PACE-FIB, His/LBBAP) — controls rate and preserves ventricular activation but, by design, does not restore the atrial kick because AF persists.
Interatrial shunts (Corvia, V-Wave) — these decompress the left atrium; they don't restore contraction. Both pivotal sham-controlled trials failed in HFpEF, and RELIEVE-HF showed a net-harm signal in HFpEF specifically (win ratio 0.70, p=0.001).
Left atrial assist devices / atrial pumps (Cleveland Clinic LAAD; Pumpinheart "PReduction") — these mechanically bypass or augment atrial pumping and are the closest engineering analog to "restoring the atrial contribution," but remain preclinical (bench/animal only).
-
SGLT2 inhibitors are now Class I in HFmrEF/HFpEF (EMPEROR-Preserved, DELIVER) and the 2024 ESC AF guideline recommends them for HF+AF regardless of EF; they modestly reduce AF burden and post-ablation recurrence. Finerenone (FINEARTS-HF) is newly positive in HFmrEF/HFpEF, with benefit unmodified by AF.
Bottom line: restoring the atrial rhythm (and with it, native contraction) is the only validated route to recovering atrial contribution to output today, with the best hemodynamic proof in small AF-HFpEF ablation RCTs and outcome support from subgroup/meta-analytic data pending CABA-HFPEF (2026). True device-based atrial contractile restoration is still preclinical, and the success ceiling is set by how much atrial myopathy/fibrosis is already established, which argues for early intervention
-