Ventricular–Arterial Coupling in Heart Failure with Preserved Ejection Fraction: A Bibliometric Review and Clinical Implications

Heart failure (HF) affects approximately 55 million people worldwide by 2021 – more than doubling since 1990.¹ Its rising prevalence, high mortality, and frequent hospitalisations demand better prevention and therapy. Population ageing and arterial stiffening disrupt ventricular–arterial coupling (VAC) before overt HF.^2–4 Early compensation via higher left ventricular (LV) end-systolic elastance (Ees) may offset arterial load, but this reserve wanes, especially in older women, leading to elevated filling pressures and HF with preserved ejection fraction (HFpEF).^5,6 VAC, commonly expressed as the ratio of effective arterial elastance (EA) to LV Ees, links arterial load with LV performance and provides a unifying lens from subclinical dysfunction to clinical risk.^7,8

Accordingly, this review combines bibliometric mapping with a thematic appraisal to summarise measurement approaches and clinical applications, and to set practical priorities for standardisation and validation.

Methods

Data Source and Search Strategy

All bibliometric data were retrieved from the Web of Science Core Collection (WoSCC, Clarivate Analytics). A comprehensive search was conducted on 1 March 2025, covering publications from database inception through December 2024. The full Boolean query and synonym expansions are provided in Supplementary File S1; in the main text we use the standard term “ventricular–arterial coupling (VAC)” for consistency. Two reviewers (TL and YL) independently screened titles/abstracts and full texts; disagreements were resolved by a third investigator (BZ). Conference abstracts, editorials, and non-peer-reviewed content were excluded; non-English and animal studies were retained to preserve the historical breadth (potential bias addressed in Limitations). A PRISMA flow diagram (Figure 1 ) summarises the selection: 4,235 records were identified; 1,050 studies were included. Language was recorded and is reported at screening, inclusion, and exclusion (PRISMA; Supplementary Table 1 ). Open access status was extracted from the WoSCC Open Access Designations field (gold/hybrid/green/bronze/closed). In the final inclusion set (n=1,050), we compared age-normalised annual citation rates by open-access (OA) category (Supplementary Table 2). Age normalisation was defined as times cited, all databases divided by years since publication (cut-off 1 March 2025); for 2025 publications, the denominator was set to 1 year. All citation counts, co-citation links, and network metrics were computed within WoSCC; no cross-database normalisation (e.g. Scopus/PubMed harmonisation) was performed.

Definitions

VAC is expressed as Ea/Ees, where effective EA and LV Ees are derived from pressure–volume relations (invasive pressure–volume loop or validated single-beat methods). Right ventricular–pulmonary arterial (RV–PA) coupling is commonly approximated by the tricuspid annular plane systolic excursion/pulmonary artery systolic pressure (TAPSE/PASP) ratio; the mean pulmonary artery pressure–cardiac output slope (mPAP/CO slope) reflects pulmonary vascular/exercise reserve. Global longitudinal strain (GLS) and pulse wave velocity (PWV) index LV deformation and large-artery stiffness, respectively. Left atrial reservoir strain (LASr) and left atrial stiffness index (LASI) = (E/e′)/LASr quantify atrial reservoir function and stiffness. Epicardial adipose tissue (EAT) denotes pericardial visceral fat implicated in myocardial remodelling and altered ventricular mechanics. Artificial intelligence (AI) and machine learning (ML) refer to data-driven approaches used here for measurement standardisation and prognostic modelling.

Data Extraction and Software Tools

Records (full metadata and cited references) were exported in plain text. Bibliometric analyses used VOSviewer 1.6.20 (co-authorship networks, country collaborations, keyword co-occurrence), CiteSpace 6.1.R6 (citation burst detection, co-citation analysis, clustering), OriginPro 2025 (annual publication and citation trends), and Bibliometrix (R, 4.1.0; thematic evolution, three-field plots). Scientometric metrics were prespecified: h-index (cumulative influence), g-index (highly cited output), and total link strength (TLS) from VOSviewer (collaboration intensity), following established practice.^9,10 Parameter settings for CiteSpace included 1-year slices, Top 50 items per slice, cosine similarity and pathfinder pruning to optimise clarity. Risk of bias used a Quality in Prognostic Studies-based traffic-light chart (Supplementary Figures 1 and 2 and Supplementary Table 3). All analyses followed established protocols.

Data Preprocessing and Standardisation

Institution names, author names, and keywords were standardised to minimise duplication and ensure consistency across data sets. Geocoding of institution affiliations was conducted using the Google Geocoding API to consolidate multiple campuses under unified university identifiers. Author affiliations, institutional collaborations, and keyword variations were normalised according to WoS thesaurus guidelines to improve the robustness. Two researchers processed the data independently, resolving discrepancies by consensus.

Analytical Framework

The study analysed publication output, citation trends, author productivity, country and institution collaborations, core and co-cited journals, keyword clustering, thematic evolution, citation bursts, and highly cited articles. For authors, we computed h-index, g-index and TLS; for countries/institutions/journals, we summarised TLS to quantify collaboration centrality. Co-citation networks and timeline visualisations traced the historical development and emerging themes in VAC.

Results

Annual Publication Outputs and Trends

VAC-related publications have increased steadily since 1983 (Figure 2). Mann–Kendall analysis confirmed a highly significant upward trend (Kendall tau, 0.992, p<0.001). Early studies (1983–1990, <5 papers/year) primarily explored fundamental physiological concepts. Interest markedly accelerated after 2006, coinciding with rising attention to HFpEF and advances in non-invasive imaging (echocardiography, MRI). Output peaked in 2024.

Global Contributions and Institutional Collaborations

The US leads VAC research (375 publications; 11,114 citations; TLS 93,771), followed by Italy (145; 1,944; TLS 50,974) and Japan (84; 1,213). China (72; 1,235) has recently expanded contributions, notably in arterial stiffness and diastolic dysfunction. Belgium (63; 2,419) and France (59; 2,071) have influenced pulmonary hypertension and HFpEF substantially. International collaboration intensified across North America–Europe (Supplementary Figures 3–5 ). At the institutional level, Mayo Clinic (26; 1,622; TLS 10,555) and University of Pennsylvania (24; 1,061; TLS 10,365) are prominent hubs; University of Colorado and University of Pisa (21 each; TLS 6,776 and 9,668) emphasise HFpEF clinical implications. Erasme University Hospital (14; 1,237) has been highly influential in pulmonary vascular coupling (Supplementary Figures 6–7 ). Publication count correlated with TLS across countries (Spearman r=0.593, p<0.001), underscoring the link between output and collaborative depth.

Influential Journals and Key Author Networks

Leading journals included European Heart Journal (81 publications; 547 citations), Circulation (75; 944) and Journal of the American College of Cardiology (57; 2,553), emphasising translational impact. Specialty titles, such as European Journal of Heart Failure (47; 638) and American Journal of Physiology–Heart and Circulatory Physiology (25; 1,938), bridge fundamental and clinical application. Journal co-citation analysis demonstrated growing interdisciplinarity (Supplementary Figures 8 and 9).

Author-network analysis identified Barry A Borlaug (16 publications; 1,908 citations; TLS 5,961; h=13; g=18) and Naomi C Chesler (15; 324; TLS 4,441; h=13; g=21) as central figures. Borlaug’s research significantly advanced understanding of cardiovascular reserve dysfunction in HFpEF, while Levine’s contributions were instrumental in elucidating exercise-related haemodynamic responses. Other influential authors include Naomi C Chesler, Julio A Chirinos, and Robert Naeije, who notably enhanced the understanding of RV–PA interactions, a rapidly expanding subfield within VAC research (Supplementary Figures 10 and 11 ).

Keywords: Analysis and Research Frontiers

Keyword co-occurrence and thematic clustering delineated a transition from mechanics/afterload (1990–2005) to clinical applications (2006–2015), and, most recently (2016–2024), to clinically actionable concepts: HFpEF, RV–PA coupling, and echocardiography (Supplementary Figures 12–14). Citation-burst analysis (Supplementary Figure 15 ) revealed rising interest in exercise interventions (burst 7.8; 2018–2023), AI (6.5; 2019–2023), and preserved ejection fraction (EF) (9.2; 2017–2023), indicating a shift toward personalised diagnostics and predictive analytics. Figures 3 and 4 visualise venue overlays and thematic timelines.

Highly Cited Literature and Landmark Studies

Co-citation network analysis (Supplementary Figure 16) identified critical landmark studies shaping current VAC knowledge: Borlaug et al. (2010; 545 citations) on cardiovascular reserve dysfunction in HFpEF; Reymond et al. (2009; 481) on arterial modelling; Tello et al. advancing RV–PA coupling assessments. Recent influential studies by Kobayashi et al. and Pugliese et al. further reinforced the clinical significance of RV–PA coupling indices (TAPSE/PASP) across HF populations.^11–13 Four primary clusters emerged (clinical haemodynamics, pulmonary hypertension, computational modelling, and vascular–ventricular dynamics) mirroring keyword-based themes. Temporal analysis (Supplementary Figure 17 ) shows a shift from early mechanical work (pre-2000, depicted in blue) to translational/clinical focus (2015 onward, outlined in red-yellow). The citation landscape is skewed, with a median of 60 citations (IQR 39–107), underscoring the outsized impact of these pivotal works.

Discussion

Bibliometric Insights into the Evolution of Ventricular–Arterial Coupling Research

Synthesis of the bibliometric mapping (search on 1 March 2025; coverage through December 2024) shows accelerated growth since 2006 across four dominant clusters-clinical haemodynamics, RV–PA coupling, arterial stiffness, and imaging–aligning with expanding interest across HF subtypes.^14,15 These patterns coincide with the rising prognostic use of TAPSE/PASP and the complementary, but methodologically heterogeneous, behaviour of EA/Ees versus PWV/GLS indicated in population cohorts.^13,16–20 To avoid duplication with Results, detailed geographic/author metrics and co-citation networks are interpreted there and in the Supplement; here we focus on clinical synthesis across LA–LV–RV–PA and on priorities for standardisation and prospective validation, building on translational advances in HFpEF and pulmonary vascular physiology.^21–25 To anchor this clinical synthesis, Table 1 contrasts commonly used VAC metrics and echocardiographic biomarkers across screening, mechanistic interpretation, and prognostic utility in HFpEF, with summarised representative evidence.^{9–11,13,14,17–19,21–41}

Risk of Bias

As summarised in Table 1 and detailed in Supplementary Figures 1 and 2 and Supplementary Table 3, high-risk ratings clustered in D8 (confounding control) and D10 (response rate/data-collection completeness) more often graded as “some concerns/unclear” and occasional high-risk flags in D3–D4 (exposure window and sampling continuity). Contributing factors included:

1. non-standardised acquisition/derivation of GLS, PWV, LASr/LASI and TAPSE/PASP without prespecified protocols;
2. inconsistent outcome definitions and follow-up windows across cohorts; and
3. analytical practices prone to optimism–post-hoc thresholding, univariable screening, low events-per-variable, limited handling of missing data (D9), and sparse calibration or external validation.

These limitations may inflate or attenuate the independent associations of VAC-related metrics. Therefore, we advocate harmonised measurement workflows, preregistered analysis plans, transparent reporting of discrimination and calibration, and external replication, consistent with contemporary risk-of-bias guidance for prognostic factor research.⁴² Priorities are harmonised protocols (Ea/Ees, PWV/GLS, LASr/LASI, TAPSE/PASP, mPAP/CO), multicentre prospective validation (including exercise-based strategies), and adoption of AI and ML for cross-vendor calibration, particularly for the RV–PA and LA phenotypes that carry the strongest clinical signal. We examine chamber-specific implications in the following section.

Ventricular–Arterial Coupling as a Central Modulator of Cardiac Function during Heart Failure Progression

Impact of Ventricular–Arterial Coupling on Left Atrial Function

Impaired VAC in HFpEF critically impacts left atrial (LA) function and accelerates disease progression. Abnormal VAC, characterised by increased arterial stiffness and compromised LV compliance, results in elevated haemodynamic stress and chronically raised filling pressures in the LA.^26,31 Initially, the LA compensates with augmented contractility and dilation, but chronic pressure overload triggers maladaptive fibrotic remodelling and loss of compliance, eroding its reservoir and booster-pump functions.^43–45 Recent 3D modelling indicates that LA roof dilation can precede global atrial enlargement in ‘masked’ HFpEF (a form in which HFpEF criteria become evident only under stress/exercise, yet remain borderline at rest), revealing an early structural phenotype of atrial myopathy.⁴⁶ These structural changes further diminish both reservoir and booster-pump roles. Consequently, LA dysfunction often predates clinically overt HFpEF and aggravates disease progression by impairing pulmonary haemodynamics and promoting AF, thus worsening outcomes.^32,33,47,48

Moreover, LA disease and possible reverse remodelling across different HF stages have emerged as a key target for prevention and therapy.³⁴ Emerging imaging biomarkers capture these early impairments: LASr and the LASI are sensitive to subclinical atrial dysfunction and predict adverse outcomes in HFpEF.^32,35 Reductions in LASr correlate strongly with diminished exercise tolerance, underscoring its value as an early indicator before symptom onset.^35,46 LASI, by integrating atrial mechanics with LV filling pressures, provides a more nuanced reflection of LA structural and functional decline.^49,50 However, measurements of LASr and LASI can vary across imaging platforms, indicating the need for consistent technical protocols. Recent cohort-imaging studies have also identified an HFpEF phenotype with severe LA dysfunction disproportionate to ventricular abnormalities, implying a primary atrial myopathy driven by fibrotic, inflammatory processes.^43,51 Interventional and review data further underscore the modifiability of this process: Edelmann et al. reported that 12 months of spironolactone therapy in early HFpEF reduced the LA volume index and improved LASr. Likewise, Inciardi et al. found that pharmacologic and lifestyle interventions across the HF spectrum can induce LA reverse remodelling with associated outcome improvements.^52,34 Collectively, these findings support the concept that early VAC-targeted strategies, such as antihypertensive, antifibrotic, and rhythm-control interventions, may arrest or even reverse maladaptive atrial remodelling if instituted before overt HFpEF develops.

Exercise-stress imaging can unmask latent atrial dysfunction, offering additional prognostic insight in early HFpEF.³⁶ Despite strong evidence for a bidirectional VAC–LA interplay, key questions remain: Can early restoration of VAC definitively preserve LA mechanics, or might primary atrial interventions forestall VAC deterioration? Prospective studies are needed to evaluate integrated strategies that simultaneously target ventricular–arterial and LA–LV (atrioventricular, AV) coupling.

Left Ventricular Dysfunction in Heart Failure Progression: Role of Ventricular–Arterial Coupling

By the time HF becomes clinically manifest, many patients have experienced years of asymptomatic LV dysfunction – sometimes termed Stage B or pre-HF, marked by structural or functional abnormalities (e.g. LV hypertrophy or diastolic dysfunction) without overt symptoms.² In hypertensive populations, subclinical LV diastolic dysfunction (LVDD) emerges early, affecting approximately 30–60% (average ~45%) of long-term hypertensive patients, as shown in large community-based studies with robust Doppler and strain imaging.^3,53 This subclinical dysfunction is clinically significant: impaired relaxation and elevated filling pressures lead to exercise intolerance and set the stage for HFpEF.^3,54 Pathophysiologically, chronic pressure overload initiates LV remodelling characterised by cardiomyocyte hypertrophy and interstitial fibrosis, increasing chamber stiffness and impairing relaxation.^27,55 Beyond structural hypertrophy, chronic wall stress and endothelial injury provoke coronary microvascular dysfunction (CMD) and rarefaction, causing subendocardial ischaemia that further promotes fibrosis; consequently, ventricular compliance declines and diastolic filling pressures rise. Persistent remodelling and fibrosis can erode systolic reserve, potentially transitioning from HFpEF to overt systolic HF (HFrEF).^28,53

Concurrent arterial stiffening commonly parallels LV remodelling, amplifying afterload. Increased arterial stiffness leads to sharper systolic pressure rises and premature return of wave reflections, elevating late-systolic load.⁵⁶ This interplay disrupts VAC, elevates mid-wall myocardial stress, and impairs ventricular relaxation.^4,56 Indeed, combined ventricular systolic and arterial stiffening is a hallmark of HFpEF, associated with limited stroke-volume reserve, elevated natriuretic peptides, and impaired functional capacity.^26,57 Advanced imaging consistently demonstrates subtle systolic dysfunction, such as abnormal global longitudinal strain (GLS) and delayed apical untwisting, coexisting with diastolic dysfunction, emphasising that ‘preserved’ EF often masks clinically relevant contractile impairments.^3,29,54 These subtle deficits have substantial prognostic implications, independently predicting adverse outcomes even in early-stage disease.³⁰

Not all hypertensive patterns confer equivalent risks; notably, isolated nocturnal hypertension accelerates arterial stiffening and LV remodelling, underscoring the need for comprehensive 24-hour blood-pressure management.⁵³ Adding to this complexity, large-scale imaging studies have shown myocardial perfusion reserve (MPR) <2.0 in ~70% of HFpEF patients, correlating with more severe LV remodelling and worse clinical outcomes.⁵⁸ Future studies should clarify whether joint targeting of arterial stiffness and CMD might provide synergistic benefits in preventing HFpEF progression.

Right Ventricular Dysfunction and Ventricular–Arterial Coupling in Heart Failure

Although much emphasis in HF has historically centred on left-sided mechanisms, there is growing recognition that the RV also plays a pivotal role. In HFpEF, abnormal VAC and chronic LVDD eventually raise LA pressure, with backward transmission into the pulmonary vasculature.^59,60 Over time, these haemodynamic shifts provoke pulmonary hypertension and increase RV afterload, ultimately leading to impairment of RV–PA coupling (i.e. RV–PA uncoupling) and subclinical right-sided dysfunction.^61,62 Although classic teaching posits that the RV remains relatively uncompromised in early HFpEF, some data challenge this view: even among higher-LVEF subgroups, distinctive LV–arterial and RV–PA coupling alterations can be detected, indicating that RV function may be less robust than previously assumed once pulmonary pressures begin to rise.^63,64 Moreover, subtle RV dysfunction, such as reduced RV global longitudinal strain (RV GLS), significantly impacts prognosis, nearly doubling the risk of HF hospitalisation and mortality.⁶⁵

Interventricular interactions further complicate this picture: elevated LV end-diastolic pressure can bulge the interventricular septum into the RV, impeding its filling and contractile efficiency.⁶⁶ This altered septal motion reduces biventricular output, emphasising the interdependent nature of ventricular mechanics in HF.⁶⁷ Advanced echocardiography and cardiac magnetic resonance (CMR) increasingly identify early RV systolic dysfunction preceding overt right-sided failure, particularly among patients with diabetes or uncontrolled hypertension.^68,69 Additionally, borderline pulmonary pressures and gas exchange abnormalities may impair RV strain before overt LV abnormalities become evident.^59,60 Impaired RV–PA coupling – quantified by TAPSE/PASP ratio (tricuspid annular plane systolic excursion / systolic pulmonary artery pressure) – has emerged as a powerful prognostic marker. In HFpEF patients, a TAPSE/PASP <0.36 predicts higher mortality and identifies combined precapillary and postcapillary pulmonary hypertension.³⁸

A recent meta-analysis confirmed that every 0.1 unit decrease in TAPSE/PASP ratio increased adverse event risk by 17%, reinforcing its clinical utility.³⁷ Kobayashi et al. further demonstrated that impaired TAPSE/PASP predicted severe pulmonary congestion and impaired decongestive response in acutely decompensated HF, emphasising its value across acute clinical scenarios. In stable HF populations, Pugliese et al. linked impaired TAPSE/PASP with exercise intolerance, supporting its central role in functional limitation across HF subtypes, particularly in HFpEF. Furthermore, a separate matched analysis comparing HFpEF and HFrEF cohorts demonstrated significantly worse RV–PA coupling in HFpEF, contributing to greater exercise-induced pulmonary hypertension and haemodynamic impairment.^10,11,70

Recent data from De Biase et al. indicated that patients with aortic stenosis and preserved EF exhibit RV–PA uncoupling comparable to HFpEF, suggesting a shared haemodynamic mechanism underlying exertional limitations in these phenotypes.^39,40 Additionally, the mPAP/CO slope has emerged as a valuable prognostic marker for RV–PA uncoupling across cardiovascular conditions, including HFpEF and aortic stenosis, providing incremental risk stratification beyond conventional measures such as peak pulmonary artery pressure.^40,41 Collectively, these findings underscore that multifactorial RV dysfunction – resulting from pulmonary pressure overload, interventricular dependence, and intrinsic myocardial impairment – critically influences outcomes in HFpEF. Given these robust findings, TAPSE/PASP and mPAP/CO slope should be incorporated as key components in both diagnostic and prognostic assessments in clinical practice and future HFpEF studies. Integrative RV evaluation combining deformation indices with coupling metrics further refines clinical risk assessment.⁷¹

Thus, even in ostensibly ‘preserved’ HF states, right-sided dysfunction confers substantial prognostic significance. Future research should prioritise rigorous validation of these emerging RV–PA coupling metrics, including prospective clinical trials assessing interventions aimed at improving RV–PA coupling and reducing RV afterload, enhancing patient outcomes. Incorporating multiparametric RV imaging and dynamic exercise-stress testing into routine practice might further enhance early detection, guide precision interventions, and improve patient outcomes.

Integrated Multichamber and Extracardiac Perspective on VAC Dysfunction in HFpEF

EAT has increasingly been recognised as a pathophysiological amplifier of HFpEF impairing myocardial stiffness and VAC through inflammatory and fibrotic mechanisms.^72,73 EAT burden is consistently higher in HFpEF than in matched controls and correlates with a more adverse haemodynamic and inflammatory profile.^72,74 In a prospective exercise-right-heart-catheterisation study of HFpEF, patients in the highest EAT tertile demonstrated ~30% higher peak pulmonary-capillary wedge pressure and ~25% lower stroke-volume reserve versus the lowest tertile, directly linking EAT to blunted exertional haemodynamics.⁷⁵ In hypertension at risk for HFpEF, excess EAT tracks with deteriorating VAC, compounding the effects of arterial stiffening and LV diastolic dysfunction on filling pressures.^76,77 EAT-derived cytokines activate profibrotic transforming growth factor-β/Smad signalling and oxidative stress pathways, accelerating myocardial fibrosis and microvascular rarefaction.⁷³ Furthermore, multichamber interactions involving LA, LV, and RV abnormalities typically occur concurrently, indicating an integrated multichamber pathophysiology rather than isolated chamber dysfunction (Figure 5 ).^{61,62,76,78–80} Thus, a whole-heart perspective encompassing EAT and integrated therapeutic strategies targeting the entire heart-VAC axis, including extracardiac tissues, to optimise HFpEF prognosis and treatment effectiveness.

Translational Implications: Standardisation and Artificial Intelligence

Building on the whole-heart and extracardiac perspective, clinical translation of VAC critically depends on reproducible, vendor-agnostic measurements across LA–LV–RV–PA. ML can help operationalise this by:

1. learning domain-robust representations and enabling cross-vendor calibration to harmonise GLS/LASr and related surrogates;
2. integrating echocardiographic waveforms, arterial tonometry and ECG to estimate Ea/Ees and infer RV–PA coupling surrogates (e.g. automated TAPSE/PASP from cine loops);
3. providing automated quality control, segmentation and imputation; and
4. building multimodal prognostic models for risk enrichment in HFpEF.

To ensure credibility and reproducibility, development and evaluation should follow reporting standards for AI prediction models and AI trials (e.g. TRIPOD-AI/TRIPOD-ML for model reporting and CONSORT-AI/SPIRIT-AI for interventional evaluation), with emphasis on external validation across vendors/sites, temporal-drift monitoring and model updating; federated learning can support multicentre standardisation without sharing raw data.^81–83

Limitations

This study has several limitations. First, our analysis used only WoSCC. Although we did not exclude non-English items a priori, non-English records accounted for 1.21% at screening (51/4,235), 0.00% among the 1,050 included studies, and 1.60% among excluded records (51/3,185) (PRISMA; Supplementary Table 1 ). This pattern suggests minimal language bias within the analysed corpus, yet an upstream indexing bias cannot be excluded because WoSCC predominantly indexes English-language journals. Non-WoSCC citation dynamics were not captured and may bias comparative influence estimates. Future work should integrate multiple databases (e.g. Scopus, Embase, regional indices) and, where feasible, machine-translation-assisted screening to improve coverage.

Second, in the final corpus, OA articles constituted 40.0% (420/1,050) and exhibited a higher age-normalised annual citation rate than closed-access articles (median 2.00 [IQR 0.66–4.33] versus 0.00 [0.00–1.09] citations per years; Supplementary Table 2). Because the closed-group median was 0, a relative median increase (Δ%) is not defined; as a sensitivity analysis, the mean difference corresponded to +284.0% (4.22 versus 1.10 citations/year). Although age was accounted for, residual confounding (e.g. journal impact, document types) may remain.

Third, we included both human and animal studies, without restriction to adult human populations, which introduces heterogeneity and limits direct clinical applicability.

Fourth, the search ended in December 2024 (conducted 1 March 2025), meaning recent or unindexed studies were not captured.

Fifth, despite standardisation, residual inaccuracies in author names and affiliations may remain. Citation metrics inherently favour older publications; we mitigated this by age-normalising citations (citations per year) and emphasising thematic patterns over raw counts. Sixth, substantial methodological heterogeneity precluded statistical meta-analysis; our narrative synthesis thus provides weaker evidence strength. Finally, variability in VAC assessment methods (e.g. Ea/Ees versus PWV/GLS; LASr/LASI; TAPSE/PASP; mPAP/CO slope) underscores the need for harmonised protocols, prospective validation (including exercise-based strategies), and cross-vendor calibration to support clinical translation.

Conclusion

Research into VAC has progressed from a mechanistic concept toward a clinically relevant framework, though significant challenges remain. Methodological heterogeneity, fragmented mechanistic insights, and limited therapeutic validation are key obstacles. To overcome these issues, future studies should focus on developing standardised protocols for measuring key VAC indices (Ea/Ees, PWV/GLS, TAPSE/PASP, and mPAP/CO slope) and conducting longitudinal mechanistic investigations and rigorous clinical trials. Exercise interventions aimed at enhancing VAC efficiency represent a promising yet underexplored area for future research. An integrated, multichamber approach that considers the interplay among the LA, LV, and RV is essential for understanding the complexity of HFpEF. Advanced imaging modalities, computational modelling, and AI analytics could further refine VAC metrics, raising their clinical applicability. The prognostic potential of RV–PA coupling indices such as TAPSE/PASP and mPAP/CO slope also warrants systematic validation in prospective studies. Ultimately, sustained interdisciplinary collaboration among physiologists, imaging specialists, computational scientists, and clinicians is essential to effectively translate VAC science into meaningful improvements in HF prevention and management. Echoing Chirinos’ caution, it is not yet “time to close the loop and catch the wave” of VAC in HF foundational questions on pulse-wave dynamics and clinical integration still await resolution.⁸⁴ While significant progress has been made, foundational questions regarding pulse-wave dynamics, methodological standardisation, and clinical integration remain open, underscoring the ongoing need for rigorous, well-designed research.

Click here to view Supplementary Material.