Prominent V-Waves Augment Right Ventricular Afterload Independent of the Lung Vessels in Heart Failure: Implications for Clinical Assessment of Pulmonary Vascular Disease

Abstract

Background:

Left atrial (LA) stiffening augments the LA V-wave, pressurizing the pulmonary artery (PA) out of proportion to left ventricular end-diastolic pressure. We characterized the pathophysiologic, diagnostic, and clinical implications of LA stiffening, as defined by prominent V-waves, in HFpEF.

Methods:

In this retrospective cohort study, patients with HFpEF and less than moderate mitral regurgitation at rest underwent invasive hemodynamic exercise testing. Patients with heightened operant LA stiffening were defined by PA wedge pressure (PAWP) V-wave exceeding mean PAWP by ≥10 mmHg. A previously-used formula for V-wave-corrected PAWP was validated against the PAWP-waveform pressure-time-area to explore impact of the V-wave on transpulmonary pressure gradients (TPG). Clinical, echocardiographic, hemodynamic, and outcome data were compared in patients with and without prominent V-waves.

Results:

Compared with patients without prominent V-waves (n=844, 79%), those with prominent V-waves (n=225, 21%) were older and more likely to be women, with more severe LA myopathy, evidenced by greater LA volume index (42.2±15.0 vs. 32.5±13.0 mL/m²; P<0.001), lower LA reservoir strain (16.4±7.5 vs 23.2±8.3%; P<0.001), more atrial fibrillation (61.3% vs. 22.9%; P<0.001), and increased operant LA reservoir stiffness (0.59±0.25 vs 0.23±0.24 mmHg/mL; P<0.001). Patients with prominent V-waves had markedly higher PA systolic pressure than those without (54±16 vs. 38±12 mmHg; P<0.001) despite no difference in end-diastolic PAWP-PA pressure gradient (3±5 vs. 3±4 mmHg; P=0.950). Incorporating the V-wave into PAWP improved correlation between directly measured PAWP pressure-time-integral compared with mid-A wave PAWP (R²=0.96 vs R²=0.86, P<0.001 for comparison). In patients with prominent V-waves, the prevalence of combined pre- and post-capillary pulmonary hypertension decreased from 60.2% to 29.3% after V-wave correction for TPG. Despite similar corrected pulmonary vascular resistances in patients with and without prominent V-waves, apparent PA compliance was much lower in patients with prominent V-waves (3.03±1.56 vs. 4.90±2.27 mL/mmHg; P<0.001), augmenting right ventricular (RV) afterload, and these changes were coupled with more severe RV dysfunction, tricuspid regurgitation, blunted cardiopulmonary exercise reserve, and increased risk of HF hospitalization or death in patients with vs without prominent V-waves (adjusted HR: 2.09; 95% CI: 1.53–2.86; P<0.001).

Conclusions:

Over 1 in 5 patients with HFpEF display prominent V-waves in excess of end-diastolic PAWP. This augments RV afterload due to backward transmission of the V-wave, contributing to right heart failure that is independent of the pulmonary vasculature. Failure to account for the V-wave in hemodynamic assessment overestimates the severity of pulmonary vascular disease present.

Introduction

It has been recognized for decades that some patients develop excessive left atrial (LA) stiffening following surgical or catheter ablation procedures, often termed the “stiff LA syndrome”^1–4. It is also known that patients with heart failure (HF) with preserved ejection fraction (HFpEF) frequently have LA remodeling, dysfunction, and rhythm disturbance (collectively termed LA myopathy), which is generally not iatrogenic, but contributes to pathophysiology and risk for adverse outcome^5–8. Assessment of LA myopathy in HFpEF has traditionally relied upon noninvasive imaging,^5,6,8,9 without direct invasive hemodynamic measurements.

The amplitude of the V-wave in the pulmonary artery (PA) wedge pressure (PAWP) tracing provides an estimate of operant LA stiffness: when the LA is operating on the steep portion of its diastolic pressure-volume relationship (DPVR), as with LA and pulmonary venous stiffening, a prominent V-wave (≥10 mmHg over mean PAWP^3,10) is observed. Retrograde transmission of this V-wave may augment pulmonary artery pressure out of proportion to left ventricular diastolic pressures, increasing right ventricular afterload. Little is known regarding the prevalence or clinical implications of such operant LA stiffening in HFpEF.

Here, we sought to identify and comprehensively characterize operant LA stiffening using a surrogate measure of prominent V-wave out of proportion to the end diastolic PAWP, in large cohort of patients with HFpEF undergoing high fidelity invasive hemodynamic exercise testing. We evaluated clinical, echocardiographic, and hemodynamic correlations of LA stiffness along with the potential impact of LA stiffening on the optimal characterization of pulmonary vascular loading, as well as HFpEF pathophysiology and prognosis.

Methods

Study Population

Consecutive adults with exertional dyspnea and preserved left ventricular (LV) ejection fraction (LVEF≥50%) who underwent simultaneous invasive hemodynamic assessment and expired gas analysis (invasive cardiopulmonary exercise testing) at the Mayo Clinic (Rochester, MN) between February 2006 and July 2025 were identified. Those with significant (greater than mild) mitral regurgitation (MR) at rest and other significant left-sided valvular heart diseases (≥ severe aortic regurgitation and ≥ moderate stenosis), prior mitral valve replacement, repair or intervention, prior episode of HF with LVEF<50%, infiltrative, inflammatory or hypertrophic cardiomyopathy, high output HF, previous heart transplantation, unstable coronary artery disease, constrictive pericarditis, congenital shunt disease, pulmonary vein stenosis, or isolated precapillary pulmonary hypertension (defined as mean pulmonary artery [PA] pressure [PAP]>20 mm Hg, PAWP<15 mm Hg at rest and <25 mmHg during exercise, and pulmonary vascular resistance [PVR]>2 WU at rest) were excluded.

Patients with confirmed HFpEF (defined as PAWP≥15 mmHg at rest and/or ≥25 mmHg during exercise) were included in the analyses¹¹. Patients who did not meet criteria for HFpEF were classified as noncardiac dyspnea (NCD) and assessed for the prevalence of prominent V-wave. All patients provided consent for their medical records to be used for research purposes, and the study was approved by the Mayo Clinic Institutional Review Board.

Assessment of cardiac structure and function

Transthoracic echocardiography was performed prior to the invasive hemodynamic assessment, and, in a subset of patients, simultaneous with catheterization, interpreted by board-certified cardiologists with advanced training in echocardiography. Chamber dimensions, LV diastolic parameters and right ventricular (RV) function were assessed in accordance with standard guidelines^12–15. LV mass, LA volume and right atrial (RA) volume were indexed to body surface area^15–17. LV global longitudinal strain (LVGLS) and LA strain were calculated as the average of the 4-chamber and 2-chamber apical views^8,18,19. To accommodate the inclusion of patients with persistent atrial fibrillation (AF) and absent P waves, LA and RA strain was measured using the QRS complex as the fiducial point. In subjects where both views were unavailable, measurements were derived solely from the 4-chamber view. The myocardial reservoir function of the left atrium was evaluated by the LA reservoir strain^9,20.

Invasive Hemodynamic Assessment and Expired Gas Analysis

Simultaneous invasive hemodynamic assessments and expired gas analyses at rest and during supine cycle ergometer exercise were performed, as previously described^21–23. RHC was performed using a 7-Fr fluid-filled catheter containing a 2-Fr high-fidelity micromanometer for the measurement of right atrial pressure (RAP), PAP, and PAWP. Pressures were recorded at end-expiration, averaged over at least 3 beats, with transducers zeroed at the mid-thoracic level. Mean PAWP was measured at the mid-A wave in patients with sinus rhythm and at the QRS complex in patients with AF. The V-wave was measured as the peak PAWP during ventricular systole, occurring near the T-wave, regardless of the rhythm. A 4- to 6-Fr radial artery cannula was utilized for arterial blood pressure monitoring and blood gas sampling. Patients with HFpEF and heightened LA stiffening, suggested by a prominent V-wave out of proportion to end-diastolic PAWP were defined as those with V-wave amplitude ≥10 mmHg above the mean PAWP measured at mid A-wave at rest^3,10, and compared to those without prominent V-waves for the primary analysis. As detailed in the Supplemental Methods, LA reservoir compliance was calculated by dividing the increase in LA volume during the reservoir phase (LA maximal volume minus LA minimal volume measured by echocardiography) by the increase in LA pressure during the reservoir phase (increase in PAWP from the nadir of the X-descent to the peak of the V wave, Figure S1), and LA stiffness was defined as the inverse of LA compliance. The quotient of LA reservoir strain to the V-wave to X-nadir PAWP difference was also used as a surrogate measure of LA compliance (Supplemental Methods).

Arteriovenous oxygen (AVO₂) difference was directly measured as the difference between systemic arterial and PA oxygen contents. Breath-by-breath expired gas analysis (Med-Graphics) was performed during the RHC procedure to measure oxygen consumption (VO₂). Cardiac output (CO) was calculated using the direct Fick method (CO = VO₂/AVO₂ difference). PVR was calculated using the equation, PVR = (MAP – mean PAWP)/CO (Wood Unit [WU] or mmHg*min/L).

To account for the disproportionate downstream pressure elevation associated with a V-wave, we calculated a V-wave-corrected PAWP using a clinically-utilized “rule of thumb” that adds one-third of the V-wave excursion to the mid-A wave pressure (corrected PAWP = mean PAWP + [V-wave – mean PAWP]/3) (Figure 1). This value was then used to calculate a “corrected PVR” to better reflect the true transpulmonary pressure gradient (TPG) throughout a cardiac cycle. While this formula has been used to approximate the true downstream pressure in the setting of prominent V-waves, it has never been formally validated. Therefore, we also sought to validate this simple physiologic estimate against the actual mean PAWP measured as the area under the curve (AUC) of the pressure tracing from a single cardiac cycle, analyzed from the digitized pressure tracings.

Figure 1: Impact of a Prominent V-Wave on Transpulmonary Pressure Gradient and Pulmonary Vascular Resistance.

Invasive hemodynamic pressure tracings from two patients with heart failure with preserved ejection fraction (HFpEF) demonstrate the impact of a prominent V-wave pulmonary artery wedge pressure (PAWP, green line) on the calculation of the transpulmonary pressure gradient (TPG) and pulmonary vascular resistance (PVR). (A) In a HFpEF patient without prominent V-wave, the V-wave magnitude is modest. Consequently, the traditional TPG (cross-hatched area) closely approximates the V-wave-corrected TPG (solid yellow area), resulting in a relatively minimal change to the calculated PVR. (B) Conversely, in a HFpEF patient with prominent V-wave, the prominent V-wave creates a large retrograde pressure surge. Since the traditional TPG calculation assumes a steady downstream pressure based on the mid A-wave PAWP, the large retrograde pressure surge (cross-hatched area) is falsely incorporated in the TPG and PVR calculation. Applying the V-wave-corrected PAWP isolates the true and physiologic TPG (solid yellow area), which results in a dramatic decrease in PVR. LV = left ventricle (red line); PA = pulmonary artery (blue line); ECG = electrocardiogram

Pulmonary hypertension (PH) was defined as mean PAP >20 mmHg. Isolated postcapillary PH (IpcPH) was defined as mean PAP > 20 and PVR ≤ 2 WU, while combined pre- and post-capillary PH (CpcPH) was defined as mean PAP > 20 and PVR > 2WU. PA compliance was determined by the ratio of stroke volume to PA pulse pressure²⁴.

Following baseline measurements, patients began graded supine cycle ergometry exercise, starting at 20 W for 3 to 5 minutes for the first stage and incrementally increasing by 20 W every 2 to 3 minutes until exhaustion. Hemodynamic assessment and expired gas analysis data were collected simultaneously at each stage.

Outcome Assessment

Patient follow-up was initiated on the day of invasive exercise testing. Primary outcome was a composite of all-cause mortality and HF hospitalization. Mortality data were ascertained via medical records, death certificates, obituaries, local newspaper notices, and annual death data from the State of Minnesota. HF Hospitalization, defined as an admission requiring intravenous diuretic therapy or intensification of outpatient oral diuretic regimen for acute decompensated HF, was identified using the Mayo Clinic electronic medical record (EPIC), capturing events within the Mayo system and other institutions utilizing the EPIC platform. Patients were censored at the time of last follow-up contact or on July 24, 2025, whichever occurred first.

Statistical Analysis

The data that support the findings of this study are available from the corresponding author upon reasonable request. Continuous variables are expressed as median (interquartile range) and were compared using the Kruskal-Wallis test, while categorical variables are reported as frequencies (percentages) and compared using the chi-square test. Linear regression analysis was used to assess associations between PAWP AUC, mid-A and V-wave-corrected PAWP. Linear mixed models were used to formally test whether changes in LA pressure and volume during atrial filling, V-wave-corrected PAWP, CO and VO₂ from rest to peak exercise differed between groups. Cumulative event rates were estimated using the Kaplan–Meier analysis with differences among the groups assessed by the log-rank test. Univariable and multivariable Cox proportional hazards models were used to further evaluate the prognostic implications of prominent V-waves with adjustment for age, sex, body mass index (BMI) and relevant comorbidities including hypertension, coronary artery disease, diabetes mellitus, and serum creatinine (Model 1), as well as adjustment for resting mean PAWP (Model 2). Sensitivity analyses were performed applying different V-wave cut-offs to define prominent V-wave, including V-wave amplitudes of ≥ 5, 8, and 12 mmHg in excess of end diastolic PAWP. Additionally, a sensitivity analysis excluding patients with a history of catheter ablation for AF or a surgical Cox-Maze procedure was performed to determine if the prognostic difference between patients with and without prominent V-waves persisted in the absence of iatrogenic atrial scarring and decreased LA compliance. We also compared the prognostic significance of the standard prominent V-wave (V wave≥10 mmHg above mean PAWP) and a severe prominent V-wave, defined using an arbitrary cut-off of V-wave≥20 mmHg above mean PAWP, to assess if more profound LA dysfunction or stiffness was associated with incrementally higher risk. No imputation was performed. All analyses used available data, and the number of patients contributing to each analysis is reported where it differs from the full cohort. Correction for multiple comparisons was not performed, and, therefore, the findings should be considered exploratory. A two-sided P-value of <0.05 was considered statistically significant. All data were analyzed using BlueSky Statistics software v. 10.3.4 (BlueSky Statistics LLC, Chicago, IL, USA), which leverages the R programming language (R Foundation for Statistical Computing, Vienna, Austria).

Results

Between February 2006 and July 2025, a total of 1,860 patients with LVEF≥50% underwent initial invasive hemodynamic assessment and expired gas analysis (at rest and during supine exercise) for exertional dyspnea and provided consent. After excluding those not fulfilling inclusion criteria, there were 1,095 patients with confirmed HFpEF and 342 patients with NCD. From this sample, 26 patients with HFpEF and 2 with NCD were excluded due to uninterpretable PAWP V-waves at rest, leaving 1,069 patients with confirmed HFpEF and 340 patients with NCD in the final analysis set. Based on the definition of a PAWP V-wave amplitude exceeding PAWP at mid A-wave ≥10 mmHg, prominent V-wave was present in 225 (21%) HFpEF patients while only 5 (1.5%) NCD patients had a prominent V-wave meeting this definition. The remaining 844 HFpEF and 335 NCD patients were classified as not having prominent V-waves (Figure S2). Among 5 NCD patients with prominent V-wave, mid-A wave PAWP ranged from 12–14 mmHg at rest and 20–22 mmHg at peak exercise, and two of these patients had a history of AF and AF procedures. Of the total of 230 patients with prominent V waves across all patients, 225 (97.8%) had HFpEF based upon an elevated mid-A wave PAWP (Figure S3).

Baseline clinical characteristics of patients with or without prominent V-waves are summarized in Table 1. Patients with prominent V-waves were older (median age 72 vs. 67 years; P<0.001), more likely to be women (61.8% vs. 48.8%; P<0.001), and had higher burden of comorbidities, including hypertension, diabetes mellitus, and chronic kidney disease, along with higher N-terminal pro-B-type natriuretic peptide (NT-proBNP). In contrast, BMI and the prevalence of coronary artery disease were not significantly different between groups (Table 1).

Table 1:

Baseline Characteristics

	No prominent V-wave (n=844)	Prominent V-wave (n=225)	P Value
Age (years)	67 (59–74)	72 (65–77)	<0.001
Women	412 (48.8)	139 (61.8)	<0.001
Body mass index, kg/m2	32.9 (28.7–38.0)	32.3 (27.8–36.6)	0.093
Comorbidities
Hypertension	658 (78.0)	195 (87.4)	0.002
Coronary artery disease	282 (33.4)	69 (30.3)	0.485
Diabetes mellitus	196 (23.2)	70 (31.4)	0.012
Chronic kidney disease	289 (34.2)	111 (49.3)	<0.001
Stage 3a	175 (20.7)	54 (24.0)
Stage 3b	97 (11.5)	42 (18.7)
Stage 4	14 (1.7)	14 (6.2)
Stage 5	3 (0.4)	1 (0.4)
AF	193 (22.9)	138 (61.3)	<0.001
Paroxysmal AF	146 (17.3)	65 (28.9)
Persistent/permanent AF	47 (5.6)	73 (32.4)
Surgical Cox-Maze or AF ablation	47 (5.6)	55 (24.4)	<0.001
Once	33 (3.9)	34 (15.1)
More than once	14 (1.6)	21 (9.3)
Medications
RAAS inhibitors	363 (43.0)	115 (51.1)	0.030
Beta blockers	377 (44.7)	133 (59.1)	<0.001
Diuretics	418 (49.5)	143 (63.6)	<0.001
Laboratories
Hemoglobin, g/dL	13.4 (12.4–14.4)	12.5 (11.4–13.7)	<0.001
eGFR, mL/min	67 (55–81)	60 (45–72)	<0.001
NT-proBNP, pg/mL	160 (64–374)	672 (240–1402)	<0.001

Values are median (interquartile range) or n (%)

Abbreviations: AF = atrial fibrillation; eGFR = estimated glomerular filtration rate; LA = left atrium; NT-proBNP = N-terminal pro-B-type natriuretic peptide; RAAS = renin-angiotensin-aldosterone system

The prevalence of AF was notably higher in the prominent V-wave cohort compared to the patients without prominent V-waves (61.3% vs 22.9%, P<0.001; Table 1, Figure S3). A history of catheter ablation for AF or surgical Cox-Maze procedures was rare, seen in a minority of the overall cohort (9.5%), though prior LA procedures were more common in patients with prominent V-waves than those without prominent V-waves (24.4% vs 5.6%; P<0.001; Figure S3). The use of medications, including renin-angiotensin-aldosterone system (RAAS) inhibitors, beta-blockers and diuretics was also more common in patients with prominent V-waves (Table 1).

Cardiac Structure and Function

Those with prominent V-waves displayed multiple findings indicative of more severe LA stiffening and myopathy, including significantly higher LA volume index (LAVI, 39.4 vs. 30.0 mL/m²; P<0.001) and markedly impaired LA function as evidenced by lower LA reservoir strain (15.4% vs. 23.4%; P<0.001) (Table S1). Although LVEF, LV dimensions and wall thickness were similar between groups, LV global longitudinal strain (15.0% vs 16.3%; P<0.001) and LV diastolic indices were more abnormal in the prominent V-wave group compared to those without prominent V-waves, as evidenced by higher septal E/e’ (15.0 vs. 10.0; P<0.001) and average E/e’ ratios (13.3 vs. 9.7; P<0.001). The higher E/e’ ratio was driven by higher E wave velocity, since e’ values were comparable (Table S1). The prevalence of mild MR, another indicator of LA myopathy, was nearly two-fold greater in patients with prominent V-waves (56.0% vs. 29.6%; P<0.001; Figure S3). LA reservoir stiffness was weakly correlated with mean PAWP (r = 0.42; P<0.001), but correlated much more strongly with V-wave amplitude (r = 0.71; P<0.001) and V-wave to nadir X-descent difference (r = 0.78; P<0.001; Figure S4).

Importantly, right heart failure was more profound in those with prominent V-waves. RV function was more impaired in the prominent V-wave group, with lower tricuspid annular plane systolic excursion (TAPSE: 19 vs. 22 mm; P<0.001) and RV s’ velocity (11 vs. 13 cm/s; P<0.001), as well as lower RV free wall strain (21.0 vs. 23.2; P<0.001). Patients with prominent V-waves also had significantly more tricuspid regurgitation (moderate or greater in 25.4% vs 9.1%; P<0.001), and greater RA enlargement (RA volume index: 24.9 vs. 15.3 mL/m²; P<0.001).

Differences in Resting Pulmonary Vascular Load in Patients With vs Without Prominent V-Waves

In patients with prominent V-waves, elevated pulmonary artery pressures reflected backward transmission of left atrial pressure and increased pulsatile RV load rather than greater resistive pulmonary vascular load. To test this, we first evaluated how the transpulmonary gradient is best estimated. The transpulmonary gradient is defined as the pressure drop between mean PA and the downstream LA, and the latter is most accurately estimated by the averaged area under the pressure-time curve (AUC) in the PAWP tracing for a single cardiac cycle. We measured the PAWP AUC from digitized pressure tracings in a randomly-selected subset of 272 patients from the study cohort. The mid-A wave PAWP was well correlated with the PAWP AUC (R²=0.86, P<.0001), but the V-wave corrected PAWP displayed a notably stronger correlation with the area under the PAWP pressure-time curve as compared with the conventional mid-A-wave PAWP (R²=0.96, P<.0001, P for comparison<0.001).

Invasive hemodynamic data at rest are presented in Table S2. At rest, patients with prominent V-waves had higher mean PAP (32 vs. 24 mmHg; P<0.001), which was predominantly driven by significantly higher PA systolic pressure (Figure 2A), with less difference in PA diastolic pressure as reflected in a significantly wider PA pulse pressure in patients with prominent V-waves compared with those without prominent V-waves (26 [21–35] versus 17 [14–21] mmHg; P<0.001). The magnitude of the V-wave was strongly correlated with PA systolic pressure and PA compliance, but there was less correlation with PA diastolic pressure, and little to no correlation with the end-diastolic TPG (Figure 2B), which did not significantly differ between the two groups (2 [0–6] mmHg in patients with prominent V-waves vs. 3 [1–5] mmHg in those without prominent V-waves; P=0.950).

Figure 2: Impact of Left Atrial Stiffening on Left Atrial Pressure-Volume Relationships and Pulmonary Hemodynamics.

(A) Comparison of pulmonary artery (PA) diastolic and systolic pressures. The elevation in the PA systolic pressure in the heart failure with preserved ejection fraction and prominent V-wave group was more pronounced than the difference in PA diastolic pressure between the groups. (B) Linear correlations between the magnitude of the prominent V-wave (PA wedge pressure [PAWP] V-wave minus mid-A wave) and PA pressures and diastolic transpulmonary pressure gradient (TPG). While larger V-wave magnitude positively correlated with both PA systolic (green line) and diastolic pressures (orange line), the relationship was stronger with PA systolic pressure. Furthermore, the V-wave magnitude did not correlate with diastolic TPG (black line). (C) Pulmonary vascular resistance (PVR) calculated using the traditional mid-A wave PAWP versus a V-wave corrected PAWP. The significant difference in PVR observed in the prominent V-wave group using the standard calculation was attenuated when the TPG was corrected for the prominent V-wave. (D) Impact of the V-wave correction on pulmonary hypertension (PH) classification in heart failure with preserved ejection fraction (HFpEF). When using the traditional mean or mid-A wave PAWP to calculate PVR, patients with prominent V-waves had a significantly higher prevalence of combined pre- and post-capillary PH (CpcPH) compared to patients with HFpEF and no prominent V-waves. However, when applying the V-wave-corrected PAWP, the difference in CpcPH prevalence between the two groups is attenuated. (E) PA compliance was significantly lower in patients with prominent V-waves compared with those without prominent V-waves. (F) LA compliance was assessed during ventricular systole by indexing LA volume change (LA max volume – LA min volume) to LA pressure change (pulmonary artery wedge pressure [PAWP] V-wave – PAWP nadir X-descent). Patients with prominent V-waves (red) had significantly lower LA compliance than those without prominent V-waves (blue) both at rest and peak exercise (P<0.001 for group difference; P=0.055 for interaction). A linear mixed model was fitted to the PAWP rise during LA filling indexed to LA filling volume (PAWP v-x/LA vol max-min), with participant ID as random effect and exercise phase (rest/peak), presence of prominent V-waves, and their interaction as a fixed effect. Error bars reflect standard error of the mean.

A resting mean PAWP (mid-A wave) indicative of HFpEF (≥15 mmHg) was present in 213 patients (94.7%) in the prominent V-wave group, compared with 547 patients (64.8%) in those without prominent V-waves (P<0.001). When applying the V-wave-corrected PAWP, the proportion of patients meeting these resting criteria increased to 99.6% (n=224) in the prominent V-wave cohort and 68.1% (n=575) in the cohort without prominent V-wave.

The prominent V-wave group had higher PVR using end-diastolic PAWP for calculation of TPG (2.36 vs. 1.56 WU; P<0.001), along with higher prevalence of combined pre- and post-capillary PH (60.2% vs 29.7%). However, after incorporating V-wave pressure in the calculation of the TPG, there were no significant differences in corrected PVR (1.38 vs. 1.25 WU; P=0.249; Figure 2C) or prevalence of CpcPH (29.3% vs 25.2%; P=0.296; Figure 2D). Despite the absence of difference in corrected PVR, patients with prominent V-waves displayed notably lower calculated PA compliance compared with those without prominent V-waves (2.67 vs. 4.55 mL/mmHg; Figure 2E), indicating greater pulsatile loading on the RV.

Differences in Exercise Hemodynamics and Cardiopulmonary Reserve

During exercise, patients with prominent V-waves were unable to accommodate increased pulmonary venous return, resulting in greater elevation in effective LA pressure, blunted cardiac output reserve, and impaired augmentation of CO relative to metabolic demand. Invasive hemodynamic data during exercise are presented in Table S2. Expired gas data are presented in Table S3. Figure 2F shows LA pressure-volume relationships at rest and peak exercise in the subset of participants with simultaneous invasive assessment along with echocardiographic imaging (n=403). During ventricular systole, patients with prominent V-waves had a greater restriction in the ratio of LA volume change during the reservoir phase (LA max volume – LA min volume), despite (by definition) greater LA pressure change (PAWP V-wave – PAWP nadir X-descent), indicating significantly decreased LA volume compliance compared to those without prominent V-waves, both at rest (1.56 [1.28–1.84] vs 4.85 [2.47–8.63] ml/mmHg; P=0.002) and exercise (0.86 [0.49–1.35] vs 1.92 [1.07–3.61] ml/mmHg; P<0.001) with P<0.001 for group difference and P=0.055 for interaction. Similarly, LA strain compliance at rest was also significantly decreased in patients with prominent V-waves (1.19 [0.67–1.37] vs 3.72 [2.48–6.45] %/mmHg; P<0.001).

Patients with prominent V-waves had higher V-wave-corrected PAWP but relatively similar CO at rest, compared to patients without prominent V-waves. At peak exercise, patients with prominent V-waves had significantly more limited CO reserve and higher V-wave-corrected PAWP, compared to patients without prominent V-waves (P<0.001 for group difference; P<0.001 for interaction; Figure S5A). The prominent V-wave group displayed a blunted CO with exercise (peak CO 7.7 vs. 9.7 L/min; P < 0.001), with diminished ability to augment CO in response to metabolic demand (P<0.001 for group difference; Figure S5B). Patients with prominent V-waves had higher PVR with exercise (2.24 vs. 1.42 WU; P<0.001), although differences were again attenuated after correction for the V-wave (1.16 vs. 0.82 WU; P=0.035)(Table S2).

Sensitivity Analysis in Patients With and Without Atrial Fibrillation

Prominent V-waves and LA stiffening were not merely a consequence of AF-related atrial myopathy, given that over one-third of patients with prominent V-waves had no history of AF, although the phenotype in this subgroup was milder, with less LA remodeling and less severe pulmonary vascular loading. Among the 225 patients with prominent V-waves, 87 (38.7%) did not have AF. Compared to those with AF, patients without AF had less LA remodeling (LA volume index 34.9 vs 44.0 mL/m²; P<0.001) and less severe impairment in LA reservoir function (LA reservoir strain 20.7% vs 12.1%; P<0.001), despite similar mean PAWP (21 vs 21 mmHg; P=0.775) and slightly lower V-wave amplitude (35 vs 37 mmHg; P=0.017). Pulmonary vascular loading was less severe in those without AF, with lower PA systolic pressure (46 vs 53 mmHg; P<0.001), higher PA compliance (3.28 vs 2.41 mL/mmHg; P<0.001), and lower PVR (2.74 vs 1.83 WU; P<0.001). After correction for the V-wave, corrected PVR was lower (0.96 vs 1.59; P<0.001), and CpcPH was less prevalent in patients with prominent V-waves without AF compared to those with AF (14.8% vs 37.5%; P<0.001) (Table S4).

Clinical Outcomes

Prominent V-waves identified patients at increased risk of adverse outcomes, independent of comorbidities or mean PAWP, and in a graded fashion with V-wave amplitude. Over a median follow-up of 4.2 (1.5, 6.2) years, there were 200 (18.5%) composite events, including 146 (13.5%) all-cause deaths and 83 (7.7%) HF hospitalizations in the cohort of 1,057 patients with available follow up data (223 patients with prominent V-waves and 834 patients without prominent V-waves). Compared with patients without prominent V-waves, patients with prominent V-waves displayed a significantly higher risk of HF hospitalization or death (hazard ratio [HR]: 2.31; 95% CI: 1.71–3.11; P<0.001). In Model 1, the association remained significant after adjusting for age, sex, BMI, hypertension, coronary artery disease, diabetes mellitus, and serum creatinine (adjusted HR: 2.09; 95% CI: 1.53–2.86; P<0.001; Figure 3A). In Model 2, the association persisted after adjusting for resting mean PAWP (adjusted HR: 1.61; 95% CI: 1.16–2.23; P=0.004).

Figure 3: Prognostic Implications of Left Atrial Stiffening in HFpEF.

Kaplan-Meier curves for the composite endpoint of all-cause mortality and heart failure hospitalization. Patients with heart failure with preserved ejection fraction and prominent V-wave (red line) demonstrated significantly higher cumulative event rates compared to the HFpEF patients without prominent V-wave (blue line) (A). Sensitivity analysis demonstrated a step-wise relationship between severity of prominent V-wave and prognosis. Patients with severe prominent V-waves (V-wave ≥ 20 mmHg above mean PAWP, green line) had the highest cumulative event rates, followed by the milder prominent V-wave group (V-wave 10–19 mmHg above mean PAWP, red line), compared to those wihout prominent V-waves (B). Hazard ratios (HR) and 95% confidence intervals (CI) were from multivariable Cox proportional hazards models with adjustment for age, sex, body mass index, hypertension, coronary artery disease, diabetes mellitus, and serum creatinine.

Sensitivity analyses excluding patients with prior MAZE or AF ablation procedures showed that these differences are not simply ascribable to patients with prior LA procedures (adjusted HR: 2.14; 95% CI: 1.53–3.01; P<0.001, Figure S6). In addition, all clinical, echocardiographic and hemodynamic characteristics of the non-iatrogenic prominent V-wave subgroup were replicated in a sensitivity analysis where iatrogenic prominent V-wave was excluded (Table S5).

In sensitivity analyses defining prominent V-wave by V-wave amplitude exceeding end-diastolic PAWP by ≥ 5, 8, and 12 mmHg, findings on key group differences were replicated (Table S6), suggesting that the findings are not dependent upon the threshold selected to define groups. In a separate analysis where patients were stratified into a milder prominent V-waves group (V-wave ≥10–19 mmHg above mean PAWP) and a more severe prominent V-wave group (V-wave ≥20 mmHg above mean PAWP) as shown in Table S7, the less severe prominent V-wave group again displayed higher risk of death or HF hospitalization (adjusted HR: 1.90; 95% CI: 1.34–2.70; P<0.001), while patients with severe prominent V-waves displayed even greater risk for HF hospitalization or death (adjusted HR: 2.75; 95% CI: 1.71–4.43; P<0.001) compared to patients without prominent V-waves (Figure 3B).

Discussion

In this large cohort of patients with invasively confirmed HFpEF, a prominent V-wave in the PAWP tracing, an indirect indicator of heightened operant LA stiffness, was common, most often non-iatrogenic, and associated with three major clinical consequences. First, failure to account for the V-wave inflated the calculated PVR, thereby doubling the apparent prevalence of CpcPH compared with measurements including the V-wave amplitude. Second, despite comparable V-wave corrected PVR, patients with prominent V-waves had markedly increased pulsatile RV afterload, which was associated with greater RV dysfunction, tricuspid regurgitation, and right atrial remodeling. Third, prominent V-waves identified patients with more severely impaired cardiopulmonary reserve and higher risk of death or HF hospitalization. Collectively, these findings show that presumed LA stiffening, defined here by prominent PAWP V-wave in excess of the end-diastolic PAWP, is a common feature in HFpEF that pressurizes the PA and right heart out of proportion to left ventricular end diastolic pressures, contributing to RV dysfunction independently of intrinsic pulmonary vascular disease. These data underline the importance of considering the V-wave in diagnostic evaluation of PH, while emphasizing the priority for new research to identify mechanisms causing atrial stiffening, as well as novel treatments to improve LA compliance properties in HFpEF.

Historically, “stiff LA syndrome” was described as a rare, iatrogenic complication following surgical or catheter-based ablation for atrial fibrillation, driven by extensive atrial scarring³. In these cases, reduction in LA compliance leads to prominent V-waves out of proportion to the mean A-wave PAWP, which approximates LVEDP in the absence of mitral valve disease. The present study shows that a hemodynamically identical phenomenon occurs frequently in patients with HFpEF, who were all defined by elevation in mid-A wave PAWP (i.e., all would have had elevated LVEDP). While patients with prominent V-waves were more likely to have undergone prior AF procedures (24.4% vs. 5.6%), it is striking that over 75% of those with prominent V-waves had no history of prior AF procedures. Instead, this phenomenon appears to be non-iatrogenic in most patients, apparently a consequence of LA myopathy in HFpEF. In this context, prominent V-waves are the result of a left atrium operating on the steep portion of its pressure-volume curve, where a small increase in venous return causes disproportionate rises in pressure¹⁰ (Figure S1). It seems likely that over time, sustained exposure to this pressure overload promotes greater LA dilatation and fibrosis, making LA enlargement a potential surrogate marker of LA stiffness.

To calculate PVR, it is necessary to accurately determine the pressure downstream of the PA in the LA, which is estimated by the PAWP. By convention, this pressure is taken at the mid-point of the A wave, a surrogate of LVEDP²⁵ (Figure 1). However, when the V-wave amplitude is markedly increased, this pressurizes the pulmonary vasculature out of proportion to LVEDP. Adding one-third of the difference between the V wave peak and mid-A wave PAWP has been used as “rule of thumb” to estimate downstream LA pressure in such patients in clinical practice, but the veracity of this approach has not been formally tested. Here, we show that this method better reflects the actual mean PAWP derived from the AUC of the pressure tracing throughout the cardiac cycle, outperforming the standard mid-A PAWP approach, and this estimate can serve as a simple, accurate and easily accessible tool for evaluating the true TPG during the right heart catheterization. Application of this method reduces the TPG, and accordingly the PVR, reclassifying many patients from combined pre- and postcapillary PH to isolated postcapillary PH.

The diagnostic impact is illustrated in the present study. Patients with prominent V-waves had significantly higher calculated PVR using mid-A wave PAWP, with a two-fold higher prevalence of CpcPH compared to the group without prominent V-waves. However, the simple ratio of conventionally defined TPG to CO cannot distinguish pulmonary vascular remodeling, vasoconstriction, perivascular congestion, or retrograde transmission of a prominent V-wave. This was shown in our study when the difference in CpcPH prevalence disappeared when V-wave-corrected PAWP was used. If the downstream V-wave is not accounted for in the calculation of TPG, the patient is more likely to be labelled as having precapillary PH, which may have therapeutic implications, as novel treatments for pulmonary vascular disease emerge in HFpEF²⁶. However, despite similar PVR after correction, RV afterload was still increased in patients prominent V-waves compared with those without prominent V-waves due to retrograde transmission of the V-wave, which preferentially augments PA systolic pressure (with no impact on end-diastolic pressure gradients), reducing calculated PA compliance even in the absence of intrinsic pulmonary vascular disease. The net result exposes the right heart to afterload mismatch like that observed with combined pre- and postcapillary PH. However, the treatment targets are much different, focusing on enhancing LA compliance in those with LA stiffening, rather than pulmonary vascular remodeling as in those with CpcPH.

The impact of intra-cardiac cycle variations in PAWP on the pulmonary circulation has been discussed²⁷ and was revisited to address the limitations of relying solely on PVR to diagnose pulmonary vascular disease^28,29. In the 2015 ESC/ERS guidelines, a diastolic pressure gradient of ≥7 mmHg was recommended to diagnose CpcPH (2015 ESC/ERS guideline), but this component was removed from the most recent guidelines³⁰. In our cohort, the end-diastolic pressure gradient did not differ between the two groups (2 vs. 3 mmHg; P=0.950), and group means both fell below the previously endorsed diagnostic threshold for CpcPH.

These hemodynamic differences were accompanied by greater risk for adverse clinical outcomes. Patients with prominent V-waves had higher resting end-diastolic LV filling pressures than those without prominent V-waves, raising the question of higher risk for clinical events could simply be related to more severe congestion. However, prominent V-waves remained associated with death or HF hospitalization even after adjustment for end-diastolic PAWP, indicating that the prominent V-wave carries independent prognostic information.

Among patients with prominent V-waves, 87 (38.7%) were not in AF, indicating that while AF and LA stiffening are related sequalae of LA myopathy, AF is not required for this phenotype. At similar LV filling pressures, those with AF had larger left atria and markedly lower reservoir strain. While this could indicate a more severe cardiac phenotype related to the dysrhythmia, it seems perhaps more likely that the presence of AF is a consequence and not cause of more advanced LA myopathy in HFpEF.

These findings also help contextualize the results of prior studies. Patients with HFpEF and AF have been shown to display more severe LA myopathy²⁰, and both prevalent and incident AF are independently associated with greater risk for new onset RV dysfunction³¹. The present results would suggest that the association between AF and RV dysfunction is more likely to be mediated by LA stiffening that accompanies AF, rather than the rhythm disturbance per se. This may also have implications for rhythm-control strategies to treat HFpEF, as with catheter ablation, which could worsen LA stiffening by increasing tissue fibrosis even further in such patients.

It is notable that LA involvement in patients with prominent V-waves was not coupled with congruent differences in the LV compared with those without prominent V-waves. Specifically, there were no differences in LV mass, wall thickness, dimensions, or diastolic relaxation (e’ velocity). LVGLS was lower in patients with prominent V-waves than those without prominent V-waves, but this could relate in part to lower LV preload owing to LA dysfunction, as well as the blunted deformation of the mitral annulus caused by LA dysfunction (as the annulus is “shared” by the LA and LV). Notably, the relative difference in LA reservoir strain (15.4 vs 23.4%) was more than 4-fold greater than the relative difference in LVGLS (15.0 vs 16.3%). While early diastolic (e’) annular velocities did not differ, E/e’ was much higher in the prominent V-wave group, indicating much greater LA filling pressure during early diastole. During exercise, there was more severe pulmonary hypertension and poorer cardiac output reserve, though end-diastolic PAWP did not differ in those with prominent V-waves and without prominent V-waves during exertion. Collectively, these findings indicate LA disease out of proportion to LV dysfunction, which has important implications for structures upstream of the left atrium.

The mechanisms driving LA myopathy in HFpEF remain understudied and likely extend to include the adjacent pulmonary venous system, which shares an identical embryological origin and histology with the LA³². This contiguous, thin-walled pulmonary vein-LA continuity is heavily surrounded by epicardial adipose tissue, particularly at the left atrium-pulmonary vein junction³³, which may be more vulnerable to paracrine effects of secreted factors when compared to the thicker-walled LV³⁴. Further research is warranted to determine whether excess epicardial adipose tissue may serve as a mechanical and molecular contributor to the progression of LA and pulmonary vein stiffening in HFpEF, along with chronic LA pressure-volume overload itself due to sustained congestion, along with circulating molecules that may promote dysfunction, remodeling, and fibrosis.

Although we cannot directly evaluate the degree of pulmonary venous disease present in patients with prominent V waves, the present data does impugn LA disease as a key driver, because LA reservoir compliance and strain compliance (direct measures of the LA) were markedly reduced, indicating impaired distensibility of the left atrium itself. The left atrium and pulmonary veins function as an integrated reservoir unit, and dedicated imaging and histopathologic studies are needed to define the relative contribution of each compartment.

Limitations

This study has limitations related to its retrospective, single-center design. First, the cohort consists of patients referred for invasive cardiopulmonary exercise testing at a tertiary care center, which may introduce bias. Second, clinical outcomes were ascertained through retrospective review of electronic medical records. While data linkage between different health systems allowed for the capture of events across a broad geographic area, it is possible that some deaths or HF hospitalizations occurring outside these systems were not entirely identified. The cross-sectional nature of the data precludes insights into causality in the relationships between LA stiffening, pulmonary hypertension severity, and right heart function, though it is self-evident that an increase in the downstream V-wave pressurizes the PA out of proportion to end diastolic LA pressures.

Conclusion

A stiff left atrium, defined invasively by the presence of prominent V-waves out of proportion to end diastolic PAWP, is common, almost universally associated with HFpEF, most often not procedure-related, and associated with multiple features of advanced HFpEF. The prominent V-waves associated with LA stiffening are an important contributor to worsening PH and clinical right heart failure that does not necessarily require intrinsic pulmonary vascular disease, and failure to consider the V-wave contribution overestimates the burden of precapillary pulmonary vascular disease that is present. Further study is indicated to determine what causes LA stiffening out of proportion to LVEDP to inform identification of novel treatments that can improve LA compliance in HFpEF, and V-wave amplitude should be incorporated to measurement of PAWP during hemodynamic catheterization when calculating measures of TPG and PVR.

Clinical Perspective.

What Is New?

• Over 20% of patients with HFpEF have prominent V-waves (≥10 mmHg above mean pulmonary artery wedge pressure [PAWP]), reflecting operant left atrial stiffening, most often non-iatrogenic.

• Correcting PAWP for the V-wave (mean PAWP plus one third of the V-wave amplitude) closely matched the PAWP area under the pressure-time curve, and incorporating this data halved the prevalence of combined precapillary and postcapillary pulmonary hypertension in these patients.

• Despite similar pulmonary vascular resistance, patients with prominent V-waves had lower pulmonary artery compliance, worse right heart failure, and higher risk of death or heart failure hospitalization independent of mean PAWP.

What Are the Clinical Implications?

• During right heart catheterization, V-wave amplitude should be incorporated into PAWP when calculating the transpulmonary gradient and pulmonary vascular resistance to avoid overestimation of pulmonary vascular disease, which could misguide treatment as pulmonary vascular therapies emerge for heart failure.

• A prominent V-wave identifies high-risk HFpEF patients in which right heart failure can develop from left atrial stiffening without intrinsic pulmonary vascular disease.

• Further study is required to better characterize the pathogenesis of left atrial stiffening and identify novel treatments to improve compliance properties.