Prognostic Value of Echocardiographic Coupling Metrics in Systemic Sclerosis-Associated Pulmonary Vascular Disease

Abstract

Background:

Ineffective right ventricular (RV) adaptation to increasing pulmonary arterial (PA) afterload in pulmonary vascular disease (PVD) significantly contributes to morbidity and mortality. PVD in systemic sclerosis (SSc) arises through various mechanisms, yet detecting abnormal contractile response remains challenging. Here, we examine whether echocardiographic RV-PA coupling metrics correlate with invasive pressure-volume (PV) loops, enhancing the prediction of adverse clinical outcomes in SSc-PVD patients.

Methods:

Prospectively enrolled patients with SSc-PVD with paired echocardiogram and PV loops were included. Linear regression and receiver-operating curve (ROC) analysis were used to assess the relationship between tricuspid annular plane systolic excursion (TAPSE)/PA systolic pressure (PASP), fractional area change (FAC)/PASP, tissue Doppler velocity (TDI S’)/PASP, RV free wall strain (RVFWS)/PASP and coupling thresholds defined by end-systolic to end-arterial elastance (Ees/Ea), obtained by the multi-beat method. The contribution of right atrial strain (RAS) to RV-PA coupling parameters was also investigated. Kaplan-Meier analysis was used to identify the relationship between coupling ratios and composite outcomes including clinical worsening, lung transplant, and death.

Results:

42 patients with SSc were studied with mean age 59 ± 12 years, 91% female and varying degrees of PVD: mPAP 29.5 ± 12.8 mmHg, PVR 4.7 ± 4.2 WU, PCWP 10.3 ± 4.1 mmHg. Echocardiographic coupling metrics including TAPSE/PASP, FAC/PASP, TDI S’/PASP, RVFWS_global and RVFWS_basal/PASP, and RAS_reservoir/PASP were linearly associated with Ees/Ea. At cut-points obtained through ROC analysis, all ratios were predictive of RV-PA uncoupling, defined by Ees/Ea, and composite outcomes. Additionally, RAS_reservoir/RVFWS correlated with Ees/Ea even after adjustment for PASP, suggesting that diminished RAS further impacts RV performance and coupling.

Conclusion:

Echocardiographic RV-PA coupling ratios strongly correlate with invasive Ees/Ea and predict adverse clinical outcomes in SSc patients across the spectrum of PVD. Further, we demonstrate how RAS impacts RV performance. These findings may refine risk stratification and prognostication in this at-risk cohort.

Introduction

Systemic sclerosis (SSc) is a disease characterized by immune system dysregulation, leading to multiorgan fibrosis and vasculopathy.[1] Cardiopulmonary involvement in SSc can lead to pulmonary hypertension (PH) across clinical classifications,[2] which develops through multiple mechanisms including small vessel vasculopathy leading to Group 1 pulmonary arterial hypertension (PAH), Group 2 PH from left ventricular systolic and/or diastolic dysfunction, and Group 3 from SSc-associated interstitial lung disease (ILD).[3] Given the intricate and diverse pathways through which PH may manifest in SSc, a range of clinical tools, such as comprehensive clinical assessment and disease subtyping, measurement of pertinent biomarkers, pulmonary function testing (PFTs), and echocardiography, are employed for the screening and early identification.[4–7] In particular, echocardiography plays a crucial role in identifying cardiopulmonary manifestations in SSc, while also providing assessment of right heart adaptation to emerging pulmonary vascular disease (PVD).

The ability of the right ventricle (RV) to adapt to increasing afterload is a primary determinant of morbidity and mortality in PH.[8, 9] The echocardiographic evaluation of effective RV compensation depends significantly on understanding the intricate hemodynamic relationship between RV contractile function and pulmonary artery (PA) impedance, known as RV-PA coupling. Validated against gold-standard invasive pressure-volume (PV) loops, the echocardiographic ratio of tricuspid annular plane systolic excursion (TAPSE) to PA systolic pressure (PASP) has emerged as an important determinant of outcomes in patients with pre- and post-capillary PH, heart failure, chronic lung disease, and aging.[10–16]

However, TAPSE has several known limitations, as it only estimates the longitudinal shortening of the basal RV free wall and can be affected by RV geometry, angulation and preload.[17] In light of these limitations, several alternative echocardiographic indices of RV-PA coupling have been described in non-PH populations, employing various metrics of RV function, including fractional area change (FAC), tissue Doppler imaging (TDI) S’ velocity, and RV free wall strain (RVFWS). In fact, our previous research demonstrates that among these ratios, the FAC/PASP ratio is most strongly associated with functional outcomes in healthy individuals and effectively accounts for age- and sex-specific geometric remodeling.[14, 16] This finding has important clinical implications for the generalizability of previous studies that have focused solely on the TAPSE/PASP ratio as the marker of RV-PA coupling. It also underscores a critical gap in understanding how alternative echocardiographic coupling metrics, which use different RV functional numerators, correlate with gold-standard invasive measures of RV-PA coupling across the PVD spectrum in SSc, a disease characterized by significant phenotypic heterogeneity and deficient right heart adaptation. Additionally, the contribution of right atrial (RA) phasic function to adaptative and maladaptive right heart remodeling in PVD remains insufficiently explored. In this study, we examine well-characterized SSc patients with paired echocardiograms and gold-standard invasive PV loops to assess whether echocardiographic metrics correlate with invasive coupling measures across the PVD spectrum and are predictive of adverse clinical outcomes. We also investigate the impact of RA phasic function on RV performance and RV-PA coupling. We hypothesize that echocardiographic measures, employing various RV functional numerators, can reliably serve as noninvasive correlates of invasively derived RV-PA coupling, offering a practical and accessible alternative for assessing right heart adaptative response in this at-risk population.

Methods

Study Participants.

The Johns Hopkins Pulmonary Hypertension Program is a large IRB-approved longitudinal database (IRB00027124) consisting of well-characterized adults with PVD. The present study is an analysis of SSc patients[18–21] prospectively referred from the Johns Hopkins Scleroderma Center Registry (IRB00226995) for right heart catheterization (RHC) due to suspicion of PH from standard cardiopulmonary screening metrics with PV loop data and same-day 2D echocardiography (2DE).[20, 22] Survival was determined by reviewing the electronic record and searching the Social Security Death Index. All included participants provided informed consent to participation.

Clinical Variables.

We conduct comprehensive annual cardiopulmonary screening for patients with SSc including patient-reported outcome measures, documentation of World Health Organization (WHO) functional status, N-terminal pro brain natriuretic peptide (NT-proBNP), 2DE, PFTs, and 6-minute walk distance (6MWD) when applicable. In the present study, adult patients with SSc were prospectively recruited for RHC due to suspicion of PVD. In addition to extraction of standard cardiopulmonary screening tests closest to RHC, Creatinine, and hemoglobin values at the time of RHC were recorded. PH was defined by mean pulmonary artery pressure (mPAP) > 20 mmHg,[2] and participants were further classified as having no PH (mPAP<20 mmHg), PAH (mPAP>20 mmHg, pulmonary artery wedge pressure (PCWP) ≤ 15 mmHg, pulmonary vascular resistance (PVR)>2 Wood Units). SSc with heart failure with preserved ejection fraction (SSc-HFpEF-PH) was defined as mPAP > 20 mmHg, PVR < 2 Wood units, and PCWP > 15 mmHg, and echocardiographic evidence of left ventricular systolic or diastolic dysfunction with preserved ejection fraction. SSc with interstitial lung disease (ILD) and PH (SSc-ILD-PH) was defined as mPAP > 20 mmHg, PVR > 3 Wood units, and PCWP < 15 mmHg with radiographic evidence of ILD based on established definitions,[2] and further adjudicated by our multidisciplinary clinical team. Participants were prospectively followed, and clinical worsening was defined as 10% or more reduction in 6MWD, and/or worsening WHO functional class, and/or hospitalization, and/or escalation of PAH therapy, and/or if applicable, death or lung transplantation.[23]

Echocardiographic Methods.

Conventional analysis of left and right chamber size and function, as well as left ventricular diastology, were performed following established guidelines[24, 25] using ProSolv (FujiFilm 4.0, Indianapolis, Indiana). Standard 2DE measures of RV function included TAPSE (normally defined as >1.7 cm), tissue Doppler S’ velocity (normal defined as >9.5 cm/s), and FAC (>35%). RV systolic pressure (RVSP) was estimated from the peak tricuspid regurgitant (TR) jet velocity, using the simplified Bernoulli equation, and adding this value to an estimate of the RAP pressure, derived from the diameter and respirophasic variation of the inferior vena cava. In the absence of a transpulmonary gradient, PASP is equal to RVSP.[26] RVFWS was calculated as the average of regional strain from the basal, mid, and apical RV free wall segments using commercially available vendor-independent STE software (Epsilon Version 3.1.03358, Milwaukee, WI)[27] from the 4-chamber RV-focused apical view at 70–90 frames per second, and compared to standard values.[28, 29] STE-derived RA strain (RAS) was analyzed from the RA-focused apical view throughout the cardiac cycle, and included reservoir, conduit, and booster phases.[30]

PV Loop Analysis.

The full details of the RHC and multi-beat PV loop analysis protocol have been previously described.^14,15 Effective arterial elastance (Ea) was calculated in a blinded manner by dividing the end-systolic pressure by the stroke volume. End-systolic elastance (Ees) was calculated using perpendicular regression line of multiple end-systolic points from a family of PV loops generated using preload reduction. RV-PA coupling was defined utilizing the gold standard Ees/Ea ratio derived from multi-beat PV loops.[23] Echocardiograms were read by two independent readers blinded to hemodynamics and clinical parameters, and conventional echocardiographic and speckle-derived echocardiographic right atrial and ventricular functional parameters were tabulated. Coupling metrics included TAPSE/PASP, FAC/PASP, TDI S’ velocity/PASP, RVFWS_global/PASP, RVFWS_basal/PASP, and RAS_reservoir/PASP.

Statistical Analysis.

Demographic, baseline clinical variables, right heart catheterization data, and echocardiography data were tabulated. Associations between echocardiographic surrogates of RV-PA coupling including TAPSE/PASP, FAC/PASP, TDI S’/PASP, global RVFWS/PASP, and basal RVFWS/PASP and Ees/Ea were assessed using Pearson correlation coefficient (r). Variables between groups were compared with a Mann-Whitney U or Kruskal-Wallis test. Statistical significance was defined as p-value < 0.05. Receiver operating curve (ROC) analysis was used to identify cut-offs of echocardiographic ratios which discriminated RV-PA uncoupling, defined as Ees/Ea < 0.805 in accordance with prior work.¹⁶ Kaplan-Meier analysis and log-rank tests were used to assess time to clinical worsening of echocardiographic ratios at cut-offs which were determined by ROC analysis.

Results

Study Cohort and Right Heart Catheterization.

We identified 42 patients with SSc at risk of PH with complete echocardiography, RHC, and PV loop data. Participants were mostly female (91%) and aged mean 59 ± 12 years. Most had WHO functional class 2 (50%) or 3 (43%) at baseline, and the majority had the limited SSc subtype (81%). Following adjudication of the World Symposium on Pulmonary Hypertension (WSPH), nine participants had no PH, 20 had SSc-PAH (WSPH Group 1), five had SSc-HFpEF-PH (WSPH Group 2), and eight had SSc-ILD-PH (WSPH Group 3). Clinical characteristics are demonstrated in Table 1 and Supplemental Table 1.

Table 1. Clinical characteristics of the Systemic Sclerosis Cohort.

Abbreviations: 6MWD=six-minute walk distance, BMI=body mass index, DLCO=diffusion lung capacity of carbon monoxide, FEV1=forced expiratory volume in one minute, FVC=forced vital capacity, HFpEF=heart failure preserved ejection fraction, ILD=interstitial lung disease, NT-proBNP=end-terminal beta-naturetic peptide, PAH=pulmonary arterial hypertension, PH=pulmonary hypertension, SSc=systemic sclerosis, TLC=total lung capacity, WHO=World Health Organization.

	Study Cohort (n=42)	SSc no PH (n=9)	SSc-PH (n=20+5+8)	SSc-PAH (n=20)	SSc-HFpEF-PH (n=5)	SSc-ILD-PH (n=8)	p-value*	p-value#
Female (%)	38 (91)	9 (100)	29 (88)	18 (90)	4 (80)	7 (88)	0.56	0.63
Age (y)	59 ±12	66 ± 9.9	57 ± 12	60 ± 12	53 ± 15	52 ± 8	0.03	0.04
Race, n (%)
White	33 (79)	8 (88)	25 (76)	15 (75)	2 (40)	8 (100)	0.17	0.06
African American	7 (17)	0 (0)	7 (21)	4 (20)	3 (60)	0 (0)
Other	2 (4)	1 (12)	1 (3)	1 (5)	0 (0)	0 (0)
BMI, kg/m2	28.8 ± 6.4	26.7 ± 4.3	28.4 ± 6.9	27.8 ± 7.4	30.4 ± 5.7	28.5 ± 6.9	0.70	0.69
Scleroderma subtype, n (%)
Limited	34 (81)	9 (100)	25 (76)	18 (90)	5 (100)	2 (25)	0.17	0.001
Diffuse	8 (19)	0 (0)	8 (24)	2 (10)	0 (0)	6 (75)
WHO Functional Class at Enrollment, n (%)
1	3 (7)	1 (11)	2 (6)	2 (10)	0 (0)	0 (0)	0.06	0.31
2	21 (50)	6 (67)	15 (45)	9 (45)	2 (40)	4 (50)
3	18 (43)	2 (22)	16 (48)	9 (45)	3 (60)	4 (50)
4	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)
6MWD (meters)	327 ± 117	465 ± 41	349 ± 119	348 ± 133	380 ± 65	334 ± 112	0.007	0.06
NT-Pro-BNP (pg/mL)	491 ± 531	316 ± 329	532 ± 565	644 ± 616	256 ± 107	428 ± 565	0.35	0.43

^*p-value represents comparison between SSc no PH and SSc-PH

^#p-value represents SSc no PH vs. SSc-PAH vs. SSc-HFpEF-PH vs. SSc-ILD-PH

Bolded font signifies statistical significance, p-value<0.05

Echocardiography and Invasive PV Loops.

Hemodynamic and PV loop data, including Ea, Ees, and Ees/Ea are shown in Table 2. Across the study cohort, participants had mPAP of 29.5 ± 12.8 mmHg, PCWP of 10.3 ± 4.1 mmHg, PVR of 4.7 ± 4.2 WU, and cardiac index (CI) of 2.6 ± 0.6 L/min/m². Average Ea across the study cohort was 0.72 ± 0.47 mmHg/mL, Ees was 0.63 ± 0.32 mmHg/mL, and Ees/Ea was 1.13 ± 0.83. Echocardiographic RV functional parameters and echocardiographic surrogates of RV-PA coupling are shown in Table 2. Across the study cohort, TAPSE/PASP was 0.49 ± 0.26 mm/mmHg, FAC/PASP was 1.09 ± 0.61 %/mmHg, TDI S’/PASP was 0.31 ± 0.18 cm/s/mmHg, global RVFWS was −0.43 ± 0.22 %/mmHg, and basal RVLSS/PASP was −0.55 ± 0.34 %/mmHg. All echocardiographic measures were linearly correlated with Ees/Ea, p <0.05, Figure 1.

Table 2. Hemodynamic and Echocardiographic Characteristics by WSPH Classification.

Abbreviations: CI=cardiac index, CO=cardiac output, Ea=effective arterial elastance, Ees= end-systolic elastance, RA=right atrium, RV=right ventricle, PA=pulmonary artery, PAH=pulmonary arterial hypertension, PCWP=pulmonary capillary wedge pressure, PH=pulmonary hypertension, SSc=systemic sclerosis, WSPH=World Symposium of Pulmonary Hypertension.

	Entire Study Cohort (n=42)	SSc no PH (n=9)	SSc PH (n=20+5 +8)	SSc-PAH (n=20)	SSc-HFpEF-PH (n=5)	SSc-ILD-PH (n=8)	p-value*	p-value#
Hemodynamic Parameters
Heart Rate (bpm)	73 ± 10	75 ± 10	73 ± 10	74 ± 10	69 ± 7	73 ± 13	0.63	0.81
Systolic Blood Pressure (mmHg)	131 ± 20	144 ± 11	127 ± 20	127 ± 19	138 ± 24	123 ± 19	0.02	0.07
Diastolic Blood Pressure (mmHg)	71 ± 10	69 ± 8	71 ± 10	71 ± 12	73 ± 12	71 ± 7	0.48	0.90
RA pressure (mmHg)	6.3 ± 4.1	3.4 ± 0.9	6.9 ± 4.3	7.5 ± 4.7	8.6 ± 1.9	4.6 ± 3.5	0.02	0.02
RV systolic pressure (mmHg)	48.3 ± 21.6	30.7 ± 5.3	53.8 ± 21.8	61.0 ± 24.6	40.2 ± 7.9	40.6 ± 8.0	0.003	0.001
RV diastolic pressure (mmHg)	7.3 ± 4.8	3.7 ± 1.2	8.6 ± 5.4	9.3 ± 5.4	9.4 ± 3.3	5.3 ± 3.4	0.01	0.009
PA systolic pressure (mmHg)	48.9 ± 21.3	30.6 ± 6.3	53.9 ± 21.2	61.7 ± 23.8	41.2 ± 7.4	42.3 ± 7.4	0.003	0.0005
PA diastolic pressure (mmHg)	18.2 ± 8.0	10.2 ± 1.9	20.2 ± 7.9	22.6 ± 8.6	18.6 ± 4.5	16 ± 4.1	0.0006	0.0004
Mean PA pressure (mmHg)	29.5 ± 12.8	17.3 ± 3.1	32.8 ± 12.4	37.3 ± 13.6	27 ± 6.5	25.3 ± 5.8	0.0007	0.0002
PCWP (mmHg)	10.3 ± 4.1	9.1 ± 2.4	10.7 ± 4.4	10.2 ± 3.9	16.2 ± 4.3	8.6 ± 3.0	0.30	0.003
PVR (WU)	4.7 ± 4.2	1.9 ± 0.9	5.6 ± 4.3	7.0 ± 4.9	2.5 ± 1.8	3.9 ± 1.4	0.02	0.004
CO (L/min)	4.7 ± 1.3	4.6 ± 0.8	4.7 ± 1.3	4.6 ± 1.3	5.4 ± 2.4	4.5 ± 0.8	0.88	0.58
CI (L/min/m2)	2.6 ± 0.6	2.6 ± 0.3	2.6 ± 0.7	2.5 ± 0.6	2.7 ± 1.2	2.5 ± 0.5	0.79	0.96
PA saturation (%)	68.3 ± 5.2	70.6 ± 3.1	67.7 ± 5.5	66.6 ± 5.9	68.7 ± 6.1	70.1 ± 3.7	0.19	0.24
Ea (mmHg/mL)	0.72 ± 0.47	0.42 ± 0.14	0.82 ± 0.56	0.95 ± 0.56	0.43 ± 0.17	0.66 ± 0.24	0.03	0.02
Ees (mmHg/mL)	0.63 ± 0.32	0.81 ± 0.42	0.58 ± .28	0.57 ± 0.17	0.52 ± 0.12	0.66 ± 0.51	0.07	0.02
Ees/Ea	1.13 ± 0.83	2.05 ± 1.06	0.91 ± 0.56	0.78 ± 0.48	1.29 ± 0.42	1.03 ± 0.74	0.0001	0.0005
Echocardiographic Parameters
LV Ejection Fraction, %	60 ± 8	61 ± 8	60 ± 8	61 ± 7	56 ± 11	59 ± 8	0.78	0.72
Mitral E/A ratio	1.1 ± 0.4	0.9 ± 0.2	1.2 ± 0.5	1.1 ± 0.3	1.7 ± 0.9	1.1 ± 0.4	0.12	0.24
E/e’ (n=27)	11.8 ± 5.3	10.6 ± 6.6	12.1 ± 5.6	12.4 ± 5.8	12.2 ± 3.6	11.1 ± 6.8	0.60	0.66
RA Area, cm2	15.1 ± 4.4	13.1 ± 3.4	15.6 ± 4.6	15.3 ± 4.3	18.7 ± 6.7	14.1 ± 2.8	0.14	0.26
RV basal diameter, cm	3.8 ± 0.7	3.7 ± 0.8	3.8 ± 0.6	4.0 ± 0.7	3.7 ± 0.3	3.5 ± 0.6	0.49	0.33
RVEDd, cm	3.1 ± 0.5	3.0 ± 0.6	3.2 ± 0.4	3.1 ± 0.4	3.1 ± 0.4	3.2 ± 0.5	0.72	0.95
RVEDA, cm2	20.2 ± 6.6	17.3 ± 6.7	21.0 ± 6.4	21.2 ± 6.6	22.0 ± 6.5	19.4 ± 6.6	0.10	0.33
RVESA, cm2	12.4 ± 5.8	10.2 ± 6.4	12.9 ± 5.6	13.8 ± 6.0	10.9 ± 2.5	12.0 ± 6.2	0.10	0.30
TAPSE, mm, n=32	17.9 ± 5.0	18.8 ± 6.3	17.6 ± 4.6	17.1 ± 3.9	21.0 ± 8.0	16.8 ± 3.3	0.18	0.29
PASP, mmHg, n=29	46.0 ± 22.1	28.3 ± 5.6	50.6 ± 22.4	60.6 ± 25.7	35 ± 4.2	39.3 ± 3.7	0.0001	0.0009
TDI S’ Velocity, cm/s, n=21	11.5 ± 4.0	10.9 ± 3.0	11.7 ± 4.2	9.7 ± 1.7	16.2 ± 6.0	11.6 ± 3.3	.65	.27
FAC, %	40.7 ± 11.1	44.3 ± 11.7	39.7 ± 10.8	36.9 ± 10.6	49.2 ± 7.3	40.5 ± 10.5	0.25	0.07
Basal RVFWS, %	−22.4 ± 7.7	−25.8 ± 6.9	−21.6 ± 6.9	−22.1 ± 5.9	−23.0 ± 11.3	−19.5 ± 6.6	0.06	0.24
Midventricular RVFWS, %	−18.3 ± 6.1	−18.7 ± 3.0	−18.2 ± 6.6	−18.9 ± 7.2	−17.8 ± 4.1	−16.7 ± 6.5	0.85	0.97
Apical RVFWS, %	−12.3 ± 6.6	−11.1 ± 6.0	−12.5 ± 6.8	−12.2 ± 7.6	−13.6 ± 6.7	−12.6 ± 5.4	0.67	0.94
Global RVFWS, %	−17.7 ± 5.2	−18.5 ± 4.3	−17.5 ± 5.5	−17.9 ± 5.5	−18.2 ± 6.5	−16.0 ± 5.2	0.65	0.79
RA reservoir strain, %	25.9 ± 8.1	22.7 ± 5.0	26.7 ± 8.7	26.0 ± 7.8	32.2 ± 9.6	24.8 ± 10.4	0.28	0.32
RA booster strain, %	−12.7 ± 5.9	−9.5 ± 4.9	−13.5 ± 5.9	−13.1 ± 5.9	−15.8 ± 6.8	−13.2 ± 5.9	0.12	0.37
RA conduit strain, %	−13.8 ± 5.3	−14.1 ± 5.9	−13.6 ± 5.3	−13.6 ± 5.1	−16.2 ± 4.8	−12 ± 6.2	0.92	0.68
TAPSE/PASP, mm/mmHg, n=25	0.49 ± 0.26	0.75 ± 0.25	0.41 ± 0.20	0.30 ± 0.12	0.62 ± 0.31	0.42 ± 0.09	0.001	0.003
FAC/PASP, %/mmHg, n=29	1.09 ± 0.61	1.79 ± 0.50	0.89 ± 0.48	0.72 ± 0.47	1.41 ± 0.39	0.94 ± 0.26	0.0009	0.0002
TDI S’/PASP, cm/s/mmHg, n=17	0.31 ± 0.18	0.41 ± 0.11	0.29 ± 0.19	0.16 ± 0.09	0.48 ± 0.21	0.27 ± 0.05	0.10	0.02
Global RVFWS/PASP, %/mmHg, n=29	−0.43 ± 0.22	−0.72 ± 0.19	−0.39 ± 0.19	−0.33 ± 0.20	−0.56 ± 0.10	−0.39 ± 0.14	0.01	0.01
Basal RVFWS/PASP, %/mmHg, n=29	−0.55 ± 0.34	−0.85 ± 0.53	−0.48 ± 0.24	−0.42 ± 0.24	−0.69 ± 0.26	−0.48 ± 0.17	0.11	0.12
RA reservoir strain/PASP, %/mmHg, n=21	0.69 ± 0.35	0.67 ± 0.36	0.62 ± 0.38	0.94 ± 0.33	0.51 ± 0.23	0.51 ± 0.23	0.37	0.28
Pericardial effusion, n (%)
None, n (%)	32 (76)	8 (89)	24 (73)	14 (70)	0 (100)	5 (63)	0.76	0.53
Mild, n (%)	9 (21)	1 (11)	8 (24)	6 (30)	0 (0)	2 (25)
Moderate, n (%)	1 (3)	0 (0)	1 (3)	0 (0)	0 (0)	3 (12)

^*p-value represents comparison between SSc no PH and SSc-PH

^#p-value represents SSc no PH vs. SSc-PAH vs. SSc-HFpEF-PH vs. SSc-ILD-PH

Bolded font signifies statistical significance, p-value<0.05

Figure 1. Relationship between echocardiographic and invasive PV loop measures of RV-PA coupling.

(A) TAPSE/PASP, (B) FAC/PASP (C) TDI S’/PASP, (D) Global RVFWS/PASP, (E) Basal RVFWS/PASP. Abbreviations: Ea=effective arterial elastance, Ees= end-systolic elastance, FAC=fractional area change, PASP=pulmonary artery systolic pressure, RA=right atrium, RVFWS=right ventricular free wall strain, TAPSE=tricuspid annular plane systolic excursion, TDI=tissue Doppler imaging.

ROC Analysis and RV-PA Uncoupling.

ROC analysis showed cut-points of 0.40 mm/mmHg for TAPSE/PASP, 0.83 %/mmHg for FAC/PASP, 0.22 cm/s/mmHg for TDI S’/PASP, 0.46 %/mmHg for RVFWS_global/PASP, 0.54 %/mmHg for RVFWS_basal/PASP, and 0.36 %/mmHg for RAS_reservoir/PASP to discriminate RV-PA uncoupling as defined by Ees/Ea < 0.805, Table 3. There were no statistically significant differences in AUC between echocardiographic coupling ratios.

Table 3. Receiver-Operating Curve Analysis for Prediction of RV-PA Uncoupling.

Abbreviations: AUC=area under the curve, FAC=fractional area change, PASP=pulmonary artery systolic pressure, RA=right atrium, RVFWS=right ventricular free wall strain, TAPSE=tricuspid annular plane systolic excursion, TDI=tissue Doppler imaging.

Ratio	Cut-Point	Sensitivity (%)	Specificity (%)	Positive Predictive Value (PPV)	Negative Predictive Value (NPV)	AUC (95% CI)	Standard Error
TAPSE/PASP (mm/mmHg)	0.40	80	70	0.80	0.70	0.81 (0.640.98)	0.09
FAC/PASP (%/mmHg)	0.83	82	64	0.70	0.77	0.82 (0.660.97)	0.08
TDI S’/PASP (cm/s/mmHg)	0.22	90	57	0.80	0.75	0.80 (0.581.0)	0.11
Global RVFWS/PASP (%/mmHg)	0.46	67	75	0.64	0.76	0.73 (0.540.93)	0.10
Basal RVFWS/PASP (%/mmHg)	0.54	60	94	0.60	0.93	0.79 (0.650.93)	0.07
RA Reservoir Strain/PASP (%/mmHg)	0.36	56	78	0.90	0.70	0.78 (0.570.99)	0.11

RA phasic function and RV-PA coupling.

RAS_conduit and RAS_reservoir both correlated with Ees, but not Ees/Ea ratio (Figure 2A). RAS_reservoir/PASP correlated with Ees/Ea ratio (Figure 2B), however, there was no correlation between RVFWS_global and RAS_reservoir as independent variables. We then sought to determine the impact of load on these metrics, and after adjusting for PASP, RAS_reservoir/RVFWS had a strong correlation with Ees/Ea ratio (Figure 2C). In sensitivity analyses accounting for the degree of TR, 13 subjects had trace TR, 27 subjects had mild TR, and 2 subjects had moderate TR. After excluding individuals with moderate TR, the association between RA mechanics and coupling remained significant (p-value<0.01).

Figure 2. Relationships between RA phasic function and RV-PA coupling.

(A). Correlation between RA conduit strain (left) and RA reservoir strain (right) with Ees. (B) Correlation between Ees/Ea and RA reservoir strain/PASP (left), and scatter plot showing no relationship between RVFWS and RA reservoir strain (right). (C) Correlation between RVFWS and RA reservoir strain, adjusted for PASP (left). A positive residual corresponds to preserved RA function, while a negative residual to reduced RA function. Correlation between RA reservoir strain to PASP adjusted for RVFWS and Ees/Ea (right). Abbreviations: Ea=effective arterial elastance, Ees= end-systolic elastance, PASP=pulmonary artery systolic pressure, RA=right atrium, RVFWS=right ventricular free wall strain.

Outcomes and RV-PA Uncoupling.

Patients were followed for a mean time of 3.3 ± 2.4 years. TAPSE/PASP, FAC/PASP, global RVFWS/PASP, and RAS_reservoir/PASP dichotomized by defined cut points, were predictive of time to clinical worsening in Kaplan Meier analysis (log-rank p<.05) as shown in Figure 3.

Figure 3.

Relationship of Echocardiographic Coupling Ratios and Composite Outcomes in SSc. Outcomes stratified by (A) TAPSE/PASP, (B) FAC/PASP, (C) TDI S’/PASP, (D) RVFWS_global/PASP, (E) RVFWS_basal/PASP, and (F) RAS_reservoir/PASP. Abbreviations: Ea=effective arterial elastance, Ees= end-systolic elastance, FAC=fractional area change, PASP=pulmonary artery systolic pressure, RA=right atrium, RVFWS=right ventricular free wall strain, TAPSE=tricuspid annular plane systolic excursion, TDI=tissue Doppler imaging. made with Biorender.

Discussion

In a cohort of patients with SSc with varying degrees and severity of PVD, echocardiographic coupling ratios strongly correlated with invasive RV-PA coupling measurements across patients with variable mPAP and WSPH classifications. In particular, TAPSE/PASP < 0.40 mm/mmHg, FAC/PASP < 0.83%/mmHg, RVFWS_global/PASP < 0.46 %/mmHg, and RVFWS_basal/PASP < 0.54 %/mmHg define RV-PA uncoupling and are predictive of adverse clinical outcomes. To our knowledge, these findings are the first to represent echocardiographic thresholds of RV-PA uncoupling, validated against gold-standard multi-beat invasive PV loops, across the phenotypic spectrum of SSc-PVD. We additionally found significant relationships between RA phasic function and markers of RV-PA coupling, and demonstrate that after adjustment for PASP, RAS_reservoir/RVFWS_global correlates with Ees/Ea ratio, signifying the importance of RA phasic function on right heart performance in SSc patients, a cohort at exceedingly high risk for adverse clinical outcomes due to right heart maladaptation to pulmonary vascular disease.

SSc is a remarkably diverse syndrome marked by intricate interactions among overlapping risk factors, leading to varied pathophysiologic manifestations. The prevalence of SSc-associated PH spans multiple WHO classifications, and recent research indicates an improvement in the 5-year transplant-free survival for SSc-PAH over the last decade.[31] Nevertheless, the prognosis for SSc-PAH remains poor, characterized by disproportionately poor clinical outcomes, including reduced treatment response and worse functional status when compared to other PAH causes, despite comparable hemodynamic abnormalities.[32–34] Moreover, mortality rates in SSc-associated PH due to WHO Groups 2 and 3 remain unacceptably high, with minimal improvements in survival despite advancements in our understanding of these disease subtypes.[31]

Regardless of the etiology, the capacity of the RV to augment contractility in response to increased afterload is the primary determinant outcome in PH.[35, 36] We have previously demonstrated that SSc patients with and without known PH have heterogeneous patterns of regional RV contractility inadequately captured by conventional echocardiographic techniques,[37] that progressively worsen in SSc-associated PAH, when compared to other PAH subtypes with similar severity of hemodynamic derangements.[38] During exercise provocation, SSc patients exhibit echocardiographic impairments in RV contractile reserve with dilatation of the right heart chambers to maintain cardiac output. We further demonstrated that these STE-derived contractile abnormalities mirror invasive evidence of RV maladaptive response to progressive PVD,[39, 40] related to sarcomeric deficiencies in calcium handling derived from RV myocytes in patients with SSc.[41] Given the inherent abnormalities in the scleroderma myocardium, the ability to utilize echocardiographic metrics to detect effective versus ineffective RV adaptation to PVD at an earlier stage in the disease has significant clinical implications including the prompt initiation of directed therapeutics.

The echocardiographic TAPSE/PASP ratio is a validated and integrative index of RV-PA coupling that correlates with invasive hemodynamics, functional class, and clinical outcomes predominately in precapillary PH, and those with severe disease.[10, 11, 13, 15, 42] In a cohort of 52 patients with pre-capillary PH due to WSPH Group 1 PAH or WSPH Group 3 chronic thromboembolic pulmonary hypertension (CTEPH), a TAPSE/PASP ratio cutoff of 0.31 mm/mmHg was identified as an independent discriminator of uncoupling, as defined by Ees/Ea.[13] The lower threshold of 0.31 mm/mmHg observed in this study compared to 0.40 mm/mmHg may be attributed to a sicker PAH cohort with more pronounced hemodynamic derangements (median mPAP 47 mmHg versus 29.5 ± 12.8 mmHg in our cohort) with a lower median Ees/Ea of 0.70 compared to 1.13 ± 0.83 in our study. In a series of WSPH Group 2 patients with PH secondary to chronic lung disease,[43] a TAPSE/PASP cut-off of 0.26 mm/mmHg discriminated severity of PH and predicted mortality. In non-PH populations absent of cardiovascular disease (CVD), the TAPSE/PASP ratio has also been shown to correlate with functional parameters such as 6MWD, NT-proBNP, and patient-reported outcomes.[14]

However, prior investigations of the TAPSE/PASP ratio in SSc patients have predominantly concentrated on SSc-PAH and are limited by the absence of correlation with gold-standard PV loop hemodynamics. In a study of 60 SSc-PAH patients, a lower cutoff value for TAPSE/PASP, set at 0.194 mm/mmHg, was employed to predict “high risk,” as defined by ESC/ERS criteria.[44] Yet, determining the clinical relevance of these findings proves challenging due to the lack of invasive validation. In a separate study of 70 SSc patients, a TAPSE/PASP cut-off of < 0.7 mm/mmHg improved the precision of PAH diagnosis when added to the DETECT algorithm,[6, 45] and in another study drawing from the EUSTAR cohort, TAPSE/PASP ≤ 0.32 mm/mmHg was the most significant predictor of mortality in multivariate analysis.[46] These studies, however, similarly lacked PV validation, thereby limiting the generalizability. Our study significantly strengthens previous attempts by conducting a thorough investigation of SSc patients encompassing various classifications of PH with varying severity as well as SSc patients without a known diagnosis of PH who may be at risk, with PV loop hemodynamic validation.

Further, we present alternative echocardiographic numerators of RV contractility within the coupling ratio and evaluate their performance against PV loop-derived Ees/Ea gold-standard measures of RV-PA coupling utilizing a multi-beat method. Several prior studies in non-SSc cohorts have suggested alternative echocardiographic numerators. We have previously shown that in a large non-PH population without prevalent CVD, that each coupling ratio may have distinct advantages and disadvantages based on both age- and sex-related geometric and functional assumptions.[16] For example, TAPSE/PASP and FAC/PASP were strongly associated with 6MWD, NT-proBNP, and Kansas City Cardiomyopathy Questionnaire (KCCQ) scores compared to TDI S’/PASP and RVFWS/PASP. Due to our sample size in the present study, we were unable to determine if one echocardiographic parameter had superior performance or predictive capacity for clinical worsening. However, in PH populations, a study by Ünlü and colleagues demonstrated that the RVFWS/PASP score was the only one to associate with the combined endpoint of death or heart/lung transplantation in pre-capillary PH patients with multivariable analysis, exhibiting a more robust predictive capability compared to TAPSE/PASP.[47] Given the lack of PV validation, in a follow-up cohort study by Richter et al, retrospective analysis of 29 patients with PAH, demonstrated a strong correlation between RVFWS/PASP and Ees/Ea utilizing a single-beat analytic method.[48] Our study builds on these previous investigations by assessing the predictive value of all echocardiographic coupling metrics correlated with multi-beat invasive hemodynamics. Specifically, we demonstrate the highest negative predictive value for RVFWS_basal/PASP and relatively similar AUC for all coupling ratios (0.78–0.82, Table 3) with the lowest AUC for RVFWS_global/PASP. These findings provide a framework for reliably assessing RV-PA coupling across the continuum of SSc-PVD. Our approach further enables the identification of validated thresholds and prognostic values for all echocardiographic functional indices within the coupling ratios, accounting for differing degrees of RV adaptation. The proposed echocardiographic parameters have applicability across the spectrum of PH, and our study adds to the literature by demonstrating correlation between invasive hemodynamic assessment of RV-PA coupling across a range of classes of PH.

Lastly, we are the first to demonstrate the contribution of RA mechanics to right heart performance in SSc, a disease marked by striking phenotypic heterogeneity and overlapping cardiopulmonary subtypes. RA phasic function reflects RV hemodynamic conditions, systolic and diastolic abnormalities, and can be simplified into three key components: the RA acts as a reservoir for systemic venous return during RV systole, facilitates blood flow to the RV during early diastole, and contracts to augment late diastolic RV filling.[30] In patients with SSc, a prior study demonstrated impaired RA function, even before clinically evident PH or RV dysfunction.[49] A separate analysis of 70 patients with SSc demonstrated that RA stiffness, as measured by the ratio of E/e’ to RA reservoir strain, is increased in patients with SSc and linked to functional capacity as measured by 6MWD.[50] Further, RA reservoir strain and contractile strain were significantly lower in patients with SSc. We extend these findings by demonstrating that RAS_reservoir/RVFWS correlated with Ees/Ea even after adjustment for PASP, suggesting that diminished RAS further impacts RV performance and coupling. Future investigations focused on the dynamic interplay between RV diastolic parameters and RA phasic function in SSc may provide mechanistic insights underlying these observations.

Our study has several limitations. First, our study is a cross-sectional analysis of SSc patients undergoing a dedicated and highly specialized research protocol that may limit reproducibility. Nevertheless, it is crucial to highlight that our study population accurately mirrors a genuine clinical sample of SSc patients susceptible to PH, and the variability of pulmonary pressures within our cohort suggests a spectrum of disease and phenotypic heterogeneity. We also provide a clinically relevant framework for the noninvasive assessment of RV-PA coupling utilizing different RV functional parameters as the numerator. Additionally, established standardization of post-processing strain software and vendor-specific variability may limit the generalizability of these results.[51] Finally, the limited sample size of 42 patients in our study may restrict the broader application of the identified phenotypes to larger cohorts, emphasizing the necessity for additional validation in non-SSc populations as a critical extension of our work.

Conclusion

We demonstrated that noninvasive echocardiographic ratios including TAPSE/PASP, FAC/PASP, TDI S’/PASP, and strain measures, validated against multi-beat invasive PV hemodynamics, can serve as noninvasive measures of RV-PA coupling in patients with SSc at risk for and with varying degrees of PVD. We also show the importance of RA phasic function on RV performance. Patients with SSc routinely undergo screening echocardiography, and due to the myriad ways PH can develop in this condition, these parameters may provide powerful prognostic information to detect ineffective right heart adaptation at an early stage of disease.

Central Illustration.

Relationship between pressure volume loop measurement of RV-PA coupling and echocardiography in patients with scleroderma at risk for pulmonary vascular disease. Abbreviations: Ea=effective arterial elastance, Ees= end-systolic elastance, FAC=fractional area change, PASP=pulmonary artery systolic pressure, RVFWS=right ventricular free wall strain, TAPSE=tricuspid annular plane systolic excursion, TDI=tissue Doppler imaging.

Highlights.

• Echocardiographic RV-PA coupling correlates with invasive hemodynamics

• RV-PA coupling metrics apply to scleroderma patients with diverse types of PH

• RA and RV strain measures have value in patients with scleroderma

• Echocardiographic RV-PA coupling metrics have prognostic significance