Authors: Esteban Kosak Lopez (Department of Medicine, Jefferson Einstein Philadelphia Hospital, Philadelphia, Pennsylvania, USA), Andrew Geller (Department of Medicine, Jefferson Einstein Philadelphia Hospital, Philadelphia, Pennsylvania, USA), Raul Leguizamon (Department of Medicine, Jefferson Einstein Philadelphia Hospital, Philadelphia, Pennsylvania, USA), Justin Lam (Department of Medicine, Jefferson Einstein Philadelphia Hospital, Philadelphia, Pennsylvania, USA), Thitiphan Srikulmontri (Department of Medicine, Jefferson Einstein Philadelphia Hospital, Philadelphia, Pennsylvania, USA), Michael Vera Ricaurte (Department of Medicine, Jefferson Einstein Philadelphia Hospital, Philadelphia, Pennsylvania, USA), Maria Siqueira Tavares de Melo (Department of Medicine, Jefferson Einstein Philadelphia Hospital, Philadelphia, Pennsylvania, USA), Thomas Stavola (Department of Medicine, Jefferson Einstein Philadelphia Hospital, Philadelphia, Pennsylvania, USA), John Malin (Department of Medicine, Jefferson Einstein Philadelphia Hospital, Philadelphia, Pennsylvania, USA), Kevin Bryan Lo (Division of Cardiovascular Medicine, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts, USA), Jose M. Martinez‐Manzano (Division of Pulmonary and Critical Care Medicine, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts, USA), Enrique Pacheco (Department of Pulmonary, Critical Care, and Sleep Medicine, Jefferson Einstein Philadelphia Hospital, Philadelphia, Pennsylvania, USA), Zurab Azmaiparashvili (Department of Medicine, Jefferson Einstein Philadelphia Hospital, Philadelphia, Pennsylvania, USA)
Categories: Research Article, chronic obstructive, hemodynamics, hypertension, obstructive, pulmonary, pulmonary disease, sleep apnea, vascular resistance
Source: Pulmonary Circulation
Doi: 10.1002/pul2.70328
Prevalence, Characteristics, and Outcomes Associated With Early Vasculopathy in Patients With Chronic Lung Disease Undergoing Right Heart Catheterization
Authors: Esteban Kosak Lopez, Andrew Geller, Raul Leguizamon, Justin Lam, Thitiphan Srikulmontri, Michael Vera Ricaurte, Maria Siqueira Tavares de Melo, Thomas Stavola, John Malin, Kevin Bryan Lo, Jose M. Martinez‐Manzano, Enrique Pacheco, Zurab Azmaiparashvili
Abstract
Pulmonary hypertension associated with chronic lung disease (PH‐CLD) meets precapillary PH criteria. Less is known CLD patients with “early vasculopathy,” a phenotype characterized by elevated pulmonary vascular resistance (PVR) and normal mean pulmonary artery pressure (mPAP). In this study, we aimed to determine the prevalence, characteristics, and outcomes of the early vasculopathy phenotype among CLD patients. We performed a retrospective cohort analysis of adults with chronic obstructive pulmonary disease or obstructive sleep apnea undergoing right heart catheterization between 2018 and 2023. Patients were categorized into three non‐PH cohort (mPAP ≤ 20 mmHg, PVR ≤ 2 WU), early vasculopathy cohort (mPAP ≤ 20 mmHg, PVR > 2 WU), and precapillary PH cohort (mPAP > 20 mmHg, PVR > 2 WU). The primary endpoint was a 1‐year composite of mortality and major adverse cardiovascular events. Of 243 patients with CLD, 7.4% (n = 18) had early vasculopathy, whereas 35.4% (n = 86) had non‐PH, and 57.2% (n = 139) had precapillary PH. Among CLD patients with early vasculopathy, the median age was 67 years, 56% were males, and 67% were African Americans. Compared to other groups, the early vasculopathy cohort had the lowest right atrial pressure, pulmonary artery wedge pressure, and cardiac output. On multivariable analysis, the early vasculopathy group was at increased risk of 1‐year composite outcome compared to the non‐PH group (adjusted hazard ratio [aHR] = 2.37, [95% confidence interval [CI] 1.11–5.06]; p = 0.025). The association was attenuated after adjusting for left ventricular systolic dysfunction (aHR = 1.24, 95% CI [0.57–2.71]; p = 0.581). Early vasculopathy is an uncommon hemodynamic pattern potentially associated with adverse outcomes seemingly mediated by subclinical heart failure.
Introduction
Pulmonary hypertension associated with chronic lung disease (PH‐CLD) is a well‐recognized complication associated with adverse outcomes in patients with chronic obstructive pulmonary disease (COPD) and obstructive sleep apnea (OSA) [1]. PH‐CLD meets precapillary PH criteria, including an elevated mean pulmonary artery pressure (mPAP) driven by an elevated pulmonary vascular resistance (PVR) and a normal pulmonary artery wedge pressure (PAWP).
The recent 2022 ESC/ERS guidelines redefined precapillary PH criteria, now including patients with elevated mPAP (> 20 mmHg) with a lower PVR (2–3 Wood units [WU]), acknowledging that PVR elevation is a more specific marker of pulmonary vascular disease than mPAP alone [2]. Less is known about the clinical significance of patients who exhibit elevated PVR (> 2 WU) but maintain a “normal” mPAP (≤ 20 mmHg) [3].
Such a hemodynamic profile—normal pressure with elevated resistance—has been termed “early vasculopathy” in other instances, such as heart failure, suggesting an early phase of vascular remodeling in which the mPAP remains unaffected [3]. In the context of PH‐CLD, the pulmonary vasculature is affected in several ways, including vasoconstriction (mediated by alveolar hypoxia), vascular pruning (owing to lung parenchyma destruction and loss of vasculature), and vascular remodeling leading to vessel wall thickening (due to intimal fibrosis, smooth muscle proliferation and hypertrophy, and adventitial thickening) [4]. Therefore, identifying early pulmonary vascular disease among CLD patients could be clinically relevant [5].
To our knowledge, early vasculopathy has not been described in CLD patients. Hereby, we aim to determine the prevalence, clinical characteristics, and clinical outcomes of early vasculopathy among CLD patients with COPD, OSA, or both undergoing right heart catheterization (RHC). We sought to investigate whether these patients have different survival trajectories compared to those without PH.
Methods
We performed a single‐center, retrospective cohort analysis of adult patients ( ≥ 18 years) with a clinical diagnosis of COPD, OSA, and ILD who underwent RHC between January 2018 and December 2023 at Jefferson Einstein Philadelphia Hospital. In patients with multiple RHC, the earliest within the study timeframe was used. The study data were collected using the Jefferson REDCap database. This study was approved by Jefferson's Institutional Review Board (iRISID‐2024‐1574).
Patient Selection
Patients undergoing RHC were identified using Current Procedural Terminology (CPT) codes. RHC indications were based on common clinical scenarios, including volume status clarification, heart failure or valvular heart disease, known or suspected precapillary PH, liver or kidney pre‐transplant evaluation, myocardial ischemic work‐up (as part of left heart catheterization), or other reasons. For patients who had a complete RHC, the following hemodynamic parameters were right atrial pressure (RAP), mPAP, PAWP, cardiac output (CO), and cardiac index (CI). CO and CI were calculated using the Fick principle. The pulmonary vascular resistance (PVR) was calculated as ([mPAP – PAWP]/CO) [6].
A total of 3631 patients undergoing RHC during this period were captured. An initial screening for clinical diagnoses of COPD, OSA, and overlap syndrome (both) using the electronic health records and ICD‐10 coding was done, identifying 825 patients. Then, we excluded patients with incomplete RHC data or without confirmatory ancillary testing, including chest CT, spirometry, or polysomnography (n = 281), and patients not meeting hemodynamic study criteria for this study (n = 301), leaving a total of 243 patients for final analysis (Figure 1).
To differentiate the physiological effects of intrinsic pulmonary vascular disease from left‐sided heart failure, we intentionally excluded patients demonstrating postcapillary PH hemodynamics—defined as isolated postcapillary PH (mPAP of > 20 mmHg, PAWP > 15 mmHg, and a PVR of ≤ 2 WU) or combined pre‐ and postcapillary PH (mPAP of > 20 mmHg, PAWP > 15 mmHg and a PVR of > 2 WU). A detailed flow diagram outlining initial screening, explicit reasons for exclusion, and final hemodynamic allocation is provided in Figure 1.
Furthermore, patients were categorized into three study cohorts according to their invasive hemodynamic profiles. These groups were mutually Cohort 1 (non‐PH) with mPAP of ≤ 20 mmHg, PVR of ≤ 2 WU, and PAWP of ≤ 15 mmHg; Cohort 2 (Early Vasculopathy) with mPAP of ≤ 20 mmHg, PVR of > 2 WU, and PAWP of ≤ 15 mmHg; and Cohort 3 (precapillary PH) with mPAP of > 20 mmHg, PVR of > 2 WU, and PAWP of ≤ 15 mmHg.
Baseline Information and Definitions
Baseline demographics and medical history, including history of asthma, smoking status, pack‐year history, use of supplemental home oxygen, and prevalence of cardiovascular comorbidities such as essential hypertension, diabetes mellitus, hyperlipidemia, coronary artery disease (CAD), and chronic kidney disease were obtained through chart review based on ICD‐10 coding.
The prevalence of emphysema was determined by chest CT scan readings as part of routine clinical care by board‐certified thoracic radiologists at Jefferson Einstein Philadelphia Hospital. We used the closest available spirometry values to RHC to evaluate for airflow obstruction, which was defined as a ratio of forced expiratory volume in 1‐s (FEV1) to forced vital capacity (FVC) of < 0.7. Spirometry were performed in the pulmonary function testing laboratory at Jefferson Einstein Philadelphia Hospital following the American Thoracic Society guideline standards [7].
Transthoracic echocardiography (TTE) interpretations, including assessments for right ventricular (RV) strain and dilation, were performed by board‐certified cardiologists as part of routine clinical care. The left ventricular ejection fraction (LVEF) was reported as % and routinely estimated using the modified Simpson method. Evaluation of RV systolic function varied among echocardiographers, typically involving a quantitative (TAPSE or S') and a qualitative (visual) parameter for RV systolic function assessment. RV dilation was defined as an RV‐to‐LV ratio ≥ 1:1.
Statistical Analysis
All statistical analyses were performed using SPSS Statistics, Version 26.0 (IBM Corp., Armonk, NY). Data were reported as median (25th‐75th percentile) for continuous variables and as frequencies (n) and percentages (%) for categorical variables. Continuous variables that were non‐normally distributed, as assessed by visual inspection (histograms), were compared using the Kruskal–Wallis H test. Categorical variables were compared using the Pearson Chi‐Square test or Fisher's exact test as appropriate. A two‐tailed p‐value of < 0.05 was considered statistically significant for all baseline comparisons.
The primary endpoint was a composite outcome including all‐cause mortality and major adverse cardiovascular events (MACE) at 1‐year follow‐up. MACE included cardiovascular death, acute heart failure, acute coronary syndrome (ACS), stroke, and arrhythmia. The secondary endpoints included the incidence of 1‐year pulmonary outcomes, including COPD exacerbations, respiratory failure, and use of non‐invasive or invasive mechanical ventilation. The date of RHC was defined as the entry point for all analyses. We used unadjusted Kaplan–Meier curves and Log‐Rank test to compare the incidence of the primary outcome across hemodynamic cohorts.
To further evaluate the association between the early vasculopathy cohort and adverse outcomes, we used a multivariable Cox proportional hazards model, using cohort 1 (non‐PH) as the reference category. Data was adjusted for pre‐defined clinical confounders, including demographics (age, gender, and body mass index) and CV risk factors (diabetes mellitus, hypertension, and renal dysfunction [glomerular filtration rate]). To address the potential confounding effect of underlying left ventricular systolic dysfunction on the composite endpoint, we performed a sensitivity analysis using a multivariable Cox regression model maintaining an appropriate events‐per‐variable ratio to avoid overfitting. A sensitivity model was restricted to the primary grouping variable, demographic baselines (age and sex), and LVEF. Data were reported as hazard ratios (HRs) with 95% confidence intervals (CI). Missing data were handled using pairwise exclusion, which applied solely to a single missing LVEF value in the sensitivity analysis, as primary covariates were 100% complete.
Results
Of 243 patients with PH‐CLD, 68% (n = 165) had only COPD, 47% (n = 115) had only OSA, and 15% (n = 37) had overlap syndrome. A total of 35.4% (n = 86) were categorized as having “non‐PH”, 7.4% (n = 18) as having “early vasculopathy”, and 57.2% (n = 139) as having “precapillary PH”. Their median mPAP (in mmHg) was 17 (13–19), 18 (16–19), and 30 (26–35), and their median PVR (in WU) was 1.1 (0.7–1.7), 2.5 (2.1–2.9), and 3.5 (2.7–4.7), respectively. Their complete baseline characteristics are shown in Table 1.
Characteristics of the “Early Vasculopathy” Cohort
Their median age was 67 years (IQR 61–75), 56% were males, and 67% (n = 12) were African American. Eighty‐nine percent (n = 16) had COPD, 17% (n = 3) had OSA, and none had ILD. Their median smoking history was 40 pack‐years; 50% (n = 9) were current smokers, 39% (n = 7) were former smokers. Only 11% (n = 2) were on supplemental oxygen at home. Their median PAWP and RAP were normal, whereas their CO and CI were low. Further baseline characteristics are shown in Table 1.
Comparisons Between Hemodynamic Cohorts
The age and gender distribution between cohorts was comparable. Compared to the other cohorts, the early vasculopathy cohort had a higher prevalence of COPD (p = 0.03) and heart failure (p < 0.001). This group also had a lower BMI, lower left‐and right‐sided filling pressures (PAWP and RAP), and cardiac indices (CO, CI, and LVEF) (each p < 0.001). Conversely, the precapillary‐PH cohort had a higher prevalence of supplemental home oxygen use (p = 0.033) and echocardiographic parameters of RV dysfunction (RV dilation and RV strain), higher right‐sided filling pressures (RAP, mPAP, and TPG), and a lower diffusion capacity of carbon monoxide (each with p < 0.001). The prevalence of CV comorbidities such as hypertension, hyperlipidemia, and diabetes mellitus was elevated and not significantly different between cohorts.
Clinical Outcomes
A total of 38% (n = 93) of patients had composite outcomes within a year after RHC. The median time from RHC to outcomes 282 days (IQR 266–298). The early vasculopathy cohort had the highest prevalence of composite outcomes at 56% (n = 10), primarily driven by MACE (n = 9) (Table 2). Unadjusted Kaplan–Meier analysis showing a higher incidence of 1‐year composite outcomes (log‐rank p‐value = 0.03) and of heart failure (log‐rank p‐value = 0.01) in the early vasculopathy cohort is shown in Figures 2 and 3.
In univariable Cox‐regression analysis, compared to the non‐PH cohort, the early vasculopathy cohort had a significantly higher risk of the 1‐year composite outcome (HR = 2.41, 95% CI 1.15–5.02, p = 0.019) (Table 2). In multivariable Cox‐regression analysis, the early vasculopathy group remained at increased risk of developing the composite outcome within 1 year of follow‐up (adjusted HR = 2.37, 95% CI 1.11–5.06; p = 0.025) (Table 2). On sensitivity analysis, when adjusting for LVEF, the association between early vasculopathy and adverse outcomes was no longer significant (adjusted HR = 1.24, 95% CI 0.57–2.71; p = 0.581) (Table 3).
Discussion
In this retrospective cohort study of CLD patients undergoing invasive hemodynamic evaluation, we identified 7% of patients fitting the “early vasculopathy” phenotype, characterized by a normal mPAP ( ≤ 20 mmHg), elevated PVR ( > 2 WU), and normal left‐ and right‐sided filling pressures. In multivariable analysis, the early vasculopathy group was associated with approximately a 2.4‐fold increased risk of the 1‐year composite outcome, including MACE and all‐cause mortality, compared with the non‐PH group. Such association appears mediated by the high prevalence of left ventricular dysfunction. Our findings suggest that while CLD patients with early vasculopathy do not meet traditional criteria for PH, their hemodynamic profile unmasks a highly vulnerable cardiopulmonary phenotype wherein adverse clinical outcomes are primarily driven by underlying left‐heart dysfunction rather than by isolated pulmonary vascular pathology.
The term “early vasculopathy” has been described among CLD patients with ILD and non‐CLD patients with chronic liver disease, systemic sclerosis, and others [8, 9]. In ILD patients, those with early vasculopathy were defined as having a normal mPAP ( ≤ 20 mmHg) and elevated PVR ( > 2 WU). In these patients, the prevalence of “early vasculopathy” was 26%, and it was an independent predictor of mortality after adjusting for ILD severity features [5]. Similarly, an elevated PVR ( > 2 WU) among patients with borderline mPAP (21–24 mmHg) was associated with increased risk of mortality in patients with chronic liver disease and of progression to more severe PH in patients with systemic sclerosis [9].
Our findings expand the concept of “early vasculopathy” to include CLD patients with COPD and OSA, populations in which pulmonary vascular involvement has traditionally been recognized as clinically significant only after the development of exceedingly elevated pulmonary artery pressures (mPAP > 35 mmHg) and resistance (PVR > 5 WU) [10, 11]. The poorer outcomes observed in our study among CLD patients with “early vasculopathy,” primarily driven by HF exacerbations, reinforce the high sensitivity of PVR as an indicator of early pulmonary vascular disease and suggest that relying on mPAP alone may delay the recognition of clinically relevant pulmonary vascular disease in these patients.
Notably, the rate of adverse outcomes among CLD patients with “early vasculopathy” seemed comparable to that of patients with overt precapillary PH. Such similarity despite patients with established precapillary PH exhibiting higher right‐sided filling pressures, greater rates of RV failure, and more severe hypoxemia. This finding likely reflects a more aggressive management of patients with overt precapillary PH, whereas those with early vasculopathy stages remain underappreciated. For instance, as expected, only a few patients with overt precapillary PH, and none with early vasculopathy, were treated with pulmonary vasodilators in this study.
Alternatively, “early vasculopathy” may represent a subgroup with a particularly vulnerable cardiopulmonary phenotype, characterized by disproportionately low cardiac indices (on RHC and echo) yet normal left‐ and right‐sided filling pressures, resulting in normal pulmonary artery pressures despite mildly elevated PVR. In these patients, it's plausible that mild fluctuations in left ventricular preload, afterload, and contractility act as a “second hit” to an already compromised pulmonary vasculature, which is likely limited in its ability to accommodate additional intravascular volume via distention and recruitment, ultimately leading to cardiovascular events such as pulmonary edema. Our sensitivity analysis confirmed the impact of subclinical left‐heart dysfunction as a mediator of adverse outcomes of this seemingly vulnerable hemodynamic phenotype [12].
Our study results have potential clinical implications. First, PVR should be carefully evaluated in CLD patients with COPD and OSA undergoing RHC, even among patients with normal mPAP. Second, invasive hemodynamic assessment in CLD patients and prompt identification of early pulmonary vascular disease could provide valuable prognostic information and an opportunity for closer surveillance, aggressive management of comorbidities, and future research of targeted therapeutic strategies aimed at preventing progression to overt precapillary PH. Given the small sample size of our early vasculopathy cohort, larger future studies are warranted to compare outcomes between patients with and without preserved LVEF to better isolate the independent prognostic impact of early pulmonary vascular disease.
Our study has several limitations inherent to its retrospective and single‐center design. First, the sample size of the “early vasculopathy” cohort was small (n = 18), limiting the statistical power for granular subgroup analyses. Likewise, the high prevalence of reduced ejection fraction in the cohort of interest challenges our ability to distinguish the net contribution of mildly elevated PVR from primary cardiac dysfunction. Second, the diagnosis of comorbidities such as COPD, OSA, and HF relied on electronic health records and ICD‐10 coding, which may introduce classification bias; however, we mitigated the risk through manual chart review and inclusion of pulmonary function testing, chest CT, spirometry, polysomnography, and echocardiography data. Third, fluid challenges were not performed, so we cannot definitively rule out occult post‐capillary hypertension in patients with borderline wedge pressures. Finally, because this study was conducted at a community‐based hospital without an active lung transplant program, we were unable to capture lung transplantation events as a competing risk in our composite endpoint. However, this lack of tertiary referral bias inherently allows our cohort to reflect a highly representative, real‐world community population.
In summary, early vasculopathy was present in about 1 out of 14 CLD patients undergoing invasive pulmonary hemodynamic assessment, and its presence was associated with adverse outcomes within 1 year of follow‐up. The association seemed driven by their high prevalence of left ventricular dysfunction. Further research is needed on this relatively uncommon but highly relevant hemodynamic occurrence.
Author Contributions
Esteban Kosak Lopez, Zurab Azmaiparashvili, and Enrique Pacheco conceived the study. Esteban Kosak Lopez, Andrew Geller, Raul Leguizamon, Justin Lam, Thitiphan Srikulmontri, Michael Vera Ricaurte, Maria Siqueira Tavares de Melo, Thomas Stavola, and John Malin were responsible for data collection. Esteban Kosak Lopez, Kevin Bryan Lo, and Jose M. Martinez‐Manzano performed the data analysis. Esteban Kosak Lopez, and Jose M. Martinez‐Manzano drafted the manuscript. All authors reviewed and approved the final version of the manuscript.
Funding
The authors have nothing to report.
Ethics Statement
The study was approved by the Institutional Review Board of Jefferson Einstein Philadelphia Hospital (iRISID‐2024‐1574).
Conflicts of Interest
The authors declare no conflicts of interest.
Guarantor
Esteban Kosak Lopez is the guarantor of the content of the manuscript, including the data and analysis.