Authors: Justin K. Lui (aThe Pulmonary Center, Boston University School of Medicine, Boston, MA, USA; bDepartment of Medicine, Boston University School of Medicine, Boston, MA, USA), Ruchika A. Sangani (aThe Pulmonary Center, Boston University School of Medicine, Boston, MA, USA; bDepartment of Medicine, Boston University School of Medicine, Boston, MA, USA), Kari R. Gillmeyer (aThe Pulmonary Center, Boston University School of Medicine, Boston, MA, USA; bDepartment of Medicine, Boston University School of Medicine, Boston, MA, USA), Jasmine Vakhshoorzadeh (bDepartment of Medicine, Boston University School of Medicine, Boston, MA, USA), Marcin A. Trojanowski (bDepartment of Medicine, Boston University School of Medicine, Boston, MA, USA; cArthritis and Autoimmune Diseases Research Center, Boston University School of Medicine, Boston, MA, USA), Andreea M. Bujor (bDepartment of Medicine, Boston University School of Medicine, Boston, MA, USA; cArthritis and Autoimmune Diseases Research Center, Boston University School of Medicine, Boston, MA, USA), Deepa M. Gopal (bDepartment of Medicine, Boston University School of Medicine, Boston, MA, USA; dSection of Cardiovascular Medicine, Boston University School of Medicine, Boston, MA), Renda Soylemez Wiener (aThe Pulmonary Center, Boston University School of Medicine, Boston, MA, USA; bDepartment of Medicine, Boston University School of Medicine, Boston, MA, USA; eCenter for Healthcare Organization & Implementation Research, VA Boston Healthcare System, Boston, MA, USA), Michael P. LaValley (cArthritis and Autoimmune Diseases Research Center, Boston University School of Medicine, Boston, MA, USA; fDepartment of Biostatistics, Boston University School of Public Health, Boston, MA, USA), Elizabeth S. Klings (aThe Pulmonary Center, Boston University School of Medicine, Boston, MA, USA; bDepartment of Medicine, Boston University School of Medicine, Boston, MA, USA)
Categories: Article, Pulmonary arterial hypertension, cardiac hemodynamics, pulmonary vasodilators, scleroderma
Source: Cardiovascular drugs and therapy
Authors: Justin K. Lui, Ruchika A. Sangani, Kari R. Gillmeyer, Jasmine Vakhshoorzadeh, Marcin A. Trojanowski, Andreea M. Bujor, Deepa M. Gopal, Renda Soylemez Wiener, Michael P. LaValley, Elizabeth S. Klings
Although classified as Group 1 pulmonary arterial hypertension (PAH), patients with systemic sclerosis-related pulmonary hypertension (SSc-PH) experience poorer clinical response to PAH therapy and increased mortality compared to those with idiopathic PAH. Due to heterogeneity in phenotypes, identifying patients likely to respond to therapy is challenging. The goal of this study was to determine clinical factors associated with hemodynamic response, defined by a > 20% reduction in pulmonary vascular resistance on repeat right heart catheterization.
We applied a time-to-event model using a retrospective cohort of 39 patients with pre-capillary SSc-PH, defined by a mean pulmonary artery pressure of ≥ 25 mmHg and pulmonary arterial wedge pressure (PAWP) ≤ 15 mmHg on right heart catheterization.
Patients with PAWP ≤ 8 mmHg were nearly four-fold more likely to achieve a hemodynamic response compared to those with PAWP > 8 mmHg (HR 3.88; 95% CI: 1.20, 12.57); each 1 mmHg increase in PAWP was associated with a decreased hazard for hemodynamic response (HR 0.84; 95% CI: 0.70, 1.00).
In patients with pre-capillary SSc-PH, PAWP was associated with time to hemodynamic response, suggesting the importance of subclinical cardiac disease in determining hemodynamic response to oral vasodilator therapy.
Classified as group I pulmonary arterial hypertension (PAH), pulmonary hypertension (PH) affects 8–12% of patients with systemic sclerosis (SSc) and is a risk factor for morbidity and mortality [1]. Diagnosis is established by right heart catheterization (RHC), defined hemodynamically by a mean pulmonary artery pressure (mPAP) > 20 mmHg [2]. Treatment of SSc-PH includes use of PAH-targeted vasodilating agents which reduce pulmonary artery pressures and pulmonary vascular resistance (PVR) while increasing cardiac output. The decision of which medication(s) to use and the need for systemic therapy necessitates a comprehensive risk assessment utilizing symptoms, noninvasive clinical testing, and cardiopulmonary hemodynamics by RHC to determine mortality risk and to monitor response to therapy [3, 4].
However, in contrast to those with idiopathic PAH, patients with SSc-PH have worse outcomes on the same therapies [5]. Why this occurs is thought to be related to the heterogeneity of SSc phenotypes and systemic organ involvement [6]. For example, patients with SSc-PH may have contributions from left-sided disease due to SSc cardiac involvement [7, 8], in which use of PAH therapies can worsen left ventricular (LV) dysfunction. While cardiac disease is clinically evident in only 10–30% of patients with SSc, > 70% may have subclinical disease [9]. The goal of this study was to investigate factors associated with hemodynamic response to oral PAH therapy among patients with SSc-PH. We hypothesized that cardiopulmonary hemodynamic response to PAH therapy may differ among individual patients with SSc-PH and that this may reflect subclinical left-sided cardiac disease.
We conducted a retrospective, single-center study of patients with SSc enrolled into the Scleroderma biorepository at Boston University School of Medicine. At the time of enrollment, patients were consented for inclusion in this longitudinal clinical registry. The study was approved by the Boston University Medical Campus and Boston Medical Center Institutional Review Board. We included patients with SSc-PH diagnosed between 2004 and 2018. PH was defined by a resting mPAP ≥ 25 mmHg on RHC (definition of PH during the study period) [2]. We included only treatment naïve patients with SSc-PH who had a RHC both prior to and after initiation of PAH therapy. We excluded those commenced on upfront systemic prostacyclin or investigational therapies given their different baseline hemodynamic profiles and greater potential for variable treatment response. Those with a pulmonary arterial wedge pressure (PAWP) > 15 mmHg reflective of post-capillary PH at the time of diagnosis were excluded. The primary outcome was hemodynamic response from the time of PAH therapy initiation, defined by a decrease in the PVR > 20% on repeat RHC [10]. Patients were followed until the following conditions were met (whichever came first): 1) Hemodynamic response; 2) The final RHC available in the registry; or, 3) The final RHC immediately prior to starting systemic prostacyclins. If hemodynamic response was not achieved, patients were censored at the time of either the final RHC available in the registry or the final RHC immediately prior to starting systemic prostacyclins. Furthermore, all patients were censored if any of the conditions were met at the end of the study period (January 1, 2021). We collected the following data at the time of 1) Demographics; 2) Tobacco use history; 3) SSc subtype; 4) Renal insufficiency; 5) Systolic blood pressure; 6) Heart rate; 7) Hospitalizations for any cause within the preceding 6 months; 8) Echocardiography data (LV ejection fraction, LV hypertrophy, left atrial dilation, right atrial dilation, pericardial effusion); 9) Pulmonary function testing; 10) Brain natriuretic peptide (BNP); 11) Cardiopulmonary hemodynamics; 12) New York Heart Association (NYHA) functional classification; and 13) Classes of PAH therapy. Additionally, we collected cardiopulmonary hemodynamics and BNP following initiation of treatment.
We described continuous variables by the mean (± standard deviation) and categorical variables by frequencies and percentages for each group. We determined the median time-to-hemodynamic response utilizing the Kaplan Meier method. We applied univariable and multivariable Cox Proportional Hazard regression models to determine factors associated with hemodynamic response. For our adjusted analysis, we explored the impact of the following fixed covariates, determined a priori both by clinical pertinence and differences in clinical characteristics (Table 1) defined at the time of initiation of PAH 1) Age; 2) Male sex; 3) Diffuse cutaneous SSc; 4) NYHA functional class III; 5) Presence of severe PH (defined by mPAP ≥ 35 mmHg); 6) Dual therapy use; and 7) PAWP. To delineate the optimal cutoff for PAWP, we calculated the HR and the 95% CI at each mmHg ranging from 3 to 12 mmHg (Figure 1). The highest PAWP below which there was a significantly increased hazard for achieving hemodynamic response was 8 mmHg. We assessed all covariates for proportional hazards assumption by Schoenfeld residuals. Finally, we conducted a sensitivity analysis excluding those who were lost to follow-up to determine whether this was informative of hemodynamic response. Statistical analyses were conducted in RStudio® (Boston, MA) using the ‘survival’ package in R version 4.2.0 (R Foundation for Statistical Computing, Vienna, Austria).
We identified 39 patients with SSc-PH who underwent RHC before and after initiation of PAH therapy; 15 (38.5%) achieved hemodynamic response. Patient demographics and clinical characteristics are summarized in Table 1. For initial PAH therapy, 34 (87.2%) were placed on monotherapy reflective of the timing of the study. Endothelin receptor antagonists (56.4%) and phosphodiesterase-5 inhibitors (56.4%) were the most frequently utilized oral PAH therapies.
From the unadjusted Cox Proportional Hazards model, each 1-mmHg increase in PAWP was associated with decreased hazards of achieving hemodynamic response (HR 0.84; 95% CI: 0.70, 1.00). When adjusted for age, male sex, diffuse cutaneous SSc, NYHA functional class III presence of severe PH, and dual therapy use, the association between PAWP and hemodynamic response remained unchanged (HR 0.92; 95% CI: 0.71, 1.20). Our sensitivity analysis in which we excluded those lost to follow-up (n = 10) led to similar results. Patients with a PAWP ≤ 8 mmHg had increased hazards of achieving hemodynamic response compared to those with a PAWP > 8 mmHg (HR 3.88; 95% CI: 1.20, 12.57). Age (HR: 0.99; 95% CI: 0.93, 1.05), male sex (HR: 1.35; 95% CI: 0.23, 7.89), diffuse cutaneous SSc (HR: 0.76; 95% CI: 0.16, 3.62), NYHA functional class III (HR: 2.10; 95% CI: 0.60, 7.28), presence of severe PH (HR: 0.65; 95% CI: 0.16, 2.65), and dual therapy use (HR: 2.10; 95% CI: 0.48, 9.25) were not associated with hemodynamic response.
Patients with a PAWP ≤ 8 mmHg had a median time to hemodynamic response of 18.8 months compared to 87.4 months in those with a PAWP > 8 mmHg (log-rank p = 0.02). The pretreatment mPAP was similar between those with a pretreatment PAWP ≤ 8 mmHg compared to those with a pretreatment PAWP > 8 mmHg (36.9 ± 7.6 mmHg vs. 38.0 ± 7.4 mmHg); however, the pretreatment PVR was higher among those with a pretreatment PAWP ≤ 8 mmHg compared to a pretreatment PAWP > 8 mmHg (544 ± 183.8 dyn·s/cm^5^ vs. 423 ± 207.8 dyn·s/cm^5^). Following initiation of treatment, there was a greater reduction in mPAP (35.3 ± 13.7 mmHg vs. 41.2 ± 12.3 mmHg), PAWP (10 ± 7.2 mmHg vs. 11.8 ± 6.1 mmHg), and PVR (446.4 ± 246.1 dyn·s/cm^5^ vs. 620.7 ± 443.7 dyn·s/cm^5^) in patients with a pretreatment PAWP ≤ 8 mmHg compared to those with a pretreatment PAWP > 8 mmHg.
In our SSc-PH cohort, patients with a PAWP of > 8 mmHg were less likely to achieve hemodynamic response compared to those with a PAWP of ≤ 8 mmHg. Given our small cohort, it is possible that the PAWP threshold for hemodynamic response may fall within the range of 8–10 mmHg, which is notably within normal limits (≤ 15 mmHg). Patients who did not exhibit hemodynamic response also had more frequent pericardial effusions on echocardiography and a higher BNP (> 200 pg/mL, on average) while exhibiting similar cardiopulmonary hemodynamics compared to those who did experience hemodynamic response with PAH therapy. Following initiation of PAH therapy, patients who did not meet hemodynamic response also had a greater increase in BNP. Following therapy, patients with a pretreatment PAWP ≤ 8 mmHg also had a greater reduction in mPAP, PAWP, and PVR compared to those with a pretreatment PAWP > 8 mmHg. Taken together, despite a PAWP in the normal range, these findings suggest possible subclinical cardiac involvement in SSc that can contribute to inadequate hemodynamic response. This finding may play a role in explaining why patients with SSc-PH have poorer prognosis than those with idiopathic PAH despite use of the same therapies. Interestingly, the presence of severe PH by hemodynamics and NYHA functional class III, both prognosticators of mortality in PAH [3], were not associated with hemodynamic response. These findings may, in part, be due to the exclusion of patients requiring upfront systemic prostacyclin therapy. Furthermore, upfront dual oral therapy was not associated with hemodynamic response. However, this is likely due to the small number (n = 5) of those on dual therapy.
In the pre-capillary PH characteristic of Group 1 PAH, the rationale behind use of PAH therapy is to reduce PVR and right ventricular afterload and increase pulmonary blood flow. However, in post-capillary PH reflective of left-sided cardiac disease and characterized by a PAWP > 15 mmHg, that increase in pulmonary blood flow can lead to increased left-sided filling pressures which may potentiate LV dysfunction leading to cardiogenic pulmonary edema. Within SSc-PH, there often exists a heterogeneity of clinical phenotypes and systemic multiorgan involvement [6]. Patients with SSc-PH may have contributions from left-sided cardiac disease due to SSc cardiac involvement in which use of PAH therapies can worsen LV dysfunction. While cardiac involvement clinically occurs in only 10–30% of patients with SSc, many more may have subclinical cardiac disease [9]. These patients may not necessarily carry the typical post-capillary hemodynamic profile. Our findings suggest that the optimal PAWP to achieve a positive hemodynamic response may indeed be lower than the PAWP of 15 mmHg used to define post-capillary PH.
Our study has limitations. First, our cohort was derived from retrospective data of a single specialist referral center which led to inconsistencies in data availability and timing, specifically in time intervals between RHCs. We also did not have access to six-minute walk test data commonly used in risk stratification [3, 11, 12]. However, we are reassured that missing six-minute walking distance did not significantly alter the ability of the REVEAL 2.0 score to discriminate between risk categories [13]. To address inconsistencies in data timing, we purposely used time-to-event models to account for some of the time differences. Secondly, while there were no missing data, we were limited to RHC data only available within our registry. This led to the possibility of patients obtaining RHCs at outside institutions, particularly in 10 (25.6%) who were lost to follow-up. However, exclusion of these patients from our analysis did not significantly change our findings. Third, because the timing of diagnosis of PH occurred prior to 2019 when new definitions for PH were established, we utilized the older definition (mPAP ≥ 25 mmHg) [3]. Finally, given the longitudinal nature of the database, standards of care changed numerous times between 2004 and 2018 as many more therapies were FDA-approved for our patients and the importance of upfront dual combination therapy was established.
Among patients with SSc-PH, PAWP was inversely associated with hemodynamic response. These findings suggest that even among patients with pre-capillary SSc-PH, subclinical cardiac involvement may play a role in the pathogenesis of PH and may contribute to the poorer clinical response to vasodilator therapy in patients with SSc-PH.