Early parenteral prostacyclin therapy enables near-normal haemodynamics and treatment de-escalation in patients with haemodynamically severe pulmonary arterial hypertension

Extract

Parenteral prostacyclins including epoprostenol or treprostinil are cornerstone therapies for haemodynamically severe pulmonary arterial hypertension (PAH). Although not double-blind or placebo-controlled, the original epoprostenol trial reported reduced mortality [1]. Prostacyclin facilitates smooth muscle relaxation, improves endothelial barrier function, and mitigates pro-proliferative and fibrotic signalling pathways [2]. However, continuous intravenous administration imposes a substantial treatment burden, including catheter-related complications and adverse effects [3, 4].

Shareable abstract

∼20% of patients with haemodynamically severe PAH achieved near-normal haemodynamics and safely de-escalated parenteral prostacyclin. Early intensive treatment may be key to maintaining near-normal haemodynamics after prostacyclin discontinuation. https://bit.ly/3R6iqmc

To the Editor:

Parenteral prostacyclins including epoprostenol or treprostinil are cornerstone therapies for haemodynamically severe pulmonary arterial hypertension (PAH). Although not double-blind or placebo-controlled, the original epoprostenol trial reported reduced mortality [1]. Prostacyclin facilitates smooth muscle relaxation, improves endothelial barrier function, and mitigates pro-proliferative and fibrotic signalling pathways [2]. However, continuous intravenous administration imposes a substantial treatment burden, including catheter-related complications and adverse effects [3, 4]. Given these challenges, parenteral prostacyclin de-escalation is desirable when clinically feasible. Previous studies of prostacyclin withdrawal have reported heterogeneous results; however, successful discontinuation has been described in patients with near-normal haemodynamics [5–9].

We hypothesised that early and intensive treatment with parenteral prostacyclin may allow a subset of patients with severe PAH to achieve near-normal haemodynamics, thereby enabling subsequent treatment de-escalation. This study aimed to describe haemodynamic changes and treatment trajectories in patients who achieved near-normal haemodynamics and underwent parenteral prostacyclin de-escalation, and to compare patient characteristics with and without treatment de-escalation in a single-centre cohort.

In this retrospective, single-centre cohort study, we included consecutive patients with PAH who were referred to our centre and required parenteral prostacyclin between 1999 and 2023. Parenteral prostacyclin was initiated in patients with a mean pulmonary arterial pressure (mPAP) ≥50 mmHg, World Health Organization functional class III or IV, or an inadequate response to oral combination therapy [10]. In general, prostacyclin was up-titrated to 20 ng·kg⁻¹·min⁻¹ within 3 weeks and 40 ng·kg⁻¹·min⁻¹ within 3 months. Thereafter, the titration rate was reduced to 2–4 ng·kg⁻¹·min⁻¹ per month. Post-partum pulmonary hypertension (PPPHT) was defined as idiopathic or heritable PAH diagnosed within 6 months of childbirth [11]. Patients receiving stable parenteral prostacyclin were considered eligible for de-escalation if near-normal haemodynamics (mPAP <25 mmHg and cardiac index >2.2 L·min⁻¹·m⁻²) were confirmed on repeated right heart catheterisation performed ≥3 months apart, without treatment escalation [5, 8]. Prostacyclin was gradually tapered until discontinuation (2–4 ng·kg⁻¹·min⁻¹ per month). All patients on parenteral prostacyclin underwent serial right heart catheterisation at regular intervals as part of routine mPAP-guided follow-up; therefore, eligibility for de-escalation was evaluated systematically across the whole cohort. Variables are presented as numbers (%), medians (interquartile range (IQR)), or mean±sd. Comparisons were performed using the Chi-squared test, Mann–Whitney U-test, or paired t-test. Statistical analyses were performed using R version 4.2.0 (R Foundation for Statistical Computing, Vienna, Austria).

Between 1999 and 2023, parenteral prostacyclin therapy was initiated in 112 patients. After excluding patients who initiated prostacyclin therapy at other hospitals (n=4), were lost to follow-up (n=2), had pulmonary veno-occlusive disease (n=3), or discontinued prostacyclin following liver transplantation or atrial septal defect closure (n=2), 101 patients were included in the analysis. The mean age was 34.1±14.1 years, and 83 (82%) patients were female. Aetiologies included idiopathic PAH (n=47), heritable PAH (n=15), connective tissue disease-associated PAH (n=23), congenital heart disease-associated PAH (n=8), and other forms of pulmonary hypertension (n=8). 10 patients had PPPHT, of whom seven had idiopathic PAH, and three had heritable PAH. The median (IQR) interval from the time of first physician visit to parenteral prostacyclin initiation was 15.0 (2.4–72.4) months. At baseline, the patients presented with haemodynamically severe PAH, with a median (IQR) mPAP of 55 (49–68) mmHg, pulmonary vascular resistance of 14 (10–19) Wood Units, and cardiac index of 2.4 (1.0–3.0) L·min⁻¹·m⁻².

Among the 101 patients, 22 (22%) achieved near-normal haemodynamics and met the pre-specified criteria for prostacyclin de-escalation. All eligible patients opted to be carefully weaned from prostacyclin after a comprehensive discussion with the attending physician. De-escalation was initiated after a mean of 4.1±3.7 years, with 13 (59%) patients undergoing de-escalation after 2 years of treatment. All patients successfully discontinued therapy after a mean of 7.7±5.4 years of treatment without adverse events. Baseline characteristics were broadly similar between patients with and without de-escalation, including age, sex, aetiology and haemodynamics, except for a higher prevalence of PPPHT (seven (31.8%) versus three (3.8%), p=0.001) and a shorter interval from the time of first physician visit to parenteral prostacyclin initiation (median 2.4 (IQR 1.2–16.1) versus 28.7 (3.8–79.3) months, p=0.011) in the de-escalation group. Patients who were successfully de-escalated received lower peak prostacyclin doses (median 52 (IQR 41–74) versus 90 (60–132) ng·kg⁻¹·min⁻¹, p=0.003). Parenteral prostacyclin initiation in 2015 or later was observed in nine (41%) patients, of whom six (67%) received triple therapy at baseline. At the time of prostacyclin de-escalation, most patients remained on dual or triple oral PAH therapy (n=6 and n=15, respectively). At the final follow-up, 12 and seven patients received dual and triple therapies, respectively (figure 1). Using the European Society of Cardiology/European Respiratory Society risk stratification model, patients shifted from intermediate- and high-risk at baseline (n=17 and n=5, respectively) to low- and intermediate–low-risk at the final follow-up (n=20 and n=2, respectively).

FIGURE 1.

Treatment and haemodynamic changes over time in patients with parenteral prostacyclin de-escalation. a) Sankey diagram illustrating changes in treatment over time. b–d) Changes in mean pulmonary artery pressure (mPAP), pulmonary vascular resistance (PVR), and cardiac index from baseline to the time of prostacyclin de-escalation and at final follow-up in patients who underwent de-escalation. Haemodynamic improvements achieved at the time of de-escalation were sustained over a mean follow-up of 5 years. IQR: interquartile range.

Haemodynamics improved substantially from baseline to the time of de-escalation and remained stable during long-term follow-up (figure 1). The mean duration from prostacyclin discontinuation to final follow-up was 5.1±2.8 years, with 14 (64%) patients achieving normal haemodynamics (mPAP <21 mmHg) at final assessment. Among the 22 patients who achieved near-normal haemodynamics and underwent prostacyclin de-escalation, no deaths or lung transplantations occurred after discontinuation.

In this single-centre cohort of patients with haemodynamically severe PAH, near-normal haemodynamics, and subsequent prostacyclin de-escalation were observed in a clinically relevant subset. All patients who underwent successful de-escalation remained haemodynamically stable during long-term follow-up (>5 years).

In the present study, patients who underwent successful de-escalation tended to receive parenteral prostacyclin earlier during the disease course, and PPPHT was more frequently observed. These observations suggest that early diagnosis and timely initiation of intensive therapy may facilitate haemodynamic recovery, thereby enabling subsequent treatment de-escalation. Notably, patients who underwent successful prostacyclin de-escalation were treated with lower peak prostacyclin doses than those without de-escalation, which may reflect enhanced treatment responsiveness when therapy was initiated earlier in the disease course. This finding may have important clinical implications, as lower prostacyclin doses could reduce the treatment burden and healthcare costs and improve quality of life. In addition, most patients remained on dual or triple oral PAH therapy at the time of prostacyclin de-escalation and final follow-up, indicating that de-escalation occurred within the context of sustained multidrug therapy. Emerging and less burdensome prostacyclin pathway therapies such as treprostinil palmitil inhalation powder may provide additional opportunities for treatment de-escalation in selected patients [12].

This study had several limitations. First, the favourable outcomes in the de-escalation group should not be interpreted as a survival benefit of de-escalation; because de-escalation required survival to the point of confirmed near-normal haemodynamics, any comparison of outcomes between groups is inherently affected by immortal-time bias. Secondly, PPPHT was overrepresented in the de-escalation group; although near-normal haemodynamics and successful de-escalation were also observed in 15 non-PPPHT patients, we cannot exclude that the post-partum phenotype contributed to this association, which requires confirmation in larger cohorts. Thirdly, the study spans 1999–2023, during which diagnostic criteria, risk-assessment tools, and the availability of oral combination therapy evolved substantially. Lastly. although the prevalence of Bone Morphogenetic Protein Receptor Type 2 mutations appears to be similar between Japanese and Western PAH cohorts, differences in patient characteristics and treatment practices may limit the generalisability of our findings to other regions [13].

In this single-centre, retrospective cohort, a subset of patients with haemodynamically severe PAH achieved near-normal haemodynamics and discontinued parenteral prostacyclin. These hypothesis-generating findings suggest that an early and intensive treatment strategy may facilitate haemodynamic recovery and subsequent de-escalation in selected patients. De-escalation was offered only to patients with confirmed near-normal haemodynamics, and its value in patients who did not reach this threshold remains untested; prospective, multicentre validation is warranted.