By Michael H. Crawford, MD, Editor
Synopsis: A large French national registry observational study of the use of anticoagulants in patients with pulmonary artery hypertension of diverse etiologies enrolled over 11 years, and a meta-analysis of this study and four other national registry studies, showed that there was no association between anticoagulant use and survival.
Source: Bertoletti L, Escal J, Boucly A, et al. Association between anticoagulant therapy and survival in pulmonary arterial hypertension: A registry report and updated meta-analysis. J Am Coll Cardiol. 2025;86(13):982-995.
Oral anticoagulant (AC) use in patients with pulmonary arterial hypertension (PAH) is common but the association of such use with survival is unclear. Thus, these investigators designed a study to assess the association between ACs from the time of PAH diagnosis and survival in the French national PAH registry and to update a meta-analysis of other prospective national registries, including theirs.
Between 2009 and 2020, PAH patients from 17 French centers with a minimum follow-up of three years in those who survived were enrolled. Diagnostic criteria for PAH were mean pulmonary artery pressure (mPAP) ≥ 25 mmHg, pulmonary capillary wedge pressure (PCWP) ≤ 15 mmHg, and pulmonary vascular resistance (PVR) > 3 Wood units. Three patient etiology subgroups were identified: idiopathic, hereditary, or anorexigen use-associated; connective tissue disease-related; and repaired congenital heart disease-related PAH. Patients with other causes of PAH were excluded. The primary endpoint was survival.
After excluding patients with missing data, 1,597 patients were included from the French registry. Of these, 380 were taking ACs and their median survival was 5.6 years. The median survival of the 1,598 without AC use was 5.3 years (hazard ratio [HR], 0.97; 95% confidence interval [CI], 0.84 to 1.14; P = 0.74). After propensity score adjustment to balance patient characteristics, there were 1,591 subjects in the pseudo-AC group and 1,598 subjects in the no-AC group. Survival in the pseudo-AC group was 5.6 years and was 5.4 years in the no-AC group (HR, 0.997; 95% CI, 0.84 to 1.18; P = 0.97). These findings were consistent across all the etiologic subgroups of PAH. The updated meta-analysis of their study plus four other national observational studies, which included 2,337 patients, also showed no significant association between AC use and survival overall (HR, 0.99; 95% CI, 0.79 to 1.25) nor in any etiologic subgroup. The authors concluded that ACs had no significant effect on survival in their national observational study of patients with PAH and that these results were supported by an updated meta-analysis of their study and four other national registries.
Commentary
AC use in PAH has long been advised based on the observation of thrombi in the pulmonary arteries at autopsy and early retrospective studies in the 1980s, which showed lower mortality with AC use. As the pathophysiology of PAH became better understood and more targeted therapies were developed, the rationale for ACs became weaker, especially considering the risk of bleeding. In 2009, guidelines considered ACs a class IIa, level of evidence C recommendation; by 2015 it was IIb-C; and the 2022 guidelines suggested that ACs be considered on a case-by-case basis. The results of this national registry study and meta-analysis of five national registry studies support this recommendation.
A post-hoc analysis of 312 patients in the French registry study showed that 32 (10%) had a major bleeding event and 24 (75%) of them were taking ACs. So, bleeding risk is a serious concern. Thus, the authors recommended considering anticoagulants in those with strong indications for them, such as venous thrombosis, thromboembolism, and atrial fibrillation. When such patients were excluded, there continued to be no association with ACs and survival (HR, 0.96; 95% CI, 0.80 to 1.16; P = 0.71).
The principal strength of this observational registry study is the use of propensity score-based analysis incorporating 13 variables, which likely led to better control of confounders. Also, the bonus of an updated meta-analysis of similar national registry studies that supports their conclusions is reassuring.
There were limitations to consider. AC use was only determined at study entry and termination, but 85% still were taking ACs at the end of the study. The authors did not specify which AC agents were used or their doses. This may be more of an issue with the meta-analysis, since different countries could have different AC practices. Also, it cannot be extrapolated to other types of PAH, which the researchers excluded. In addition, the investigators excluded patients with incomplete data, which introduces a selection bias. Finally, this was not a randomized controlled trial, and the discriminative ability was modest: an area under the curve of 0.67 (> 0.70 is considered strong).
Michael H. Crawford, MD, is Professor Emeritus of Medicine and Consulting Cardiologist, UCSF Health, San Francisco.
