ABSTRACT
The role of antithrombotic therapy in pulmonary hypertension (PH) remains uncertain, with conflicting adult data and limited pediatric evidence. This study aimed to describe real‐world antithrombotic use in pediatric PH utilizing Pediatric Pulmonary Hypertension Network (PPHNet) registry data and a survey of PPHNet centers. We analyzed data of 1,716 patients from the PPHNet registry (2014–2024), comparing characteristics of those who received aspirin or warfarin with those who did not. A survey on real‐world antithrombotic use was also distributed to 12 North American PPHNet centers in 2023. Eighteen percent of subjects received aspirin and 7% received warfarin. Patients who received aspirin or warfarin were more likely to have idiopathic/heritable PH. Aspirin or warfarin‐treated patients demonstrated higher‐risk features as evidenced by use of PH‐targeted triple therapy (aspirin vs. no aspirin: 22% vs. 9%, p < 0.01; warfarin vs. no warfarin: 31% vs. 10%; p < 0.01) and higher mortality over a median follow‐up of 3.2 years (aspirin vs. no aspirin: 14% vs. 9%, p = 0.008; warfarin vs. no warfarin: 19% vs. 9%, p = 0.001). Center‐specific aspirin and warfarin use varied widely. Survey data demonstrated general agreement that antithrombotic therapy was not indicated for patients at lowest risk and probably indicated for patients at highest risk, but management of intermediate‐risk patients varied considerably. Antithrombotic therapy is used infrequently in pediatric PH and primarily reserved for high‐risk patients, with significant variation across centers. These findings highlight the need for further research to define which patients may benefit and guide evidence‐based practice.
Keywords: anticoagulation, antiplatelet, pediatric, pulmonary vascular disease
1.
Pulmonary hypertension (PH) is a rare disorder marked by elevated pulmonary artery pressures, which can lead to right ventricular dysfunction, diminished quality of life, and significant morbidity and mortality [1]. Pediatric PH shares common features with adult PH, but unique characteristics and a smaller evidence base often require a customized approach to management, with extrapolation from studies in adults [2]. While targeted PH medications remain central to the treatment of PH, comprehensive management, including thoughtful consideration of antithrombotic therapy, is essential.
In the context of PH, antithrombotic therapies generally refer to either anticoagulants or antiplatelet agents. Although use of anticoagulation was previously common in patients with World Symposium on Pulmonary Hypertension (WSPH) Group 1 pulmonary arterial hypertension (PAH), clinical experience and evidence suggest its routine use has declined in recent years [3, 4]. There are no pediatric‐specific data to inform whether routine anticoagulation is beneficial for pediatric PAH patients. Evidence from randomized controlled trials in adult PAH remains lacking as well, and the current body of literature, largely composed of observational studies, registry data, and meta‐analyses has produced inconsistent results regarding efficacy [3, 5, 6, 7, 8, 9, 10]. Current adult PAH guidelines do not recommend routine anticoagulation, however, they emphasize that decision making should be individualized and that anticoagulation could be appropriate in the context of specific risk factors [1, 11].
Due to the lack of pediatric‐specific literature on the use of antithrombotic therapy in PH, pediatric recommendations are extrapolated from adult data [2, 12]. According to the WSPH 2024 pediatric PH treatment recommendations, anticoagulant therapies and antiplatelet agents are generally not indicated in pediatric patients unless there are specific concerns for thromboembolic complications or a known history of thrombosis [2]. However, there is no guidance on whether patients with certain clinical features may be at increased risk of thrombosis or whether those patients may benefit from antithrombotic therapy. Given the unique anatomic, physiological and clinical challenges in the pediatric population, the lack of guidance remains a significant gap in care.
We present the first real‐world data on the use of antithrombotic therapy in pediatric PH, drawing from two key sources: an analysis of the PPHNet (Pediatric Pulmonary Hypertension Network) Registry and a survey of current clinical practices across PPHNet centers. Given the absence of strong guidelines and evidence, we hypothesized that the use of antithrombotic therapy in pediatric PH is uncommon, limited primarily to patients with highest risk features, and is highly variable among centers.
2. Methods
2.1. PPHNet Registry
The PPHNet Registry database used for this report is comprised of comprehensive clinical data from 12 academic pediatric PH centers in the United States and Canada that enrolled patients and submitted associated follow‐up information between October 2014 and July 2024. The study protocol was approved by the Institutional review boards/Ethics Committee of all participating centers, and all study subjects, parents, and/or guardians provided informed consent prior to entry of patient information into the registry. Assent was obtained from patients in accordance with institutional guidelines. The PPHNet Registry includes WSPH groups 1–5 newborns, infants, children, and adolescents diagnosed with PH by 18 years of age, as defined by the joint American Heart Association/American Thoracic Society (AHA/ATS) guidelines for pediatric PH [12].
2.2. PPHNet Registry Data Collection and Analysis
This study included PPHNet Registry subjects with a completed medication form indicating whether the subject received antiplatelet therapy or anticoagulation. Patient data relevant to the study, including demographics, PH diagnosis, risk factors, PH medication regimen, antithrombotic therapy, markers of disease severity and hemodynamics from right heart catheterization (RHC), were used in the analysis.
Antithrombotic therapy data were collected systematically for warfarin, aspirin, and clopidogrel only. Other antithrombotic agents, including direct oral anticoagulants (DOAC), were not recorded consistently in the registry case report forms and therefore were not analyzed in this study. To evaluate antithrombotic therapy practices, two separate analyses were performed. First, patients were grouped based on aspirin use: those who received aspirin at any time from PH diagnosis through the end of the study period (aspirin+ group) and those who did not (aspirin− group). A second analysis classified patients based on warfarin use: those who received warfarin at any point from PH diagnosis through the end of the study period (warfarin+ group) and those who did not (warfarin− group). A separate analysis of clopidogrel was not completed due to low numbers of patients receiving clopidogrel (n = 4) [13].
For patients in the aspirin− and warfarin− groups, clinical characteristics were extracted from the enrollment visit. For patients in the aspirin+ and the warfarin+ groups, clinical characteristics were extracted from the study visit during which the respective medication was first reported. Hospitalization data (PH‐related hospitalization, hemoptysis) were reported if they occurred at any time for all groups. PH medications were included only for patients in the aspirin− and warfarin‐ groups who had a completed PH Medications form at enrollment and for patients in the aspirin+ group and warfarin+ group with a PH Medications form completed at the visit where aspirin or warfarin was reported, respectively. PH medications that were tabulated included calcium channel blockers (CCB), phosphodiesterase‐5‐inhibitors (PDE‐5i), endothelin receptor antagonists (ERA), soluble guanylate cyclase (sGC) stimulators, prostacyclins, and inhaled nitric oxide. Monotherapy was defined as the use of one of these six agents, dual therapy was defined as the use of two of these agents, and triple therapy was defined as the use of three of these agents. For the aspirin− and warfarin− groups, catheterization data closest to enrollment were utilized (allowing that the catheterization itself could have been performed at or before the enrollment visit). For the aspirin+ and warfarin+ groups, catheterization data at the first visit where aspirin or warfarin was reported, respectively, were utilized if available. RV systolic pressure (RVSP) elevation severity was calculated based on systemic arterial systolic pressure and RVSP, when both data points were available from RHC.
2.3. Statistical Analysis
Registry data were tabulated as mean ± standard deviation (SD), median (interquartile range) or frequency and percentage, as appropriate for the distribution of the data. Student's t‐test, the Wilcoxon rank sum test and Fisher exact tests were used, as appropriate, to compare demographic, diagnostic, and treatment factors between subjects who received the antithrombotic medication in question and those who did not. Statistical significance was defined as a p < 0.05. The analysis was performed using SAS 9.4 (SAS institute, Inc., Cary, NC).
2.4. PPHNet Center Antithrombotic Practice Survey
To understand practice patterns regarding antithrombotic therapy in pediatric patients with PH, we developed a survey of pediatric PH providers and distributed it to the 12 member‐centers of the PPHNet (Supplement). The survey was conducted, and data were stored in a RedCap database hosted at the University of Colorado. The survey was designed to evaluate clinical use of antithrombotic therapy, incorporating scenarios based on varying levels of PH severity and associated risk factors such as RV dysfunction, intracardiac shunts, and suprasystemic pressures. The full list of questions distributed can be found in the Supplement. The survey was conducted between December 2022 and March 2023.
Survey responses were analyzed using descriptive statistics, and antithrombotic regimen choices were compared across clinical scenarios to identify patterns and variability in practice.
3. Results
3.1. PPHNet Registry Analysis
3.1.1. Overall Cohort
Of the 1807 patients who were enrolled in the PPHNet Registry between October 2014 and June 2024, medication data were available in 1,716 patients (95%), comprising the analytic cohort (Tables 1 and 3). Among these, 46% had a primary classification of WSPH Group 1% and 47% as Group 3. The median follow‐up following PH diagnosis was 6.0 years (IQR, 2.7–9.7) and the median follow‐up following enrollment was 3.2 years (IQR, 1.2–6.1). After attrition due to missing medication forms at certain time points, 1688 subjects and 1685 of the 1716 patients were included for the PH medications portion of the analysis for the aspirin and warfarin groups, respectively (Tables 1 and 3). Regarding PH therapy, 36% of patients were not on targeted treatment, 36% received monotherapy, 17% dual therapy, and 11% triple therapy. Overall, 18% of subjects (n = 308) received aspirin and 7% (n = 120) received warfarin. Qualifying cardiac catheterization data were available for 385 subjects in the aspirin analysis and 384 for warfarin. Hemodynamic parameters for patients with available cardiac catheterization data are shown in Tables 2 and 4, with the overall cohorts for each comparison demonstrating moderate PH. During the follow‐up period, of the 1,716 patients, 10% died, 2% underwent lung transplantation, and 1% received a Potts shunt.
Table 1.
Demographic and clinical characteristics: aspirin+ group versus aspirin− group.
| Variable | Overall Cohort (n = 1716) | Aspirin+ (n = 308) | Aspirin− (n = 1408) | p |
|---|---|---|---|---|
| Demographics | ||||
| Male | 865 (50.4%) | 152 (49.4%) | 713 (50.6%) | 0.706 |
| Race | 0.398 | |||
| White | 1010 (58.9%) | 188 (61.0%) | 822 (58.4%) | |
| Black | 215 (12.5%) | 28 (13.0%) | 187 (13.3%) | |
| Asian | 164 (9.6%) | 31 (9.1%) | 133 (9.4%) | |
| Other | 134 (7.8%) | 25 (8.1%) | 109 (7.7%) | |
| Unknown | 193 (11.2%) | 36 (11.7%) | 157 (11.2%) | |
| Hispanic ethnicity | 0.794 | |||
| Yes | 284 (16.6%) | 51 (16.6%) | 233 (16.5%) | |
| No | 1274 (74.2%) | 232 (75.3%) | 1042 (74.0%) | |
| Unknown | 158 (9.2%) | 25 (8.1%) | 133 (9.4%) | |
| Diagnoses | ||||
| Age at PH diagnosis, median (IQR) | 0.5 (0.1, 3.7) | 1.9 (0.3, 6.4) | 0.4 (0.1, 2.9) | < 0.001 |
| Primary WSPH group | < 0.001 | |||
| Group 1 | 793 (46.2%) | 214 (69.5%) | 579 (41.1%) | |
| 1.1 IPAH | 186 (10.8%) | 49 (15.9%) | 137 (9.7%) | 0.002 |
| 1.2 HPAH | 60 (3.5%) | 18 (5.8%) | 42 (3.0%) | 0.024 |
| 1.4.4 CHD‐PAH | 448 (26.1%) | 131 (42.5%) | 317 (22.5%) | < 0.001 |
| 1.4.4 A Eisenmenger syndrome | 27 (1.6%) | 7 (2.3%) | 20 (1.4%) | 0.308 |
| Group 2 | 68 (4.0%) | 31 (10.1%) | 37 (2.6%) | |
| Group 3 | 804 (46.9%) | 53 (17.2%) | 751 (53.3%) | |
| 3.7.1 BPD‐PAH | 349 (20.3%) | 26 (8.4%) | 323 (22.9%) | < 0.001 |
| 3.7.3 CDH‐PAH | 277 (16.1%) | 16 (5.2%) | 261 (18.5%) | < 0.001 |
| Group 4 | 10 (0.6%) | 0 (0%) | 10 (0.7%) | |
| 4.1 CTEPH | 10 (0.6%) | 0 (0%) | 10 (0.7%) | 0.224 |
| Group 5 | 41 (2.4%) | 10 (3.2%) | 31 (2.2%) | |
| Additional variables | ||||
| History of clotting disorder | 32 (1.9%) | 12 (3.9%) | 20 (1.4%) | 0.009 |
| Trisomy 21 | 184 (10.9%) | 26 (8.5%) | 158 (11.4%) | 0.156 |
| Progressive deterioration of right heart failure | 42 (2.4%) | 14 (4.5%) | 28 (2.0%) | 0.014 |
| WHO Functional Class, III/IV | 385 (22.4%) | 91 (29.5%) | 294 (20.9%) | < 0.001 |
| Pulmonary vein stenosis | 85 (5.0%) | 30 (9.7%) | 55 (3.9%) | < 0.001 |
| IV prostacyclin therapy | 75 (4.4%) | 18 (5.8%) | 57 (4.0%) | 0.167 |
| Bidirectional intracardiac shunt | 51 (3.0%) | 15 (4.9%) | 36 (2.6%) | 0.04 |
| Warfarin ever | 120 (7.0%) | 40 (13.0%) | 80 (5.7%) | < 0.001 |
| RV FAC (Mean ± SD) | 40.2 ± 11.5 (n = 229) | 37.3 ± 13.0 (n = 42) | 40.9 ± 11.1 (n = 187) | 0.071 |
| Hospitalization at any time | 1226 (71.5%) | 235 (76.3%) | 991 (70.4%) | 0.037 |
| PH‐related hospitalization | 927 (54.0%) | 192 (62.3%) | 735 (52.2%) | 0.001 |
| Hospitalization with hemoptysis | 18 (1.0%) | 5 (1.6%) | 13 (0.9%) | 0.348 |
| Outcomes | ||||
| Lung transplant | 29 (1.7%) | 11 (3.6%) | 18 (1.3%) | 0.011 |
| Potts shunt | 14 (0.8%) | 6 (1.9%) | 8 (0.6%) | 0.026 |
| Death | 167 (9.7%) | 43 (14.0%) | 124 (8.8%) | 0.008 |
| PH medications | Overall Cohort (n = 1688) | Aspirin + (n = 307) | Aspirin‐ (n = 1381) | < 0.001 |
| None | 606 (35.9%) | 52 (16.9%) | 554 (40.1%) | < 0.01 |
| Monotherapy | 599 (35.5%) | 116 (37.8%) | 483 (35.0%) | 0.352 |
| Dual therapy | 289 (17.1%) | 71 (23.1%) | 218 (15.8%) | < 0.01 |
| ≥ Triple therapy | 194 (11.5%) | 68 (22.1%) | 126 (9.1%) | < 0.01 |
| ≥ Triple with inhaled prostacyclin | 32 (1.9%) | 6 (2.0%) | 26 (1.9%) | 1 |
| ≥ Triple with oral prostacyclin or analog | 13 (0.8%) | 6 (2.0%) | 7 (0.5%) | 0.019 |
| ≥ Triple with SQ or IV prostacyclin | 123 (7.3%) | 47 (15.3%) | 76 (5.5%) | < 0.001 |
Note: Data are shown as n(%) unless otherwise specified. p values compare the aspirin+ group to the aspirin− group for each variable. Bold values are statistically significant p values < 0.05.
Abbreviations: BPD‐PAH, bronchopulmonary dysplasia associated pulmonary arterial hypertension; CDH‐PAH, congenital diaphragmatic hernia‐associated pulmonary hypertension; CHD‐PAH, congenital heart disease associated pulmonary arterial hypertension; CTEPH, chronic thromboembolic pulmonary hypertension; HPAH, heritable pulmonary arterial hypertension; IPAH, idiopathic pulmonary arterial hypertension; IV, intravenous; PH, pulmonary hypertension; RV FAC, right ventricular fractional area change; SQ, subcutaneous; WHO, World Health Organization; WSPH, World Symposium on Pulmonary Hypertension.
Table 3.
Demographic and clinical characteristics: warfarin+ group versus warfarin− group.
| Variable | Overall Cohort (n = 1716) | Warfarin+ (n = 120) | Warfarin− (n = 1596) | p |
|---|---|---|---|---|
| Demographics | ||||
| Male | 865 (50.4%) | 44 (36.7%) | 821 (51.5%) | 0.002 |
| Race | 0.373 | |||
| White | 1010 (58.9%) | 79 (65.8%) | 931 (58.3%) | |
| Black | 215 (12.5%) | 16 (13.3%) | 199 (12.5%) | |
| Asian | 164 (9.6%) | 10 (8.3%) | 154 (9.6%) | |
| Other | 134 (7.8%) | 6 (5.0%) | 128 (8.0%) | |
| Unknown | 193 (11.2%) | 9 (7.5%) | 184 (11.5%) | |
| Hispanic ethnicity | 0.999 | |||
| Yes | 284 (16.6%) | 20 (16.7%) | 264 (16.5%) | |
| No | 1274 (74.2%) | 89 (74.2%) | 1185 (74.2%) | |
| Unknown | 158 (9.2%) | 11 (9.2%) | 147 (9.2%) | |
| Diagnosis | ||||
| Age at PH diagnosis, median (IQR) | 0.5 (0.1, 3.7) | 6.3 (2.9, 11.7) | 0.4 (0.1, 2.9) | < 0.001 |
| Primary WSPH group | < 0.001 | |||
| Group 1 | 793 (46.2%) | 94 (78.3%) | 699 (43.8%) | |
| 1.1 IPAH | 186 (10.8%) | 44 (36.7%) | 142 (8.9%) | < 0.001 |
| 1.2 HPAH | 60 (3.5%) | 8 (6.7%) | 52 (3.3%) | 0.065 |
| 1.4.4 CHD‐PAH | 448 (26.1%) | 39 (32.5%) | 409 (25.6%) | 0.106 |
| 1.4.4 A Eisenmenger syndrome | 27 (1.6%) | 5 (4.2%) | 22 (1.4%) | 0.036 |
| Group 2 | 68 (4.0%) | 9 (7.5%) | 59 (3.7%) | |
| Group 3 | 804 (46.9%) | 9 (7.5%) | 795 (49.8%) | |
| 3.7.1 BPD‐PAH | 349 (20.3%) | 3 (2.5%) | 346 (21.7%) | < 0.001 |
| 3.7.3 CDH‐PAH | 277 (16.1%) | 3 (2.5%) | 274 (17.2%) | < 0.001 |
| Group 4 | 10 (0.6%) | 5 (4.2%) | 5 (0.3%) | |
| 4.1 CTEPH | 10 (0.6%) | 5 (4.2%) | 5 (0.3%) | < 0.001 |
| Group 5 | 41 (2.4%) | 3 (2.5%) | 38 (2.4%) | |
| Additional variables | ||||
| History of clotting disorder | 32 (1.9%) | 2 (1.7%) | 30 (1.9%) | 1 |
| Trisomy 21 | 184 (10.9%) | 2 (1.7%) | 182 (11.6%) | < 0.001 |
| Progressive deterioration of right heart failure | 42 (2.4%) | 5 (4.2%) | 37 (2.3%) | 0.211 |
| Pediatric Functional Class, III/IV or WHO Functional Class, III/IV | 378 (22.0%) | 32 (26.7%) | 346 (21.7%) | 0.209 |
| Pulmonary vein stenosis | 85 (5.0%) | 2 (1.7%) | 83 (5.2%) | 0.122 |
| IV prostacyclin therapy | 78 (4.5%) | 28 (23.3%) | 50 (3.1%) | < 0.001 |
| Bidirectional intracardiac shunt | 51 (3.0%) | 1 (0.8%) | 50 (3.1%) | 0.257 |
| Aspirin ever | 308 (17.9%) | 40 (33.3%) | 268 (16.8%) | < 0.001 |
| RV FAC | 40.2 ± 11.5 (n = 229) | 30.8 ± 7.7 (n = 13) | 40.8 ± 11.3 (n = 216) | 0.002 |
| Hospitalization at any time | 1226 (71.5%) | 87 (72.5%) | 1139 (71.4%) | 0.835 |
| PH‐related hospitalization | 927 (54.0%) | 75 (62.5%) | 852 (53.4%) | 0.058 |
| Hospitalization with hemoptysis | 18 (1.0%) | 5 (4.2%) | 13 (0.8%) | 0.006 |
| Outcomes | ||||
| Lung transplant | 29 (1.7%) | 7 (5.8%) | 22 (1.4%) | 0.003 |
| Potts shunt | 14 (0.8%) | 6 (5.0%) | 8 (0.5%) | < 0.001 |
| Death | 167 (9.7%) | 23 (19.2%) | 144 (9.0%) | 0.001 |
| PH medications | Overall (n = 1685) | Warfarin+ (n = 120) | Warfarin− (n = 1565) | < 0.001 |
| None | 607 (36.0%) | 8 (6.7%) | 599 (38.3%) | < 0.01 |
| Monotherapy | 599 (35.5%) | 37 (30.8%) | 562 (35.9%) | 0.263 |
| Dual therapy | 290 (17.2%) | 38 (31.7%) | 252 (16.1%) | < 0.01 |
| ≥ Triple therapy | 189 (11.2%) | 37 (30.8%) | 152 (9.7%) | < 0.01 |
| ≥ Triple with inhaled prostacyclin | 36 (2.1%) | 10 (8.3%) | 26 (1.7%) | < 0.001 |
| ≥ Triple with oral prostacyclin or analog | 12 (0.7%) | 6 (5.0%) | 6 (0.4%) | < 0.001 |
| ≥ Triple with SQ or IV prostacyclin | 116 (6.9%) | 21 (17.5%) | 95 (6.1%) | < 0.001 |
Note: Data are shown as n (%) unless otherwise specified. p values compare the warfarin+ group to the warfarin‐ group for each variable. Bold values are statistically significant p values < 0.05.
Abbreviations: BPD‐PAH, bronchopulmonary dysplasia associated pulmonary arterial hypertension; CDH‐PAH, congenital diaphragmatic hernia‐associated pulmonary hypertension; CHD‐PAH, congenital heart disease associated pulmonary arterial hypertension; CTEPH, chronic thromboembolic pulmonary hypertension; HPAH, heritable pulmonary arterial hypertension; IPAH, idiopathic pulmonary arterial hypertension; IV, intravenous; PH, pulmonary hypertension; RV FAC, right ventricular fractional area change; SQ, subcutaneous; WHO, World Health Organization; WSPH, World Symposium on Pulmonary Hypertension.
Table 2.
Catheterization findings: aspirin+ group versus aspirin− group.
| Variable | Overall Cohort (n = 385) | Aspirin+ (n = 128) | Aspirin− (n = 257) | p |
|---|---|---|---|---|
| Hemodynamics | ||||
| RAP, mmHg | 7.48 ± 7.24 (n = 339) | 9.09 ± 10.83 (n = 112) | 6.69 ± 4.35 (n = 227) | 0.004 |
| RVSP, mmHg | 62.98 ± 26.02 (n = 227) | 68.28 ± 24.48 (n = 76) | 60.31 ± 26.44 (n = 151) | 0.029 |
| RVEDP, mmHg | 8.86 ± 3.82 (n = 220) | 9.96 ± 4.52 (n = 74) | 8.31 ± 3.28 (n = 146) | 0.002 |
| PA systolic, mmHg | 58.92 ± 25.89 (n = 346) | 63.28 ± 27.18 (n = 116) | 56.73 ± 24.99 (n = 230) | 0.026 |
| PA diastolic, mmHg | 25.79 ± 14.57 (n = 346) | 27.06 ± 16.16 (n = 116) | 25.15 ± 13.69 (n = 230) | 0.251 |
| PA mean, mmHg | 39.55 ± 18.32 (n = 356) | 41.10 ± 19.13 (n = 121) | 38.75 ± 17.87 (n = 235) | 0.253 |
| PCWP, mmHg | 9.89 ± 4.12 (n = 292) | 11.09 ± 4.80 (n = 105) | 9.22 ± 3.51 (n = 187) | < 0.001 |
| Systemic arterial systolic pressure, mmHg | 82.59 ± 18.67 (n = 308) | 81.36 ± 18.35 (n = 108) | 83.26 ± 18.85 (n = 200) | 0.395 |
| Systemic arterial diastolic pressure, mmHg | 46.99 ± 12.54 (n = 308) | 46.44 ± 12.38 (n = 108) | 47.28 ± 12.65 (n = 200) | 0.578 |
| Systemic arterial mean pressure, mmHg | 61.12 ± 14.18 (n = 306) | 60.46 ± 13.60 (n = 107) | 61.48 ± 14.50 (n = 199) | 0.549 |
| mPAP: mSAP | 0.67 ± 0.31 (n = 295) | 0.70 ± 0.31 (n = 104) | 0.65 ± 0.31 (n = 191) | 0.252 |
| PVRi, Woods units X m2 | 9.15 ± 8.61 (n = 321) | 10.27 ± 10.03 (n = 110) | 8.57 ± 7.74 (n = 211) | 0.095 |
| SVRi, Woods units X m2 | 17.17 ± 8.48 (n = 291) | 17.32 ± 9.25 (n = 101) | 17.09 ± 8.07 (n = 190) | 0.829 |
| PVRi: SVRi | 0.65 ± 1.85 (n = 286) | 0.87 ± 3.10 (n = 98) | 0.53 ± 0.45 (n = 188) | 0.140 |
| QP:QS | 1.27 ± 0.75 (n = 252) | 1.14 ± 0.52 (n = 80) | 1.33 ± 0.83 (n = 172) | 0.057 |
| CI, L/min/m2 | 3.55 ± 1.22 (n = 312) | 3.50 ± 1.31 (n = 111) | 3.58 ± 1.17 (n = 201) | 0.568 |
| RVSP elevation severity | Overall (n = 195) | Aspirin + (n = 65) | Aspirin‐ (n = 130) | 0.101 |
| < 50% systemic | 28 (14.4%) | 5 (7.7%) | 23 (17.7%) | |
| 50%–100% systemic | 122 (62.6%) | 41 (63.1%) | 81 (62.3%) | |
| Suprasystemic | 45 (23.1%) | 19 (29.2%) | 26 (20.0%) |
Note: Continuous variables are expressed as mean ± SD. RVSP elevation severity was calculated based on systemic arterial systolic pressure and RVSP, when both data points available. Bold values are statistically significant p values < 0.05.
Abbreviations: CI, cardiac index; mPAP, mean pulmonary artery pressure; mSAP, mean systemic arterial pressure; PA, pulmonary artery; PCWP, pulmonary capillary wedge pressure; PH, pulmonary hypertension; PVRi, pulmonary vascular resistance index; QP:QS, pulmonary‐to‐systemic flow ratio; RAP, right atrial pressure; RVEDP, right ventricular end‐diastolic pressure; RVSP, right ventricular systolic pressure; SVRi, systemic vascular resistance index.
Table 4.
Catheterization findings: warfarin+ group versus warfarin− group.
| Variable | Overall Cohort (n = 384) | Warfarin+ (n = 56) | Warfarin− (n = 328) | p |
|---|---|---|---|---|
| Hemodynamics | ||||
| RAP, mmHg | 7.10 ± 4.35 (n = 341) | 8.02 ± 3.68 (n = 55) | 6.92 ± 4.45 (n = 286) | 0.087 |
| RVSP, mmHg | 64.48 ± 26.48 (n = 227) | 87.52 ± 29.42 (n = 23) | 61.88 ± 24.89 (n = 204) | < 0.001 |
| RVEDP, mmHg | 8.67 ± 3.55(n = 217) | 11.23 ± 3.85 (n = 22) | 8.38 ± 3.41 (n = 195) | < 0.001 |
| PA systolic, mmHg | 60.71 ± 26.77 (n = 344) | 81.06 ± 30.01 (n = 51) | 57.17 ± 24.55 (n = 293) | < 0.001 |
| PA diastolic, mmHg | 27.29 ± 15.45 (n = 344) | 41.53 ± 19.25 (n = 51) | 24.81 ± 13.24 (n = 293) | < 0.001 |
| PA mean, mmHg | 41.56 ± 19.41 (n = 354) | 58.67 ± 22.54 (n = 52) | 38.62 ± 17.22 (n = 302) | < 0.001 |
| PCWP, mmHg | 9.71 ± 3.77 (n = 283) | 10.49 ± 4.32 (n = 49) | 9.55 ± 3.63 (n = 234) | 0.113 |
| Systemic arterial systolic pressure, mmHg | 83.06 ± 17.61 (n = 312) | 90.67 ± 17.84 (n = 51) | 81.57 ± 17.21 (n = 261) | < 0.001 |
| Systemic arterial diastolic pressure, mmHg | 47.32 ± 12.56 (n = 312) | 54.51 ± 13.77 (n = 51) | 45.91 ± 11.83 (n = 261) | < 0.001 |
| Systemic arterial mean pressure, mmHg | 61.54 ± 13.86 (n = 310) | 69.02 ± 13.96 (n = 50) | 60.10 ± 13.40 (n = 260) | < 0.001 |
| mPAP: mSAP | 0.70 ± 0.32 (n = 300) | 0.89 ± 0.34 (n = 49) | 0.66 ± 0.31 (n = 251) | < 0.001 |
| PVRi, Woods units X m2 | 9.66 ± 8.08 (n = 313) | 17.21 ± 10.31 (n = 47) | 8.33 ± 6.82 (n = 266) | < 0.001 |
| SVRi, Woods units X m2 | 17.62 ± 8.31 (n = 285) | 22.12 ± 10.38 (n = 45) | 16.78 ± 7.60 (n = 240) | < 0.001 |
| PVRi: SVRi | 0.57 ± 0.45 (n = 278) | 0.83 ± 0.51 (n = 45) | 0.52 ± 0.42 (n = 233) | < 0.001 |
| QP:QS | 1.29 ± 0.75 (n = 241) | 1.08 ± 0.43 (n = 34) | 1.32 ± 0.79 (n = 207) | 0.087 |
| CI, L/min/m2 | 3.48 ± 1.16 (n = 305) | 3.09 ± 1.02 (n = 46) | 3.55 ± 1.17 (n = 259) | 0.012 |
| RVSP elevation severity | Overall cohort (n = 200) | Warfarin+ (n = 21) | Warfarin− (n = 179) | 0.062 |
| < 50% systemic | 31 (15.5%) | 2 (9.5%) | 29 (16.2%) | |
| 50%–100% systemic | 118 (59.0%) | 9 (42.9%) | 109 (60.9%) | |
| Suprasystemic | 51 (25.5%) | 10 (47.6%) | 41 (22.9%) |
Note: Continuous variables are expressed as mean ± SD. RVSP elevation severity was calculated based on systemic arterial systolic pressure and RVSP, when both data points available. Bold values are statistically significant p values < 0.05.
Abbreviations: CI, cardiac index; mPAP, mean pulmonary artery pressure; mSAP, mean systemic arterial pressure; PA, pulmonary artery; PCWP, pulmonary capillary wedge pressure; PH, pulmonary hypertension; PVRi, pulmonary vascular resistance index; QP:QS, pulmonary‐to‐systemic flow ratio; RAP, right atrial pressure; RVEDP, right ventricular end‐diastolic pressure; RVSP, right ventricular systolic pressure; SVRi, systemic vascular resistance index.
3.2. Aspirin Use
Demographic characteristics for the aspirin+ and aspirin− subjects are presented in Table 1. Among those in the aspirin+ group, median age at aspirin start date was 3.0 (IQR: 0.7, 9.3) years, mean time from PH diagnosis to aspirin initiation was 1.3 ± 3.7 years, and 34% were either on aspirin or started on aspirin at time of diagnosis. Aspirin use varied across primary WSPH groups (p < 0.001), with higher rates seen in WSPH Group 1 patients (27%) and WSPH Group 2 patients (46%) and lower rates in WSPH Group 3 patients (6.6%) and WSPH Group 4 (0%) patients. Compared to aspirin− patients, aspirin+ patients were more likely to have diagnoses of congenital heart disease associated pulmonary arterial hypertension (CHD‐PAH, 43% vs. 23%, p < 0.001), idiopathic pulmonary arterial hypertension (IPAH, 16% vs. 10%, p = 0.002), heritable pulmonary arterial hypertension (HPAH, 6% vs. 3%, p = 0.024), and less likely to have congenital diaphragmatic hernia‐associated pulmonary hypertension (CDH‐PAH, 5% vs. 19%, p < 0.001) or bronchopulmonary dysplasia associated pulmonary arterial hypertension (BPD‐PAH, 8% vs. 23%, p < 0.001).
Hemodynamics from RHC in the aspirin+ (n = 128) and aspirin− (n = 257) groups are compared in Table 2. In the aspirin+ group, the median time from initiation of aspirin to RHC was 22 days (IQR: ‐1, 387). In the aspirin+ group, markers of disease severity generally indicated a higher‐risk cohort, including WHO Functional Class III/IV (30% vs. 21%, p < 0.001) and a greater rate of PH‐related hospitalization (62% vs. 52%, p = 0.001). Additionally, rates of lung transplantation (4% vs. 1%, p = 0.011), Potts shunt (2% vs. 1%, p < 0.001), and death (14% vs. 9%, p = 0.008) were all higher in the aspirin+ group compared with rates in the aspirin− group.
Aspirin use varied by PH medication regimen (p < 0.001). The aspirin+ group was more likely to be on dual therapy (23% v. 16%) or triple therapy (22% vs. 9%) and was less likely to not be receiving PH‐targeted therapy (17% vs. 40%). Aspirin+ patients also more commonly received warfarin compared to aspirin− patients (13% vs. 6%, p < 0.001).
3.3. Warfarin Use
Demographic characteristics for the warfarin+ and warfarin‐ subjects are presented in Table 3. Among the 120 in the warfarin+ group, median age at warfarin start date was 9.0 (IQR 4.8, 14.7) years, mean time from PH diagnosis to warfarin start was 3.2 ± 6.5 years, and 33% were either on warfarin or started on warfarin at the time of diagnosis. Warfarin use varied across primary WSPH groups (p < 0.001), with higher rates seen in WSPH Group 1 patients (12%), WSPH Group 2 patients (13%), and WSPH Group 4 patients (50%) and lower rates in WSPH Group 3 patients (1.1%). Compared with warfarin‐ patients, warfarin+ patients were more likely to have a diagnosis of IPAH (37% vs. 9%, p < 0.001), Eisenmenger syndrome (4% vs. 1%, p = 0.036), and chronic thromboembolic pulmonary hypertension (CTEPH, 4% vs. 0.3%, p < 0.001), and less likely to have CDH‐PAH (3% vs. 17%, p < 0.001) or BPD‐PAH (3% vs. 22%, p = < 0.001).
Hemodynamics from RHC in the warfarin+ (n = 56) and warfarin− (n = 328) group are compared in Table 4. In the warfarin+ group, the median time from initiation of warfarin to RHC was 0 days (IQR: ‐7, 673). Patients that received warfarin had higher mPAP (58.67 ± 22.54 mmHg vs. 38.62 ± 17.22 mmHg, p < 0.001), and lower CI (3.09 ± 1.02 L/min/m2 vs. 3.55 ± 1.17 L/min/m2, p = 0.012) compared to those that did not. Warfarin+ patients also had higher rates of lung transplant (6% vs. 2%, p = 0.003), Potts shunt (5% vs. 0.5%, p < 0.001) and death (19.2% vs. 9.0%, p = 0.001).
Warfarin use also varied by PH medication regimen (p < 0.001). The warfarin+ group was more likely to be on dual therapy (32% vs. 16%, p < 0.01) or triple therapy (31% vs. 10%, p < 0.01) and was less likely to not be receiving PH‐targeted therapy (7% vs. 38%, p < 0.01). Warfarin+ patients also more commonly received aspirin compared with warfarin‐ patients (33% vs. 17%, p < 0.001).
3.4. PPHNet Center Practice Variability
Center practices varied greatly in the use of aspirin and warfarin (Table 5). The percentage of patients who ever received warfarin or aspirin ranged by center from 0% to 14% and from 0% to 28%, respectively. The center variation remained even after adjusting for differing distributions of WSPH groups amongst the centers (p < 0.001 for aspirin, p < 0.001 for warfarin). Generally, centers more likely to have treated patients with aspirin were also more likely to have treated patients with warfarin.
Table 5.
Variability of warfarin use (n = 120) and aspirin use (n = 308) across PPHNet centers (N = 1716).
| Variable | Warfarin+ | p | Aspirin+ | p |
|---|---|---|---|---|
| Center | < 0.001 | 0.008 | ||
| 1 | 3.6% | 14.4% | ||
| 2 | 7.3% | 20.9% | ||
| 3 | 3.0% | 14.0% | ||
| 4 | 4.9% | 19.5% | ||
| 5 | 9.4% | 20.2% | ||
| 6 | 13.9% | 14.5% | ||
| 7 | 0% | 0% | ||
| 8 | 0% | 0% | ||
| 9 | 5.8% | 20.3% | ||
| 10 | 3.8% | 28.2% | ||
| 11 | 2.8% | 18.1% | ||
| 12 | 0% | 0% | ||
| All | 7.0% | 17.9% |
Note: p values reflect comparisons of the prevalence of warfarin use among study centers and the prevalence of aspirin use among study centers. Bold values are statistically significant p values < 0.05.
3.4.1. PPHNet Survey of Centers
Survey responses were submitted by 10 of 12 PPHNet Centers (Figure 1). The degree of agreement varied widely among scenarios. All center responses agreed not to use antithrombotic therapy for patients with sub‐systemic RV pressures and without known thrombosis risk factors. Most centers (80%) reported they would use anticoagulation in patients with all posited risk factors (systemic IPAH, atrial shunt, decreased right ventricular function, and presence of a central line). However, answers varied on use of antiplatelet agents, anticoagulation, or a combination according to the other risk factors surveyed (Figure 1).
Figure 1.
PPHNet Center 2023 Survey Responses on Antithrombotic therapy (n = 10); The figure presents survey responses from 10 of 12 PPHNet Centers regarding their selection of antithrombotic therapy across various risk‐factor scenarios. For some risk factors, total n is more than ten due to centers reporting combination therapy (n = 2 systemic IPAH + Iatrogenic Atrial Shunt) or differing agents based on patient factors (n = 1 for Systemic IPAH + Decreased RV Function); IPAH = idiopathic pulmonary arterial hypertension; RV = right ventricular.
The responses regarding antithrombotic therapy for valved and non‐valved Potts shunts and a transcatheter device in an atrial shunt (i.e flow restrictor or stent in situ) also varied. For valved Potts shunts, three centers reported use of anticoagulation, five reported use of antiplatelet therapy, one reported use of both, and one reported not applicable. For non‐valved Potts shunts, one center reported use of anticoagulation, seven reported use of antiplatelet therapy, one reported use of none, and one reported use of dual antiplatelet therapy initially. The responses for an atrial shunt device included four centers that used dual antiplatelet therapy or anticoagulation plus antiplatelet therapy, four centers that used antiplatelet therapy with aspirin alone, one center that used anticoagulation alone, and one center reported no use of these devices. However, many comments suggested that patient complexity and other patient specific factors may result in variation in practice within a center and the absence of a consistent protocol to guide decision making at each individual center.
In patients for whom anticoagulation was indicated, warfarin was selected as an anticoagulant of choice by all 10 centers. The most common INR goal chosen was 1.5–2, but 1.5–2.5 or 2–3 was also chosen depending on indication and risk factors. Three centers selected additional agents as anticoagulants of choice: one center selected enoxaparin, one center selected rivaroxaban, and one center selected rivaroxaban and apixaban.
4. Discussion
We present the first real‐world data on antithrombotic therapy use in pediatric PH, based on a large North American cohort enrolled in the PPHNet registry, together with survey findings that detail current clinical practices across PPHNet Centers. We found that warfarin use was relatively uncommon (7%) and aspirin use was generally low (18%). However, use varied by PH subgroup, with higher rates in WSPH Group I (aspirin 27%; warfarin 12%) and substantially lower rates observed in WSPH Group 3 patients (aspirin 6.6%; warfarin 1.1%). Both registry and survey data revealed considerable variation in antithrombotic therapy practices across institutions and based on patient‐specific factors, underscoring the need for clearer clinical guidance. Both data sources consistently indicated that pediatric PH patients were more likely to receive antithrombotic therapy when they had clinical features associated with higher‐risk profiles or features that may represent independent indications and were unlikely to receive such therapy in the absence of these risk factors or indications.
Evidence regarding anticoagulation in adult PH remains conflicting, while pediatric‐specific data are scarce, leaving little clinical guidance on patient selection. Prospective and retrospective adult PH studies have demonstrated inconsistent associations between anticoagulation and clinical outcomes [3, 5, 6, 14, 15]. Several systematic reviews and meta‐analyses have similarly yielded conflicting conclusions [4, 8, 9, 10]. Wang et al. analyzed eight observational studies involving adults with IPAH and reported no significant difference in mortality between anticoagulated and non‐anticoagulated groups [8]. In contrast, a large systematic review and meta‐analysis that included 12,512 patients demonstrated a significant mortality reduction among the anticoagulated subjects in the overall PAH cohort and the IPAH subgroup [9]. Two recent meta‐analyses of adults with PAH found that anticoagulation did not improve survival [4, 10]. However, most of these adult studies are specific to WSPH Group I patients and no studies have evaluated antithrombotic use in pediatric PH. Due to limited supporting evidence and the known bleeding risks associated with antithrombotic therapy, current pediatric and adult PH guidelines state that antithrombotic therapy is generally not indicated in the absence of specific thromboembolic risk factors, emphasizing individualized decision‐making [1, 2, 11]. However, strong clinical guidance on patient selection remains lacking.
4.1. Rates of Antithrombotic Therapy
The use of antithrombotic therapy in our PPHNet Registry cohort was generally low and substantially lower than those observed in adult populations, which likely in part reflect the high proportion of WSPH Group 3 patients in our cohort, a subgroup in which antithrombotic use was notably uncommon. Data on antithrombotic use from pediatric PH registries is limited to the PPHNet Registry, which includes a previously analyzed cohort from 2014 to 2020 that overlaps with our study group. That earlier analysis reported warfarin use at 5%, closely aligning with our current finding of 7% [13]. Aspirin use in our cohort was relatively low but higher than in the earlier PPHNet analysis (18% vs. 10%), although slightly below the 24% reported in a 2023 parent‐reported survey on therapy burden in pediatric PH [13, 16].
In adult registries, anticoagulation use has been reported at significantly higher rates [3, 5]. The COMPERA registry, an analysis of patients from 2007 to 2013, demonstrated 58% (738/1283) of incident PAH patients and 66% (528/800) of those with IPAH received anticoagulation [5]. A subsequent analysis of data from 2007 to 2016 reported lower rates, with 43% of IPAH patients receiving anticoagulation [17]. In comparison, the PHAR registry reported that 27% (540/1175) of the overall adult PH cohort had ever received anticoagulation, including 33% (176/540) of those with IPAH/HPAH. Notably, anticoagulation use in PHAR declined over time based on the year of registry entry from 29% to 19% between 2015 and 2019 [3]. In our analysis, children with WSPH Group I PAH in the PPHNet Registry had higher rates of warfarin use than the registry population overall, which included a substantial proportion of patients with WSPH Group 3 PH. Still, anticoagulation among PPHNet participants with WSPH Group I PAH remained below the levels reported in adult PAH cohorts.
This discrepancy between pediatric and adult anticoagulation rates potentially reflects both a shift in PH management away from routine anticoagulation, and the lack of evidence supporting its use in pediatric PH. Additionally, the challenges of managing anticoagulation in children who often have an inconsistent diet and greater difficulty tolerating regular blood draws, particularly with vitamin K antagonists, may further contribute to the lower utilization observed in our cohort compared to adults, underscored by the older age of patients in the warfarin+ group. We also found that rates of anticoagulation were lower in CDH‐PH and BPD‐PH, who may have been more likely to enter the registry at younger ages.
4.2. Antithrombotic Risk‐Based Utilization
Our PPHNet survey demonstrated that current clinical decision making aligns with guideline recommendations for the lowest risk patients, with full agreement across centers that antithrombotic therapy would generally not be used without identifiable thrombotic risk factors. PPHNet registry data further supported this as patients with variables associated with less severe disease were less likely to receive antithrombotic therapy. Notably, the lower utilization of antithrombotic therapy in our cohort may reflect a population with less severe disease, as suggested by the substantial proportion of patients who were either not receiving targeted therapy or were receiving monotherapy.
Conversely, we found that patients in the PPHNet registry who received antithrombotic therapy were more likely to have high‐risk features consistent with selective use among patients with greater disease severity. It is challenging to determine whether antithrombotic therapy was used broadly in patients perceived as high risk overall, or instead targeted to those with specific high‐risk features that represented independent indications for use. Similarly, antithrombotic therapy in our cohort was more frequently used in patients receiving advanced PH‐targeted treatments, particularly those involving parenteral prostacyclins. Notably, patients in the warfarin+ group, but not the aspirin+ group, were more likely to be treated with intravenous prostacyclin, potentially reflecting the decision of some providers to use warfarin for the indication of central line thrombosis prevention in this population.
These findings align with responses from the PPHNet center survey, where 80% of respondents reported recommending anticoagulation for pediatric patients with multiple high‐risk features, including systemic IPAH, atrial shunt, impaired RV function, and the presence of a central venous catheter. Data from adult registries including PHAR, COMPERA, REVEAL, and the French PH Registry have also demonstrated that anticoagulation is more frequently prescribed in patients with high‐risk features, specifically patients with higher clinical risk scores, worse functional class, and those receiving advanced PH‐targeted therapies, likely reflecting an increased concern for thromboembolic risk in this population [3, 4, 5, 6]. Higher rates of combination targeted therapy, lung transplantation, Potts shunt, and mortality in patients receiving antithrombotic therapy in our cohort support the idea that risk profile is a major driver of antithrombotic therapy choice. In this scenario, we suggest that the greater incidence of PH‐related adverse outcomes in treated patients more likely reflects the underlying etiology and disease severity rather than an adverse effect from the antithrombotic medications themselves. Similarly, these higher‐risk features and PH‐related adverse outcomes are more prevalent in WSPH Group I patients, who were also more likely to receive antithrombotic therapy.
4.3. Antithrombotic Use and Diagnosis
Our analysis of PPHNet registry data demonstrated that antithrombotic therapy use varied by primary WSPH group. Aspirin use was more frequent in WSPH Group 1 patients, with slightly higher use in IPAH/HPAH and substantially higher use in CHD‐PAH, likely reflecting underlying CHD‐related indications. Antithrombotic use was overall low in WSPH Group 3 patients, while aspirin use was relatively high in WSPH Group 2 patients, likely representing use in pulmonary vein stenosis. Patients with a Potts shunt were also more likely to receive antithrombotic therapy, which in this context may represent a separate indication rather than PH severity alone. However, the specific indications for antithrombotic therapy were not captured in the registry, nor were contraindications or concerns for adverse effects and comorbidities. Additionally, some patients receiving antithrombotic therapy were not on PH‐targeted therapy, further suggesting use for non‐PH indications.
Warfarin use was more common among patients with CTEPH, consistent with the condition's distinct pathophysiology and the long‐standing recommendation for lifelong anticoagulation with a vitamin K antagonist [11, 18]. However, only half of the CTEPH patients in the registry received warfarin, which may reflect practice variability, missing or incomplete data from non‐PH medication forms in the registry, or use of DOACs not recorded in the case report forms, as emerging adult data suggest they may offer comparable safety and outcomes [19, 20]. These findings may suggest that antithrombotic therapy use in pediatric PH may be influenced not only by PH subtype but also by coexisting conditions that warrant treatment. This highlights the complexity of clinical decision‐making in this population and the need for further investigation into prescribing practices and indications to help guide future studies.
4.4. Variability in Practice
One of the most significant findings in this study is the variation in clinical practice across PPHNet centers, as shown by both the registry data and the survey on clinical practice. The use of aspirin and warfarin varied significantly by center, with some centers not using either medication and others with rates as high as 28% for aspirin and 14% for warfarin. This variation remained even after accounting for center differences in case mix defined by WSPH group. These notable differences may reflect a combination of institutional protocols, provider interpretation of the risks and benefits, and differences in patient populations.
The survey data reinforces this variability, showing that approaches to antithrombotic therapy differ based on clinical scenarios and perceived risk. While there was general agreement for patients at the lowest and highest ends of the risk spectrum, management of intermediate‐risk patients differed considerably. This emphasizes the individualized nature of decision‐making and absence of a standardized approach in pediatric PH and highlights the need for more data to provide clearer guidance and consensus on antithrombotic therapy.
4.5. Anticoagulants
The PPHNet survey data indicate that warfarin remains the standard of care for anticoagulation, as it was selected as an anticoagulant of choice by all 10 participating centers. However, several centers also reported using additional agents, including rivaroxaban and apixaban. Although the registry data did not capture DOAC use and therefore did not offer additional insight, the survey may reflect growing clinical interest in use of DOACs in pediatric PH. This potential evolving practice could be linked to the FDA approval of dabigatran, rivaroxaban, and apixaban for pediatric use, as well as the publication of several randomized control trials in pediatrics, including use for thromboprophylaxis in CHD [21, 22, 23, 24, 25, 26, 27]. Future studies evaluating the safety and efficacy of the individual anticoagulants in pediatric PH are needed.
5. Limitations
This study has several limitations typical of both registry and survey‐based design. Survey data can be affected by response bias, variability in interpretation, and incomplete reporting, which may impact the reliability and generalizability of the results. Similarly, our registry data are limited by missing or incomplete entries; for example, catheterization data were only available for some patients, and hospitalization data depended on the completion of specific forms. Patient selection and timing of enrollment into the PPHNet registry occurred at the discretion of the treating clinicians, introducing the potential for bias. Medication use between diagnosis and enrollment was collected at the time of enrollment introducing the potential for recall bias, with a median time of 1.2 years between diagnosis and enrollment.
Because patients were enrolled at varying time points after diagnosis and throughout their treatment course, the data primarily reflected the prevalence of antithrombotic therapy use rather than the incidence of initiation in relation to specific clinical risk factors. This limitation was compounded by the study design, which required classification by “ever” versus “never” use of warfarin and aspirin, thereby limiting the ability to evaluate patient‐specific factors at the time of therapy initiation. To partially address this, clinical features such as medication therapy and RHC data closest to the visit where antithrombotic therapy was first reported was used in the “ever” group. However, even with this approach, we cannot determine whether aspirin or warfarin was initiated in response to high‐risk features, clinical deterioration, or unrelated factors, nor could we distinguish whether patients who received both agents were treated concomitantly or sequentially.
It is also important to note that the registry included patients across a wide range of disease severity and WSPH groups, which may shift the overall findings toward a milder phenotype. This is reflected in the substantial proportion of patients managed on less intensive therapy regimens, as well as the high representation of WSPH Group 3 patients, both of which may have contributed to lower overall utilization rates. Analyses limited to higher‐risk patients, particularly those with WSPH Group I PAH, would likely demonstrate higher rates of use. These findings should therefore be interpreted in the context of this heterogeneous cohort.
Furthermore, the registry also only includes data from specialized centers with expertise in pediatric PH, which might limit generalizability. Differences in patient populations, enrollment approaches, and clinical management across centers reflect real‐world variations in antithrombotic therapy use among PPHNet centers. Additionally, the registry spans a broad time period during which clinical practices and medication regimens have evolved. Notably, the use of anticoagulation may have declined over time, and newer agents such DOACs, which were not captured in the registry, have become available. These changes may contribute to an unmeasured era effect that influences outcomes and complicates interpretation across cohorts, and inclusion of DOACs could have resulted in a higher percentage of patients receiving anticoagulation.
Another important limitation of this analysis is the absence of data on specific indications for antithrombotic therapy within the registry. It was not possible from registry data to determine whether therapy was initiated specifically due to PH or a different indication, particularly in the CHD population without specific information on subtyping. Finally, outcomes data, including thrombotic or bleeding events, are not collected in the PPHNet Registry. Although patients treated with aspirin or warfarin appeared to represent a cohort with higher‐risk features, the study was unable to determine whether antithrombotic therapy contributed to outcomes in specific patient subgroups.
6. Conclusion
Our analysis offers the first real‐world perspective on antithrombotic therapy use in pediatric PH, drawing from both registry data and survey responses across PPHNet centers. Importantly, both data sources suggest that antithrombotic therapy was used most commonly in pediatric PH patients with clinical features associated with higher‐risk profiles or features that may represent a distinct indication, whereas its use was uncommon in patients without identifiable risk factors or indications. The absence of high‐quality randomized trials and limited pediatric‐specific evidence continue to hinder the development of clear clinical guidance. Further research is important to define which pediatric PH patients may benefit from antithrombotic therapy and to help inform evidence‐based recommendations for the most appropriate medication regimens.
Author Contributions
All authors contributed to the conception, design, and/or editing of this work. Lynn A. Sleeper and Lu Minmin conducted the PPHNet Registry data analysis. Meredith J. O'Neil and Benjamin S. Frank performed the survey analysis, interpreted the data, and drafted and revised the manuscript. All authors provided critical revisions to manuscript drafts regarding both content and language, and all authors have read and approved the final version of the manuscript.
Ethics Statement
The work reflected in this manuscript was carried out according to The Code of Ethics of the World Medical Association (Declaration of Helsinki). Informed consent was obtained from each subject's family prior to any research activities, and all center institutional review boards approved the study.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Supporting File
Acknowledgments
We gratefully acknowledge the contributions of the Pediatric Pulmonary Hypertension Network (PPHNet) investigators, coordinators, nurses, and research staff across all participating centers for their dedication to advancing the care of children with pulmonary hypertension. Their commitment to collaboration and data sharing made this project possible. Most importantly, we thank the patients and their families for their participation and trust, without whom this research could not be accomplished.
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.
References
- 1. Chin K. M., Gaine S. P., Gerges C., et al., “Treatment Algorithm for Pulmonary Arterial Hypertension,” European Respiratory Journal 64, no. 4 (2024): 2401325. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Ivy D., Rosenzweig E. B., Abman S. H., et al., “Embracing the Challenges of Neonatal and Paediatric Pulmonary Hypertension,” European Respiratory Journal 64, no. 4 (2024): 2401345. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Garry J. D., Kolaitis N. A., Kronmal R., et al., “Anticoagulation in Pulmonary Arterial Hypertension ‐ Association With Mortality, Healthcare Utilization, and Quality of Life: The Pulmonary Hypertension Association Registry (PHAR),” Journal of Heart and Lung Transplantation 41, no. 12 (2022): 1808–1818. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Bertoletti L., Escal J., Boucly A., et al., “Association Between Anticoagulant Therapy and Survival in Pulmonary Arterial Hypertension,” Journal of the American College of Cardiology 86, no. 13 (2025): 982–995. [DOI] [PubMed] [Google Scholar]
- 5. Olsson K. M., Delcroix M., Ghofrani H. A., et al., “Anticoagulation and Survival in Pulmonary Arterial Hypertension: Results From the Comparative, Prospective Registry of Newly Initiated Therapies for Pulmonary Hypertension (COMPERA),” Circulation 129, no. 1 (2014): 57–65. [DOI] [PubMed] [Google Scholar]
- 6. Preston I. R., Roberts K. E., Miller D. P., et al., “Effect of Warfarin Treatment on Survival of Patients With Pulmonary Arterial Hypertension (PAH) in the Registry to Evaluate Early and Long‐Term PAH Disease Management (REVEAL),” Circulation 132, no. 25 (2015): 2403–2411. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Roldan T., Rios J. J., Villamañan E., and Waxman A. B., “Complications Associated With the Use of Oral Anticoagulation in Patients With Pulmonary Arterial Hypertension From Two Referral Centers,” Pulmonary Circulation 7, no. 3 (2017): 692–701. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Wang P., Hu L., Yin Y., et al., “Can Anticoagulants Improve the Survival Rate for Patients With Idiopathic Pulmonary Arterial Hypertension? A Systematic Review and Meta‐Analysis,” Thrombosis Research 196 (2020): 251–256. [DOI] [PubMed] [Google Scholar]
- 9. Khan M. S., Usman M. S., Siddiqi T. J., et al., “Is Anticoagulation Beneficial in Pulmonary Arterial Hypertension?,” Circulation: Cardiovascular Quality and Outcomes 11, no. 9 (2018): e004757. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Chen X., Zhai Z., Lin L., et al., “The Role of Anticoagulation on the Long‐Term Survival of Patients With Pulmonary Arterial Hypertension: A Meta‐Analysis of 15 Cohort Studies,” Thrombosis Research 243 (2024): 109173. [DOI] [PubMed] [Google Scholar]
- 11. Humbert M., Kovacs G., Hoeper M. M., et al., “2022 ESC/ERS Guidelines for the Diagnosis and Treatment of Pulmonary Hypertension,” European Heart Journal 43, no. 38 (2022): 3618–3731. [DOI] [PubMed] [Google Scholar]
- 12. Abman S. H., Hansmann G., Archer S. L., et al., “Pediatric Pulmonary Hypertension: Guidelines From the American Heart Association and American Thoracic Society,” Circulation 132, no. 21 (2015): 2037–2099. [DOI] [PubMed] [Google Scholar]
- 13. Abman S. H., Mullen M. P., Sleeper L. A., et al., “Characterisation of Paediatric Pulmonary Hypertensive Vascular Disease From the PPhNet Registry,” European Respiratory Journal 59, no. 1 (2022): 2003337. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Roldan T., Villamañán E., Rios J. J., and Waxman A. B., “Assessment of the Quality of Anticoagulation Management in Patients With Pulmonary Arterial Hypertension,” Thrombosis Research 160 (2017): 83–90. [DOI] [PubMed] [Google Scholar]
- 15. Jose A., Eckman M. H., and Elwing J. M., “Anticoagulation in Pulmonary Arterial Hypertension: A Decision Analysis,” Pulmonary Circulation 9, no. 4 (2019): 1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Nelson E. J., Cook E., Nelson S., et al., “Quantifying Side Effects and Caregiver Burdens of Pediatric Pulmonary Hypertension Therapies,” BMC Pediatrics 23, no. 1 (2023): 43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Hoeper M. M., Kramer T., Pan Z., et al., “Mortality in Pulmonary Arterial Hypertension: Prediction by the 2015 European Pulmonary Hypertension Guidelines Risk Stratification Model,” European Respiratory Journal 50, no. 2 (2017): 1700740. [DOI] [PubMed] [Google Scholar]
- 18. Kim N. H., D'Armini A. M., Delcroix M., et al., “Chronic Thromboembolic Pulmonary Disease,” European Respiratory Journal 64, no. 4 (2024): 2401294. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Jain H., Odat R. M., Ahmed M., et al., “Safety and Outcomes With Direct Oral Anticoagulants Versus Vitamin‐K Antagonists in Chronic Thromboembolic Pulmonary Hypertension: A Systematic Review, Meta‐Analysis, and Meta‐Regression,” Cardiology in Review, ahead of print, June 4, 2024. [DOI] [PubMed] [Google Scholar]
- 20. Wu D., Liang H., Kaisaier W., et al., “The Use of Non‐Vitamin K Antagonist Oral Anticoagulants in Chronic Thromboembolic Pulmonary Hypertension: An Updated Meta‐Analysis,” Journal of Thrombosis and Thrombolysis 57, no. 7 (2024): 1256–1267. [DOI] [PubMed] [Google Scholar]
- 21. Monagle P., Lensing A. W. A., Thelen K., et al., “Bodyweight‐Adjusted Rivaroxaban for Children With Venous Thromboembolism (EINSTEIN‐Jr): Results From Three Multicentre, Single‐Arm, Phase 2 Studies,” Lancet Haematology 6, no. 10 (2019): e500–e509. [DOI] [PubMed] [Google Scholar]
- 22. Male C., Lensing A. W. A., Palumbo J. S., et al., “Rivaroxaban Compared With Standard Anticoagulants for the Treatment of Acute Venous Thromboembolism in Children: A Randomised, Controlled, Phase 3 Trial,” Lancet Haematology 7, no. 1 (2020): e18–e27. [DOI] [PubMed] [Google Scholar]
- 23. Röshammar D., Huang F., Albisetti M., et al., “Pharmacokinetic Modeling and Simulation Support for Age‐ and Weight‐Adjusted Dosing of Dabigatran Etexilate in Children With Venous Thromboembolism,” Journal of Thrombosis and Haemostasis 19, no. 5 (2021): 1259–1270. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Halton J., Brandao L. R., Luciani M., et al., “Dabigatran Etexilate for the Treatment of Acute Venous Thromboembolism in Children (Diversity): A Randomised, Controlled, Open‐Label, Phase 2b/3, Non‐Inferiority Trial,” Lancet Haematol 8, no. 1 (2021): e22–e33. [DOI] [PubMed] [Google Scholar]
- 25. Portman M. A., Jacobs J. P., Newburger J. W., et al., “Edoxaban for Thromboembolism Prevention in Pediatric Patients With Cardiac Disease,” Journal of the American College of Cardiology 80, no. 24 (2022): 2301–2310. [DOI] [PubMed] [Google Scholar]
- 26. Payne R. M., Burns K. M., Glatz A. C., et al., “Apixaban for Prevention of Thromboembolism in Pediatric Heart Disease,” Journal of the American College of Cardiology 82, no. 24 (2023): 2296–2309. [DOI] [PubMed] [Google Scholar]
- 27. McCrindle B. W., Michelson A. D., Van Bergen A. H., et al., “Thromboprophylaxis for Children Post‐Fontan Procedure: Insights From the UNIVERSE Study,” Journal of the American Heart Association 10, no. 22 (2021): e021765. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supporting File
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.