Comparative efficacy and safety of advanced pharmacological therapies for pediatric pulmonary hypertension: a systematic review and meta-analysis with dose–response analysis of phosphodiesterase-5 inhibitors

Pulmonary hypertension (PH) in pediatric patients presents unique challenges due to limited evidence on dose-response and age-specific drug responses. This systematic review and meta-analysis aims to compare the safety and efficacy of phosphodiesterase type 5 inhibitors (PDE5i), endothelin receptor antagonists (ERAs), and prostacyclin (PGI2) on hemodynamic and oxygenation parameters in pediatric PH, addressing the need for comprehensive evidence-based insights. To date, limited randomized control trials (RCTs) and evidence-based investigations have focused on specific dose-response and age-specific drug responses in pediatric PH. This study provides a comprehensive overview of the safety and efficacy of PDE5i, ERAs, and PGI2, evaluating their impact on hemodynamic and oxygenation parameters. Related databases including PubMed/Medline, Web of Science, Scopus and Cochrane Library and Embase were searched to find relevant studies up to May 2025. We examined the RCTs evaluating the impact of PDE5i, ERAs and PGI2 on patient’s mPAP, sPAP, OI, SpO2, PaO2 and PA/AO, mortality, ventilation time, ICU stay or hospital stay. Effect size was expressed as weighted mean difference (WMD) and 95% confidence interval (CI). The meta-analysis of 31 randomized controlled trials (32 arms) including 1,498 participants showed significant reductions in mPAP with PDE5i (WMD: -13.49 mmHg, 95% CI: -18.04, -8.95, p < 0.001) and ERAs (WMD: -8.95 mmHg, 95% CI: -14.20, -3.69, p = 0.001). A non-linear dose-response analysis revealed a significant correlation between PDE5i and mPAP reductions (coefficient = -1.49, p = 0.024), with stronger effects at doses > 2 mg/kg/day. Random-effects meta-analysis of 10 studies indicated significant reductions in OI for PDE5i (WMD: -16.11, 95% CI: -25.06, -7.17, p < 0.001). ERAs showed a non-significant reduction in OI (WMD − 10.22; 95% CI − 23.92 to 3.49, p = 0.144, I² = 91.5%). SpO₂ increased significantly with PDE5i (WMD: 35.10%, 95% CI: 31.08, 39.12, p < 0.001) and ERAs (WMD: 12.66%, 95% CI: 3.33, 21.99, p = 0.008), and PGI2 (WMD: 4.05%, 95% CI: 0.61, 7.49, p = 0.021). PaO2 levels also increased significantly with PDE5i (WMD: 30.91 mmHg, 95% CI: 21.34, 40.48, p < 0.001) and ERAs (WMD: 10.02 mmHg, 95% CI: 5.45, 14.58, p < 0.001), alongside a significant reduction in PA/AO (WMD: -0.10, 95% CI: -0.14, -0.05, p < 0.001). The pooled relative risk showed lower mortality with PDE5i, ERAs and PGI2 compared with controls (RR 0.55, 95% CI 0.32–0.93); this finding was driven by a small number of trials and should be regarded as exploratory. GRADE assessment indicated very low to low certainty of evidence. Our meta-analysis supports a greater effect of PDE5i compared to ERAs and PGI2 on hemodynamic and oxygenation parameters in pediatric PH. Given substantial heterogeneity and very low to low certainty of evidence, these findings are hypothesis-generating; adequately powered trials in etiologically uniform populations with standardized hemodynamic assessment are required.