Skip to main content

Table 2 Characteristics of evidence synthesis studies

From: Evidence synthesis in pulmonary arterial hypertension: a systematic review and critical appraisal

Study ID (author year) Patient population Type of evidence synthesis Number of studies included Treatments included Outcomes included Quality assessment tool (used for included trials)
Avouac 2008 [17] Patients with PAH (including idiopathic, secondary to CTD or CHD) MA 10 Oral ERAs (bosentan, sitaxentan) and PDE-5I (oral sildenafil) 6MWD Jadad scores
Badiani 2016a [18] Patients with PAH (including associated PAH and IPAH) NMA 17 Oral ERAs (bosentan, ambrisentan and macitentan), oral PDE-5Is (sildenafil, tadalafil and vardenafild), prostanoids (oral beraprost and oral treprostinil) sPRA (oral selexipag), sGCS (oral riociguat) Composite clinical worsening Not reported
Bai 2011 [19] Patients with PAH MA 6 Oral PDE-5Is (tadalafil and sildenafil), ERA (bosentan), prostanoids (inhaled iloprost and IV epoprostenol) developed and approved for PAH; combination therapies only included with 2 or 3 drugs. 6MWD, clinical worsening, NYHA FC, mPAP, RAP, PVR and cardiac output, SAEs, all-cause mortality Quality assessment completed, tool not stated.
Barnes 2019 [20] Patients with PH (all groups 1–5) MA 3 (for PAH) Oral PDE5Is (sildenafil, tadalafil), oral ERAs (ambrisentan, bosentan) Primary outcomes: WHO FC, 6MWD and mortality. Secondary outcomes: Haemodynamic parameters, quality of life/health status, dyspnoea, clinical worsening (hospitalisation/intervention), and AEs Cochrane’s risk of bias
Biondi-Zoccai 2013 [21] Patients with PAH NMA 6 First line oral drugs: oral ERAs (bosentan, sitaxentan, ambrisentan, prostacyclin analogues (oral beraprost), oral PDE-5Is (tadalafil, sildenafil) All-cause mortality, clinical improvement and clinical worsening Cochrane’s risk of bias
Chen 2009 [22] Patients with PAH and subgroups (i.e. idiopathic PAH, CTD-associated PAH) MA 20 Any of epoprostenol (IV), iloprost (inhaled), bosentan (oral), sitaxentan (oral) and sildenafil (oral) Survival, time to clinical deterioration, HRQoL, 6MWD, symptomatic improvement, frequency and duration of hospitalization and outpatient/GP visits, SAEs, AEs, withdrawal, haemodynamic assessment Quality assessment completed, tool not stated
Coeytaux 2014 [23] [McCrory 2013 full report] Patients with PAH NMA 28 Oral PDE-5I (sildenafil and tadalafil), oral ERAs (bosentan and ambrisentan), prostanoids (IV epoprostenol, inhaled iloprost and IV or SC treprostinil) and calcium channel blockers Mortality, 6MWD, hospitalization, hemodynamic measures (i.e. PVR, PAP, cardiac index), and commonly reported AEs. Quality appraisal approach as described in the US Agency for Healthcare Research and Quality’s “Methods Guide for Effectiveness and Comparative Effectiveness Reviews.
Dranitsaris 2009 [24] Patients with PAH MA 9 Oral treatments: ambrisentan, bosentan, sitaxentan and sildenafil 6MWD, BDI, NYHA Functional Class and clinical worsening Not reported
Duo-Ji 2017 [25] Patients with symptomatic PAH, idiopathic PAH or PAH associated with other diseases NMA 10 Oral ERAs only (ambrisentan, bosentan, sitaxentan and macitentan) 6MWD, clinical worsening, SAE, mortality and all-cause discontinuation Jadad scores
Fox 2011b [26] Patients with PAH (including idiopathic PAH, familial PAH, CTD associated PAH, pulmonary shut, portal hypertension, HIV infection and thyroid disease) MA 6 Oral PDE-5I (sildenafil and tadalafil, ERA (oral bosentan), prostanoids (IV epoprostenol, inhaled iloprost and inhaled treprostinil developed and approved for PAH 6MWD, clinical worsening, mortality, hospitalization for PAH deterioration, lung transplantation, escalation of treatment and safety outcomes Jadad scores
Fox 2016 [27] Patients with PAH MA 18 Prostanoids (IV epoprostenol, inhaled iloprost and inhaled/oral treprostinil) oral ERAs (bosentan, ambrisentan, sitaxsentan and macitentan), oral PDE-5I (sildenafil and tadalafil), sGCS (oral riociguat), sPRA (oral selexipag) with their approved dose Primary outcomes: all-cause mortality (analysed separately) and composite clinical worsening. Secondary outcomes: 6MWD, PAP, cardiac index, WHO Functional Class. Cochrane’s risk of bias
Gabler 2012 [28] Patients with PAH (including idiopathic PAH, CTD-associated PAH, CHD-associated PAH, HIV infection) MA 10 Oral PDE-5I (sildenafil and tadalafil), oral ERAs (ambrisentan, bosentan and sitaxentan) and prostanoids (inhaled iloprost and SC treprostinil) 6MWD, mortality, lung transplantation, atrial septostomy, hospitalization due to PAH worsening, withdrawal for worsening right-sided heart failure, or addition of other PAH medications Not reported
Galie 2009b [29] Patients with PAH MA 21 Both approved and not approved treatments for PAH (oral ambrisentan, oral bosentan, oral sitaxentan, oral sildenafil, inhaled iloprost, oral beraprost, IV epoprostenol, SC treprostinil, oral terbogreld) Primary outcome: all-cause mortality Secondary outcomes: PAH-related hospitalizations to PAH, 6MWD, NYHA/ WHO Functional Class, RAP, PAP, cardiac index, and PVR Not reported
Gao 2017 [30] Patients with PAH NMA 32 Prostanoids (IV epoprostenol, inhaled iloprost, oral beraprost and oral/inhaled/SC treprostinil), oral ERAs (bosentan, ambrisentan and macitentan), oral PDE-5Is (sildenafil, tadalafil and vardenafild), sGCS (oral riociguat), and combination therapy regardless of drug dosage forms Primary endpoint: 6MWD
Secondary endpoints: PAP, PVR, all-cause mortality, and composite clinical worsening. Safety endpoint: SAEs
Jadad scores
He 2010 [31] Patients with PAH MA 11 Oral bosentan, oral sildenafil and inhaled iloprost Clinical worsening, NYHA/WHO Functional Class, 6MWD, and hemodynamic parameters including systolic PAP, PAP, PVR, cardiac output and cardiac index, treatment-related SAEs. Juni scale
Igarashi 2016 [32] Patients with PAH NMA 7 5 oral PAH treatments: ambrisentan, bosentan, sildenafil, tadalafil, and beraprost 6MWD, WHO Functional Class and PAP Cochrane’s risk of bias
Jain 2017 [33] Patients with symptomatic PAH NMA 31 All US-FDA approved PAH-specific drugs: oral ERAs (bosentan, ambrisentan and macitentan), oral PDE-5Is (sildenafil and tadalafil), prostanoids (oral/inhaled/SC/IV treprostinil, inhaled iloprost and IV epoprostenol), sGCS (oral riociguat) and sPRA (oral selexipag) Primary efficacy outcome: composite clinical worsening
Secondary efficacy outcomes: PAH-related hospitalization and all-cause mortality Safety outcome: treatment-related AEs leading to drug discontinuation
Cochrane’s risk of bias
Khouri 2018 [34] Patients with PH in the main analysis; patients with PAH in the sensitivity analysis MA and a disproportionality analysis 13 (7 in PAH patients) Oral PDE-5Is (sildenafil and tadalafil) and sGCS (oral riociguat) AEs Cochrane’s risk of bias and GRADE for evidence
Kirtania 2019 [35] Patients with PAH of any aetiology MA 7 Combination of oral ERAs (ambrisentan, bosentan, macitentan, sitaxentan) with oral PDE5Is (sildenafil or tadalafil), ERA or PDEI monotherapies Primary outcome: 6MWD
Secondary outcomes: Clinical worsening (death, hospitalisation, WHO FC, lung transplantation, clinical deterioration of PAH requiring additional therapy, PVR and NT-proBNP
Cochrane’s risk of bias
Kuwana 2013 [36] Patients with PAH and CTD-associated PAH MA 19 Oral PDE-5I (sildenafil and tadalafil), oral ERAs (bosentan and ambrisentan), prostacyclin analogues (IV epoprostenol, oral beraprost, inhaled iloprost and IV/SC/inhaled treprostinil) 6MWD Cochrane’s risk of bias
Lajoie 2016 [37] Patients with PAH (including idiopathic PAH, associated PAH, or hereditary PAH) MA 17 Prostanoids (IV epoprostenol, inhaled/oral treprostinil, inhaled iloprost), oral ERAs (bosentan, ambrisentan and macitentan), oral PDE-5I (sildenafil, tadalafil and vardenafild) or sGCS (oral riociguat) Primary outcome: clinical worsening Secondary outcomes: all-cause mortality, PAH-related mortality, PAH-related hospitalizations, lung transplantation, treatment escalation, symptomatic progression, WHO Functional Class, exercise capacity, treatment discontinuation, and treatment duration Cochrane’s risk of bias
Lajoie 2018 [38] Patients with PAH (including idiopathic PAH and associated PAH) MA 15 Currently licensed PAH-specific therapies: prostanoids (IV epoprostenol, inhaled iloprost, inhaled/oral treprostinil), oral ERAs (ambrisentan, bosentan, and macitentan), oral PDE-5Is (sildenafil, tadalafil, and vardenafild), sGCS (oral riociguat), and a sPRA (oral selexipag) Clinical worsening Cochrane’s risk of bias
Lei 2020 [39] Patients with CTD-associated PAH or SSc-PAH MA 27 Combination of oral ERAs (ambrisentan, bosentan) with oral PDE5Is (sildenafil or tadalafil), oral ERA or oral PDEI monotherapies 6MWD, hemodynamics parameters (PVR, PAP) not analysed due to insufficient data Cochrane’s risk of bias
Li 2013 [40] Patients with PAH MA 14 Prostanoids (IV epoprostenol, inhaled iloprost, SC/inhaled treprostinil, oral beraprost) Efficacy or safety endpoints (e.g. 6MWD, NYHA Functional Class, PAP, PVR, or all-cause mortality) Jadad scores
Li 2020 [41] Patients with PAH NMA 9 Oral ambrisentan, oral bosentan, oral sildenafil 6MWD, PAP, cardiac index, PVR, RAP and mortality Cochrane’s risk of bias and Jadad score
Lin 2018 [42] Patients with PAH NMA 43 Oral ERAs (bosentan, macitentan, sitaxentan and ambrisentan) sGCS (oral riociguat), oral PDE-5Is (sildenafil, tadalafil and vardenafild), Prostanoids (IV epoprostenol, IV/inhaled/oral/SC treprostinil, inhaled iloprost and oral beraprost), and sPRA (oral selexipag) monotherapy or in combination 6MWD, Functional Class amelioration, mortality, clinical worsening, SAEs, withdrawal, PVR, PAP, cardiac index, and RAP Jadad scores
Liu 2016 [43] Patients with PAH MA 35 Prostanoids (IV epoprostenol, inhaled/IV/SC/oral treprostinil, inhaled iloprost, oral beraprost and oral selexipag), oral ERAs (bosentan, ambrisentan and macitentan), oral PDE-5I (sildenafil, tadalafil and vardenafild) sGCSs (oral riociguat) and rho-kinase inhibitor (fasudild; ROA unclear) Primary outcomes: Mortality, 6MWD, WHO/NYHA Functional Class Secondary outcomes: Cardiopulmonary hemodynamics including PAP, PVR, cardiac index, withdrawal due to AEs Cochrane’s risk of bias
Macchia 2007 [44] Patient with PH (including primary PH due to CTD and PH related to thromboembolic disease) MA 16 Prostanoids (IV epoprostenol, SC treprostinil, inhaled iloprost and oral beraprost), oral ERAs (sitaxentan and bosentan, and PDE-5I (oral sildenafil) Total mortality, NYHA Functional Class and 6MWD Not reported
Macchia 2010 [45] Patients with PH (including idiopathic PAH and PAH-related conditions) MA 26 Prostanoids (inhaled iloprost, SC treprostinil and IV epoprostenol), oral ERAs (bosentan, ambrisentan and sitaxentan), and oral PDE-5I (sildenafil and tadalafil) Total mortality, NYHA Functional Class and 6MWD Not reported
Pan 2017 [46] Patients with different diseases including PAH MA 33 All oral ERAs (atrasentand, avosentand, ambrisentan, bosentan, darusentand, macitentan, sitaxentan and zibotentand) Mortality, CVD increased risk, AEs The Newcastle–Ottawa scale
Pan 2018 [47] Patients with CTD-associated PAH only MA 6 Prostanoids (IV epoprostenol, inhaled treprostinil, and inhaled iloprost), oral ERAs (ambrisentan, bosentan and macitentan), oral PDE-5Is (sildenafil and tadalafil, vardenafild), sGCSs (oral riociguat) and sPRA (oral selexipag) Primary outcome: composite clinical worsening
Secondary outcomes: 6MWD, N-terminal pro-B type natriuretic peptide (NT-proBNP), WHO/NYHA Functional Class or cardiopulmonary hemodynamics
Cochrane’s risk of bias
Paramothayan 2009 [48] Patients with primary PH and its variant MA 9 Prostanoids (IV/inhaled Iloprost, IV epoprostenol, IV/SC/oral treprostinil and oral beraprost) Primary outcomes: 6MWD NYHA Functional Class
Secondary outcomes: Mortality and AEs
The Cochrane approach and the Jadad score
Petrovic 2020a [49] Patients with PAH NMA 16 Oral ERAs (ambrisentan, bosentan, macitentan), oral PDE5Is (sildenafil, tadalafil), prostanoids (IV epoprostenol, oral/inhaled treprostinil, inhaled iloprost, oral beraprost), sGCSs (oral riociguat), sPRA (oral selexipag) as add-on therapies 6MWD, all-cause mortality, discontinuation due to AEs Cochrane’s risk of bias
Petrovic 2020b [50] Patients with PAH NMA 21 Oral ERAs (ambrisentan, bosentan, macitentan), oral PDE5Is (sildenafil, tadalafil), prostanoids (IV epoprostenol, SC treprostinil, inhaled iloprost, oral beraprost), sGCSs (oral riociguat), sPRA (oral selexipag) Efficacy outcomes: 6MWD, all-cause mortality
Safety outcome: discontinuation due to AEs
Cochrane’s risk of bias
Ryerson 2010 [51] Patients with PAH MA 24 Approved prostanoids (IV/inhaled/SC treprostinil, IV epoprostenol and inhaled iloprost oral) ERAs (ambrisentan, bosentan and sitaxentan) and PDE-5I (sildenafil and tadalafil) Total mortality and other clinical endpoints, including dyspnea, 6MWD, hemodynamics and AEs The Jadad score and the Cochrane Collaboration’s tool
Savarese 2012 [52] Patients with PAH MA 22 Prostanoids (IV epoprostenol, inhaled iloprost, oral beraprost and IV/SC treprostinil), oral ERAs (bosentan, ambrisentan and sitaxentan), oral PDE-5Is (sildenafil, tadalafil and vardenafild) and other drugs (oral imatinib, aspirin; ROA unclear) Primary endpoint: 6MWD Secondary endpoints: all-cause mortality, hospitalization for PAH and/or lung or heart-lung transplantation, initiation of PAH rescue therapy Detsky method
Savarese 2013 [53] Patients with PAH MA 16 Oral PDE-5I (sildenafil and vardenafild), prostanoids (SC treprostinil, IV epoprostenol and inhaled iloprost), oral ERAs (sitaxentan and bosentan), oral imatinibd Hemodynamic parameters (PAP, PVR, RAP and cardiac index), and clinical events (all-cause mortality, hospitalization for PAH and/or lung or heart-lung transplantation, initiation of PAH rescue therapy) Detsky method
Silva 2017 [54] Patients with idiopathic PAH and associated or secondary etiologies (heart failure, CTD-associated, anorexigen use, sickle-cell disease, and HIV) NMA 20 Prostanoids (IV epoprostenol, SC/oral treprostinil, oral beraprost and inhaled iloprost), oral ERAs (ambrisentan, bosentan and macitentan), oral PDE-5Is (sildenafil, tadalafil and vardenafild), sGCS (oral riociguat) 6MWD, Cardiac index, PAP, PVR, clinical worsening, and mortality Oxford quality scoring system
Steele 2010 [55] Patients with idiopathic PAH, or PAH associated with CTD, CHD or HIV MA 10 Oral bosentan, oral sitaxentan, inhaled iloprost, IV epoprostenol, sildenafil, oral ambrisentan, oral beraprost, inhaled/SC treprostinil, oral tadalafil and oral vardenafild Primary outcomes: 6MWD Functional Class Secondary outcomes: mortality, AEs Not reported
Thom 2015 [56] Patients with PAH NMA 16 (10 RCTs, 6 observational studies) Imatinib (oral) as add-on therapy to ERA (oral bosentan), oral PDE-5Is (sildenafil and tadalafil) or prostanoids (IV epoprostenol, inhaled iloprost, inhaled/SC treprostinil and oral beraprost) 6MWD NICE checklist for RCTs
Tran 2015 [57] [CADTH report] Patients with PAH NMA and MA 20 Prostanoids (IV epoprostenol and SC/IV treprostinil), oral ERAs (bosentan, ambrisentan and macitentan), and oral PDE-5Is (sildenafil and tadalafil), sGCS (oral riociguat) Clinical outcomes: mortality (all-cause, PAH-related), hospitalization, clinical worsening, NYHA/WHO heart failure Functional Class, 6MWD, and BDI and hemodynamic parameters (PVR, PAP, and cardiac index)
HRQoL
Safety outcomes: AEs, SAEs and treatment discontinuation due to AEs.
A standardized table based on major items from the SIGN 50 instrument. Further critical appraisal performed based on input from clinical experts.
Vizza 2018 [58] Patients with PAH MA 6 Oral bosentan, oral ambrisentan, oral riociguat, oral tadalafil and oral/inhaled treprostinil 6MWD Not reported
Wang 2018e [59] Patients with PAH NMA 45 Oral ERAs (ambrisentan, bosentan, macitentan, sitaxsentan), oral PDE5Is (sildenafil, tadalafil, vardenafild), prostanoids (IV epoprostenol, oral/IV/inhaled/SC treprostinil, inhaled iloprost, oral beraprost), sGCSs (oral riociguat), sPRA (oral selexipag) 6MWD, WHO FC, BDI, cardiac index, PAP, RAP, PVR, clinical worsening, hospitalization, death, SAEs, and withdrawal Not reported
Wei 2016 [60] Patients with different diseases including PAH MA 24 Oral ERAs (bosentan, ambrisentan and macitentan); EU authorised AEs Cochrane’s risk of bias and GRADE for evidence
Xing 2011 [61] Patients with PAH (including idiopathic PAH, familial PAH, as well as CTD-associated PAH, pulmonary shut, portal hypertension, HIV infection and thyroid disease) MA 10 Prostanoids (IV epoprostenol, IV/SC treprostinil, oral beraprost and inhaled iloprost) 6MWD, BDI, cardiac index, mean PAP, PVR, mortality, clinical worsening and AEs Jadad scores
Zhang 2015 [62] Patients with PAH MA 21 Oral treatments (ambrisentan, bosentan, macitentan, sitaxentan, sildenafil, tadalafil, riociguat, beraprost, epoprostenol, treprostinil, terbogreld and imatinibd) CCW or at least all-cause mortality Cochrane’s risk of bias
Zhang 2016 [63] Patients with PAH NMA 14 Prostanoids (IV epoprostenol, inhaled/IV/oral/SC treprostinil, oral beraprost and inhaled iloprost) 6MWD, mortality, Functional Class, and discontinuation Not reported
Zhang 2019 [64] Patients with PAH NMA 10 Oral ERAs (bosentan, ambrisentan, macitentan) Safety outcomes: abnormal liver function, peripheral edema and anemia Cochrane’s risk of bias
Zheng 2014ac [65] Patients with PAH MA 18 Oral targeted therapies: prostanoids (beraprost and treprostinil), ERAs (bosentan, ambrisentan and macitentan), PDE-5Is (sildenafil, tadalafil and vardenafild), and sGCS (riociguat) Primary efficacy outcome: all-cause mortality Secondary efficacy outcomes: clinical worsening, WHO Functional Class, 6MWD Safety outcome: withdrawal due to AEs Jadad scores
Zheng 2014b [66] Patients with PAH MA 14 Prostanoids (IV epoprostenol, inhaled, inhaled/IV/oral/SC treprostinil, inhaled and oral beraprost) Primary efficacy outcome: all-cause mortality Secondary efficacy outcomes: clinical worsening, 6MWD, and hemodynamic parameters, including PAP, PVR, cardiac index, and mixed venous oxygen saturation.
Safety outcome: withdrawal due to AEs
Jadad scores
Zheng 2018 [67] Patients with PAH MA 25 Oral prostanoids (treprostinil, beraprost), oral ERAs (ambrisentan, bosentan, macitentan), oral PDE-5Is (sildenafil, tadalafil, vardenafild), sGCSs (oral riociguat), sPRA (oral selexipag) Primary outcome: composite clinical worsening
Secondary outcomes: all-cause mortality, lung transplantation, admission to hospital, treatment escalation, WHO FC improvement, symptomatic progression and 6MWD
Jadad score
Zhu 2012 [68] Patients with PAH MA 7 Oral PDE-5Is (sildenafil and tadalafil) oral ERAs (bosentan, sitaxentan and ambrisentan), prostanoids (IV epoprostenol, inhaled iloprost and IV treprostinil) 6MWD, clinical worsening, mortality (data not shown) Moher 1998 reference provided for the quality assessment
  1. AEs Adverse events, BDI Borg dyspnea index, CCW Combined clinical worsening, CHD Congenital heart disease, CTD Connective tissue disease, ERAs Endothelin Receptor Antagonists, FC Functional class, FPAH Familial PAH, HRQOL Health related quality of life, IPAH Idiopathic PAH, MA Meta-analysis, NMA Network meta-analysis, NT-proBNP N-terminal probrain natriuretic peptide, NYHA/ WHO New York Heart Association/World Health Organization, IV Intravenous, PAH Pulmonary arterial hypertension, PAP Pulmonary arterial pressure, PCAs Prostacyclin analogues, PDE-5Is Phosphodiesterase 5 Inhibitors, PH Pulmonary hypertension, PVR Pulmonary vascular resistance, RAP Right atrial pressure, SAEs Severe adverse events, sGCSs Soluble guanylate cyclase stimulators, sPRAs Selective non-prostanoid prostacyclin receptor agonists, SC subcutaneous, SSc-PAH Pulmonary arterial hypertension related to systemic sclerosis, 6MWD Six minute walking distance
  2. aAlthough Badiani 2015 reported that prostanoids with IV/inhaled/SC ROA were considered for evaluation, trials on prostanoids with these ROAs were not included in the analysis. No justification provided. bIn Fox 2011, sitaxsentan, ambrisentan and vardenafil were included in the search strategy of the review, however, trials with these therapies were not included in the analysis. No justification provided. c In Zheng 2014a, trials on sitaxentan were excluded from the analysis as it was withdrawn from the market due to liver toxicity. The trial on selexipag was also excluded but provided no justification for the exclusion. dTreatments that have not been approved or made to any markets for adult patients with PAH. eIn Wang 2018, a subgroup analysis excluding sitaxsentan was conducted for network comparison of drugs in use on the market