Table 2. Comparison of health-related quality of life (HRQoL) outcomes in pulmonary arterial hypertension (PAH) clinical trials
DrugFirst author [ref.]Comparator backgroundPatients nBaseline NYHA FCHRQoL measureDuration weeksPositive outcome componentsp-value
Target: endothelin pathway
    AmbrisentanGaliè [36]versus placebo202 ARIES-1I (2%)#
II (38%)#
III (55%)#
IV (5%)#
SF-3612Physical functioning0.005
192 ARIES-2
    BosentanGaliè [60]versus placebo
(+/- sildenafil)
185II (100%)#SF-3624Health transition index0.0244+
    MacitentanMehta [73]versus placebo
(+/-PDE5i or prostanoid)
742I (<1%)#
II (52%)#
III (46%)#
IV (2%)#
SF-36v26 monthsPCS, MCS and in seven out of eight domains<0.05§
EOTPCS3 mg 0.008ƒ
10 mg 0.001##
MCS3 mg 0.085¶¶
10 mg 0.053++
Target: nitric oxide pathway
    RiociguatGhofrani [74]versus placebo
(+/- prostanoid or ERA)
443I (3%)#
II (42%)#
III (53%)#
IV (1%)#
Missing (<1%)
EQ-5D§§12 (+4)No significant improvement0.07
LPHƒƒNominal improvement seen versus placebo0.002
    SildenafilPepke-Zaba [75]versus placebo
(+ background therapy)
278I (<1%)#
II (39%)#
III (58%)#
IV (3%)#
SF-3612Physical function
General health
EQ-5DUtility index
Current health status
Sastry [76]versus placebo22II (82%)
III (18%)
Emotional function
Simonneau [77]versus placebo
(+ epoprostenol)
267I (1%)#
II (25%)#
III (66%)#
IV (6%)#
Missing (2%)
SF-36v116Physical function
Physical role
General health
Social functioning
Mental health
    TadalafilPepke-Zaba [78]versus placebo
(+/- bosentan)
405I (1%)#
II (32%)#
III (65%)#
IV (2%)#
SF-3616Physical function###
Role physical
Bodily pain
General health
Social function###
EQ-5D UK utility
Target: prostacyclin pathway
    BeraprostBarst [57]versus placebo116II (53%)#
III (47%)#
MLHFQNo improvement shownns/data not shown
    EpoprostenolBarst [34]versus conventional therapy81III (74%)
IV (26%)
Emotional function
NHPEmotional reaction
Dyspnoea-fatigue rating
    Iloprost (inhaled)Olschewski [38]versus placebo203III (59%)
IV (41%)
EQ-5D12EQ-5D VAS0.026§§§
SF-12No improvement shownns
    TreprostinilMcLaughlin [61]versus placebo
(+ bosentan or sildenafil)
235III (98%)
IV (2%)
MLHFQ12Global score
Physical score
  • NYHA: New York Heart Association; FC: functional class; ARIES: Ambrisentan in Pulmonary Arterial Hypertension, Randomized, Double-Blind, Placebo-Controlled, Multicenter, Efficacy Studies; SF: short-form health survey; PDE5i: phosphodiesterase type-5 inhibitors; PCS: physical component summary; MCS: mental component summary; EOT: end of treatment; ERA: endothelin receptor antagonist; EQ-5D: EuroQol 5D questionnaire; LPH: Living with Pulmonary Hypertension questionnaire; CHFQ: chronic heart failure questionnaire; VAS: visual-analogue scale; MLHFQ: Minnesota Living with Heart Failure questionnaire; ns: nonsignificant; NHP: Nottingham Health Profile. #: World Health Organization FC; : for combined ambrisentan groups (2.5 and 5 mg), improvements in physical function were noted for 2.5 mg (p = 0.005) and 5 mg (p = 0.040), similar trends were observed without statistical significance for 5 and 10 mg of ambrisentan; +: SF-36 as compared with relative risks (bosentan versus placebo) and the treatment effect was tested with Fisher’s exact test, a higher proportion of patients on bosentan (57%) versus placebo (38%) improved their SF-36 at 24 weeks (p = 0.02) (HRQoL was an exploratory end-point only); §: except for general health perception; ƒ: hazard ratio (HR) 0.70 (95% CI 0.54–0.92); ##: HR 0.65 (95% CI 0.50–0.85); ¶¶: HR 0.81 (95% CI 0.63–1.03); ++: HR 0.79 (95% CI 0.61–1.01); §§: scores range from -0.6 to 1.0, higher scores indicate better QoL; ƒƒ: scores range from 0 to 105, higher scores indicate worse QoL; ###: were also improved in the 20 mg group; ¶¶¶: ANCOVA was used to evaluate changes from baseline to week 16 for SF-36 and EQ-5D, no adjustment of significance values for multiple testing was performed; +++: Hodges–Lehmann estimate was used to estimate the true median changes from baseline, whether statistical analysis was properly corrected for multiple comparisons (e.g. for each of the CHFQ and NPH domains) was not described; §§§: ANCOVA was used to compare score at baseline and at the end of study; ƒƒƒ: Hodges–Lehmann estimates, whether statistical analysis was properly corrected for multiple comparisons (e.g. for each of the MLHFQ domains) was not described. Reproduced from [79] with permission from the publisher.