The added value of cardiopulmonary exercise testing in the follow-up of pulmonary arterial hypertension

BACKGROUND: The added value of cardiopulmonary exercise testing (CPET) in the follow-up of patients with stable pulmonary arterial hypertension (PAH) remains undefined. METHODS: Idiopathic, heritable, and drug-induced PAH patients free from clinical worsening (CW) after 1 year of treatment were enrolled in derivation (n = 80) and validation (n = 80) cohorts at an interval of 6 years and followed for 3 years. Prognostic models were constructed and validated in low-risk patients in World Health Organization (WHO) Functional Class I or II with cardiac index (CI) ≥2.5 liters/min/m2 and right atrial pressure (RAP) <8 mm Hg. Discrimination and calibration were assessed. RESULTS: Forty-one derivation cohort patients had CW (51.2%) during 722 ± 349 days. Changes (∆) in WHO classification and CI and absolute value of RAP were independent predictors of CW. With addition of CPET variables, peak oxygen uptake (VO2 peak) and ∆CI independently improved the power of the prognostic model. Receiver operating characteristic (ROC)-derived cut-off values for ∆CI and VO2 peak were 0.40 liter/min/m2 and 15.7 ml/kg/min (≥60% predicted value), respectively. Twenty-nine validation cohort patients had CW (36.2%) during 710 ± 282 days. Different combinations of cut-off values of VO2 peak and ∆CI defined 4 groups. The event-free survival rates at 1, 2, and 3 years were 100%, 100%, and 100%, respectively, for the high ∆CI with high VO2 peak combination; 100%, 88%, and 71% for low ∆CI/high VO2 peak; 80%, 54%, and 40% for high ∆CI/low VO2 peak; and 72%, 54%, and 33% for low ∆CI/low VO2 peak. CONCLUSIONS: The combinations of baseline VO2 peak and change in CI during follow-up is important in prognostication of low-risk patients with idiopathic, heritable, and drug-induced PAH.