Continuous infusion of intravenous epoprostenol (EPO) improves exercise capacity and survival in pulmonary arterial hypertension (PAH); however, it is associated with side effects. This study aimed to describe our experience on safe EPO withdrawal with the aid of an exercise stress test. This retrospective review included patients with PAH who were successfully withdrawn from EPO. Haemodynamic data were obtained before EPO administration, at withdrawal, and after discontinuation. After the mean pulmonary arterial pressure (mPAP) decreased to <25 mmHg for at least 1 year, an exercise test under right heart catheterisation was performed. If exercise pulmonary hypertension was not record (mPAP - cardiac output slope < 3), EPO was withdrawn. Of the 99 patients who received EPO, ten were identified as having undergone withdrawal or termination. mPAP decreased from 61 (54–71) mmHg before treatment to 19 (17–21) mmHg before withdrawal and remained unchanged, at 19 (14–23) mmHg, 1 year after EPO discontinuation. After a median follow-up of 32 months, all the patients survived. Patients with PAH who recovered their normal haemodynamic function were safely withdrawn from EPO with the aid of an exercise stress test. • We reported our experience with pulmonary arterial hypertension (PAH) patients who accomplished normalised haemodynamics at rest and succeeded in weaning the epoprostenol (EPO) dosage to termination using the exercise stress test. • The criteria for EPO withdrawal were as follows: (1) persistence of WHO-FC I and normal haemodynamic status (mPAP <25 mmHg and PVR <3 Wood units at rest) at least for 1 year, (2) consistent dose of EPO for the last 1 year, and (3) no exercise pulmonary hypertension. • Patients with PAH who were in stable clinical condition and had normal haemodynamics were safely withdrawn from intravenous EPO with an aid of the exercise stress test. [ABSTRACT FROM AUTHOR]