The short-term vasodilator response to prostacyclin (PGI2) in patients with primary pulmonary hypertension (PPH) is not only unpredictable but also extremely variable in magnitude. In this retrospective study, we attempted to evaluate in a nonselected population of patients with PPH the degree of vasodilatation achieved during short-term infusion of PGI2 and to investigate whether patients with PPH differed in terms of baseline characteristics and prognoses, according to the level of vasodilatation achieved during initial testing with PGI2. Between 1984 and 1992, 91 consecutive patients with PPH underwent catheterization of the right side of the heart with a short-term vasodilator trial with PGI2 (5 to 10 ng.kg-1.min-1). According to the level of vasodilatation achieved during PGI2 infusion, patients were divided into three groups: nonresponding (NR, n = 40), moderately responding (MR, n = 42), and highly responding (HR, n = 9) patients. All three groups were defined by a decrease in total pulmonary resistance index (TPRi) of < 20%, between 20% and 50%, and > 50%, respectively, relative to control values. Prolonged oral vasodilator therapy was subsequently started only in MR and HR patients. All patients had long-term oral anticoagulant therapy. The survival rate at 2 years (transplant recipients excluded) was significantly higher in HR patients compared with NR and MR patients (62% versus 38% and 47% survivors, respectively; P < .05). Comparisons between groups showed no significant differences in baseline hemodynamics or clinical characteristics except for a longer time between onset of symptoms and diagnosis (ie, first catheterization) of PPH in HR patients than in NR and MR patients (71 +/- 61 versus 35 +/- 34 and 21 +/- 21 months, respectively; P < .05). In this study, patients with PPH exhibiting a decrease in TPRi > 50% during short-term PGI2 challenge at the time of diagnosis had longer disease evolutions and better prognoses than patients with a lower vasodilator response.