Survival in Primary Pulmonary Hypertension with Long-Term Continuous Intravenous Prostacyclin

Primary pulmonary hypertension is characterized by a progressive elevation in pulmonary arterial pressure that eventually leads to right ventricular failure and death [1-3]. The median survival of patients who were prospectively entered into the National Institutes of Health (NIH) Registry on Primary Pulmonary Hypertension was 2.8 years after diagnosis [4]. No known cure exists for primary pulmonary hypertension; however, treatment for this disease has improved substantially over the past decade. Medical approaches include therapy with vasodilators [5-12], anticoagulant agents [13], inotropic agents [7], and diuretic agents and oxygen [14]. Nevertheless, some patients are refractory to medical therapy and require transplantation (heart and lung, single lung, and bilateral lung) [15-17]. In these patients with severe pulmonary vascular disease, the waiting time for transplantation may exceed their expected survival. Prostacyclin is a potent, short-acting vasodilator and inhibitor of platelet aggregation that is produced by the vascular endothelium. Prostacyclin decreases pulmonary vascular resistance and increases cardiac output and systemic oxygen delivery when acutely administered to patients with primary pulmonary hypertension [18]. We have previously reported that at the end of an 8-week randomized study, patients treated with prostacyclin had increased exercise capacity and improved hemodynamics compared with those who received conventional therapy [19]. The objective of this study was to evaluate the effects of long-term intravenous infusion of prostacyclin on exercise capacity, hemodynamics, and survival in patients with primary pulmonary hypertension. Methods In a previous 8-week randomized study [19], 11 of 25 patients were randomly assigned to receive prostacyclin plus conventional therapy, and 14 of the 25 patients were randomly assigned to receive conventional therapy alone. At the completion of the 8-week study, all survivors were eligible to enter this open, multicenter, uncontrolled trial of long-term intravenous infusion of prostacyclin regardless of their baseline hemodynamic response to prostacyclin or the treatment group to which they were initially assigned. The clinical diagnosis of primary pulmonary hypertension was established in all patients before entry on the basis of the criteria of the NIH Registry [3]. Thromboembolic disease was excluded on clinical grounds by perfusion lung scanning, or, when this was inconclusive, by pulmonary angiography. Patients with associated conditions such as portal hypertension, human immunodeficiency virus infection, collagen vascular diseases, and pulmonary vasculitides were excluded from this study. Sterile, lyophilized prostacyclin sodium powder (Flolan, epoprostenol sodium), synthesized by the Upjohn Co. (Kalamazoo, Michigan) and formulated by the Wellcome Research Laboratories (Beckenham, Kent, United Kingdom), was refrigerated until use. Immediately before administration, prostacyclin was reconstituted with a sterile buffer (pH, 10.5) at a concentration of 5 mg/mL and was filtered. All patients were treated with warfarin. Hemodynamic measurements were obtained by catheterization of the right side of the heart using standard techniques. After baseline hemodynamic measurements were obtained, an intravenous infusion of prostacyclin was begun at a rate of 2 ng/kg of body weight per minute and increased by increments of 2 ng/kg per minute every 10 to 15 minutes. Hemodynamic measurements were repeated at the end of the infusion of each dose. The acute infusion dose was not further increased when one or more of the following occurred: a greater than 40% decrease in systemic arterial pressure, a greater than 40% increase in heart rate, or symptoms such as nausea, vomiting, or severe headache. After the acute dose-ranging study, the dose of prostacyclin was decreased to a dose that did not result in any adverse effects. The long-term infusion dose was adjusted based on results of the most recently completed dose-ranging study or clinical needs of the patient. Patients were allowed to continue prostacyclin therapy until transplantation or death. Six-minute walk tests were done to assess exercise capacity [19]. Tests were carried out before long-term prostacyclin therapy was initiated and after 6, 12, and 18 months of long-term therapy. Venous access for the infusion of prostacyclin was obtained by inserting a permanent intravenous catheter into a jugular or subclavian vein and tunneling subcutaneously. Prostacyclin was infused continuously through a portable pump (Autosyringe AS2F, Travenol Inc., Hooksett, New Hampshire or CADD-1 Model 5100 HF, Pharmacia Deltec Inc., St. Paul, Minnesota). Before hospital discharge, patients were thoroughly trained in catheter care, sterile technique, and drug preparation and administration. Safety was monitored by physical examination, routine hematologic and chemical profiles, electrocardiograms, and adverse experience assessments. The prostacyclin doses and hemodynamic and exercise data are presented as the mean SD. Changes in exercise capacity and hemodynamics were analyzed using the Wilcoxon signed-rank test for paired data [20]. Survival analysis was done using a proportional hazards regression model [21] that models survival time against group (the 17 NYHA class III and IV patients treated with prostacyclin compared with the 31 NYHA class III and IV patients [historical controls] who were treated with anticoagulant agents). The model stratifies patients according to transplantation status and NYHA class and controls through a confounder score for baseline mean pulmonary artery pressure, mean right atrial pressure, and cardiac index. These three hemodynamic variables have been shown to be associated with survival in patients with primary pulmonary hypertension [4], and the particular covariate used here as a confounder score is the predicted survival at 1 year, which is a function of the three hemodynamic variables mentioned above. Patients were censored at the time of transplantation. We constructed Kaplan-Meier curves for the patients treated with prostacyclin and for the historical controls [22]. A two-sided P value of 0.05 was considered statistically significant. The Equation developed from the NIH Primary Pulmonary Hypertension Registry [4] was used to predict survival for the patients treated with prostacyclin as well as for the historical controls. The formula P(t) = H(t) A(x,y,z) H(t) = 0.88 0.14 t + 0.01t2 A(x,y,x) = e(0.007325x + 0.0526y 0.3275 z) where x = mean pulmonary artery pressure; y = mean right-atrial pressure; z = cardiac index; and t = 1, 2, or 3 years, estimates a patient's chances for survival (P[t]) at 1, 2, and 3 years given the values of the patient's three hemodynamic variables: mean pulmonary artery pressure, mean right-atrial pressure, and cardiac index. Results Patients Eighteen patients with primary pulmonary hypertension (17 adults and 1 child) were entered into this long-term study after informed consent was obtained. From the 25 patients enrolled in the preceding 8-week randomized study [19], the 10 surviving patients already receiving prostacyclin plus conventional therapy and 8 of the 11 survivors from the group receiving conventional therapy alone elected to enter this long-term study. The clinical and demographic characteristics are shown in Table 1. The mean age was 35.9 13.4 years. Twelve patients were female and 6 patients were male. Seventeen of the 18 patients were NYHA class III or IV despite conventional therapy, which consisted of vasodilators, oxygen, diuretic agents, cardiac glycosides, and anticoagulant agents as deemed necessary. Six patients were receiving oral vasodilator agents before entering this study, and 5 of these patients continued this therapy. Table 1. Clinical Characteristics and Baseline Hemodynamics Treatment Regimen Before long-term treatment with continuous intravenous prostacyclin was started and during the baseline acute dose-ranging study, the range of maximal tolerated doses of prostacyclin for the 18 patients was 4 to 22 ng/kg per minute. After the acute dose-ranging study, the dose of prostacyclin was decreased until it did not result in any adverse effects. The initial dose of the long-term prostacyclin infusion ranged from 2 to 8 ng/kg per minute (mean, 6.9 3.0 ng/kg per minute). Sixteen patients had repeat hemodynamic evaluation and dose-ranging study after 6 months of therapy, and the hemodynamics of 14 patients were re-evaluated at 12 months. Three patients had transplantation at 6, 10, and 12 months, respectively, before repeat hemodynamic measurements were obtained. One patient declined catheterization at 12 months. One patient had catheterization at 12 months but not at 6 months. The mean dose of long-term prostacyclin was 17.6 11.2 ng/kg per minute (n = 14) at 1 year, 36.7 21.2 ng/kg per minute (n = 11) at 2 years, and 52.9 30.2 ng/kg per minute (n = 7) at 3 years. Exercise Capacity Exercise endurance evaluated on the basis of the 6-minute walk test is shown in Figure 1. At 6 and 18 months, patients could walk, on average, more than 100 meters farther than they could before prostacyclin therapy was started. The length of the 6-minute walk increased from 264 160 meters at baseline to 370 119 meters at 6 months, 348 142 meters at 12 months, and 408 138 meters at 18 months (P < 0.001 at 6 months and P = 0.02 at 18 months compared with baseline). Figure 1. Effect of prostacyclin on exercise capacity. P P Hemodynamics Mean hemodynamic measurements at baseline (n = 18), 6 months (n = 16), and 12 months (n = 14) are shown in Table 2. In the patients who had follow-up cardiac catheterization at 6 months, the cardiac index increased 18% (CI, 0.1% to 36.7%), mean pulmonary artery pressure decreased 9% (CI, 1.4% to 15.7%), and total pulmonary resistance improved 26% (CI, 6.1% to 46.3%). Among patients re-evaluated at 12 months, a 27% increase remain