EPOPROSTENOL

Fetal Risk Summary

The naturally occurring prostaglandin, epoprostenol (prostacyclin, PGI2, PGX), is a metabolite of arachidonic acid. Available in the United States since 1995, it is a potent direct vasodilator of pulmonary and systemic arterial vascular beds and an inhibitor of platelet aggregation. Epoprostenol is indicated for the long-term IV treatment of primary pulmonary hypertension (1).

Reproduction studies have revealed no evidence of impaired fertility or fetal harm in pregnant rats and rabbits at 600 µg/m2/day (2.5 times the recommended human dose based on body surface area [RHD]) and 1180 µg/m2/day (4.8 times the RHD), respectively (1).

In animals, epoprostenol causes dose-related changes on the heart rate: bradycardia at low doses and reflex tachycardia secondary to vasodilation and hypotension at higher doses. Other pharmacologic effects in animals include bronchodilation, inhibition of gastric acid secretion, and decreased gastric emptying (1).

No reports describing the placental transfer of epoprostenol have been located. The molecular weight (about 374) of the drug is low enough for placental transfer, but the clinical significance of this is unknown because the agent is rapidly hydrolyzed and undergoes enzymatic degradation in the plasma. In addition, other factors (e.g., ionization, placental metabolism) may limit transfer to the fetus. Continuous infusions of the prostaglandin, which reach steady-state plasma levels within 15 minutes in animals and are proportional to infusion rates, could potentially provide a reservoir to allow placental transfer. But the in vivo plasma half-life of tritium-labelled epoprostenol in animals is very short (2.7 minutes) (1). In humans, the in vitro half-life in blood is slightly longer, approximately 6 minutes, and the in vivo half-life, although it has not yet been determined, is expected to be no longer than this time (1). Thus, even if placental transfer does occur the amounts reaching the fetus may be clinically insignificant. The effects of adding to the endogenous concentrations of the prostaglandin in the fetus are unknown.

In an in vitro study, prostacyclin was shown to be a potent inhibitor of human fetal and maternal platelet aggregation and more potent than S-nitroso-N-acetylpenicillamine (a releaser of nitric oxide) (2). Moreover, fetal platelets were more sensitive to both agents than were maternal platelets.

The effects of various drugs on the production of maternal and fetal endogenous prostacyclin have been studied in in vivo and in vitro systems. For ritodrine, one study found an increase in the synthesis of the prostaglandin (3), whereas a second reported no effect (4). Ethanol was shown to decrease prostacyclin synthesis in a dose-related manner (5). No effect on prostacyclin production was shown with aspirin (6,7 and 8) or magnesium sulfate (9).

The enzymes involved in the synthesis of prostacyclin (PGI2) and thromboxane A2 (TXA2) were studied in maternal and fetal platelets and venous endothelium in a 1991 publication (10). Three groups of pregnant women were studied: normal controls; mild preeclampsia; and severe preeclampsia. Based on their findings, the authors concluded that the production of both PGI2 and TXA2 was equal in mild preeclampsia, but in severe preeclampsia, the rise in TXA2 exceeded that of PGI2.

A number of reports have described the use of epoprostenol for the treatment of severe preeclampsia (11,12,13,14,15,16 and 17) or pulmonary hypertension (18,19). In a 1980 case report, a 29-year-old woman at 27 weeks' gestation with hypertension uncontrolled with oral methyldopa (3 g/day) and hydralazine (150 mg/day) was started on a continuous infusion of epoprostenol at 8 ng/kg/minute (11). Diastolic pressure fell from 140 mm Hg to 80 mm Hg and remained at or below 95 mm Hg during the next 3 days. Adverse effects in the mother included flushing, tachycardia (95–100 beats per minute [bpm]), and transient, nonspecific chest and abdominal pain. On day 4 of therapy, prolonged (1–2 minutes) fetal bradycardia occurred (60 bpm) and delivery by cesarean section was initiated. During the induction phase of anesthesia, the epoprostenol infusion was stopped for 4 minutes. During this interval, the blood pressure rose to 200/100 mm Hg but returned to 130/80 mm Hg when the infusion was restarted. More than normal bleeding occurred during the procedure and the infusion was discontinued. The 730-g newborn was alive and making good progress at 3 weeks of age.

A continuous IV infusion of epoprostenol was started at 25 weeks' gestation in a 27-old-woman with preeclampsia only partially responsive to methyldopa, aspirin (300 mg every other day), and hydralazine (12). The woman had a history of two previous pregnancies complicated by pregnancy-induced hypertension, severe intrauterine growth retardation, and fetal death at 27 weeks' gestation. Epoprostenol was chosen because of a rise in her diastolic pressure to 90 mm Hg, a slowing of fetal growth, and a falling maternal plasma concentration of 6-oxo-PGF1a (6-keto-prostaglandin F1a), the hydrolyzed product of epoprostenol (12). Mild nausea, facial flushing, and headache limited the dose tolerated by the patient to 4 ng/kg/minute. The maternal diastolic blood pressure stabilized at or below 85 mm Hg and her pulse rate increased from 69 bpm to 84 bpm, but the FHR remained unchanged. In spite of almost continuous FHR monitoring, the fetus died on the 4th day of the infusion, after which a macerated 430-g female fetus was delivered. The authors speculated that intervention with epoprostenol came to late to save the pregnancy (12).

A study published in 1985 examined the effect of IV infusions of epoprostenol in five women (gestational ages 24 to 30 weeks) with severe preeclampsia that was unresponsive to various combinations of other antihypertensive therapy (methyldopa, hydralazine, labetalol, and atenolol) (13). The maximum continuous IV infusion doses tolerated by the patients ranged from 1 to 6 ng/kg/minute with durations varying from 5 hours to 11 days. Facial flushing, nausea and vomiting, and headache were dose-limiting adverse effects. Four infants were delivered by cesarean section when the infusions (duration 5 hours to 7 days) were stopped, but one newborn (mother treated for 4 days) died 5 hours after delivery (birth weight 480 g). Another fetus (600 g) was stillborn at 26 weeks' gestation on the 11th day of epoprostenol (2.5 ng/kg/minute). Ultrasound and FHR monitoring did not detect any significant changes in the fetus prior to death. The authors thought that the two deaths were not related to the drug. The blood pressures of the mothers were reduced, but dose-limiting adverse effects prevented satisfactory control in one patient. Epoprostenol had no effect on uterine activity in any of the cases (13).

No significant changes in placental (intervillous) and umbilical vein blood flow were measured in a study of 13 women (gestational ages 28–38 weeks) with either superimposed preeclampsia (N=6) or preeclampsia (N=7) (14). The infusion dose was increased in four increments (1, 2, 4, and 8 ng/kg/minute), each given for 20 minutes. All women experienced facial flushing and two complained of headache. Significant decreases in maternal blood pressure and consistent increases in maternal plasma or urinary 6-keto-prostaglandin F1a levels were measured. In contrast, no changes in maternal or fetal pulse rates or uterine activity were observed. The fetuses were delivered 1 to 13 days after the study. Except for one 850-g newborn delivered at 30 weeks' who died at age 2 days from respiratory distress syndrome, the other newborns (birth weights 980 g to 2840 g) were well at birth and survived (14).

Two vasodilators, epoprostenol (N=22) or dihydralazine (N=25), were compared in a prospective randomized study of women with proteinuria and acute diastolic blood pressures >110 mm Hg immediately before delivery (15). The mean gestational age in both groups was 36 weeks (range 32–40 weeks). The epoprostenol infusion was started at 0.5 ng/kg/minute and was increased to a maximum of 10 ng/kg/minute within 120 minutes, if needed to control the blood pressure. The initial dihydralazine dose was 0.5 mg/kg/minute and increased to a maximum of 1.5 mg/kg/minute, if needed. Both drugs were continued for 24 hours after delivery and during delivery for those requiring cesarean section. Increases in maternal pulse rate occurred in both groups, but the rise in the epoprostenol patients (6 bpm) was significantly less than the rise with dihydralazine (22 bpm) (p=0.0024). Blood pressures were reduced in both groups, but the difference between them was not significant. Cesarean sections were performed in 13 (60%) (11 for FHR decelerations) of the epoprostenol group and in 20 (80%) (14 for FHR decelerations) of those receiving dihydralazine (difference not significant). One neonatal death occurred in each group, but neither was related to the drug therapy (15).

A brief 1992 communication reported the treatment of a 35-year-old woman with thrombotic microangiopathy superimposed on preeclampsia at about 26 weeks' gestation (16). Epoprostenol was started at 2 ng/kg/minute and increased to 20 ng/kg/minute over 24 hours with reduction of her blood pressure and a rise in the platelet count. After stabilization, therapy was stopped but restarted later when her condition worsened. Improvement was again achieved, but she was delivered at 28 weeks' because of oligohydramnios and growth retardation. The baby girl died a week later from respiratory arrest. In addition, the authors described, without details, the successful treatment of a 27-year-old patient with HELLP (hemolysis, elevated liver enzymes, and low platelets) syndrome (16). Epoprostenol has also been used to treat a severe case of HELLP syndrome in the immediate postpartum period (17).

A 1999 publication described the use of epoprostenol for the treatment of pulmonary hypertension in three women in the 3rd trimester (18). One patient, at 28 weeks' gestation, died from severe disease within hours of the diagnosis. The other two women delivered newborns of 3333 g and 2905 g, respectively, but the condition and status of the newborns was not mentioned.

In another case of primary pulmonary hypertension, a 34-year-old woman became pregnant with twins after 1.5 years of continuous IV infusion of epoprostenol (40 ng/kg/hour) (19). She was also taking warfarin, digoxin, furosemide, spironolactone, and ferrous sulfate. She first presented at 15 weeks' gestation at which time the warfarin was discontinued. An ultrasound examination 2 weeks later revealed the death of twin A and hydrocephalus and bilateral clubfeet in twin B. She refused pregnancy termination. She was treated twice for central catheter-related septicemia with vancomycin and gentamicin. The dose of epoprostenol was increased as pregnancy progressed, eventually reaching a dose of 60 ng/kg/hour. Betamethasone was given for lung maturity. Nitric oxide via nasal cannula was started just prior to cesarean section for breech presentation. A live, 2155-g male infant was delivered with Apgar scores of 7 and 8 at one and five minutes, respectively. The infant had severe hydrocephalus and facial anomalies consistent with fetal warfarin syndrome (19). He was alive but still hospitalized at the time of the report. The nitric oxide was discontinued after 40 days and the mother was discharged home 43 days after delivery on epoprostenol 77 ng/kg/hour.

In summary, epoprostenol is not teratogenic in animals but the human data are too limited, only one human case of exposure during early gestation has been reported, to assess. The placental transfer of the prostaglandin has not been characterized, but it is unlikely if clinically significant amounts of exogenous epoprostenol reach the fetus. Because the prostaglandin occurs naturally in the fetus, it is also unlikely that maternal administration would produce a direct adverse effect on the embryo or fetus. The adverse fetal outcomes that have been noted (bradycardia, FHR decelerations, and death) were most likely due to severe maternal disease or other factors. Although the maternal benefits obtained from use of epoprostenol for severe preeclampsia have not been well documented, its use for pulmonary hypertension is beneficial and appears to far outweigh any potential risks to the fetus.

Breast Feeding Summary

No reports describing the use of epoprostenol during lactation have been located. However, there is potential for its use during breast feeding, such as in postpartum women with pulmonary hypertension on long-term infusions of the drug. However, this situation must be exceedingly rare. Because of its rapid degradation at physiologic pH and probably in the gut, the clinical significance of any transfer into milk to a nursing infant is probably nil. Moreover, the prostaglandin has been given directly by inhalation to a premature neonate with beneficial effects (20).

References

1. Product information. Flolan. Glaxo Wellcome, 2000.
2. Varela AF, Runge A, Ignarro LJ, Chaudhuri G. Nitric oxide and prostacyclin inhibit fetal platelet aggregation: a response similar to that observed in adults. Am J Obstet Gynecol 1992;167:1599–604.
3. Jouppila P, Kirkinen P, Koivula A, Ylikorkala O. Ritodrine infusion during late pregnancy: effects on fetal and placental blood flow, prostacyclin, and thromboxane. Am J Obstet Gynecol 1985;151:1028–32.
4. Ekblad U, Erkkola R, Uotila P, Kanto J, Palo P. Ritodrine infusion at term: effects on maternal and fetal prostacyclin, thomboxane and prostaglandin precursor fatty acids. Gynecol Obstet Invest 1988;25:106–12.
5. Randall CL, Saulnier JL. Effect of ethanol on prostacyclin, thromboxane, and prostaglandin E production in human umbilical veins. Alcohol Clin Exp Res 1995;19:741–6.
6. Martin C, Varner MW, Brance DW, Rodgers G, Mitchell MD. Dose-related effects of low dose aspirin on hemostasis parameters and prostacyclin/thromboxane ratios in late pregnancy. Prostaglandins 1996;51:321–30.
7. Tulppala M, Marttunen M, Soderstrom-Anttila V, Foudila T, Ailus K, Palosuo T, Ylikorkala O. Low-dose aspirin in prevention of miscarriage in women with unexplained or autoimmune related recurrent miscarriage: effect on prostacyclin and thromboxane A2 production. Hum Reprod 1997;12:1567–72.
8. Vainio M, Maenpaa J, Riutta A, Ylitalo P, Ala-Fossi SL, Tuimala R. In the dose range of 0.5–2.0 mg/kg, acetylsalicylic acid does not affect prostacyclin production in hypertensive pregnancies. Acta Obstet Gynecol Scand 1999;78:82–8.
9. Walsh SW, Romney AD, Wang Y, Walsh MD. Magnesium sulfate attenuates peroxide-induced vasoconstriction in the human placenta. Am J Obstet Gynecol 1998;178:7–12.
10. Satoh K, Seki H, Sakamoto H. Role of prostaglandins in pregnancy-induced hypertension. Am J Kidney Dis 1991;17:133–8.
11. Fidler J, Bennett MJ, De Swift M, Ellis C, Lewis PJ. Treatment of pregnancy hypertension with prostacyclin. Lancet 1980;2:31–2.
12. Lewis PJ, Shepherd GL, Ritter J, Chan SMT, Bolton PJ, Jogee M, Myatt L, Elder MG. Prostacyclin and pre-eclampsia. Lancet 1981;1:559.
13. Belch JJF, Thorburn J, Greer IA, Sarfo S, Prentice CRM. Intravenous prostacyclin in the management of pregnancies complicated by severe hypertension. Clin Exp Hypertens Preg 1985;B4:75–86.
14. Jouppila P, Kirkinen P, Koivula A, Ylikorkala O. Failure of exogenous prostacyclin to change placental and fetal blood flow in preeclampsia. Am J Obstet Gynecol 1985;151:661–5.
15. Moodley J, Gouws E. A comparative study of the use of epoprostenol and dihydralazine in severe hypertension in pregnancy. Br J Obstet Gynaecol 1992;99:727–30.
16. De Belder AJ, Weston MJ. Epoprostenol infusions in thrombotic microangiopathy of pregnancy. Lancet 1992;339:741–2.
17. Huber W, Schweigart U, Classen M. Epoprostenol and plasmapheresis in complicated HELLP syndrome with pancreatitis. Lancet 1994;343:848.
18. Easterling TR, Ralph DD, Schmucker BC. Pulmonary hypertension in pregnancy: treatment with pulmonary vasodilators. Obstet Gynecol 1999;93:494–8.
19. Badalian SS, Silverman RK, Aubry RH, Longo J. Twin pregnancy in a woman on long-term epoprostenol therapy for primary pulmonary hypertension. A case report. J Reprod Med 2000;45:149–52.
20. Soditt V, Aring C, Groneck P. Improvement of oxygenation induced by aerosolized prostacyclin in a preterm infant with persistent pulmonary hypertension of the newborn. Intensive Care Med 1997;23:1275–8.