Peripartum Substitution of Inhaled for Intravenous Prostacyclin in a Patient with Primary Pulmonary Hypertension

A 36-yr-old woman, gravida 4, para 3, weighing 77 kg was diagnosed with severe primary pulmonary hypertension during her sixth week of pregnancy and was receiving intravenous prostacyclin at 11.78 ng kg⁻¹min⁻¹. She presented at 36 weeks gestation for elective induction of labor with intravenous oxytocin, to be followed by a forceps-assisted vaginal delivery.

We obtained special permission from our institutional review board to use inhaled prostacyclin on our existing Food and Drug Administration Investigational New Drug Permit. Written, informed consent was obtained from the patient. Invasive monitoring was established with an arterial line and a PA catheter. An initial attempt at obtaining a wedge pressure was unsuccessful, and no further attempts were made. Initial vital signs showed a PA pressure of 80/37 mmHg, with a systemic blood pressure of 125/74 mmHg. We then started inhaled prostacyclin via  facemask, while continuing intravenous prostacyclin. We used a concentration-based regimen for prostacyclin inhalation: 20,000 ng/ml nebulized at 8 ml/h (which corresponded to 35 ng kg⁻¹min⁻¹). A continuous nebulization system (MiniHEART^®nebulizer; Westmed, Tuscon, AZ) was attached to a facemask with a Venturi attachment. Oxygen was administered to the Venturi attachment (to achieve an inspired oxygen concentration of 50%) and to the MiniHEART^®nebulizer (prefilled with 15 ml prostacyclin) at 2 l/min. A constant infusion of prostacyclin at 8 ml/h was maintained to the nebulizer to replace nebulized losses. PA pressures decreased further, to 68/25 mmHg, and the intravenous prostacyclin was carefully weaned off over 30 min.

One hour later, a lumbar epidural catheter was placed, and a T12 sensory level was established with incremental doses of 0.25% bupivacaine. The patient remained stable, with a blood pressure of 139/82 mmHg, a PA pressure of 71/30 mmHg, and a cardiac index of 4.0 l min⁻¹m⁻². Two hours after epidural placement, we restarted intravenous prostacyclin at half the previous dose and rapidly titrated the dose upward as tolerated. This was done because there were no published reports of the efficacy of inhaled prostacyclin in this specific setting. Inhaled prostacyclin was stopped when a dose of 10 ng kg⁻¹min⁻¹intravenous prostacyclin was reached.

Over the next 3–5 h, the patient’s PA pressures increased to levels above baseline (88/21 mmHg), while her systemic blood pressure (103/52 mmHg) and cardiac index (3.4 l min⁻¹m⁻²) decreased. We decided that the reported benefits of selective PA vasodilation with inhaled prostacyclin would be desirable. 1–3Inhaled prostacyclin was restarted at the previous concentration-based dose (20,000 ng/ml at 8 ml/h), and intravenous prostacyclin was rapidly weaned off. There was an immediate favorable response. The PA pressure decreased to 68/27 mmHg, and arterial blood pressure subsequently improved to 129/80 mmHg.

Six hours later, labor was induced with oxytocin. PA pressures increased somewhat during uterine contractions (peaks of 70–80/24–30 mmHg) but declined to baseline during uterine diastole (60–64/12–24 mmHg). A healthy infant was delivered by forceps 3 h after the start of the labor induction. The postpartum uterine tone was normal, and there was no excessive bleeding. The epidural catheter was removed shortly after the delivery. Inhaled prostacyclin was continued for 8 more hours to allow for smooth reintroduction of intravenous prostacyclin. The patient’s postpartum course was uneventful, and she was discharged home on postpartum day 4.

Primary pulmonary hypertension during pregnancy is a disease with significant morbidity, and a mortality rate as high as 50% has been reported. 4Death may occur during or immediately after cesarean or vaginal delivery. 5,6Most deaths, however, occur within the first post-partum week and are due to acute right ventricular failure and cardiovascular collapse. Successful management of right ventricular failure may be guided by invasive hemodynamic monitoring to optimize intravascular volume, reduce right ventricular afterload, and support right ventricular contractility. In this patient, a high forceps delivery was planned to avoid excessive pushing during delivery, which we anticipated would have deleterious effects on PA pressures and right ventricular function. We did not plan an elective cesarean delivery because maternal mortality is higher with cesarean than vaginal delivery. 4 

Inhaled and intravenous prostacyclin have been extensively studied, and their efficacy and side effects have been well described. We elected to switch this patient to inhaled prostacyclin for several important reasons: (1) it is as effective as inhaled nitric oxide, 3,7,8is less costly, and does not carry the potential risk of toxicity seen with inhaled nitric oxide; (2) it has minimal systemic absorption 9,10and therefore would potentially limit the anti-platelet activity seen with intravenous prostacyclin; and (3) the patient had already demonstrated a favorable response to intravenous prostacyclin.

Inhaled prostacyclin has been studied over a dose range of 0–50 ng kg⁻¹min⁻¹. No systemic side effects such as systemic hypotension or platelet dysfunction was observed over this dose range even though prostacyclin metabolites, with known antiplatelet effects, increased significantly. 11We used a concentration-based regimen for prostacyclin inhalation: 20,000 ng/ml nebulized at 8 ml/h (which corresponded to 35 ng kg⁻¹min⁻¹). Prostacyclin and its major metabolite, 6-keto-prostaglandin F1α, are potent inhibitors of platelet aggregation in vitro , 9,10raising concern about the use of regional anesthesia in patients receiving systemic prostacyclin. No clinical evidence of platelet dysfunction or excessive bleeding has been reported with inhaled use. Significant in vitro  inhibition of platelet function does not usually occur until prostacyclin has been inhaled for 4–6 h. 9In human studies, inhaled prostacyclin has not been shown to have a significant clinical effect on postoperative bleeding in postoperative cardiac surgery patients. 9Prostacyclin is rapidly hydrolyzed (T1/2 = 5 min) to 6-keto prostaglandin F1α at acid or neutral pH. Prolonged antiaggregatory effects (up to 48 h) are seen only at alkaline pH (pH > 7.8). Platelets incubated with (or exposed to) prostacyclin recover spontaneously and respond to aggregating agents within 15–60 min, depending on the initial prostacyclin concentration. 12 

We carefully considered both the potential risk of epidural hematoma due to the antiplatelet effects of prostacyclin and the risk of rebound pulmonary hypertension seen with intravenous drug withdrawal. At the time, we did not have access to test for potential anti-platelet function effects during the transition of intravenous to inhaled prostacyclin. Despite a normal pro-thrombin time, a normal activated partial thromboplastin time, and a normal platelet count antepartum, we could not fully exclude a potential or residual antiplatelet effect. A prolonged antiplatelet effect, however, seemed unlikely at normal pH, 12and we anticipated plasma concentrations to be lower with inhaled than with intravenous prostacyclin, which would further minimize the risk of an epidural hematoma due to an antiplatelet effect. 12Although we do not know if our approach was entirely safe, under the circumstances, our plan seemed reasonable.

In conclusion, inhaled prostacyclin seems to be a safe and effective alternative to intravenous prostacyclin in the peripartum period. Its hemodynamic profile is favorable, it is relatively inexpensive, and it is easy to administer to spontaneously breathing patients. It may offer advantages over intravenous prostacyclin in those patients requiring a regional anesthetic technique.