Pulmonary hypertension (PHTN), primary or secondary, is a challenge to manage in the operating room (OR) and the intensive care unit (ICU). Over time, various IV therapies have been evaluated, including calcium channel blockers, nitrate vasodilators, β agonists, various prostaglandins, angiotensin-converting enzyme inhibitors, α-blockers, and direct vasodilators. Systemic hypotension is the limiting factor in escalating dosage to a therapeutic level because of lack of pulmonary selectivity (1).
The use of inhaled nitric oxide (NO) to treat PHTN has demonstrated the advantage of applying a drug with a very short half-life in close proximity to the site of desired action. However, although NO can improve pulmonary hemodynamics and gas exchange with a minimum of systemic effects, improvements in outcome have been elusive. Using NO requires significant financial investment in administration and monitoring equipment. The products of its reaction with molecular oxygen are toxic, and the safety of long-term use is unclear (2). The practical difficulties in providing inhaled NO to patients in the OR on an urgent basis given personnel and space limitations make its use problematic in many settings.
Prostacyclin or epoprostenol (PGI2) is an endogenous prostaglandin produced via the cyclooxygenase arm of the arachidonic acid metabolic pathway. It is a potent vasodilator, inhibits neutrophil activation, stabilizes cell membranes, inhibits platelet aggregation, and enhances myocardial inotropy (3). It has a half-life of two to three minutes and is metabolized primarily in the liver. As with other IV drugs used to treat PHTN, its systemic use is severely limited by systemic hypotension. Recently, it has been used as an aerosolized or nebulized drug. We present case reports of five patients in whom inhaled PGI2 was used on a compassionate basis to treat PHTN and right ventricular failure in the OR.
Inhaled PGI2 was administered by using a common, disposable nebulizer system (Nellcor Puritan Bennett REF 6-001140-00; St. Louis, MO). The device was placed in the inspiratory limb of the anesthetic circuit approximately 12 cm from the Y-connector with an oxygen flow of 4 L/min (Figure 1). Epoprostenol is packaged as an off-white powder and is accompanied with a specific diluent. The drug was diluted to a concentration of 10 μg/mL and given as a continuous aerosol.
Four of our five patients were undergoing coronary artery bypass surgery (CABG) with or without valve replacement. The fifth underwent open adrenalectomy. All were managed with use of standard American Society of Anesthesiologists monitors, urinary bladder catheter, arterial catheter, pulmonary artery catheter (thermodilution catheter; Abbott, Critical Care Systems, Abbott Park, IL; List:41223-04-01), and transesophageal echocardiography (TEE). Data were compared by using a paired student’s t-test. Results were presented as mean ± sem with P values < 0.05 considered significant.
A 53-yr-old, 70-inch, 98-kg man with coronary artery disease (CAD), hypertension (HTN), and asthma experienced an inferior wall infarct 5 days previously and several ventricular fibrillation arrests during the intervening days. Cardiac catheterization revealed severe three-vessel disease and pulmonary artery pressures (PAP) of 48/22 mm Hg. TEE estimated his ejection fraction at 30%. Medications included lisinopril, isosorbide dinitrite, atenolol, furosemide, ranitidine, and pravastatin. He underwent CABG with two saphenous veins and a left internal mammary graft. He was separated from cardiopulmonary bypass (CPB) with the use of dopamine, milrinone, and norepinephrine, and anticoagulation was reversed with protamine. Cardiac index (CI) measured after CPB was 3.3 L · min−1 · m−2, PAP was 44/20 mm Hg, and pulmonary capillary wedge pressure was 19 mm Hg. Within 20 min, however, CI declined to 2.6 L · min−1 · m−2, and TEE showed hypokinesis and mild dilation of the right ventricle. Inhaled PGI2 was begun at this time in an attempt to reduce right ventricular afterload. CI increased from 2.6 to 3.15 L · min−1 · m−2, and PAP decreased slightly from 44/22 to 42/20 mm Hg. The pulmonary vascular resistance (PVR), however, decreased from 138 to 97 dynes · s · cm−5. The PGI2 was stopped because of difficulties in administering the drug during transport, and by arrival in the ICU, PVR had increased to 127 dynes · s · cm−5. PGI2 was restarted with a resultant decrease in PVR to 103 dynes · s · cm−5. The patient was weaned off of PGI2 after several hours, and his trachea was successfully extubated 8 h later. After discontinuation of treatment, PVR was 92 dynes · s · cm−5.
A 64-yr-old, 66-inch, 65-kg woman with HTN, CAD, unstable angina, and chronic obstructive lung disease (COPD) secondary to tobacco use had a 1-s forced expiratory volume and functional vital capacity that were 58% and 53% of predicted, respectively. Her medications included metoprolol, estrogens, aspirin, nitroglycerin, and heparin. She had undergone a CABG and now had significant stenoses of both vein grafts and the mammary graft and underwent repeat CABG. The prebypass period was remarkable for extreme instability, recurrent ventricular ectopy, and a large tear in the anterior ventricular wall on chest opening. After a prolonged CPB period, she was separated from CPB with great difficulty by using dopamine, milrinone, norepinephrine, epinephrine, and intraaortic balloon counterpulsation. Her CI after CPB was 2.6 L · min−1 · m−2 with PAP of 44/22 mm Hg and PVR of 154 dynes · s · cm−5, and TEE revealed a dilated and globally hypokinetic right ventricle. Inhaled PGI2 was initiated at that time because of impending right ventricular failure. CI increased to 3.7 L · min−1 · m−2, and the PVR decreased to 78 dynes · s · cm−5. The PAP was minimally altered (42/20 mm Hg). Three hours after separating from CPB and before chest closure, she developed severe dyskinesis of most of the anterior ventricular wall and septum. All vasoactive support was continued, and the grafts were confirmed to be open and to have adequate flow. The new wall motion abnormalities were felt to be secondary to the myocardial tear. She was transferred to the ICU and died 12 h later.
A 62-yr-old, 73-inch, 95-kg man presented with a history of COPD, HTN, and severe three-vessel CAD with moderate ventricular dysfunction. Tobacco use included smoking 3 packs per day for 25 yr, and pulmonary function testing revealed one-second forced expiratory volume and functional vital capacity of 44% and 54%, respectively. At cardiac catheterization, his ejection fraction was estimated to be 40% with inferior akinesis. Medications included metoprolol, aspirin, and albuterol multidose inhaler. He underwent CABG and was separated from CPB with the use of dopamine. CI after CPB was 3.2 L · min−1 · m−2, PAP was 44/12 mm Hg, and PVR 131 was dynes · s · cm−5. Therapy with inhaled PGI2 was initiated in response to right atrial and ventricular dilation noted on TEE. Twenty minutes after starting the drug, CI had increased to 4.1 L · min−1 · m−2, and PVR had decreased to 106 dynes · s · cm−5 with no change in PAP. The patient was transferred to the ICU and tracheally extubated successfully 7 h later.
A 70-yr-old, 76-inch, 113-kg man presented with adult-onset diabetes mellitus, nonspecific pancytopenia, chronic atrial fibrillation, CAD, severe mitral regurgitation, and congestive heart failure (New York Heart Association Class III). At cardiac catheterization, PAP was 42/17 mm Hg. He had severe two-vessel disease, apical and anterior hypo/akinesis, and an estimated ejection fraction of 30%. Medications included digoxin, atenolol, furosemide, glyburide, atorvastatin, ranitidine, losartan, heparin, sliding scale insulin, and nitroglycerin patch. Before initiation of CPB, PAP increased to 47/24 mm Hg and CI decreased to 1.86 L · min−1 · m−2 with a PVR of 143 dynes · s · cm−5. TEE revealed increased mitral regurgitation by color flow Doppler. Inhaled PGI2 was begun with an increase in CI to 2.34 L · min−1 · m−2, a small change in PAP (43/20 mm Hg), and a decrease in PVR to 100 dynes · s · cm−5. He underwent two-vessel CABG with mitral valve annuloplasty by using a ring prosthesis. The patient was separated from CPB with milrinone, norepinephrine, and dopamine. PGI2 was continued throughout the postbypass period and discontinued before extubation the next morning.
A 48-yr-old, 67-inch, 155-kg woman with a history of HTN, morbid obesity with hypoventilation syndrome of obesity, and an adrenal mass presented for adrenalectomy. Her serum catecholamines were marginally elevated but the mass was not thought to be a pheochromocytoma. She also had a history of tobacco use of 30 packs per year, but no use in the last 6 yr. Medications included quinapril and levothyroxine. She underwent an open adrenalectomy under general anesthesia. An epidural catheter was placed but not used during the procedure. After placement of the pulmonary artery catheter, mean PAP was 42 mm Hg and PVR was 118 dynes · s · cm−5 with a CI of 3.32 L · min−1 · m−2. After initiation of PGI2 therapy for attempted direct pulmonary vasodilation, mean PAP decreased to 37 mm Hg, PVR decreased to 67 dynes · s · cm−5, and CI increased to 3.92 L · min−1 · m−2. An attempt to discontinue therapy was made, but PVR increased to 124 dynes · s · cm−5, and mean PAP to 41 mm Hg. She was managed with PGI2 throughout the remainder of the procedure and overnight. She was tracheally extubated on postoperative day one.
Clinical hemodynamic data from our five patients are summarized in Table 1. Patients exhibited a mean decrease in PVR of 35% (P < 0.004), with a mean increase in CI of 26% (P < 0.003). This was accompanied by a small but significant decrease in mean PAP by 7% (P < 0.03) and systemic vascular resistance by 23% (P < 0.03). There were no significant changes in systemic arterial pressure or effective preload as measured by central venous pressure or pulmonary capillary wedge pressure. We did not detect any changes in oxygenation but did not monitor specifically for that other than pulse oximetry. In four of these patients, impending right ventricular failure rather than gas exchange or preexisting PHTN was the worrisome factor prompting the therapeutic intervention.
Clinical use of inhaled PGI2 or epoprostenol was proposed more than a decade ago (4–5). Various animal models, including toxin-induced PHTN and hypoxic pulmonary vasoconstriction models in rats, sheep, dogs, and pigs, have yielded results similar to those presented here (6–13). In addition, several case reports have been published concerning isolated patients treated with inhaled PGI2 in the critical care setting. These include infants with respiratory distress syndrome, idiopathic PHTN, persistent PHTN of the newborn, and one adult with PHTN resulting from acute-on-chronic pulmonary embolism (14–17). Larger prospective series have included adults with adult respiratory distress syndrome, septic shock, primary PHTN, PHTN secondary to severe connective tissue disorders or after heart transplantation and have included up to 16 patients. All have shown similar results (18–24). Mean PAPs decreased by 9% to 19%, and PVR by 29% to 38%. In one study of patients after transplantation, transpulmonary gradient decreased by 29%, and in another, shunt fraction decreased from 33.5% to 26% (20,24). Our results concur with these studies with a mean decrease in mean PAP and PVR of 7% and 35%, respectively.
As both NO and PGI2 are proposed as inhaled therapy for PHTN, it is important to note significant differences. NO and PGI2 act via different signaling pathways: NO via guanylyl cyclase and PGI2 via increasing intracellular cyclic adenosine monophosphate. NO and PGI2 are additive without increased toxicity (24–26). PGI2 is not avidly bound on entering the vascular space as is NO, but requires metabolic elimination in the liver. Therefore, its longer biologic half-life may carry the potential for systemic side effects. Significant systemic hypotension has not been reported, nor did we encounter this in our patients. However, one study addresses potential systemic effects. Sixteen patients with PHTN and septic shock were randomized to receive inhaled NO or inhaled PGI2, and gastric intramucosal pH was monitored as a measure of splanchnic perfusion. Pulmonary hemodynamics were similar in both groups, but PGI2 was able to significantly increase gastric pH (22). Given the morbidity and mortality thought to result from hypoperfusion of the gut in critically ill patients, this peripheral effect may be beneficial.
The potential for toxicity, including bronchoconstriction and bleeding complications, from pulmonary exposure to PGI2 has been of concern, and indeed, in vivo and in vitro platelet aggregation defects have been documented (27–28). However, in animal studies and available published series, no bleeding complications have been reported (29). We encountered no bleeding problems associated with use of PGI2. With the exception of Patient 2, neither fresh-frozen plasma or platelets nor additional protamine were administered, and no increased mediastinal drainage was noted. The issue of irritation of the airways in response to inhaled therapy with PGI2 is not supported by the literature or by our experience. No incidents of bronchospasm or increased inspiratory pressures were noted. Light and electron microscopy studies have confirmed the safety of PGI2 inhaled therapy in animals, and human trials of large-dose PGI2 (10 times the current dose) failed to cause significant bronchoconstriction (4–5,30).
Precise dosing of inhaled drugs is difficult. In the literature, various methods and dosing schedules are described. Several reports have documented that observed clinical response is dose-dependent in animals and humans, with an apparent ceiling of effect (6,8–9,20). Clinical administration is not standard, some centers dosing in ng · kg−1 · min−1, and others as μg/mL of nebulized solution given continuously. By using a standard, widely available nebulizer, 68% of the particles produced are less than 4 microns in diameter. Delivery of 0.2–0.3 mL/min is achieved with 4.5 L/min oxygen flow. According to the manufacturer, deposition of drug is 90% with lung retention of 50% (31). Doses listed in the literature range from 0.9 to 250 ng · kg−1 · min, with most clustered in the range of 1–50 ng · kg−1 · min−1. Delivering 0.25 mL/min of nebulized solution of 10 μg/mL yields a dose of 36 ng · kg−1 · min−1 for an average 70-kg adult. We choose to deliver inhaled PGI2 in the most simple, easily accessible manner possible. The nebulizer system is always available, and assembly requires no special equipment or significant amounts of time. The advantages of simple, easily adjustable drug delivery systems in the OR are obvious.
Indications for treatment with inhaled PGI2 include PHTN as well as impending or progressive right ventricular failure. In four of our patients, impaired right ventricular function was caused by a variety of factors including recent myocardial infarction (Patient 1), significantly decreased biventricular function (Patients 3 and 4), and prolonged CPB and cross-clamp times (Patient 2). This was combined with increased PVR caused by a variety of factors (chronic hypoventilation syndrome, COPD). Once severe right ventricular failure has manifested, success of treatment decreases dramatically. Hence, it is prudent to prevent progression by early intervention. The “usual” treatments (preventing hypoxemia, hypercarbia, hypothermia, acidosis, light anesthesia, administration of vasodilators) are attempted first. Three of our patients did not manifest increased pulmonary pressures, but did have high calculated PVR in association with low cardiac output. TEE findings showed evidence of right ventricular failure (hypokinesis, dilation, wall motion abnormalities) with near-normal pulmonary pressures, and in our judgement, this warranted attempts at optimizing oxygen supply and demand, thus, improving function. This, more than the degree of PHTN itself, was the motivating factor in these four patients. Myocardial performance immediately after CPB is in a state of flux, with disturbances in temperature and perfusion, ischemia-reperfusion injury, myocardial stunning, inadequate preservation, and derangements in vascular smooth muscle function all contributing (32). In addition, the right ventricle is especially susceptible to injury during the CPB and reperfusion periods (33–34). These factors are usually fairly short-lived, resolving within hours. Keeping the right ventricle unloaded as much as possible during this period can prevent a lethal cycle of right ventricle failure, decreasing cardiac output, decreased coronary perfusion, and progressive myocardial failure. This explains the relatively short period of treatment in these patients, terminating with successful extubation.
In summary, we report a series of patients in whom inhaled PGI2 therapy was used to treat PHTN and right ventricular failure in the OR. We used a simple, easily available, and quickly assembled apparatus to achieve reliable, effective reductions in PVR and mean PAPs with increases in cardiac performance and minimal systemic effects. Given the limited availability, cost, and technical complexity of using inhaled NO in the OR and the limited ability of IV drugs to effectively dilate the pulmonary vasculature, inhaled PGI2 provides a viable, effective alternative or perhaps a method to augment inhaled NO efficacy by using multiple metabolic pathways.
1. Presberg K, Wood LD. Critical Illness due to pulmonary hypertension and cor pulmonale. In: Hall JB, Schmidt GA, Wood LD, eds. Principles of critical care. New York: McGraw-Hill, 1992: 1456–75.
2. Tsukahara H, Ishida T, Mayumi M. Gas-phase oxidation of nitric oxide: chemical kinetics and rate constant. Nitric Oxide 1999; 3:191–8.
3. Silverman H. Pharmacologic approach in patients with pulmonary failure. In: Chernow B, ed. Essentials of critical care pharmacology. Baltimore: Williams and Wilkins, 1994: 114–38.
4. Hardy C, Bradding P, Robinson C, Holgate S. Bronchoconstrictor and antibronchoconstrictor properties of inhaled prostacyclin in asthma. J Appl Physiol 1988; 64:1567–74.
5. Hardy C, Robinson C, Lewis R, et al. Airway and cardiovascular responses to inhaled prostacyclin in normal and asthmatic subjects. Am Rev Respir Dis 1985; 131:18–21.
6. Hill L, Pearl R. Combined inhaled nitric oxide and inhaled prostacyclin during experimental chronic pulmonary hypertension. J Appl Physiol 1999; 86:1160–4.
7. Kleen M, Habler O, Hofstetter C, et al. Efficacy of inhaled prostanoids in experimental pulmonary hypertension. Crit Care Med 1998; 26:1103–9.
8. Welte M, Zwissler B, Habazettl H, Messmer K. PGI2 aerosol versus nitric oxide for selective pulmonary vasodilation in hypoxic pulmonary vasoconstriction. Eur Surg Res 1993; 25:329–40.
9. Booke M, Bradford D, Hinder G, et al. Effects of inhaled nitric oxide and nebulized prostacyclin on hypoxic pulmonary vasoconstriction in anesthetized sheep. Crit Care Med 1996; 24:1841–8.
10. Zwissler B, Welte M, Habler O, et al. Effects of inhaled prostacyclin as compared with inhaled nitric oxide in a canine model of pulmonary microembolism and oleic acid edema. J Cardiothoracic Vasc Anesth 1995; 9:634–40.
11. Zwissler B, Welte M, Messmer K. Effects of inhaled prostacyclin as compared with inhaled nitric oxide on right ventricular performance in hypoxic pulmonary vasoconstriction. J Cardiothoracic Vasc Anesth 1995; 9:283–9.
12. Walmrath D, Schermuly R, Pilch J, et al. Effects of inhaled versus intravenous vasodilators in experimental pulmonary hypertension. Eur Respir J 1997; 10:1084–92.
13. Zobel G, Dacar D, Rodl S, Friehs I. Inhaled nitric oxide versus inhaled prostacyclin and intravenous versus inhaled prostacyclin in acute respiratory failure with pulmonary hypertension in piglets. Pediatr Res 1995; 38:198–204.
14. 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.
15. Webb S, Stott S, van Heerden P. The use of inhaled aerosolized prostacyclin (IAP) in the treatment of pulmonary hypertension secondary to pulmonary embolism. Intensive Care Med 1996; 22:353–5.
16. Santak B, Schreiber M, Kuen P, et al. Prostacyclin aerosol in an infant with pulmonary hypertension. Eur J Pediatr 1995; 154:233–5.
17. Bindl L, Fahnenstich H, Peukert U. Aerosolised prostacyclin for pulmonary hypertension in neonates. Arch Dis Child Fetal Neonatal Ed 1994; 71:F214–6.
18. Mikhail G, Gibbs J, Richardson M, et al. An evaluation of nebulized prostacyclin in patients with primary and secondary pulmonary hypertension. Eur Heart J 1997; 18:1499–504.
19. Zwissler B, Kemming G, Habler O, et al. Inhaled prostacyclin (PGI2) versus inhaled nitric oxide in adult respiratory distress syndrome. Am J Respir Crit Care Med 1996; 154:1671–7.
20. Haraldsson A, Kieler-Jensen N, Ricksten S. Inhaled prostacyclin for treatment of pulmonary hypertension after cardiac surgery or heart transplantation: a pharmacodynamic study. J Cardiothorac Vasc Anesth 1996; 10:864–8.
21. van Heerden P, Blythe D, Webb S. Inhaled aerosolized prostacyclin and nitric oxide as selective pulmonary vasodilators in ARDS: a pilot study. Anaesth Intensive Care 1996; 24:564–8.
22. Eichelbronner O, Reinelt H, Wiedeck H, et al. Aerosolized prostacyclin and inhaled nitric oxide in septic shock: different effects on splanchnic oxygenation. Intensive Care Med 1996; 22:880–7.
23. Olschewski H, Walmrath D, Schermuly R. Aerosolized prostacyclin and iloprost in severe pulmonary hypertension. Ann Intern Med 1996; 124:820–4.
24. Walmrath D, Schneider T, Schermuly R, et al. Direct comparison of inhaled nitric oxide and aerosolized prostacyclin in acute respiratory distress syndrome. Am J Respir Crit Care Med 1996; 153:991–6.
25. Forrest I, Small T, Corris P. Effect of nebulized epoprostenol (prostacyclin) on exhaled nitric oxide in patients with pulmonary hypertension due to congenital heart disease and in normal controls. Clin Sci (Colch) 1999; 97:99–102.
26. Ikeda S, Shirai M, Shimouchi A, et al. Pulmonary microvascular responses to inhaled prostacyclin, nitric oxide, and their combination in anesthetized cats. Jpn J Physiol 1999; 49:89–98.
27. van Heerden P. Systemic levels of 6-keto-prostaglandin F1 alpha following administration of inhaled aerosolized prostacyclin. Anaesth Intensive Care 1997; 25:701–3.
28. van Heerden P, Gibbs J, Michalopoulos N. Effect of low concentrations of prostacyclin on platelet function in vitro. Anaesth Intensive Care 1997; 25:343–6.
29. Habler O, Kleen M, Zwissler B, et al. Inhalation of prostacyclin (PGI2) for 8 hours does not produce signs of acute pulmonary toxicity in healthy lambs. Intensive Care Med 1999; 22:426–33.
30. Habler O, Kleen M, Takenaka S, et al. Eight hours’ inhalation of prostacyclin (PGI2) in healthy lambs: effects on tracheal, bronchial, and alveolar morphology. Intensive Care Med 1996; 22:1232–8.
31. Haraldsson A, Kieler-Jensen N, Nathorst-Westfelt U, et al. Comparison of inhaled nitric oxide and inhaled aerosolized prostacyclin in the evaluation of heart transplant candidates with elevated pulmonary vascular resistance. Chest 1998; 114:780–6.
32. Mangano CM, Cartwright CR, Hindman BJ. Cardiopulmonary bypass and the anesthesiologist. In: Kaplan JA, Reich SN, Konstadt DL, eds. Cardiac anesthesia. Philadelphia: WB Saunders, 1999: 1061–110.
33. Fisk RL, Ghaswalla D, Guilbeau EJ. Asymmetrical myocardial hypothermia during hypothermic cardioplegia. Ann Thorac Surg 1082; 34:318–23.
34. Chen YF, Lin YT, Wu SC. Inconsistent effectiveness of myocardial preservation among cardiac chambers during hypothermic cardioplegia. J Thorac Cardiovasc Surg 1991; 102:684–7.