Inhaled nitric oxide as a therapy for pulmonary hypertension after operations for congenital heart defects

Transcription

1 Inhaled nitric oxide as a therapy for pulmonary hypertension after operations for congenital heart defects Seventeen infants were treated with inhaled nitric oxide for critical pulmonary artery hypertension after operations for congenital heart defects. In au 17 patients conventional medical therapy consisting of hyperventilation, deep sedation/analgesia, and correction of metabolic acidosis had failed. AlI children were monitored with a transthoracic pulmonary artery catheter inserted at operation. Pulmonary artery hypertension was defined as an acute rise in pulmonary pressure associated with a decrease in oxygen arterial or venous saturation. After failure of conventional medical therapy, 20 ppm of inhaled nitric oxide was administered to the patient. In au patients the pulmonary pressures decreased (mean pulmonary arterial pressure decreased by -34 % ± 21 % ) without significant change in systemic arterial pressure, whereas the oxygen arterial saturation and oxygen venous saturation increased by 9.7% ± 12% and 37% ± 28%, respectively. Fifteen children were discharged from the intensive care unit at 10 ± 6 days (range 3 to 26 days) and two died. This study demonstrates that inhaled nitric oxide exerts a selective pulmonary vasodilation without decreasing systemic arterial pressure in children with congenital heart disease. The increased values of mixed venous oxygen saturation and urinary output suggest that this selective lowering of pulmonary vascular resistance iqlproved the overall hemodynamics. The potential toxic effects of nitric oxide and nitrogen dioxide necessitate careful consideration of the risks and benefits of inhaled nitric oxide therapy. (J THoRAc CARDIOVASC SURG 1994;107:1129-) Didier Journois, MD,a Philippe Pouard, MD,a Philippe Mauriat, MD,a Thierry Malhere, MD,a Pascal Vouh6, MD,b and Denis Safran, MD,a Paris, France Ater neonatal cardiac operations for certain congenital heart defects, a critical rise in pulmonary artery pressure (PAP) may occur. 1 3 Despite aggressive therapy with manual hyperventilation, deep levels of analgesia and sedation, muscle paralysis and vasodilator therapy, the mortality remains usually high when major hypertensive crises occur.2, 3 Furthermore, pulmonary artery vasodilator therapy is often nonselective and may produce systemic hypotension, partially contributing to the morbidity.2-4 From the Departments of Anesthesia and Intensive Care Medicine a and,b H6pital Laennec, Paris, France. Received for publication Feb. 22, Accepted for publication Oct. 5, Address for reprints: Didier lournois, MD, H6pital Laennec, 42, rue de Sevres, 3 Paris, France. Copyright 1994 by Mosby-Year Book, Inc /94 $ /1/52225 Recent reports have demonstrated that nitric oxide is a selective pulmonary vasodilator in adults 5, 6 and in newborn infants. 7 The usefulness of nitric oxide has been suggested by several isolated case reports after pediatric cardiac operations We therefore systematically evaluated the hemodynamic changes during inhalation of low concentrations of nitric oxide in the treatment of postoperative pulmonary hypertensive crises after repair of congenital heart defects. Patients and methods A pulmonary arterial catheter was inserted transthoracically via the right ventricular myocardium in 42 children thought to be at risk for postoperative pulmonary hypertension.! Twenty of these catheters were 4F fiberoptic pulmonary artery catheters that allowed continuous mixed venous oxygen saturation measurement (SV02; Oximetrix Lab., Abbott Laboratories, North Chicago, 111.).11 After operation, central venous and left atrial pressures, systemic arterial pressure, systemic arterial saturation by pulse oximetry, and body temperature were continuously monitored. Blood gas analysis was done at least every 6 hours. 1 I 29

2 I I Journois et al. The Journal of Thoracic and April 1994 Table I. Demographic data Weight Stay in/cu Extubation timing Patient No. Diagnosis Age (kg) Survival Complication (days) (days) I Total APVR 6 days 3.2 Yes No VSD lib II rno 6.5 Yes No TotalAPVR days 3.0 Yes No Total APVR 5 days 3.0 Yes Infection TA+VSD 20 days 3.4 Yes No TotalAPVR days 3.6 Yes No 3 7 APVR + VSD + IAA 10 days 4.5 Yes No VSD + IAA 15 days 3.1 Died Tracheornalacia 9 9 AVCD 2rno 4.0 Yes No AVCD 3 rno 4.5 Died Infection 12 II LAO 18 rno 9.2 Yes Infection VSD days 3.5 Yes No TotalAPVR 5 days 3.2 Yes No AVCD 4rno 4.9 Yes No AVCD 2 yr 12 Yes No TA 14 days 3.3 Yes Infection VSD 2rno 3.7 Yes No 5 3 Mean ± SD 4.6 ± ± 6 6±6 ICU. Intensive care unit;apvr, anomalous pulmonary venous return; VSD, ventricular septal defect; T A, truncus arteriosus; IAA, interrupted aortic arch; AVCD, atrioventricular canal defects; LAO, left atrium obstruction after a Senning operation; SD, standard deviation. The arterial pulse oximetry was continuously monitored. The lungs were ventilated with a Siemens Servo 900C ventilator (Siemens-Elema, Division of Elema-Schiinander, Inc., Solna, Sweden) with an inspired oxygen fraction (Fio2) of 1 and slightly hyperventilated to reach an arterial carbon dioxide tension of approximately mm Hg. Continuous infusions of pancuronium and fentanyl were administered during the first 12 postoperative hours. Dobutamine or dopamine, or both, infused at rates less than or equal to 5 f.lg kg-i. min- 1 were used as inotropic agents in several patients at the end of cardiopulmonary bypass. Conventional management of acute pulmonary hypertension crisis was instituted if PAP exceeded % of systemic pressure in association with a decrease in SV02 and/or arterial (Sao2) oxygen saturation. This therapy included 100% oxygen with manual ventilation, deepening the level of anesthesia, and correction of the metabolic acidosis with sodium bicarbonate administration. After 5 minutes of unsuccessful therapy, inhaled nitric oxide was administered. The measured variables during nitric oxide administration were heart rate; systolic, diastolic, and mean systemic arterial pressures (AP); systolic, diastolic, and mean PAPs; central venous pressure; left atrial pressure; urinary output; and venous and arterial blood gases. A complete set of measurements was recorded before (To) and 20 minutes after administration of 20 ppm of inhaled nitric oxide (Tl). Nitric oxide administration was continued until pulmonary artery hypertension was absent for 6 hours in the face of tracheal suctioning, progressive weaning of sedation, and progressive reduction of Fio2. Persistence of a low SV02, a low Sao2 or a high mean PAP/mean AP ratio was treated by increasing the inspired nitric oxide fraction from 20 ppm to ppm. Further sets of measurements were recorded every hour or in the event of paroxysmal pulmonary hypertension crisis. Methemoglobin concentration was measured twice daily. A tank containing 225 ppm of nitric oxide in nitrogen (Compagnie Fran~aise des Produits Oxygenes, Paris, France) was connected to a specially designed low-flow blender. The mixture of nitric oxide and nitrogen was continuously delivered through a needle inserted at the end of the inspiratory limb into the breathing circuit. A 1.0 (16 cases) or a 1.5 mm (1 case) external diameter catheter was inserted into the tracheal tube and positioned 1 cm from its distal extremity. The catheter was connected to an oxygen/nitrogen analyzer (Servo gas monitor 120, Siemens) via an electrochemical nitric oxide/nitrogen dioxide analyzer (Polytron NO/N02, Drager, Antony, France), which allowed continuous measurements of Fi02, nitric oxide, nitrogen dioxide, and nitrogen. The gas aspiration rate for analysis of 105 ml. min- 1 was compensated by an equivalent increase in ventilation. The flow of the nitric oxide/ nitrogen mixture was adjusted to achieve the desired inhaled nitric oxide concentration. After extubation, if PAP remained elevated, nitric oxide was given by means of a facial mask. Expired and analyzer's gases were scavanged. This investigation was done with approval of the hospital review board and informed consent was obtained from each infant's family. Data are expressed as mean ± standard deviation and were compared between To and Tl by Wilcoxon test for paired values. Further measurements were compared by analysis ofvariance for repeated measures followed by a two-tailed Dunnett's post-hoc test with To as control value. Comparisons between the two patient groups used Wilcoxon or Fisher's exact test. The relationship between nitric oxide and nitrogen dioxide concentrations was studied by linear regression. A p value less than 0.05 was considered statistically significant. Results Seventeen children with pulmonary artery hypertension, from 5 days to 24 months of age (median days), in whom conventional medical therapy had failed, were

3 The Journal of Thoracic and Volume 107, Number 4 Journois et ai, spap mpap ~ 25 T T1 T1 Fig. 1. Evolution of PAPs between To (before nitric oxide) and TJ (after nitric oxide). spap, Systolic PAP; mpap, mean PAP; dp AP, diastolic PAP. sap map 95 dap T1 T1 T1 Fig. 2. Evolution of systemic APs between To (before nitric oxide) and TJ (after nitric oxide). sap, Systolic AP; map, mean AP; dap, diastolic AP. studied. Demographic data and description of congenital heart diseases are reported in Table I. Inhaled nitric oxide resulted in significant decreases in systolic PAP (-32% ± 16%), mean PAP (-34% ± 21 %), and diastolic PAP (-43% ± 20%) (Fig. I), whereas no significant change was observed in systolic, mean, or diastolic AP (Table II, Fig. 2). Significant increases were observed in Sao2 (+9.7% ± 12%) and SV02 ( + 37% ± 28%) despite a reduction of Fio2 from 1 to 0.92 ± 0.1 as a result of inhaled nitric oxide administration (Fig. 3 and Table III). The mean PAP/mean AP ratio decreased from 0.76 ± 0.21 to 0.49 ± 0.18 at Tl (-34%; Fig. 4). Urinary output increased from 1.4 ± 1.2 to 2.8 ± 1.3 ml. kg-i. hr- 1 at the third hour after the institution of inhaled nitric oxide (p < 0.05). In three children, a dramatic decrease in mean AP was observed (Fig. 2) in association with an increase in Sao2, SV02, and urinary output. A secondary rise in PAPs was ob-

4 J oumois et al. The Journal of Thoracic and April % Sa02 Sv02 T1 T1 Fig. 3. Evolution of Saoz and SV02 between To (before nitric oxide) and Tl (after nitric oxide). 70 Y served during sedation weaning in six children. Increasing of inhaled nitric oxide concentration from 20 ppm to ppm failed to normalize PAPs in five of these six. Patients were treated by inhaled nitric oxide for 44 ± 69 hours (range 4 to 2). All patients were eventually weaned from inotropic drugs, vasodilators, highlevel sedation, and high Fio 2 In three patients pulmonary infections subsequently developed after the initial period of hemodynamic instability ended. Nine of the children recovered and two children died (one of infection and the other of upper respiratory tract problems). In one child, the autopsy revealed a severe pulmonary infection. Except for minor suction wounds, the trachea was histologically normal. Fifteen children were discharged from the intensive care unit at 10 ± 6 days (range 3 to 26 days). In 238 measurements near the tracheal outlet of the endotracheal tube, nitric oxide concentration was 22.2 ± 14 ppm (range 8.5 to ) and nitric oxide 1.41 ± 1.8 ppm (range 0.1 to 8.2). Linear regression showed a strong relationship between nitric oxide and nitrogen dioxide concentrations (nitrogen dioxide = X nitric oxide - 0.1; P = ). The highest methemoglobin concentration was 1.8%. No other side effects were observed. Table II. Evolution of hemodynamic data between To and T] To TJ (before NO) (20 min after NO) p Value HR (beats. min-l) 153 ± 22 1 ± Systolic PAP (mm ± 14 ± Mean PAP (mm 42 ± ± Diastolic PAP (mm 31 ± ± Systolic AP (mm 68 ± ± Mean AP (mm ± ± Diastolic AP (mm 44 ± ± LAP () 7.2 ± ± CVP () 6.6 ± ± SV02 (%) ± ± Sao2 (%) 87 ± 8 96 ± Data are expressed as means ± standard deviation. NO, Nitric oxide; HR, heart rate; PAP, systolic pulmonary arterial pressure; LAP, left atrial pressure; CVP, central venous pressure; SV02, mixed venous blood oxygen saturation; Sao2, arterial oxygen saturation. Discussion This study confirms that inhaled nitric oxide exerts a selective pulmonary vasodilation without decreasing systemic AP even in neonates and infants with pulmonary artery hypertension after repair of congenital heart disease. The improvement in SV02 and urinary output suggests that the selective lowering of pulmonary vascular resistance promoted a better hemodynamic condition. Because oxygen consumption and hemoglobin levels are likely to remain nearly constant, this SV02 increase is probably caused by both increases in Sao2 and cardiac output. Although cardiac output was not measured, an improvement in the right ventricular function is probably responsible for this effect. l This phenomenon, in association with an increased Sao2, might improved arterial oxygen transport. Furthermore, inhaled nitric oxide may specifically dilate pulmonary vessels next to ventilated alveoli, enhancing ventilation-perfusion matching. Prolonged postoperative anesthesia has been demonstrated to control episodes of pulmonary artery hypertension in response to tracheal stimulation. l2 This technique requires several days for pulmonary artery reactivity to decrease and may facilitate pulmonary infections. In our study, inhaled nitric oxide seemed to be effective as a sole therapy and allowed discontinuation of vasodilators, inotropic support, and lightening of anesthesia. We speculate that this therapy may reduce durations of ventilatory support and intensive care unit stay in patients at risk for pulmonary artery hypertension.

5 The Journal of Thoracic and Volume 107, Number 4 Journois et al. 1 I 3 3 Table III. Evolution of blood gases and respiratory data between To and TJ To (before NO) TJ (20 min after NO) p Value Fioz 0.99 ± ± O.oI5 < Bicarbonate 25.6 ± ± (mmoljl) Paco2 (mm 24.8 ± ± ph ± ± Respiratory rate 43.9 ± ± (counts/min) Pressure plateau (em 23. ± ± 3. < H2O) Pressure peak (em ± ± 3.9 < H2O) Minute ventilation 2917± ± (mljrnin) Comparisons with paired Student's I test. Data are expressed as means ± standard deviation. Paeo], Arterial carbon dioxide tension; 0.9 ~ m vs, P = Fig. 4. Evolution of mean PAP / mean AP ratio (mp AP/mAP) over time after inhaled nitric oxide administration. Data are expressed as mean and standard deviation. * p < 0.05 versus To..' An important characteristic of pulmonary artery hypertension is its paroxysmal nature. I This may be due to the auto-intensification of pulmonary artery hypertension, mediated by hypoxemia, that is a major cause of pulmonary vasoconstriction in young children. 13 Pulmonary artery hypereactivity seems to persist several days after a complete surgical correction that should eliminate the cause of pulmonary hypertension. 3 This phenomenon suggests the coexistence of an abnormality in arterial reactivity induced by the congenital heart disease. 14 It seems to occur in association with an increase in medial thickness resulting from preexisting elevated pulmonary arterial pressure. 14,15 Furthermore, hypoxemia and increased pulmonary blood flow induce the persistence of fetal pulmonary vascular muscularization in neonates. 16 Abnormalities in endothelin-derived relaxing factor or in endothelin releaseshave also been suggested. 17, 18 Most of these abnormalities usually do not persist more than a few days after surgical correction. 3, 18 A therapeutic aim of inhaled nitric oxide administration, therefore, is stabilization until pulmonary arterial reactivity is reduced after operation. 1 Oxygen inhalation is an effective pulmonary vasodilator when pulmonary hypertension is due to a hypoxic stimulation of precapillary arteriolar vasculature. 4 The reduction in Fio2 that inhaled nitric oxide allows is of particular interest in neonates in whom high Fio2s are deleterious. Correction of acidosis is also necessary to repress the pulmonary arterial reactivity. I This is achieved by prevention of low cardiac output, which inhaled nitric oxide provides. Many pharmacologic agents have been reported to have pulmonary vasodilating properties, including a-blockers,19 tolazoline,4 (3-agonists,2D nitrates,21,22 calcium channel blockers,23 and.prostaglandins. 24 All these agents induce tissue edema and a systemic vasodilation. I, 7, 25 These deleterious effects can be partially counteracted by combining their infusion with norepinephrine. 24 Furthermore, the frequent association of atrial or ventricular right-to-left shunts during pulmonary hypertension crisis leads to shunting of increased levels of the administered vasodilators toward the systemic circulation.! This may lead to a predominant systemic vasodilation that is particularly undesirable inasmuch as left ventricular output is simultaneously impaired by the pulmonary hypertension-induced preload reduction. Ultimately, when pulmonary hypertension crisis persists despite the optimal use of these pharmacologic agents, extracorporeal membrane oxygenator support has been proposed as a last resort to avoid the complications of these therapies.! Nitric oxide produces vasodilation by directly activating guanylate cyclase, which increases the intracellular cyclic guanosine 3,5-monophosphate levels in smooth muscle cells and produces vasodilation. 26 The rapid nitric oxide inactivation by the hemoglobin contained in the pulmonary blood vessels 27 explains its short-acting and local effects. Inhaled nitric oxide doses in human beings appear safe. 5, 7, 8, 25, 28, 29 Nevertheless, nitric oxide is known to be a toxic gas. 3D Major toxic effects have been described with high nitric oxide concentrations (,000 ppm).3! They involve bronchial and tracheal damage,3! methemoglobinemia,3! and nitrogen dioxide and peroxynitrite pro-

6 lournois et al. The Journal of Thoracic and April 1994 duction. 32, 33 Methemoglobinemia levels were low in our study. Because exposure to only 2 ppm of nitrogen dioxide during 24 hours induces ciliary depletion and bronchial dysplasia,34 maximal continuous inhaled nitric oxide concentration should be limited to 20 ppm or strictly restricted to crisis treatment. Administration of nitric oxide can be beneficial in treatment of patients with certain congenital heart defects. The small number of patients in this study does not permit any subgroup analysis to determine which congenital heart defects are appropriate for therapy with inhaled nitric oxide. At low doses, inhaled nitric oxide is effective, simple to use, and had no serious adverse effect. Inhaled nitric oxide has therapeutic potential in postoperative pulmonary artery hypertension after operations for congenital heart disease. Nevertheless, risks and benefits of inhaled nitric oxide need careful consideration. Further tolerance studies are required. We thank W. 1. Greeley, MD, Associate Professor of Anesthesiology and Pediatrics, Duke University Medical Center, Durham, N.C., for his valuable advice and comments; and the staff and the nurses of our Intensive Care Unit for their support. REFERENCES I. Hopkins R, Bull C, Haworth S, de Leval MR, Stark 1. Pulmonary hypertensive crises following surgery for congenital heart defects in young children. Eur 1 Cardiothorac Surg 1991;5: nes 0, Shore D, Rigby M, et al. The use of tolazoline hydrochloride as a pulmonary vasodilator in potentially fatal episodes of pulmonary vasoconstriction after cardiac surgery in children. Circulation 1981;64: Wheller 1, George B, Muller D, larmakani 1. Diagnosis and management of postoperative pulmonary hypertensive crisis. Circulation 19;: Goetzman B, Sunshine P, 10hnson 1, et al. Neonatal hypoxia and pulmonary vasospasm response to tolazoline. 1 Pediatr 1976;89: Rich G, Murphy G, Roos C, 10hns R. Inhaled nitric oxide: selective pulmonary vasodilatation in cardiac surgical patients. Anesthesiology 1993;78: Pepke-Zaba 1, Higenbottam T, Dinh-Zuan A. Inhaled nitric oxide as a cause of selective pulmonary vasodilatation in pulmonary hypertension. Lancet 1991;338: Roberts 1, Polaner D, Lang P, Zapol W. Inhaled nitric oxide in persistent pulmonary hypertension of the newborn. Lancet 1992;3: Sellden H, Winberg P, Gustafsson L, Lundell B, Book K, Frostell C. Inhalation of nitric oxide reduced pulmonary hypertension after cardiac surgery in a 3.2 kg infant. Anesthesiology 1993;78: Berner M, Beghetti M, Ricou B, Rouge 1, Pretre R, Friedli B. Relief of severe pulmonary hypertension after closure of a large ventricular septal defect using low dose inhaled nitric oxide. Intensive Care Med 1993;19: Haydar A, Mauriat P, 10urnois D, Pouard P, Safran D, Vouhi: P. Inhaled nitric oxide for postoperative pulmonary hypertension in patients with congenital heart defects [Abstract). Lancet 1992;3: Rah K, Dunwiddie W, Lower R. A method for continuous postoperative measurement of mixed venous saturation in infants and children after open heart procedures. Anesth Analg 1984;63: Hickey P, Hansen D, Wessel D, Lang P, 10nas R, Elixson E. Blunting of stress responses in the pulmonary circulation in infants by fentanyl. Anesth Analg 1985;64: lames L, Rowe R. The pattern of response of pulmonary and systemic arterial pressures in newborn and older infants to short periods of hypoxia. 1 Pediatr 1957;51: Haworth S. Normal pulmonary vascular development and its disturbance in congenital heart disease. In: Goodman M, ed. Pediatric cardiology. New York: Churchill-Livingstone, 1981: Meyrick B, Reid L. Ultrastructural findings in lung biopsy material from children with congenital heart defects. Am 1 PathoI19;101: Emmanouilides G, Moss A, Duffie E. Pulmonary arterial pressure changes in human newborn infants from birth to 3 days of age. 1 Pediatr 1964;: Abman S, Chatfield B, Hall S, McMurtry I. Role ofedrf during transition of pulmonary circulation at birth. Am 1 Physiol 1 990;259:H Yoshibayashi M, Nishioka K, Nakao K, et al. Plasma endothelin concentrations in patients with pulmonary hypertension associated with congenital heart defects. Circulation 1991;84: Artman M, Parrish M, Boerth R, Boucek R, Graham T. Short-term hemodynamic effects of hydralazine in infants with complete atrioventricular canal defects. Circulation 1984;69: Pietro D, Bresh KL, Shulman R, Folland E, Parisi A, Sasahara A. Sustained improvement in primary pulmonary hypertension during six years of treatment with sublingual isoproterenol. N Engl 1 Med 1984;310: Damen 1, Hitchcock 1. Reactive pulmonary hypertension after a switch operation: successful treatment with glyceryl trinitrate. Br Heart ;53: Bixler T, Gott V, Gardner T. Reversal of experimental pulmonary hypertension with sodium nitroprusside. 1 THO RAC CARDIOVASC SURG 1981;81: De Freyter Pl, Kerkkamp H, long ld. Sustained beneficial effect of nifedipine in primary pulmonary hypertension. Am Heart ;105: D' Ambra M, Rain PL, Philbin D, Watkins W, Hilgenberg A, Buckley M. Prostaglandin EI: a new therapy for refractory right heart failure and pulmonary hypertension after mitral valve replacement. 1 THORAC CARDIOV ASC SURG 1985;89: Kinsella 1. Low-dose inhalational nitric oxide in persistent

7 The Journal of Thoracic and Volume 107, Number 4 lournois et al pulmonary hypertension of the newborn. Lancet 1992; 3: Ignarro L. Biological actions and properties of endothelialderived nitric oxide formed and released from artery and vein. Circ Res 1989;: Gruetter C, Gruetter D, Lyon J, Ignarro L. Relationship between cyclic guanosine 3,5-monophosphate formation and relaxation of coronary arterial smooth muscle by glyceryl trinitrate, nitroprusside, nitrite and nitric oxide. J Pharmacol Exp Ther 1981;219: Girard C, Lehot J, Pannetier J, Filley S, French P, Estanove S. Inhaled nitric oxide after mitral replacement in patients with chronic pulmonary artery hypertension. Anesthesiology 1992;77: Frostell C, Blomqvist H, Hedepstierna G, Lundberg J, Zapol W. Inhaled nitric oxide selectively reverses human hypoxic pulmonary vasoconstriction without causing systemic vasodilation. Anesthesiology 1993;78: Clutton-Brock J. Two cases of poisoning by contamination of nitrous oxide with higher oxides of nitrogen during anaesthesia. Br J Anaesth 1967;39: Greenbaum R, Bay J, Hargreaves M, et al. Effects of higher oxides of nitrogen on the anaesthetized dog. Br J Anaesth 1967;39: Austin A. The chemistry of the higher oxides of nitrogen as related to the manufacture, storage and administration of nitrous oxide. Br J Anaesth 19;39: Beckman J, Beckman T, Chen J, Marshall P, Freeman B. Apparent hydroxyl radical production by peroxynitrite: implications for endothelial injury from nitric oxide and superoxide. Proc Natl Acad Sci USA 1990;87: Stephens R, Freeman G, Evans M. Early response of lungs to low levels of nitrogen dioxide. Arch Environ Health 1972;24: 1-79.