Elevated immunoreactive endothelin-1 levels in newborn infants with persistent pulmonary hypertension

Elevated immunoreactive endothelinlevels in newborn infants with persistent pulmonary hypertension A d a m A. Rosenberg, MD, Jan K e n n a u g h , MD, Stacia L. K o p p e n h a f e r , M a r y Loomis, RN, Barbara A, C h a t f i e l d , MD, a n d S t e v e H. A b m a n , MD From the Divisions of Neonatology and Pulmonary Medicine, Department of Pediatrics, University of Colorado School of Medicine and Children's Hospital, Denver To study the potential role of endothelin-1, a potent endothelium-derived vasoconstrictor peptide, in the p a t h o p h y s i o l o g y of persistent pulmonary hypertension of the newborn (PPHN), we measured arterial concentrations of immunoreactive endothelin-1 (irET-1) in 24 neonates with PPHN. Secondary diagnoses i n c l u d e d meconium aspiration syndrome (13 patients), sepsis (2), c o n g e n i t a l d i a p h r a g m a t i c hernia (1), asphyxia (1), pulmonary hemorrhage (I), aspiration of b l o o d (1), and respiratory distress syndrome (1), C o m p a r e d with irETT1 levels in umbilical cord b l o o d in normal infants (15.1 _+ 4.1 pg/ml; m e a n + SEM) and in newborn infants with hyaline m e m b r a n e disease who were s u p p o r t e d by m e c h a n i c a l ventilation (11.8 • 1.2 pg/ml), infants with PPHN had markedly elev a t e d circulating irET-1 levels (27.6 + 3.6 pg/ml; p <0.01 vs cord blood, hyaline m e m b r a n e disease). Infants with severe PPHN requiring extracorporeal membrane o x y g e n a t i o n (ECMO) therapy had higher irET-1 levels than infants with milder disease (31.0 _+ 4.7 for ECMO-treated infants vs 21.2 • 2.0 for non-ECMOtreated infants; p <0.05). In patients treated without ECMO, irET-1 progressively d e c r e a s e d during the following 3 to 5 days, paralleling clinical improvement. In contrast, irET-1 concentrations remained e l e v a t e d in infants with severe PPHN during ECMO therapy. We c o n c l u d e that circulating irET-1 levels are e l e v a t e d in newborn infants with PPHN, are positively correlated with disease severity, and d e c l i n e with resolution of disease in patients who do not require ECMO therapy. Whether endothelin-1 contributes directly to the p a t h o p h y s i o l o g y of PPHN or is simply a marker of disease activity remains speculative. (J PED|ATR1993;123:109-14)

Persistent pulmonary hypertension of the newborn infant is characterized by marked elevation of pulmonary vascular resistance, causing right-to-left shunting at the foramen ovale, ductus arteriosus, or both and severe hypoxemia, l' 2 PPHN is a clinical syndrome associated with a wide variSupported by grants HL46481, HL41012, and M01 RR0069 from the National Institutes of Health, a March of Dimes basic research grant, and the Kempe Center at Children's Hospital. Submitted for publication Nov. 5, 1992; accepted Feb. 5, 1993. Reprint requests: Adam A. Rosenberg, MD, Associate Professor, Divisionof N eonatology(B070), Children's Hospital, 1056 E. 19th Ave., Denver, CO 80218-1088. Copyright | 1993 by Mosby-Year Book, Inc. 0022-3476/93/$1.00 + .10 9/23/46189

ety of neonatal cardiopulmonary disorders, and its pathogenesis and pathophysiology are incompletely understood. Despite the diversity of clinical settings, the pathophysiology of PPHN can be characterized by the presence of variable degrees of vascular remodeling, decreased arterial number, and active vasoconstriction.3, 4 However, mechanisms contributing to the abnormal vasoreactivity and structural changes of PPHN remain unclear. During the past decade, experimental and clinical studies have demonstrated the important role of the endothelium in regulation of vascular tone and structure in normal and hypertensive circulations.5, 6 In addition to releasing vasodilators, such as prostacyclin and endothelium-derived relaxing factor, the endothelium also releases vasoconstrictors, including a re

109

110

Rosenberg et al.

ANOVA ECMO EDRF HFJV HMD irET- 1 OI PPHN P(A-a)o2

Analysis of Variance Extracorporeal membrane oxygenation Endothelium-derived relaxing factor High-frequency jet ventilation Hyaline membrane disease Immunoreactive endotbelin-1 Oxygenation index Persistent pulmonary hypertension of the .newborn Alveolar-arterial difference in partial pressure of oxygen

cently described peptide, endothelin-1.7-9 Endothelin-1 is a potent vasoconstrictor, is mitogenic for vascular smooth muscle cells, l~ and heightens vasoreactivity by augmenting the effects of other constrictor agonists at subthreshold concentrations of endothelin-1.11 Increased stretch, pressure, thrombin, inhibition of EDRF, and other stimuli have been shown to increase release of endothelin-1.12' 13 Although its role in regulation of tone in the perinatal pulmonary circulation remains unclear, endothelin-I infusions cause tone- and time-dependent hypertension in the ovine fetus]4, 15 and endothelin-1 level is elevated in umbilical cord blood from asphyxiated newborn infants. 16In addition, high levels of immunoreactive endothelin-1 have been reported in adults with pulmonary hypertension.17 Whether endothelin-1 contributes to the pulmonary hypertension, increased vasoreactivity, or structural remodeling of P P H N is not known. To determine whether endothelin-1 contributes to the pathophysiology of PPHN, we measured arterial irET-1 levels in newborn infants with mild and severe clinical disease, and compared these levels with measurements of blood samples obtained from neonates with hyaline membrane disease, but without pulmonary hypertension, and with measurements of umbilical cord samples from normal infants. To examine further the relationship of endothelin- 1 to the severity of PPHN, we measured irET-1 levels in serial arterial blood samples from patients with P P H N who required extracorporeal membrane oxygenation therapy and from those whose disease resolved without ECMO. METHODS This study was approved by the human subjects committees of the University of Colorado Health Sciences Center and Children's Hospital, Denver. After obtaining informed consent, we collected serial blood samples from a group of 24 infants with a clinical diagnosis of PPHN. The diagnosis of P P H N was characterized by persistent hypoxemia in the presence of associated clinical conditions (e.g., meconium aspiration syndrome, sepsis, asphyxia) and without evidence of structural heart disease. In all cases the diagnosis was confirmed by echocardiographic evidence of in-

The Journal of Pediatrics July 1993

creased pulmonary artery pressure with right-to-left shunting at the level of the foramen ovale, ductus arteriosus, or both. Initial samples were collected a t the time of diagnosis from an indwelling arterial line (umbilical, radial, or posterior tibial arteries) on the first day of life. All samples were obtained between 12 and 24 hours of ag e. o f the 24 infants, 12 required treatment with ECMO. Criteria for ECMO at Children's Hospital, Denver, included weight >2000 gm, gestational age >34 weeks, --<7 days of assisted ventilation with reversible lung disease, failure of maximal medical support (alkalinization, pressor support, vasodilator trial), no intracranial hemorrhage, no severe asphyxia, and no multisystem disease. In addition, one of the following was required: an alveolar-arterial oxygen gradient _>500 on two determinations 6 hours apart or an oxygenation index >__40, or a ventilat0ry index (mean airway pressure x rate) _> 1500, or uncontrolled air leak. Subsequent samples were obtained after 24 hours of ECMO, on days 5 and 7, and after ECMO was terminated. Non-ECMO-treated infants had samples collected on days 1 (between 12 and 24 hours of age), 3, and 5. In addition, samples were obtained on day 1 (between 12 and 24 hours of age) from 8 infants given ventilatory support for surfactant deficiency and 10 well infants (cord blood) who were products of normal, spontaneous vaginal deliveries. Clinical information was obtained on al ! patients by chart review. Samples of blood (1 ml) were collected in tubes of ethylenediaminetetaacetic acid and spun down immediately at 2500 rpm in a refrigerated centrifuge; plasma was separated and frozen at - 7 0 ~ C. Samples were run in batches. Plasma levels of endothelin-1 were measured with slight modification of a commercial radioimmunoassay kit (Peninsula Laboratories, Inc., Belmont, Calif.). SepPak C18 cartridges (Waters Chromatography Division, Millipore, Milford, Mass.) were prepared with serial 10 ml washes of methanol and 0.1% trifiuoroacetic acid. At assay, samples were thawed and extracted by injecting 0.5 ml of plasma into the cartridges, which were then washed with 5 ml of 0.1% trifluoroacetic acid, deionized water, hex~me, and 1% methanol ethyl acetate, and eluted with 2 ml of 60%/40% acetonitrile and 0.1% trifluoroacetic acid. Eluted samples were concentrated by evaporation (SpeedVac; Savant Instruments, Inc., Farmingdale, N.Y.), reconstituted in 0.25 ml radioimmunoassay buffer, and incubated for 24 hours with rabbit anti-endothelin- 1 serum at 4 ~ C. The addition of endothelin- 1 labeled with iodine 125 was followed by a second 24-hour incubation. Bound and free radioligands wer e separated by use of the second antibody method. Bound radioactivity data were evaluated after logit-log transformation. The sensitivity of the radioimmunoassay was 1 pg per tube. Recovery of authentic endothelin- 1 added to serum samples was consistently >80%. Dilution of known endothelin-1

The Journal o f Pediatrics Volume 123, Number 1

R o s e n b e r g et al.

111

T a b l e I. Clinical data in patient groups HMD 8)

PPHN 12)

(n =

Gestational age (wk) Birth weight (gm) Sex ( M / F ) Diagnosis

34.8 + 0.5 2433 _+ 217 7/1 HMD = 8

Oxygenation index (cm H 2 0 / m m Hg) P(A-a)Oz (mm Hg)

10.3 • 1.3 434 • 32

PPHN with ECMO ( n = 12)

(n =

39.6 _+ 0.4* 3130 • 134t 9/3 MAS = 5 Idiopathic PPHN = 3 CDH = 1 Asphyxia = 1 Sepsis = 1 Blood aspiration = 1 14.2 + 2.3 471 + 8

39.6 • 0.5* 3329 • 167" 3/9 MAS = 8 HMD = 1 Idiopathic PPHN = 1 Sepsis = I Pulmonary hemorrhage = 1 49.0 • 5.9",:~ 505 • 11w

Values are expressed as mean + SEM. MAS, meconium aspiration syndrome; CDH, congenital diaphragmatic hernia. *p <0.001 compared with HMD. tP <0.01 compared with HMD. Sp <0.001 compared with PPHN. w <0.005 compared with HMD.

concentrations in serum were parallel to the s t a n d a r d curve. T h e antibody used for this assay cross-reacts with the precursor of endothelin-1, big endothelin-1 (35%), and a n o t h e r of the three endothelins coded for in m a m m a l i a n genomes, endothelin-3 (7%). T h e most significant cross-reactivity is with big endothelin-1, which has physiologic effects completely d e p e n d e n t on its conversion by an endothelinconverting e n z y m e to endothelin-1. ~8 D a t a analysis c o m p a r i n g cord blood controls, H M D controls, and P P H N with and without E C M O was done with a one-way analysis of variance with post hoc testing using t tests with a Bonferroni correction. Significance was determined a t the p <0.05 level. Response of irET-1 levels over time within the P P H N groups with and without E C M O was compared by use of A N O V A with a repeated measures design. Oxygenation index was calculated with the following formula: OI = MAP X FIO2X 100/Pa02 where M A P is the m e a n airway pressure, FI02 is the fraction of inspired oxygen concentration, a n d Pa02 is the arterial oxygen tension. T h e P(A-a)O2 was calculated as follows: P(A-a)O2 = (578 X FIO2) -Pac02 - Pa02 where P a t 0 2 is the arterial carbon dioxide tension. T h e O I a n d P(A-a)O2 were correlated with irET-1 by use of linear regression analysis. RESULTS Clinical patient data are presented in T a b l e I. The H M D group h a d lower gestational age and weight, as would be expected. T h e P P H N group t h a t required E C M O was sig-

Table

II. Plasma i m m u n o r e a c t i v e endothelin-I levels Cord blood (n = t0)

HMD (n = 8)

PPHN, no ECMO (n = t2)

PPHN, with ECMO (n = t2)

irET-1 15.1 • 4.1 11.8 + 1.2 21.2 + 2.0 31.0 _+ 4.7",t,:~ (pg/ml) Values are expressed as mean + SEM. *p <0.003 compared with cord blood control. ~'p <0.002 compared with HMD. ~:p <0.05 compared with PPHN, no ECMO.

nificantly more ill t h a n either of the other two groups as gauged by OI, P(A-a)o2, a n d the need for high-frequency jet ventilation. Pressors to m a i n t a i n the systemic arterial blood pressure were required in all 24 infants with P P H N and in 5 (63%) of 8 infants with H M D . All patients with P P H N who required E C M O were treated with H F J V ; 4 (33%) of 12 infants with P P H N who did not require E C M O needed H F J V . N o patient with H M D qualified for H F J V . Because H F J V was not part of a research protocol, use of this ventilator strategy was at the discretion of the attending physician in infants in whom " m a x i m a l " conventional therapy failed. Endothelin levels in infants with P P H N who required E C M O were significantly higher t h a n levels in the other three patient groups (Table II). T h e m e a n ( + S E M ) irET- 1 concentration for all the infants with P P H N was 27.6 _+ 3.6 p g / m l (p <0.01 vs cord blood and H M D infants). The irET-1 levels correlated with disease severity as assessed by P(A-a)O2 and OI. T h e relationship between P(A-a)O2 as assessed by linear regression was significant at the p <0.02 level (r = 0.44), and OI versus i r E T - I was significant at the

1 12

Rosenberg et al.

A.

40

The Journal of Pediatrics July 1993

30

irET-1

20

(pg/ml) 10

ECMOTreatment I

0

I

i

I

i

I

B| 2O

irET-1 (pg/ml)

,.

I0

i

0

l

I

2

4

Time

i

l

6

i

8

I

10

(days)

Figure. A, Immunoreactive endothelin-1 (irET-1) levels in ECMO-treated infants. Day 1 sample is before ECMO; day 3 sample is after 24 hours of ECMO; day 10 sample is after decannulation from ECMO. B, Levels of irET-1 in non-ECMO-treated patients with persistent pulmonary hypertension (*p <0.001 compared with day 1 value). p <0.01 level (r = 0.49). The Figure shows the changes in plasma levels of irET-1 with time in infants with PPHN treated with and those treated without ECMO. In nonECMO-treated patients, values significantly decreased by day 3 of life. In patients requiring ECMO, irET-I levels decreased after ECMO was started but remained elevated as the course proceeded. The average age at initiation of ECMO was 33.8 hours of age; thus the 24-hour ECMO sample was obtained on the third day of life in most cases. Average time of decannulation was 8 days of age. DISCUSSION Although PPHN has been welt recognized as a clinical syndrome for more than 20 years, l, 2 the mechanisms contributing to its pathogenesis and pathophysiology remain

poorly understood. PPHN is associated with a wide spectrum of clinical disorders, but common features include increased pulmonary vasoreactivity and abnormal vascular remodeling.3, 4, 19-22Structural changes can include smooth muscle cell proliferation and increased adventitia, decreased arterial number, vascular occlusion caused by thromboemboli or endothelial cell swelling, and perivascular edema. 3' 19-22Despite the presence of structural changes, infants with PPHN also have marked vasolability, especially early in the course of the disease, as reflected by swings in arterial oxygenation.23 Links between altered vasoreactivity and structure of the neonatal pulmonary circulation in PPHN are unclear, but recent discoveries in vascular biology have demonstrated the crucial role of endothelium-derived products in the regulation of vascular tone and growth. 513 Endothelin-1, an endothelium-derived peptide with potent vasoconstrictor properties, 9 can also stimulate smooth muscle cell proliferation1~ and augment vasoconstriction to other agonists even when present in subthreshold concentrations.11 These physiologic effects of endothelin-1 mimic the central characteristics of the pulmonary circulation in PPHN. The association of high circulating irET-1 levels in children with severe PPHN suggests that increased endothelin-1 activity may contribute to its pathophysiologic changes. It remains possible, however, that high irET-1 levels are merely markers reflecting disease activity. Testing the hypothesis that endothelin-1 contributes to the pathophysiology of PPHN will require further studies examining lung endothelin-1 in babies dying with PPHN, or the physiologic effects of selective endothelin antagonists currently under development. Circulating irET-1 levels in children with PPHN are much higher than levels that we measured previously in normal children (2.9 + 0.2 pg/ml) or in older children with pulmonary hypertension (12.3 + 3.4 pg/ml). 24 The relatively high cord blood values measured in normal neonates are consistent with those from another laboratory that used the same assay. 25 Although endothelin-I is present in lategestation fetal rat lungs,26 causes umbilical cord constriction,27 and is vasoactive in the fetus, 14, 15its physiologic role in the fetus is unknown. It has been speculated that endothelin-1 may contribute to high pulmonary vascular resistance in the normal f e t u s . 14' 15 The rate of decline of blood irET-1 levels from cord blood values through the first days after birth in the normal newborn infant is not known, but levels are similar to adult levels by 7 months of age. 24 Although the postnatal decline in irET-1 levels appears delayed in sick infants with PPHN, mechanisms of increased circulating irET-1 levels in PPHN are uncertain. Factors that may contribute to enhanced endothelin-1 production include increased transmural pressure, severe by-

The Journal o f Pediatrics Volume 123, Number 1

poxia, low flow or shear stress, and decreased E D R F release. 13,2s,29 Increased lung production has been suggested in adults with primary pulmonary hypertension 17 and in some children with pulmonary hypertension and congenital heart disease. 3~ In addition to increased production, decreased clearance may lead to high circulating irET-I levets. The normal pulmonary circulation clears up to 67% of circulating endothelin- 1 in a single passage31; thus the low pulmonary blood flow that characterizes P P H N and E C M O therapy could result in diminished endothelin-1 clearance. We found that irET-1 levels progressively decreased with clinical improvement in mild P P H N that did not require E C M O treatment. In contrast, infants with more severe P P H N continued to have high irET-1 levels during E C M O therapy. There was no correlation between p r e - E C M O irET-1 level and duration of E C M O course, nor was there a correlation between day 5 irET- 1 level and time to decannulation. Thus these persistently elevated levels during E C M O do not appear to be related to the initial disease process that necessitated E C M O therapy. Persistent elevation of irET- 1 levels may be caused by low pulmonary blood flow during E C M O , resulting in either low clearance or activation of endothelin- 1 production. Increased irET- 1 levels have been reported in older children with congenital heart disease and pulmonary hypertension after cardiac bypass) 2 Studies have identified elevated concentrations of lipid mediators in P P H N , including leukotrienes in tracheal effluent, 33 serum thromboxane and prostacyclin, 34 and circulating platelet-activating factor. 35 It is possible that multiple vasoactive mediators are released with P P H N or with lung injury during its treatment. The relative effects of each vasoactive product on pulmonary vascular tone and the basic mechanisms underlying the failure of postnatal adaptation in P P H N are not known. Although increased production of potent vasoconstrictors, such as endothelin-1, may contribute to P P H N , the inability to sustain vasodilator release or decreased smooth muscle responsiveness to dilator stimuli may also play an important role. For example, release of E D R F at birth lowers pulmonary vascular resistance in the ovine fetus, supporting the speculation that P P H N may be associated with decreased E D R F release at birth. 36 Decreased E D R F activity has been reported from in vitro studies of pulmonary arteries from adults with pulmonary hypertension, 37 but it is not known whether this is also true in P P H N . This concept is supported by recent studies demonstrating clinical efficacy of inhaled nitric oxide (identified as E D R F ) in infants with P P H N , suggesting that the pulmonary circulation in P P H N remains responsive to nitric oxide and that insufficient endogenous E D R F production contributes to P P H N . 38' 39 We conclude that circulating irET-1 levels are higher in

R o s e n b e r g et al.

113

newborn infants with P P H N than in other sick neonates, and that levels are strongly correlated with severity of illness. Whether high irET-1 level directly contributes to the pathophysiology of P P H N or is simply a marker of disease severity requires further study. We express our appreciation to John T. Reeves, MD, for his guidance and support, and to Susan Moreland, RN, and Audrey Howard, RN, for their enormous help in data collection. REFERENCES

1. Levin DA, Heymann MA, Kitterman JA, Gregory GA, Phibbs RH, Rudolph AM. Persistent pulmonary hypertension of the newborn. J PEOIATR 1976;89:626-30. 2. Siassi B, Goldberg S J, Emmanouilides GC, Higashino SM, Lewis E. Persistent pulmonary vascular obstruction in newborn infants. J PEDIATR 1971;78:610-5. 3. Geggel RL, Reid LM. Structural basis of persistent pulmonary hypertension of the newborn. Clin Perinatol 1984;11:52549. 4. Stenmark KR, Abman SH, Accurso FJ. Etiologic mechanisms in persistent pulmonary hypertension of the newborn. In: Weir EK, Reeves JT, edsl Pulmonary vascular physiology and pathophysiology. New York: Marcel Dekker, 1988:355-402. 5. Brenner BM, Troy JL, Ballermann BJ. Endothelium-dependent vascular responses. J Clin Invest 1989;84:1373-8. 6. Vane JR, Anggard EE, Botting RM. Regulatory functions of the vascular endothelium. N Engl J Med 1990;323:27-37. 7. Hickey KA, Rubyani GM, Paul R J, Highsmith RF. Characterization of a coronary vasoconstriction produced by cultured endothelial cells. Am J Physiol 1985;248:C550-6. 8. Gillespie MN, Owasoyo JO, McMurtry IF, O'Brien RF. Sustained coronary vasoconstriction provoked by a peptidergic substance released from endothelial cells in culture. J Pharmacol Exp Ther 1986;239:339-43. 9. Yanagisawa M, Kurihara H, Kimura S, et al. A novel potent vasoconstrictor peptide produced by vascular endothelial ceils. Nature 1988;332:411-5. 10. Komura I, Kurihara H, Sugiyama T, Yoshizumi M, Takaku F, Yazaki Y. Endothelin stimulates c-fos and c-myc expression and proliferation of vascular smooth muscle cells. FEBS Lett 1988;238:249-52. 11. Yang Z, Richard V, Von Segesser L, et al. Threshold concentrations of endothelin-1 potentiate contractions to norepinephrine and serotonin in human arteries. Circulation 1990;82:18895. 12. Doherty AM. Endothelin: a new challenge. J Med Chem 1992;35:1493-508. 13. Rubanyi GM. Endothelium-derived vasoconstrictor factors: an overview. In: Ryan US, Rubanyi GM, eds. Endothelial regulation of vascular tone. New York: Marcel Dekker, 1992:37586. 14. Cassin S, Kristova V, Davis T, Kadowitz P, Gause G. Tone dependent responses to endothelin in'the isolated perfused fetal sheep pulmonary circulation in situ. J Appl Physiol 1991;70:1228-34. 15. Chatfield BA, McMurtry IF, Huel SL, Abman SH. Hemodynamic effects of endothelin-I in ovine fetal pulmonary circulation. Am J Physiol 1991;261:R182-7. 16. Isozaki-Fukuda Y, Kojima T, Hirata Y, et al. Plasma immu-

1 14

17.

18.

19.

20. 21.

22.

23.

24.

25.

26.

27.

Rosenberg et al.

noreactive endothelin-1 concentration in human fetal blood: its relation to asphyxia. Pediatr Res 1991;30:244-7. Stewart D J, Levy RD, Cernacek P, Langleben D. Increased plasma endothelin-1 in pulmonary hypertension: marker or mediator of disease? Ann Intern Med 1991;114:464-9. Sakurai T, Yanagasawa M, Masaki T. Molecular characterization of endothelin receptors. Trends Pharmacol Sci 1992; 13:103-8. Murphy JD, Rabinovich M, Goldstein J, Reid LM. The structural basis of persistent pulmonary hypertension of the newborn infant. J PEDIATR 1981;98:962-7. Murphy JD, Vawter GF, Reid LM. Pulmonary vascular disease in fatal meconium aspiration. J PEDIATR 1984;104:758-62. Arnold J, O'Brodovich H, Whyte R, Coates G. Pulmonary thromboemboli after neonatal asphyxia. J PEDIATk 1985; 106:806-9. Haworth SG. Pulmonary vascular remodeling in neonatal pulmonary hypertension. State of the art. Chest 1988;93:133S8S. Drummond WH, Gregory GA, Heymann MA, Phibbs RA. The independent effects of hyperventilation, tolazoline, and dopamine on infants with PPHN. J PEDIATR 1981;98:603-11. Allen SW, Chatfield BA, Koppenhafer SL, et al. Circulating immunoreactive ET-1 and hypoxic pulmonary vasoconstriction in childhood pulmonary hypertension [Abstract]. Pediatr Res 1992;31:16A. Nakamura T, Kasai K, Konuma S, et al. lmmunoreactive endothelin concentration in maternal and fetal blood. Life Sci 1990;46:1045-50. MacCumber MW, Ross CA, Glaser BM, Snyder SH. Endothelin visualization of mRNAs by in situ hybridization provides evidence for local action. Proc Natl Acad Sci USA 1989;86:7285-9. Haegerstand A, Hemsen A, Gillis C, Larsson O, Lundberg JM. Endothelin presence in human umbilical vessels, high levels in fetal blood and potent vasoconstrictor effect. Acta Physiol Scand 1989;137:541-2.

The Journal of Pediatrics July 1993

28. Kourembanas S, Marsden PA, McQuillan LP, Fuller DV. Hypoxia induces endothelin gene expressions and secretion in cultured human endothelium. J Clin Invest 1991;88:1054-7. 29. Boulanger C, Luscher TF. Release of endothelin from the porcine aorta. J Clin Invest 1990;85:587-90. 30. Yoshibayashi M, Nishioka K, Nakao K, et al. Plasma endothelin concentrations in patients with pulmonary hypertension associated with congenital heart defects. Circulation 1991; 84:2280-5. 31. de Nucci G, Thomas R, D'Orleans-Justa P, et al. Pressor effects of circulatory endothe!in are limited by its removal in the pulmonary circulation and by the release of prostacyclin and endothelium derived relaxing factor. Proc Natl Acad Sci USA 1988;85:9797-9800. 32. Komai H, Adatia IT, Elliott M J, et al. Increased endothelin plasma levels following cardiopulmonary bypass in congenital heart disease. Am Rev Respir Dis 1992;145:A641. 33. Stenmark KR, James SL, Voelkel NF, Toews WH, Reeves JT, Murphy RC. Leukotriene C4 and D4 in neonates with hypoxemia and pulmonary hypertension. N Engl J Med 1983;309:7780. 34. Hammeman C, Lass N, Strates E, Komar K, Bui K-C. Prostanoids in neonates with persistent pulmonary hypertension. J PEBIATR 1987;110:470-2. 35. Caplan MS, Hsuch W, Sun X-M, Gidding SS, Hageman JR. Circulating plasma platelet activating factor in persistent pulmonary hypertension of the newborn. Am Rev Respir Dis 1990;142:1258-62. 36. Abman SH, Chatfield BA, Hall SL, McMurtry RF. Role of EDRF during transition of pulmonary circulation at birth. Am J Physiol 1990;259:H1921-7. 37. Cremona J, Dinh Xuan AT, Higenbottam TW. EDRF and the pulmonary circulation. Lung 1991;169:185-202. 38. Roberts JO, Polaner DM, Lang P, Zapol WM. Inhaled nitric oxide in PPHN. Lancet 1992;340:818-9. 39. Kinsella JP, Neish SR, Shaffer E, Abman SH. Low dose inhalational nitric oxide in PPHN. Lancet 1992;340:819-20.