Persistent pulmonary hypertension of the newborn

Abstract

Persistent pulmonary hypertension of the newborn (PPHN) is a severe pulmonary disorder which occurs one in every 500 live births. About 10–50% of the victims will die of the problem and 7–20% of the survivors develop long term impairments such as hearing deficit, chronic lung disease, and intracranial bleed. Most of the adult survivors show evidence of augmented pulmonary vasoreactivity suggesting a phenotypical change. Several animal models have been used to study the pathophysiology and help to develop new therapeutic modality for PPHN. The etiology of PPHN can be classified into three groups: [A] abnormally constricted pulmonary vasculature due to parenchymal diseases; [B] hypoplastic pulmonary vasculature; [C] normal parenchyma with remodeled pulmonary vasculature. Impaired vasorelaxation of pulmonary artery and reduced blood vessel density in lungs are two characteristic findings in PPHN. Medical treatment includes sedation, oxygen, mechanical ventilation, vasorelaxants (inhaled nitric oxide, inhaled or intravenous prostacyclin, intravenous prostaglandin E1, magnesium sulfate), and inotropic agents. Phosphodiesterase inhibitor has recently been studied as another therapeutic agent for PPHN. Endothelin-1 (ET-1) inhibitor has been studied in animal and a case of premature infant with PPHN successfully treated with ET-I inhibitor has been reported in the literature. Surfactant has been reported as an adjunct treatment for PPHN as a complication of meconium aspiration syndrome. Even with the introduction of several new therapeutic modalities there has no significant change in survival rate. Extracorporeal membrane oxygenator is used when medical treatment fails and patient is considered to have a recoverable cause of PPHN.

Introduction

Persistent pulmonary hypertension of the newborn (PPHN) is a frequent cause of hypoxemic respiratory failure in term and late preterm infants affecting 0.43 to 6.8 per 1000 live births¹. Severe hypoxemia usually develops right after birth but occasionally can be a consequence of other diseases, such as severe respiratory distress syndrome, or secondary to the management of other perinatal disorder, such as body cooling for hypoxic-ischemic encephalopathy and slowly develops several hours or even days of birth. With the introduction of inhaled nitric oxide (INO) the management of PPHN becomes easier nowadays than a decade ago but the mortality rate remains unchanged and a high percentage of the PPHN survivors carry long term sequels including chronic oxygen dependence, stroke, or impaired hearing. In this review we describe the physiology of the perinatal transition, some commonly used animal models that help us to understand the pathophysiology of PPHN, etiology and risk factors for PPHN, and recent strategies in PPHN management.

Animal models of PPHN

Several animal models have been used to elucidate the pathophysiology of PPHN including intrauterine ductus arteriosus ligation of fetal lamb or piglet, acute or chronic hypoxic pulmonary hypertension of newborn animals^2,3, intra-tracheal meconium instillation⁴, creation of aorto-pulmonary shunt⁵, and intrauterine non-steroid anti-inflammatory drug (NSAID) or selective serotonin re-uptake inhibitor (SSRI) exposure⁶. Each model has its pros and cons and readers should refer to each published model for details. Most animal models show smooth muscle layer thickening of the pulmonary arteries and thickened right ventricular wall. Other described findings include decreased blood vessel density of the lungs, impaired endothelial nitric oxide synthase (eNOS) activity, increased ET-1 formation, increased reactive oxygen species (ROS) formation⁷, and impaired alveolar formation. The increased ROS formation can stimulate the proliferation of smooth muscle cells⁸ and increases activity of phosphodiesterase (PDE) that enhance the hydrolysis of cGMP.

Physiology of perinatal transition

Many structural and functional changes occur during fetal lung development to prepare the lung for the transition to air breathing. The development of pulmonary vasculature is genetically controlled and pulmonary vessels acquire increased vasoreactivity with advancing gestation. An extensive review was recently published and readers who are interested in this topic are encouraged to read this article⁹. Before birth, the lungs are filled with fluid and the pulmonary artery resistance is very high, due to low oxygen tension in the alveoli, with most of the blood returns to the right atrium goes through foramen ovale into left atrium. Blood goes into right ventricle also shunts through ductus arteriosus into the aorta leaving only small amount of blood flows into the lungs. The first breath after birth allows air enters into alveoli with a dramatic reduction in pulmonary artery resistance secondary to increased oxygen tension. The sudden increase of oxygen tension from 20 torr to 150 torr increases mitochondrial oxidative phosphorylation and ATP production. The surge in blood ATP levels during the postnatal transition can stimulate eNOS function with production of NO and pulmonary vasorelaxation¹⁰.

Several vasoactive substances are known to modulate the vasomotor tone of pulmonary artery. Endothelin-1 (ET-1), nitric oxide (NO), and prostacyclin (PGI2) are most extensively studied in PPHN. Thromboxane A₂ is another product of the cyclo-oxygenase and thromboxane synthase that plays some role mainly during infections associated PPHN. Voltage gated potassium (Kv) channel which modulate vascular smooth muscle contraction also play important role in PPHN¹¹. ET-1 is a vasoconstrictor to the pulmonary arteries and enhances O₂⁻ formation that depletes NO bioavailability and promotes the growth of pulmonary artery muscular layer. When pulmonary artery resistance failed to decrease during the perinatal transition the deoxygenated blood will shunt from right to left through either foramen ovale (no differential cyanosis) or ductus arteriosus (with differential cyanosis) with PPHN. Most venous return from inferior vena cava goes through foramen ovale whereas most blood returns from the superior vena cava tends to go into main pulmonary trunk (Figure 1).

Figure 1.

Blood flow pattern of PPHN. Two potential right-to-left shunts, either through foramen ovale or ductus arteriosus, can be observed in PPHN. There will be a differential cyanosis if the right-to-left shunt is through the ductus. Venous blood from superior vena cava preferentially flow through right ventricle into the main pulmonary artery so intravenous vasorelaxant should be given via an intravenous catheter placed in the upper part of the body. Most venous blood from inferior vena cava flow directly through foramen ovale when PPHN is present so inotropic agent should be given through catheter placed in the lower part of the boby.

Endothelial NOS, or nitric oxide synthase type 3 (NOS3) is the most extensively studied enzyme in PPHN. Un-stimulated eNOS is believed to bind to caveolin-1¹². When activated by shear stress or ATP, eNOS departs caveolin and become phosphorylated to form dimer. The eNOS dimer then associated with heat-shock-protein-90 (hsp90) in the presence of BH4, calcium, FAD, FMN, iron and L-arginine to be “coupled” to convert L-arginine into NO and L-citrullin (Figure 2). BH4 depletion and hypocalcemia can both uncouple eNOS and lead to O₂⁻ formation. Hypoglycemia leads to low FAD or FMN levels in endothelial cells may also uncouple eNOS. L-arginine is not only the substrate for eNOS, its presence can also help eNOS coupling. If arginase-II activity is increased, L-arginine will be converted into asymmetric di-methyl-arginine (ADMA) that will inhibit eNOS function.

Figure 2.

Controls of the eNOS function. Shear stress in blood vessel activates eNOS expression and activity. When activated, eNOS is free from the association with caveolin-1 and form dimer and becomes phosphorylated. eNOS then binds to heat-shock-protein-90 (hsp90), calmodulin, and in the presence of BH4, calcium, FAD, FMN to be activated. Once activated, eNOS converts L-arginine into citrullin and NO.

Etiology of PPHN

The etiology of PPHN can be classified into three main categories (Table 1). The most common one is the PPHN secondary to parenchymal diseases including meconium aspiration syndrome (MAS), severe respiratory distress syndrome) RDS and pneumonia. This is mainly due to poor oxygen entry into the alveolar space, especially in MAS with obstruction in the airways. Inadequate blood vessel density with decreased total cross section of pulmonary vasculature and increased pulmonary vascular resistance is the cause of PPHN in congenital diaphragmatic hernia. The least common etiology is normal parenchyma with remodeled pulmonary vasculature such as idiopathic PPHN, congenital heart disease, and chronic intrauterine hypoxia. Some congenital heart diseases are associated with obstructed pulmonary venous return which can lead to secondary increased pulmonary artery resistance. Hypoxic-ischemic encephalopathy due to chronic intrauterine hypoxia may remodel the pulmonary vasculature with either eNOS uncoupling or increased ET-1 production that increases pulmonary vascular resistance. Idiopathic PPHN is the rarest cause of PPHN usually with normal chest x-ray findings. There are some metabolic, or genetic, disorders that can present with PPHN. Pearson et al reported heterozygote T1405N genotype for carbamoyl-phosphotate synthatase, an enzyme that determine the blood levels of arginine and citrulline, is associated with PPHN possible due to lack of substrate for endothelial nitric oxide synthase (eNOS)¹³. Epidemiologic study demonstrated black and Asian maternal race is associated with significant higher risk for PPHN. Male gender also leads to higher incidence of PPHN¹⁴. Although it is hard to decipher the observations the evidence suggests some genetic predisposition to PPHN and deserve further exploration.

Table 1.

Classification of PPHN

■ Abnormally constricted pulmonary vasculature due to parenchymal diseases ■ meconium aspiration syndrome ■ respiratory distress syndrome [31] ■ pneumonia ■ Hypoplastic pulmonary vasculature ■ congenital diaphragmatic hernia ■ lung hypoplasia ■ Normal parenchyma with remodeled pulmonary vasculature ■ idiopathic PPHN ■ congenital heart disease ■ Hypoxic-ischemic encephalopathy, chronic ■ Others

Pathology of PPHN

Decreased vascular density and thickened smooth muscle layer of the pulmonary artery are two most common pathological findings in PPHN. Morphometric analysis of the PPHN lungs reveals extension of muscle into small pulmonary arteries, all alveolar duct and wall arteries (<30 µm external diameter), normally non-muscular, are fully musularized. The medial wall thickness of the normally muscular intra-acinar arteries is doubled; arterial size and number, however, are normal in all¹⁵. Decreased number of alveoli is seen in lung hypoplasia and congenital diaphragmatic hernia.

Symptoms and signs

In developing countries post-term infant and intrauterine growth restriction (IUGR) are the major groups to have PPHN so dry and peeled skin is commonly seen. Meconium passage or lack of subcutaneous fat tissue is commonly observed in IUGR and post-term neonates. General cyanosis is the typical presentation of PPHN but sometime we can observe the so-called differential cyanosis with pre-ductal skin less cyanotic than the post-ductal skin unless transposition of great arteries (TGA) is also present. When PPHN is associated with TGA then reversed differential cyanosis can be seen. The routine hyperoxia challenge usually cannot help the diagnosis since severe PPHN can behave like typical cyanotic congenital heart disease and does not respond to high oxygen challenge.

Diagnosis and evaluation

Echocardiography is the most convenient and reliable method to establish the diagnosis. Poor myocardial contractility, poor movement of inter-ventricular septum, deviation of inter-atrial septum to the left, turbulent flow for tricuspid regurgitation, or shunt through the ductus arteriosus can be used to evaluate the cause and severity of the PPHN. Pulmonary arterial accelerating time, and maximal velocity of the tricuspid regurgitation can be used to estimate the pulmonary artery pressure. Major complex congenital heat disease has to be ruled out by echocardiography prior to the extracorporeal membrane oxygenator (ECMO) treatment. Infusion of normal saline through a venous line placed below the diaphragm which creates a microbubble picture can be used to demonstrate the inter-atrial right-to-left shunt, either through patent foramen of ovales or patent ductus arteriosus, as a supportive evidence for PPHN. Very rarely the PPHN can be the consequence of obstructive type total anomalous pulmonary venous return, that does not respond to usual PPHN treatment, can only be ruled-out by echocardiography. Echocardiography is also a mandatory study before initiation of ECMO because it is necessary to rule-out possible lethal cyanotic congenital heart disease that can not be corrected by heart surgery.

Risk factors

Some risk factors for PPHN have been reported in the literature. However, the true mechanism remains obscure for most of them.

Intrauterine growth restriction

Intrauterine growth restriction has been reported to associate with increased risk of PPHN¹⁶. It is believed that uteroplacental insufficiency may lead to postnatal pulmonary hypertension by two mechanisms: oligohydramnios and chronic fetal hypoxia. Hypoxia increases endothelial synthesis of vasoconstrictors and smooth muscle mitogens such as endothelin-1, platelet-derived growth factor-β and vascular endothelial growth factor; it also inhibits endothelial nitric oxide synthase (eNOS). Increased thickness of right ventricular wall, a hallmark of pulmonary hypertension, is also commonly seen in animal models of intrauterine growth restriction.

Maternal SSRI exposure

SSRI, a commonly prescribed anti-depressant, has been reported to associate with PPHN, especially during the late trimester, in a case-control study¹⁷. Some later studies also demonstrate the increased risk of PPHN in SSRI exposed neonates. Similar finding is seen in a rat study and is believed to be due to smooth muscle cell proliferation in pulmonary arteries¹⁸.

In utero NSAID exposure

In utero NSAID exposure is considered a risk factor for PPHN due to the fact that most NSAIDs inhibit prostaglandin synthesis and in utero exposure is believed to close ductus arteriosus prenatally. Alano et al studied the meconium and found out an association between the existence of NSAID metabolites and the development of PPHN¹⁹.

Genetic risk factors

The only genetic risk factor reported is the association between heterozygote T1405N genotype of carbamoyl-phosphotate synthatase and PPHN possible due to lack of substrate for endothelial nitric oxide synthase (eNOS)¹³. Epidemiologic studies suggest black maternal race and male newborns have higher chance to develop PPHN.

Management (Table 2)

Table 2.

PPHN Management

■ General ■ Reduce stimulation: noise control, dim ambient light, thermoneutral control ■ Sedation and/or muscle relaxation ■ Empirical antibiotics ■ Avoid hypoglycemia ■ Avoid hypocalcemia ■ Nutritional support ■ Alkalosis: questionable benefit ■ Inotropic agent ■ Mechanical ventilation ■ Conventional mechanical ventilation: pressure limited or volume-controlled ventilator ■ High frequency ventilator: HFOV or HFJV ■ Vasorelaxant ■ Inhaled NO ■ Prostaglandin: PGE1 or PGI2 ■ Others: sildenafil, MgSO4, milrinone ■Extracorporeal membrane oxygenator: VA- and VV-type

Current therapy for PPHN includes mechanical ventilation, muscle paralysis/relaxation, sedation, alkalosis and vasorelaxants²⁰. Inhaled nitric oxide (INO) is currently regarded as the gold standard therapy, but as many as 30% of cases are non-responder to INO treatment¹.

General management

Quiet environment with minimal stimulation is recommended for PPHN management. It is known that bright light or loud noise can affect the oxygenation. Body temperature should be maintained at thermoneutral range (37.0±0.5°C). Appropriate hydration and hematocrit (40–50%) should be maintained. Polycythemia (hematocrit > 55%) can increase the blood viscosity and increase the pulmonary vascular resistance. Hypoglycemia and hypocalcemia should be avoided. Hypoglycemia may lead to reduced ATP formation and ATP is a known agonist for eNOS. Calcium is one of the critical cofactor for eNOS activity and hypocalcemia may impair eNOS function and should be corrected. Preductal oxygenation should be used to adjust the ventilator support and SpO₂ above 95% may be appropriate. Some centers prefer to monitor the difference between pre- and post-ductal SpO₂ as an indicator for the severity of PPHN which we do not regularly use. For severe PPHN shunting through foramen ovale we can observe fluid infusion through intravenous catheter below the diaphragm goes through foramen ovale so it is reasonable to provide pressor(s) via such line in order to reduce the effect to the pulmonary arteries. Empirical antibiotics, ampicillin and gentamicin, are recommended before infection, especially group B streptococcus, can be ruled out as the cause of PPHN.

Severity of PPHN

Two parameters, alveolar-arterial oxygen difference (AaDO2) and oxygenation index (OI), are used most frequently in PPHN management to judge the severity and progress of the disease. AaDO2 is the difference between alveolar oxygen content and arterial oxygen content and the formula to calculate AaDO2 is

• AaDO2 = (ATM-P_H2O)×FiO₂-PaO₂-PaCO₂/RQ

ATM is the atmospheric pressure which is usually equal to 760 torr at sea level but needs to be adjusted in high altitude. P_H2O is the pressure of water vapor in one ATM which is usually considered to be 47 torr. FiO2 is the fraction of inspired air provided by the ventilator. The reason to correct for PaCO2 is due to the fact that some oxygen used to produce CO₂ is from the nutrient inside of the body. RQ is the respiratory quotient and equal to 1 is the energy source is purely sugar or equal to 0.8 when the nutritional source is a combination of glucose, protein, and lipid. The clinical decision to use ECMO is usually depends on AaDO₂. When AaDO₂ is above 600 twice with maximal support then ECMO can be considered. Oxygenation index is more commonly used during medical management of PPHN since it also takes into the consideration of ventilator support. Oxygenation index is calculated as

• OI = MAP×FiO₂×100/PaO₂

MAP is the mean airway pressure provided by the ventilator. OI below 10 is usually considered to be mild PHN, OI between 10 and 20 is considered to be moderate PPHN, whereas OI above 20 is considered to be severe PPHN. Under maximal medical support and the OI remains above 20 we usually recommend ECMO treatment if it is available.

Mechanical ventilation

Mild PPHN can be managed by nasal cannula whereas moderate and severe PPHN requires positive pressure ventilator. High oxygen concentration and low PaCO₂ are commonly used for PPHN under the belief that both can help to relax the pulmonary arteries. However, it is recommended that PaCO₂ levels should not be lower than 35 torr since also CO₂ controls cerebral perfusion. Aggressive hyperventilation with hypocapnia is known to be a significant risk factor for hearing impairment in PPHN survivors²¹. Both conventional mechanical ventilator and high frequency (oscillator or jet) ventilator can be used. In severe PPHN most centers prefer HFOV since this model usually will take away patients’ spontaneous breathing.

Muscle paralysis/relaxation

Agitation usually aggravates PPHN with initial transient increase of oxygenation follows by precipitous drop of oxygenation. To remove this flip-flop phenomenon in PPHN management some centers will paralyze the PPHN patients.

Sedation

Continuous sedation, either by bezodiazepine or narcotic agent, is a common practice in PPHN management. Sedation can decrease the frequency of desaturations.

Alkalosis

Induction of alkalosis by either infusion of sodium bicarbonate or hyperventilation has frequently been used as part of the PPHN treatment. However, there is no solid evidence to show that this practice is effective. A multicenter observational study demonstrated an increased ECMO requirement for patients received NaHCO₃ infusion before the INO era. It is hypothesized that NaHCO₃ infusion increases CO₂ formation and leads to increases in ventilator support²⁰.

Inotropic agents and vassopressor

Increased right-to-left shunt is believed to be the main reason for the severe hypoxemia in PPHN. To decrease the right-to-left shunt may be beneficial in PPHN. β-Adrenergic agonists can decrease pulmonary vascular resistance more than systemic vascular resistance might have a more favorable effect in PPHN especially with poor myocardial function such as in birth asphyxia associates with PPHN²². Dopamine increases both systemic and pulmonary vascular resistance and reduces left-to-right ductal shunting in preterm infants which suggests that it may not be a good choice for premature infants with patent ductus arteriosus and PPHN²³.

Vasorelaxants

The most effective vasorelaxant(s) for PPHN is the one(s) that work specifically to the pulmonary vasculature. But, unfortunately, there is so far no specific vasorelaxant to pulmonary arteries. Several vasorelaxants have been used for the past 4 decades including epinephrine (β-adrenergic receptor agonist), tolazoline (non-selective competitive α-adrenergic receptor antagonist or histamine release), magnesium sulfate, etc. Prostanoids (PGE1 or PGI2) have been suggested recently when the cause of severe hypoxemia remains uncertain before ductal-dependent cyanotic heart disease can be ruled out. Prostanoids help to relax the vascular smooth muscle cells, and maintain the patency of ductus arteriosus, though cAMP formation. But owing to their non-specificity, and potential side effect of apnea, a secure airway is needed and the possibility of low blood pressure should always be condidered. Prostacyclin (PGI₂) and its analogues have been used more commonly than PGE₂ recently. Epoprostanol is an intravenously administered analogue of PGI₂ whereas iloprost is inhaled analogue. Presently the experience of PGI₂ analogues for PPHN remains limited but inhaled iloprost is considered to be a more appropriate route to provide more selective effect.

Nitric oxide (NO) is considered to be the most specific pulmonary artery vasodilator due to its method of administration. By increasing the intracellular cGMP in the smooth muscle cells of the pulmonary arteries NO can decrease the pulmonary vascular resistance. Inhaled nitric oxide (INO) at a dose of ≥ 5 ppm significantly reduces the combined outcome of death and need for ECMO by 35% in infants with oxygen index of ≥ 25. However, long-term follow-up studies (12–24 months) indicate that INO does not alter either the incidence of chronic lung disease or neurodevelopmental impairment^{24, 25}. Early use of INO in the disease course does not reduce the use of ECMO, mortality, or improve other outcomes. INO above 20 ppm does not provide more benefit and should be avoided. Special arrangement is needed if INO is provided when patient is on high frequency jet ventilator. NO can oxidize hemoglobin into methemoglobin and levels of methemoglobin should be monitored regularly since some patient may not detoxify it due to enzyme deficiency. INO is not universally effective and about 30% of the severe PPHN does not respond to INO.

Original approved INO use is when OI is above 25 but recent trend is to start it when OI reaches 20. There is no consensus regarding the OI threshold to initiate INO and most centers establish their own criteria according to their own experience and the availability of the ECMO facility. When ECMO locates far away from the care center then some centers will initiate INO when OI reaches 15 twice several (2–6) hours apart. The initial dose is usually 10 ppm and escalates, or decreases, according to the clinical response. Lower concentration of INO (5–10ppm) has been studied and result remains unclear.

Extracorporeal membrane oxygenator (ECMO)

When all the medical treatment fails then ECMO should be considered when the PPHN is caused by a reversible cause. Chromosomal anomaly, lethal congenital malformation, uncorrectable heart defect, and major intracranial bleed prohibit the use of ECMO. The course of ECMO is usually set between 10 to 14 days according to the individual institutional guidelines. If patient does not respond to ECMO during this period then (s)he will be de-cannulated and put back on the original management. Two types of ECMO can be used including veno-venous (VV) and veno-arterial (VA). VA type ECMO needs cannulatation of one vein and one artery, usually one external jugular vein and one internal carotid artery will be used. The problem for VA-ECMO is that an internal carotid artery will be sacrificed and the increased chance of intracranial bleeding. VV-ECMO can be performed using double-lumen catheter without sacrificing any artery. However, a bigger size catheter is required for VV-ECMO and a good pumping heart is mandatory for the VV-ECMO. Patient’s size should be considered for ECMO use. The patient needs to be at least two kilograms in weight and usually more than 34 weeks’ gestation in order to be cannulated. During ECMO treatment the patient has to be maintained by low ventilator setting to keep the alveoli open. Coagulation profiles should be checked several times a day to avoid massive bleeding. The chance of intracranial bleed is about 10–15% for ECMO which should be mentioned to the family before the procedure. Well-trained perfusionsits are required and regular wet-run should be performed routinely to maintain the skill by the ECMO centers. Since the introduction of INO the number of ECMO centers has declined dramatically due to the cost of maintenance.

Others

L-Arginine is the substrate for eNOS and has been shown to improve eNOS function in vitro but fails to show benefit in vivo. However, L-citrulline, a precursor of L-arginine, can go into endothelial cells effectively and has been shown to ameliorate the severity of hypoxia induced pulmonary hypertension in animal model²⁶. It is believed the differential effect is due to fact the L-arginine can not enter endothelial cell to recouple eNOS function. Endothelin receptor antagonists have been studied to be effective vasorelaxants but their clinical usefulness remains unknown. Bosentan is the only commercially available endothelin receptor antagonist and a few case reports are available in the literature. Magnesium inhibits calcium entry into smooth muscle cells and, as a result, is an effective vasorelaxants²⁷. Tolazoline is an active vasorelaxant especially administered through inhalation. There are also case series, or reports, about the effect of magnesium and tolazoline use in PPHN but large scale randomized control trial is lacking for both. Adenosine is a vasorelaxant with an extremely short half-life and is an eNOS agonist that increases the formation of NO. Due to its short half-life we can expect less systemic side effect to use adenosine. Presently there is limited experience with adenosine in PPHN²⁸.

Recently there is an interest in phosphodiesterase (PDE) inhibitor especially the type 5 inhibitors²⁹. There are at least more than 11 types of PDE that hydrolyze cAMP, cGMP, or both. Type 5 is the main PDE in the pulmonary vasculature, genital organ, and auditory system. Inhibiting PDE type 5 can maintain intracellular cGMP levels and prolong the effect of NO in pulmonary arteries. There have been several cases series and one randomized control trial of using sildenafil (0.5–2.0 mg/kg every 6 hours), PDE-5 inhibitor, in managing PPHN with success³⁰. Milrinone is a type 3 PDE inhibitor that decreases hydrolysis of cAMP. Dosage between 0.2 to 1 µg/kg/min with or without loading (20–50 µg/kg) of milrinone has been reported when patient fail to respond INO. Since PGI₂ relax pulmonary artery via increasing intravascular cAMP levels so milrinone may be helpful in treating PPHN. There are a few case series demonstrated efficacy of milrinone use in PPHN. However, since milrinone also decreases the systemic vasculature resistance may aggravate right-to-left shunt so its use is recommended only when poor systemic perfusion is considered to affect the PPHN treatment. ET-1 is another important pulmonary vasoconstrictor that contributes to PPHN and ET-1 dual antagonist, Bosentan, has been reported in a few severe PPHN neonates. A recent single center, randomized, blinded, controlled trial of Bosentan 1 mg/kg twice daily via feeding tube was reported against placebo without inhaled NO treatment showed a dramatic decrease in mortality and better neurological outcome³¹.

Surfactant has been studied in animal model of meconium aspiration induced PPHN or combine with NO donor, or sildenafil, in other type of PPHN with some success. Clinical trial of surfactant in PPHN has been reported in human neonates with MAS under the belief that meconium can inhibit surfactant protein function or used as an airway toilet to wash out the meconium in the airway³². Recent Australian multicenter study reconfirmed the efficacy of surfactant in MAS especially in places where ECMO is not available³³. Superoxide dismutase (SOD), or its mimetics, has been studied by both Steinhorn and Black groups. Both groups show some promise of using SOD in animal model of PPHN³⁴. The mechanism may be secondary to the removal of O₂⁻ that leads to an increased bioavailability of NO and increased apoptosis of pulmonary artery smooth muscle cells^35–37.

Tetrahydrobiopterin (BH4) is not only a vital cofactor for eNOS function but also an important intracellular antioxidant. Depletion of BH4, or oxidation of BH4 into dihydrobiopterin (BH2), can uncouple eNOS function and shift eNOS from NO formation into O₂⁻ formation. NO not only relax arteries but also mediate blood vessel formation (angiogenesis). Method to increase endothelial cell BH4 content might improve eNOS function and help angiogenesis. We recently demonstrated that sepiapterin, a precursor for BH4, can recouple eNOS and correct the impaired angiogenesis in PPHN pulmonary artery endothelial cells³⁸. Some other antioxidants such as N-acetylcyeteine, apocynin, and ascorbate can either decrease intracellular ROS formation or increases BH4 levels that may help to improve NO bioavailability but their efficacy has not been demonstrated in human infants.

Outcome

Survivors have high morbidity in the forms of neurodevelopmental and audiological impairment, cognitive delays, hearing loss, and a high rate of rehospitalization. Low PaCO₂ level, usually due to hyperventilation, is considered to be a contributing factor for hearing loss especially when combined with the prolonged use of aminoglycoside treatment.

Conclusion

Even with the advance in PPHN treatment the mortality rate of this lung disease remains high. Animal models of PPHN provide us insight into potential treatment modalities and pathophysiology but cannot completely reflect the disease. Decrease ambient stimulation, continuous sedation, appropriate mechanical ventilation, and inhaled NO treatment is the cornerstone for PPHN management. ECMO should be provided when maximal medical treatment fails. Head ultrasound and echocardiography should be performed before the discussion about ECMO use. Veno-veno-ECMO is the preferred mode of ECMO when myocardial function is adequate. Potential therapeutic modalities such as PDE inhibitor, intravenous citrullin infusion, or BH4 treatment require more studies before clinical application.