Pharmacology & Therapeutics 114 (2007) 318 – 326
`
`www.elsevier.com/locate/pharmthera
`
`Associate editor: R.M. Wadsworth
`Therapy of pulmonary hypertension in neonates and infants
`Thomas Hoehn ⁎
`
`Neonatology and Pediatric Intensive Care Medicine, Department of General Pediatrics, Heinrich-Heine-University, Moorenstr. 5 D-40225 Duesseldorf, Germany
`
`Abstract
`
`Pulmonary hypertension (PH) in newborns and infants can present in its idiopathic form or complicate a long list of other diseases. Most of
`these conditions are either pulmonary or cardiovascular in origin. In the present review our currrent knowledge regarding pathophysiology,
`structural changes, diagnosis, and available treatment options for PH in the age group below 1 year of age is summarized. New treatment options
`available in adults including endothelin receptor antagonists (ETRA) and phosphodiesterase (PDE) inhibitors are presented and the need for
`randomized controlled trials in newborns and infants is emphasized. Future candidates for pharmacotherapy of PH in infants include among others
`vasoactive intestinal polypeptide (VIP), PDE-3 and PDE-4 inhibitors, hydroxymethylglutaryl coenzyme A (HMG-CoA) reductase inhibitors, and
`adrenomedullin (ADM).
`© 2007 Elsevier Inc. All rights reserved.
`
`Keywords: Persistent pulmonary hypertension of the newborn; Pulmonary arterial hypertension; Nitric oxide; Prostacyclin; Sildenafil; Bosentan
`
`Abbreviations: 6MWD, 6-min walking distance; ADM, adrenomedullin; EPC, endothelial progenitor cells; ETRA, endothelin receptor antagonist; HMG-CoA,
`hydroxymethylglutaryl coenzyme A; MCT, monocrotaline; NO, nitric oxide; PAH, pulmonary arterial hypertension; PDE, phosphodiesterase; PH, pulmonary
`hypertension; PPHN, persistent pulmonary hypertension of the newborn; PVR, pulmonary vascular resistance; VEGF, vascular endothelial growth factor; VIP,
`vasoactive intestinal polypeptide.
`
`Contents
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`Introduction .
`1.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`Physiology of perinatal and postnatal changes .
`2.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`3. Definition and incidence of pulmonary hypertension.
`4.
`Structural changes in persistent pulmonary hypertension of the newborn/pulmonary hypertension
`5.
`Functional changes in persistent pulmonary hypertension of the newborn/pulmonary hypertension
`6. Animal models .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`7. Neonatal disease .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`7.1.
`Primary persistent pulmonary hypertension of the newborn .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`7.2.
`Secondary persistent pulmonary hypertension of the newborn .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`7.3.
`Clinical presentation of persistent pulmonary hypertension of the newborn .
`.
`.
`.
`.
`.
`.
`.
`Pulmonary hypertension in infancy .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`8.
`9. Diagnosis of pulmonary hypertension .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`9.1. Newborns .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`9.2.
`Infants .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`319
`319
`320
`320
`320
`321
`321
`321
`322
`322
`322
`322
`322
`322
`
`⁎ Tel.: +49 211 81 18091; fax: +49 211 81 19786.
`E-mail address: thomas.hoehn@uni-duesseldorf.de.
`
`0163-7258/$ - see front matter © 2007 Elsevier Inc. All rights reserved.
`doi:10.1016/j.pharmthera.2007.03.006
`
`318
`
`

`
`T. Hoehn / Pharmacology & Therapeutics 114 (2007) 318–326
`
`319
`
`10.
`
`Therapy of pulmonary hypertension .
`10.1. Newborns .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`10.2.
`Infants.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`Future options .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`11.
`Summary .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`12.
`References .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`
`. 322
`. 322
`. 322
`. 323
`. 324
`. 324
`
`1. Introduction
`
`Pulmonary hypertension (PH) covers a broad clinical
`spectrum ranging from transient neonatal condition to perma-
`nent disabling disease in infancy or childhood. A variety of
`perinatal conditions can trigger persistent PH of the newborn
`(PPHN; Dakshinamurti, 2005). In the majority of cases, PPHN
`can successfully be reversed by specific treatment of the
`underlying condition in addition to treatment of PH. Primary
`PPHN is a rare event, in which cases no triggering condition of
`PPHN can be identified. Treatment of PPHN consists of
`preferably selective pulmonary vasodilation, which was first
`made available by the introduction of inhaled nitric oxide (NO)
`into neonatal clinical practice in the early 1990s (Kinsella et al.,
`1992; Roberts et al., 1992). While starting with considerable
`enthusiasm for the new drug, neonatologists began to realize,
`that a certain proportion of newborns with clinical PPHN did
`not respond favorably to NO (Hoehn & Krause, 2001; Travadi
`& Patole, 2003). Clinical management of this subgroup of
`patients combined with economic considerations and transient
`decreased availability of NO combined to boost the develop-
`ment of alternative drugs, like phosphodiesterase-5 inhibitors
`(PDE) or endothelin-1 receptor antagonists (ETRA) to success-
`fully treat PH. Advantages of the latter medications include a
`prolonged half-life and independence from inhalational equip-
`ment. Unfortunately, due to the low number of neonates and
`infants treated with these substances, there are currently neither
`
`randomized controlled trials nor much experience with long-
`term application of these drugs available.
`
`2. Physiology of perinatal and postnatal changes
`
`A prerequisite for efficient postnatal gas exchange in any
`newborn infant is the clearance of substantial amounts of fetal
`lung fluid (Jain & Dudell, 2006). Rapid clearance of fetal lung
`fluid is a key part of perinatal adaptation and is mediated in
`large part by transepithelial sodium reabsorption through amilo-
`ride-sensitive sodium channels in the alveolar epithelial cells
`(Jain & Eaton, 2006). Failure to achieve this adaptation results
`in respiratory morbidity presenting as transient tachypnea of the
`newborn (term infant) or respiratory distress syndrome (preterm
`infant), depending on gestational age of
`the newborn.
`Particularly when uterine contractions are absent immediately
`prior to delivery, as in selective Cesarean section, there is no
`activation of the amiloride-sensitive sodium channels in the
`alveolar epithelial cells and therefore an increased risk of
`respiratory morbidity.
`Regulation of pulmonary arterial resistance thus determining
`pulmonary blood flow is achieved by the interaction of 3 main
`players: NO, endothelin, and prostaglandins (Ziegler et al.,
`1995). Among these, endothelin increases pulmonary vascular
`resistance (PVR), whereas NO and prostaglandins lead to
`vasodilation and reduced vascular resistance (for details, see
`Fig. 1). Other genes and their products involved in the
`
`Fig. 1. Schematic interactions of endothelin, NO, and prostaglandins between pulmonary endothelial and smooth muscle cells.
`
`319
`
`

`
`320
`
`T. Hoehn / Pharmacology & Therapeutics 114 (2007) 318–326
`
`pathogenesis of PH include prostacyclin synthase, serotonin
`transporters, serine elastases, matrix metalloproteinases
`(MMP), voltage-gated potassium (Kv) channels, angiotensin-
`converting enzyme (ACE), vascular endothelial growth factor
`(VEGF), carbamoyl phosphate synthase, and plasminogen
`activator inhibitor type 1 (PAI-1; Runo & Loyd, 2003). The
`ultimate effect on pulmonary vascular remodeling is a result of
`the interplay of these genetic factors, modifying genes, and
`environmental factors.
`PVR is high in utero when pulmonary blood flow is limited
`to ∼ 8% of total cardiac output and fetal oxygen requirements
`are met by placental blood flow (Lakshminrusimha & Stein-
`horn, 1999).
`Immediately after birth PVR decreases and
`continues to do so for the following 3 months (Haworth,
`1995). The decrease in PVR leads to increases in pulmonary
`blood flow. Higher pressures in the left atrium and in the aorta
`lead to functional closure of the foramen ovale and reverse the
`intrauterine right-to-left shunting across the duct into a left-to-
`right shunting until the duct eventually closes. Any disruption in
`arterial oxygenation can reverse this process and lead to
`increased PVR and subsequent right-to-left shunt with clinical
`cyanosis.
`
`3. Definition and incidence of pulmonary hypertension
`
`The definition of PH is derived from adult patients and
`includes all individuals with mean pulmonary arterial pressures
`N25 mm Hg at rest or N30 mm Hg with exercise no matter what
`age (British Cardiac Society Guidelines and Medical Practice
`Committee, 2001). Since most of these measurements are
`performed by echocardiography, tricuspid regurgitation with a
`Doppler velocity of more than 2.5 m/sec has been used for the
`screening for PH. Most pediatric cardiologists would agree on a
`definition of PH where systolic pulmonary artery pressure
`exceeds 50% of systolic systemic pressure. These measure-
`ments are usually taken from either tricuspid regurgitation
`or from any known connection between systemic and pulmo-
`nary circulation (i.e., patent ductus arteriosus, ventricular septal
`defect; Tulloh, 2006).
`The WHO classification of PH, which has been modified
`lastly in 2003 (Proceedings of the 3rd World Symposium on
`Pulmonary Arterial Hypertension, 2004) is shown in Table 1.
`Conflicting data have been published concerning the in-
`cidence of PPHN. Whereas Farrow et al. (2005) quantified the
`incidence at 0.2% of live-born term infants, others gave a
`higher range of 0.43–6.8 per 1000 (Walsh-Sukys et al., 2000).
`The associated mortality rate of PPHN at the beginning of the
`21st century was given at 10–20%, whereas earlier investiga-
`tions reported mortality rates of up to 50% (Fox et al., 1977).
`According to a single center experience over almost 2 de-
`cades, a high proportion of infants who died from clinically
`suspected idiopathic PPHN were later found to have alveolar
`capillary dysplasia at autopsy (Tibballs & Chow, 2002). A
`decreasing incidence of PPHN could be expected from the
`falling incidence of meconium aspiration syndrome (Yoder
`et al., 2002), a disease which is frequently associated with
`PPHN.
`
`4. Structural changes in persistent pulmonary
`hypertension of the newborn/pulmonary hypertension
`
`Most prominent histologic changes in PPHN include hyper-
`trophy of the perivascular muscular layer in small and large
`pulmonary arteries (see Fig. 2A,B). Ultimately all 3 layers of the
`vascular wall are affected by thickening and extracellular matrix
`deposition, which is summarized by the term ‘pulmonary vascular
`remodeling’ (Jeffery & Wanstall, 2001). The latter condition
`consists of precocious development of muscle in intraacinar
`arteries, proliferation of adventitial connective tissue, and medial
`hypertrophy of preacinar arteries (Geggel et al., 1986).
`
`5. Functional changes in persistent pulmonary
`hypertension of the newborn/pulmonary hypertension
`
`Functional changes in PPHN/PH are mainly related to
`endothelial dysfunction and result in a dysbalance between
`
`Table 1
`Revised clinical classification of PH (adapted from Farber and Loscalzo, 2004;
`Simonneau et al., 2004)
`
`Group I
`Pulmonary arterial hypertension (PAH)
`Idiopathic (IPAH)
`Familial (FPAH)
`Associated with (APAH)
`•Collagen vascular disease
`•Congenital systemic-to-pulmonary shunts
`•Portal hypertension
`•HIV infection
`•Drugs and toxins
`•Other (thyroid disorders, glycogen storage disease, Gaucher disease,
`hereditary hemorrhagic telangiectasia, hemoglobinopathies,
`myeloproliferative disorders, splenectomy)
`Associated with significant venous or capillary involvement
`•Pulmonary veno-occlusive disease (PVOD)
`•Pulmonary capillary hemangiomatosis (PCH)
`PPHN
`
`Group II
`PH with left heart disease
`Left-sided atrial or ventricular heart disease
`Left-sided valvular heart disease
`
`Group III
`PH associated with lung diseases and/or hypoxemia
`Chronic obstructive pulmonary disease
`Interstitial lung disease
`Sleep-disordered breathing
`Alveolar hypoventilation disorders
`Chronic exposure to high altitude
`Developmental abnormalities
`
`Group IV
`PH due to chronic thrombotic and/or embolic disease
`Thromboembolic obstruction of proximal pulmonary arteries
`Thromboembolic obstruction of distal pulmonary arteries
`Nonthrombotic pulmonary embolism (tumor, parasites, foreign material)
`
`Group V
`Miscellaneous
`Sarcoidosis, histiocytosis X, lymphangiomatosis, compression of pulmonary
`vessels (adenopathy, tumor, fibrosing mediastinitis)
`
`320
`
`

`
`T. Hoehn / Pharmacology & Therapeutics 114 (2007) 318–326
`
`321
`
`by ductal ligation in the fetal lamb VEGF expression was
`markedly reduced (Grover et al., 2003). The postnatal model
`of hypoxia-induced PH found increased levels of VEGF
`but abolished vasodilation to VEGF potentially secondary to
`decreased expression of the VEGF receptor 2 (Nadeau et al.,
`2005).
`
`6. Animal models
`
`Several models of PH exist, the most frequently used among
`these are the injection of monocrotaline (MCT) and exposure
`to hypoxia (Campian et al., 2006). These animal models have
`not only been used to characterize the pathophysiology of PH
`and its sequelae such as right ventricular hypertrophy and
`failure but also to test novel therapeutic strategies. Others have
`used a specific rat strain, the Fawn-hooded rat, which tends to
`spontaneously develop PH under certain conditions (Le Cras
`et al., 1999; Tyler et al., 1999). The analysis of Campian et al.
`(2006) suggests that all approaches which have been
`successful in patients (most notably prostacyclin and ETRA)
`are also effective in various animal models. This is not true the
`other way around: results of animal experiments can often not
`been translated into efficacy in clinical studies, which presents
`a valid argument to perform the human studies in adults first,
`and thereafter in infants and children. Additional factors iden-
`tified to affect pulmonary vascular resistance include carbon
`dioxide and pH. In a neonatal lamb model it was shown that
`elevated pH rather than decreased PaCO2 during hyperventi-
`lation appears to be the major factor in moderating the response
`of the pulmonary vessels to acute hypoxia (Lyrene et al., 1985).
`Similar data have been obtained from infants after cardiopul-
`monary bypass for cardiac surgery. In this investigation in-
`creasing the arterial pH by the administration of sodium
`bicarbonate both lowered the pulmonary arterial pressure
`and increased the cardiac index, resulting in a decrease in
`pulmonary vascular resistance. These changes were observed
`without alteration in PaCO2 (Chang et al., 1995). These ob-
`servations should provide sufficient evidence to refrain from
`using hyperventilation in the clinical management of PH,
`whereas running the risk of decreasing cerebral blow flow
`below a critical threshold. Metabolic alkalosis apparently rep-
`resents a highly efficient modulator of PH.
`
`7. Neonatal disease
`
`7.1. Primary persistent pulmonary hypertension of the newborn
`
`The absence of any known trigger of PPHN in a term
`newborn presenting with cyanosis due to PH immediately
`postnatally suggests the presence of primary PPHN. In this
`condition PH appears to be caused by an abnormal pulmonary
`vascular bed rather than functional vasoconstriction due to other
`causes (Murphy et al., 1981). Intrauterine conditions like
`chronic hypoxia have been shown to result in muscularization
`of the pulmonary vasculature in animals and humans and to
`cause sustained hypertension after hypoxia is reversed (Rabi-
`novitch et al., 1981; Stenmark et al., 1987).
`
`Fig. 2. (A) Hematoxylin–eosin staining of small pulmonary artery in control
`infant (magnification 400×, own data). (B) Hematoxylin–eosin staining of
`small pulmonary artery in an infant with PPHN (magnification 400×, own
`data).
`
`vasodilation and vasoconstriction, in which vasoconstriction
`prevails. One factor contributing to vasoconstriction is high
`levels of endothelin-I, which have been shown in patients with
`PPHN (Vitali & Arnold, 2005). Another explanation for pul-
`monary arterial vasoconstriction in PPHN is the finding of
`diminished NO synthesis (Endo et al., 2001). Although endo-
`thelial NO synthase (eNOS) is upregulated in rapid PPHN
`(RPPHN), a particularly severe and fulminant form of PPHN
`(Hoehn et al., 2003), the net effect for the vessel diameter may
`still be vasoconstriction. Once hypoxia is present in PPHN it
`induces synthesis and release of VEGF (Liu et al., 1998).
`Increased levels of VEGF and smooth muscle proliferation
`can both be suppressed by NO. Others have shown beneficial
`effects of recombinant VEGF treatment in a lamb model of
`experimental PPHN (Grover et al., 2005). In this investigation
`VEGF improved endothelium-dependent vasodilation and
`reduced the severity of pulmonary vascular remodeling. Al-
`though VEGF expression is increased in newborns with
`PPHN (Lassus et al., 2001), the effects of VEGF may be
`altered by impaired signaling. An alteration of VEGF sig-
`naling has been suggested by animal studies of PH induced by
`either hypoxia or VEGF inhibition (Grover et al., 2003;
`Nadeau et al., 2005). In an intrauterine model of PH induced
`
`321
`
`

`
`322
`
`T. Hoehn / Pharmacology & Therapeutics 114 (2007) 318–326
`
`7.2. Secondary persistent pulmonary hypertension of the newborn
`
`10. Therapy of pulmonary hypertension
`
`A variety of perinatal and postnatal conditions can cause
`secondary PPHN: asphyxia, hypoxia, acidosis, hypoglycemia
`(Dakshinamurti, 2005), cold stress, infection, sepsis, meconium
`aspiration, lung hypoplasia, congenital diaphragmatic hernia
`(Perreault, 2006), respiratory distress syndrome, congenital
`pulmonary lymphangiectasia (Hoehn et al., 2006b), and
`congenital alveolar capillary dysplasia (Tibballs & Chow,
`2002). Symptoms occur either immediately postnatally, such
`as in perinatal asphyxia or can be delayed by hours, like in
`evolving neonatal sepsis.
`
`7.3. Clinical presentation of persistent
`pulmonary hypertension of the newborn
`
`Neonates with PPHN present with severe cyanosis due to
`extrapulmonary shunting across the duct and the foramen ovale.
`Depending on the underlying condition,
`the associated
`tachydyspnea is more or less pronounced.
`
`8. Pulmonary hypertension in infancy
`
`Symptoms of infantile PH include breathlessness, fainting or
`chest pain during exercise, and exercise-induced syncope
`(Tulloh, 2006). A minority of infants presents with cyanosis,
`hemoptysis, or
`right heart
`failure with ankle edema or
`hepatomegaly (Rosenzweig & Barst, 2005).
`
`9. Diagnosis of pulmonary hypertension
`
`9.1. Newborns
`
`PPHN can be most easily diagnosed by echocardiography
`even by the less experienced echocardiographer. Should this
`expertise be unavailable, simultaneous (or rapid sequential)
`transcutaneous measurements of pre- and postductal oxygen
`saturation can be used. A 5% or greater saturation decrease
`from pre- to postductal values is highly suggestive of the
`presence of extrapulmonary right-to-left shunting (Macdonald
`& Yu, 1992). High oxygen requirements without radiographic
`evidence of parenchymal pulmonary disease in any newborn
`infant should lead the clinician to suspect the presence of
`PPHN.
`
`9.2. Infants
`
`Echocardiography is the most frequently used investigation
`to screen for PH or to confirm or refute clinically suspected PH.
`Not only can cardiac causes of PH be identified (if present), the
`measurement of tricuspid regurgitation velocity enables the
`estimation of right ventricular pressures (Tulloh, 2006). Car-
`diac catheterization remains the gold standard for measurement
`of pulmonary artery pressures. This investigation specifical-
`ly allows the quantification of the effects of PVR-lowering
`medications like NO (Atz et al., 1999) or prostacyclin (Mikhail
`et al., 1997).
`
`10.1. Newborns
`
`Therapy of PPHN in preterm and term infants is primarily
`directed at correction of the underlying condition (e.g., sepsis,
`hypoxia). Additionally oxygen saturation can be kept higher
`than in healthy term infants in order to decrease PVR by making
`use of the vasodilatory effect of oxygen. Due to the toxic effects
`of hyperoxia (retinopathy of prematurity; McColm et al., 2004)
`this is not an option in preterm infants. Prostacyclin (PGI2) is a
`potent vasodilator of both pulmonary and systemic circulation
`acting via induction of cyclic adenosine monophosphate
`(Howard & Morrell, 2005). The nonselective effect of
`prostacyclins resulted in the development of the PGI2 analogue
`iloprost, which can be applied by nebulization and therefore acts
`predominantly on the pulmonary circulation. The discovery of
`the identity of endothelium-derived relaxing factor (EDRF) as
`NO (Ignarro et al., 1987; Furchgott & Vanhoutte, 1989) led to
`the rapid clinical application of NO (Kinsella et al., 1992;
`Roberts et al., 1992). This molecule causes selective pulmonary
`vasodilation not only in adults but also in near-term and term
`infants; data on a large number of newborns have been summa-
`rized in a recent Cochrane analysis (Finer & Barrington, 2006).
`Altogether 14 randomized controlled trials have been included
`in this analysis comparing inhaled NO to either standard care
`without NO or allowing NO for rescue treatment. Inhaled NO
`significantly reduced the combined endpoint need for extracor-
`poreal membrane oxygenation (ECMO) or death. Whereas
`mortality was hardly influenced, the main effect of inhaled NO
`was the avoidance of ECMO. No beneficial effect has been
`shown for subgroups of PPHN as for example infants with
`congenital diaphragmatic hernia (Finer & Barrington, 2006).
`Other medications studied in a small number of infants only
`include the PDE-5 inhibitor sildenafil (Baquero et al., 2006) and
`the nonselective ETRA bosentan (Galie et al., 2004). For both
`substances there is currently insufficient data in newborns in
`order to draw any conclusions for clinical practice.
`
`10.2. Infants
`
`Therapy of PH in children depends on the degree of PH in an
`individual patient and the extent of deterioration in specific
`situations (e.g., respiratory infection). One of the oldest drugs
`used for the treatment of PH is nifedipine, to which only a
`minority of infants respond favorably (Tulloh, 2006). Oxygen has
`been used traditionally for treatment not only under clinical
`conditions but also for home treatment. The vasodilatory response
`to high concentrations of oxygen belongs to the established tests
`during cardiac catheterization testing various vasodilatory condi-
`tions, including the inhalation of NO (Scheurer et al., 2006). The
`introduction of prostacyclin (PGI2) into clinical practice brought a
`sustained benefit for hemodynamics and reduced mortality in
`comparison with historical data (McLaughlin et al., 2002). The
`major disadvantage for prolonged treatment consisted of PGI2's
`short half-life of only 6 min and the necessity of an indwelling
`central line for continuous supply of PGI2. Therefore other PGI2
`
`322
`
`

`
`T. Hoehn / Pharmacology & Therapeutics 114 (2007) 318–326
`
`323
`
`analogues like treprostinil have been developed, which has a half-
`life of 3 hr and can be delivered subcutaneously (McLaughlin
`et al., 2003). Although treprostinil has been licensed for use in the
`United States since 2002, there is only limited experience in
`children. Iloprost is another PGI2 analogue which has a half-life of
`∼20 min and can be delivered both intravenously and by
`nebulization. Inhalation of iloprost can reverse pulmonary arterial
`hypertension (PAH) and vascular structural remodeling in MCT-
`treated rats (Schermuly et al., 2005) but unfortunately there seems
`to occur a rapid tolerance to this substance (Schermuly et al.,
`2006). Due to the half-life of iloprost it needs to be inhaled 6–9
`times/day, which makes it not very convenient to integrate into
`normal social life in infants. The only orally available PGI2
`analogue is beraprost, which has a half-life of ∼40 min and is
`currently licensed in Japan. Data of 2 randomized controlled trials
`in adults suggest neither significant hemodynamic effects in
`patients with PAH (Galie et al., 2002) nor was initial improvement
`in 6-min walking distance (6MWD) sustained at 9 and 12 months
`of therapy (Barst et al., 2003). There are currently no data of RCT
`of beraprost in infants; until now merely case series have been
`published (Limsuwan et al., 2005). In this case series from
`Thailand the authors found no significant effect on pulmonary
`arterial pressures. This suggests that beraprost
`is of limited
`suitability in infants since, it does not appear to reduce PH and
`reverse progression of morphologic pulmonary changes.
`Another class of medication is the group of ETRA. Most data
`exist on the nonselective ET-A/ET-B ETRA bosentan. In animal
`models of PH this substance can prevent and even reverse
`hypoxia-induced PH (Chen et al., 1995) and can also partially
`prevent MCT-induced PH (Hill et al., 1997). Human data in
`adults appear encouraging, although hepatic toxicity remains
`a serious concern (Cohen et al., 2004; Oldfield & Lyseng-
`Williamson, 2006). Retrospective data of bosentan treatment
`in children suggest safety and efficacy (Rosenzweig et al.,
`2005). Other ETRAs under investigation include sitaxsentan
`and ambrisentan; pediatric experience with both substances is
`limited (Abman, 2006). In view of the reversal of PH in animal
`models and the suggested safety and efficacy in the Rosenzweig
`data, bosentan appears a promising substance provided hepatic
`toxicity does not present a major obstacle for treatment of
`infants.
`Yet another substance class makes use of the NO-mediated
`mechanism of pulmonary vasodilation: phophodiesterases
`inactivate cyclic GMP (cGMP). cGMP in turn is generated by
`the activation of soluble guanylate cyclases and activates
`protein kinases to achieve vasodilation by relaxation of the
`smooth muscle cell. Since particularly PDE-5 is expressed in
`the pulmonary circulation, PDE-5 inhibitors, like sildenafil,
`have been shown to inhibit the development of PH in animal
`models of PAH (Sebkhi et al., 2003; Schermuly et al., 2004;
`Ladha et al., 2005; Ryhammer et al., 2006). Pediatric data
`suggest a beneficial effect of oral sildenafil on hemodynamics
`and exercise capacity (6MWD) in children with PAH (Humpl
`et al., 2005). In this investigation in 14 children 6MWD
`increased from 278 ± 114 m to 443 ± 107 m at 6 months and to
`432 ± 156 m at 12 months. Apparently a plateau was reached
`between 6 and 12 months of treatment. Mean pulmonary artery
`
`pressure decreased from a median of 60 mm Hg (range, 50–
`105) to 50 mm Hg (range, 38–84). Intravenous sildenafil has
`been shown to be as efficacious as NO in reducing pulmonary
`vascular resistance in children with congenital heart disease
`(Schulze-Neick et al., 2003). Unfortunately the number of
`infants included in these case series is too low to draw any
`reliable conclusions from these data.
`
`11. Future options
`
`Since the armamentarium of available medications for
`the treatment of PH has increased dramatically, the near future
`will bring various combinations of drugs being already on the
`market. The vast majority of ongoing trials includes patients
`above the age of 18 years (phase III study ‘Combination
`treatment with bosentan and sildenafil
`to patients with
`Eisenmengers syndrome’ from the Rigshospitalet, Denmark
`[ClinicalTrials.gov Identifier: NCT00303004]; phase IV study
`‘To assess the efficacy and safety of sildenafil when added to
`bosentan in the treatment of pulmonary arterial hypertension’
`from Pfizer Investigational Site, San Antonio, TX, United
`States [ClinicalTrials.gov Identifier: NCT00323297]; phase
`IV study ‘Combination therapy of bosentan and aerosolized
`iloprost in idiopathic pulmonary arterial hypertension (IPAH)’
`from Hannover Medical School, Germany [ClinicalTrials.gov
`Identifier: NCT00120380]). Others have also included pe-
`diatric patients above the age of 12 years (phase II study
`‘Trial of iloprost inhaled solution as add-on therapy with
`bosentan in subjects with pulmonary arterial hypertension
`(PAH)’ from CoTherix, United States [ClinicalTrials.gov
`Identifier: NCT00086463]; phase IV study ‘Effects of the
`combination of bosentan and sildenafil versus sildenafil
`monotherapy on pulmonary arterial hypertension (PAH)’
`from Actelion, United States [ClinicalTrials.gov Identifier:
`NCT00303459]). As far as the author is aware there are cur-
`rently no trials enrolling newborns or infants in studies of
`combined therapies of pulmonary antihypertensive drugs.
`Like other medications in newborns and infants, a substantial
`number of drugs is not licensed for this particular age group
`(Conroy et al., 1999). Reasons for this fact are numerous and
`include weight adapted low dosages, subsequently a negligible
`business volume for the pharmaceutical industry, and rela-
`tively high costs to conduct licensing trials in newborns and
`infants. Costs would be expected to be especially high due to
`specific outcome measures in this age group. Beyond cardio-
`vascular and echocardiographic parameters, follow-up would
`nowadays require long-term observations into school age in
`order to assess neurodevelopmental outcome of these infants
`(Hoehn et al., 2006a).
`Future trials in newborns and infants will need to address
`the efficacy of monotherapy (e.g., sildenafil, bosentan) versus
`combined therapies, including prostacyclins and NO. The adult
`experience and convincing data from animal experiments on the
`efficacy of these substances preclude the possibility of placebo-
`controlled trials in newborns and infants. Nevertheless it seems
`important to have the adult experience prior to introduction of
`medications in infants, both for numerical reasons (there are
`
`323
`
`

`
`324
`
`T. Hoehn / Pharmacology & Therapeutics 114 (2007) 318–326
`
`many more adult patients) and for reasons of safety of a given
`compound.
`The pathophysiology in this age group would not favor a
`specific combination of drugs. Potentially a combination of
`NO-dependent vasodilation (e.g., sildenafil) with NO-indepen-
`dent mechanisms of action (prostacyclin) might prove superior.
`At
`least
`in newborns and infants deposition of the active
`substance during inhalation is regarded as highly erratic
`(Everard, 1996). In order to answer the question of optimized
`drug therapy in a scientifically acceptable manner, randomized
`controlled trials are needed in this particular age group.
`Other
`treatments for PH include vasoactive intestinal
`polypeptide (VIP), which has been shown to induce both
`intracellular cAMP and cGMP and which requires intact
`endothelial function including synthesis of endothelial NO
`(Shahbazian et al., 2006). In an investigation in adults with
`PAH-inhaled VIP led to significant improvements in pulmonary
`hemodynamics and 6MWD (Petkov et al., 2003).
`Besides PDE-5 inhibitors significant pulmonary vasodilation
`can be achieved by PDE-3 and PDE-4 inhibitors when either
`combined with each other or in conjunction with a prostacyclin
`analogue (Phillips et al., 2005). In a case series of 4 PPHN
`patients, treatment with the PDE-3 inhibitor milrinone was
`reported to efficiently reduce PH following nonresponse to
`inhaled NO. However, 3 quarters of patients developed
`intraventricular hemorrhage, 2 of which were classified as
`severe (Bassler et al., 2006). This potential side effect of
`milrinone needs to be taken very seriously, since intraventric-
`ular hemorrhage is extremely uncommon in term newborns (as
`opposed to preterm infants).
`Hydroxymethylglutaryl coenzyme A (HMG-CoA) reductase
`inhibitors, called as statins, have been shown to improve
`endothelial
`function and exert antiproliferative effects on
`vascular smooth muscle cells of systemic vessels (Rakotoniaina
`et al., 2006). In a rat model of MCT-induced PH, pravastatin
`was superior to atorvastatin regarding the degree of PH and the
`restoration of endothelium-dependent relaxation.
`Adrenomedullin (ADM) is a potent vasodilator and acts by
`cAMP and NO-dependent mechanisms. In patients with PAH,
`ADM decreases pulmonary arterial pressures and pulmonary
`vascular
`resistance following intratracheal or
`intravenous
`delivery (Murakami et al., 2006). Intravenously administered
`ADM gene-modified endothelial progenitor cells (EPC) were
`incorporated into lung tissues and attenuated MCT-induced PH
`in rats (Zhao et al., 2005). In this rat model of MCT-induced PH,
`endothelial-like progenitor cells were transduced with human
`eNOS, which