Endogenous nitric oxide (NO) is important for the regulation of vascular tone. The enzyme NO synthase generates NO from L-arginine. NO then activates guanyl cyclase and increases the production of cyclic guanosine monophosphate (cGMP). This in turn activates a cascade of GMP-dependent protein kinases resulting in an efflux of calcium from cells and in the relaxation of the vascular smooth muscle. Inhaled nitric oxide (iNO) reaches the alveolar space and diffuses into vascular smooth muscle of the adjacent pulmonary arteries, where it causes vasodilation. When iNO enters the vascular lumen it is rapidly bound to hemoglobin, restricting its effect to the pulmonary circulation. iNO is distributed to the ventilated segments of the lungs and causes increased perfusion in those areas, leading to an improved ventilation/perfusion ratio and improved oxygenation.
iNO therapy causes potent, selective, sustained pulmonary vasodilation and improves oxygenation in term newborns with severe hypoxemic respiratory failure and persistent pulmonary hypertension. Multicenter randomized clinical studies have demonstrated that iNO therapy reduces the need for extracorporeal membrane oxygenation (ECMO) treatment in term neonates with hypoxemic respiratory failure.
The role of iNO therapy has been extensively studied, leading to regulatory approval in 1999 by the U.S. Food and Drug Administration of the treatment of near-term and term newborns with hypoxemic respiratory failure and evidence of persistent pulmonary hypertension of the newborn (PPHN). In this chapter, we review an approach to the initial evaluation of the hypoxemic newborn for treatment with iNO, summarize the clinical experience with iNO in near-term and term newborns, and propose guidelines for the use of iNO in this population. We also review controversies and current evidence for the use of iNO in the premature newborn, although its use remains investigational in this population.
The physiologic rationale for iNO therapy in the treatment of neonatal hypoxemic respiratory failure is based upon its ability to achieve potent and sustained pulmonary vasodilation without decreasing systemic vascular tone ( Fig. 32-1 ). PPHN is a syndrome associated with diverse neonatal cardiac and pulmonary disorders that are characterized by high pulmonary vascular resistance (PVR) causing extrapulmonary right-to-left shunting of blood across the ductus arteriosus and/or foramen ovale (see Chapter 33 ) ( Fig. 32-2 ). Extrapulmonary shunting due to high PVR in severe PPHN can cause critical hypoxemia that is poorly responsive to inspired oxygen or pharmacologic vasodilation. Vasodilator drugs administered intravenously in the past to lower PVR, such as tolazoline and sodium nitroprusside, were often unsuccessful because of systemic hypotension and an inability to achieve or sustain pulmonary vasodilation. Thus the ability of iNO therapy to selectively lower PVR and decrease extrapulmonary venoarterial admixture without affecting blood pressure accounts for the acute improvement in oxygenation observed in newborns with PPHN.
As described in children and adults with severe respiratory failure, oxygenation can improve during iNO therapy in some newborns who do not have extrapulmonary right-to-left shunting. Hypoxemia in these cases is primarily due to intrapulmonary shunting caused by continued perfusion of lung units that lack ventilation (e.g., atelectasis), with variable contributions from ventilation–perfusion inequality. Distinct from its ability to decrease extrapulmonary right-to-left shunting by reducing PVR, low-dose iNO therapy can improve oxygenation by redirecting blood from poorly aerated or diseased lung regions to better aerated distal air spaces (“microselective effect”).
In addition to its effects on vascular tone and reactivity, other physiologic targets for iNO therapy in hypoxemic respiratory failure may include direct effects of NO on lung inflammation, vascular permeability, and thrombosis in situ. Although some laboratory studies have suggested that NO can potentiate lung injury by promoting oxidative or nitrosative stress, inactivating surfactant, and stimulating inflammation, other studies have demonstrated striking antioxidant and antiinflammatory effects in models of lung injury. Thus clinical benefits of low-dose iNO therapy may include reduced lung inflammation and edema, as well as potential protective effects on surfactant function, but these effects remain clinically unproven ( Box 32-1 ).
Pulmonary vasodilation → decreased extrapulmonary right-to-left shunting
Enhanced matching of alveolar ventilation with perfusion
↓ Inflammation (↓ lung neutrophil accumulation)
↓ Vascular leak and lung edema
Preservation of surfactant function
↓ Oxidant injury (inhibition of lipid oxidation)
Preservation of vascular endothelial growth factor expression
Altered proinflammatory gene expression
Finally, the diagnostic value of iNO therapy is important because failure to respond to iNO raises important questions about the specific mechanism of hypoxemia. Poor responses to iNO should lead to further diagnostic evaluation for “unsuspected” anatomic cardiovascular or pulmonary disease.
Physiology of Nitric Oxide in the Pulmonary Circulation
The fetal circulation is characterized by high PVR. Pulmonary blood flow accounts for less than 10% of combined ventricular output in the late-gestation ovine fetus. Mechanisms responsible for maintaining high fetal PVR and causing sustained pulmonary vasodilation at birth are incompletely understood; however, studies in fetal and transitional pulmonary vasoregulation have led to increased understanding of the normal physiologic control of PVR. Fetal and neonatal pulmonary vascular tone is modulated through a balance between vasoconstrictor and vasodilator stimuli, including mechanical factors (e.g., lung volume) and endogenous mediators.
The pharmacologic activity of nitrovasodilators derives from the release of NO, which was recognized as a potent vascular smooth muscle relaxant as early as 1979. In 1987 investigators from two separate laboratories reported that the endothelium-derived relaxing factor was NO or an NO-containing substance. NO modulates basal pulmonary vascular tone in the late-gestation fetus; pharmacologic NO blockade inhibits endothelium-dependent pulmonary vasodilation and attenuates the rise in pulmonary blood flow at delivery, implicating endogenous NO formation in postnatal adaptation after birth. Increased fetal oxygen tension augments endogenous NO release, and the increases in pulmonary blood flow in response to rhythmic distension of the lung and high inspired oxygen concentrations are mediated in part by endogenous NO release. However, in these studies the pulmonary circulation was structurally normal. Studies using a model of PPHN in which marked structural pulmonary vascular changes are induced by prolonged fetal ductus arteriosus compression demonstrated that the structurally abnormal pulmonary circulation also was functionally abnormal. Despite the progressive loss of endothelium-dependent (acetylcholine) vasodilation with prolonged ductus compression in this model, the response to endothelium-independent (atrial natriuretic peptide, NO) vasodilation was intact.
Exogenous (inhaled) NO causes potent, sustained, selective pulmonary vasodilation in the late-gestation ovine fetus. Based on the chronic ambient levels considered to be safe for adults by regulatory agencies in the United States, studies were performed in near-term lambs using iNO at doses of 5, 10, and 20 ppm. iNO caused a dose-dependent increase in pulmonary blood flow in mechanically ventilated newborn lambs. iNO at 20 ppm did not decrease coronary arterial or cerebral blood flow in this model.
Roberts et al. studied the effects of iNO on pulmonary hemodynamics in mechanically ventilated newborn lambs. iNO reversed hypoxic pulmonary vasoconstriction, and maximum vasodilation occurred at doses greater than 80 ppm. They also found that the vasodilation caused by iNO during hypoxia was not attenuated by respiratory acidosis in this model. Berger et al. investigated the effects of iNO on pulmonary vasodilation during group B streptococcal sepsis in piglets. iNO at 150 ppm for 30 minutes caused marked pulmonary vasodilation but was associated with physiologically significant increases in methemoglobin concentrations. Corroborating studies in other animal models support the observations that iNO is a selective pulmonary vasodilator at low doses (less than 20 ppm).
Initial Evaluation of the Term Newborn for Inhaled Nitric Oxide Therapy
Although extensive reference material is available to the clinician when a specific diagnosis has been determined for the hypoxemic term newborn, an approach to the initial evaluation of the cyanotic newborn has received less attention. In this section, we propose an approach to the evaluation of the hypoxemic newborn that may be useful in clarifying the etiology of hypoxemia and in assessing the need for iNO treatment ( Fig. 32-3 ).
Evaluation of the newborn with cyanosis begins with an approach designed to assess the primary cause of hypoxemia. Marked hypoxemia in the newborn can be caused by parenchymal lung disease with mismatch or intrapulmonary shunting, pulmonary vascular disease causing extrapulmonary right-to-left shunting (PPHN), or anatomic right-to-left shunting associated with congenital heart disease. Evaluation should begin with the history and assessment of risk factors for hypoxemic respiratory failure. Relevant history may include the results of prenatal ultrasound studies. Lesions such as congenital diaphragmatic hernia (CDH) and congenital cystic adenomatoid malformation are diagnosed prenatally with increasing frequency. Although many anatomic congenital heart diseases can be diagnosed prenatally, vascular abnormalities (e.g., coarctation of the aorta, total anomalous pulmonary venous return) are more difficult to diagnose with prenatal ultrasound. A history of a structurally normal heart by fetal ultrasonography should be confirmed by echocardiography in the newborn with cyanosis (see Chapter 33 ).
Other historical information that may be important in the evaluation of the cyanotic newborn includes a history of severe and prolonged oligohydramnios causing pulmonary hypoplasia. Absent or a marked decrease in fetal movement over several days and a nonreactive fetal heart rate from time of admission may be indicators of chronic fetal hypoxia and acidosis. Prolonged fetal bradyarrhythmia and/or tachyarrhythmia and marked anemia (caused by hemolysis, twin–twin transfusion, or chronic hemorrhage) may cause congestive heart failure, pulmonary edema, and respiratory distress. Maternal illness (e.g., diabetes mellitus), medication use (e.g., aspirin or medications containing nonsteroidal antiinflammatory drugs causing premature constriction of the ductus arteriosus, association of Ebstein malformation with maternal lithium use), and drug use may contribute to acute cardiopulmonary distress in the newborn. Risk factors for infection that cause sepsis/pneumonia should be considered, including premature or prolonged rupture of membranes, fetal tachycardia, maternal leukocytosis, uterine tenderness, and other signs of intra-amniotic infection.
Events at delivery may provide clues to the etiology of hypoxemic respiratory failure in the newborn. For example, if positive-pressure ventilation is required in the delivery room, the risk of pneumothorax increases. A history of meconium-stained amniotic fluid, particularly if meconium is present below the cords, is the sine qua non of meconium aspiration syndrome. Birth trauma (e.g., clavicular fracture, phrenic nerve injury) or acute fetomaternal or fetoplacental hemorrhage may cause respiratory distress in the newborn.
The initial physical examination provides important clues to the etiology of cyanosis. Marked respiratory distress in the newborn (retractions, grunting, nasal flaring) suggests the presence of pulmonary parenchymal disease with decreased lung compliance. However, it is important to recognize that upper airway obstruction (e.g., Pierre Robin sequence or choanal atresia) and metabolic acidemia also can cause severe respiratory distress. In contrast, the newborn with cyanosis alone or cyanosis plus tachypnea (“nondistressed tachypnea”) typically has cyanotic congenital heart disease, most commonly transposition of the great vessels (TGV) or idiopathic PPHN.
The presence of a heart murmur in the first hours of life is an important sign in the newborn with cyanosis or respiratory distress. In that setting, it is unusual for the common left-to-right shunt lesions (patent ductus arteriosus, atrial septal defect, ventricular septal defect) to produce an audible murmur because PVR remains high and little turbulence is created across the defect. A murmur that sounds like a ventricular septal defect in the first hours of life is most commonly caused by tricuspid regurgitation (associated with PPHN or asphyxiated myocardium).
Interpretation of Pulse Oximetry Measurements
The interpretation of preductal (right hand) and postductal (lower extremity) saturation by pulse oximetry provides important clues to the etiology of hypoxemia in the newborn. Right-to-left shunting across the ductus arteriosus (but not the patent foramen ovale) causes postductal desaturation (i.e., greater than 5% difference). However, it is important to recognize that variability in oximetry readings may be related to differences in available devices and affected by local perfusion. If the measurements of preductal and postductal SaO 2 are equivalent, this suggests either that the ductus arteriosus is patent and PVR is subsystemic (i.e., the hypoxemia is caused by parenchymal lung disease with intrapulmonary shunting or cyanotic heart disease with ductal-dependent pulmonary blood flow) or that the ductus arteriosus is closed (precluding any interpretation of pulmonary artery pressure without echocardiography). It is uncommon for the ductus arteriosus to close in the first hours of life in the presence of systemic or suprasystemic pulmonary artery pressures.
The most common cause of preductal–postductal gradients in oxygenation is suprasystemic PVR in PPHN. However, ductal-dependent systemic blood flow lesions (hypoplastic left-heart syndrome, critical aortic stenosis, interrupted aortic arch, coarctation) may also present with postductal desaturation. Moreover, anatomic pulmonary vascular disease (alveolar–capillary dysplasia, pulmonary venous stenosis, anomalous venous return with obstruction) can cause suprasystemic PVR with right-to-left shunting across the ductus arteriosus and postductal desaturation.
Finally, the unusual occurrence of markedly lower preductal SaO 2 compared to postductal measurements suggests one of two diagnoses: TGV with pulmonary hypertension or TGV with coarctation of the aorta.
Laboratory and Radiologic Evaluation
One of the most important tests to perform in the evaluation of the newborn with cyanosis is the chest radiograph (CXR). The CXR can demonstrate the classic findings of respiratory distress syndrome (air bronchograms, diffuse granularity, underinflation), diffuse parenchymal lung disease in pneumonia, meconium aspiration syndrome, and CDH. Perhaps the most important question to ask when viewing the CXR is whether the severity of hypoxemia is out of proportion to the radiographic changes ( Table 32-1 ). In other words, marked hypoxemia despite supplemental oxygen in the absence of severe pulmonary parenchymal disease radiographically suggests the presence of an extrapulmonary right-to-left shunt (idiopathic PPHN or cyanotic heart disease). The diagnosis of PPHN without CXR evidence of pulmonary parenchymal disease is sometimes called black lung PPHN .
|Mechanism||Associated Conditions||Response to 100% Oxygen|
|Ventilation–perfusion disturbances (high ratios indicate increased dead space; low ratios indicate alveolar underventilation)||Meconium aspiration, retained lung fluid, pulmonary interstitial emphysema, effects of positioning on gas exchange (i.e., decreased ventilation in dependent lung)||PaO 2 ↑↑|
|Intrapulmonary right-to-left shunt (= 0, blood that passes through nonventilated segments of lung)||Atelectasis, alveolar filling (meconium, blood), bronchial collateral circulation||Little change in PaO 2|
|Extrapulmonary right-to-left shunt||PPHN (right-to-left shunting at FO and DA), cyanotic heart disease||Little change in PaO 2|
Other essential measurements include an arterial blood gas to determine the blood gas tensions and pH, a complete blood count to evaluate for signs of infection, and blood pressure measurements in the right arm and a lower extremity to determine aortic obstruction (interrupted aortic arch, coarctation).
Response to Supplemental Oxygen
Marked improvement in SaO 2 (increase to 100%) with supplemental oxygen (100% oxygen by hood, mask, or endotracheal tube) suggests the presence of intrapulmonary shunt or mismatch resulting from lung disease or reactive PPHN. The response to mask continuous positive airway pressure is also a useful discriminator between severe lung disease and other causes of hypoxemia. Most patients with PPHN have at least a transient improvement in oxygenation in response to interventions such as high inspired oxygen and/or mechanical ventilation. If the preductal SaO 2 never reaches 100%, the likelihood of cyanotic heart disease is high.
Echocardiography has become a vital tool in the clinical management of newborns with hypoxemic respiratory failure. The initial echocardiographic evaluation is important to rule out structural heart disease causing hypoxemia (e.g., coarctation of the aorta, total anomalous pulmonary venous return). Moreover, it is critically important to diagnose congenital heart lesions for which iNO treatment would be contraindicated. In addition to the lesions mentioned earlier, congenital heart diseases that can present with hypoxemia unresponsive to high inspired oxygen concentrations (i.e., dependent on right-to-left shunting across the ductus arteriosus) include critical aortic stenosis, interrupted aortic arch, and hypoplastic left-heart syndrome. Decreasing PVR with iNO in these conditions could lead to systemic hypoperfusion, worsening the clinical course and delaying definitive diagnosis.
Echocardiographic evaluation is an essential component in the initial evaluation and ongoing management of the hypoxemic newborn. Not all hypoxemic term newborns have echocardiographic signs of PPHN. As noted earlier, hypoxemia can be caused by intrapulmonary right-to-left shunting or disturbances associated with severe lung disease. In unusual circumstances, right-to-left shunting can occur across pulmonary-to-systemic collaterals. However, extrapulmonary right-to-left shunting at the foramen ovale and/or ductus arteriosus (PPHN) also complicates hypoxemic respiratory failure and must be assessed to determine initial treatments and evaluate the response to those therapies.
PPHN is defined by the echocardiographic determination of extrapulmonary venoarterial admixture (right-to-left shunting at the foramen ovale and/or ductus arteriosus), not simply evidence of increased PVR (i.e., elevated PVR without extrapulmonary shunting does not directly cause hypoxemia). Echocardiographic signs suggestive of pulmonary hypertension (e.g., increased right ventricular systolic time intervals, septal flattening) are less helpful ( Table 32-2 ).
|Measurement||Findings in PPHN|
|Estimate of PA pressure using Doppler estimate of tricuspid regurgitation jet: 4( V 2 ) + CVP, where V is the peak velocity of tricuspid regurgitation jet (in m/s) and CVP is the central venous pressure||Elevated PA pressure reliably estimated (mm Hg); compare with simultaneous systemic pressure|
|Direction of PDA shunt (by pulsed and color Doppler)||Right-to-left or bidirectional PDA shunting|
|Direction of atrial shunt (by pulsed and color Doppler)||Right-to-left or bidirectional shunting through PFO|
Doppler measurements of atrial and ductal level shunts provide essential information when managing a newborn with hypoxemic respiratory failure. For example, left-to-right shunting at the foramen ovale and ductus arteriosus with marked hypoxemia suggests predominant intrapulmonary shunting, and interventions should be directed at optimizing lung inflation.
Finally, the measurements made with echocardiography can be used to predict or interpret the response or lack of response to various treatments. For example, in the presence of severe left ventricular dysfunction with pulmonary hypertension, pulmonary vasodilation alone may be ineffective in improving oxygenation. The echocardiographic findings in this setting include right-to-left ductal shunting (caused by suprasystemic PVR) and mitral insufficiency with left-to-right atrial shunting. In this setting, efforts to reduce PVR should be accompanied by targeted therapies to increase cardiac performance and decrease left ventricular afterload.
This constellation of findings suggests that left ventricular dysfunction may contribute to pulmonary venous hypertension, such as occurs in congestive heart failure. In this setting, pulmonary vasodilation alone (without improving cardiac performance) will not cause sustained improvement in oxygenation. Careful echocardiographic assessment will provide invaluable information about the underlying pathophysiology and help guide the course of treatment.
The initial echocardiographic evaluation determines both structural and functional (i.e., extrapulmonary right-to-left shunting in PPHN, left ventricular performance) causes of hypoxemia. Serial echocardiography is important to determine the response to interventions (e.g., pulmonary vasodilators) and to reevaluate cases in which specific interventions have not resulted in improvement or have resulted in progressive clinical deterioration. For example, in a patient with extrapulmonary right-to-left shunting and severe lung disease, pulmonary vasodilation might reverse the right-to-left venous admixture with little improvement in systemic oxygenation. These observations unmask the critically important contribution of intrapulmonary shunting to hypoxemia (see also the discussion in Chapter 26 ).
Whom to Treat
Guidelines for the use of iNO therapy are given in Box 32-2 .
Patient profile: Near-term/term newborn of 34 weeks’ or greater gestation in the first week of life with echocardiographic evidence of extrapulmonary right-to-left shunting and OI greater than 25 after effective lung recruitment.
Starting dose: 20 ppm
Monitoring for methemoglobinemia: Monitor percentage methemoglobin by co-oximetry within 4 hours of starting iNO and at 24-hour intervals.
Duration of treatment: Typically less than 5 days.
Discontinuation: FiO 2 less than 0.60 with increase in FiO 2 of less than 0.15 after discontinuation.
ECMO availability: If used in a non-ECMO center, arrangements should be in place to continue iNO during transport.
ECMO , extracorporeal membrane oxygenation; iNO , inhaled nitric oxide; OI , oxygenation index.
Because of its selective pulmonary vasodilator effects, iNO therapy is an important adjunct to available treatments for term newborns with hypoxemic respiratory failure. However, hypoxemic respiratory failure in the term newborn represents a heterogeneous group of disorders, and disease-specific responses have clearly been described.
Several pathophysiologic disturbances contribute to hypoxemia in the newborn infant, including cardiac dysfunction, airway and pulmonary parenchymal abnormalities, and pulmonary vascular disorders. In some newborns with hypoxemic respiratory failure a single mechanism predominates (e.g., extrapulmonary right-to-left shunting in idiopathic PPHN), but more commonly several of these mechanisms contribute to hypoxemia. For example, in a newborn with meconium aspiration syndrome, meconium may obstruct some airways, decreasing ratios and increasing intrapulmonary shunting. Other lung segments may be overventilated relative to perfusion and increase physiologic dead space. Moreover, the same patient may have severe pulmonary hypertension with extrapulmonary right-to-left shunting at the ductus arteriosus and foramen ovale. Not only does the overlap of these mechanisms complicate clinical management, but time-dependent changes in the relative contribution of each mechanism to hypoxemia require continued vigilance as the disease progresses. Therefore, understanding the relative contribution of these different causes of hypoxemia becomes critically important as the inventory of therapeutic options expands.
Considering the important role of parenchymal lung disease in many cases of PPHN, pharmacologic pulmonary vasodilation alone would not be expected to cause sustained clinical improvement. The effects of iNO may be suboptimal when lung volume is decreased in association with pulmonary parenchymal disease. Atelectasis and airspace disease (e.g., pneumonia, pulmonary edema) will decrease effective delivery of iNO to its site of action in terminal lung units, and PVR increases at lung volumes above and below functional residual capacity. In PPHN associated with heterogeneous (“patchy”) parenchymal lung disease, iNO may be effective in optimizing matching by preferentially causing vasodilation within lung units that are well ventilated. The effects of iNO on matching appear to be optimal at low doses (less than 20 ppm). However, in cases complicated by homogeneous (diffuse) parenchymal lung disease and underinflation, pulmonary hypertension may be exacerbated because of the adverse mechanical effects of underinflation on PVR. In this setting, effective treatment of the underlying lung disease is essential (and sometimes sufficient) to resolve the accompanying pulmonary hypertension ( Fig. 32-4 ).