6th Bengt Robertson Memorial Lecture Neonatology 2014;105:323–331 DOI: 10.1159/000360646
Published online: May 30, 2014
Oxygen Supplementation in the Neonatal Period: Changing the Paradigm Máximo Vento Neonatal Research Unit, Division of Neonatology, University and Polytechnic Hospital La Fe, Valencia, Spain
Abstract Oxygen is one of the most widely used drugs in the neonatal period. A lack of knowledge of oxygen metabolism and toxicity has prompted guidelines to fluctuate from liberal use to treat respiratory distress to restriction to avoid retinopathy of prematurity. In recent years, studies performed in the immediate postnatal period have revealed that newly born infants achieve a stable saturation only several minutes after birth. Moreover, the time needed to reach a saturation plateau is inversely proportional to a newborn’s gestational age. As a consequence, guidelines have changed and recommend an individualized supplementation in the first minutes after birth with the inspiratory fraction of oxygen titrated against preductal pulse oximetry. However, randomized controlled trials have concluded that, after postnatal stabilization, keeping preterm babies within a low-saturation target range (85–89%) may lead to increased mortality while keeping them in a higher saturation range (91–95%) increases the risk of retinopathy of prematurity. The present state of the art in the management of oxygen supplementation recommends that caregivers in the delivery room allow preduc-
© 2014 S. Karger AG, Basel 1661–7800/14/1054–0323$39.50/0 E-Mail [email protected] www.karger.com/neo
tal oxygen saturation to spontaneously increase in the first minutes of life; however, if supplemented, it should be titrated according to pulse oximeter readings and kept within the safe margins of the nomogram. Thereafter, if oxygen is still needed, it should be kept within stringent security margins (90–95%) to avoid deleterious consequences. Importantly, in babies with chronic lung disease, oxygen should be supplemented to allow the patient to grow and develop. © 2014 S. Karger AG, Basel
Introduction
Oxygen (O2) is probably the most widely used drug in the neonatal period. Since it was first employed in preterm infants in 1930 [1], the policies regulating its use have changed periodically, reflecting the limited knowledge of O2 metabolism and toxicity [2]. Hence, for a certain period, liberal use was recommended based on the beneficial effects of O2 on patients with respiratory distress and, conversely, when its toxic effects (especially those on the retina) were realized, the recommendation of a drastic reduction of O2 supplementation led to increased mortality [2, 3]. In recent years, major steps towards a better understanding of O2 metabolism [4] and the generalized use of pulse oximetry have extensively Máximo Vento, MD, PhD Neonatal Research Unit, Division of Neonatology University and Polytechnic Hospital La Fe Avenida de Campanar 21, ES–46009 Valencia (Spain) E-Mail maximo.vento @ uv.es
Downloaded by: NYU Medical Center Library 128.122.253.228 - 4/22/2015 6:16:25 PM
Key Words Oxidative stress · Oxygen saturation · Pulse oximetry · Antioxidants · Biomarkers
Aerobic Metabolism, Oxidative Stress, Antioxidant Defenses and Redox Regulation
Multicellular organisms require a substantial amount of energy to accomplish their biologic functions, and this can only be provided by a highly efficient mitochondrial combustion of glucose, amino acids and free fatty acids in the presence of O2 [9]. Highly energized electrons from the tricarboxylic acid cycle (Krebs cycle) are used to generate a transmembrane potential across the electron transport chain. These reducing equivalents maintain the electrochemical gradient that drives adenosine triphosphate synthesis. O2 is the final electron acceptor at complex IV of the respiratory chain. The process involves the complete tetravalent reduction of O2 to H2O without the production of reactive oxygen species (ROS) [6]. The efficiency of aerobic metabolism is significantly greater than anaerobic metabolism; hence, 1 mol of glucose will produce 36 mol in the presence of O2 and 2–4 mol of adenosine triphosphate in the absence of O2 [10]. Under physiologic conditions, approximately 98% of O2 undergoes a complete reduction to form H2O. However, a small percentage (2%) of electrons will leak, causing a partial reduction of O2-producing ROS. Thus, the monovalent reduction of O2 elicits the production of anion superoxide. Additional monovalent reduction of anion superoxide generates hydrogen peroxide (H2O2), while a third monovalent reduction will produce the highly reactive hydroxyl radical [11]. The production of ROS under physiologic and also pathologic conditions is very highly dependent on the balance between O2 concentration in the tissue and the antioxidant capacity. ROS can cause direct structural and/or functional damage and/or interfere with essential redox regulatory elements when acting as free radicals. A free radical can be defined as any molecule capable of independent existence with one or more unpaired electrons in the outer shell (e.g. anion superoxide and hydroxyl radical). However, other ROS such as 324
Neonatology 2014;105:323–331 DOI: 10.1159/000360646
H2O2, that are lacking unpaired electrons in the outer shell, will not behave as free radicals. Free radicals are extremely aggressive towards adjacent molecules whether these are free radicals or not. Thus, they react with nonradical molecules in chain reactions causing damage to DNA, proteins and lipids or by promoting the formation of adducts with DNA. The range of molecular damage produced by free radicals is rather remarkable and encompasses, for instance, lipid peroxidation and nitration, protein oxidation and nitration, protein-thiol depletion, nucleic acid hydroxylation and nitration, DNA strand breakage and DNA adduct formation. Under stressful conditions such as ischemia-reperfusion, inflammation or prolonged hyperoxia ROS may induce apoptosis and/ or necrosis (for reviews see [11–13]). However, antioxidant defenses, present in cellular structures and interstitial tissue, are capable (under normal conditions) of balancing the generation of ROS and the maintenance of cellular redox status. Antioxidant defenses can include enzymatic and nonenzymatic mechanisms. Antioxidant enzymes catalytically remove ROS, thereby decreasing ROS reactivity, and protect proteins through the use of chaperones, transition metal-containing proteins (e.g. transferrin, ferritin and ceruloplasmin) and low-molecular-weight compounds that purposely function as oxidizing or reducing agents to maintain intracellular redox stability [6]. Among the most relevant and ubiquitous antioxidant defenses are the superoxide dismutases which catalyze the dismutation of superoxide anions to H2O2. Catalases and glutathione peroxidases convert H2O2 into H2O and O2. Glutathione peroxidase couples H2O2 reduction to H2O with the oxidation of glutathione to glutathione disulfide. Finally, nonenzymatic antioxidants such as reduced glutathione, vitamin C, vitamin E and β-carotene also function to protect cells from the damaging effects of ROS (for reviews see [4, 6, 11–13]). An adequate balance between pro-oxidant and antioxidant elements is essential for biologic processes. Redox (transmembrane electron potential) regulation depends on control elements, which are functionally organized in redox circuits controlled by glutathione/ glutathione disulfide, thioredoxins, cysteine/cystine and other control nodes. These circuits are isolated from each other and are organelle-dependent (mitochondria, endoplasmic reticulum, nucleus, cytoplasm and the extracellular milieu), highly responsive to redox conditions and can function independently in the signaling and regulation of different biologic processes. Hence, oxidative stress would be the consequence of disruption of these circuits by different mechanisms [14]. Vento
Downloaded by: NYU Medical Center Library 128.122.253.228 - 4/22/2015 6:16:25 PM
contributed to reducing O2-derived toxicity [5]. In this scenario, the results from a series of clinical studies in the delivery room [6] and during hospitalization [7, 8] have contributed to setting the safety limits for O2 supplementation in the neonatal period. The aims of this review are to provide an updated overview of recent advances in the basic and applied knowledge of O2 metabolism at different stages of postnatal life and to make recommendations for the clinical use of O2 in the neonatal period.
Hypoxia/reoxygenation
Mitochondria ETC H2O2
CYTOPLASM p38 MAPK
Hydroxylases
Fig. 1. Hypoxia and reoxygenation both ଯ HIF-1į
HIF-1į NUCLEUS
Oxygen-Sensing Mechanisms
Hypoxia is detected by O2 sensors to activate hypoxiainducible factor (HIF)-dependent gene expression in all metazoans. It has been proposed that, in mammals, the mitochondrial respiratory chain acts as an O2 sensor that activates hypoxia-dependent gene expression through the activation of oxidant-dependent intracellular signaling. The respiratory chain increases ROS production during hypoxia, stimulating the p38α mitogen-activated protein kinase signaling pathway to induce HIF-dependent gene expression. Interestingly, during anoxia, prolyl-hydroxylase activity and factor-inhibiting HIF-1 are blocked because of the lack of O2 (fig. 1) [15]. In adults, the arterial chemoreceptors, especially the carotid bodies, respond to hypoxia almost immediately and trigger autonomic reflexes that ensure the adequate delivery of O2 to the tissues. In neonates, however, immature carotid bodies barely respond to hypoxia, but adrenal medullary chromaffin cells are highly responsive to low O2, secreting catecholamines into the general circulation and thus eliciting an adaptive response [16]. In addition, neonatal carotid bodies are greatly responsive to intermittent hypoxia, displaying hyperplasia of glomus cells, and the hypersensitivity to hypoxia is long-lasting [17]. Intermittent hypoxia increases the ROS content of carotO2 Supplementation in the Neonatal Period: Changing the Paradigm
VHL-mediated degradation
HIF-1DŽ
Target gene transcription
Angiogenesis Cell survival Glucose metabolism invasion
id bodies, which induces increased synthesis and stability of hypoxia-inducible factor 1α (HIF-1α) and calpain-dependent degradation of HIF-2α. Thus, intermittent hypoxia will cause a disruption of the equilibrium between HIF-1α and HIF-2α, causing a loss of redox homeostasis (fig. 2) [18]. Intermittent hypoxia-evoked exaggerated hypoxic sensitivity might explain the persistent sympathetic excitation and elevated plasma catecholamines that manifest under this condition [16]. Preterm babies are especially prone to intermittent hypoxia and respiratory instability requiring caffeine therapy and intermittent O2 supplementation. Hypoxia reoxygenation causes an increase in ROS production that could lead to persistent sympathetic excitation, the alteration of glomus cells and hypertension with cardiocirculatory and respiratory consequences [6].
Hypoxia in utero Is Relative
Fetal life develops in an environment that is hypoxic, relative to the atmosphere. In utero, the arterial partial pressure of O2 is approximately 25–30 mm Hg (3.3–3.9 kPa) compared to 80–90 mm Hg (10.5–11.8 kPa) in the mother. However, O2 delivery to tissues does not differ in the fetus, the newborn or the adult. Hence, the presence Neonatology 2014;105:323–331 DOI: 10.1159/000360646
325
Downloaded by: NYU Medical Center Library 128.122.253.228 - 4/22/2015 6:16:25 PM
cause the generation of a burst of ROS in the electron transport chain (ETC) of the mitochondria. H2O2 acting as a signaling molecule activates mitogen-activated protein kinases (MAPK), which block prolylhydroxylases/von Hippel-Lindau protein (VHL)-mediated HIF-1α degradation. HIF-1α translocates to the nucleus and binds with HIF-1β, promoting targeted gene transcription (modified from [15]).
(+) HIF-1į NoX2
(+) ROS
[Ca++]
(++)
mTOR PHD
Carotid body and adrenal medulla
Calpain
HIF-2į
Breathing alterations
Increased systemic blood pressure
and/or nitrosative stress and may cause serious complications [21]. Under hypoxic conditions, fetal production of erythropoietin (EPO) increases significantly. EPO does not cross the placenta or accumulate in the tissues, and its concentration in amniotic fluid correlates with fetal oxygenation status. EPO levels in the amniotic fluid correlate highly with biomarkers of oxidative stress (meta-tyrosine/phenylalanine ratio), damage to DNA (8-oxodG/2dG ratio) and nitrosative stress (nitro-tyrosine). Moreover, newly born infants with high levels of EPO in the amniotic fluid have increased perinatal morbidity [22].
SOD2
Fig. 2. Antagonism between HIF-1α and HIF-2α controls redox
status and establishes the set point for oxygen sensing in the carotid bodies and adrenal medulla chromaffin cells and thus regulates catecholamine secretion and influences the breathing pattern and the blood pressure. Intermittent hypoxia causes oxidative stress and leads to an imbalance of the equilibrium between HIF1α and HIF-2α with an increase of the former and decrease of the latter that is manifested by alterations in the respiratory pattern and a tendency towards systemic hypertension (adapted from [18]).
of fetal hemoglobin with a greater affinity for O2 facilitates placental O2 uptake and increases SpO2 for a given arterial partial pressure of O2. Furthermore, the fetus has an extraordinarily high cardiac output (250–300 ml/kg/ min) and, in addition, cardiac afterload is shunted to the most O2-demanding organs, the heart and brain [19]. Oxygenation of the fetus is clearly dependent on the partial pressure gradients between the maternal blood, the placental tissue and the fetal blood and tissue. Interestingly, the level of placental O2 varies with gestation time. Studies performed in human fetuses have shown that before the 12th week of gestation, the intervillous partial pressure of O2 has a median value of approximately 18–20 mm Hg (2.4–2.6 kPa). The embryo is highly sensitive to ROS, which could be teratogenic or cause abortion. The intervillous partial pressure of O2 rises to 60 mm Hg (7.9 kPa) at 14–16 weeks of gestation and thereafter slowly decreases, reaching values of 45–48 mm Hg (5.9–6.3 kPa) at 36 weeks of gestation [19, 20]. The fetus is permanently under physiologic oxidative and nitrosative stress necessary for the morphogenesis and activation of specific metabolic pathways. However, conditions causing fetal hypoxia such as maternal hypertension or insulin-dependent diabetes increase oxidative 326
Neonatology 2014;105:323–331 DOI: 10.1159/000360646
Color has been traditionally used to assess the oxygenation status of the newly born infant immediately after birth, as reflected by the Apgar score [23]. However, studies that have recorded arterial O2 saturation (SpO2) using pulse oximetry have shown that healthy term or nearterm newborn infants do not achieve a stable saturation >90% until several minutes after clamping the cord. Of note, while some babies only need 2–3 min, others will need ≥10 min to achieve SpO2 ≥90%, reflecting a great interindividual variability [24]. To establish a normality range for SpO2, Dawson et al. [25] merged three databases, which included SpO2 retrieved during the first 10 min after birth in term and preterm infants not needing resuscitation. Based on this nomogram, the 2010 American Academy of Pediatrics resuscitation guidelines recommended targeted SpO2 of 60–65% at 1 min, 65–70% at 2 min, 70–75% at 3 min, 75–80% at 4 min, 80–85% at 5 min and 85–95% at 10 min [26]. At present, this is probably the best available guide for keeping the O2 inspiratory fraction (FiO2) within a safety range, but there are still questions that remain unanswered. Thus, recently, it was shown that preterm babies receiving continuous positive pressure with air achieved higher SpO2 in the first minutes after birth than the babies in the nomogram (fig. 3). Intriguingly, female newborn infants were significantly faster at achieving stable saturations ≥90% than paired males of the same gestational age [27]. Another factor possibly influencing postnatal SpO2 evolution is delayed cord-clamping. At present, there is no doubt that even a brief delay in clamping the cord can be beneficial for the newborn infant [28]. Moreover, experimental studies in lambs have shown that delaying cord-clamping until the initiation of ventilation favors stabilization of the cardiopulmonary circulation after birth [29]. The blood gases determined in the cord blood Vento
Downloaded by: NYU Medical Center Library 128.122.253.228 - 4/22/2015 6:16:25 PM
Oxygenation after Birth Is a Gradual Process
O2 Supplementation in the Neonatal Period: Changing the Paradigm
Neonatology 2014;105:323–331 DOI: 10.1159/000360646
100
**
90
** ** *
80 SpO2 (%)
70 60 50 40 30 20
Spontaneously breathing in 21% oxygen CPAP in 21% oxygen
10 0
1
2
3
4 5 6 7 Minutes after birth
8
9
10
Fig. 3. Arterial SpO2 measured by pulse oximetry, in preterm ba-
bies spontaneously breathing with continuous positive airway pressure (CPAP) and air in the first 10 min after birth, plotted against Dawson’s nomogram of preterm babies spontaneously breathing air without respiratory support (modified from [28]).
of healthy newborn infants are significantly different if the umbilical cord is clamped immediately after birth or is delayed until the spontaneous cessation of blood flow [30, 31]. Partial pressure of O2 measured sequentially in arterial cord blood shows a highly significant increase in the first 90 s after birth, but there are no changes in venous cord blood. Moreover, delayed cord-clamping also increased cerebral oxygenation as measured by near infrared spectroscopy in preterm babies at 4 and 24 h after birth [32]. For these reasons, the generation of a nomogram following delayed cord-clamping, in term and especially preterm infants, would provide interesting information to optimize the interventions by caregivers in the delivery room.
Initial FiO2 and O2 Titration in Ill-Adapted Newly Born Infants in the Delivery Room
327
Downloaded by: NYU Medical Center Library 128.122.253.228 - 4/22/2015 6:16:25 PM
In 1993, Ramji et al. [33] performed a small pilot trial showing that it was feasible to resuscitate asphyxiated neonates with room air. Thereafter, a series of studies have corroborated the effectiveness of 21% O2 for the resuscitation of asphyxiated term neonates; moreover, room air drastically reduces oxidative stress [4, 6]. In 2008, Saugstad et al. [34] published an updated review and metaanalysis showing that the use of 21% instead of 100% O2
significantly reduced mortality in term newborn babies with asphyxia. In 2010, the ILCOR (International Liaison Committee on Resuscitation) guidelines were changed: ‘In term infants receiving resuscitation with intermittent positive-pressure ventilation, 100% oxygen conferred no advantage over air in the short term and resulted in increased time to first breath or cry or both. Meta-analyses of these studies showed a decrease in mortality with the group for whom resuscitation was initiated with air’ [35]. Postnatal adaptation of very preterm infants is hampered by the immaturity of both their respiratory and antioxidant systems. To overcome these difficulties, caregivers sometimes need to provide support with positivepressure ventilation and O2. However, given the fragility of the lungs and the lack of antioxidant response, recent updates have advocated gentle management of respiratory insufficiency in this group of patients [36, 37]. O2 supplementation should be guided by preductal pulse oximetry and titrated to keep SpO2 within recommended ranges [24–26, 35]. However, identification of the appropriate initial fraction of inspired oxygen, i.e. the FiO2, is still a matter of debate. Different studies have approached this issue, starting with a lower (50%) initial FiO2. What apparently matters is the total O2 load received by the infant at the end of resuscitation because the capacity of counteracting the deleterious effects of O2 free radicals is limited in very preterm infants. Resuscitation starting with a high initial FiO2 risks delivering too much O2 compared to a low initial FiO2. Hence, in studies comparing higher (90–100%) versus lower (21–30%) initial FiO2, newborn infants receiving a higher O2 load had significantly higher concentrations of oxidative stress biomarkers and were more likely to develop bronchopulmonary dysplasia (BPD) [38, 39]. However, in a recent updated review and meta-analysis of randomized or quasi-randomized controlled studies comparing high (>50%) and low (
Published online: May 30, 2014
Oxygen Supplementation in the Neonatal Period: Changing the Paradigm Máximo Vento Neonatal Research Unit, Division of Neonatology, University and Polytechnic Hospital La Fe, Valencia, Spain
Abstract Oxygen is one of the most widely used drugs in the neonatal period. A lack of knowledge of oxygen metabolism and toxicity has prompted guidelines to fluctuate from liberal use to treat respiratory distress to restriction to avoid retinopathy of prematurity. In recent years, studies performed in the immediate postnatal period have revealed that newly born infants achieve a stable saturation only several minutes after birth. Moreover, the time needed to reach a saturation plateau is inversely proportional to a newborn’s gestational age. As a consequence, guidelines have changed and recommend an individualized supplementation in the first minutes after birth with the inspiratory fraction of oxygen titrated against preductal pulse oximetry. However, randomized controlled trials have concluded that, after postnatal stabilization, keeping preterm babies within a low-saturation target range (85–89%) may lead to increased mortality while keeping them in a higher saturation range (91–95%) increases the risk of retinopathy of prematurity. The present state of the art in the management of oxygen supplementation recommends that caregivers in the delivery room allow preduc-
© 2014 S. Karger AG, Basel 1661–7800/14/1054–0323$39.50/0 E-Mail [email protected] www.karger.com/neo
tal oxygen saturation to spontaneously increase in the first minutes of life; however, if supplemented, it should be titrated according to pulse oximeter readings and kept within the safe margins of the nomogram. Thereafter, if oxygen is still needed, it should be kept within stringent security margins (90–95%) to avoid deleterious consequences. Importantly, in babies with chronic lung disease, oxygen should be supplemented to allow the patient to grow and develop. © 2014 S. Karger AG, Basel
Introduction
Oxygen (O2) is probably the most widely used drug in the neonatal period. Since it was first employed in preterm infants in 1930 [1], the policies regulating its use have changed periodically, reflecting the limited knowledge of O2 metabolism and toxicity [2]. Hence, for a certain period, liberal use was recommended based on the beneficial effects of O2 on patients with respiratory distress and, conversely, when its toxic effects (especially those on the retina) were realized, the recommendation of a drastic reduction of O2 supplementation led to increased mortality [2, 3]. In recent years, major steps towards a better understanding of O2 metabolism [4] and the generalized use of pulse oximetry have extensively Máximo Vento, MD, PhD Neonatal Research Unit, Division of Neonatology University and Polytechnic Hospital La Fe Avenida de Campanar 21, ES–46009 Valencia (Spain) E-Mail maximo.vento @ uv.es
Downloaded by: NYU Medical Center Library 128.122.253.228 - 4/22/2015 6:16:25 PM
Key Words Oxidative stress · Oxygen saturation · Pulse oximetry · Antioxidants · Biomarkers
Aerobic Metabolism, Oxidative Stress, Antioxidant Defenses and Redox Regulation
Multicellular organisms require a substantial amount of energy to accomplish their biologic functions, and this can only be provided by a highly efficient mitochondrial combustion of glucose, amino acids and free fatty acids in the presence of O2 [9]. Highly energized electrons from the tricarboxylic acid cycle (Krebs cycle) are used to generate a transmembrane potential across the electron transport chain. These reducing equivalents maintain the electrochemical gradient that drives adenosine triphosphate synthesis. O2 is the final electron acceptor at complex IV of the respiratory chain. The process involves the complete tetravalent reduction of O2 to H2O without the production of reactive oxygen species (ROS) [6]. The efficiency of aerobic metabolism is significantly greater than anaerobic metabolism; hence, 1 mol of glucose will produce 36 mol in the presence of O2 and 2–4 mol of adenosine triphosphate in the absence of O2 [10]. Under physiologic conditions, approximately 98% of O2 undergoes a complete reduction to form H2O. However, a small percentage (2%) of electrons will leak, causing a partial reduction of O2-producing ROS. Thus, the monovalent reduction of O2 elicits the production of anion superoxide. Additional monovalent reduction of anion superoxide generates hydrogen peroxide (H2O2), while a third monovalent reduction will produce the highly reactive hydroxyl radical [11]. The production of ROS under physiologic and also pathologic conditions is very highly dependent on the balance between O2 concentration in the tissue and the antioxidant capacity. ROS can cause direct structural and/or functional damage and/or interfere with essential redox regulatory elements when acting as free radicals. A free radical can be defined as any molecule capable of independent existence with one or more unpaired electrons in the outer shell (e.g. anion superoxide and hydroxyl radical). However, other ROS such as 324
Neonatology 2014;105:323–331 DOI: 10.1159/000360646
H2O2, that are lacking unpaired electrons in the outer shell, will not behave as free radicals. Free radicals are extremely aggressive towards adjacent molecules whether these are free radicals or not. Thus, they react with nonradical molecules in chain reactions causing damage to DNA, proteins and lipids or by promoting the formation of adducts with DNA. The range of molecular damage produced by free radicals is rather remarkable and encompasses, for instance, lipid peroxidation and nitration, protein oxidation and nitration, protein-thiol depletion, nucleic acid hydroxylation and nitration, DNA strand breakage and DNA adduct formation. Under stressful conditions such as ischemia-reperfusion, inflammation or prolonged hyperoxia ROS may induce apoptosis and/ or necrosis (for reviews see [11–13]). However, antioxidant defenses, present in cellular structures and interstitial tissue, are capable (under normal conditions) of balancing the generation of ROS and the maintenance of cellular redox status. Antioxidant defenses can include enzymatic and nonenzymatic mechanisms. Antioxidant enzymes catalytically remove ROS, thereby decreasing ROS reactivity, and protect proteins through the use of chaperones, transition metal-containing proteins (e.g. transferrin, ferritin and ceruloplasmin) and low-molecular-weight compounds that purposely function as oxidizing or reducing agents to maintain intracellular redox stability [6]. Among the most relevant and ubiquitous antioxidant defenses are the superoxide dismutases which catalyze the dismutation of superoxide anions to H2O2. Catalases and glutathione peroxidases convert H2O2 into H2O and O2. Glutathione peroxidase couples H2O2 reduction to H2O with the oxidation of glutathione to glutathione disulfide. Finally, nonenzymatic antioxidants such as reduced glutathione, vitamin C, vitamin E and β-carotene also function to protect cells from the damaging effects of ROS (for reviews see [4, 6, 11–13]). An adequate balance between pro-oxidant and antioxidant elements is essential for biologic processes. Redox (transmembrane electron potential) regulation depends on control elements, which are functionally organized in redox circuits controlled by glutathione/ glutathione disulfide, thioredoxins, cysteine/cystine and other control nodes. These circuits are isolated from each other and are organelle-dependent (mitochondria, endoplasmic reticulum, nucleus, cytoplasm and the extracellular milieu), highly responsive to redox conditions and can function independently in the signaling and regulation of different biologic processes. Hence, oxidative stress would be the consequence of disruption of these circuits by different mechanisms [14]. Vento
Downloaded by: NYU Medical Center Library 128.122.253.228 - 4/22/2015 6:16:25 PM
contributed to reducing O2-derived toxicity [5]. In this scenario, the results from a series of clinical studies in the delivery room [6] and during hospitalization [7, 8] have contributed to setting the safety limits for O2 supplementation in the neonatal period. The aims of this review are to provide an updated overview of recent advances in the basic and applied knowledge of O2 metabolism at different stages of postnatal life and to make recommendations for the clinical use of O2 in the neonatal period.
Hypoxia/reoxygenation
Mitochondria ETC H2O2
CYTOPLASM p38 MAPK
Hydroxylases
Fig. 1. Hypoxia and reoxygenation both ଯ HIF-1į
HIF-1į NUCLEUS
Oxygen-Sensing Mechanisms
Hypoxia is detected by O2 sensors to activate hypoxiainducible factor (HIF)-dependent gene expression in all metazoans. It has been proposed that, in mammals, the mitochondrial respiratory chain acts as an O2 sensor that activates hypoxia-dependent gene expression through the activation of oxidant-dependent intracellular signaling. The respiratory chain increases ROS production during hypoxia, stimulating the p38α mitogen-activated protein kinase signaling pathway to induce HIF-dependent gene expression. Interestingly, during anoxia, prolyl-hydroxylase activity and factor-inhibiting HIF-1 are blocked because of the lack of O2 (fig. 1) [15]. In adults, the arterial chemoreceptors, especially the carotid bodies, respond to hypoxia almost immediately and trigger autonomic reflexes that ensure the adequate delivery of O2 to the tissues. In neonates, however, immature carotid bodies barely respond to hypoxia, but adrenal medullary chromaffin cells are highly responsive to low O2, secreting catecholamines into the general circulation and thus eliciting an adaptive response [16]. In addition, neonatal carotid bodies are greatly responsive to intermittent hypoxia, displaying hyperplasia of glomus cells, and the hypersensitivity to hypoxia is long-lasting [17]. Intermittent hypoxia increases the ROS content of carotO2 Supplementation in the Neonatal Period: Changing the Paradigm
VHL-mediated degradation
HIF-1DŽ
Target gene transcription
Angiogenesis Cell survival Glucose metabolism invasion
id bodies, which induces increased synthesis and stability of hypoxia-inducible factor 1α (HIF-1α) and calpain-dependent degradation of HIF-2α. Thus, intermittent hypoxia will cause a disruption of the equilibrium between HIF-1α and HIF-2α, causing a loss of redox homeostasis (fig. 2) [18]. Intermittent hypoxia-evoked exaggerated hypoxic sensitivity might explain the persistent sympathetic excitation and elevated plasma catecholamines that manifest under this condition [16]. Preterm babies are especially prone to intermittent hypoxia and respiratory instability requiring caffeine therapy and intermittent O2 supplementation. Hypoxia reoxygenation causes an increase in ROS production that could lead to persistent sympathetic excitation, the alteration of glomus cells and hypertension with cardiocirculatory and respiratory consequences [6].
Hypoxia in utero Is Relative
Fetal life develops in an environment that is hypoxic, relative to the atmosphere. In utero, the arterial partial pressure of O2 is approximately 25–30 mm Hg (3.3–3.9 kPa) compared to 80–90 mm Hg (10.5–11.8 kPa) in the mother. However, O2 delivery to tissues does not differ in the fetus, the newborn or the adult. Hence, the presence Neonatology 2014;105:323–331 DOI: 10.1159/000360646
325
Downloaded by: NYU Medical Center Library 128.122.253.228 - 4/22/2015 6:16:25 PM
cause the generation of a burst of ROS in the electron transport chain (ETC) of the mitochondria. H2O2 acting as a signaling molecule activates mitogen-activated protein kinases (MAPK), which block prolylhydroxylases/von Hippel-Lindau protein (VHL)-mediated HIF-1α degradation. HIF-1α translocates to the nucleus and binds with HIF-1β, promoting targeted gene transcription (modified from [15]).
(+) HIF-1į NoX2
(+) ROS
[Ca++]
(++)
mTOR PHD
Carotid body and adrenal medulla
Calpain
HIF-2į
Breathing alterations
Increased systemic blood pressure
and/or nitrosative stress and may cause serious complications [21]. Under hypoxic conditions, fetal production of erythropoietin (EPO) increases significantly. EPO does not cross the placenta or accumulate in the tissues, and its concentration in amniotic fluid correlates with fetal oxygenation status. EPO levels in the amniotic fluid correlate highly with biomarkers of oxidative stress (meta-tyrosine/phenylalanine ratio), damage to DNA (8-oxodG/2dG ratio) and nitrosative stress (nitro-tyrosine). Moreover, newly born infants with high levels of EPO in the amniotic fluid have increased perinatal morbidity [22].
SOD2
Fig. 2. Antagonism between HIF-1α and HIF-2α controls redox
status and establishes the set point for oxygen sensing in the carotid bodies and adrenal medulla chromaffin cells and thus regulates catecholamine secretion and influences the breathing pattern and the blood pressure. Intermittent hypoxia causes oxidative stress and leads to an imbalance of the equilibrium between HIF1α and HIF-2α with an increase of the former and decrease of the latter that is manifested by alterations in the respiratory pattern and a tendency towards systemic hypertension (adapted from [18]).
of fetal hemoglobin with a greater affinity for O2 facilitates placental O2 uptake and increases SpO2 for a given arterial partial pressure of O2. Furthermore, the fetus has an extraordinarily high cardiac output (250–300 ml/kg/ min) and, in addition, cardiac afterload is shunted to the most O2-demanding organs, the heart and brain [19]. Oxygenation of the fetus is clearly dependent on the partial pressure gradients between the maternal blood, the placental tissue and the fetal blood and tissue. Interestingly, the level of placental O2 varies with gestation time. Studies performed in human fetuses have shown that before the 12th week of gestation, the intervillous partial pressure of O2 has a median value of approximately 18–20 mm Hg (2.4–2.6 kPa). The embryo is highly sensitive to ROS, which could be teratogenic or cause abortion. The intervillous partial pressure of O2 rises to 60 mm Hg (7.9 kPa) at 14–16 weeks of gestation and thereafter slowly decreases, reaching values of 45–48 mm Hg (5.9–6.3 kPa) at 36 weeks of gestation [19, 20]. The fetus is permanently under physiologic oxidative and nitrosative stress necessary for the morphogenesis and activation of specific metabolic pathways. However, conditions causing fetal hypoxia such as maternal hypertension or insulin-dependent diabetes increase oxidative 326
Neonatology 2014;105:323–331 DOI: 10.1159/000360646
Color has been traditionally used to assess the oxygenation status of the newly born infant immediately after birth, as reflected by the Apgar score [23]. However, studies that have recorded arterial O2 saturation (SpO2) using pulse oximetry have shown that healthy term or nearterm newborn infants do not achieve a stable saturation >90% until several minutes after clamping the cord. Of note, while some babies only need 2–3 min, others will need ≥10 min to achieve SpO2 ≥90%, reflecting a great interindividual variability [24]. To establish a normality range for SpO2, Dawson et al. [25] merged three databases, which included SpO2 retrieved during the first 10 min after birth in term and preterm infants not needing resuscitation. Based on this nomogram, the 2010 American Academy of Pediatrics resuscitation guidelines recommended targeted SpO2 of 60–65% at 1 min, 65–70% at 2 min, 70–75% at 3 min, 75–80% at 4 min, 80–85% at 5 min and 85–95% at 10 min [26]. At present, this is probably the best available guide for keeping the O2 inspiratory fraction (FiO2) within a safety range, but there are still questions that remain unanswered. Thus, recently, it was shown that preterm babies receiving continuous positive pressure with air achieved higher SpO2 in the first minutes after birth than the babies in the nomogram (fig. 3). Intriguingly, female newborn infants were significantly faster at achieving stable saturations ≥90% than paired males of the same gestational age [27]. Another factor possibly influencing postnatal SpO2 evolution is delayed cord-clamping. At present, there is no doubt that even a brief delay in clamping the cord can be beneficial for the newborn infant [28]. Moreover, experimental studies in lambs have shown that delaying cord-clamping until the initiation of ventilation favors stabilization of the cardiopulmonary circulation after birth [29]. The blood gases determined in the cord blood Vento
Downloaded by: NYU Medical Center Library 128.122.253.228 - 4/22/2015 6:16:25 PM
Oxygenation after Birth Is a Gradual Process
O2 Supplementation in the Neonatal Period: Changing the Paradigm
Neonatology 2014;105:323–331 DOI: 10.1159/000360646
100
**
90
** ** *
80 SpO2 (%)
70 60 50 40 30 20
Spontaneously breathing in 21% oxygen CPAP in 21% oxygen
10 0
1
2
3
4 5 6 7 Minutes after birth
8
9
10
Fig. 3. Arterial SpO2 measured by pulse oximetry, in preterm ba-
bies spontaneously breathing with continuous positive airway pressure (CPAP) and air in the first 10 min after birth, plotted against Dawson’s nomogram of preterm babies spontaneously breathing air without respiratory support (modified from [28]).
of healthy newborn infants are significantly different if the umbilical cord is clamped immediately after birth or is delayed until the spontaneous cessation of blood flow [30, 31]. Partial pressure of O2 measured sequentially in arterial cord blood shows a highly significant increase in the first 90 s after birth, but there are no changes in venous cord blood. Moreover, delayed cord-clamping also increased cerebral oxygenation as measured by near infrared spectroscopy in preterm babies at 4 and 24 h after birth [32]. For these reasons, the generation of a nomogram following delayed cord-clamping, in term and especially preterm infants, would provide interesting information to optimize the interventions by caregivers in the delivery room.
Initial FiO2 and O2 Titration in Ill-Adapted Newly Born Infants in the Delivery Room
327
Downloaded by: NYU Medical Center Library 128.122.253.228 - 4/22/2015 6:16:25 PM
In 1993, Ramji et al. [33] performed a small pilot trial showing that it was feasible to resuscitate asphyxiated neonates with room air. Thereafter, a series of studies have corroborated the effectiveness of 21% O2 for the resuscitation of asphyxiated term neonates; moreover, room air drastically reduces oxidative stress [4, 6]. In 2008, Saugstad et al. [34] published an updated review and metaanalysis showing that the use of 21% instead of 100% O2
significantly reduced mortality in term newborn babies with asphyxia. In 2010, the ILCOR (International Liaison Committee on Resuscitation) guidelines were changed: ‘In term infants receiving resuscitation with intermittent positive-pressure ventilation, 100% oxygen conferred no advantage over air in the short term and resulted in increased time to first breath or cry or both. Meta-analyses of these studies showed a decrease in mortality with the group for whom resuscitation was initiated with air’ [35]. Postnatal adaptation of very preterm infants is hampered by the immaturity of both their respiratory and antioxidant systems. To overcome these difficulties, caregivers sometimes need to provide support with positivepressure ventilation and O2. However, given the fragility of the lungs and the lack of antioxidant response, recent updates have advocated gentle management of respiratory insufficiency in this group of patients [36, 37]. O2 supplementation should be guided by preductal pulse oximetry and titrated to keep SpO2 within recommended ranges [24–26, 35]. However, identification of the appropriate initial fraction of inspired oxygen, i.e. the FiO2, is still a matter of debate. Different studies have approached this issue, starting with a lower (50%) initial FiO2. What apparently matters is the total O2 load received by the infant at the end of resuscitation because the capacity of counteracting the deleterious effects of O2 free radicals is limited in very preterm infants. Resuscitation starting with a high initial FiO2 risks delivering too much O2 compared to a low initial FiO2. Hence, in studies comparing higher (90–100%) versus lower (21–30%) initial FiO2, newborn infants receiving a higher O2 load had significantly higher concentrations of oxidative stress biomarkers and were more likely to develop bronchopulmonary dysplasia (BPD) [38, 39]. However, in a recent updated review and meta-analysis of randomized or quasi-randomized controlled studies comparing high (>50%) and low (
