Chapter Contents
Part 1: Stabilisation and resuscitation of the newborn 223
Epidemiology of neonatal resuscitation 224
The approach to newborn resuscitation 224
Assessment of the newborn 225
Techniques of resuscitation 228
Supporting the circulation 235
Chest compressions 235
Drugs 235
Meconium-stained liquor 236
A baby born under maternal general anaesthesia 236
Congenital anomalies 236
Upper airway anomalies 236
Congenital diaphragmatic hernia 236
Gastroschisis 236
Hydrops fetalis 236
Babies still attached to the placenta 236
Ethics issues 236
Teaching and training 237
Research in neonatal resuscitation 237
Part 2: Neonatal transport 239
Remit of neonatal transport 240
Clinical and physiological stresses and pretransport stabilisation 240
Clinical care during transport 240
Care at receiving hospital 241
Communication 241
Organisation of services 241
Stabilisation and resuscitation of the newborn
- Colm O’Donnell
- Colin J Morley
Many interventions, including pinching, shaking and electrocution, have been used to revive apparently lifeless newborns over the years, with claims for success made for all ( Fig. 13.1 ; ). A series of animal experiments by a group of physiologists led by Geoffrey Dawes in the 1950s and 1960s demonstrated the importance of positive-pressure ventilation (PPV) in neonatal resuscitation ( ). In the seminal experiment, a term rhesus monkey was delivered and subjected to an acute and total asphyxial insult (its umbilical cord was ligated and a saline-filled bag placed over its head). The monkey made initial rapid breathing efforts but these were ineffective because of the bag. Figure 13.3 is a hand-drawn representation of what was recorded. The heart rate (HR) fell, the animal became more acidaemic and hypotensive and breathing efforts ceased. The animal slid inexorably toward death unless it was revived by endotracheal PPV. This resulted in a prompt increase in HR with recovery in acid–base status and blood pressure and was followed by a return of spontaneous respiration. These studies were responsible for PPV becoming the mainstay of neonatal resuscitation and underpin neonatal resuscitation practice in the delivery room to this day.
Formal courses teaching neonatal resuscitation evolved in the 1980s, notably the Neonatal Resuscitation Program in the USA ( ) and the Newborn Life Support Course in the UK ( ). These courses taught the causes and physiology of neonatal asphyxia and response to resuscitation described by Dawes. They taught a structured approach to assessing and assisting newborns that was agreed to be ‘accepted practice’ by ‘experts in the field’ but based on little hard evidence. Doctors, midwives and nurses have attended these courses in ever-increasing numbers, and attendance at such courses is mandatory for physician training in many countries. In 1996, the International Liaison Committee on Resuscitation (ILCOR), a body representing resuscitation societies around the world, made recommendations for resuscitation of all age groups. The first guidelines to make specific recommendations for newborns were published in 1999 ( ). First convened in 2000 ( ), the Neonatal Task Force of ILCOR has reconvened at 5-yearly intervals to assess neonatal resuscitation research and update the guidelines accordingly (The ; ).
Epidemiology of neonatal resuscitation
It is estimated that up to 10% of newborns receive some help to establish regular breathing at birth, making respiratory support one of the most commonly performed medical interventions ( ). The rates of late fetal death, stillbirth, birth asphyxia and perinatal sepsis are higher in developing countries than in developed countries; consequently, term infants are more frequently resuscitated in developing countries. The more immature an infant, the higher the chance that respiratory support after birth will be needed. In developed countries, preterm birth occurs more frequently than birth asphyxia; hence premature infants constitute the majority who receive respiratory support and resuscitation.
The approach to newborn resuscitation
While most babies require either no or minimal resuscitation, it is important to be prepared for babies who may need resuscitation. Ask about the history of the pregnancy and labour, as many easily identifiable features increase the risk that the baby will need help after birth ( Box 13.1 ).
| Prematurity | The more immature the baby, the more likely he or she will need help |
| Antepartum haemorrhage | This is usually mother’s blood. However, it can be baby’s blood. Rarely, the baby may have exsanguinated (vasa praevia) Some fetal haemorrhages are concealed, e.g. fetomaternal, retroplacental or subgaleal (subaponeurotic) |
| Signs of fetal compromise | Oxygenation of the baby’s vital organs may be compromised |
| Heavily meconium-stained liquor | The baby may be very hypoxic and depressed. He or she may have been gasping in utero and inhaled meconium |
| Very small-for-dates babies | These babies may not cope with the stress of labour |
| Maternal general anaesthetic | The baby may not breathe spontaneously |
| Congenital abnormality affecting the lungs | For example diaphragmatic hernia, cystic adenomatoid malformation. Baby may not breathe effectively |
| Hydrops fetalis | The chest wall may be oedematous and stiff and pleural effusions and ascites may compromise breathing |
| Intrauterine infection | The lungs may be stiff with congenital pneumonia |
| Mechanical factors | Prolapsed cord, difficult forceps delivery, cephalopelvic disproportion, breech delivery, shoulder dystocia, precipitate delivery and prolonged labour all increase the chance of oxygenation being compromised |
| Prolonged rupture of the membranes >3 weeks | If the membranes have been ruptured with little liquor the lungs may not develop properly and lung hypoplasia can occur |
| A large baby | The delivery may be difficult |
About 25% of depressed babies are undiagnosed before birth. This is why everybody who has the responsibility of caring for a neonate must be trained in basic resuscitation techniques.
Always check the equipment before the baby is born.
Newborns have problems that make respiratory support and resuscitation fundamentally different from other ages:
- •
Their lungs are full of ‘lung liquid’ and have never been aerated.
- •
Preterm infants have a very compliant chest wall that distorts with their inspiratory efforts.
- •
In the first few minutes the oxygen saturation is lower and they appear cyanosed.
Many adults needing resuscitation do not have a heart beat or breathe adequately, because they have cardiac dysrhythmia caused by an acute coronary event. Circulatory support with chest compressions, cardioversion and/or drugs is the priority when resuscitating adults ( ). When a baby’s heart beat is slow or absent it is due to progressive myocardial acidaemia and hypoxia, not an acute dysrhythmia. The acidaemia and hypoxia cannot be corrected by chest compressions, adrenaline (epinephrine) or cardioversion alone or in combination. In newborns the acidaemia and hypoxia are treated by aerating the fluid-filled lungs and establishing gas exchange. The priority in neonatal resuscitation – even when faced with an absent heart beat – is always to provide respiratory support.
Assessment of the newborn
In 1953, an American anaesthetist, Dr Virginia Apgar, proposed a method of evaluating newborns to ‘compare the effect of obstetric and anesthetic interventions’ ( Fig. 13.3 ). The Apgar score is the sum of values 0–2 assigned for each of five items (breathing, HR, colour, tone and reflex irritability) 1 minute after birth. Later the score was routinely assessed at 5 and 10 minutes of life. At the outset, Apgar stated that each element of the score was not of equal importance, but thought that weighting the score would make it cumbersome and less likely to be applied ( ). She believed the score would be useful for comparing populations of infants but that it would be neither sensitive nor specific enough to prognosticate in individual infants ( ). In the early 1950s, newborns were largely ignored in the delivery room (Apgar remarked at the time that ‘surely nine months’ observation of the mother merits one minute’s observation of the baby’; ). The Apgar score was hugely important in drawing attention to infants in the delivery room and, more than 50 years later, it is still universally recorded.
The Apgar score is now used for purposes for which it was not intended. Although persistently low Apgar scores are associated with poor neurological outcome in populations, the predictive value of low ). It is relatively easy to calculate the Apgar score for healthy infants but there is disagreement on how to score premature infants and those receiving respiratory support ( ). As the Apgar score is subjective there is considerable interobserver variability in how scores are assigned, which makes it somewhat inaccurate ( ).
Muscle tone
In adults and older children, tone is assessed by examining passive resistance to movement of the limbs over one or more joints for a minute or more. In contrast, a newborn’s tone is assessed in seconds by observing posture and movements. Tone is the first sign that can be evaluated. Though the quality of an infant’s tone is subjective and varies with gestational age, it is an important and frequently underrated sign. Healthy newborns have a flexed posture that is most notable at the hips, knees and elbows ( Ch. 40.1 ); the degree of flexion increases with maturity. Whatever the gestation, a baby with flexed posture or who is moving the limbs has reasonable tone and is not seriously asphyxiated. Clinicians presented with a baby with a flexed posture can thus be reassured and proceed with less haste. A baby who is limp is more likely to be in poor condition and needs immediate close attention.
Spontaneous breathing
For such a fundamental act it is surprising how little is known about how babies begin to breathe after birth. Many infants, even the most immature, cry shortly after delivery. Others may not cry but breathe independently, albeit irregularly, at birth. Breathing movements in babies may be subtle and movement of the abdomen, due to diaphragmatic contraction, is often more easily seen than chest excursion. The majority of newly born very premature babies breathe at birth. reported that 72% of babies of 27–28 weeks’ gestation, born in the early 1970s, had sustained breathing at 1 minute of age without PPV. Among infants <28 weeks born in the modern era, about 70% cry and 80% breathe spontaneously after birth ( ).
Relatively little is known about the nature of the initial breathing. reported that among 18 well term newborns the time of onset of breathing varied from 3 to 80 seconds; the inspired tidal volume varied from 12 to 67 mL (approximately 3.5–14.7 mL/kg), and they often used a negative pressure up to 70 cmH 2 O. described patterns of breathing where preterm and term infants take quick large inspirations, followed by prolonged expiration, usually achieved by crying or laryngeal adduction, associated with contraction of the abdominal muscles causing a high intrathoracic pressure during most of expiration. The presumption is that this ‘expiratory braking’ helps to establish and maintain a functional residual capacity (FRC). A baby’s respiratory efforts are rarely directly measured during neonatal resuscitation and the efficacy of breathing is largely judged by its effect on the infant’s HR. An infant’s breathing efforts must be sufficient to ensure a HR >100 beats per minute (bpm) ( ).
Breathing difficulty may present at or shortly after birth. Many of the signs – intercostal, subcostal and sternal recession – are due to the powerful diaphragm exerting its influence on the newborn’s pliable thoracic cage as it tries to draw air into relatively stiff (poorly compliant) lungs. Expiratory grunting during expiration has long been recognised as a sign of respiratory difficulty. that when infants who were grunting were intubated the grunting ceased; however, the oxygen in their arterial blood ( Pa o 2 ) fell. When the endotracheal tube (ETT) was removed, grunting resumed and the P a o 2 increased. This demonstrated that grunting is a mechanism infants use to preserve their lung volume and create an FRC by generating positive end-expiratory pressure (PEEP) by partially closing their larynx during a forced expiration (see PEEP below).
Heart rate
HR is the most sensitive indicator of a newborn’s condition and it increases rapidly in bradycardic infants with effective resuscitation ( Fig. 13.3 ). Clinically, HR is best determined by listening to the chest with a stethoscope (auscultation). The number of beats in 6 seconds is multiplied by 10 to give the HR in bpm ( ). A short interval – far shorter than the 15–60 seconds used to count HR in older children or adults – is thought necessary because the HR is more rapid and changes quickly in newborns. This has the disadvantage that errors are increased 10-fold and so can be inaccurate ( ). The HR can sometimes be counted by feeling umbilical cord pulsations, but, as they are often impalpable, auscultation is superior to cord palpation, which is no longer recommended. Neither provides information readily available to the resuscitation team, leading to some advocating that the person listening to the HR ‘tap it out’ with a finger as a visual cue. Resuscitative efforts may be interrupted to allow the HR to be determined by auscultation, which is not ideal. It is taught that the HR should be counted every 30 seconds. However, this does not enable a dynamic assessment of the HR changes. Many of these problems may be addressed by using a pulse oximeter (see below), which is now recommended by ILCOR ( ).
It is taught that an infant’s HR should be >100 bpm after birth; that respiratory support should be given if the HR is <100 bpm; and that chest compression should also be given if the HR is <60 bpm ( ). In reality, well term newborns who are not resuscitated frequently have a HR <100 bpm in the first 2 minutes after birth, which rapidly increases with time ( Fig. 13.4 ; ). For babies who have reasonable tone and who are crying or making breathing efforts, HR <100 bpm on its own is not an indication for intervention. However, if a baby is limp, not breathing and the HR is <100 bpm, PPV is needed. If the HR does not increase promptly, the most likely reason is that the PPV is ineffective. Before chest compression is started, it is imperative that ventilation is optimised. If the baby remains bradycardic despite ventilation with PEEP (see later) moving the chest, then chest compressions should be given. However, the most likely reason for continued bradycardia is that the respiratory support is inadequate ( ).
Colour
Compared with adults, children and older infants, the fetus has low P a o 2 (~2–3 kPa) and oxygen saturation of haemoglobin in arterial blood ( S a o 2 , ~50%); hence newborns appear cyanosed immediately after birth. As they aerate their lungs and establish gas exchange, the peripheral oxygen saturation ( S p o 2 ) increases over the next few minutes and they look increasingly pink. They first look pink centrally – i.e. their tongue, lips, face and torso – while their peripheries (arms and legs) become pink later. There is substantial interobserver variability in the determination of pinkness and this determination correlates poorly with zan infant’s S p o 2 ( ). It takes normal newborns several minutes to achieve an S p o 2 >90% ( Fig. 13.5 ; ). It is less important to decide whether or not an infant is pink than to determine oxygenation is improving (becoming ‘pinker’) with time. In the minutes after birth, the oxygenation of a baby receiving respiratory assistance is best measured using a pulse oximeter with the sensor on the right wrist (see below).
Occasionally babies are born very pale. These babies are usually also limp and bradycardic. This sickly yellowish colour signifies poor circulation and acidosis. These babies must be resuscitated without delay, and help will be required. Even more rarely, babies are born a ghostly, milky-white colour. These infants may have better tone and be more responsive than one would expect for an asphyxiated infant. Despite adequate PPV some have bradycardia that responds quickly to infusion of volume. These babies may have had an acute haemorrhage, which is usually heralded by a large antepartum haemorrhage (e.g. placental abruption, vasa praevia), but may be concealed because it was due to a fetomaternal haemorrhage. These babies may require urgent blood transfusion, and on occasion a transfusion of uncrossmatched O-negative blood in the delivery room can be life-saving.
Pulse oximetry
A pulse oximeter measures both HR and S p o 2 using a sensor that has a light-emitting diode and a detector placed in opposition around a baby’s hand or foot. See Chapter 19 for more information on this method of monitoring.
When placed preductally (on the right wrist) oximeters can give data about 20 seconds after the sensor has been applied correctly. If placed immediately after birth a signal can usually be obtained by 90 seconds ( ). Pulse oximeters count the HR more accurately than clinicians listening with a stethoscope or feeling the umbilical cord ( ). In addition, the HR is counted accurately ( ) and displayed continuously, non-invasively and without the need for resuscitation to be interrupted. Fuelled by uncertainty about using oxygen during neonatal resuscitation, pulse oximeters in the delivery room are now advocated by ILCOR ( ).
Interpreting the changing S p o 2 data immediately after birth requires knowledge of the normal changes with time. The preductal S p o 2 of newborns who do not need resuscitation has a wide range and increases during the first 10 minutes. At 1 minute the median (10th and 90th centiles) is 66% (33%, 85%), at 5 minutes it is 89% (72%, 97%), and at 10 minutes it is 96% (87%, 99%) ( Fig. 13.5 ; ). Babies born prematurely or by caesarean section (CS) have lower S p o 2 than those born vaginally at term and it increases more slowly. The appropriate S p o 2 for an infant at any given time – or more pertinently, the value which suggests that the S p o 2 is too low and O 2 should be given – is not yet clear. The recent publication of centile charts ( Fig. 13.5 ) has been very helpful in illustrating the normal range for babies who did not require resuscitation ( ). The appropriate inspired oxygen for infants (particularly preterm) who receive respiratory support at birth is still being investigated. We suggest a target is to keep the S p o 2 >10th centile and <90th centile targets. There is good evidence from animal studies that hypoxia on its own does not cause brain damage. Hypoxia with hypotension and bradycardia is potentially damaging.
Techniques of resuscitation
Delivery-room stabilisation and resuscitation comprise interventions to minimise the infant’s heat loss, and provide respiratory and circulation support, as determined by ongoing assessment of their clinical condition. Measures are taken to reduce heat loss for all babies. Respiratory support is the key to neonatal resuscitation and is always the priority for infants in poor condition at birth. Circulatory support is rarely needed ( ).
Figure 13.6 shows the ILCOR algorithm, on which most other algorithms are based.
Thermoregulation
Hypothermia can cause respiratory compromise, acidosis, circulatory collapse and death in newborns of any gestation. It is of particular concern for premature babies who have less insulation (body fat) and bulk (energy stores) and are more skinny (greater body surface area : body mass ratio) than their term counterparts. Maintaining the body temperature of preterm infants in ‘the thermoneutral zone’ (36–37°C, the range in which the infant’s metabolic demands were least) substantially reduced mortality, and remains a key plank of neonatal care ( Ch. 15 ) ( ).
Though less common than hypothermia, there is an association between hyperthermia after birth and poor neurodevelopmental outcome. Whether this is because hyperthermia is largely due to chorioamnionitis, itself associated with adverse neurological outcomes, or because there is an independent effect of high temperature is unclear.
Therapeutic hypothermia for term babies with a recent hypoxic–ischaemic insult is an effective therapy which is discussed in Chapter 40, part 4 ( ). If a baby is born who is likely to fit the criteria for treatment with therapeutic hypothermia, i.e. apnoeic, requiring respiratory support through an ETT at 10 minutes, low pH of cord blood, it is reasonable to turn off external heat sources (e.g. radiant warmers) to allow the infant to cool passively. However, the baby’s temperature must be closely monitored as it is surprisingly easy to cool these infants more than to the intended level of 33.5°C.
Strategies to keep newborns in the thermoneutral range usually combine measures to provide heat and to reduce heat loss. The simplest external heat source is the mother’s body. Babies may be placed directly (skin to skin) on the mother’s chest and covered with a blanket or towel.
To minimise heat loss, babies should be delivered in a warm room (the World Health Organization recommends 25°C). Evaporative heat loss is a particular concern for all newborns as they are naked and wet at birth ( Ch. 15 ). Term babies should be dried and covered with a hat. If there is concern about keeping a baby warm, heated blankets or a radiant heat source can be used. In general these measures are adequate for term babies. However, despite these measures, hypothermia on admission to the neonatal intensive care unit (NICU) was common among extremely preterm babies in the UK in the mid-1990s (40% of infants <26 weeks <35°C) and the risk of death and disability was increased among these infants ( ). Several randomised trials have examined the effect of placing very premature babies, directly after birth, in clear plastic (food-grade polyethylene) bags without first drying them. These showed significant increases in temperature on admission to the NICU ( ). Polyethylene wrapping exploits a simple principle – light in the infrared spectrum (i.e. heat) is admitted through the polyethylene, while evaporative heat loss is reduced (impressive misting is seen on the inside of the wrap). Polyethylene wrapping has been recommended for reducing hypothermia among preterm infants ( ). It is worth noting, however, that overheating is possible, so it is prudent to check the baby’s temperature if he or she remains under radiant heat in a plastic bag for protracted periods. More recently, demonstrated that a polyethylene cap was as effective as polyethylene wrapping and both were more effective than drying for preventing hypothermia in preterm infants placed under radiant heat. Exothermic mattresses are used as a heat source to prevent hypothermia in preterm infants. The effect of any of these interventions – polyethylene bags, polyethylene caps and exothermic mattresses – in combination is unclear.
Airway management
Compared with humans of other ages, newborns have relatively large heads and floppy airways. Though babies with respiratory difficulty are often placed prone in the nursery to improve their breathing, it is recommended that, immediately after birth, infants are placed supine on a firm surface ( ), although this is not evidence-based. It is recommended that the head is held in a ‘neutral’ (neither flexed nor extended) or ‘sniffing’ position; and, if the infant is not breathing, that ‘jaw thrust’ – gently pushing the baby’s mandible forward with fingers placed behind the angles of the jaw – is used as a means of opening the airway ( ; ). When attempting to maintain an open airway during spontaneous breathing or respiratory support, try not to press on the soft tissues under the chin as this may predispose to airway obstruction by the tongue.
Newborns are frequently suctioned because clear fluid comes from their mouths. Fetal lungs are filled with liquid. Production of lung liquid ceases during labour and infants expel the liquid from the airways and reabsorb it from the lungs after birth. It is normal for clear fluid to well up from the trachea and appear in the oropharynx. There is no evidence that routine suction of the airways after birth aids a baby’s breathing. The use of a standard suction catheter vaguely poked into a baby’s mouth will not clear blood clots or pieces of meconium. Blindly suctioning the naso- and oropharynx of well infants with a suction catheter may cause harm by inducing vagally mediated bradycardia, laryngospasm or local trauma and should be avoided. If blood or meconium is thought to be obstructing the airways it should be removed under direct vision, using a laryngoscope and a large suction catheter. Complete airway obstruction is rarely a cause of collapse in newborns, and, when it does occur, the obstruction is usually due to an anatomical anomaly (e.g. large cystic hygroma, tracheal agenesis) that is not remediable by suction. Suction may be needed if fluid or meconium obscures the view when attempting endotracheal intubation.
Suction may be indicated when a non-vigorous infant is delivered through meconium-stained liquor (see Special circumstances, meconium-stained liquor, below).
Oropharyngeal (Guedel) airways have previously been recommended for use in newborns. These were developed for adults with depressed consciousness. While models for term babies are available, they are not suitable for very preterm infants. Studies of their use in newborns have not been reported. Laryngeal masks have been used for near-term infants with upper airway anomalies. Infants with the Pierre Robin sequence have very small chins (micrognathia). When these infants are placed supine, the tongue may obstruct the airway (glossoptosis). These infants should be nursed prone to help reduce the obstruction. Should this prove ineffective, a nasopharyngeal airway should be used. This may be a specifically designed device or an ETT of diameter appropriate for the infant’s weight shortened to the distance from the infant’s nose to the angle of the mandible. The aim of these devices is to maintain airway patency by relieving tongue obstruction.
Respiratory support
At birth, placental support is removed and the newborn must use the lungs to exchange gas. To do so effectively, the lung fluid must be rapidly cleared, air must enter the lungs and the blood flow through their lungs must increase. Gas exchange is a continuous process, with oxygen and carbon dioxide diffusing across pulmonary capillary membranes throughout the respiratory cycle, whether the baby is breathing in or out. For this to occur it is critical that the infant has an adequate volume of gas in the lungs at the end of expiration, i.e. an FRC. Mechanical stretching as gas enters and is retained within the lung and increased oxygen tension in the alveoli provoke profound vasodilation in the pulmonary vascular bed. The consequent drop in the pulmonary blood pressure results in a large increase in pulmonary blood flow, which allows gas exchange to occur. These physiological changes are most commonly achieved by spontaneous breathing. Occasionally infants are apnoeic or breathe inadequately. The key to successful stabilisation or resuscitation is to establish effective ventilation.
The aim of respiratory support is to assist spontaneous breathing with continuous positive airway pressure (CPAP) through a mask or nasal prongs, or, if the baby is apnoeic, give PPV with a face mask, through nasal prongs or an ETT.
It is critical for effective gas exchange that newborns quickly achieve and maintain an FRC. The key determinant of FRC in ventilated infants is the mean airway pressure (MAP). MAP is determined mainly by the CPAP or PEEP as it is applied for about two-thirds of the respiratory cycle. The peak inflating pressure (PIP) also affects MAP, but to a lesser extent than PEEP, as it is applied for about one-third of the respiratory cycle.
For adequate gas exchange babies need to breathe in and out a tidal volume ( V T ) of gas of ~4–8 mL/kg body weight. The V T that enters the lung is determined principally by the spontaneous inspiratory effort and to a lesser extent by the PIP during PPV. See Chapter 27 part 1 for more information on respiratory physiology.
Expired air ventilation
Mouth to mouth, or more often mouth to mouth and nose, assisted ventilation was advocated in early resuscitation guidelines. It was recommended that resuscitators placed their mouth around the mouth, or mouth and nose, of apnoeic newborns and breathe into the baby sufficient to produce a gentle rise of the baby’s chest. demonstrated that most mothers’ mouths are not large enough to encircle their infant’s mouth and nose, and speculated that mouth to nose resuscitation may be a better alternative. No human studies of any of these techniques have been reported and they are rarely used in developed countries, where manual ventilation devices are available. In places where manual ventilation devices are not available, mouth to mask resuscitation, rather than mouth to mouth/nose, is recommended to reduce the risks of transmitting infections ( ).
Continuous positive airway pressure
Most babies, even the most immature, breathe or cry at birth ( ). If they are very preterm or have obvious respiratory difficulty, with marked sternal recession or irregular breathing, it is important to help them quickly aerate their lungs and facilitate early gas exchange. This is done by providing CPAP as soon as they start breathing. CPAP aids lung aeration, lung fluid clearance and the formation of an FRC and improves oxygenation. CPAP can be given through a face mask or nasal prongs attached to a T-piece device. The pressure should be no less than 5 cmH 2 O and immediately after birth, while the lung fluid is being cleared, a pressure up to 8 cmH 2 O may help. Self-inflating bags (SIBs) and flow-inflating bags (FIBs) cannot be used to provide consistent CPAP.
Aerating the lung and forming an FRC are the most important first steps in any resuscitation. Because this improves gas exchange it should be used before the inspired oxygen is increased. Treating a baby with supplementary oxygen will not be effective if the lung volume is very low. Several randomised controlled trials, enrolling very preterm infants, have shown that early CPAP is very effective for at least half the babies who do not require intubation ventilation or surfactant and have an outcome that is at least as good as babies who are routinely intubated at birth ( ; ).
Positive-pressure ventilation
PPV is indicated from birth if the infant is apnoeic. In general, PPV is started with a face mask. Mask ventilation is usually only needed for a few minutes. Most infants who receive mask ventilation subsequently breathe. Those who do not breathe should be intubated and ventilated.
Face masks
Face masks are the interface most commonly used to give PPV to newborns. While there are few studies to determine which type of mask is superior, round silicone masks with a cushioned rim are the most commonly used model ( ). Round masks are better than Rendell-Baker masks (solid triangular masks without a rim that were developed for inhalational anaesthesia) ( ). Face masks should encircle the infant’s mouth and nose, but not encroach on the eyes ( ); and different sizes are available. During inflations, in mannikin studies ( ; ; ) and in infants during resuscitation ( ) there are gas leaks from face masks which are often large and variable and can interfere with effective resuscitation.
The way the mask is positioned and held is important ( ). With a round Laerdal-type face mask it is important to use the two-finger-top hold with chin lift. The mask is rolled on to the face from the chin and held in place with the thumb and first finger placed either side of the stem, exerting an even pressure vertically downwards. The second and third fingers are placed under the mandible and pull the jaw upwards with an equal pressure to the downward force on the mask. This can be learned by good education and practice ( ). The mask must not be held around the rim as this can cause it to kink and leak badly. If ventilation is still is not effective a good seal may be obtained with a two-handed technique to hold the mask in position and a second person doing the inflations ( ). While tidal volumes adequate for gas exchange are often not delivered to infants during mask PPV ( ), it may be sufficient to stimulate the baby to breathe spontaneously.
Airway obstruction occurs during mask ventilation ( ; ). When babies do not respond to mask ventilation, measures should be taken to ensure the airway is patent (e.g. checking the mask position and hold, repositioning the head, jaw thrust, suctioning the airway) and that enough PIP is being used before continuing with mask ventilation. Infants who still do not respond to mask ventilation with a pressure of 30 cmH 2 O should be intubated for further PPV ( ).
Nasal airways
Infants are primarily nose breathers. Though CPAP is often given to preterm infants via nasal prongs for prolonged periods in the neonatal nursery, respiratory support is less frequently given to newborns with nasal prongs in the delivery room. The use of a nasal tube (an ETT of appropriate internal diameter shortened to ~5 cm; also known as a nasopharyngeal tube, single nasal prong, short tube) in preference to a face mask for preterm infants was associated with a halving in the rate of intubation in the delivery room in a retrospective cohort study ( ) and with a reduction in the rate of mechanical ventilation within 72 hours in a randomised trial ( ). As the nasal tube was only one of a number of differences in the treatment of the groups in both studies, it is difficult to determine the contribution of the nasal tube to the differences in outcomes observed between the groups.
If a short nasal tube is used, it is important that it is placed in the nostril at right angles to the face and angled back over the soft palate. Nasal tubes should not be pushed up the nose (caudally) as they can pierce the cribriform plate and cause brain injury. Binasal cannulae were superior to the seldom-used Rendell-Baker face mask in a randomised trial enrolling term and near-term infants ( ). Studies comparing nasal airways to face masks for respiratory support in the delivery room are ongoing.
Laryngeal mask airway
Laryngeal mask airways (LMAs) were developed as an alternative to ETTs to support the breathing of adults undergoing short periods of general anaesthesia. They are designed to fit over the laryngeal inlet and can be placed quickly without the need for a laryngoscope. Once inserted, a cushioned rim is inflated to achieve a seal. LMAs offer a more stable airway than a face mask and are easier to place correctly than ETTs, even by relatively inexperienced operators. They were designed to be used with patients with normally compliant lungs, not with the stiff lungs frequently encountered in newborns. The use of LMAs has been reported in case series of term infants ( ) and in small numbers of moderately preterm infants ( ). The use of LMAs in newborns is limited, at least in part owing to the lack of sizes small enough for infants weighing less than 2 kg, who constitute the majority of infants being resuscitated. Successful use of LMAs in the setting of upper airway anomalies has been reported ( ). It may thus be prudent to have a neonatal LMA available for use in exceptional circumstances and for people to be trained in its use.
Endotracheal tubes
When infants fail to respond to PPV by face mask, or are judged to need ongoing respiratory support, they should be intubated and ventilated. ETTs used in newborns are sterile, single-use plastic tubes with a uniform diameter. In contrast to those used for children and adults, ETTs for newborns do not have an inflatable cuff. ‘Shouldered’ Cole’s ETTs – tubes with a narrow distal diameter that passes through the vocal cords, and a wider proximal diameter – are now rarely used. ETTs are sized by their internal diameter, with the appropriate size determined by the weight or gestation of the infant (2.5 mm for infants <1000g / ≤27 weeks; 3.0 mm for 1–2 kg/28–36 weeks; 3.5 mm for >2 kg/near term). They are usually marked at 1-cm intervals from the tip and have a black line that acts as a guide to insertion depth. The insertion depth is often determined by the infant’s weight (depth of insertion at lips for oral ETTs (cm) = 6 + (infant’s weight in kg), e.g. for a 1.5-kg baby, approximate depth of insertion = 7.5 cm at the lips); or may be more accurately estimated from the gestational age ( Table 13.1 ; ).
