Reese H. Clark, MD

Differential Diagnosis of Neonatal Pulmonary Disorders

See Table 18-1.

No. Although apnea is often believed to be a provocative factor for SIDS, this relationship has never been causally established. It appears that premature infants with apnea of prematurity are no more likely to die as a result of SIDS than those of comparable gestational age who do not have apnea of prematurity. Premature infants do, however, have a higher SIDS rate than do term infants, suggesting that immaturity of respiratory control may be a component of SIDS. Furthermore, several studies have indicated that unless respiratory patterns of premature infants are recorded, apnea will not be detected because the respiratory abnormalities in these babies are very difficult to see clinically. ¹

Of infants who die as a result of bacteremia, 90% have evidence of pneumonia on postmortem examination. Many of these infants, however, will not have positive blood cultures during life, making the bacteriologic diagnosis of pneumonia difficult. If pneumonia is suspected on the basis of the clinical examination or chest x-ray, it should be treated aggressively until it has clinically resolved or until the child has been treated for a minimum of 10 days.

In premature infants GBS often mimics respiratory distress syndrome (RDS) with a diffuse reticulogranular pattern and air bronchograms. It is unclear whether this pattern indicates simultaneous disease processes (RDS and GBS) or whether GBS disease causes a secondary surfactant deficiency that produces a radiographic appearance similar to that of RDS when a premature infant is infected.

In term neonates the most common GBS appearance mimics that of transient tachypnea of the newborn, with increased perihilar interstitial markings, hyperexpanded lung fields, and small pleural effusions.

Neonates who have low oxygen saturations or arterial oxygen levels (arterial PO₂), normal carbon dioxide levels (PaCO₂), and no signs of respiratory distress usually have cyanotic congenital heart disease. Low arterial PO₂ and a rising PaCO₂ in a neonate with labored breathing (e.g., grunting, retractions, tachypnea) suggest intrinsic lung disease and its attendant intrapulmonary shunt. A high level of PaCO₂ in association with severe retractions, normal arterial PO₂ in minimal oxygen support, and signs of gas trapping on chest radiograph is most consistent with upper airway obstruction. Therefore if an infant is easy to oxygenate and impossible to ventilate, think airway; if the infant is easy to ventilate and impossible to oxygenate, think heart; and if both oxygenation and ventilation are problems, think lung disease.

Neonatal Resuscitation

Airway, airway, airway: Managing the airway is always the most critical aspect of resuscitation. Most neonates who require support in the delivery room will respond to stimulation, opening of the airway, and gentle ventilation with a bag and mask.

Only about 40% oxygen can be delivered without a reservoir. Each time a self-inflating bag is squeezed, room air is drawn into the bag, diluting any oxygen that is connected. When a reservoir is connected, concentrations up to 90% or more of oxygen may be delivered. One of the limitations of the self-inflating bag is that the desired concentration of oxygen cannot be altered easily.

These sizes are reasonable approximations for most infants, but attention should be paid to the ease of introduction of the ET tube into the airway. A 2.5-mm ET tube may be too small for some babies weighing less than 1000 g, and it may be too large for a few infants with birth weights greater than 1000 gm. The ET tube should slide easily into the airway, and a small leak should be audible around the ET tube when pressures of 25 to 30 cm H₂O are exceeded. An ET tube that is too snug may lead to tracheal inflammation and stenosis, whereas an ET tube that is too small simply may not allow adequate gas delivery to the lungs.

The “tip-to-lip” rule for placement is the distance from the ET tube tip to the centimeter marking on the tube itself ( Fig. 18-1). Good approximations are as follows:

No. Compared with expectant management, intubation and suctioning of the apparently vigorous meconium-stained infant does not result in a decreased incidence of meconium aspiration syndrome (MAS) or other respiratory disorders. In addition, it may provoke airway injury, especially if the child is active and moving after delivery. ³⁴

Only (C), epinephrine for heart rate below 60 bpm is currently recommended by the Neonatal Resuscitation Program. The other therapies have their advocates, but most studies have not shown them to be effective adjuncts for neonatal resuscitation.

Two meta-analyses of several randomized controlled trials comparing neonatal resuscitation initiated with room air versus 100% oxygen showed increased survival when resuscitation was initiated with room air. ⁵⁶⁷

Voltaire, Samuel Johnson, Johann Wolfgang von Goethe, Thomas Hardy, Pablo Picasso, and Franklin D. Roosevelt—the world would have been a very different place had these individuals not had the benefit of resuscitation, rudimentary as it was. Remember, there were no board-certified neonatologists until the mid-1970s.

James Blundell (1790-1878), a Scottish obstetrician, used a “silver tracheal pipe” that had a blunt distal end with two side holes. He would slide his fingers over the tongue to feel the epiglottis and guide the tube into the trachea. Blundell would blow air into the tube approximately 30 times per minute to ventilate the baby.

Term gestation, crying or breathing, and good muscle tone are three useful characteristics identifying infants who do not require resuscitation.

A regular sequence of events occurs when an infant becomes hypoxemic and acidemic. Initially, gasping respiratory efforts increase in depth and frequency for up to 3 minutes, followed by approximately 1 minute of primary apnea ( Fig. 18-2). If oxygen (along with stimulation) is provided during the apneic period, respiratory function spontaneously returns. If asphyxia continues, gasping then resumes for a variable period of time, terminating with the “last gasp” and is followed by secondary apnea. During secondary apnea the only way to restore respiratory function is with positive pressure ventilation and high concentrations of oxygen. Thus a linear relationship exists between the duration of asphyxia and the recovery of respiratory function after resuscitation. The longer that artificial ventilation is delayed after the last gasp, the longer it will take to resuscitate the infant. Clinically, however, the two conditions are indistinguishable, although an infant’s cyanosis will become progressively worse over time.

The first breath of an infant has been measured in the delivery room and is reported to be between −30 and −140 cm H₂O. These pressures are needed to overcome the substantial fluid and elastic forces present in the airway at the time of delivery. As surfactant is deposited, however, subsequent breaths rapidly decrease to −4 to −10 cm H₂O. If surfactant is decreased, as is the case in RDS, the baby must continue to exert the original very high effort to continue to inflate the lung. With limited energy reserves this effort soon deteriorates, and respiratory failure ensues.

Studies support both routes of intubation for newborn infants. The oral intubation school argues that because neonates are obligate nose breathers, they will demonstrate increased work of breathing and atelectasis after removal of nasotracheal tubes. On the other hand, nasal intubation proponents assert that orotracheal intubation results in grooving of the palate with subsequent orthodontic problems. Nasal tubes, however, have been associated with injury to the nasal cartilage. Therefore operator skill and institutional tradition are primary considerations in this clinical decision. After extubation, however, there does appear to be a higher incidence of atelectasis with nasal ET tubes. ⁷³

Transitional Physiology and the Asphyxiated Fetus

The correct answer is (C), metabolic acidosis. “Asphyxia” has become a controversial term because difficult deliveries of babies have resulted in so much litigation. The term asphyxia often is used inappropriately to describe infants who experience transient depression or delayed transition, much to the dismay of obstetricians, because of the medicolegal problems associated with birth asphyxia. In general, it is better not to label infants as “asphyxiated,” but simply to describe numerically the metabolic derangements in the blood gases that are present after birth.

Tricuspid regurgitation is the most likely source of the murmur. Tricuspid regurgitation is due to increased pulmonary pressure and the backflow of blood into the right atrium. Although two fetal channels often remain open in this situation of transitional circulation (i.e., the foramen ovale and the ductus arteriosus), the source of heart murmurs is most likely from the associated tricuspid regurgitation.

A neonate who has had hypoxia proximal to delivery that was sufficiently severe to result in hypoxic-ischemic encephalopathy should show evidence of all of the following:

True. The etiology of mental retardation and seizures is not known in most cases.

Dr. Little was an orthopedic surgeon. He saw children with the spasticity and mobility problems associated with CP. Only in the past two decades has it been recognized that only 4% to 10% of CP can be attributed to intrapartum events. That understanding has not, however, prevented the initiation of litigation in many cases of CP, even when no obstetric or neonatal malpractice exists.

Sigmund Freud. Although he is most famous for his work in psychiatry, Freud was a prominent neurologist who made many astute observations in the field.

False. Electronic fetal monitoring has not been shown to be any better than intermittent auscultation of the fetal heart rate. There are no well-controlled trials that show any decline in deaths or CP rates that can be attributed to electronic fetal heart rate monitoring. Although the use of fetal heart rate monitoring has become a standard practice, its prognostic value is currently unclear.

If arterial blood gases were taken from a fetus, the PaO₂ (arterial PO₂) would be in the range of 25 to 35 mm Hg. Although seemingly low, the strong affinity of fetal hemoglobin for oxygen results in a highly saturated blood that is sufficient to meet the metabolic needs of the fetus. There is, however, little additional room for the PO₂ to decrease, and the fetus whose oxygen level begins to decrease even a small amount may develop problems quickly.

No. Fetal distress often manifests as nonreassuring fetal heart rate patterns, meconium staining of the amniotic fluid, or a low 1-minute Apgar score. None of these has any predictive value for long-term neurologic outcome. However, the presence of signs of fetal distress is a good predictor of the need for resuscitation after delivery.

Shorter and less severe periods of asphyxia often reverse spontaneously and may not lead to any long-term damage unless they occur repeatedly. However, complete failure of gas exchange can cause death in as little as 10 minutes.

The outcome of infants with asphyxia depends on several factors:

The significance of ischemia, in particular, cannot be overstated. Unless circulation is restored, the administration of oxygen will not be effective, and acidemia will increase. The ABCs of resuscitation—airway, breathing, and circulation—remain the key to successful outcome in resuscitation.

Virginia Apgar, an anesthesiologist at Columbia Presbyterian Medical Center in New York City, introduced the Apgar scoring system in 1953 to assess newborn infants’ responses to the stress of labor and delivery.

Apgar scores are useful for assessing and describing the condition of neonates after birth and their subsequent transition to an extrauterine state ( Table 18-2). The Apgar scores are generally obtained and totaled at 1 minute and 5 minutes after birth; however, scores should be recorded for longer periods (at 10, 15, and even 20 minutes) if they are low (until the score is ≥7). Low Apgar scores are useful in identifying neonates who are depressed, and the change in score at 1 minute, 5 minutes, and subsequent time intervals is often helpful in assessing the efficacy of resuscitation. Low Apgar scores (<3) that persist beyond 5 minutes have a better correlation with a poor long-term outcome than Apgar scores at 1 minute. ¹¹

The mortality among severely asphyxiated infants is high and can vary from 50% to 75%. Among survivors, long-term neurodevelopmental sequelae are common and occur in approximately one third of infants. Currently, there are no dependable predictors of long-term outcome. The presence and extent of neurologic abnormalities in the early postasphyxial phase and the persistence of abnormal neurologic findings at the time of discharge are the simplest and most effective predictors of long-term outcome. One measure of the severity of early neurologic dysfunction is the clinical staging system developed by Sarnat. Infants with Sarnat stage I encephalopathy are the ones who have mild asphyxia and recover without any significant neurologic sequelae. However, among infants with Sarnat stages II and III encephalopathy, the incidence of long-term neurodevelopmental handicaps can range anywhere from 50% to 100%. In one study of infants who had no detectable heart rate at birth and at 1 minute of age, two thirds died before discharge and 33% of the survivors had severe neurologic handicaps ( Table 18-3).

Younger animals have been shown to have greater resistance to hypoxic-ischemic injury than older animals. Certain areas of the brain, however, appear to be more vulnerable to injury in neonates than in adults and in preterm compared with term infants. The neonatal brain is often described as having some degree of “plasticity,” in which some areas may assume the function of other areas of the brain after injury. To what degree this phenomenon actually takes place is not known, but it may explain why prediction of outcome after neurologic injury in neonates is so fraught with error.

In preterm infants who survive with hypoxic-ischemic injury, periventricular leukomalacia is the most common (and most devastating) neuropathologic lesion. A large percentage of infants with periventricular leukomalacia develop spastic diplegia later in life. In term infants patterns of neuropathologic injury commonly seen are “watershed infarcts” and diffuse cortical necrosis. These infants are at high risk for developing spastic monoplegia, hemiplegia, or quadriplegia.

In asphyxiated infants who have been successfully resuscitated, the central nervous system (CNS) is the most frequently involved site (72%), followed by the kidneys (62%), heart (29%), intestines (29%), and lungs (26%). Multiple organ involvement occurs even as an asphyxiated fetus or neonate tries to redistribute blood to vital organs as a part of the “diving reflex.” Fortunately, injury to these organs (except the CNS) is not permanent, and complete recovery of function can be expected in most infants who survive. However, the presence of multiorgan failure can seriously jeopardize chances of survival in the immediate postnatal period. ¹²

Oliguria is commonly seen in asphyxiated infants and can result from one or more of the following causes:

Although recovery from acute tubular necrosis is common, acute cortical necrosis is usually fatal. Infants with asphyxiated bladder syndrome, with marked distention, usually recover within a few days, as do most infants with SIADH, unless there has been a pituitary infarct.

Until recently, the presence of hypothermia in resuscitated infants was thought to correlate with poor survival and a higher occurrence of complications. However, recent studies have shown that selective cooling of the brain in infants suspected of having severe hypoxic-ischemic brain damage can improve long-term outcome. Multicenter trials from nurseries around the world have been very promising in this regard. It appears, however, that the primary beneficiary is a child with mild to moderate perinatal asphyxia. Infants with more severe forms of injury do not appear to benefit as much. In addition to improving neurodevelopmental outcomes, cooling reduces mortality risk. Analysis of data from all 10 trials that reported mortality rates showed that infants treated with prolonged moderate hypothermia were less likely to die than those who received normal care. A total of 169 (26%) of the 660 infants treated with therapeutic hypothermia died, compared with 217 (33%) of the 660 infants who received standard care (relative risk 0.78, 95% CI 0.66 to 0.93, P=0.005), with a number needed to treat of 14 (95% CI 8 to 47). ¹³¹⁴

The outcome of depressed infants is usually determined by the degree of resuscitative efforts that are necessary. In one study infants who required chest compressions and epinephrine had the worst outcome, with up to 56% dying in the neonatal period and 21% having an intracranial hemorrhage. Other complications noted in recipients of chest compressions included seizures (18%), respiratory distress (68%), and pneumothorax (24%). ¹⁵

Very-low-birth-weight (VLBW) infants have the greatest need for resuscitation at birth, with up to two thirds of infants weighing less than 1500 g requiring some form of resuscitation. The morbidity and mortality rates in VLBW infants requiring cardiopulmonary resuscitation are inversely related to their birth weight. Recent data indicate that VLBW infants do better if they are delivered and cared for at tertiary care centers. The speed of an in-house response team to the delivery room for resuscitation unquestionably is a great advantage of the tertiary care center compared with a community hospital.

Immediate bag-and-mask ventilation is contraindicated when there is thick meconium in the hypopharynx and trachea or if a congenital diaphragmatic hernia is known or suspected. In all instances, however, the resuscitator must weigh the advantages of bag-and-mask therapy with the risks. At times, immediate intubation for suctioning or to avoid abdominal distention may be required.

Make sure the oxygen source is turned on to the bag apparatus (“The heart and lungs can’t run if there’s no gas”).

In a depressed infant with gasping or absent respirations, 100% oxygen should be given by positive pressure ventilation. Depending on the extent of asphyxia (and depression of heart rate), cardiac compressions are usually initiated within 30 seconds. If there is no response (i.e., increased heart rate) after at least 30 seconds of positive pressure ventilation with 100% oxygen and chest compressions, epinephrine is indicated. As Kattwinkel et al. explain, “The recommended IV dose is 0.01 to 0.03 mg/kg per dose. Higher IV doses are not recommended because animal and pediatric studies show exaggerated hypertension, decreased myocardial function, and worse neurologic function after administration of IV doses in the range of 0.1 mg/kg. If the endotracheal route is used, doses of 0.01 or 0.03 mg/kg will likely be ineffective. Therefore, IV administration of 0.01 to 0.03 mg/kg per dose (0.1-0.3 mL of the 1:10,000 solution), is the preferred route.” ¹⁶

Unless ventilation is adequate, the carbon dioxide produced by the buffering reaction will not be eliminated and will act as a weak acid, further reducing the pH (“closed flask” phenomenon). It is therefore inappropriate and even dangerous to give bicarbonate until ventilation has been established and is found to be adequate.

The relative risks of sodium bicarbonate therapy in infants are related to dosage (higher > lower), rapidity of administration (faster > slower), and osmolality (higher > lower). Physiologic complications include a transient increase in PaCO₂ and fall in arterial PO₂. The sudden expansion of blood volume and increase in cerebral blood flow may increase the risk of periventricular and intraventricular hemorrhage in preterm infants. Other potential complications include the development of hypernatremia and metabolic alkalosis. For these reasons sodium bicarbonate is not recommended as a routine part of the Neonatal Resuscitation Program (NRP).

Generally, it is safest to correct half the base deficit initially and then reassess acid–base status to determine whether further correction is necessary. Administration of bicarbonate to improve acidosis will often expand the circulation and provide additional acid washout, reducing the need for further therapy. Under optimal circumstances, sodium bicarbonate should be infused in small doses over 10 to 20 minutes as a dilute solution (0.5 mEq/mL). It is sometimes not possible to take that much time to administer bicarbonate.

Naloxone has a history of being remarkably free of adverse effects, except for the possible precipitation of sudden drug withdrawal in infants born to drug-addicted mothers. Other reported side effects relate to the sudden release of catecholamines, which can cause hypertension, sudden cardiac arrest, and cardiac dysrhythmias. It is important to remember that the half-life of naloxone is significantly shorter than that of opioids. As a result, new guidelines from NRP are as follows: “Administration of naloxone is not recommended as part of initial resuscitative efforts in the delivery room for newborns with respiratory depression. Heart rate and oxygenation should be restored by supporting ventilation.”

Heart rate is the main variable through which an infant can increase cardiac output. A baby cannot significantly change stroke volume. Bradycardia will therefore significantly reduce a newborn’s cardiac output.

NRP 2010 states that “chest compressions are indicated for a heart rate that is less than 60 per minute despite adequate ventilation with supplementary oxygen for 30 seconds. Because ventilation is the most effective action in neonatal resuscitation and because chest compressions are likely to compete with effective ventilation, rescuers should ensure that assisted ventilation is being delivered optimally before starting chest compressions. Compressions should be delivered on the lower third of the sternum to a depth of approximately one third of the anterior-posterior diameter of the chest. Two techniques have been described: compression with 2 thumbs with fingers encircling the chest and supporting the back (the 2 thumb–encircling hands technique) or compression with 2 fingers with a second hand supporting the back. Because the 2 thumb–encircling hands technique may generate higher peak systolic and coronary perfusion pressure than the 2-finger technique, 76–80, the 2 thumb–encircling hands technique is recommended for performing chest compressions in newly born infants.” ¹⁷

An umbilical venous catheter (UVC) can be placed quickly by trimming the umbilicus to approximately 0.5 to 1 cm in length and inserting the catheter just far enough to obtain blood flow (usually about 3 to 5 cm in term infants). All medications (including vasopressor agents) and fluids can be given through this line. This source of access is often available for many days after birth with appropriate preparation of the cord. The catheter, however, should not be left in this position for a prolonged period of time. If the field has remained sterile (a not terribly common situation in the haste of resuscitation), the catheter should soon be advanced into the inferior vena cava, just below the level of the right atrium. A chest radiograph should be obtained to determine the position. If the catheter has been contaminated, it should be replaced after sterile preparation of the field.

If the heart rate remains below 60 beats per minute despite adequate ventilation with 100% oxygen and chest compressions, administration of epinephrine or volume expansion (or both) may be indicated. Rarely, buffers, narcotic antagonists, or vasopressors may be useful after resuscitation; they are not recommended in the delivery room. The recommended IV dose of epinephrine is 0.01 to 0.03 mg/kg per dose. Higher doses are not recommended.

“Volume expansion should be considered when blood loss is known or suspected (pale skin, poor perfusion, weak pulse) and the baby’s heart rate has not responded adequately to other resuscitative measures. An isotonic crystalloid solution or blood is recommended for volume expansion in the delivery room. The recommended dose is 10 mL/kg, which may need to be repeated. When resuscitating premature infants, care should be taken to avoid giving volume expanders rapidly, because rapid infusions of large volumes have been associated with intraventricular hemorrhage.” ⁷²

Hypoglycemia can be very damaging to the developing nervous system. It can result when hepatic glycogen stores are depleted as a result of stress. A blood glucose level below 40 mg/dL warrants treatment. The clinician should infuse 10% dextrose in water at a dose of 2 mL/kg over 10 to 15 minutes in an attempt to correct hypoglycemia. The target glucose concentration is greater than 45 to 50 mg/dL before each feeding. It is not necessary to use higher concentrations of glucose (e.g., D₂₅W) in such circumstances. After the hypoglycemia has been corrected, normoglycemia can usually be maintained by an infusion rate of 5 to 8 mg/kg/min. In some circumstances hypoglycemia may not be corrected until the infusion rate is 8 to 12 mg/kg/min. Infants receiving levels this high who continue to demonstrate hypoglycemia may have an islet cell adenoma of the pancreas that is producing hyperinsulinemia. ¹⁹

The immediate response to acute blood loss is vasoconstriction to maintain blood pressure. The blood that has been lost contains the same percentage of red blood cells as the blood that is retained. The hematocrit will not drop until fluid repletion of the intravascular volume occurs.

These clinical signs are pulse rate and quality, capillary refill time, and urine output.

No precise answer is possible because clinical circumstances and responses are variable. However, in one study of 58 newborns with an Apgar score of 0 at 10 minutes despite appropriate resuscitative efforts, only 1 of 58 survived, and that infant had profound CP. Studies of therapeutic hypothermia, however, show that initiaton of cooling within 6 hours improves outcomes. According to NRP guidelines (see first reference below), “In a newly born baby with no detectable heart rate, it is appropriate to consider stopping resuscitation if the heart rate remains undetectable for 10 minutes. The decision to continue resuscitation efforts beyond 10 minutes with no heart rate should take into consideration factors such as the presumed etiology of the arrest, the gestation of the baby, the presence or absence of complications, the potential role of therapeutic hypothermia, and the parents’ previously expressed feelings about acceptable risk of morbidity.”

Prolonged resuscitation has a very high risk of ischemic injury to the brain, resulting in cystic encephalomalacia, CP, severe microcephaly, and developmental delay. Failure of response after more than 10 to 15 minutes should prompt the clinician to consider cessation of therapy, as difficult as that always is to do. ²⁰²¹²²

Radiology of Pulmonary Disorders of the Neonate

There are two major schools of thought on this subject. For many years the preferred position was between the third and fourth lumbar vertebrae, as projected on an AP radiograph. The tip lies below the take-off points for the renal and mesenteric arteries, theoretically reducing the risk of injecting fluids or drugs directly into those vessels. With this catheter placement, however, it has been shown that even with relatively low pressure, injectable material can ascend retrograde into the aorta for quite some distance. Other neonatologists prefer a higher placement, in the thoracic aorta at approximately T10 to T12, again avoiding placement of the catheter near the major tributaries off of the descending aorta. Positioning the tip there, however, means that anything injected will flow past major vessels. Several papers have argued for one placement instead of the other, but both are probably safe as long as the clinician takes the following precautions:

Umbilical catheters may be left in place for many days as long as the aforementioned conditions are satisfactorily met. The typical goal is to remove umbilical lines within 7 days of birth. In extreme cases a catheter can be kept in place for 3 weeks without complication. One of the most common errors that is made in neonatal medicine, however, is to leave a catheter in place that is no longer necessary. Because it is not needed, the catheter is often not checked religiously and the risk of complications rises dramatically.

The UVC should be kept at the lower margin of the cardiac silhouette, approximately at the level of the right diaphragm, which would correspond to the junction of the inferior vena cava and right atrium of the heart. UVCs should not be allowed to remain below this level or within any of the branches of the portal system of the liver. Infusion of calcium or hyperalimentation into catheters in these incorrect positions may lead to liver toxicity, portal necrosis, cirrhosis, and cavernous transformation of the portal vein. Umbilical venous lines may also inadvertently cross the foramen ovale and enter the left side of the heart if inserted too far. Catheters in this location occasionally cause rhythm disturbances of the heart. This incorrect placement can be detected by the high levels of PO₂ obtained on a venous sample of blood. A single AP film can be misleading because the left atrium is posterior to the right atrium. Line placement may appear to be appropriate because the AP film does not demostrate how far posterior the line is placed. Lateral films of the chest and abdomen and echocardiograms can be used to confirm appropriate placement.

The optimal position for an ET tube is approximately halfway between the thoracic inlet (look for the medial ends of the baby’s clavicles to get a good approximation) and the carina or level of tracheal bifurcation. In small neonates ET tubes often enter the right mainstem bronchus and produce left-sided atelectasis. They may also exert vagal effects and cause bradycardia or irritation if they strike the carina. Tubes that are excessively high also may produce vagal effects and loss of effective ventilation.

The classic RDS picture in a premature neonate has a diffuse increase in lung density (opacity) with a fine, reticulogranular (grainy) or ground-glass appearance, air bronchograms (a darker appearance to the branching central airway in contrast to the opacity of the lungs), and low lung volumes ( Fig. 18-4).

GBS pneumonia in a premature infant is reported to have an appearance similar to that of RDS. However, premature babies with GBS disease may also have surfactant inactivation or deficiency with true RDS as well as GBS septicemia. Although Sir William Osler might not like the concept of two diagnoses in one little patient, it probably happens more often than not.

Affected babies often have a coarse, irregular increase in lung markings accompanied by hyperinflation of the lungs. The pneumonic process here is one of patchy atelectasis and overdistention ( Fig. 18-5). Pneumomediastinum and pneumothorax are frequent accompanying abnormalities as well.

The textbook description of this clinical condition is a hazy-appearing lung, often with fluid in the right horizontal fissure and increased perihilar markings. The babies have rapid, shallow breathing, in contrast to the retractions of RDS or MAS. It usually is reported to last approximately 24 hours, with 48 to 72 hours considered the maximum. Some infants, however, seem to have this clinical problem for many more days, with subsequent uneventful recovery. What turns off fetal lung fluid production at birth has not been established. One theory is that certain babies may continue to produce a low level of lung fluid for some time after they are born.

A patient with congenital diaphragmatic hernia rarely presents at birth as the diagnostic dilemma in the delivery room, as once was the case. With fetal ultrasound most of these infants are diagnosed before birth. Radiographically, they have a complex pattern of lucency in one hemithorax (usually the left side, and reflecting air-containing loops of intestine), contralateral shift of the heart and other mediastinal structures, and a lack of expected air-containing intestine in the abdomen.

Early air leaks are often difficult to diagnose. The most obvious finding is a separation of the edge or margin of the lung from the inner margin of the chest wall, with no lung markings definable in that space. An AP decubitus view of the chest with the side of suspected pneumothorax to the top (nondependent) is also helpful. For example, if you suspect a left-sided pneumothorax, you should order a “right decubitus AP chest radiograph,” which means the right side of the patient will be dependent and the left side nondependent. If a pneumothorax is present, look for a zone of lucency representing pleural air collecting between the lateral chest margin and the adjacent lung ( Fig. 18-6).

Respiratory Distress Syndrome

The small one collapses more quickly because of surface tension. The LaPlace relationship states P = 2T/R, where P is the pressure across the wall of the sphere, T is surface tension of the substance forming the bubble (i.e., its tendency to collapse), and R is the radius of the sphere. The smaller the radius, the greater the collapsing pressure ( Fig. 18-7).

When the bubbles are alveoli without surfactant, pressure on the alveolar surface is quite high because the surface tension is high. As the alveolus collapses without surfactant during exhalation, pressure increases as the radius of the alveolus decreases.

Avery and Mead described the absence of a surface tension–reducing substance in the alveolar fluid of infants who died of hyaline membrane disease. The substance turned out to be the complex substance known as surfactant, which greatly lowers the alveolar surface tension and therefore the tendency of the alveolus to collapse. Surfactant also lowers surface tension as the diameter of the alveolus decreases, allowing for stable alveoli at end-expiratory volumes.

Within minutes after delivery, pulmonary artery pressure falls, and blood flow increases in response to birth-related stimuli, such as ventilation, increased PO₂, and shear stress. Physical stimuli, including increased shear stress, lung inflation, ventilation, and increased oxygen, cause pulmonary vasodilation in part by increasing production of vasodilators, nitric oxide, and prostacyclin. Pretreatment with the nitric oxide synthase inhibitor, nitro-L-arginine, attenuates pulmonary vasodilation after delivery by 50% in near-term fetal lambs. These findings suggest that a significant part of the rise in pulmonary blood flow at birth may be related directly to the acute release of nitric oxide. Each of the birth-related stimuli can stimulate nitric oxide release independently, followed by vasodilation through cyclic guanosine monophosphate kinase–mediated stimulation of K channels. Although the endothelial isoform of nitric oxide synthase III has been presumed to be the major contributor of nitric oxide at birth, recent studies suggest that other isoforms (neuronal type I and inducible type II) may be important sources of nitric oxide release in utero and at birth. Other vasodilators, especially prostacyclin, also modulate changes in pulmonary vascular tone. Rhythmic lung distention and shear stress stimulate both prostacyclin and nitric oxide production in late gestation. Increasing oxygen tension also triggers nitric oxide activity and overcomes the effects of prostacyclin inhibition at birth. Thus, although nitric oxide does not account for the entire fall in pulmonary vascular resistance at birth, nitric oxide synthase activity appears important in achieving postnatal adaptation of lung circulation.

More recently, vasculoendothelial growth factor (VEGF), which is really a family of growth factors, and angiopoietins 1 through 4 have been also shown to have important angiogenic roles during fetal development of the pulmonary vascular bed. These agents also appear to be closely related to the release and influence of nitric oxide in pulmonary development and vasodilation. Additional proteins have been elaborated that may play critical roles in this entire process, demonstrating the increasing complexity of our understanding of neonatal lung development. ²³

Surfactant, from “surface active material,” is 80% phospholipids and 8% neutral lipids. The phospholipid most responsible for surface tension reduction is dipalmitoylphosphatidylcholine. About 12% of surfactant is protein, half of which most likely comprises serum contaminants. Surfactant proteins A, B, C, and D (SP-A, SP-B, SP-C, SP-D) are active in surfactant’s surface tension reduction, secretion, absorption, and immune function. SP-A works with other proteins and lipids to improve surface actions and regulate secretion and reuptake. It also works with host defense in the alveolus. Lipophilic SP-B and SP-C facilitate adsorption and spread of lipid across the alveolar surface. SP-D is known to be a ligand for pathogens.

Surfactant is made in alveolar type II cells. The endoplasmic reticulum and Golgi apparatus package the lipid and protein precursors. Lamellar bodies are formed, including more protein with increasing gestational age. Catecholamines, corticosteroids, and other hormones stimulate the type II cells to secrete lamellar bodies. These unravel to form tubular myelin. Tubular myelin then adsorbs as a lipid-protein monolayer on the alveolar surface, giving maximum surface support to the alveolus. The interaction between intact protein and phospholipid allows optimal surfactant functioning.

Surfactant is inactivated in the alveolar space without large changes in amounts of its components. The monolayer does break down as protein and lipid dissociate. Surfactant changes to a small aggregate form that minimally reduces surface tension. These aggregates are then absorbed by macrophages and type II cells, which recycle lipid and protein components. ²⁴

The meta-analysis of studies conducted before the routine application of continuous positive airway pressure (CPAP) demonstrated a decrease in the risk of air leak and neonatal mortality associated with prophylactic administration of surfactant. By prophylaxis, most agree that for babies younger than 27 to 29 weeks’ gestation, treatment should be given soon after birth (within 20 minutes), after initial stabilization.

Analyses of more recent studies challenge this view. When studies that allowed for routine stabilization on CPAP are evaluated by meta-analysis, they demonstrated a decrease in the risk of chronic lung disease or death in infants stabilized on CPAP. Recent large trials that reflect current practice (including greater utilization of maternal steroids and routine postdelivery stabilization on CPAP) do not support the differences seen in earlier studies and demonstrate less risk of chronic lung disease or death when using early stabilization on CPAP with selective surfactant administration to infants requiring intubation. ²⁵²⁶

The goal of therapy is to maintain minute volume by maintaining functional, open alveoli for gas exchange. When atelectasis occurs in infants with RDS, CO₂ cannot get out and O₂ cannot get in. To maintain alveolar volume and therefore gas exchange, positive end-expiratory pressure (PEEP) is essential. CPAP can be used to maintain alveolar volume during exhalation despite inadequate surfactant. It works if the pressure delivered to the alveoli prevents closing pressure (remember, P = 2T/R) from completely collapsing alveoli, but it should not be so great that it hinders adequate exhalation. Early institution of CPAP appears to prevent the need for intubation in a significant percentage of VLBW infants.

When alveolar collapse is too rapid or widespread, positive pressure ventilation is the best tool. Positive pressure opens the alveoli for inhalation. End-expiratory pressure maintains alveolar volume during exhalation. Positive pressure ventilation will be necessary until adequate surfactant reduces surface tension and a sufficient number of alveoli are recruited for adequate minute ventilation.

Should infants with RDS receive high-frequency ventilation? The primary pathology of RDS comes from an inability to maintain lung inflation and fluid leak into the alveolar space. Secondary pathology originates from positive pressure reexpansion of collapsed alveoli. A ventilation strategy that maintains lung volume and avoids large distending pressure seems ideal. That is the idea behind high-frequency ventilation for RDS. The lung is inflated, and lung volumes are maintained while gas exchange occurs, using tidal volumes less than dead space. High-frequency ventilator technology is improving, and its applicability as the first-line treatment for RDS continues to be evaluated. Consensus has not yet been reached, however. ²⁷²⁸²⁹

U.S. mortality rates associated with RDS and prematurity declined significantly with the introduction of exogenous surfactant. By 1994 the combination of congenital and chromosomal defects had become the leading cause of infant mortality, and RDS with prematurity fell, for the first time, to number two on the list. Although bronchopulmonary dysplasia (BPD) has not significantly decreased in frequency, the severity of this chronic lung disease has declined for most surviving premature infants with RDS.

The only pulmonary complication that appears to have increased with therapy is a small but detectable rise in pulmonary hemorrhage. Other nonpulmonary complications have not been significantly affected.

CPAP is continous positive airway pressure. CPAP is applied to an infant’s airway using a variety of devices to maintain positive pressure in the airway during spontaneous breathing. These devices include a head hood, face chamber, face mask, several types of nasal cannulae, nasopharyngeal tube, and ET tube. Nasal cannulae inserted into the nares are used most often. Not all CPAP devices are equal, and they have varying degrees of success. They have been associated with a number of problems, such as difficulty with gaining access to the baby and maintaining connection to the airway, increase in dead space, and increase in airway resistance.

As with many things in life, the right amount is beneficial and too much is detrimental. When the proper amount of positive pressure is used, CPAP will accomplish the following:

If too much CPAP is applied, however, it can cause overdistention of the alveoli; worsen ventilation–perfusion match; increase pulmonary vascular resistance; decrease compliance; and impede venous return to the right side of the heart, thereby decreasing cardiac output.

These include but are not limited to the following:

The oxygenation index (OI) is used to express the severity of the respiratory disease.

where MAP = mean airway pressure, FiO₂ = fractional concentration of oxygen in inspired gas, PaO₂ = partial pressure of oxygen in arterial blood, PIP = peak inspiratory pressure, PEEP = positive end-expiratory pressure, T_I = inspiration time, and T_E = expiration time.

Note that the MAP is influenced by all respirator controls except the FiO₂. However, without a uniform ventilation strategy, the oxygenation index cannot be universally applied as an expression of severity of respiratory disease. This statement is especially true in the NICU, where patients may be hyperventilated; in these patients the MAP, and thus the oxygenation index, is elevated regardless of the severity of disease.

Meconium-Stained Amniotic Fluid and Meconium Aspiration Syndrome

No. Those who have an initial heart rate greater than 100 bpm, good respiratory effort, and reasonable tone will not benefit from intubation and suctioning. In fact, some vigorous infants may be injured in the process of suctioning because they are so difficult to restrain. ³⁰³¹

Gross staining of the infant is a surface phenomenon proportional to the length of exposure and meconium concentration. With heavy meconium, staining of the umbilical cord begins in as little as 15 minutes, and with light meconium, it begins after 1 hour. Yellow staining of the newborn’s nails requires 4 to 6 hours. Yellow staining of the vernix caseosa takes between 12 and 14 hours.

Because between 10% and 20% of all deliveries have in utero passage of meconium, meconium staining alone is not a good marker for neonatal asphyxia. For an infant to pass meconium, there does need to be a period of hypoxemia that initiates increased bowel contractility before birth. Simply having hypoxemia, however, is not the same thing as having perinatal asphyxia.

MAS is associated with the majority of cases of PPHN. Other associated disorders include RDS, sepsis or pneumonia, idiopathic PPHN, and lung hypoplasia (including congenital diaphragmatic hernia). In all instances the pulmonary artery pressure remains near systemic levels and results in right-to-left shunting of blood.

Causes of aspirated meconium are as follows:

Infants with MAS make up between 30% and 40% of infants who are treated with ECMO. Unfortunately, the circumstances that lead to MAS in many cases are precipitous and unavoidable. As a result, by the time therapy can be started, the pathophysiology is sufficiently far advanced and can be halted only by the use of ECMO. Other disorders that are managed with ECMO include sepsis, pneumonia, pulmonary hypoplasia (most often caused by congenital diaphragmatic hernia), and RDS. Patients with MAS tend to have the shortest ECMO courses and the highest survival rates, approaching 97% in the most experienced ECMO centers. However, the use of ECMO in recent years has declined significantly with the introduction of inhaled nitric oxide and improved ventilatory management.

Farmers and veterinarians grab newborn animals by their hindquarters and swing them in a circular motion. Centrifugal forces move MSAF outward into the upper airway and oropharynx. Caretakers then manually remove the material. Of course, it is not recommended that infants be swung by the legs to remove meconium from the airway.

No. The thicker the consistency of MSAF, the greater the likelihood of MAS or other respiratory distress. There is at least a sevenfold increase in the incidence of respiratory disorders among infants born through “pea-soup” MSAF compared with those born through watery-consistency MSAF.

Meconium-induced lung injury is associated with many pulmonary changes that contribute to respiratory failure. These include airway obstruction, inflammation with release of vasoactive substances, and surfactant dysfunction. Meconium has the ability to inactivate surfactant both in vivo and in vitro and has direct effects on type II pneumocyte function. In both animal models and human infants who have aspirated meconium and who are undergoing pulmonary fluid analysis, inflammatory cell numbers and total protein are significantly elevated compared with infants in the control group. Various inflammatory mediators, including myeloperoxidase and interleukin-8, are increased. Maximal influx of inflammatory cells occurs by 16 hours of age with some recovery by 72 hours. These findings support the role of surfactant replacement in infants with MAS that requires ventilatory support. ³²³³

Persistent Pulmonary Hypertension of the Newborn

Successful transition from intrauterine to extrauterine life requires that the pulmonary vascular resistance decreases precipitously at birth. In infants with PPHN, this decrease does not occur. Pulmonary arterial pressure remains elevated, and blood continues to shunt right to left across the ductus arteriosus and foramen ovale, resulting in significant hypoxemia.

In 1969 Gersony and coworkers described a group of term infants without structural heart disease who became cyanotic shortly after birth and who had only mild respiratory distress. These infants all had suprasystemic pulmonary arterial pressures with right-to-left shunting across persistent fetal pathways (ductus arteriosus and foramen ovale). Hence this condition was called persistent fetal circulation.

The shunting across the foramen ovale and ductus arteriosus as a result of suprasystemic pulmonary arterial pressure seen in PPHN is very similar to fetal circulation. However, the exclusion of placental circulation and the fact that ductus venosus may or may not be patent preclude the use of the term persistent fetal circulation to describe this condition. The term persistent pulmonary hypertension of the newborn describes the pathophysiology of the disease more accurately, indicating that the critical problem in this situation is the failure of the pulmonary circulation to decrease to normal pressures. ^34353637

Infants with PPHN are usually delivered at term or post term. Often they are born through MSAF. The typical clinical manifestations of a neonate with PPHN are as follows:

The common causes of PPHN are summarized in the mnemonic DIAPHRAGMATIC:

The causes of PPHN can also be classified according to the predominant abnormality involved (see Fig. 18-2).

In some infants the left subclavian artery arises from the arch of the aorta just distal to the level of the insertion of the ductus arteriosus. In these infants a pulse oximetry probe applied to the left hand indicates postductal saturations. Therefore it is always better to obtain preductal oxygen saturation from the right upper limb, a site that indicates preductal saturation.

Persistent elevation of pulmonary arterial pressure in PPHN results from active constriction of pulmonary vessels (as in pneumonia), underdevelopment of the pulmonary vessels (as in congenital diaphragmatic hernia), or maldevelopment of the pulmonary vasculature (as in prenatal ductal closure caused by maternal ingestion of nonsteroidal antiinflammatory drugs and idiopathic PPHN).

Vascular remodeling: In infants dying as a result of PPHN caused by maldevelopment of the pulmonary vessels, pulmonary arterial smooth muscle hypertrophies and extends from pre-acinar arteries into normally nonmuscular intra-acinar arteries, even to the level of the alveolus. This thickened muscle encroaches on the vessel lumen and results in mechanical obstruction to blood flow.

Functional abnormalities in the pulmonary vessels (e.g., reduced nitric oxide synthase, reduced soluble guanylyl cyclase, and increased levels of vasoconstrictors such as endothelin) have been described.

Persistently elevated pulmonary vascular resistance increases right ventricular afterload and oxygen demand and impairs oxygen delivery to cardiac muscle. Ischemic damage to the myocardium, papillary muscle necrosis, and tricuspid regurgitation can occur. Increased right ventricular pressure displaces the septum into the left ventricle, impairs left ventricular filling, and decreases cardiac output. Myocardial dysfunction is an important cause for mortality in PPHN ( Fig. 18-9).

It is often very difficult clinically to differentiate between these two conditions. Patients with PPHN are more labile and exhibit wide swings in oxygen saturations. A significant difference between the preductal and postductal oxygen saturations is also a clinical finding in favor of PPHN. An additional test that is sometimes used in this clinical situation is the hyperoxia test. The child to be tested is placed on an inspired oxygen level of 100%. On an arterial blood gas determination, if the arterial PO₂ level rises above 100 mm Hg, it is unlikely that the infant has significant cyanotic heart disease and more likely has pulmonary hypertension or pulmonary parenchymal disease. This test, however, is not infallible, and some children with PPHN may not be able to increase their arterial PO₂ above 100 mm Hg. In addition, it may be necessary to give positive pressure ventilation to a baby to be sure that one is ventilating the lungs of a child with pulmonary disease adequately to maximize the arterial oxygen levels. The best way to differentiate between these two entities is by echocardiography.

In the past the mortality for infants with PPHN ranged from 20% to 40%, and the incidence of neurologic handicap ranged from 12% to 25%. With recent advances in conservative management, survival and neurodevelopmental outcome have improved considerably. In most centers inhaled nitric oxide and ECMO have further reduced mortality associated with severe PPHN. Survival rates between 76% and 93% have been reported for infants with pneumonia, meconium aspiration, and idiopathic PPHN who require ECMO. The outlook for infants with diaphragmatic hernia requiring ECMO has not been as dramatic, and survival is still only about 60% to 80%.

Most infants treated for PPHN have few respiratory symptoms or neurologic or developmental sequelae by 1 year of age. However, the following also must be considered:

Although ECMO has considerably reduced the mortality rates associated with PPHN, it is an invasive procedure limited to a few tertiary care centers. Inhaled nitric oxide has reduced the use of ECMO. Unfortunately, this reduction has not been associated with an improvement in long-term outcome.

Inhaled Nitric Oxide

Nitric oxide is an important regulator of vascular muscle tone at the cellular level. Nitric oxide is generated enzymatically by nitric oxide synthases from L-arginine. Nitric oxide activates guanylyl cyclase by binding to its heme component, leading to the production of cyclic guanosine monophosphate (GMP). The mechanism by which cyclic GMP relaxes vascular smooth muscle is not clear. It appears to involve inhibition of activation-induced elevation in cytosolic calcium concentration.

Several randomized control trials indicate that inhaled nitric oxide reduces the incidence of the combined end point of death or need for ECMO compared with patients not offered treatment with inhaled nitric oxide. This reduction seems to be entirely due to a reduction in the use of ECMO insofar as mortality is not reduced. ⁴⁰

Nitric oxide has a high affinity for the iron of all heme proteins, including reduced hemoglobin, with which it forms nitrosyl hemoglobin (NOHb). The NOHb is then oxidized to methemoglobin with the production of nitrate. As a result, when given by inhalation, nitric oxide is inactivated before acting on any systemic vascular bed, while relaxing the pulmonary vascular smooth muscle through the cyclic GMP production. In normal development endogenous nitric oxide produced in endothelial cells from oxygen and L-arginine diffuses into smooth cells in the vascular wall and causes vasodilation. Nitric oxide that diffuses into the blood vessel lumen is avidly bound by hemoglobin and does not cause systemic vasodilatation.

Meta-analysis showed that infants with diaphragmatic hernia do not appear to share the benefits of inhaled nitric oxide that infants with other causes of hypoxemic respiratory failure experience. Indeed, there are suggestions that outcomes may be worse in infants with congenital diaphragmatic hernia who received inhaled nitric oxide compared with control subjects. This analysis showed that the incidence of death or requiring ECMO was 40 of 46 among control patients and 36 of 38 among patients treated with nitric oxide (relative risk, 1.09; 95% confidence interval [CI], 0.95 to 1.26). Mortality rates were similar in control and treatment patients (18 of 46 in the control group compared with 18 of 38 in the treatment group; relative risk of death, 1.20; 95% CI, 0.74 to 1.96), but there was a significant increase in the requirement for ECMO in infants treated with inhaled nitric oxide (31 of 46 in the control group compared with 32 of 38 in the treatment group; relative risk, 1.27; 95% CI, 1.00 to 1.62). ⁴¹

When nitric oxide and O₂ come into contact, peroxynitrite (ONOO⁻), a potent oxidant, is formed. The relative amount of nitric oxide, O₂⁻,ONOO⁻, and antioxidants in the airway will determine whether nitric oxide will be beneficial or potentially toxic. These oxidants can contribute to lung injury by enhancing lung inflammation, producing pulmonary edema, and reducing surfactant function. Furthermore, recent findings have shown that abrupt withdrawal of inhaled nitric oxide, even in infants with minimal or no response, can induce worsening pulmonary hypertension. The potential for pulmonary inflammatory injury can be decreased as the concentrations of inhaled nitric oxide and O₂ are lowered. Most late preterm and term infants can be weaned off inhaled nitric oxide within 4 days. ⁴²

An individual patient meta-analysis indicated that routine use of inhaled nitric oxide for treatment of respiratory failure in preterm infants cannot be recommended. Further research is necessary to determine the optimal starting dose and duration of therapy.

One population that may be an exception is preterm infants born after prolonged rupture of membranes. Multiple small studies suggest that preterm infants with severe respiratory failure born after premature and prolonged rupture of membranes may benefit from inhaled nitric oxide ⁴³⁴⁴

Major Anomalies that Alter Pulmonary Function

Infants with Pena–Shokeir phenotype (also termed arthrogryposis multiplex congenita with pulmonary hypoplasia) have gracile ribs and reduced thoracic volume. Also present are a lack of fetal breathing activity; polyhydramnios resulting from a lack of fetal swallowing; and intrauterine constraint, resulting in muscular hypoplasia involving both intercostal and diaphragmatic musculature. Thoracic wall weakness, hypotonia of the muscles of respiration, and anterior horn cell atrophy or deficiency lead to reduced ventilatory drive, which may improve over time for some infants.

The major causes of extrinsic fetal airway obstruction are cervical lymphangioma, teratoma, and vascular rings (e.g., double aortic arch, pulmonary vascular sling).

As fetuses with fetal airway obstruction reach viability, they should be monitored closely for development or progression of hydrops (for intrinsic obstruction cases) or polyhydramnios (when extrinsic obstruction is present). The fetus should be delivered by using the ex utero intrapartum treatment procedure, with maintenance of uteroplacental circulation and gas exchange. This approach provides time to perform procedures such as direct laryngoscopy, bronchoscopy, or tracheostomy to secure the fetal airway, thereby converting an emergent airway crisis into a controlled situation.

Mechanical Ventilation of the Neonate

The chief disadvantage is the fact that tidal volume is not directly controlled. The delivered tidal volume is determined by the interaction between PIP and lung compliance. Consequently, as compliance changes, so will the delivered tidal volume. Improving lung compliance can lead to excessive tidal volume and can cause lung injury. Conversely, worsening compliance can lead to hypoventilation and loss of lung volume. In addition, if an infant is breathing asynchronously with the ventilator, peak pressures are reached quickly, and volume is reduced. This situation may result in a serious deterioration of blood gases.

Uncuffed ET tubes that are used in newborn infants result in a variable degree of air leakage around the tube, causing variable loss of tidal volume. Additional tidal volume is lost through gas compression within the relatively large volume of gas in the ventilator circuit and humidifier and to stretching of the relatively compliant circuit during inspiration. As a result, the tiny premature infant with poorly compliant lungs receives only a small and variable fraction of the tidal volume generated by the ventilator. In essence, the circuit is ventilated rather than the baby.

There is an upper limit to the effective respiratory rate. An excessively rapid IMV rate may lead to inadequate expiratory time with incomplete exhalation and air trapping. Thus, paradoxically, when the IMV is greater than 90 to 120 breaths per minute, further increases in rate may lead to CO₂ retention. This situation is most likely to occur in infants with increased airway resistance and prolonged time constants. In such infants the best way to improve ventilation is to decrease the IMV rate.

Tidal volume is proportional to the difference between PIP and PEEP. This is referred to as 蜐P. Thus lowering PEEP will increase 蜐P and improve ventilation (although it can lead to loss of lung volume with deterioration of oxygenation). Occasionally, excessively high PEEP in a patient with relatively compliant lungs can lead to incomplete exhalation and CO₂ retention. This is not a common problem but should be considered in a patient with improving oxygenation and a worsening respiratory acidosis.

Paw has been shown to be a major determinant of oxygenation. Adequate distending pressure is needed to maintain lung volume and prevent the diffuse microatelectasis that leads to ventilation–perfusion imbalance with consequent hypoxemia. ⁴⁵

Paw is the area under the pressure curve ( Fig. 18-10). Increasing the PEEP is usually the most effective means of increasing the Paw. The least recognized factor affecting the area under the curve is the slope of the upstroke of pressure, which determines the shape of the pressure waveform. Higher flow leads to more rapid upstroke and a more square-shaped curve, which has a larger area than one with a gradual upstroke and a more triangular shape.

For any given PIP the delivered tidal volume will be determined by the compliance of the baby’s lungs. Select a pressure based on the best estimate of what the infant will need, and observe the result. If adequate chest rise, good breath sounds, and oxygenation are apparent, the pressure is appropriate. If the chest rise is excessive, reduce the PIP, and if the chest movement is inadequate, higher PIP is needed (assuming the ET tube is correctly positioned).

Most modern infant ventilators now have the means to directly measure tidal volume (V_T), eliminating the dependence on subjective assessment of adequacy of chest wall movement and allowing more accurate determination of optimal PIP. The target V_T measured at the airway opening should be 4 to 6 mL/kg in the acute phase of the disease.

Note: Some devices measure V_T at the point where the circuit attaches to the ventilator. This position is undesirable because it will give an artificially large V_T measurement, ignoring the loss of V_T to compression of gas in the circuit and circuit stretching. Furthermore, gadgets do malfunction, so continue to use your eyes and ears to verify that the “numbers” are believable. ⁴⁶⁴⁷

Most of these should be readily recognizable clinically. If the chest is not moving, the first priority is to make sure that the airway is patent, the ET tube is in place, and the ventilator is cycling. Many modern infant ventilators have the ability to display flow and pressure waveforms, which should help diagnose or confirm the problem. When in doubt, the clinician should reintubate. Manual ventilation may be appropriate if a circuit or ventilator problem is suspected, but be careful not to use excessive pressure, which may cause lung injury.

Some degree of impairment of venous return to the heart is inevitable because, unlike spontaneous breathing, intrathoracic pressure rises above ambient pressure during positive pressure ventilation. The problem becomes more severe when high or excessive pressures are used. Intraventricular hemorrhage can be triggered by hemodynamic instability, elevated venous pressure, and sudden increases in cerebral blood flow (as might occur with retention of CO₂). Periventricular leukomalacia is associated with hypotension and with marked respiratory alkalosis.

(A) is correct. Hypercarbia, hemodynamic impairment, and air leak caused by incomplete exhalation occur when the expiratory time is too short to allow complete exhalation before the next mechanical breath occurs. This situation is most likely to occur in infants who have increased airway resistance, such as is seen in meconium aspiration with acute airway obstruction or in chronic lung disease in which airway edema, copious secretions, and bronchospasm are present.

A time constant is the product of lung compliance and airway resistance (Tc = R × C). Conceptually, time constants reflect the time it takes for gas flow to cease and pressure to be fully equilibrated between the large airways and the alveoli when a sudden pressure change is applied to the airway opening (three time constants are needed for 95% equilibration) ( Fig. 18-11).

In acute RDS, compliance is low and airway resistance is also low (normal). Therefore short inspiratory times can be used. In addition, time constants are also a function of size (total compliance, not compliance per kilogram, is used). Consequently, large subjects such as adults or horses have long time constants, and small premature infants and hummingbirds have short time constants. Time constants are a major determinant of resting respiratory rate, which turns out to fall exactly where work of breathing is lowest. This is why adults at rest breathe at a rate of 14 breaths per minute, term infants breathe at 40 breaths per minute, and small premature infants breathe at about 60 breaths per minute. Mice and hummingbirds breathe significantly more quickly. In infants with acute respiratory distress, tachypnea is a reflection of shorter time constants as lung compliance decreases because of various causes. Asthmatics, on the other hand, prefer to breathe rather slowly because of their prolonged expiratory phase. The bottom line is this: Consider the underlying disease process and its pathophysiology before making decisions about ventilator settings.

A/C ventilation is a form of mechanical ventilation in which the infant triggers the ventilator to cycle with each breath ( Fig. 18-12). With a small triggering effort, therefore, the baby can achieve a much higher level of ventilatory support than with spontaneous breathing. In general, A/C ventilation can be used very successfully to treat VLBW babies with RDS or pulmonary insufficiency of prematurity. It has become the most common way to initiate mechanical ventilation therapy in these clinical situations. It often enables patients to be ventilated at lower PIP levels than with conventional mechanical ventilation or SIMV. It differs from SIMV in that, with A/C, the baby will trigger a ventilator breath with each respiratory effort. In SIMV the ventilator is synchronized to the baby’s respiratory cycle so as to avoid stacking of the ventilator and infant breaths, but the baby is given only a preset amount of synchronized breaths. With modern ventilators, if the baby becomes apneic during either A/C ventilation or SIMV, the machine will deliver a preset number of breaths per minute. ⁴⁸

C is correct. In A/C mode, every breath that the infant takes triggers a ventilator breath—that is, every breath is supported. As a result, the baby is in control of the ventilatory rate. The IMV rate is only a back-up rate in case the infant is apneic or the triggering mechanism is not functioning. Decreasing the IMV rate does not actually decrease the level of support the infant is receiving. Weaning occurs by lowering the degree of support for each breath (i.e., the PIP). Ultimately, when the PIP is down to where the ventilator is generating only enough pressure to overcome the resistance of the ET tube and circuit, the baby is ready for extubation (usually about 10 cm H₂O).

With manual ventilation, much higher PIP levels are used than anyone would dare set on the ventilator. In a crisis it is frighteningly easy to inadvertently generate pressures above 40 cm H₂O. The clinician must beware of the risk of pneumothorax. Using a manometer may be helpful, but most of the mechanical gauges grossly underestimate the actual PIP and the actual duration of inspiration, especially when the ventilatory rate is rapid. This explains, in part, why it is that when you place the baby back on the ventilator, ostensibly on the same settings as the pressures that were used with hand ventilation, the saturation usually drifts down again (because the ventilator PIP is actually lower than that with which you were bagging). It is sometimes preferable to maintain the infant on the ventilator and simply increase the level of support (PIP and IMV) as needed to achieve the desired result. This approach allows you to continue to use the monitoring function of the ventilator to provide feedback regarding the tidal volume and other parameters, and it provides controlled and accurate pressure delivery. However, if the baby is still doing poorly, hand ventilation is an acceptable alternative.

When a baby is doing poorly on a ventilator, the clinician should remove the baby from the machine and hand ventilate with an anesthesia (preferably) or self-inflating bag. The chest excursion should be carefully examined and breath sounds auscultated to ensure that the ET tube is still well positioned and not plugged. If there is any question about the tube, it should be replaced promptly. A chest radiograph is often helpful to ensure proper positioning of the tube and to confirm that no air leak is present. Translumination of the chest can also be helpful in detecting pneumothoraces. If the tube seems fine and there are no radiologic changes, the ventilator itself must be carefully checked for malfunction. Respiratory therapists should be available around the clock in any intensive care nursery in which infants are ventilated.

Although there is a great deal of literature on neonatal intubation, few articles describe the risks of extubation. Nothing is more frustrating than successfully completing a course of neonatal mechanical ventilation on a sick baby only to have a serious setback because of a poor effort at extubation. When a child has reached the predetermined levels for extubation, the following should be done:

When the child is ready to be extubated, the tube should be carefully untaped from the face to prevent any abrasions. An anesthesia bag should be attached to the ET tube, and a long, slow, low-level (15 cm H₂O), positive pressure breath should be administered as the ET tube is withdrawn from the airway. This breath overcomes the natural negative pressure created as the tube is withdrawn from the airway. The child should be given CPAP or oxygen and observed closely. Stridor or hoarseness is common and typically indicates upper airway edema. Marked retractions also may be seen and are worrisome, indicating either volume loss in the lung or upper airway obstruction. Adequate humidification of inspired gas is essential after extubation. Because of the initial inability to oppose the vocal cords, feeding should not be resumed for at least 6 to 12 hours after extubation. Clinical deterioration that occurs 24 to 48 hours after extubation may be caused by a number of factors, including increased atelectasis, upper airway edema and obstruction, and muscular fatigue. If reintubation is deemed necessary, it should be carried out promptly.

Neonatal High-Frequency Ventilation

Neonatal high-frequency ventilation uses devices that provide respiratory support for critically ill neonates with the use of small tidal volume, rapid rate assisted ventilation. Generally, this means rates above 150 breaths per minute and tidal volumes below 2 to 3 mL/kg.

Oscillation exchanges gas by producing positive and negative flow in the ventilator circuit through the use of a vibrating diaphragm. Jet ventilation delivers high-frequency breaths through the interruption of a continuous gas flow directly into the airway through a unique ET tube located in the airway. The interruption takes place in a patient box located close to the baby, by a pinch valve that opens and closes on a piece of plastic tubing. With jet ventilation, inspiration is active; exhalation is passive. High-frequency flow interruption generates the signal by interrupting the flow of gas. It is similar to the jet ventilator except that the interruption of the gas flow occurs at a site much farther from the infant.

No. Because there have been no comparison trials, each type has its advocates and critics.

It decreases. Impedance of the airway and ET tube are frequency dependent. As rate is increased, impedance to transmission of pressure swings increases. Thus tidal volume decreases as frequency is increased.

With standard mechanical ventilation or spontaneous breathing, minute ventilation = frequency × tidal volume. In high-frequency ventilation, minute ventilation = (frequency) × (tidal volume)²

This question emphasizes the importance of understanding the differences between high-frequency oscillation and conventional ventilation. In conventional ventilation increasing the rate will increase carbon dioxide elimination in most cases. With high-frequency ventilation turning up the rate generally causes a decrease in minute ventilation owing to the loss of tidal volume delivery. When ventilation is inadequate during high-frequency ventilation, turning the rate down can increase carbon dioxide elimination.

No one really knows. Modeling of the wave flow in high-frequency ventilation is exceedingly complex. Several theories have been proposed, however, to explain high-frequency ventilation:

Just as in conventional ventilation, changes in respiratory system impedance affect carbon dioxide elimination during high-frequency ventilation. The important distinction is that high-frequency

ventilation is more sensitive to changes in impedance than conventional modes of ventilation. Changes in ET tube size, respiratory system compliance, airway patency, and mucus plugging all can have a profound effect on tidal volume delivery and therefore ventilation. Because of the frequencies used and the small tidal volumes, these changes seem to be significantly magnified with high-frequency ventilation compared with conventional ventilation.

The strategy with which high-frequency oscillation is used is important. Patients with diffuse loss of lung volume (i.e., atelectasis) should be treated with a lung recruitment strategy. High-frequency oscillation allows the use of higher Paws than conventional ventilation because the small tidal volumes promote ventilation without causing lung overinflation. This approach has been studied in animal models of hyaline membrane disease and has been shown to improve lung inflation, decrease acute lung injury, decrease pulmonary air leaks, and promote survival. Often referred to as a “high mean airway pressure strategy,” the real goal is not a high Paw but rather optimal lung inflation. Unfortunately, measures of optimal lung inflation are not available. Clinically, the goal is to promote lung recruitment while avoiding lung overinflation, cardiac compromise, and lung atelectasis. ⁴⁹⁵⁰

The chest radiograph and the arterial PO₂/FiO₂ ratio can be used to help guide therapy. If the chest radiograph shows more than nine posterior ribs of inflation, flattened diaphragms, a small heart, or very clear lung fields, the lung may be overinflated. Similarly, if the Paw is high and the FiO₂ is low, then Paw should be decreased before FiO₂. If the chest radiograph shows fewer than seven posterior ribs of inflation, domed diaphragms, a normal heart size, or diffuse radiopacification, the lung may be underinflated. Therefore the Paw should be increased if the Paw is low and the FiO₂ is high. The assessment of cardiac function is also important for the safe use of high-frequency ventilation. Monitoring heart rate, blood pressure, urine output, and capillary refill can help alert the care provider to changes in cardiac output.

Several studies have shown evidence of increased brain injury (i.e., periventricular leukomalacia and intraventricular hemorrhage) associated with high-frequency ventilation, particularly when initiated as an initial treatment modality in a VLBW baby. Although meta-analysis does not confirm this finding, the concern remains, and further studies are needed in this regard. The complication of necrotizing tracheobronchitis was reported with early models of high-frequency ventilation. This complication has disappeared with the development of improved humidification systems.

Altering Paw to optimal levels will change lung volume, improve ventilation–perfusion matching, and decrease intrapulmonary shunt. FiO₂ is used to change the alveolar oxygen concentration.

In oscillatory ventilation Paw can be altered directly by changing that setting on the ventilator. With jet ventilation Paw is a measured value that is a combination of several factors: PIP, PEEP, duration of inspiratory phase (jet valve on time), and background sigh rate.

No. Only anecdotal evidence exists to support the efficacy of high-frequency ventilation in neonates with MAS. In fact, in neonates with MAS and signs of air trapping, high-frequency ventilation may be dangerous. Reported success in MAS with high-frequency ventilation is between 30% and 40%. ⁵¹

Volutrauma occurs most rapidly when the lung is repeatedly cycled from a low volume to a high volume. Use of zero end-expiratory pressure and excessive tidal volumes can create acute lung injury within minutes. Application of end-expiratory pressure reduces “atelectotrauma” by preserving functional residual capacity at the end of each assisted breath. Lung overinflation is avoided by using small tidal volumes. Thus the extremes of low and high lung volumes are avoided with high-frequency ventilation.

The use of end-expiratory pressure, surfactant, prone positioning, and liquid ventilation all promote lung recruitment over time. They work by stabilizing recruited alveoli at the end of exhalation.

Hyperventilation must be avoided. Data on brain injury in neonates suggest that hyperventilation may cause brain injury through ischemia as CO₂ is lowered. This finding has been observed in a number of published studies, both with conventional and high-frequency ventilation.

Lung overinflation or underinflation also may have adverse affects on the baby. Currently, no good methods are available for defining optimal lung volume during high-frequency ventilation. Evaluating cardiac performance, chest radiographs, and arterial PO₂/FiO₂ ratio can help the clinician avoid extremes, but the Holy Grail of high-frequency ventilation is defining when the lung is optimally inflated.

The most dramatic improvements in oxygenation have been reported in patients with poor lung inflation. In general, this means neonates with RDS or pneumonia. Lung disease in which there is a significant amount of airway debris or resistance does not seem to respond as well to high-frequency ventilation.

Neonatal Extracorporeal Membrane Oxygenation

ECMO is a modification of standard cardiopulmonary bypass techniques used in the operating room during open heart surgery. It was adapted in a simplified circuit to provide artificial life support to pulmonary patients in an intensive care unit setting. Neonatal ECMO was the first clinically successful application of this technology to treat severe and progressive cardiorespiratory failure caused by MAS and complicated by persistent pulmonary artery hypertension occurring in the first week of life. At the core of ECMO technology are the heart–lung pump (a semiocclusive roller device) and the innovative Kolobow polycarbonate-spooled, silicone membrane oxygenator ( Fig. 18-14). Both devices are sufficiently powerful to completely support cardiac output and lung function in neonates.

ECLS includes ECMO, hemofiltration, hemodialysis, and indwelling oxygenator filaments (i.e., intravenous oxygenator). Many of these other techniques can be incorporated with an ECMO circuit or can be applied separately. ⁵²⁵³⁵⁴

The definitive randomized trial establishing the effectiveness of neonatal ECMO was conducted by the National Health Service in the United Kingdom. Of 93 infants referred to ECMO centers, 30 (or 32%) died compared with 54 of 92 (59%) receiving conventional care. The relative risk for reduced mortality with ECMO was 0.55 (95% CI, 0.39 to 0.77; P <0.0005). Of the survivors, one child in each group was severely disabled at 1 year, and ten ECMO patients (compared with six conventionally treated infants) were disabled to a lesser degree. The UK Collaborative ECMO Trial Group concluded that ECMO support should be actively considered for mature neonates with severe but potentially reversible respiratory failure. ⁵⁵⁵⁶

The success of ECMO relies on the physician’s ability to recognize, within the first week of illness, those near-term or term newborn infants with reversible pulmonary disease and to exclude infants with irreversible pulmonary disease. According to the ECLS Organization’s Registry data estimates, only one in approximately 1700 infants can benefit from ECMO. Criteria for ECMO patient selection have been widely debated during the past decade, and two controversial questions have arisen: (1) Is less invasive therapy likely to succeed? (2) With constantly improving neonatal ventilatory and pharmacologic techniques, must physicians continually reassess ECMO criteria? In general, the earlier the ECMO physician can identify the infant with a high probability of dying as a result of disease (before iatrogenic consequences of conventional therapy), the better the patient selection and outcome will be. The following inclusion and exclusion criteria provide general neonatal ECMO guidelines that are currently widely accepted:

Once the aforementioned inclusion and exclusion criteria have been considered, one of several pulmonary indices is used to assess the severity of respiratory illness and the likelihood of death if the infant is treated conventionally. The simplest and most popular index is the oxygenation index ( Fig. 18-15). Briefly, the oxygenation index is equivalent to the Paw generated during mechanical ventilation multiplied by the FiO₂ (both of these values indicate the level of conventional ventilatory support) divided by the postductal arterial oxygen tension in the blood (a sensitive indicator of both ventilation and perfusion of the baby’s lung). The resulting value is multiplied by 100. The relative importance of the ratio between Paw and arterial oxygen tension in the calculation of oxygenation index performed at 1.00 (FiO₂) is further demonstrated graphically in Figure 18-15. Once the arterial PO₂ is below 40 mm Hg in the denominator of the oxygenation index equation, a geometric rise in the oxygenation index occurs. This rise parallels increased pulmonary vascular resistance with increased right-to-left shunting in the patient with severe pulmonary arterial hypertension.

VA Bypass: The gold standard for ECMO therapy is VA bypass. An internal jugular drainage cannula and a second common carotid arterial infusion cannula are placed surgically through a right neck incision performed at the bedside. VA ECMO provides complete cardiopulmonary support to an infant’s native heart and lungs when either or both are failing.

Advantages:

Disadvantages:

VV Bypass: A less invasive technique for augmenting systemic oxygenation using ECMO is VV bypass. In neonates a novel double-lumen cannula (12 or 14 French) is surgically inserted into the internal jugular vein and positioned within the right atrium. Blood is withdrawn from the lateral lumen, reoxygenated, and infused back into the medial lumen. The right atrial admixture of oxygenated and deoxygenated blood then crosses through fetal channels (the foramen ovale and the ductus arteriosus) in the infant with severe pulmonary arterial hypertension to supply systemic oxygenation via shunt flow. Because systemic blood supply is delivered entirely by the infant’s native left ventricle, sufficient ventricular force must be available to circulate this oxygenated admixture against systemic vascular resistance, which is usually increased in critically ill patients. Frequently, both cardiotonic pressors and generous volume infusions of saline, albumin, or plasma along with blood transfusions are required to maintain an infant’s circulation on VV ECMO. VV access does not require invasion of the carotid artery; therefore systemic embolism is less risky, and the right common carotid artery is left intact after decannulation from bypass.

Advantages:

Disadvantages:

SvO₂ from the jugular venous cannula drain is monitored continuously during bypass using a fiberoptic device inserted directly into the blood path coming out of the patient. SvO₂ does not so much reflect pulmonary function (as does the systemic arterial saturation) as it represents the adequacy of tissue oxygen delivery from the native heart and the ECMO circuit combined. If the oxygen delivered by ECMO is sufficient to meet tissue oxygen demand, then the SvO₂ is generally greater than 70%. Failure to meet tissue oxygen demand results in the progressive desaturation of venous blood returning from the capillary beds into the right atrium. An SvO₂ below 65% to 70% indicates marginal oxygen delivery, and an SvO₂ below 60% may be associated with lactic acid production through anaerobic metabolism. Therefore the single most important parameter monitored during ECMO and used to assess the adequacy of bypass is the SvO₂. Notably, during VV ECMO the SvO₂ may be artificially elevated because of recirculation of arterialized blood back into the drainage side of the double-lumen cannula; however, trends in SvO₂ may still be useful, and the patient may be taken off bypass briefly to assess a true SvO₂.

The typical ECMO course transpires over 3 to 7 days, while awaiting spontaneous lung recovery.

Cardiac recovery and mobilization of capillary leak edema usually precede lung recovery and weaning the ECMO pump flow rate. As the tissue edema is mobilized, fluid is transferred back into the intravascular space, increasing the baby’s native cardiac output. Therefore the infant’s systemic arterial saturation and arterial PO₂ may actually decrease during recovery (as ECMO support is weaned and the infant’s native cardiac output drives right-to-left shunting of deoxygenated blood through fetal channels in an accelerated fashion). During this early improvement phase on ECMO, diuretic therapy (e.g., furosemide, mannitol) or hemofiltration may assist in reducing this native circulation of desaturated blood. Thereafter, as the mixed venous saturation improves in the jugular venous cannula (above 80%), the ECMO pump flow is reduced in 10 mL/min decrements until a pump idle rate is reached of approximately 100-mL/min minimum flow (to prevent stasis and clotting within the circuit). Frequent arterial and venous blood gas assessments are important during the weaning process. Recent reports have suggested that pulmonary function testing demonstrating increased functional residual capacity (>15 mL/kg) and improved dynamic lung compliance may be useful in determining more exactly when lung recovery is sufficient to warrant coming off bypass.

Air Leak Syndromes

Infants with RDS and MAS are at highest risk for air-leak syndrome, and it is estimated that between 1% and 2% of all newborns may have a spontaneous pneumothorax. The incidence of air leak increases with decreasing birth weight and gestational age, and it increases with more severe lung disease. The reason for the increase with MAS is the viscosity of meconium, which results in a ball-valve mechanism that leads to air trapping. Newborns in general have a higher incidence of air leaks than the general population because of the high transpulmonary pressure (−30 to −150 cm H₂O) associated with the onset of breathing.

Pneumothorax is the most common form of air leak, and, fortunately, pneumopericardium is the least common. Pneumopericardium must be recognized promptly because of its high morbidity and mortality risk. In the era before surfactant, pulmonary interstitial emphysema was more common and in many cases preceded other forms of air leak. Pneumomediastinum is uncommon but the most difficult to treat because there is no easy way to evacuate mediastinal air.

One of the major factors has to be the “kinder, gentler” approach to neonatal ventilation. Permissive hypercapnia was a popular approach during the 1990s, and this led to more conservative ventilatory management strategies.

A second important change was the introduction of surfactant replacement therapy toward the end of the 1980s. Most of the early surfactant trials documented a 30% to 50% reduction in the rate of neonatal air leaks.

Your suspicion should be high for a tension pneumothorax in this clinical situation. Before you place a needle into the chest, however, consider the following:

The primary treatment goal unilateral pulmonary interstitial emphysema is to allow the affected lung to deflate. Selective bronchial intubation will allow the contralateral lung to deflate (of course, selective left mainstem intubation may be technically difficult), but it may pose problems because the perfusion to the ventilated lung may not be sufficient for gas exchange in all cases. A randomized trial of high-frequency jet ventilation did show effectiveness in treating pulmonary interstitial emphysema by lowering the Paw, which may allow the emphysema to resolve. In infants the lung in the superior position will receive more of the ventilation. Placing the affected lung in the downward position may be helpful in deflating that lung. ⁵⁷⁵⁸

There is no specific sign. In a tension pneumothorax an ongoing air leak contributes to a progressive increase in intrathoracic pressure. Shift of the trachea or the point of maximal impulse, decreased breath sounds, pallor, or cyanosis and retractions may occur in either tension or nontension pneumothorax. In a tension pneumothorax the critical factor is the ongoing increase in cardiopulmonary embarrassment to the patient. When a pneumothorax is first detected, it is usually very difficult to tell whether a pneumothorax is under tension. If the child appears clinically stable for the moment, the clinician can wait for a time (30 to 60 minutes) and repeat a chest radiograph before inserting a chest tube. In some cases, however, waiting is impossible, and a thoracentesis must be done immediately.

Neonates are subject to air leaks because of uneven alveolar ventilation in RDS or MAS. Air trapping also frequently occurs because of small airway plugs. The areas that are more distensible receive more ventilation, which leads to high transpulmonary pressure that in turn increases the likelihood of alveolar rupture. An additional factor is that the neonate has fewer alveolar connecting channels (pores of Kohn), which allow air to redistribute between ventilated and nonventilated alveolar spaces. Lastly, resuscitation by an overzealous, inexperienced practitioner also increases the newborn’s susceptibility to air leaks.

Ideally, a chest tube should be placed in that part of the thoracic cavity where it will do the most good with the least risk to the infant. Positioning of the infant is the key to the entire procedure. All too commonly, the baby is allowed to remain supine. When the clinician enters the chest in that position, the catheter hits the lung and moves posteriorly ( Fig. 18-16). However, if the child is placed nearly vertical to start the procedure, it is easy to angle the catheter anteriorly for optimal placement. The thoracostomy tube is inserted through an incision made in the fifth interspace in the midaxillary line. After the incision is made, the clinician tunnels up an interspace with a hemostat, which is used to pop through the strong muscular wall of the chest (a remarkably tough structure even in a tiny premature infant). If a pneumothorax is present, a gush of air should be seen when the chest is opened. The catheter should be advanced so that no end holes lie outside the chest wall. If the catheter is inserted too far, it must be pulled back. The chest tube is then connected to a suction apparatus. The suction rarely needs to be greater than −10 to −15 cm H₂O.

The use of anterior catheter insertions in the second interspace is not recommended, except in rare circumstances. It is too easy to hit the breast bud, which may damage future breast development or leave unsightly scars in any patient.

Obstructive uropathies lead to oligohydramnios. Insufficient amniotic fluid volume leads to pulmonary hypoplasia. The mechanism is not completely understood but in part is due to external compression of the neonate’s thorax, which impedes fetal lung growth. It is also believed that fetal breathing movements against an intrauterine fluid volume may be critical for normal lung development.

The air in a spontaneous or nontension pneumothorax will have the same nitrogen concentration as room air. By allowing the baby to breathe pure oxygen, a gradient for nitrogen is created from the extrapulmonary to the intrapulmonary spaces. Nitrogen will naturally diffuse across this gradient, allowing the pneumothorax to reabsorb more rapidly. Caution should be used when considering this approach in preterm infants, who are more subject to oxidant injury. Recent work suggests that supplemental oxygen use may not be associated with faster resolution of spontaneous pneumothorax in term infants. Term infants with tachypnea associated with a spontaneous pneumothorax who were placed in room air did not require supplemental oxygen and did not have longer recovery times compared with infants placed in more than 60% oxygen.

Bronchopulmonary Dysplasia

BPD is the chronic lung disease that often follows RDS in VLBW babies. First described by Northway in 1967, it has become the greatest foe of all neonatologists and the focal point of perhaps more studies than any other clinical syndrome in neonatology. BPD was not a disease until the neonate became a patient. Once people attempted to save critically ill neonates with lung disease, a certain percentage developed BPD. In most nurseries the BPD rate is between 20% and 30% in infants who weigh less than 1500 g. BPD is defined as a need for oxygen at either 28 days of life or, more recently, at 36 weeks’ postconception, with radiographic changes consistent with chronic lung disease. The incidence of BPD increases with decreasing gestational age at birth, decreasing birth weight, and increasing severity of lung disease at birth. In addition, testing (physiologic test) to determine if an infant is able to tolerate being in room air at 36 weeks’ postconceptional age influences the use of the term BPD. ⁵⁹⁶⁰

In a series of open lung biopsies from VLBW infants ages 14 days to 7 weeks who were receiving ventilatory support with radiographic changes consistent with chronic lung disease, Coalson and colleagues described a consistent lack of alveolarization with variably widened alveolar septae and minimal changes in the airways. Mild to moderate septal fibrosis was also apparent. These widened alveolar septae were hypercellular with disordered capillary growth. Typically, the alveolar spaces were laden with numerous alveolar macrophages and neutrophils.

Transmission electron microscopy demonstrated poor differentiation of type I and type II lung epithelial cells. These epithelial cells had relatively abundant cytoplasm and extensive glycogen stores; however, lamellar bodies were extremely rare to totally absent. There was no progression of alveolarization with enlarged simplified terminal air spaces or minimal and focal saccular fibroplasia. The interstitium of the lung contained myofibroblasts, and there was focal deposition of elastin and collagen fibers. Most saccular walls showed blunted “outpouchings” or secondary crest formation. ^61626364

The etiology of BPD is not clear, but several factors likely contribute to its development (the six Ps of BPD):

Other factors, such as free oxygen radical exposure and sepsis, also seem to be contributory in many instances. Sepsis, in particular, has recently become an increasingly important piece of the BPD puzzle. The key to this disease, however, appears to be the chronic exposure that babies have to the six Ps.

In extremely-low-birth-weight infants, BPD appears to be caused by a combination of nutritional failure and failure of alveolarization, resulting in both diminished somatic and lung growth. These factors lead to oxygen and ventilator dependency in a manner different from the original etiology of BPD.

Both animal and human studies indicate that chronic steroid use may result in reduced amounts of neural tissue mass. Neurologic handicap rates are higher in infants treated with dexamethasone. Somatic growth may also be adversely affected. ⁶⁵⁶⁶

Children continue to add new alveoli until approximately 8 years of age. After that time surface area and volume within the lung continue to increase with growth, but new alveoli are no longer added. Although scarring does occur in the lungs of patients with BPD, there appears to be sufficient healthy tissue to regenerate an adequate new alveolar volume.

The key to recovery from BPD is growth of alveoli and overall growth. As a result, optimal nutritional support is critical in BPD, perhaps more so than anything else. The following are other therapeutic adjuncts that help:

BPD spells are acute episodes of deterioration encountered during the course of treatment of a child with BPD. The baby typically becomes increasingly cyanotic, agitated, and inconsolable, with a marked deterioration in overall pulmonary status. Oxygen and ventilatory assistance often need to be increased during these episodes. At times they may be quite acute and severe, occasionally resulting in sudden death.

BPD spells frustrate even the best neonatologists with respect to management issues. Bronchospasm is often cited as the cause of this deterioration, but personal experience suggests that many such episodes, especially the more acute, severe forms, are more commonly the result of airway collapse caused by tracheobronchomalacia. Increasing the PEEP to stabilize an airway (assuming the child is intubated) can be beneficial in such cases. If the clinician does suspect bronchospasm, prebronchodilator and postbronchodilator therapy can be evaluated with pulmonary function testing. Flexible fiberoptic bronchoscopy is also valuable to detect granulomas or tracheobronchial malacia that might be causing airway obstruction.

Furosemide is a more potent diuretic than either chlorothiazide or spironolactone. In chronic situations such as BPD in a neonate, however, calcium sparing is important to prevent rickets, and thiazide diuretics are thought to be more effective in this regard. Spironolactone helps prevent potassium loss and reduces the severity of metabolic alkalosis resulting from diuretics. It is always a good idea, however, to initiate potassium chloride supplementation whenever diuretics are initiated for BPD because so many of these children develop a significant metabolic alkalosis. Furosemide also has a greater tendency to produce nephrocalcinosis when used on a chronic basis, which is less likely to occur with thiazide diuretics. There are no randomized control trials that demonstrate the efficacy and safety of diuretics in the management of BPD.

Apnea of Prematurity

Apnea is the cessation of breathing. Although this problem affects people of all ages in many different forms, it is most prevalent in premature infants younger than 36 weeks’ gestation. Pathologic apnea refers to cessation of breathing for more than 20 seconds; cessation of breathing for less than 20 seconds and accompanied by bradycardia 20% below the baseline heart rate; or cessation of breathing for less than 20 seconds with oxygen desaturation below 80%. Apnea in a newborn is classified as central, obstructive, or mixed. Most apnea of prematurity is classified as central apnea ( Fig. 18-17), in which there is complete absence of respiratory effort. Obstructive apnea occurs when an infant makes a respiratory effort but no airflow is present because of the presence of obstruction (see Figure 18-17, bottom). Obstructive apnea can be associated with gastroesophageal reflux. Mixed apnea is a combination of central and obstructive apnea.

Periodic breathing is a type of central apnea characterized by brief pauses in breathing of less than 10 seconds, followed by periods of regular respiration of less than 20 seconds’ duration. This pattern repeats itself for at least three cycles and often many more times ( Fig. 18-18). The significance of this form of breathing is unknown at present. Many prematurely born infants demonstrate periodic breathing for as much as 20% to 30% of total sleep time. Because of the frequency of this finding, some neonatologists consider periodic breathing to be a normal maturational process. On the other hand, it also may be a reflection of significant immaturity of respiratory control and a variant of apnea.

Virtually all premature infants have some degree of apnea. At 34 to 35 weeks’ postconceptional age, about 65% of infants have demonstrable apnea. About two thirds of these children have central apnea or periodic breathing, and one third have obstructive or mixed apnea.

In the short term, particularly in the NICU, extremely premature infants can have prolonged apneic episodes that may be fatal. As they mature, most infants will have more self-limited episodes. Less clear are the long-term effects on infants who have had a history of severe apnea of prematurity. Clinical studies suggest apnea of prematurity, particularly with oxygen desaturation, may affect learning and other aspects of childhood development. ⁶⁷⁶⁸

Although in most cases apnea of prematurity is gone by 37 weeks’ postconceptional age, in many cases it can persist even beyond 45 weeks’ postconceptional age (and occasionally longer). Recent evidence indicates that apnea persists longest in the most immature infants. ⁶⁹

Caffeine is the preferred treatment because of its once-a-day dosing and fewer side effects. Caffeine therapy for apnea of prematurity reduces the rates of CP and cognitive delay at 18 months of age. The improved outcomes seen at 18 months were not seen at 5 years after birth, but the trends toward improvement in outcome still favored use of caffeine over placebo for the treatment of apnea. Patients are typically loaded with 20 mg/kg of caffeine citrate and maintained on 7 to 8 mg/kg daily. ⁷⁰

Home cardiorespiratory monitoring is a technology that was developed in the 1970s after several studies suggested a possible relationship between apnea and sudden infant death syndrome (SIDS). Since that time hundreds of thousands of premature infants have been discharged with these monitors. Although anecdotal evidence has shown that these devices are effective in decreasing the rate of SIDS, no large, controlled study has demonstrated this conclusively.

Sudden Infant Death Syndrome

SIDS is the unexpected death of an infant younger than 1 year that remains unexplained after a thorough autopsy, history, and investigation of the scene of the death. It was named SIDS in 1969. Although people of all ages die suddenly, the rate of sudden death is highest for those younger than 1 year of age. The 1-year cutoff has been arbitrarily assigned; actually, the overwhelming majority of SIDS deaths occur before 6 months of life.

In the United States approximately 3000 deaths resulted from SIDS in 1998. This represents a rate of approximately 0.75 deaths caused by SIDS per 1000 live births. The peak age of SIDS is 2 to 5 months. The rate is substantially higher in urban areas, particularly among African-American infants. Interestingly, the SIDS rates in the Hispanic and Asian populations are equal to or lower than that of the Caucasian population in the United States. In the developed nations of Europe and Asia, SIDS rates are slightly lower than in the United States. SIDS rates are also lower in Australia and New Zealand.

The SIDS rate slowly declined from 1985 to 1994, then it began to drop precipitously from about 2 deaths per 1000 births in 1992 to the present 0.55 per 1000 births. This rapid decline paralleled the institution of the “Back to Sleep” campaign sponsored by the National Institutes of Health, the SIDS Alliance, and the American Academy of Pediatrics. This initiative followed the discovery that the simple act of changing infants’ sleeping positions from the stomach to the back was responsible for a dramatic reduction in the SIDS rate in England and Australia. In the United States the rate of infants sleeping on their backs has risen from 15% to over 70% in the past 5 years. It is likely that the SIDS rate will decrease even more as increasing numbers of infants sleep on their backs. ⁷¹

In medicine, as in all aspects of life, uncomplicated and elegant observations can make great differences. Although the exact physiology is unclear, it is likely that sleeping on the back reduces the rebreathing of carbon dioxide; adjusts the position of the airway, thus reducing obstruction; and reduces the possibility of poor oxygenation and ventilation through the mattress. The effects on the baby’s thermal environment and the ability to eliminate heat may also be important.

There has been a great deal of publicity about infants originally diagnosed with SIDS who were ultimately found to be the victims of a homicidal parent. These children included the case that established the supposed link between SIDS and apnea. Again, like many news stories, this represents an extremely small number of cases and is the exception rather than the rule. Although it is impossible to quantify, it is thought that fewer than 2% of SIDS cases are probable homicides.

The greatest known risks for SIDS appear to be prone sleeping and maternal smoking, both prenatally and postnatally. The American Academy of Pediatrics’ “Back to Sleep” program, which encourages parents to put their infants to sleep lying on their backs, has led to a decrease in the number of SIDS cases reported in the United States. Other apparent risk factors include African-American race, low socioeconomic status, young maternal age, winter season, and prematurity. More recently, some evidence has suggested that there are genetic markers for SIDS in some families. The SIDS rate for premature infants is about 2.25 times that for term infants. Infants with apnea of prematurity are at no greater risk for SIDS than premature infants without apnea of prematurity.

Lung Abnormalities

Congenital cystic lesions of the lung generally include those diseases that result from a problem in the formation of mesodermal and ectodermal tissue during lung development. These lesions include pulmonary sequestrations, congenital cystic adenomatoid malformations, congenital lobar emphysema, and bronchogenic pulmonary cysts.

Pulmonary sequestration is thought to be the most common congenital lung malformation. A pulmonary sequestration is an area of nonfunctioning lung tissue with no connection to the tracheobronchial tree but with a systemic arterial supply. Pulmonary sequestrations can be diagnosed antenatally. They can be asymptomatic in the newborn or can cause respiratory distress caused by lung compression or congestive heart failure. Resection is generally recommended, even if asymptomatic, to reduce the secondary risk of recurrent infection.

Pulmonary sequestrations are either extralobar or intrapulmonary. Extralobar pulmonary sequestrations include lesions with lung tissue surrounded by its own pleura. Intrapulmonary sequestrations, also known as intralobar sequestrations, have no discernible pleural tissue.

Congenital cystic adenomatoid malformation originates as an adenomatous growth in the terminal bronchioles early in gestation. In most cases there is a connection with the tracheobronchial tree that causes these lesions to increase in size. Only one lobe of the lung is usually involved. Congenital cystic adenomatoid malformations are now frequently diagnosed in the antenatal period by sonography. The most common presentation in the postnatal period is respiratory distress. Surgical removal of the affected lobe is the treatment of choice.

PPHN is the most common cause. Lung malformations such as congenital diaphragmatic hernia and congenital cystic adenomatoid malformation can lead to lung hypoplasia and concomitant PPHN. Recent efforts have been made to identify infants at greatest risk of mortality who might be candidates for fetal surgical intervention.

The presence of nonimmune hydrops fetalis, shift of the mediastinum, bilateral lesions, and the presence of other associated congenital abnormalities all portend a poor prognosis for infants with congenital lung lesions.

Congenital cystic adenomatoid malformation has been treated with some success in the antenatal period. Antenatal surgical repair of congenital cystic adenomatoid malformations is generally limited to infants with fetal hydrops. One series of 13 infants had 8 survivors; 5 infants died in the intraoperative or perioperative period. In all survivors resection of the malformation led to resolution of fetal hydrops and increased lung growth. The principal operative concern is the initiation of maternal labor. Antenatal treatment of congenital diaphragmatic hernia has been ineffective.

Congenital lobar emphysema is caused by antenatal bronchial obstruction. This obstruction can be either intrinsic or extrinsic to the bronchiole and causes an overinflation of the pulmonary lobe. Intraluminal obstruction can result from a cartilaginous deficiency or inflammatory changes. Extrinsic causes include compression from an adjacent vascular structure or mass. Infants can present with respiratory distress or be asymptomatic in the newborn period. This lesion is more common in males, is usually seen in the left upper lobe, and is frequently associated with other congenital abnormalities of the heart and kidney. The treatment of congenital lobar emphysema is usually lobectomy.

A bronchogenic cyst results from abnormal budding of bronchial tissue during development. Bronchogenic cysts are single unilocular lesions of 2 to 10 cm in diameter. The cysts may or may not communicate with the remainder of the tracheobronchial tree. Bronchogenic cysts can be found in the mediastinum or in the peripheral lung tissue. Mediastinal cysts are thought to arise earlier in the development than those found in the periphery. Bronchogenic cysts can be asymptomatic at birth and may not present until adulthood. Other lesions are symptomatic because of compression or infection. Surgical resection is generally recommended.