Sleep-related breathing disorders in children

STEPHEN H. SHELDON, DO, FAAP

aFall 1996

DISORDERED BREATHING during sleep is a major problem affecting the health and well-being of many infants, children and adolescents. Although significant progress has been made in understanding various pathological conditions that result in sleep-disordered breathing, exact etiologies remain elusive. Sleep-related breathing disorders are common and can have a profound effect on a youngster’s health and other medical disorders; they can also result in significant morbidity and/or mortality. Daytime symptoms are frequently misinterpreted because consequences of sleep disorders are similar if not identical to many other abnormalities. Sequelae can range from sudden unexpected death at night to school failure and learning disability. Careful and systematic evaluation of a child with suspected sleep disordered breathing is essential for appropriate diagnosis and management.

Normal and Abnormal Respiratory Pauses During Sleep

Respiratory pauses during sleep can be divided into eight types. Clinical significance varies according to the conceptual age of the youngster, developmental status and the presence of other medical or congenital abnormalities. Respiratory pauses include obstructive apnea, central apnea, mixed apnea, obstructive expiratory apnea, post-sigh apnea, central hypoventilation, obstructive hypoventilation and periodic breathing.

FIGURE 1

An obstructive apnea is defined as an absence of nasal and oral airflow from the nose and mouth despite continuing chest and/or abdominal effort (see Figure 1). The absence of airflow may be brief, lasting six seconds or less. If there are at least two obstructed respiratory efforts, the apnea may be clinically significant. Occasionally, obstructive apnea is prolonged and associated with cardiac deceleration (generally occuring during the last third of the apnea), oxygen desaturation and elevated end-tidal CO₂ (EtCO₂).

FIGURE 2

Definition of central apnea varies according to the age of the patient. In premature and very young infants, central apnea is defined as absence of nasal and oral airflow and respiratory efforts lasting 20 seconds or longer. The apnea may also be clinically significant if it is shorter than 20 seconds but associated with heart-rate changes or oxygen desaturation (see Figure 2). Prolonged central apnea, related to the absence of both inspiratory and expiratory neuronal activity, appears to be quite rare in otherwise normal infants and children during non-REM/quiet sleep. Brief central apnea is normal during REM/active sleep. When prolonged central apneas occur during quiet sleep, central hypoventilation syndrome or a primary underlying central nervous system or brainstem abnormality may be present.

FIGURE 3

Mixed apnea contains polysomnographic components of both central and obstructive respiratory pauses (see Figure 3). There is some evidence, however, that the "central" component of a mixed apnea is either prolonged expiration or an obstructive expiratory apnea.¹ It may also represent a combination of both.

Obstructive expiratory apnea has received little attention. It is defined as absence of polysomnographically recorded nasal and oral airflow in the presence of continued expiratory effort against an occluded upper airway. Almost all expiratory apneas are preceded by an augmented breath or sigh.² Upper airway occlusion may occur at any level, from the oropharynx to the larynx. Preliminary evidence suggests that in some children reflex glottic adduction occurs in the presence of continued diaphragmatic contraction. Increased parasympathetic tone may be present, affecting the diaphragm via the vagus nerve, as well as increasing vocal cord adduction via the recurrent laryngeal nerve. Significant heart rate deceleration occurs during the first third of the respiratory pause in a manner similar to that seen during the Valsalva maneuver. There is little change in the SaO₂ in spite of the length of the apnea. This might be explained by a temporary increase in lung volume and positive expiratory pressure. The potential importance of recognizing obstructive expiratory apnea is evident when assessing apparent life-threatening events (ALTEs) and apnea/bradycardia during infant home monitoring.

Although obstructive expiratory apneas are typically preceded by a sigh, post-sigh apneas are associated with little to no upper airway obstruction. An augmented breath is followed by a central-appearing apnea. There is little change in heart rate, almost no oxygen desaturation, and hypercarbia is absent during and after the event. Identification of these respiratory pauses are also important in evaluating infants with ALTEs and in follow-up of infants being monitored at home.

Central hypoventilation is a rare disorder defined as inadequate ventilation due to inadequate output from brainstem centers, which control breathing. Most children with this disorder have normal respiratory rates but breathe extremely shallowly during sleep. Other children with central hypoventilation have normal or increased tidal volumes but significantly reduced respiratory rates.

Obstructive hypoventilation occurs when there is chronic upper airway obstruction associated with high upper airway resistance. Overt apnea is often not present. Respiratory rate during sleep is usually increased, and oxygen saturation is typically in the normal range. End-tidal CO₂ measurements, however, are significantly elevated and remain high because of the upper airway occlusion. It has been recommended that obstructive hypoventilation be considered when EtCO₂ measurements remain above 45 mm Hg for more than 60 percent of the total sleep time or when the EtCO₂ rises above 53 mm Hg.³

Periodic breathing is a pattern of respiration characterized by three or more respiratory pauses of three seconds or longer separated by normal breathing for less than 20 seconds.⁴ Periodic breathing is a normal pattern of respiration in premature infants during active/REM and quiet/NREM sleep. It persists in young infants, but decreases in percentage as the youngster matures. In term infants, periodic breathing is most often confined to active/REM sleep. Persistence of periodic breathing that occurs during long portions of sleep may be abnormal and reflect immaturity or an abnormality of brainstem respiratory control. Unfortunately, clear norms regarding appropriate percentages of periodic breathing across age groups are not yet available.

Clinical Syndromes

Obstructive Sleep Apnea Syndrome

The hallmark of obstructive sleep apnea (OSA) in children is snoring. Snoring, however, can be absent even in the presence of severe OSA. Snoring may be associated with pauses and snorts. Difficulty breathing during sleep, restless sleep, excessive sweating, morning headaches, excessive morning thirst, nightmares, sleep terrors and enuresis may be associated symptoms. Daytime abnormalities include sleepiness, hyperactivity, poor school performance, abnormal behavior, aggressiveness, pathological shyness and social withdrawal. Learning problems, frequent upper airway infections, failure-to-thrive or obesity may occur. In severe cases, pulmonary hypertension and cor pulmonale can develop.

OSA during childhood is often associated with an anatomical abnormality of the upper airway. The most common cause of upper airway obstruction in children is hypertrophy of the tonsils and adenoids.⁵ Malformations of the mandible and maxillae can also cause upper airway obstruction during sleep. Central and peripheral neurological abnormalities may also result in OSA due to dysfunction of pharyngeal muscular movements. Although obstructive apnea predominates, other types of apneas (e.g., central, mixed, expiratory and periodic/Cheyne-Stokes breathing) occur in children with neurological abnormalities.

Apnea of Prematurity

Apnea of prematurity (AOP) is defined as excessive periodic breathing with pathological apnea in a premature infant. Almost half of all premature infants manifest periodic breathing during the neonatal period. Periodic breathing occurs with greater frequency as gestational age decreases and is present in almost all newborns less than 28 weeks gestation.⁶ Apnea of prematurity occurs in about half of all infants with periodic breathing. It is thought to be related to immaturity of the brainstem respiratory centers, central and peripheral chemoreceptors, and pulmonary reflexes.

Even though many respiratory pauses in the premature infant are central, evidence suggests that half of all apneas in premature infants are obstructive or mixed in origin.^7–9 Complex neuromuscular events required to maintain pharyngeal patency during respiration are easily overwhelmed in these immature, and often ill, newborns.¹⁰

AOP is often highly responsive to treatment. Treatment of recurrent, clinically significant apnea in a premature infant should ensure adequate oxygenation and ventilation. Physical stimulation may be all that is necessary. Continuous positive airway pressure, supplemental oxygen and/or mechanical ventilation may also be required.

Methylxanthines are currently the most widely used medications in the treatment of AOP.^11,12 Theophylline has been shown to reduce the number of apneic episodes and the development of respiratory failure,¹³ but does not shorten the course of AOP. Caffeine has been shown to induce a significant increase in ventilation, tidal volume and mean inspiratory flow. Caffeine appears to increase ventilation mainly by increasing central inspiratory drive. Both caffeine and theophylline have similar effects on episodes of apnea and bradycardia; however, caffeine seems to have an earlier effect on respiratory rate. Side effects of tachycardia, arousal and gastrointestinal intolerance are more frequently observed with theophylline when compared to caffeine.¹⁴ Theophylline has also been shown to be associated with decrease in cerebral blood flow.¹⁵ Caffeine provides stable plasma levels and has a significantly longer plasma half-life, allowing the prescription of only one daily maintenance dose.

Nasal continuous positive airway pressure (CPAP) is quite effective in resolving obstructive apneas in premature and very young infants. Central apneas, on the other hand, are unaffected by nasal CPAP, although oxygenation has been shown to increase whether or not apnea is present.¹⁶ For central apneas in the premature infants, methylxanthine treatment appears to be significantly superior to CPAP.¹⁷

Doxapram, a potent respiratory stimulant has been used experimentally to treat AOP. It appears to affect both peripheral chemoreceptors and central respiratory centers.¹⁸ Doxapram infusions have been used in preterm infants when therapeutic concentrations of theophylline failed to control episodes of central apnea.^{19, 20}

Apnea of Infancy and Apparent Life-Threatening Events

Apnea of prematurity typically resolves spontaneously by about 36 weeks postconceptual age. Periodic breathing may persist during active sleep for up to 3 months post term.^{21, 22} Some periodic breathing and brief apneic events are normal during active/REM sleep at any age (Figure 4). In contrast to AOP, apnea of infancy is more difficult to define and quite controversial. Although the definitions of apneic events are the same, etiology and pathophysiological significance are quite elusive. It is generally agreed that a pathological apnea during infancy is present when breathing ceases for 20 seconds or longer or when breathing cessation is less than 20 seconds but associated with bradycardia, cyanosis, abrupt marked pallor and/or profound hypotonia. Often the etiology for the apneic event(s) cannot be readily identified. Infants with apnea secondary to septicemia, seizures, metabolic abnormalities, gastroesophageal reflux or anatomical abnormalities are not classified in this diagnostic category. In contrast to apnea of prematurity, apnea of infancy infrequently responds to methylxanthine therapy. Accurate diagnosis is essential in order to appropriately direct treatment.

FIGURE 4

Frequently, parents or caretakers observe a significant apneic episode that frightens them and often results in resuscitative efforts. These episodes have been termed apparent life-threatening events (ALTEs). Interestingly, most infants who have been diagnosed with apnea of infancy or who have suffered an ALTE do not die of sudden infant death syndrome (SIDS).⁴ The term ALTE describes a clinical syndrome that can be due to a variety of identifiable disorders including but not limited to upper airway obstruction, central hypoventilation, gastroesophageal reflux, systemic infection, central nervous system tumors and seizures. Further complicating the picture, a variety of "normal" respiratory pauses during sleep can appear quite similar. For example, obstructive expiratory apneas occur in premature infants, term infants, toddlers and children. These apneas are associated with prolonged respiratory pauses (up to 27 seconds in our laboratory); facial hyperemia, pallor, or cyanosis; and a profound fall in the heart rate. Post-sigh apneas can also be prolonged in infants. Clinical significance of post-sigh apnea and obstructive expiratory apnea is as yet unknown.

Diagnostic Methods

Accurate diagnosis of respiratory pauses during sleep is based on continuous monitoring of respiratory, cardiovascular, and electroencephalographic parameters across the habitual sleep period. Severity of breathing disorders during sleep often varies in a circadian manner. The highest incidence of pathological apnea even in very young infants occurs during the early morning hours.²³ Monitoring is most reliable when performed during the major sleep period at night. Premature newborns and neonates may be evaluated adequately during daytime hours if several continuous sleep wake cycles are monitored.

Parameters continuously monitored during polysomnography include nasal and oral airflow by capnography, chest and abdominal respiratory effort, oxygen saturation, and measurement of EtCO₂. Oxygen saturation is monitored using pulse oximetry. Continuous EKG (Lead II) is also recorded. This provides continuous feedback of the cardiovascular effects of respiratory pauses. Continuous monitoring of EEG, chin muscle EMG, electrooculogram and patient behavior during the recording is essential. When indicated, esophageal pH is also continuously monitored. Recording multiple physiological variables provides accurate identification of sleep state-related respiratory changes as well as identification of apneas associated with other significant sleep-related disorders (e.g., sleep-related seizure activity, gastroesophageal reflux).

Pneumography (respiratory effort monitoring by thoracic impedance and heart rate monitoring) should be avoided as a diagnostic method.⁴ Pneumograms have been widely used for screening in asymptomatic premature and term infants. Unfortunately, integrity of respiration during sleep can be significantly underestimated by this type of recording. Sensitivity and specificity of pneumography is poor. False-positive and false-negative results are common. Partial airway obstruction may cause false breath detection, and breaths immediately following a sigh or biphasic augmented breaths are often missed.²⁴ In addition, unless nasal and oral airflow are recorded, only central apneas can be detected. Unfortunately, central events are only some of the breathing-related abnormalities that occur during the sleep period. Thoracic impedance measurement may fail to detect obstructive apneas and can confuse cardiac pulse artifact with respiratory effort.

Impedance pneumography done in the home is almost as costly as comprehensive technician-attended polysomnography conducted in the laboratory! Pneumograms should be reserved for evaluation of infants who trigger frequent home monitor alarms. Pneumography may help by differentiating true from false activation of the monitor. Polysomnography is still considered the most reliable and effective method available for describing respiration during sleep and for identification of normal versus abnormal upper airway function during sleep.

Home Monitoring

Home monitoring has been recommended for certain infants at increased risk of sudden unexpected death.⁴ Some data suggest, however, that infants with apnea during sleep who were perceived to require resuscitation may have a mortality rate as high as 10% despite the use of home monitors. Effectiveness of home monitoring of infants depends on the proper choice of instrumentation, appropriate training of caretakers in acceptable intervention and cardiopulmonary resuscitation, adequate compliance and continued professional support. Children who have had one or more ALTEs, infants with AOP, siblings of two or more SIDS victims and infants with certain conditions such as central hypoventilation should be monitored. Although monitoring does not guarantee prevention of unexpected death during sleep, evidence shows that lives of infants at extraordinarily high risk may be saved.

Decisions to discontinue home monitoring should be based on clinical criteria. When infants with ALTEs have had two to three months free of significant respiratory events and have shown the ability to tolerate stress from intercurrent illness and immunizations, discontinuation of home monitoring should be considered. Many programs recommend discontinuing the home monitor at six months of age if the infant has been free of significant apnea for at least two months. Requiring infants to have one or more normal pneumograms before discontinuing home monitoring may needlessly prolong the monitoring period. Use of home monitors is not without complications, and prolonged use may expose the infant to unnecessary risk,²⁵ testing and hospitalization; parents also may be subjected to unnecessary stress. It has been emphasized, however, that recommendations for the duration of monitoring of infants have not been based on conclusive data. Decisions to begin and discontinue monitoring should rest on clinical grounds and be made in close collaboration with the infants’ parents.

SIDS and Sleeping Position

A possible relationship between sleeping position and SIDS was presented as early as 1965.²⁶ A variety of subsequent studies supported this hypothesis. During the 1970s, one study reported an abrupt increase in the incidence of SIDS deaths when the predominant sleeping position changed from supine to prone.²⁷ Studies conducted in New Zealand^{28, 29} and the United Kingdom³⁰ suggested a 50% decrease in SIDS deaths after a change in sleeping position from prone to supine or lateral.

In 1985, the concept of sleeping position-dependent cerebral hypoxemia as an etiological factor in SIDS was described.³¹ Head positioning in these experiments was determined from common head angles seen in infants while sleeping in the prone position. Other data suggested that the supine postion was not associated with increased incidence and frequency of obstructive apneas regardless of sleep state.

Several other hypotheses regarding etiology appear to be significant in SIDS. Posterior displacement of the mandible with resultant obstruction of the narrow, relatively vulnerable, pharyngeal airway of the infant³² may be precipitated by facial pressure,³³ which is more likely to occur in the prone position. Obstruction of the nose and distortion of the nasal cartilage of the young infants may occur when prone.³⁴ Indeed, the neonate and young infant may spend a considerable amount of time in the face down postion when placed to sleep prone. Rebreathing from a pocket formed in the infant’s bedding and from the infant in the face down position may be a significant etiology in the susceptible infant. Soft, compliant sleeping surfaces may contribute to this problem.

Increased airway resistance may occur when an infant is placed to sleep in the prone position. Increased resistance, hypercapnea and hypoxemia typically results in arousal and movement. Increased ventilation also occurs. A disorder of arousal may exist in children who die of SIDS. Hypercarbia from rebreathing and hypoxia may develop, but arousal does not occur.

Several conditions may place an otherwise normal appearing infant at increased risk. First, cellular abnormalities within the arcuate nucleus, the area of the brainstem responsible for respiratory control, have been recently demonstrated in infants who have died from SIDS.³⁵ In addition, anomalies have been demonstrated in the carotid body, which is most sensitive to hypoxemia. During sleep, there is normally a decreased responsiveness to hypercapnia and hypoxemia. This may be exaggerated in youngsters at risk for SIDS.

Many neonates are considered to be obligate nasal breathers, although oral breathing can occur. Premature infants, however, have rare episodes of spontaneous oronasal breathing during sleep.³⁶ The frequency of oral breathing in response to nasal occlusion increases with post-conceptual age, but it is dissimilar to oronasal breathing seen in term infants. It is characterized by intermittent airway obstruction leading to significant decreased respiratory rate, tidal volume, minute ventilation and transcutaneous oxygen pressure (tcPO2). Therefore, the ability of preterm infants to use the oral route for breathing increases with postnatal maturation; however its effectiveness may remain limited by high oral airway resistance.

It is interesting that few infants who die of SIDS, ever have a history of apnea.⁴ The converse is also true; most youngsters with a history of apnea rarely die of SIDS. Neither polysomnography, nor any other testing or screening technique can identify the infant who will subsequently die of SIDS. There are, however, several subtle changes in the polysomnogram that by themselves appear to be normal but that may be indirect clues suggesting high risk. For example, youngsters who have undergone polysomnography and subsequently died of SIDS have shown long uninterrupted sleep episodes (suggesting a disorder of arousal), decreased beat-to-beat variability in heart rate, more frequent episodes of periodic breathing and a greater frequency of brief (seemingly insignificant) obstructive sleep apneas. The latter variables might suggest an abnormality in the autonomic control of cardiorespiratory function during sleep. Nonetheless, these findings are not predictive. Taken together, however, they might suggest the presence of increased risk, especially when coupled with additional risk factors such as sleeping postion, maternal cigarette smoking, prematurity, soft compliant sleeping surface and lack of breast feeding.

Probably the most important reason for performing polysomnography in young infants is to identify treatable conditions that can result in sleep-related morbidity and mortality (e.g., youngsters with obstructive sleep apnea, gastroesophageal reflux, apnea from seizure disorders, obstructive expiratory apnea). Limited home studies are unsatisfactory in evaluating infants with sleep-related breathing disorders, and evaluations unattended by a technologist cannot perform this function. View Figure 4.

REFERENCES

1. Sanders MH, Rogers RM, Pennock BE: Prolonged expiratory phase in sleep apnea: A unifying hypothesis. Am Rev Respir Dis 1985;131:401–408.

2. Sheldon SH, Onal E, Lilie J, Spire JP: Sleep-related post-inspiratory upper airway obstruction in children. Sleep Res 1993;22:270.

3. Marcus CL, Omlin KJ, Basinki DJ, et.al.: Normal polysomnographic valuse for children and adolescents. Am Rev Respir Dis 1992;146:1235–1239.

4. National Institutes of Health Consensus Development Conference: Infantile apnea and home monitoring. Bethesda, MD, US Department of Health and Human Services, Oct 1, 1987, NIH Pub No 87–2950.

5. Sheldon SH, Spire JP, Levy HB: Pediatric Sleep Medicine Philadelphia: WB Saunders, 1992, 142.

6. Bouterline-Young HJ, Smith CA: Respiration of full-term and premature infants. Am J Dis Child 1953;80:753.

7. Dransfield DA, Spiter AR Fox WW: Episodic airway obstruction in premature infants. Am J Dis Child 1983;137:441.

8. Thach BT, Start AR: Spontaneous neck flexion and airway obstruction during apneic spells in preterm infants. J Pediatr 1979;94:275.

9. Milner AD, et al: Upper airway obstruction and apnea in preterm babies. Arch Dis Child 1980;55:22.

10. Wilson ST, et al: Upper airway patency in the human infant: Influence of airway pressure and posture. J Appl Physiol 1980;48:500.

11. Kelly DH, Shannon DC: Treatment of apnea and excessive periodic breathing in the full-term infant. Pediatrics 1981;68:183.

12. Shannon DC, et al: Prevention of apnea and bradycardia in the low birth weight infant. Pediatrics 1975;55:589.

13. Sims ME, et al: Limitations of theophylline in the treatment of apnea of prematurity. Am J Dis Child 1985;139:567.

14. Bairam A, et al: Theophylline versus caffeine: comparative effects in treatment of idiopathic apnea in the preterm infant. J Pediatr 1987;110:636.

15. Rosenkrantz TS, Oh W: Aminophylline reduces cerebral blood flow velocity in low birth weight infants. Am J Dis Child 1984;138:489.

16. Miller MJ, Carlo WA, Martin RJ: Continuous positive airway pressure selectively reduces obstructive apnea in preterm infants. J Pediatr 1985;106:91.

17. Jones RAK: Apnea of immaturity. I. A controlled trial of theophylline and face mask continuous positive airway pressure. Arch Dis Child 1982;57:761.

18. Hirsch K, Wang SC: Selective respiratory stimulating action of doxapram compared to pehtylene-tetrazol. J Pharmacol Exp Ther 1974;189:1–11.

19. Barrington KJ, et al: Physiologic effects of doxapram in indiopathic apnea of prematurity. J Pediatr 1986;108:125.

20. Sagi E, et al: Idiopathic apnoea of prematurity treated with doxapram and aminophylline. Arch Dis Child 1984;59:281.

21. Hoppenbrouwers T, et al:Polygraphic studies of normal infants during the first six months of life. III. Incidence of apnea and periodic breathing. Pediatrics 1977;60:418.

22. Ariagno RL, et al: Apnea and periodic breathing in control term infants. Clin Res 1984;32:120A.

23. Guilleminault C: Sleep apnea in infancy. In: Guilleminault C (ed): Sleep and Its Disorders in Children, New York: Raven Press; 1987:213.

24. Brouillette RT, Morrow AS, Weese-Mayer DE, Hunt CF: Comparison of respiratory inductive plethysmography and thoracic impedance for apnea monitoring. J Pediatr 1987;111:377.

25. Katcher ML, Shapiro MM, Guist C: Severe injury and death associated with home infant cardiorespiratory monitors. Pediatrics 1986;78:775.

26. AAP Task Force on Infant Positioning and SIDS: Positioning and SIDS. Pediatrics 1992;89:1120.

27. Engelberts AC, deJonge Ga: Choice of sleeping position for infants: Possible association with cot death. Arch Dis Child 1990;65:462.

28. Taylor BJ, Nelson EAS, Mackay S: Changing child care practice in an area with a high postneonatal mortality. Biol Neonate April 1992, Abstract.

29. Mitchell EA, Taylor JB, Ford RPK, Stewart AW: Further evidence supporting a causal relationship between prone sleeping position and SIDS. Biol Neonate April 1992, Abstract.

30. Wigfield RE, Fleming PJ, Berry PJ, Rudd PT, Golding J: Can the fall in Avon’s sudden infant death rate be explained by changes in sleeping position? Br Med J 1991:304:282.

31. Saternus KS, Adam G: Sudden infant death. Postmortem flow measurements in the large vessels of the neck for demonstration of posture-dependent cerebral hypoxemia. Deutsche Medizinische Wochenschrift 1985;110:297.

32. Tonkin S: Sudden infant death syndrome: hypothesis of causation. Pediatrics 1975;55:650.

33. Cross KW, Lewis SR: Upper respiratory obstruction and cot death. Arch Dis Child 1971;46:211.

34. Harding R: Nasal obstruction in infancy. Aust Paediatr J 1986;22(S):59.

35. Kinney HC, Filiano JJ, Sleeper LA, Mandell F, Valdes-Dapena M, White: Decreased muscarinic receptor binding in the arcuate nucleus in sudden infant death syndrome. Science 1995;269:1446.

36. Miller MJ, et al: Effect of maturation on oral breathing in sleeping premature infants. J Pediatr 1986;109:515.

TOP