Paediatric Respiratory Reviews 15 (2014) 312–318
Contents lists available at ScienceDirect
Paediatric Respiratory Reviews
Mini-Symposium: Sudden Infant Death Syndrome
The role of physiological studies and apnoea monitoring in infants Rosemary S.C. Horne *, Gillian M. Nixon The Ritchie Centre, Monash Institute of Medical Research and Prince Henry’s Institute and Department of Paediatrics, Monash University, Level 5, Monash Medical Centre, 246 Clayton Rd, Clayton, Victoria, Australia 3168
EDUCATIONAL AIMS Understand the effect of sleep on infant cardiorespiratory physiology, including the effect of different sleep states, and how the major risk factors for SIDS including prone sleeping, maternal smoking, and prematurity adversely affect cardio-respiratory control. Understand the difference between SIDS and ALTE and that previous studies have not identified an association between these events. Understand the role for sleep studies including polysomnography in infants who have suffered an ALTE, given evidence that PSG abnormalities are neither sufficiently distinctive nor predictive to support routine use of PSG for children at risk for SIDS. Understand the indications for home monitoring in infants who have suffered an ALTE or are at risk of SIDS, and the limitations of such monitoring in preventing subsequent life threatening events.
A R T I C L E I N F O
S U M M A R Y
Article history:
There is evidence that failure of cardio-respiratory control mechanisms plays a role in the final event of the Sudden Infant Death Syndrome (SIDS). Physiological studies during sleep in both healthy term born infants and those at increased risk for SIDS have been widely used to investigate how the major risk and protective factors for SIDS identified from epidemiological studies might alter infant physiology. Clinical polysomnography (PSG) in infants who eventually succumbed to SIDS however demonstrated abnormalities that were neither sufficiently distinctive nor predictive to support routine use of PSG for infants at risk for SIDS. PSG findings have also been shown to be not predictive of recurrence of Apparent Life Threatening Events (ALTE) and thus international guidelines state that PSG is not indicated for routine evaluation in infants with an uncomplicated ALTE, although PSG may be indicated when there is clinical evidence of a sleep related breathing disorder. A decision to undertake home apnoea monitoring should consider the potential advantages and disadvantages of monitoring for that individual, in the knowledge that there is no evidence of the efficacy of such devices in preventing SIDS. ß 2014 Elsevier Ltd. All rights reserved.
Available online xxx Keywords: Sudden infant death syndrome Apparent life threatening events Physiological studies Polysomnography Home monitoring
INTRODUCTION From a physiological and pathophysiological perspective, the Sudden Infant Death Syndrome (SIDS) is considered to be multifactorial in origin and the ‘‘triple risk hypothesis’’ proposed by Filiano and Kinney in 1994, has been widely used as a model to organise current knowledge about SIDS. [1] The hypothesis proposes that SIDS results when three factors coincide: 1) a
* Corresponding author. Fax: +61 3 9594 681; Tel.: +61 3 9594 5100. E-mail address: [email protected] (Rosemary S.C. Horne). http://dx.doi.org/10.1016/j.prrv.2014.09.007 1526-0542/ß 2014 Elsevier Ltd. All rights reserved.
vulnerable infant, such as a preterm infant or one whose mother smoked during pregnancy, 2) a critical developmental period in homeostatic control–90% of infants die under 6 months of age with a peak incidence between 2-4 months, and 3) an exogenous stressor, such as being placed in the prone position for sleep. The triple risk hypothesis suggests that an infant will die of SIDS only if s/he possesses all three factors, and that the infant’s vulnerability lies latent until s/he enters the critical developmental period and is exposed to an exogenous stressor. SIDS occurs during sleep or in close association to sleep. [2] The final pathway to SIDS is now widely believed to involve immature autonomic control of the cardio-respiratory system, together with a failure of arousal from sleep. [3]
R.S.C. Horne, G.M. Nixon / Paediatric Respiratory Reviews 15 (2014) 312–318
During fetal life autonomic control of the cardio-respiratory system undergoes rapid maturation and this continues after birth across the first year of life. This underlying immaturity of cardiorespiratory control leads to an increased risk for cardio-respiratory instabilities and resultant hypoxaemia during infancy. In infants born preterm, the risks of cardiovascular instability, apnoea resulting in profound hypoxaemia, and even death are significantly greater than those born at term, [4] with the risks inversely related to gestational age at birth. The risks of these instabilities are most marked during sleep, as sleep has a marked influence on cardiorespiratory control. [4] This is of particular importance during infancy as term infants spend up to 70% of each 24 hours asleep, with those infants born preterm spending an even greater proportion of time asleep. During the first year of life, and particularly the first 6 months after birth, sleep also undergoes rapid maturation. [4] Sleep states and sleep architecture, the alternating pattern of individual sleep states, are quite different in infants to those in adults. In infants, sleep states are defined as active sleep (AS) and quiet sleep (QS), which are the precursors of adult rapid eye movement sleep (REM sleep) and non rapid eye movement sleep (NREM sleep), respectively. QS is characterised by high voltage low amplitude electroencephalograph activity, the absence of eye movements and regular heart rate and respiration. In contrast, AS is characterised by low amplitude high frequency electroencephalograph activity, eye movements, and irregular heart rate and respiration (Figure 1). At birth infants spend about equal amounts of time in each sleep state, with the two states alternating throughout each sleep period. The proportion of QS increases with age, while the amount of AS decreases and the sleep periods increase in duration. Cardiorespiratory disturbances occur predominantly in AS sleep, so the predominance of AS in early infancy likely increases the risk of cardio-respiratory disturbances during this period. PHYSIOLOGICAL STUDIES IN INFANCY IN RESEARCH SETTINGS Polysomnography (PSG) involves the simultaneous recording of multiple physiological parameters, including electroencephalogram (EEG), electromyogram (EMG), electroocculogram (EOG),
313
respiratory movements, air flow, heart rate and oxygen saturation. Physiological studies during sleep in both healthy term born infants and those at increased risk for SIDS, such as those born preterm or exposed to maternal smoking have been widely used to investigate how the major risk and protective factors for SIDS identified from epidemiological studies might alter infant physiology. The focus has been on cardio-respiratory control, given evidence that failure of cardio-respiratory control mechanisms plays a role in the final event of SIDS. We will review here the studies of normal infant cardio-respiratory physiology first, then address studies examining risk factors for SIDS directly. Effects of sleep state on cardio-respiratory control There are marked differences in cardio-respiratory control between the two sleep states (Table 1). Assessment of cardiovascular control is commonly made by studies of heart rate and heart rate patterning or variability (HRV) and blood pressure and its variability. [5] The short term or high frequency (HF) variability in heart rate is related to parasympathetic vagal activity and long term or low frequency (LF) variability depends on both sympathetic and parasympathetic branches of the autonomic nervous system. The parasympathetic or vagal and sympathetic activities constantly interact, however under resting conditions vagal tone dominates. Vagal afferent stimulation leads to reflex excitation of vagal efferent activity and inhibition of sympathetic efferent activity resulting in a decrease in heart rate and blood pressure. The opposite reflex effects are mediated by the stimulation of sympathetic afferent activity. Analysis of the changes in heart rate pattern has been used to assess the state and function of the central oscillators, sympathetic and parasympathetic vagal activity, humoral factors and the sinus node. [5] In infants, heart rate is higher in AS compared to QS, and this sleep state difference is evident as early as 2-3 weeks after birth in term infants. [6] HRV is also higher in AS compared to QS, as a result of the predominance of sympathetic activity in AS. [6] Sleep state also affects blood pressure with higher values in AS and increased variability. [7]
Figure 1. Cardio-respiratory parameters in Active and Quiet sleep in an infant. ECG electrocardiograph; EOG electrooculograph; EMG electromyograph; EEG electroencephalograph; BP blood pressure; RESP ABDO abdominal respiratory effort; RESP THOR thoracic respiratory effort; SpO2 oxygen saturation; HR heart rate. Note regular breathing and heart rate in quiet sleep compared to active sleep.
314
R.S.C. Horne, G.M. Nixon / Paediatric Respiratory Reviews 15 (2014) 312–318
Table 1 The effects of sleep state on cardio-respiratory variables in infants Cardio-respiratory Variable
Active Sleep
Quiet Sleep
Heart rate Blood pressure Heart rate variability Blood pressure variability Baroreflex sensitivity Respiratory rate Respiratory variability Hypoxic ventilatory response
increased increased elevated elevated similar increased increased immature
decreased decreased decreased decreased similar decreased decreased immature
In terms of respiratory control in infants, respiratory rate decreases from wake to sleep and breathing is more variable in AS, with short apnoeas occurring more frequently in this state, than in QS. [8] In infants the majority of apnoeas are central and obstructive events are uncommon. Periodic breathing is common in newborn infants and decreases with increasing postnatal age. Respiratory instability is more common in preterm infants and the vast majority of these infants will suffer from recurrent prolonged pauses in breathing associated with arterial desaturation and bradycardia, termed apnoea of prematurity. Preterm infants also have significantly more episodes of periodic breathing (defined as three or more sequential apnoeas lasting more than 3 s) than those born at term. This immature pattern of breathing is associated with less severe desaturation than apnoea of prematurity, but may be very frequent during sleep (Figure 2). Respiratory rate and variability decreases during the first 6 months of postnatal life. [8] An increased respiratory rate in younger infants may be necessary due to the increased demand for carbon dioxide clearance (due to increased metabolic rate), while permitting the infant to maintain ventilation with minimal effort. [9] Attempting to modulate ventilation by changing tidal volume
would require increased effort due to the highly pliable ribcage and decreased diaphragmatic efficacy. [9] In addition to sleep state dependant changes, marked effects of postnatal age on cardio-respiratory control have also been shown. Effects of postnatal age on cardiovascular control during sleep Autonomic function increases with gestational age in the fetus during pregnancy. [10,11] Both parasympathetic and sympathetic activities increase during gestation, but not in the same manner. The largest increase in parasympathetic activity occurs during the last trimester, while the largest increase in sympathetic activity occurs early on, with smaller changes occurring during the last trimester. [12] Preterm birth has been associated with immaturity of autonomic nervous system control of the cardiovascular system. This is manifest with higher heart rates [13,14], reduced heart rate variability [14–16], and decreased baroreflex sensitivity compared to infants born at term. [17,18] After birth at term, heart rate increases initially over the first month of life before declining gradually as a result of an increase in parasympathetic dominance of autonomic control of heart rate. [7,19] Studies of the maturation of heart rate control using HRV have shown an increasing dominance of parasympathetic control across the first 6 months of life. [6] Longitudinal studies examining the maturation of blood pressure and its control during sleep in healthy term infants are limited, primarily because of the difficulty of measuring blood pressure continuously and non-invasively. From large studies using intermittent blood pressure measurements it has been identified that during the first 6 weeks after birth, systolic blood pressure rises rapidly to reach a steady level which is maintained during infancy. [20] Diastolic blood pressure has been shown to fall after birth, reaching a nadir at approximately 2 months of age, followed by a gradual increase until 1 year of life. [20] In healthy
Figure 2. Polysomnographic example of the effects of periodic breathing (apnoeas marked with red boxes) which is associated with repetitive desaturations which worsen over time. Marked changes in cerebral tissue oxygenation index as measured with Near Infrared Spectroscopy are also evident with a 20% fall over time.
R.S.C. Horne, G.M. Nixon / Paediatric Respiratory Reviews 15 (2014) 312–318
term infants studied during sleep using continuous blood pressure recordings, blood pressure was found to be lower at 2-3 months when compared with both 2-4 weeks and 5-6 months, the age when the risk of SIDS is greatest. [7] There are even fewer studies of maturation of blood pressure variability (BPV) in infants. BPV decreases with increasing gestational age until term, suggesting a reduction in sympathetic modulation to term equivalent age. [21] BPV continues to decrease with postnatal age after term. [22] Taken together these findings suggest that while sympathetic vascular control is predominant in the newborn period, there is a shift to that of predominantly parasympathetic control with increasing postnatal age. The arterial baroreflex is the most important autonomic regulatory mechanism for the short term control of blood pressure, heart rate and cardiac contractility. This reflex minimises changes in blood pressure primarily by altering heart rate and arterial vascular tone. Thus, when there is an increase in blood pressure, it is countered by a decrease in both heart rate and arterial vascular tone. The responses of heart rate and vascular tone are mediated by the efferent parasympathetic and sympathetic limb of the baroreflex respectively. As both systems are involved, studies of the baroreflex provide information on the sympathovagal balance of control of the autonomic nervous system. The baroreflex is present and functional from early fetal life and undergoes significant maturation in utero. The effectiveness of the baroreflex, termed baroreflex sensitivity, increases significantly with postnatal age from 5-6 ms/mmHg at 2-4 weeks of age to 11-16 ms/mmHg at 5-6 months of age, a value similar to that reported in adults. [23] Whilst immature, infants may be at increased vulnerability to hypotensive or hypertensive events. Effects of postnatal age on respiratory control In order to maintain blood oxygen levels adult humans exposed to lowered oxygen levels experience a prompt increase in ventilation that peaks within 3-5 minutes. This period of hyperventilation is sustained for approximately 15-30 minutes before a subsequent decline to pre-hypoxic baseline values. This response is referred to as the hypoxic ventilatory response (HVR). The HVR in infants is quite different to that of adults. Following exposure to low oxygen levels, infants exhibit a ‘‘biphasic’’ HVR response which typically consists of a transient hyperventilation (within the first two minutes) termed the augmented phase, followed by a sustained reduction in ventilation towards or below normoxic levels, termed the depressive phase. This biphasic HVR has been demonstrated in term and preterm infants during both wakefulness and sleep and studies have observed the immature HVR in human infants up to 2 months after birth [24,25], while others have shown that it remains immature up to 6 months of age. [26] The findings that cardio-respiratory control is immature during infancy are important, as a failure of cardio-respiratory control mechanisms is believed to play a role in the final event of SIDS.
THE ROLE OF PHYSIOLOGICAL STUDIES OF CARDIORESPIRATORY CONTROL IN SIDS RESEARCH Recordings of infants on cardio-respiratory monitors during the time of death suggest that autonomic failure plays a role in the fatal event. Among the characteristic findings prior to the SIDS event are a short lasting yet profound bradycardia before the advent of central apnoea or sometimes in the presence of continued respiratory effort. [3,27] In support of this idea, infants who had been studied previously and who subsequently died from SIDS had higher baseline heart rates [28], lower HRV [29], a prolonged
315
QT interval [30], and low parasympathetic tone and/or high sympathovagal balance. [31,32] Physiological studies during sleep in both healthy term born infants and those at increased risk for SIDS, such as those born preterm or exposed to maternal smoking have been widely used to investigate how the major risk and protective factors for SIDS identified from epidemiological studies might alter infant physiology. Prone sleeping The prone sleeping position has been identified as the major risk factor for SIDS with some studies suggesting a causal relation between prone sleep and SIDS. [33] Importantly, cardio-respiratory control is significantly altered in the prone sleeping position. Compared to the supine sleeping position, HR in the prone position is increased during sleep [7,34–36] and HRV is reduced. [35] Recent studies which utilised non-invasive measurements of blood pressure have demonstrated that blood pressure is lower despite an elevated HR in the prone position in both AS and QS across the first 6 months of life. [7] The effects of prone sleeping were most marked at 2-3 months of age in QS when systolic blood pressure fell by 6 mmHg. [7] Furthermore, control of blood pressure is also impaired. [37] The prone position is also associated with reduced cerebral oxygenation, measured by near infrared spectroscopy, in both sleep states across the first 6 months, with a significant nadir in cerebral oxygenation at 2-3 months of age. [38] Maternal smoking Numerous studies have examined maternal smoking and its association with SIDS. [33] Infants exposed to maternal smoking have been shown to exhibit altered heart rate [39] and blood pressure [40] control when compared with non exposed infants. Other studies have however found no differences in cardiovascular control. [35] Maternal smoking may also be a confounding risk factor for SIDS due to its association with other major risk factors such as low birth-weight, prematurity and/or intra-uterine growth restriction. Maternal smoking alters autonomic cardiovascular control, but these effects are dose dependant, with some studies of normal birth weight infants and light maternal smoking reporting no significant effects. [41] Prenatal smoke exposure has the potential to decrease lung volume and compliance. In summary, physiological studies during sleep in both healthy low risk for SIDS infants and those at increased risk have provided important insights into the underlying mechanisms of SIDS. These studies support the theory that impaired cardio-respiratory control may underpin the final pathway to SIDS. THE RELATIONSHIP BETWEEN APPARENT LIFE THREATENING EVENTS (ALTE) AND SIDS An apparent life threatening event (ALTE) is ‘‘an episode that is frightening to the observer and that is characterised by some combination of apnoea (central or occasionally obstructive), colour change (usually cyanotic, but occasionally erythematous or plethoric), marked change in muscle tone (usually marked limpness), choking, or gagging.’’[42] In the 1970’s it was thought that ALTEs were precursors of SIDS and these events were referred to as Near-Miss for SIDS events. [43] During the 1980’s and 1990’s numerous studies were carried out investigating infants who presented with ALTE to try and identify causes and risk factors for SIDS, however a link between the two has never been proven and there is substantial evidence that the two conditions are in fact not related. [42] Findings that do not support an association include
316
R.S.C. Horne, G.M. Nixon / Paediatric Respiratory Reviews 15 (2014) 312–318
the absence of a temporal occurrence of the two events and failure of the incidence of ALTE to change despite large reductions in the incidence of SIDS. SIDS invariably occurs during sleep and this finding has been included in the recent definition of the event. [2] In contrast to SIDS, most ALTE have been reported to occur during the daytime [44–46] and whilst awake. [47] In most cases observers of an infant experiencing an ALTE report that the event appeared life threatening or that they thought that the infant had died, but that prompt intervention resulted in normalisation of the child’s appearance. [48] Further lack of association is supported by the evidence that SIDS incidence has significantly decreased since the early 1990’s in western countries where reducing the risks campaigns were introduced, whilst the incidence of ALTE has remained unchanged. [49] In a study of 153 cases of ALTE who were enrolled in the Collaborative Home Infant Monitoring Evaluation (CHIME) between 1994-1998, it was reported that ALTE infants differed significantly from SIDS infants in four respects. [46] Fewer ALTE infants (9%) were small for gestational age at birth compared with 19% of SIDS infants [50], fewer ALTE infants (19%) were born to teenage mothers and this distribution was similar to the general population compared to 25% of SIDS mothers [50], and ALTE infants were of a younger age with 74% under 2 months of age compared to 27% [51] and 24% [52] of SIDS infants. SIDS and ALTE infants had similar rates of prematurity ( 20%) and exposure to maternal smoking (ALTE infants 36% [46] and SIDS infants 30-66% [50–54]). One estimate has suggested that 7% of SIDS cases are preceded by ALTE [42], however in the CHIME study only 1 of 153 ALTE infants subsequently died [46], and in a review of 8 studies and 643 infants who had experienced an ALTE, 5 deaths (0.8%) were reported, with all those infants having an underlying medical problem. [55] The incidence of ALTE has been estimated to be between 0.6 and 2.46 per 1,000 live births. [49,56,57] ALTEs account for 0.6% to 1.7% of all emergency department visits of patients less than one year of age [56–58] and 2.3% of paediatric hospitalisations in the USA. [59] The majority of ALTEs occur in infants younger than 1 year of age, with a median age of 1-3 months. [55,60,61] Infants born prematurely are at increased risk. [62] Over 80% of infants with ALTE appear to have no acute distress by the time they are seen at the emergency department [63] and no specific diagnosis can be found in up to 30% of those infants seen. [60] In a systematic review of 8 non randomised descriptive studies [56,57,64–68] the most common diagnoses made in cases of ALTE were gastro-oesophageal reflux disease (GORD), which was reported in all studies and comprised 31% of total diagnoses, lower respiratory tract infection including pertussis and respiratory syncytial virus infection were reported in 5 studies and 8% of all diagnoses, and seizure was reported in 7 studies and 11% of all diagnoses. [55] Other diagnoses were problems with ear, nose and throat (3.6% of all diagnoses), cardiac problems (0.8%), urinary tract infection (1.1%), metabolic disease (1.5%), ingestion of drugs or toxins (1.5%), breath holding (2.3%), and factitious illness (0.3%). Unknown diagnoses were reported in 7/8 studies and made up 23% of all diagnoses and this varied widely from 9-83% between studies. [55] Evaluation and admission to hospital for cardio-respiratory monitoring
The question of whether an infant presenting at the Emergency Department with ALTE requires hospital admission routinely is controversial. In the past as it was thought that infants who had experienced an ALTE might subsequently go on to die from SIDS, infants were routinely admitted to hospital for diagnostic monitoring and frequently discharged home on an apnoea/ bradycardia monitor. [74] Some recent reviews of treatment for ALTE still strongly recommend admission. [56,75] However, a recent review suggested that in the light of findings that repeat events usually occurred within 24 hours, the majority of ALTE infants should only be admitted for at least 23 hours with continuous cardio-respiratory monitoring, ideally with pulse oximetry and event recording. [73] However, if the event was the first ALTE experienced, when infants were not premature, had suffered a single event which was brief, not severe and self resolving and if there was a probable cause such as gastroesophogeal reflux it was felt reasonable for the infant not to be admitted. [73] Sleep studies in infants who experience an ALTE Overnight PSG is the gold standard for quantification and differentiation of obstructive and central sleep disordered breathing. The goal of PSG in children who experience an ALTE would be to identify abnormalities that predict subsequent risk of SIDS, however studies of infants who eventually succumbed to SIDS [76–78] demonstrated PSG abnormalities that were neither sufficiently distinctive nor predictive to support routine use of PSG for children at risk for SIDS. PSG findings have also been shown to be not predictive of recurrence of ALTE. [79–81] International guidelines therefore state that PSG is not indicated for routine evaluation in infants with an uncomplicated ALTE. [82,83] It has been recommended however, that PSG should be considered in infants who have experienced an ALTE when there is clinical evidence of a sleep related breathing disorder such as obstructive sleep apnoea or sleep-related hypoventilation. [82,83] Clinical indications might include symptoms of obstructed breathing during sleep, hypercapnia on blood gas sampling, bradycardia on cardiac monitoring in the absence of central apnoea, or clinical evidence of desaturation. [83] Indeed, studies have suggested that infants who had experienced an ALTE may be at increased risk of having or developing OSA, especially when there is a family history of sleep related breathing disorder or facial dysmorphology. [84–87] Sleep studies in infants with recurrent central apnoea A slightly different but overlapping clinical scenario is where there is known or suspected apnoea during sleep in an infant. Given that apnoea of prematurity is virtually ubiquitous in very premature infants, PSG would not be indicated in those cases [83] and the apnoea can be managed and monitored clinically. There are selected cases of post-term infants however where other medical disorders have been ruled out and PSG may be indicated to confirm a diagnosis of idiopathic apnoea of infancy [87], define the extent of gas exchange abnormality and rule out hypoventilation. [83] In this context PSG can be useful to determine the need for treatment and/or home monitoring. THE ROLE OF HOME MONITORING
In a consensus statement from the European Society for the Study and Prevention of Infant Death in 2004 it was concluded that there is no standard minimal workup for the evaluation of an ALTE. [48] The assessment of an infant who presents with an ALTE has been well reviewed in the literature recently and a number of approaches to investigation have been published. [47,48,69–73]
Given that there is evidence that the presence or extent of central apnoea is not predictive of subsequent death from SIDS, the issue of the use of home apnoea monitors is a fraught one. Overthe-counter apnoea monitors are in widespread use in the community, used largely without medical advice or supervision.
R.S.C. Horne, G.M. Nixon / Paediatric Respiratory Reviews 15 (2014) 312–318
This speaks perhaps to the level of anxiety in the community about SIDS, however there is actually no evidence of efficacy of such devices in preventing SIDS. A decision to embark on home monitoring should be undertaken with this knowledge in the minds of both medical practitioners and parents, and the potential advantages and disadvantages of monitoring weighed up for that individual. It should be accompanied by ongoing medical and technical support, plus psychological support where indicated. Home monitoring of a baby after ALTE is no longer recommended, but may be helpful in a subset of patients. [72] Monitoring may allow more rapid response by parents to events, thereby limiting the extent of hypoxia. It may not however provide warning in time to prevent sudden death as a result of a respiratory or other event. Monitors with capability to record cardiac and respiratory events may be particularly helpful diagnostically following ALTE [88] as physiological parameters during an event can be assessed by expert staff. Monitors that detect apnoea only are not indicated in infants with upper airway obstruction because ongoing respiratory efforts against an obstructed airway will not result in an alarm. Event monitors that can detect desaturation and bradycardia are more helpful in this context; alternatively an oximeter may be indicated for home monitoring. Possible indications for home monitoring of an infant include: – documented apnoea resulting in cyanosis or pallor, bradycardia (
Contents lists available at ScienceDirect
Paediatric Respiratory Reviews
Mini-Symposium: Sudden Infant Death Syndrome
The role of physiological studies and apnoea monitoring in infants Rosemary S.C. Horne *, Gillian M. Nixon The Ritchie Centre, Monash Institute of Medical Research and Prince Henry’s Institute and Department of Paediatrics, Monash University, Level 5, Monash Medical Centre, 246 Clayton Rd, Clayton, Victoria, Australia 3168
EDUCATIONAL AIMS Understand the effect of sleep on infant cardiorespiratory physiology, including the effect of different sleep states, and how the major risk factors for SIDS including prone sleeping, maternal smoking, and prematurity adversely affect cardio-respiratory control. Understand the difference between SIDS and ALTE and that previous studies have not identified an association between these events. Understand the role for sleep studies including polysomnography in infants who have suffered an ALTE, given evidence that PSG abnormalities are neither sufficiently distinctive nor predictive to support routine use of PSG for children at risk for SIDS. Understand the indications for home monitoring in infants who have suffered an ALTE or are at risk of SIDS, and the limitations of such monitoring in preventing subsequent life threatening events.
A R T I C L E I N F O
S U M M A R Y
Article history:
There is evidence that failure of cardio-respiratory control mechanisms plays a role in the final event of the Sudden Infant Death Syndrome (SIDS). Physiological studies during sleep in both healthy term born infants and those at increased risk for SIDS have been widely used to investigate how the major risk and protective factors for SIDS identified from epidemiological studies might alter infant physiology. Clinical polysomnography (PSG) in infants who eventually succumbed to SIDS however demonstrated abnormalities that were neither sufficiently distinctive nor predictive to support routine use of PSG for infants at risk for SIDS. PSG findings have also been shown to be not predictive of recurrence of Apparent Life Threatening Events (ALTE) and thus international guidelines state that PSG is not indicated for routine evaluation in infants with an uncomplicated ALTE, although PSG may be indicated when there is clinical evidence of a sleep related breathing disorder. A decision to undertake home apnoea monitoring should consider the potential advantages and disadvantages of monitoring for that individual, in the knowledge that there is no evidence of the efficacy of such devices in preventing SIDS. ß 2014 Elsevier Ltd. All rights reserved.
Available online xxx Keywords: Sudden infant death syndrome Apparent life threatening events Physiological studies Polysomnography Home monitoring
INTRODUCTION From a physiological and pathophysiological perspective, the Sudden Infant Death Syndrome (SIDS) is considered to be multifactorial in origin and the ‘‘triple risk hypothesis’’ proposed by Filiano and Kinney in 1994, has been widely used as a model to organise current knowledge about SIDS. [1] The hypothesis proposes that SIDS results when three factors coincide: 1) a
* Corresponding author. Fax: +61 3 9594 681; Tel.: +61 3 9594 5100. E-mail address: [email protected] (Rosemary S.C. Horne). http://dx.doi.org/10.1016/j.prrv.2014.09.007 1526-0542/ß 2014 Elsevier Ltd. All rights reserved.
vulnerable infant, such as a preterm infant or one whose mother smoked during pregnancy, 2) a critical developmental period in homeostatic control–90% of infants die under 6 months of age with a peak incidence between 2-4 months, and 3) an exogenous stressor, such as being placed in the prone position for sleep. The triple risk hypothesis suggests that an infant will die of SIDS only if s/he possesses all three factors, and that the infant’s vulnerability lies latent until s/he enters the critical developmental period and is exposed to an exogenous stressor. SIDS occurs during sleep or in close association to sleep. [2] The final pathway to SIDS is now widely believed to involve immature autonomic control of the cardio-respiratory system, together with a failure of arousal from sleep. [3]
R.S.C. Horne, G.M. Nixon / Paediatric Respiratory Reviews 15 (2014) 312–318
During fetal life autonomic control of the cardio-respiratory system undergoes rapid maturation and this continues after birth across the first year of life. This underlying immaturity of cardiorespiratory control leads to an increased risk for cardio-respiratory instabilities and resultant hypoxaemia during infancy. In infants born preterm, the risks of cardiovascular instability, apnoea resulting in profound hypoxaemia, and even death are significantly greater than those born at term, [4] with the risks inversely related to gestational age at birth. The risks of these instabilities are most marked during sleep, as sleep has a marked influence on cardiorespiratory control. [4] This is of particular importance during infancy as term infants spend up to 70% of each 24 hours asleep, with those infants born preterm spending an even greater proportion of time asleep. During the first year of life, and particularly the first 6 months after birth, sleep also undergoes rapid maturation. [4] Sleep states and sleep architecture, the alternating pattern of individual sleep states, are quite different in infants to those in adults. In infants, sleep states are defined as active sleep (AS) and quiet sleep (QS), which are the precursors of adult rapid eye movement sleep (REM sleep) and non rapid eye movement sleep (NREM sleep), respectively. QS is characterised by high voltage low amplitude electroencephalograph activity, the absence of eye movements and regular heart rate and respiration. In contrast, AS is characterised by low amplitude high frequency electroencephalograph activity, eye movements, and irregular heart rate and respiration (Figure 1). At birth infants spend about equal amounts of time in each sleep state, with the two states alternating throughout each sleep period. The proportion of QS increases with age, while the amount of AS decreases and the sleep periods increase in duration. Cardiorespiratory disturbances occur predominantly in AS sleep, so the predominance of AS in early infancy likely increases the risk of cardio-respiratory disturbances during this period. PHYSIOLOGICAL STUDIES IN INFANCY IN RESEARCH SETTINGS Polysomnography (PSG) involves the simultaneous recording of multiple physiological parameters, including electroencephalogram (EEG), electromyogram (EMG), electroocculogram (EOG),
313
respiratory movements, air flow, heart rate and oxygen saturation. Physiological studies during sleep in both healthy term born infants and those at increased risk for SIDS, such as those born preterm or exposed to maternal smoking have been widely used to investigate how the major risk and protective factors for SIDS identified from epidemiological studies might alter infant physiology. The focus has been on cardio-respiratory control, given evidence that failure of cardio-respiratory control mechanisms plays a role in the final event of SIDS. We will review here the studies of normal infant cardio-respiratory physiology first, then address studies examining risk factors for SIDS directly. Effects of sleep state on cardio-respiratory control There are marked differences in cardio-respiratory control between the two sleep states (Table 1). Assessment of cardiovascular control is commonly made by studies of heart rate and heart rate patterning or variability (HRV) and blood pressure and its variability. [5] The short term or high frequency (HF) variability in heart rate is related to parasympathetic vagal activity and long term or low frequency (LF) variability depends on both sympathetic and parasympathetic branches of the autonomic nervous system. The parasympathetic or vagal and sympathetic activities constantly interact, however under resting conditions vagal tone dominates. Vagal afferent stimulation leads to reflex excitation of vagal efferent activity and inhibition of sympathetic efferent activity resulting in a decrease in heart rate and blood pressure. The opposite reflex effects are mediated by the stimulation of sympathetic afferent activity. Analysis of the changes in heart rate pattern has been used to assess the state and function of the central oscillators, sympathetic and parasympathetic vagal activity, humoral factors and the sinus node. [5] In infants, heart rate is higher in AS compared to QS, and this sleep state difference is evident as early as 2-3 weeks after birth in term infants. [6] HRV is also higher in AS compared to QS, as a result of the predominance of sympathetic activity in AS. [6] Sleep state also affects blood pressure with higher values in AS and increased variability. [7]
Figure 1. Cardio-respiratory parameters in Active and Quiet sleep in an infant. ECG electrocardiograph; EOG electrooculograph; EMG electromyograph; EEG electroencephalograph; BP blood pressure; RESP ABDO abdominal respiratory effort; RESP THOR thoracic respiratory effort; SpO2 oxygen saturation; HR heart rate. Note regular breathing and heart rate in quiet sleep compared to active sleep.
314
R.S.C. Horne, G.M. Nixon / Paediatric Respiratory Reviews 15 (2014) 312–318
Table 1 The effects of sleep state on cardio-respiratory variables in infants Cardio-respiratory Variable
Active Sleep
Quiet Sleep
Heart rate Blood pressure Heart rate variability Blood pressure variability Baroreflex sensitivity Respiratory rate Respiratory variability Hypoxic ventilatory response
increased increased elevated elevated similar increased increased immature
decreased decreased decreased decreased similar decreased decreased immature
In terms of respiratory control in infants, respiratory rate decreases from wake to sleep and breathing is more variable in AS, with short apnoeas occurring more frequently in this state, than in QS. [8] In infants the majority of apnoeas are central and obstructive events are uncommon. Periodic breathing is common in newborn infants and decreases with increasing postnatal age. Respiratory instability is more common in preterm infants and the vast majority of these infants will suffer from recurrent prolonged pauses in breathing associated with arterial desaturation and bradycardia, termed apnoea of prematurity. Preterm infants also have significantly more episodes of periodic breathing (defined as three or more sequential apnoeas lasting more than 3 s) than those born at term. This immature pattern of breathing is associated with less severe desaturation than apnoea of prematurity, but may be very frequent during sleep (Figure 2). Respiratory rate and variability decreases during the first 6 months of postnatal life. [8] An increased respiratory rate in younger infants may be necessary due to the increased demand for carbon dioxide clearance (due to increased metabolic rate), while permitting the infant to maintain ventilation with minimal effort. [9] Attempting to modulate ventilation by changing tidal volume
would require increased effort due to the highly pliable ribcage and decreased diaphragmatic efficacy. [9] In addition to sleep state dependant changes, marked effects of postnatal age on cardio-respiratory control have also been shown. Effects of postnatal age on cardiovascular control during sleep Autonomic function increases with gestational age in the fetus during pregnancy. [10,11] Both parasympathetic and sympathetic activities increase during gestation, but not in the same manner. The largest increase in parasympathetic activity occurs during the last trimester, while the largest increase in sympathetic activity occurs early on, with smaller changes occurring during the last trimester. [12] Preterm birth has been associated with immaturity of autonomic nervous system control of the cardiovascular system. This is manifest with higher heart rates [13,14], reduced heart rate variability [14–16], and decreased baroreflex sensitivity compared to infants born at term. [17,18] After birth at term, heart rate increases initially over the first month of life before declining gradually as a result of an increase in parasympathetic dominance of autonomic control of heart rate. [7,19] Studies of the maturation of heart rate control using HRV have shown an increasing dominance of parasympathetic control across the first 6 months of life. [6] Longitudinal studies examining the maturation of blood pressure and its control during sleep in healthy term infants are limited, primarily because of the difficulty of measuring blood pressure continuously and non-invasively. From large studies using intermittent blood pressure measurements it has been identified that during the first 6 weeks after birth, systolic blood pressure rises rapidly to reach a steady level which is maintained during infancy. [20] Diastolic blood pressure has been shown to fall after birth, reaching a nadir at approximately 2 months of age, followed by a gradual increase until 1 year of life. [20] In healthy
Figure 2. Polysomnographic example of the effects of periodic breathing (apnoeas marked with red boxes) which is associated with repetitive desaturations which worsen over time. Marked changes in cerebral tissue oxygenation index as measured with Near Infrared Spectroscopy are also evident with a 20% fall over time.
R.S.C. Horne, G.M. Nixon / Paediatric Respiratory Reviews 15 (2014) 312–318
term infants studied during sleep using continuous blood pressure recordings, blood pressure was found to be lower at 2-3 months when compared with both 2-4 weeks and 5-6 months, the age when the risk of SIDS is greatest. [7] There are even fewer studies of maturation of blood pressure variability (BPV) in infants. BPV decreases with increasing gestational age until term, suggesting a reduction in sympathetic modulation to term equivalent age. [21] BPV continues to decrease with postnatal age after term. [22] Taken together these findings suggest that while sympathetic vascular control is predominant in the newborn period, there is a shift to that of predominantly parasympathetic control with increasing postnatal age. The arterial baroreflex is the most important autonomic regulatory mechanism for the short term control of blood pressure, heart rate and cardiac contractility. This reflex minimises changes in blood pressure primarily by altering heart rate and arterial vascular tone. Thus, when there is an increase in blood pressure, it is countered by a decrease in both heart rate and arterial vascular tone. The responses of heart rate and vascular tone are mediated by the efferent parasympathetic and sympathetic limb of the baroreflex respectively. As both systems are involved, studies of the baroreflex provide information on the sympathovagal balance of control of the autonomic nervous system. The baroreflex is present and functional from early fetal life and undergoes significant maturation in utero. The effectiveness of the baroreflex, termed baroreflex sensitivity, increases significantly with postnatal age from 5-6 ms/mmHg at 2-4 weeks of age to 11-16 ms/mmHg at 5-6 months of age, a value similar to that reported in adults. [23] Whilst immature, infants may be at increased vulnerability to hypotensive or hypertensive events. Effects of postnatal age on respiratory control In order to maintain blood oxygen levels adult humans exposed to lowered oxygen levels experience a prompt increase in ventilation that peaks within 3-5 minutes. This period of hyperventilation is sustained for approximately 15-30 minutes before a subsequent decline to pre-hypoxic baseline values. This response is referred to as the hypoxic ventilatory response (HVR). The HVR in infants is quite different to that of adults. Following exposure to low oxygen levels, infants exhibit a ‘‘biphasic’’ HVR response which typically consists of a transient hyperventilation (within the first two minutes) termed the augmented phase, followed by a sustained reduction in ventilation towards or below normoxic levels, termed the depressive phase. This biphasic HVR has been demonstrated in term and preterm infants during both wakefulness and sleep and studies have observed the immature HVR in human infants up to 2 months after birth [24,25], while others have shown that it remains immature up to 6 months of age. [26] The findings that cardio-respiratory control is immature during infancy are important, as a failure of cardio-respiratory control mechanisms is believed to play a role in the final event of SIDS.
THE ROLE OF PHYSIOLOGICAL STUDIES OF CARDIORESPIRATORY CONTROL IN SIDS RESEARCH Recordings of infants on cardio-respiratory monitors during the time of death suggest that autonomic failure plays a role in the fatal event. Among the characteristic findings prior to the SIDS event are a short lasting yet profound bradycardia before the advent of central apnoea or sometimes in the presence of continued respiratory effort. [3,27] In support of this idea, infants who had been studied previously and who subsequently died from SIDS had higher baseline heart rates [28], lower HRV [29], a prolonged
315
QT interval [30], and low parasympathetic tone and/or high sympathovagal balance. [31,32] Physiological studies during sleep in both healthy term born infants and those at increased risk for SIDS, such as those born preterm or exposed to maternal smoking have been widely used to investigate how the major risk and protective factors for SIDS identified from epidemiological studies might alter infant physiology. Prone sleeping The prone sleeping position has been identified as the major risk factor for SIDS with some studies suggesting a causal relation between prone sleep and SIDS. [33] Importantly, cardio-respiratory control is significantly altered in the prone sleeping position. Compared to the supine sleeping position, HR in the prone position is increased during sleep [7,34–36] and HRV is reduced. [35] Recent studies which utilised non-invasive measurements of blood pressure have demonstrated that blood pressure is lower despite an elevated HR in the prone position in both AS and QS across the first 6 months of life. [7] The effects of prone sleeping were most marked at 2-3 months of age in QS when systolic blood pressure fell by 6 mmHg. [7] Furthermore, control of blood pressure is also impaired. [37] The prone position is also associated with reduced cerebral oxygenation, measured by near infrared spectroscopy, in both sleep states across the first 6 months, with a significant nadir in cerebral oxygenation at 2-3 months of age. [38] Maternal smoking Numerous studies have examined maternal smoking and its association with SIDS. [33] Infants exposed to maternal smoking have been shown to exhibit altered heart rate [39] and blood pressure [40] control when compared with non exposed infants. Other studies have however found no differences in cardiovascular control. [35] Maternal smoking may also be a confounding risk factor for SIDS due to its association with other major risk factors such as low birth-weight, prematurity and/or intra-uterine growth restriction. Maternal smoking alters autonomic cardiovascular control, but these effects are dose dependant, with some studies of normal birth weight infants and light maternal smoking reporting no significant effects. [41] Prenatal smoke exposure has the potential to decrease lung volume and compliance. In summary, physiological studies during sleep in both healthy low risk for SIDS infants and those at increased risk have provided important insights into the underlying mechanisms of SIDS. These studies support the theory that impaired cardio-respiratory control may underpin the final pathway to SIDS. THE RELATIONSHIP BETWEEN APPARENT LIFE THREATENING EVENTS (ALTE) AND SIDS An apparent life threatening event (ALTE) is ‘‘an episode that is frightening to the observer and that is characterised by some combination of apnoea (central or occasionally obstructive), colour change (usually cyanotic, but occasionally erythematous or plethoric), marked change in muscle tone (usually marked limpness), choking, or gagging.’’[42] In the 1970’s it was thought that ALTEs were precursors of SIDS and these events were referred to as Near-Miss for SIDS events. [43] During the 1980’s and 1990’s numerous studies were carried out investigating infants who presented with ALTE to try and identify causes and risk factors for SIDS, however a link between the two has never been proven and there is substantial evidence that the two conditions are in fact not related. [42] Findings that do not support an association include
316
R.S.C. Horne, G.M. Nixon / Paediatric Respiratory Reviews 15 (2014) 312–318
the absence of a temporal occurrence of the two events and failure of the incidence of ALTE to change despite large reductions in the incidence of SIDS. SIDS invariably occurs during sleep and this finding has been included in the recent definition of the event. [2] In contrast to SIDS, most ALTE have been reported to occur during the daytime [44–46] and whilst awake. [47] In most cases observers of an infant experiencing an ALTE report that the event appeared life threatening or that they thought that the infant had died, but that prompt intervention resulted in normalisation of the child’s appearance. [48] Further lack of association is supported by the evidence that SIDS incidence has significantly decreased since the early 1990’s in western countries where reducing the risks campaigns were introduced, whilst the incidence of ALTE has remained unchanged. [49] In a study of 153 cases of ALTE who were enrolled in the Collaborative Home Infant Monitoring Evaluation (CHIME) between 1994-1998, it was reported that ALTE infants differed significantly from SIDS infants in four respects. [46] Fewer ALTE infants (9%) were small for gestational age at birth compared with 19% of SIDS infants [50], fewer ALTE infants (19%) were born to teenage mothers and this distribution was similar to the general population compared to 25% of SIDS mothers [50], and ALTE infants were of a younger age with 74% under 2 months of age compared to 27% [51] and 24% [52] of SIDS infants. SIDS and ALTE infants had similar rates of prematurity ( 20%) and exposure to maternal smoking (ALTE infants 36% [46] and SIDS infants 30-66% [50–54]). One estimate has suggested that 7% of SIDS cases are preceded by ALTE [42], however in the CHIME study only 1 of 153 ALTE infants subsequently died [46], and in a review of 8 studies and 643 infants who had experienced an ALTE, 5 deaths (0.8%) were reported, with all those infants having an underlying medical problem. [55] The incidence of ALTE has been estimated to be between 0.6 and 2.46 per 1,000 live births. [49,56,57] ALTEs account for 0.6% to 1.7% of all emergency department visits of patients less than one year of age [56–58] and 2.3% of paediatric hospitalisations in the USA. [59] The majority of ALTEs occur in infants younger than 1 year of age, with a median age of 1-3 months. [55,60,61] Infants born prematurely are at increased risk. [62] Over 80% of infants with ALTE appear to have no acute distress by the time they are seen at the emergency department [63] and no specific diagnosis can be found in up to 30% of those infants seen. [60] In a systematic review of 8 non randomised descriptive studies [56,57,64–68] the most common diagnoses made in cases of ALTE were gastro-oesophageal reflux disease (GORD), which was reported in all studies and comprised 31% of total diagnoses, lower respiratory tract infection including pertussis and respiratory syncytial virus infection were reported in 5 studies and 8% of all diagnoses, and seizure was reported in 7 studies and 11% of all diagnoses. [55] Other diagnoses were problems with ear, nose and throat (3.6% of all diagnoses), cardiac problems (0.8%), urinary tract infection (1.1%), metabolic disease (1.5%), ingestion of drugs or toxins (1.5%), breath holding (2.3%), and factitious illness (0.3%). Unknown diagnoses were reported in 7/8 studies and made up 23% of all diagnoses and this varied widely from 9-83% between studies. [55] Evaluation and admission to hospital for cardio-respiratory monitoring
The question of whether an infant presenting at the Emergency Department with ALTE requires hospital admission routinely is controversial. In the past as it was thought that infants who had experienced an ALTE might subsequently go on to die from SIDS, infants were routinely admitted to hospital for diagnostic monitoring and frequently discharged home on an apnoea/ bradycardia monitor. [74] Some recent reviews of treatment for ALTE still strongly recommend admission. [56,75] However, a recent review suggested that in the light of findings that repeat events usually occurred within 24 hours, the majority of ALTE infants should only be admitted for at least 23 hours with continuous cardio-respiratory monitoring, ideally with pulse oximetry and event recording. [73] However, if the event was the first ALTE experienced, when infants were not premature, had suffered a single event which was brief, not severe and self resolving and if there was a probable cause such as gastroesophogeal reflux it was felt reasonable for the infant not to be admitted. [73] Sleep studies in infants who experience an ALTE Overnight PSG is the gold standard for quantification and differentiation of obstructive and central sleep disordered breathing. The goal of PSG in children who experience an ALTE would be to identify abnormalities that predict subsequent risk of SIDS, however studies of infants who eventually succumbed to SIDS [76–78] demonstrated PSG abnormalities that were neither sufficiently distinctive nor predictive to support routine use of PSG for children at risk for SIDS. PSG findings have also been shown to be not predictive of recurrence of ALTE. [79–81] International guidelines therefore state that PSG is not indicated for routine evaluation in infants with an uncomplicated ALTE. [82,83] It has been recommended however, that PSG should be considered in infants who have experienced an ALTE when there is clinical evidence of a sleep related breathing disorder such as obstructive sleep apnoea or sleep-related hypoventilation. [82,83] Clinical indications might include symptoms of obstructed breathing during sleep, hypercapnia on blood gas sampling, bradycardia on cardiac monitoring in the absence of central apnoea, or clinical evidence of desaturation. [83] Indeed, studies have suggested that infants who had experienced an ALTE may be at increased risk of having or developing OSA, especially when there is a family history of sleep related breathing disorder or facial dysmorphology. [84–87] Sleep studies in infants with recurrent central apnoea A slightly different but overlapping clinical scenario is where there is known or suspected apnoea during sleep in an infant. Given that apnoea of prematurity is virtually ubiquitous in very premature infants, PSG would not be indicated in those cases [83] and the apnoea can be managed and monitored clinically. There are selected cases of post-term infants however where other medical disorders have been ruled out and PSG may be indicated to confirm a diagnosis of idiopathic apnoea of infancy [87], define the extent of gas exchange abnormality and rule out hypoventilation. [83] In this context PSG can be useful to determine the need for treatment and/or home monitoring. THE ROLE OF HOME MONITORING
In a consensus statement from the European Society for the Study and Prevention of Infant Death in 2004 it was concluded that there is no standard minimal workup for the evaluation of an ALTE. [48] The assessment of an infant who presents with an ALTE has been well reviewed in the literature recently and a number of approaches to investigation have been published. [47,48,69–73]
Given that there is evidence that the presence or extent of central apnoea is not predictive of subsequent death from SIDS, the issue of the use of home apnoea monitors is a fraught one. Overthe-counter apnoea monitors are in widespread use in the community, used largely without medical advice or supervision.
R.S.C. Horne, G.M. Nixon / Paediatric Respiratory Reviews 15 (2014) 312–318
This speaks perhaps to the level of anxiety in the community about SIDS, however there is actually no evidence of efficacy of such devices in preventing SIDS. A decision to embark on home monitoring should be undertaken with this knowledge in the minds of both medical practitioners and parents, and the potential advantages and disadvantages of monitoring weighed up for that individual. It should be accompanied by ongoing medical and technical support, plus psychological support where indicated. Home monitoring of a baby after ALTE is no longer recommended, but may be helpful in a subset of patients. [72] Monitoring may allow more rapid response by parents to events, thereby limiting the extent of hypoxia. It may not however provide warning in time to prevent sudden death as a result of a respiratory or other event. Monitors with capability to record cardiac and respiratory events may be particularly helpful diagnostically following ALTE [88] as physiological parameters during an event can be assessed by expert staff. Monitors that detect apnoea only are not indicated in infants with upper airway obstruction because ongoing respiratory efforts against an obstructed airway will not result in an alarm. Event monitors that can detect desaturation and bradycardia are more helpful in this context; alternatively an oximeter may be indicated for home monitoring. Possible indications for home monitoring of an infant include: – documented apnoea resulting in cyanosis or pallor, bradycardia (
