Article Text
PDF
Abstract
Objectives To describe breathing patterns in infants with Prader-Willi Syndrome (PWS), as well as the effects of supplemental oxygen (O2) on breathing patterns. Children with PWS commonly have sleep-disordered breathing, including hypersomnolence and obstructive sleep apnoea, as well as central sleep breathing abnormalities that are present from infancy.
Design Retrospective cohort study.
Patients Infants with a diagnosis of PWS.
Setting Tertiary children's hospital.
Interventions Infants with PWS underwent full polysomnography, and in those with frequent desaturations associated with central events, supplemental O2 during sleep was started and followed with regular split-night studies (periods in both air and O2).
Results Thirty split-night studies on 10 infants (8 female) aged 0.06–1.79 (median 0.68, IQR 0.45, 1.07) years were undertaken. At baseline (ie, air), children with PWS had a median (IQR) central apnoea index (CAI) of 4.7 (1.9, 10.6) per hour, with accompanying falls in oxygen saturation (SpO2). O2 therapy led to statistically significant reductions in CAI to 2.5/hour (p=0.002), as well as a reduced central event index (CEI) and improved SpO2. No change in the number of obstructive events was noted. Central events were more prevalent in rapid-eye movement/active sleep.
Conclusions It is concluded that infants with PWS may have central sleep-disordered breathing, which, in some children, may cause frequent desaturations. Improvements in CAI and CEI as well as oxygenation were noted with O2 therapy. Longitudinal work with this patient group would help to establish the timing of onset of obstructive symptoms.
- Sleep
- Respiratory
Statistics from Altmetric.com
What is known about this topic
-
Children with Prader-Willi syndrome (PWS) are known to have sleep-disordered breathing.
-
Sleep disorders associated with PWS include both obstructive and central sleep apnoea.
What this study adds
-
This study demonstrates that central apnoeas are commonly seen in infants with Prader-Willi syndrome (PWS)
-
Treatment with oxygen therapy reduces the number of observed central events in infants with PWS.
-
The frequency of central events in PWS improves over time.
Background
Prader-Willi syndrome (PWS) can be inherited either due to a deletion of part of the paternal chromosome 15 or via uniparental disomy of the maternal chromosome 15—a phenomenon known as genomic imprinting.1 With the advent of genetic testing,2 there has been increased awareness of the condition, as well as earlier diagnosis. The estimated prevalence for the Australian population is 1 : 25 000.3
The presence of sleep-related breathing disorder in PWS is well recognised.4–6 In addition to the obstructive sleep apnoea and hypersomnolence reported with the condition,7 central respiratory control problems are noted in children with PWS.8 ,9 This is thought to be due to abnormalities in central chemoreceptor sensitivity in such children. There is evidence from rodent and in vivo human studies to support this.10–14
There is very limited information available on the sleep-related breathing disorder in infants and very young children, and the primary aim of this study was to study the frequency of central apnoeas during sleep in infants with PWS. Secondary aims were to describe the effects of supplemental oxygen on central apnoea indices and oxyhaemoglobin saturations (SpO2) during sleep in these infants.
Methods
The study is a retrospective, descriptive case series of infants attending PWS clinic at Mater Children's Hospital, Brisbane (MCH). Infants had polysomnography (PSG) in a fully equipped sleep laboratory (MCH). Electroencepahologram (EEG), electrooculogram (EOG), electromyogram (EMG) (Network Concepts, Inc), respiratory inductance plethysmography of ribcage and abdominal wall (Respitrace), oxygen saturation (Novametrix), transcutaneous carbon dioxide measurement (Radiometer Copenhagen) and nasal airflow by a pressure transducer (Sullivan) were recorded. PSG studies were analysed according to the 2007 AASM guidelines for sleep scoring,15 and central apnoea was defined as the absence of airflow for >2 successive breaths. Central hypopnoea was scored when a 50% reduction in airflow with reduced respiratory effort was witnessed. As a slight modification to the AASM rule central apnoeas were scored such that any pause that was not preceded by a sigh was scored as a central apnoea if it was >2 breaths in duration whether an accompanying desaturation, arousal or awakening was present or not. Central hypopnoeas were similarly scored.
The primary outcomes assessed were central apnoea index (CAI), central event index (CEI), obstructive event indices, duration of central apnoea, % sleep time below 90% saturations and number of desaturation episodes below 90%. Central event indices during rapid-eye movement (REM) and non-REM sleep were analysed in a subset of subjects.
All children reported in this study had episodic desaturations associated with central events, and were started on supplemental oxygen during sleep. Each child was followed at 3-monthly intervals with split-night studies (periods in both air and supplemental oxygen). For the majority (22/30) of studies, the study was commenced in room air and oxygen titrated after at least one sleep cycle was obtained. In a minority (8/30) of studies, the study began in oxygen with subsequent time in room air. A database of results was compiled using Microsoft Excel, and statistical analyses were performed using SPSS for windows V.15.0. Non-parametric (Wilcoxon signed rank) tests were used to compare data in air and oxygen arms for our subject cohort. Ethics approval was obtained from the hospital ethics committee.
Results
Thirty split-night studies on 10 infants (8 female) aged 0.06–1.79 years with a median (IQR) age of 0.68 (0.45, 1.07) years were undertaken. The baseline characteristics of the study population are shown (table 1). A median (IQR) total sleep time of 501 (448, 527) min was measured across the group.
Demographics of the Prader-Willi study population
At baseline (ie, while in room air), children with PWS had a median (IQR) CAI of 4.7 (1.9, 10.7) per hour, with accompanying falls in SpO2. Oxygen therapy led to statistically significant reductions in CAI to 2.5 (0.9, 5.6) per hour (p=0.002) (figure 1) as well as improved SpO2 (table 2). The median CEI was also higher in air (10; 4.3, 18.9) compared with 5.3 (2.4, 9.5) in oxygen (p<0.001). No change in the number of obstructive events was noted. No differences in sleep efficiency were seen between the air and oxygen arms of the study.
Effect of oxygen therapy on sleep-disordered breathing in infants with Prader-Willi Syndrome
Differences in central apnoeic events per hour between air and oxygen therapy in infants with Prader-Willi syndrome.
The association of events with sleep stage was investigated and these data are shown in table 3. Sleep stages were dichotomised as (1) REM sleep or active sleep (AS) and (2) non-REM or quiet sleep (QS). The median (IQR) central event (apnoea and hypopnoea) index (CEI) was noted in air to be 15.9 (8.6, 32.8) events.h−1 during REM/AS and 4.4 (1.5, 7.9) during non-REM/QS (p<0.001). In oxygen at 250 ml/min, the CEI was 9.1 (4.2, 19.6) during REM/AS and 1.9 (0.4, 4.3) during non-REM/QS (p<0.001). Thus events were more prominent in REM sleep in both air and oxygen arms. Significant falls in CEI from baseline (air) during REM (p=0.001) and non-REM sleep (p=0.002) were noted with oxygen therapy (see table 3).
Effects of sleep stage (REM/AS vs non-REM/QS) on central events in air and oxygen
A trend towards improvement over time was noted. Of the 10 infants studied, seven had three or more split-night studies undertaken over the study duration. Study 1 was undertaken at a median (IQR) age of 0.42 (0.12, 0.65) years, study 2 at 0.67 (0.39, 0.88) years and study 3 at 0.94 (0.68, 1.32) years. The median (range) duration between study 1 and study 2 was 13 (12–15) weeks, and was 14 (9–34) weeks between study 2 and study 3.
The evolution with age and improvement in central apnoea and central event frequency over the course of studies 1–3 for this group of seven patients are illustrated in figure 2.
(A) Evolution of central apnoea frequency with increasing age in seven infants with Prader-Willi syndrome who underwent split-night air/oxygen sleep studies. (B) Evolution of central event frequency with increasing age in seven infants with Prader-Willi syndrome who underwent split-night air/oxygen sleep studies.
Purported mechanism by which abnormalities in respiratory control result in central apnoea in children with Prader-Willi syndrome.
Discussion
Respiratory control during sleep develops rapidly through infancy and may be integral to the neurodevelopment maturation that is taking place in such babies. There is evidence in the literature for the presence of central apnoeas and dysrhythmic breathing in normal infants, which progressively improve with age,16 ,17 and also evidence of abnormal respiratory control in PWS from both human and rodent studies.9–14
The mechanisms by which such central events occur are hypothesised to be related to abnormalities in respiratory control allowing the crossing of a threshold between eupnoea and apnoea.18 It is thought that a stimulus (eg, hypoxia, arousal) results in a transient increase in ventilatory drive. In a normal individual the increased ventilation would continue until the resultant reduction in paCO2 is detected at the chemoreceptors, but for reasons of poor respiratory control (eg, PWS), this mechanism overshoots and results in a fall in paCO2 to below the eupnic level and the reaching of ‘apnoea threshold’. A central apnoea thus occurs, which may result in desaturation or be terminated by an arousal, both stimuli that may further perpetuate such a cycle of events by increasing drive.19 It is known that the NDN (necdin) gene is deleted in children with PWS, and in a murine model, knockout of the NDN gene led to abnormal development of the pre-Bőtzinger complex and a respiratory phenotype with abnormal breathing and prolonged central apnoeas.10 Furthermore, PWS patients are reported to have blunted responses to hypoxia and hypercapnia and abnormal chemoreceptor sensitivity.11–14 PWS infants appeared to have greater respiratory stability in oxygen, and the mechanism would support this, for if by eliminating hypoxia as a causal factor in the cascade by which central events are generated, one might expect event frequency to be markedly reduced. Figure 3 highlights the purported mechanism by which central apnoeas arise in children with PWS and blunted chemoreceptor sensitivity.
In our group of infants, reduced respiratory effort was noticed in REM and non-REM sleep stages, although was more marked in REM/AS, with increases in central event numbers as well as increased number of falls in SpO2. This was witnessed in both air and oxygen arms of the study and one limitation of our study is the trend towards a shorter duration of REM/AS that was captured in oxygen versus air. This is likely to reflect the fact that functional residual capacity (FRC) of the lung is lower in REM sleep, and SpO2 dips are thus more likely to result from apnoeic events, reinforcing the cyclical mechanism for aberrant respiratory control highlighted in figure 3. As infants grew older the degree of desaturation lessened, while at the same time the dysrhythmic breathing pattern steadily improved (figure 2a, 2b) which may support changes in FRC as a significant contributory factor to such events. With advanced age, and a greater FRC (due to steady improvements in muscle tone, coupled with lung growth and changes in lung mechanics), desaturation is less likely to occur in association with a short period of apnoea, and thus the substrate of hypoxia required to perpetuate events is no longer present.
In our case series, we report the presence of central apnoeas in REM sleep, which may share similarities with the REM hypoventilation reported in older children with PWS who are obese.20 ,21 Thus, a dysrhythmic breathing pattern seen in PWS may be inherent to the condition (due to the chemoreceptor insensitivity that arises from the NDN gene deletion) and which can expresses itself in the presence of appropriate factors, namely hypotonia and immature respiratory control in infancy, or obesity in the older child.
Conclusions
Infants with PWS have sleep-disordered breathing problems, which are predominantly central in origin, and cause significant hypoxia in some patients. Improvements in both central event indices and oxygenation were noted on oxygen therapy. Longitudinal work with this patient group would help to establish the timing of onset of obstructive symptoms. Whether early recognition of central hypoventilation and correction with oxygen alter the evolution of respiratory dysfunction in this patient group remain to be seen.
Acknowledgments
The authors thank Chloe Parsley for help with data collection.
References
Footnotes
-
Contributors SS and TG devised the idea for the paper. DSU and SS wrote the first draft of the paper. TG, GW, MH and ON edited the first draft and co-wrote subsequent drafts of the paper with SS and DSU.
-
Competing interests None.
-
Ethics approval Mater Children's Hospital Ethics Committee.
-
Provenance and peer review Not commissioned; externally peer reviewed.