Study objective: Prader-Willi syndrome (PWS) is characterized by a number of abnormalities of hypothalamic function, such as hyperphagia, short stature, temperature instability, hypogonadotropic hypogonadism, and neurosecretory growth hormone deficiency. Patients with PWS are reported to have sleepdisordered breathing and have blunted hypereapnic ventilatory responses secondary to abnormal peripheral chemoreceptor function. Thus, we hypothesized that hypercapnic arousal responses would be abnormal in PWS.
Design: Hypercapnic arousal responses were tested in ten nonobese children and adults with PWS, aged 17.7 [+ or -] 2.5 (SEM) years, 70% female, and nine control subjects, aged 14.2 [+ or -] 2.6 years, 67% female. Hypercapnic challenges were performed during stage 3/4 nonrapid eye movement sleep.
Results: The PWS subjects had a significantly higher arousal threshold to hypercapnia compared with the controls (53 [+ or -] 1.0 vs 46 [+ or -] 1.7 mm Hg; p<0.01). The PWS subjects had significantly higher baseline end-tidal [CO.sub.2] levels (42 [+ or -]0.8 vs 38 [+ or -] 1.l mm Hg; p<0.01) and more central apneas greater than 15 S/h of sleep [1.5 [+ or -] 0.3 vs 0.1 [+ or -] 0.1; P<0.01).
Conclusions: Elevated hypercapnic arousal thresholds during sleep are found in PWS subjects; these may be a manifestation of abnormal peripheral chemoreceptor function and may further contribute to sleepdisordered breathing in PWS patients.
(CHEST 1995; 108:1627-31)
BMI=body mass index; EMG=electromyogram; EOG= electro-oculogram; NREM=nonrapid eye movement; [PetCO.sub.2]=end tidal carbon dioxide tension; [PiCO.sub.2]=inspired carbon dioxide tension; PWS=Prader-Willy syndrome; Rem=rapid eye movement; [SaO.sub.2]=arterial oxygen saturation
Key words: arousal: hypercapia, Prader-Willi syndrome; respiratory control
Prader-Willi Syndrome (PWS) is the most common genetic disorder leading to obesity. It is further characterized by infantile hypotonia, hyperphagia, mental retardation, hypogonadism, and short stature.1 Some physiologic and behavioral manifestations such as hyperphagia, temperature instability, hypogonadotropic hypogonadism and neurosecretory growth hormone deficiency, and the characteristic emotional lability and outbursts of rage are attributed to a hypothalamic dysfunction in these individuals. A significant number of both obese and nonobese PWS subjects have various sleep disorders, including daytime somnolence, snoring, restless movements during sleep, cataplexy, hypoventilation, significant oxygen desaturation, and abnormalities in sleep architecture.[3-8] These abnormalities were attributed by some to the hypothalamic dysfunction of these subjects.[3,7,8] In addition, abnormalities of ventilatory control have been associated with this disorder and have recently been attributed to abnormal peripheral chemoreceptor, function in PWS.[9-11] Thus, abnormalities in ventilatory control may contribute to the pathogenesis of sleep-disordered breathing in patients with PWS.
Arousal in response to respiratory stimuli plays an important role in terminating life-threatening respiratory events such as airway obstruction or abnormalities of gas exchange. Hypercapnia has been shown to be a potent arousal stimulus in both humans and animals.[13-15] Delayed arousal responses in PWS subjects secondary to abnormalities in ventilatory control could predispose patients with PWS to sleep-disordered breathing. We therefore measured hypercapnic arousal responses in FWS subjects during stage 3/4 of nonrapid eye movement (NREM) sleep and compared these findings with those in age- and sex-matched controls.
The study was approved by the Institutional Review Board of Children Hospital Los Angeles. Informed consent was obtained from each subject.
The diagnosis of PWS was based on clinical diagnostic criteria recently delineated and/or high-resolution cytogenetic analysis. Ten nonobese (body mass index [BMI] <25 kg/[m.sup.2]) PWS subjects were studied. All subjects were free of intercurrent illness at the time of study. Nine control subjects were recruited from hospital personnel and had a negative history of pulmonary disease and sleep-disordered breathing.
An overnight polysomnogram was performed between 9 PM and 6 AM in a quiet dark room with an ambient temperature of 24 [degrees] C. Subjects were not sleep deprived. No sedation was used. The following parameters were measured and recorded continuously on a polygraph strip chart recorder (Gould) at a speed of 10 mm/s:(1) chest wall and abdominal movement by respiratory inductive plethysmography; (2) heart rate by ECG; (3) inspired [Pco.sub.2] ([PiCO.sub.2]) and end-tidal [Pco.sub.2] ([PetCO.sub.2]), sampled at the nose at a rate of 60 mL/min by mass spectrometry (Perkin-Elmer medical gas analyzer); (4) arterial oxygen saturation ([SaO.sub.2]), by pulse oximetry (Nellcor N 200 Pulse Oximeter; Hayward, Calif) and the pulse waveform signal for elimination of movement artifact; (5) ROC/AL and LOC/A2 electro-oculograms (EOG); (6) submental electromyogram (EMG); and (7) C3/A2 and 02/Al EEGs conforming to the International 10-20 placement system. In addition, subjects were also monitored continuously by an infrared camera.
Polysomnography; was performed without perturbations until uniquivocal stage 3/4 of NREM sleep was observed, as determined by the characteristic EEG, EMG, and EOG waveforms. Two hypercapnic challenges were performed by a modification of the methods of Van der Hal et al and Ward et;,al, as described below. Immediately prior to performing hvpercapnic challenges, inspired gas (room air) was delivered via a head hood at a flow rate of 10 L/min for a control period of 3 min. Following the control period, the [PiCO.sub.2] was increased slowly 20 mm Hg by blending a 10% [CO.sub.2]-90% [O.sub.2] mixture with room air. The high inspired oxygen concentration ensured that hypoxemia did not contribute to the arousal response during hercapnia. Once the [PiCO.sub.2] of 20 mm Hg was reached, the [PiCO.sub.2] was increased stepwise by 5 mm Hg/ min until arousal or for a maximum period of 3 min at 60 mm Hg. Stepwise increases in [PiCO.sub.2] allowed identification of discrete [CO.sub.2] arousal thresholds. Complete behavioral arousal accompanied by characteristic EEG changes was chosen as the most uniquivocal end point to determine the arousal threshold. After the subject bad recovered and stage 3/4 sleep was established again, the challenge, was repeated in an identical fashion. Data from the two responses were averaged for each subject. Following the second hypercapnic challenge, polysomnographic monitoring continued throughout the remainder of the night. All subjects were attended by a polysomnography technician throughout the study. A physician was present during the hypercapnic challenges.
Baseline [SaO.sub.2] and [PetCO.sub.2] were defined as the mean values of these parameters in slow-wave sleep when no respiratory disturbance occurred. Highest [PetCO.sub.2] and lowest [SaO.sub.2] refer to the highest and lowest values of these parameters during the entire overnight polysomnogram, excluding the hypercapnic challenge.
Numeric data were expressed as means and SEM. Parameters were compared between groups using thetwo-tailed unpairedt test. Arousal and demographic data were compared between groups using Fisher's Exact Test.
All PWS subjects had clinical manifestations typical of PWS and as defined recently by Holm et al. In addition, five PWS subjects (50%) had a recognizable deletion in chromosome 15(qll-ql3) confirmed by high-resolution cytogenetic examination.
PWS and control subjects were similar in age and sex; PWS subjects had a mean age of 17.7 [+ or -] 2.5 years and seven were female (70%). The mean age of controls was 14.2 [+ or -] 2.6 years and six were female (67%). Although PWS subjects were nonobese, their mean BMI was significantly higher than that of controls 22.7 [+ or -] 0.6 kg/[m.sup.2] vs 19.7 [+ or -] 0.7 kg/[m.sup.2], (p<0.0l).
The PWS subjects had a significant higher number of central apneas longer than 15 s/h of sleep; however, these were not associated with abnormalities in gas exchange. Obstructive sleep apneas were not observed in either PWS subjects or controls. Sleep-onset rapid eve movement (REM) was a common finding in the PWS group and was found in seven of ten PWS subjects and in none of the controls (p<0.001) (Table 1). There were no significant differences between PWS and control subjects in baseline oxygen saturation or the lowest oxygen saturation recorded. Baseline [PetCO.sub.2] and maximum [PetCO.sub.2] were significantly higher in PWS subjects than controls, 42 [+ or -] 0.8 mm Hg vs 38 [+ or -] 1.1 mm Hg (p<0.01), and 47 [+ or -] 1.0 mm Hg vs 41 [+ or -] 1.4 mm Hg (p<0.01), respectively (Table 1). Although significantly higher than controls, [PetCO.sub.2] observed in the PWS subjects falls within the normal range prexiously determined in our laboratory.
PWS subjects and controls had 20 and 18 hypercapnic challenges, respectively. All subjects and controls aroused in response to hypercapnia. However, PWS subjects had a significantly higher arousal threshold for hypercapnia than controls (53 [+ or -] 1.0 mm Hg vs 46 [+ or -] 1.7 mm Hg; p<0.01; Table 2).
[TABULAR DATA 1 AND 2 OMITTED]
This study shows that children and adults wih PWS have elevated arousal thresholds to hypercapnia during NREM sleep. In addition to blunted hypercapnic arousal responses, subjects with PWS also had more central apneas and a significantly higher [PetCO.sub.2] during sleep than control subjects. These findings are consistent with abnormal respiratory control in PWS subjects. Previous work from our laboratory by Gozal et al has shown that individuals with PWS have absent peripheral chemoreceptor function. We propose that blunted hyperecapnic arousal responses in PWS subjects are a manifestation of absent peripheral chemoreceptor function.
Our results for arousal thresholds for control subjects are similar to those reported previously. Hypercapnia is a potent stimulus to arousal, eliciting brisk and clear-cut behavioral arousal in essentially all normal human subjects tested.[14,18,22-25] The pathways for arousal responses to respiratory stimuli are controversial. Whether hypercapnic arousal responses are mediated directly via chemoreceptor input to the reticular activating system or via mechanoreceptors of the chest wall secondary to a hypereapnic ventilatory response remains unclear.[12,26.28] Fewell and coworkers have demonstrated a decreased frequency of arousal responses to hypercapnia in NREM sleep in lambs with carotid denervation, thus supporting the role of peripheral chemoreceptors in the generation of arousal responses to hypercapnia. PWS subjects may have delayed arousal responses to hypercapnia secondary to absent peripheral chemoreceptor input to the reticular activating system or because of a blunted ventilatory response with delayed input from mechanoreceptors to the reticular activating system. Our data cannot distinguish between these two possible explanations.
The relative contributions of peripheral chemoreceptors vs central chemoreceptors to the generation of an arousal response is also not known. However, the peripheral chemoreceptors contribute about one third of the stimulus for hypercapnic ventilatory responses. In PWS subjects who lack peripheral chemoreceptor function, we observed blunted, but not absent, arousal responses to hyperecapnia. This suggests that the delay is due to the lack of peripheral chemoreceptor function, but that the eventual arousal is due to central chemoreceptor input. This is in keeping with the finding of Arens et al who demonstrated an altered threshold (J point shifted to the right) but a normal slope of the hypercapnic ventilatory response in non-obese PWS subjects.
We used a hypereapnic-hyperoxic gas mixture in our challenge to ensure exposure to isolated hypercapnia. In normal individuals, hyperoxia inllibits ventilation and reduces, but not eliminates, peripheral chemoreceptor output.[11,30] However, Gozal et al demonstrated exposure to hypercapnia increased ventilation in PWS subjects. Despite this, we found delayed hyperocapnic arousal responses in PWS subjects with simultaneous hypercapnia. This suggests that arousal responses to hypercapnia in room air or with accompanying hypoxia would be more severely impaired in patients with PWS.
We studied subjects during stage 3/4 NREM sleep for several reasons. First, respiratory control is critically dependent on chemoreceptor input during NREM sleep with no contribution from behavioral centers of respiratory control. In contrast, during REM sleep, behavioral control of respiration is present while the contribution of chemoreceptors varities, being present in tonic REM and reduced in phasic REM, thus confounding the interpretation of results. Second, because REM sleep has both tonic and phasic components, it cannot be considered a steady state, further complicating testing. Therefore, by performing challenges in NREM sleep, we were able to study arousal responses in a steady state and stimulate the chemoreceptors while other avenues for respiratory input are less active.
We chose complete behavioral arousal as the end point for arousal response testing, although all patients were monitored with EEG, EOG, and EMG throughout the study. Behavioral arousal is an unequivocal, easily identifiable end point and has been used in most other previous studies of arousal to respiratory stimuli.[14,18,22-25] EEG arousal that did not occur simultaneously with behavioral arousal was not noted in PWS subjects or control subjects during hypercapnia, and in all instances, behavioral arousal was accompanied by EEG arousal and movement artifact on multiple channels.
It should be noted that the PWS patients we studied, although heavier than our control group, have the same BMI as the nonobese PWS subjects studied by Arens et al and Gozal et al who were shown to have an altered threshold to rebreathing hypercapnic ventilatory responses and absent peripheral chemoreceptor responses. Therefore we do not believe that differences in BMI account for our findings of blunted arousal responsiveness to hypercapnia.
PWS subjects had higher resting [PetCO.sub.2] during sleep than control subjects, although the level was still within the normal range for our laboratory as previously determined in a study of 50 normal children. This mild elevation in [PetCO.sub.2] could be caused by absent peripheral chemoreceptor function resulting in a slightly higher set point. Arousal responsiveness may be affected by habituation, ie, arousal to a specific stimulus may be blunted by repetitive exposure to that stimulus. Thus, blunted arousal to hypercapnia in PWS may have been potentiated by habituation to the mil increase in sleeping [PetCO.sub.2] in PWS subjects. The usual rise in [CO.sub.2] that occurs during sleep vs wakefulness has been attributed to the withdrawal of th nonspecific stimulus of wakefulness. Thus, it is also possible that the higher sleeping [CO.sub.2] in PWS subject could be caused by a more dramatic withdrawal of the wakefulness stimulus with the onset of sleep, perhaps related to hypothalamic dysfunction in patients with PWS. Our study provides no direct evidence to support or refute this possibility.
An alternative explanation for the increase in hypercapnic arousal threshold in PWS subjects could be sleep deprivation or fragmentation. Frequent arousals and short REM periods have been reported in PWS.[5-7] A higher incidence of sleep-onset REM was demonstrated in the PWS patients in the present study. Bowes and coworkers have demonstrated in dogs that one night of sleep fragmentation can depress arousal responses to respiratory stimuli. In addition, sleep fragmentation has been implicated in reducing arousal responsiveness in obstructive sleep apnea syndrome.
Gleeson et al postulated that it is the increase in ventilatory effort during challenges with respiratory stimuli sensed via mechanoreceptors of the chest wall that stimulates the reticular activating system and results in arousal. In their study on adult men, arousal occurred in response to hypoxia, hypercapnia, and increased resistive load at similar levels of ventilatory effort as measured by peak esophageal pressures. Therefore, relatively decreased ventilatory effort in response to the hypercapnic challenge in PWS patients may be an alternative explanation for the differences in the arousal thresholds between PWS patients and controls.
PWS is associated with CNS involvement. It is possible that delayed hypercapnic arousal is related to central mechanisms rather than decreased afferent input. In particular, hypothalamic dysfunction has been proposed in PWS, and the hypothalamus participates in arousal. The present study does not allow us to separate the relative contributions of absent peripheral chemoreceptor function and altered central afferent integration.
Sleep-disordered breathing has been reported in PWS subjects.[6-9] Arousal from sleep is an important protective response against respiratory events during sleep such as obstinctive apnea, hypercapnia, and hypoxia. It is possible that blunted arousal responsiveness to hypercapnia impairs protective physiologic responses during sleep to respiratory events, thus potentiating sleep-disordered breathing in PWS patients.
In summary, we found an elevated threshold to hypercapnia in patients wth PWS. This may be due to absent peripheral ehemoreceptor function and may contribute to the seveiity of sleep-disordered breathing in these individuals.
ACKNOWLEDGMENTS: We thank Daisy B. Bautista and Walter S. von Pechmann for technical assistance in this study.
 Cassidy SB, Ledbetter DH. Prader-Willi syndrome. Neurol Clin 1989; 7:37-54  Cassidy SB. Prader-Willi syndrome. Curr Probl Pediatr 1984; 14:1-55  Cassidy SB, McKillop J, Morgan W. Sleep disorders in Prader-Willi syndrome. Dysmorphol Clin Genet 1990; 4:13-7  Clark DJ, Waters J, Corbett JA. Adults with Prader-Willi syndrome: abnormalities of sleep and behavior. J R Soc Med 1989; 82:21-4  Vela-bueno A, et al. Sleep in the Prader-Willi syndrome: clinical and polygraphic findings. Arch Neurol 1984; 41:294-96  Kaplan J, Fredrickson PA, Richardson JW. Sleep and breathing in patients with the Prader-Willi syndrome. Mayo Clin Proc 1991; 66:1124-26  Harris J, Allen R. Sleep disordered breathing and circadian disturbances of REM in Prader-Willi syndrome. Sleep Res 1985; 14:235A  Hertz G, Cataletto M, Feinsilver SH, et al. Sleep and breathening patterns in patients with Prader-Willi syndrome (PWS): effects of age and gender. Sleep 1993; 16:366-71  Orenstein D, Boat TF, Owens RP, et al. The obesity hypoventilation syndrome in children with the Prader-Willi syndrome: a possible role for familial decreased response to carbon dioxide. J Pediatr 1980; 97:765-67  Arens R, Gozal D, Omlin KJ, et al. Hypoxic and hypercapnic ventilatory responses in Prader-Willi syndrome. J Appl Physiol 1994; 77:2224-30  Gozal D, Arens R, Omlin KJ, et al. Absent peripheral chemosensitivity in Prader-Willi syndrome. J Appl Physiol 1994; 77:2231-30  Philipson EA, Suilivan CE. Arousal: the forgotten response to respiratory stimuli. Am Rev Respir Dis 1978; 118:807-08  Philipson EA, Kozar LF, Rebuck AS, et al. Ventilatory and walking responses to C02 in sleeping dogs. Am Rev Respir Dis 1977; 115:251-59  Hedemark LL, Kronenberg RS. Ventilatory and heart rate responses to hypoxia and hyperecapnia during sleep in adults. J Appl Physiol 1982; 53:307-12  Fewell JE, Baker SB. Arousal and cardiopulmonary responses to hyperoxic hypercapnia in lambs. J Dev Physiol 1989; 12:21-6  Holm VA, Cassidy S, Butler MG, et al. Prader-Willi syndrome: consensus diagnostic criteria. Pediatrics 1993; 91:398-402  Rechtschaffen A, Kales A, eds. A manual of standardized terminology, techniques and scoring system for sleep stages of human sleep. Bethesda, Md: National Institute of Neurological Disease and Blindness, 1968 (NIH publication No. 204), 6-7  Van der Hal AL, Rodriguez AM, Sargent CW, et al. Hypoxic and hypercapnic arousal responses and prediction of subsequent apnea in apnea of infancy. Pediatrics 1985; 75:848-54  Ward SLD, Nickerson BG, Van der Hal A, et al. Absent hypoxic and bypercapnic arousal responses in children with myelomeningocele and apnea. Pediatrics 1986; 78:44-50  Phillipson EA, Bowes G. Control of breathing during sleep: the handbook of physiology: section 3: the respiratory system (volume 2), part 2. Baltimore: American Physiological Society, 1986; 621-48  Marcus CL, Omlin KJ, Basinki DJ, et al. Normal polysomnographic values for children and adolescents. Am Rev Respir Dis 1992; 146:1235-39  Marcus CL, Bautista DB, Amihyia A, et al. Hypercapnic arousal responses in children with congenital central hypoventilation syndrome. Pediatrics 1991; 88:993-98  Hedemark LL, Kronenberg RS. Flurazepam attenuates the arousal response to C02 during sleep in normal subjects. Am Rev Respir Dis 1983; 128:980-83  Berthon-Jones M, Sullivan CE. Ventilation and arousal responses to hypercapnia in normal sleeping humans. J Appl Physiol 1984; 57:59-67  Bellville JW, Howland WS, Seed JC, et al. The effect of sleep on the respiratory response to carbon dioxide, Anesthesiology 1959; 20:628-35  Gleeson K, Zwillich CW, White DP. The influence of increasing ventilatory effort on arousal from sleep. Am Rev Respir Dis 1990; 142:295-300  Fewell JE, Kondo CS, Dascalu V, et al. influence of carotid denervation on the arousal and cardiopulmonary responses to alveolar hypercapnia in lambs. J Dev Physiol 1989; 12:193-99  Gleeson K, Zwillich CW. Adenosine stimtdation, ventilation, and arousal from sleep. Am Rev Respir Dis 1992; 145:453-57  Heeringa J, Berkenbosch A, DeGoede J, et al. Relative contribution of central and peripheral chemoreceptors to the ventilatory response to C02 during hypercapnia. Respir Physiol 1979; 37:365-79  Carroll J, Canet E, Bureau MA. Dynamic ventilatory responses to C02 in the awake lamb: role of the carotid chemoreceptors. J Appl Physiol 1991; 71:2198-2205  Sullivan CE, Kozar LF, Murphy E, et al. Primary role of respiratory afferents in sustaining in breathing rhythm. J Appl Physiol Respir Environ Exercise Physiol 1978; 45:11-7  Sullivan CE, Murphy E, Kozar LF, et al. Ventilatory responses to [CO.sub.2] and lung inflation in tonic vs. phasic REM sleep. J Appl Physiol Respir Environ Exercise Physiol 1979; 47:1304-10  Fink BR. Influence of cerebral activity in wakefulness on regulation of breathing. J Appl Physiol 1961; 16:15-20  Bowes G, Woolf GM, Sullivan CE, et al. Effect of sleep fragmentation on ventilatory and arousal responses of sleeping dogs to respiratory stimuli. Am Rev Respir Dis 1980; 122:899-908  Guillemainault C, Rosekind M. The arousal threshold: sleep deprivation, sleep fragrnentation, and obstructive sleep apnea syndrome. Bull Eur Physiopathol Respir 1981; 17:341-49  Gleeson K, Zwillich CW, White DP. The influence of increasing ventilatory effort on arousal from sleep. Am Rev Respir Dis 1990; 142:295-300  Starz] TE, Taylor CW, Magoun HW. Ascending conduction in reticular activating system, with special reference to the diencephalon. J Neurophysiol 1951; 14:461
(*) From the Division of Neonatology and Pediatric Pulmonology, Childrens Hospital of Los Angeles, University of Southern California School 6f Medicine, Los Angeles. Supported by Grants from the National Institute of Child Health and Human Development (1 RO1 HD22696-01A1); the National Sudden Infant Syndrome Alliance; the SIDS Foundation of Southern California; the Washington State SIDS Foundation; the Los Angeles County, Orange County, and Inland Empire Chapters of the Guild for Infant Survival;'the Junior Women's Club of Orange; and the Ruth and Vernon Taylor Foundation. Manuscript received January 20, 1995; revision accepted June 16. Reprint requests: Dr. Ward, Division of Neonatology and Pediatric Pulmonology, MS #83, Childrens Hospital Los Angeles, 4650 Sunset Boulevard, Los Angeles, CA 90027
COPYRIGHT 1995 American College of Chest Physicians
COPYRIGHT 2004 Gale Group