Domiciliary investigation of sleep-related hypoxaemia in Duchenne muscular dystrophy

Transcription

1 Eur Resplr J 1991, 4, Domiiliary investigation of sleep-related hypoxaemia in Duhenne musular dystrophy N. Carrell*, R.J.I. Bain*, P.E.M. Smith**, S. Saltissi*, R.H.T. Edwards*, P.M.A. Calverley** Domiiliary investigation of sleep-related hypoxaemia in Duhenne musular dystrophy. N. Carroll, RJ.l. Bain, P.E.M. Smith, S. Saltissi, R.H.T. Edwards, P.M.A. Calverley. ABSTRACT: In Duhenne musular dystrophy (DMD) noturnal oxygen desaturation ours during rapid eye movement (REM) sleep. Polysomnography, whih requires hospital admission, will detet sleeprelated breathing abnormalities. In order to avoid the inonveniene of hospital admission for the disabled patient, we investigated overnight oxygenation in ten boys with DMD by domiiliary oximetry. In four boys the results of oximetry were ompared with those of a polysomnographi reording. The "repeatability" of domiiliary oximetry was assessed in six boys by performing oximetry on two nononseutive nights. Older boys with DMD may develop a ardiomyopathy. In order to assess ardia rhythm and ST segment hanges we performed simultaneous Holter monitoring and oximetry in seven boys with overnight hypoxaemia. Six of the initial ten boys studied demonstrated episodi noturnal hypoxaemia and there was a strong orrelation between minimum oxygen saturation overnight and daytime arterial oxygen and arbon dioxide tensions (Pao 2 r=.89; Pao 2 r=-.87). Despite adequate REM time during polysomnography, greater oxygen desaturation was found during domiiliary oximetry. No differene was found in the severity of desaturation reorded in the boys who were studied on two separate oasions. Five boys demonstrated marked heart rate variation during hypoxaemi episodes and more serious arrhythmias ourred overnight in the three most hypoxaemi boys. Domiiliary oximetry is a simple, repeatable method of assessing overnight oxygenation and ompares well with polysomnography. In boys with advaned DMD and severe noturnal hypoxaemia, 24 h eletroardiographi monitoring may detet potentially life-threatening arrhythmias. Eur Respir J., 1991, 4, Musle Researh Centre, Dept of Mediine, University of Liverpool, Liverpool, UK. Regional Thorai Unit, Fazakerley Hospital, Liverpool, UK. Correspondene: Dr N.Carroll, Dept of Mediine, University of Liverpool, PO Box 147, Liverpool L69 3BX, UK. Keywords: Arrhythmias; Duhenne musular dystrophy; hypoxaemia; sleep Reeived: Marh 19, 199; aepted after revision November 6, 199. Duhenne musular dystrophy (DMD) is assoiated with respiratory musle weakness and ardiomyopathy whih may lead to both ardia and respiratory failure and early death [1]. The respiratory abnormalities seen in DMD inlude impaired respiratory musle strength, gradual redution in lung volumes [2], and the development of a soliosis [3]. The linial management of these patients involves physiotherapy, prevention of obesity and the early treatment of respiratory infetions. More reently attention has turned to the detetion and treatment of noturnal oxygen desaturation in older boys with more advaned DMD [ 4] and noturnal ventilatory support has been offered to those who develop hroni respiratory failure [5, 6]. Sleep-related symptoms are unusual until the disease is very advaned [1 ], but noturnal hypoxaemia, ourring during rapid eye movement (REM) sleep, may be seen without the presene of daytime hypersomnolene or early morning headahe [S].It has proved diffiult to predit the severity of noturnal oxygen desaturation from daytime measurements of pulmonary funtion, although "desaturating" patients tend to be rather older and fatter, with lower lung volumes and signifiantly lower maximal expiratory mouth pressures (Psmax) than those who do not desaturate during sleep [7]. Whilst it is rare for boys with DMD to be symptomati of ardia abnormalities even in their late teens, ardia abnormalities an be deteted early in life.

2 SLEEP-RELATED HYPOXAEMIA IN DMD 435 Eletroardiographi (ECG) abnormalities are ommon even in the young [8, 9] and the inidene of ehoardiographi evidene of ardiomyopathy inreases with age, until present in all patients by the age of eighteen [1]. In other linial settings arrhythmias are ommon in patients with ardiomyopathy and may be exaerbated by hypoxia. The physiologial abnormalities leading to oxygen desaturation overnight in DMD have been defined by polysomnography but it is not known whether hospital based studies are essential in order to assess the severity of noturnal desaturation. It is also not known whether this episodi noturnal hypoxaemia is assoiated with ardia arrhythmias or ST segment hanges. We have therefore studied a group of older boys with DMD by domiiliary pulse oximetry, to determine the severity of noturnal hypoxaemia and, by using simultaneous ECG monitoring, we have looked for ardia abnormalities that ould be temporally related to periods of noturnal oxygen desaturation. vital apaity (FVC) was measured in the sitting position, using a Vitalograph respirometer. Maximal inspiratory (P1max) and maximal expiratory (PEmax) stati mouth pressures were measured using the tehnique of BLACK and HYATI [13]. Arterial blood gas tensions were measured by sampling blood from the radial artery. An Ohmeda Biox 37 pulse oximeter and a pen reorder (Rikadenki model R-3) were then taken to the patient's home, and the ear probe was attahed to the patient at their usual bedtime. The patients slept in their normal position and were turned as neessary during the night by their parents, with any hange in sleeping position or episodes of snoring being noted on the reording. The ear probe was removed on awakening the following morning and the oxygen saturation trae subsequently analysed. "Desaturation" was defined as a fall in Sao 2 by more than 4% from the awake value. Statistial analysis using linear regression by the least squares method was used to determine the relationship between daytime measures of respiratory funtion and noturnal oxygen desaturation. Study 1 (in hospital) Methods Study 3 (repeat study at home) Full polysomnography was performed in four boys aged 15-2 yrs (median 19 yrs) in the sleep laboratory. None had symptoms of noturnal hypoventilation but they were hosen beause they were older boys in whom we might expet noturnal hypoventilation to be present. The patients slept in their usual position for two onseutive nights, the first serving as an alimatization night with all reording eletrodes attahed, but with data reorded on the seond night only. The eletroenephalogram (EEG), eletro-oulogram (EOG), eletromyogram (EMG) and eletroardiogram (ECG) were monitored ontinuously and reorded on an 8 hannel EEG pen reorder (SLE loot) together with oronasal airflow, hest and abdominal wall movement (measured by indutane plethysmography) and arterial oxygen saturation (Sao ). 2 An Ohmeda Biox 37 pulse oximeter was used to measure oxygen saturation ontinuously. Sleep was staged using standard riteria [11]. The overnight oxygenation reorded during polysomnography was subsequently ompared with that obtained during domiiliary oximetry performed a few months later in the same four boys (as desribed in study 2). Study 2 (at home) Ten boys aged yrs (median 18.5 yrs) were studied, none had symptoms of noturnal hypoventilation or evidene of daytime respiratory failure. Six of the ten had previously undergone spinal stabilization using a Luque proedure [12]. Daytime pulmonary funtion was assessed during a short hospital visit. Fored To test the "repeatability" of domiiliary oximetry, six of the boys aged yrs (median 19 yrs) were studied on a seond oasion approximately 3 mths after the first study. The patients were again monitored at home using the methods desribed above. The two oxygen saturation reordings were later ompared using a paired Student's t-test. Study 4 (24 h ECG and oximetry at home) To assess whether there is any relationship between noturnal hypoxaemia and ardia arrhythmias or eletroardiographi ST segment abnormalities in patients with DMD, seven boys (aged yrs; median 19 yrs) who demonstrated noturnal hypoxaemia in Study 2 were studied on a further oasion using overnight domiiliary oximetry and simultaneous Halter monitoring of the ECG. After areful skin preparation a modified VS lead position was hosen and skin impedane was heked (Oxford Medial Systems Xl-1) to ensure a reading of less than 5 kq. To ontrol for any ST segment hanges that might our seondary to positional hanges, one minute eah was reorded with the patient supine, in the left lateral and in the right lateral position at the start of eah Halter monitor reording. Both the ear oximeter and the Halter monitor were attahed to the patient at their usual bedtime but only the oximeter was removed on awakening. Eletroardiographi monitoring was ontinued for the full 24 h, the daytime reording therefore providing a ontrol for the night time results. The magneti tapes (TDK AD6 assettes) were replayed (Oxford 45 system) to produe a hard opy report of

3 436 N. CARROLL ET AL. the ST segment and heart rate trends and to allow eletroardiogram strips to be examined at times of interest. The presene of arrhythmias was noted together with the oxygen saturation at that time. The baseline ST level was determined after taking into aount the maximum ST hange during initial postural manoeuvres. Episodes of ST depression were identified by sanning the ST trend, printed out as real time eletroardiogram strips and then validated by examining eah point of interest using a x8 magnifying lens [14]. Study 1 (in hospital) Results During polysomnography the median total sleep time was 444 min (range min) and the median perentage time spent in REM sleep was 24.5% (range %). The median awake Sao 2 was 96% (range 95-97%) and the median minimum Sao 2 overnight, whih invariably ourred during REM sleep, was 88% (range 79-93%). The median Sao 2 during non-rem (NREM) sleep being 95%. Desaturation overnight was lassified as falls in Sao 2 of either 4-1%, or >1%. The median number of 4-1% dips per hour was.86 (range -2.6) and the median number of > 1% dips per hour was.46 (range -1.23). Desaturation was assoiated with either entral apnoei or hypopnoei events. No obstrutive episodes were noted. The sleep disordered breathing was almost exlusively seen in REM sleep as shown by a median total sleep hypopnoea/apnoea index (H!A) of 1 (range ) with a median H/A duration of 2 s (range s); but the REM sleep H!A index was 45 (range ) with a median duration of 23.5 s (range s). More detailed information from these studies has previously been reported as part of a larger "in hospital" study of ventilatory abnormalities during sleep in Duhenne musular dystrophy [15]. The polysomnographi reord was then ompared with the results of domiiliary oximetry in the same four patients. During domiiliary oximetry the median awake Sao~ was 95.5% (range 95-96%), with a median mimmum Sao 2 of 78% (range 71-86%) whih was rather lower than that seen during polysomnography (fig. 1). mhp, Prmax -117 to -128 mhp [16]. Daytime arterial blood gas tensions were well-preserved with a median arterial oxygen tension (Pao ) 2 of 11.7 kpa (range kpa) and a median arterial arbon dioxide tension (PaoJ of 5. kpa (range kpa). Only two boys demonstrated daytime arterial hyperapnia (Pao 2 >6.4 kpa). Sleep quality on the study night was reported as normal by all boys and none were noted to snore during the study. Six boys demonstrated arterial oxygen desaturation overnight lasting 15 min or more (fig. 2) Three boys did not desaturate, and one boy desaturated for only 5 min. Desaturation ourred as repetitive disrete episodes of hypoxaemia (between 1 and 7 episodes per night) eah lasting 1-6 min. Between these episodes the Sao 2 was maintained at, or very lose to, waking values. The median awake oxygen saturation (Sao ) 2 was 96% (range 9-97%) with a median nadir overnight of 86% (range 53-97% ). The median time spent with an Sao 2 >4% below the awake value was 17 min (range -62 min), with a median total monitoring time of 6 min (range min). There was no relationship between sleeping position and the presene of oxygen desaturation. ';/.! 1 9 i 8 ID g 7 JC GM GH Awake Minimum Polyaomnography JC GM AC GH Awake Minimum Oximetry Study 2 (at home) Patient harateristis are given in table 1. Respiratory measurements and arterial blood gas tensions were measured in the early afternoon in all subjets. The median FVC was low at 1.1 l (range I) and maximal respiratory pressures were redued with a median PEmax of 45 mhp (range 2-7 mhp), and a median Pimax of -4 mhp (range -1 to -6 mhp). Predited values for boys of this age are PEmax Fig A omparison of the awake and minimum oxygen saturation levels ahieved in four boys during polysomnography and oximetry. Statistial analysis revealed a strong orrelation between the minimum oxygen saturation overnight and awake Pao 2 (r=.89, p<.1), awake Pao 2 (r=-.87, p<.1), and age (r=-.65, p<.5); with a weaker orrelation with PEmax (r=.63, p<.5). There was no relationship, of statistial signifiane, between oxygen desaturation overnight and P1max or FVC (r=.48 and r=-.47, respetively).

4 SLEEP-RELATED HYPOXAEMlA IN DMD 437 Table 1. - Patient age, FVC, respiratory musle strength and daytime arterial blood gas tensions Subjet Age FVC PEmax P1max Pao 2 Pao 2 yrs I m Hp mhp kpa kpa N/A NIA N/A NIA Median FVC: fored vital apaity; PBmax and Pimax: maximal expiratory and inspiratory stati mouth pressure, respetively; Pao 2 and Pao : 2 arterial oxygen and arbon dioxide tension, respetively; N/A: not available. A ~ 1 i.. :I 1ii 5 1/) Gl tll ~ ~ 1 i.... :I ftl 5 1/) Gl tll ~ B Time h Time h Fig Examples of the oxygen saturation trae obtained overnight from two boys with DMD. A) Subjet 6, with normal daytime arterial blood gas tensions. B) Subjet 4, who demonstrates daytime arterial hypoxaemia and hyperapnia. Table 2. - Results of overnight oximetry in six boys studied on two nights (I and 11) at home, separated by at least three months Subjet Awake Minimum Monitoring Time Sao 2 >4% Sao 2 % Sao 2 % time mins below awake value min 1 I I II I I II I I Sao 2 : arterial oxygen saturation. oasion. Patient 3, however, spent 15 min desaturated during the first study and only 1 min during the seond study during whih the desaturation was less severe. Study 3 (repeat study at home) Oxygen saturation overnight in the six boys studied on two separate nights is given in table 2. For the group as a whole, no statistial differene was found between the severity of noturnal oxygen desaturation on eah Study 4 (24 h ECG and oximetry at home) All seven boys demonstrated episodi oxygen desaturation overnight from a median awake Sao 2 of 95% (range 94-96%) to a median noturnal nadir of 78% (range 54-92% ). The median total monitoring time was

5 438 N. CARROLL ET AL. a) ECG at rest during daytime b) ECG during hypoxaemi episode overnight Fig A omparison of the eletroardiographi (ECG) reords obtained in one boy during the daytime (a) and during an episode of noturnal hypoxaemia (b), showing marked variation in heart rate at night ompared with the smaller variation in rate seen during the day. 6 min (range min) and the median time during whih Sao 2 was >4% below the awake baseline was 3 min (range -72 min). Twenty four hour Halter monitoring did not demonstrate any arrhythmia during the daytime in any boy, but five boys developed a marked variation in heart rate during episodes of hypoxaemia overnight (fig. 3). More serious arrhythmias were seen in three boys. One suffered an episode of rapid atrial fibrillation and the seond had two sinus pauses eah lasting for two seonds. The third boy demonstrated four episodes of non-sustained ventriular tahyardia. These episodes ourred in the boys who suffered the most severe noturnal hypoxaemia (minimum Sao 2 overnight 81%, 54% and 65%, respetively) but these arrhythmias did not neessarily our at the most hypoxaemi time of the night. No signifiant ST segment depression or elevation ourred during any 24 h reording in any boy even during periods of hypoxaemia. Disussion Hospital admission for forty eight hours for two overnight sleep studies is unomfortable and disruptive for the disabled patient and his family. Sleep quality, as assessed in the sleep laboratory, may be poor beause of the hange in environment imposed upon the patient, and therefore the first night in the sleep laboratory is not used for data olletion. Domiiliary oximetry is a simple, safe tehnique whih provides reproduible results with the minimum of disomfort to the patient, and beause the patient is studied in their usual environment alimatization fators are of minor importane. In the present study the severity of desaturation obtained during domiiliary oximetry was greater than that seen in the same patients during polysomnography despite adequate amounts of REM sleep, suggesting that possibly an even greater amount of REM sleep was experiened on monitoring in the home environment, thus allowing more marked oxygen desaturation. Domiiliary oximetry is more likely, therefore, to define the extent of noturnal hypoxaemia in these patients. During study 3, there was a disparity between the proportion of the monitoring time spent desaturated on the two monitoring nights in only one patient. It is possible that the patient spent an unomfortable and restless seond night, thus reduing REM sleep time and, therefore, remaining well-saturated. There was no evidene of artefat on the domiiliary reordings obtained from these boys who are unable to move spontaneously during sleep beause of peripheral musle weakness. We found the dominant feature of the sleep disordered breathing pattern during polysomnography to be diminished hest wall movement during REM sleep. The majority of the desaturated episodes were due to hypopnoei events although apnoei episodes of the "entral" type did our. The distintion between entral and obstrutive apnoea depends on thorai and abdominal wall movement whih may be severely diminished in the very weak patient. The use of indutane plethysmography in study 1 indiated that the events observed in our patients were entral in origin but it may be impossible to differentiate between obstrutive and entral events in suh patients without the use of gastri and oesophageal pressure monitoring. Sleep-related hypoxaemia an our without daytime symptoms in DMD. We found that the minimum oxygen saturation overnight is related to inreasing age, daytime arterial blood gas tensions and redution in maximal expiratory mouth pressure. Those boys with the most

6 SLEEP-RELATED HYPOXAEMIA IN DMD abnormal daytime arterial blood gas tensions are the most likely to suffer REM-related hypoxaemia. Although vital apaity is onsidered a useful prognosti indiator in DMD [17], we found no orrelation between fored vital apaity and the severity of oxygen desaturation overnight. In addition, no relationship was seen between noturnal oxygen desaturation and Prmax refleting the relative preservation of diaphragm funtion during wakefulness in DMD. Daytime hyperapnia an be predited in patients with pure respiratory musle weakness when the respiratory musle strength (mean of PEmax and Prmax) is <3% of normal or when the vital apaity is <55% of the predited value [18], and in this study the two boys with daytime hyperapnia were those with the most severely impaired lung volumes and respiratory musle strength. REM-related hypoxaemia is assoiated with a transient inrease in pulmonary arterial pressure and may ontribute to the development of or pumonale. In addition, in patients with a ardiomyopathy, the stress of episodi hypoxaemia might inrease the risk of developing myoardial ishaemia and/or arrhythmias. No signifiant ST segment depression was seen in any boy in this study despite the presene of noturnal hypoxaemia. It is unlikely, therefore, that the arrhythmias seen were primarily related to myoardial ishaemia. Although serious arrhythmias ourred in the three boys with the most severe desaturation overnight, in only one ase was the arrhythmia assoiated with the nadir of oxygen saturation. Overnight oxygenation an be improved by the use of noturnal mehanial ventilatory support whih is now in widespread use for a variety of neuromusular disorders [19], and indeed one group advoates the use of intermittent positive pressure ventilation (IPPV) via a traheostomy in boys with DMD with an FVC of <6 ml or evidene of arbon dioxide retention before the onset of overt respiratory failure [2]. The use of intermittent positive pressure ventilation delivered by a nasal mask (NIPPY) is an effetive form of noturnal ventilatory support [21, 22] and is superior to the use of negative pressure ventilation [23] or to the delivery of IPPV via a traheostomy. In summary, we found that domiiliary oximetry adequately detets noturnal oxygen desaturation in boys with DMD, some of whom were also found to have potentially dangerous arrhythmias overnight (but not during the daytime). Studies using NIPPY to abolish noturnal hypoxaemia in suh patients are underway and the effet of this treatment on daytime ardiorespiratory funtion and on ardia rhythm are awaited. Suh treatment is diffiult pratially and requires ethial onsideration. For both of these reasons symptomati and objetive evidene of improvement would be required before ventilatory support ould be reommended for these patients. Referenes 1. Newsom-Davis J. - The respiratory system in musular dystrophy. Br Med Bull, 198, 36, Inkley SR, Oldeburg FC, Vignos PJ Jr. - Pulmonary funtion in Duhenne musular dystrophy related to stage of disease. Am J Med, 1974, 56, Kurz LT, Mubarak SJ, Shultz P, Park SM, Leah J. - Correlation of soliosis and pulmonary funtion in Duhenne musular dystrophy. J Pediatr Orthop, 1983, 3, Smith PEM, Calverley PMA, Edwards RHT, Evans GA, Campbell EJM. - Pratial problems in the respiratory are of patients with musular dystrophy. N Engl J Med, 1987, 316, Curran FJ. - Night ventilation by body respirators for patients in hroni respiratory failure due to late stage Duhenne musular dystrophy. Arh Phys Med Rehabil, 1981, 62, Rideau Y, Gatin G, Bah J, Gines G. - Prolongation of life in Duhenne's musular dystrophy. Ata Neural (Napoli), 1983, 5, Smith PEM, Calverley PMA, Edwards RHT. Hypoxemia during sleep in Duhenne musular dystrophy. Am Rev Respir Dis, 1988, 137, Manning GW, Cropp GJ. - The eletroardiogram in progressive musular dystrophy. Br Heart J, 1958, 2, Perloff JK, De Leon AC, O'Doherty D. - Cardiomyopathy of progressive musular dystrophy. Cirulation 1969, 33, Nigra G, Comi LI, Politano L, Bain RJI. - The inidene and evolution of ardiomyopathy in Duhenne musular dystrophy. Int J Cardiology, 199, 26, Rehtshaffen A, Kales A - A manual of standardised terminology, tehniques and soring systems for sleep stages. US Government Printing Offie, Washington DC, (Publi Health Servie), Luque ER. - Segmental spinal instrumentation for orretion of soliosis. Clin Orthop, 1982, 163, Blak LF, Hyatt RE. - Maximum stati respiratory pressures in generalised neuromusular disease. Am Rev Respir Dis 1971, 13, Quyyumi AA, Crake T, Wright C, Mokus L, Fox K. - The role of ambulatory ST segment monitoring in the diagnosis of oronary artery disease: omparison with exerise testing and thallium sintigraphy. Eur Heart J, 1987, 8, Smith PEM, Edwards RHT, Calverley PMA. - Ventilation and breathing pattern during sleep in Duhenne musular dystrophy. Chest, 1989, 96, Wilson SH, Cooke NT, Edwards RHT, Spiro SG. - Predited normal values for maximal respiratory pressures in auasian adults and hildren. Thorax, 1984, 39, Rideau Y, Jankowski LW, Grellet J. - Respiratory funtion in the musular dystrophies. Musle Nerve, 1981, 4, Braun NMT, Arora NS, Rohester DF. - Respiratory musle and pulmonary funtion in polymyositis and other proximal myopathies. Thorax, 1983, 38, Wiers PWJ, Le Coultre R, Dallinga OT, Van Dijl W, Meinesz AF, Sluiter HJ. - Cuirass respirator treatment of hroni respiratory failure in solioti patients. Thorax, 1977, 32, Gilgoff I, Prentie W, Baydur A. - Patient and family partiipation in the management of respiratory failure in Duhenne 's musular dystrophy. Chest 1989, 95, Kerby GR, Mayer LS, Pingleton SK. Noturnal positive pressure ventilation via nasal mask. Am Rev Respir Dis, 1987, 135, Bah JR, Alba A, Mosher R, Delaubier A - Intermittent

7 44 N. CARROLL ET AL. positive pressure ventilation via nasal aess in the management of respiratory insuffiieny. Chest, 1987, 92, Ellis ER, Bye PTP, Bruderer JW, Sullivan CE. - Treatment of respiratory failure during sleep in patients with neuromusular disease. Am Rev Respir Dis, 1987, 135, Investigation au domiile des hypoxemies en relation ave le sommeil, dans la dystrophie musulaire de Duhenne. N. Carroll, R.J.J. Bain, P.E.M. Smith, S. Saltissi, R.H.T. Edwards, P.M.A. Calverley. RESUME: Dans la dystrophie musulaire de Duhenne (DMD), les desaturations noturnes en oxygene se developpent pendant le sommeil a mouvement oulaire rapide (REM). La polysomnographie, qui exige une admission hospitaliere, permet de deteter les anomalies respiratoires en rapport ave le sommeil. Pour eviter les inonvenients de!'admission hospitaliere hez des patients handiapes, nous avons investigue!'oxygenation noturne hez 1 enfants atteints de DMD, grae a une oxymetrie realisee au domiile. Chez quatre gar;ons les resultats de l'oxymetrie ont ete ompares a eux obtenus pendant l'enregistrement polysomnographique. La "reprodutibilite" de l'oxymetrie au domiile a ete appreiee hez six gar;ons par!'exeution de mesures oxymetriques lors de 2 nuits onseutives. Des gar;ons plus ages, atteints de DMD, peuvent developper une ardiomyopathie. Afin d'appreier le rythme ardiaque et les modifiations du segment ST, nous avons pratique un monitoring au Holter et une oxymetrie simultanes hez sept gar;ons atteints d'hypoxemie noturne. Six des dix gar;ons repris initialement avaient des episodes d'hypoxemie noturne, et!'on a note hez eux une forte orrelation entre la saturation oxygenee minimale noturne et les tensions arterielles des gaz du sang diurnes (Pao 2 r=.89; Pao r=-.87). Malgre une duree de REM adequate au ours de Ii polysomnographie, des desaturations oxygenees plus importantes ont ete observees au ours de l'oxymetrie a domiile. L'on n'a trouve auune differene dans la severite de la desaturation enregistree hez les gar;ons qui ont ete'examines a deux oasions separees. Chez inq gar;ons,!'on a note des variations marquees du rythme ardiaque au ours des episodes hypoxemiques, et des arythmies plus serieuses ont ete observees pendant la nuit hez les trois gar;ons les plus hypoxemiques. L'oxymetrie au domiile est une methode simple et reprodutible pour appreier!'oxygenation noturne, et donne des valeurs omparables a elles de la polysomnographie. Chez les gar;ons atteints de DMD avanee et d'hypoxemie noturne severe, un monitoring eletroardiographique de 24 h. peut deeler des arythmies potentiellement letales. Eur Respir J., ,