Archives ofDisease in Childhood 1996; 74: 195-200

Sleep studies and supportive ventilatory treatmentin patients with congenital muscle disorders

Y Khan, J Z Heckmatt, V Dubowitz

AbstractEight ambulant children aged 6-13 years,four with congenital myopathy, two withcongenital muscular dystrophy and twowith the rigid spine syndrome, presentedwith recurrent chest infections, morningheadaches, shallow breathing at night,or respiratory failure. Polysomnographyconfirmed the presence of nocturnalhypoxaemia with oxygen saturation onaverage less than 90% for 49% ofsleep andless than 80% for 19% of sleep accompa-nied with severe hypoventilation. Addi-tionally there was sleep disturbancecharacterised by an increased number ofwake epochs from deep sleep (in compari-son to 10 non-hypoxaemic subjects). Theseverity of sleep hypoxaemia did not cor-relate with symptoms. Treatment withnight time nasal ventilation was startedand repeat polysomnography showednormal overnight oxygen saturation and areduced number of wake epochs duringdeep sleep. It is important to be vigilantfor sleep hypoventilation in these patientsand sleep studies should be part of theroutine respiratory evaluation. Treatmentwith nasal ventilation is effective inreversing the nocturnal respiratory failurewithout significant disturbance to lifestyle.(Arch Dis Child 1996; 74: 195-200)

Keywords: myopathy, polysomnography, ventilation.

Neuromuscular Unit,Department ofPaediatrics,HammersmithHospital, LondonY KhanJ Z HeckmattV Dubowitz

Correspondence to:Dr Y Khan, ChaileyHeritage, North Chailey,Near Lewes, East SussexBN8 4EF.

Accepted 29 September

Sleep hypoxaemia associated with chronicrespiratory failure is a recognised complicationof congenital muscle disorders.1-5 The risk ofrespiratory problems varies and is high in con-genital muscular dystrophy and certain typesof congenital myopathies. Treatment is partic-ularly worthwhile when the hypoxaemia occursin association with mild physical disability.3The recent introduction of non-invasive nighttime intermittent positive pressure ventilationby nasal mask (nasal ventilation) allows mucheasier management than was previouslypossible with the cuirass, iron lung, or trache-otomy. There is less disturbance of lifestyle asthe equipment is compact, portable, relativelyquiet, and easy to use.5-'1 Despite the advan-tages of early diagnosis there have been fewdetailed polysomnographic studies in thesepatients and there is little information on thepathophysiology and diagnostic features of thehypoxaemia.

In this paper we describe the effect oftreatment on a series of children with con-genital myopathy who presented with sleep

hypoventilation. We describe an approach todiagnosis that allows early detection and man-agement of these patients and study of theeffects of ventilation on sleep, respiratory func-tion, and patient well being.

MethodsPATIENTSWe studied eight children (four girls and fourboys) who presented to the muscle clinic at theHammersmith Hospital (table 1). The mean(range) age at presentation was 10 (6-14)years. Despite having a severe respiratoryproblem, general disability was mild and all thepatients were ambulant.

Their clinical presentation was as follows(table 1). Two sisters with congenital musculardystrophy (patients 1 and 2) were reported tohave shallow breathing and cyanosis at night,but without symptoms or history of chestinfection. The other six patients had a historyof recurrent chest infection (usually requiringhospital admission), four of these had morningheadaches, three required supplementaryoxygen, and two were on treatment for corpulmonale. One (patient 6) was referred forroutine evaluation of respiratory function inpreparation for scoliosis surgery without abreathing problem being suspected. Theyoungest patient (5), who had nemalinemyopathy, presented in acute respiratoryfailure during a chest infection. She had notpreviously had respiratory decompensationand her daytime gases were normal. Previoussleep studies had shown increasing sleephypoxaemia, but up until that time she hadbeen too young to tolerate nasal ventilationand cuirass ventilation was not successful.In general, the importance of the patients'symptoms and the underlying respiratoryproblems were underestimated.The main indication for treatment was symp-

toms and unequivocal evidence of hypoxaemiaeither on the basis of daytime blood gases orsleep study. The most important symptomswere chest infections and morning headaches.All the patients reported daytime lethargy andfatigue and only one patient (6) was able toattend full time education. Three others(patients 4, 7, and 8) attended school part timebut the remaining five received home tuition. Inthe eight patients there had been a total of nineprolonged admissions for chest infection duringthe preceding year, which in one patient lead toacute respiratory failure requiring ventilation.There were also symptoms ofpoor appetite andpoor weight gain. Four patients were below the3rd centile for weight, three were on the 10thcentile, and one was on the 25th.

195

on August 18, 2021 by guest. P

rotected by copyright.http://adc.bm

j.com/

Arch D

is Child: first published as 10.1136/adc.74.3.195 on 1 M

arch 1996. Dow

nloaded from

Khan, Heckmatt, Dubowitz

Table 1 Diagnostic features and effect of night time nasal ventilation

Patient No

1 2 3 4 5 6 7 8

Sex F F M M F F M MAge (years) 13 7 9 11 6 12 14 10Ambulant + + + + + + + +Diagnosis* CMD CMD M N N MC$ RS RSUnderbreathing reported + + - - - + +Cyanosis reported + + + - - - + +Chest infections - - + + + + + +Morning headaches - - + + + + +Acute respiratory failure - - + - +Cor pulmonale - - + - - - +Daytime oxygen - - + - - + +Night time oxygen - - + - - + +Scoliosist + + + + + + + + + + -Spinalfusion + - NP - - + +Abdominal squeeze + - + - - + + +Abdominal paradox + + - - - + +Vital capacity (% expected) 27 56 17 50 - 40 45 17Daytime arterial oxygen tension (kPa)t

Before ventilation 8-6 12-5 6-9 12 11 8-6 11-3§ 14-5§After ventilation 11-6 12-0 10-6 12-5 12-5 11-3 12-0 12-5

Daytime arterial carbon dioxide tension (kPa)tBefore ventilation 8-5 5-1 5-6 6-6 6 4-1 7-3 6-8After ventilation 6-7 5-0 5-3 5 0 5-1 4-5 6-2 5

Oxygen saturation <90% as % of total sleepOn diagnostic sleep study 99 15 40 14 20 100 66 45After nasal ventilation 0 0 0 0 0 2 0 0

*CMD=congenital muscular dystrophy, M=minicore myopathy, N=nemaline myopathy, MC='minimal change' myopathy,RS=rigid spine syndrome with dystrophic change on needle muscle biopsy.t+Cobb angle up to 400, + +Cobb angle >40°. NP=surgery for scoliosis was not possible.tValues at diagnosis of nocturnal hypoventilation before treatment and after regular ventilation established.$Supplementary oxygen at the time of the test.$This patient was referred for routine respiratory evaluation before scoliosis surgery.

Six patients had abnormal respiratory mus-cle activity, either abdominal paradox whenlying supine or an abdominal 'squeeze'manoeuvre in the erect position, that is, con-

traction of the abdominal muscles in expira-tion,3 or both. Seven had scoliosis, severe inthree, three of whom received subsequent sur-

gical stabilisation.12 Two patients had the rigidspine syndrome in association with a histologi-cal picture of muscular dystrophy on biopsy.All patients had daytime arterial blood gaseswhile at rest breathing air (except where sup-plementary oxygen was required as specified intable 1). This was repeated as necessary onceventilation was established. Vital capacity was

measured with a portable spirometer (Micro-medical, Rochester, Kent) and compared withnormal values.13

SLEEP STUDIESSleep studies were carried out on a relativelyquiet section of the open ward. Our practicewas to use overnight oximetry as an initialscreen, followed by non-invasive polysomnog-raphy (diagnostic sleep study). We used theOxford Medilog Multiparameter Recorder(Oxford Medical Ltd) which recorded eightchannels of information: (1) electroencephalo-gram (EEG) channel C4-A1, (2 and 3)bilateral electro-oculogram, (4) submentalelectromyogram, (5 and 6) thoracoabdominalmovements by inductance plethysmography('Respibands', Ambulatory Monitoring Inc,Ardsley, NY), (7) air flow by an oronasal ther-mistor, and (8) arterial oxygen saturation(Ohmeda Biox 3700). Although patients hadusually had one night of oximetry in thehospital before the sleep study, there was no

acclimatisation night to the polysomnography.The recorder was small, portable, and did notdisrupt ward routine. The oximetry data

reported was from the polysomnography night.We replayed the recordings on a comput-

erised Medilog 9000-111 system. We analysedrespiration during hypoxaemic episodes fromprintouts via a modified eight channel EEGMingograph. The sleep recordings andoximetry were automatically analysed usingthe Medilog (SS-90 software version 4.82)sleep stager and also visually analysed.For our analysis we divided sleep into 'light

sleep' (stages 1 and 2), 'deep sleep' (stages 3and 4), and rapid eye movement (REM)sleep.14 We analysed measures indicative ofsleep quality, namely: sleep efficiency, dura-tion ofREM sleep, and the period spent awakeafter the initial onset of sleep.14 15 The latterwas further subdivided into the number oftimes the patient had a 30 second or greaterawake epoch after each stage of sleep (wakeshifts).16 A 30 second awake epoch wasdefined as the development of alpha and/or lowvoltage mixed frequency activity on theEEG.17 Sleep efficiency was defined as the per-centage of the actual sleep time divided by thetotal recording time.15 Sleep architecture wasdefined as the cyclical alteration of REM andnon-REM sleep throughout the night.18 Theduration of hypoxaemia in deep and REMsleep were quantified from the hypnogram.The detailed respiratory data allowed breath

to breath analysis of airflow, chest and abdom-inal movements. Although these data were notquantifiable, a diminution of breath or airflowamplitude was taken as a reduction in respira-tion. We classified apnoeas and periods ofhypoventilation lasting longer than 10 secondsaccording to the criteria of Cherniak.19

ARTIFICIAL VENTILATIONAfter the diagnostic sleep study the patientswere started on night time nasal ventilation

196

on August 18, 2021 by guest. P

rotected by copyright.http://adc.bm

j.com/

Arch D

is Child: first published as 10.1136/adc.74.3.195 on 1 M

arch 1996. Dow

nloaded from

Sleep studies and supportive ventilatory treatment in patients with congenital muscle disorders

breathing air. They used the BiPAP S/T venti-lator (Respironics, USA) - that is, biphasicpositive airways pressure. The ventilator islightweight, compact, and portable withadjustable voltage power supply.

Patients were established on the spontane-ous/timed mode with inspiratory pressuresvarying from 10-14 cm H20, expiratoryat atmospheric, and a respiratory rate of12-1 8/min. This allowed triggering of theventilator while the patient was awake andmandatory ventilation when asleep if thepatient failed to breathe spontaneously.Delivery was via a comfortable silicon nasalmask (Medic-Aid Ltd, Sussex) or alternatelythe Adams' circuit with nasal pillows (PuritanBennett, Hounslow). The patients adapted tothe ventilator during the day for short periodsand then used it regularly overnight before dis-charge home. Parents were taught ventilatormanagement (for example adjustment of con-trols and cleaning) before discharge. Theimmediate response to ventilation was assessedby overnight oximetry. We repeated poly-somnography after an average of eight months(range 1-24 months) to determine the effec-tiveness of ventilation but without recordingoronasal airflow because of the nasal mask.Although two patients slept with supplementaloxygen at presentation this was not required onventilation.

FOLLOW UPIn addition to their routine follow up, patientswere contacted personally by the authors 2-6years after treatment to determine clinicalprogress and ventilator usage. Additionally, theauthors perused the latest clinical reports.

SLEEP CONTROLSThe purpose of studying sleep controls was toestablish the effect, if any, of hospitalisation onsleep. The controls were therefore studied onthe ward under exactly the same circumstancesas the hypoxaemic patients. We chose controlswho were unlikely to have sleep hypoxaemia.Ten patients (four boys, six girls) attending theHammersmith Hospital had identical poly-somnographic studies on the first or secondnight of admission. Five had a neuromuscular

Oxygen saturation(%)

100

0IOronasalairflowChest

Abdomen

10 secondsFigure 1 Printout showing an episode of hypoventilation in association with profoundhypoxaemia in patient 1 (congenital muscular dystrophy). During the episode respiratoryexcursions diminish gradually in amplitude until they are barely detectable. The fall inamplitude is accompanied by a corresponding gradual reduction in airflow. The oxygensaturation fals from 50% to much lower levels during the period of hypoventilation. Thispatient presented because parents had noted shallow breathing and cyanosis at night.

disorder and five a non-neurological disorder.The mean (range) age was 10-8 (6-22) years.

ETHICAL PERMISSIONThis study was approved by the hospital ethicalcommittee. Parents of hypoxaemic patientsand control subjects signed informed consentfor the sleep studies.

STATISTICSDifferences in the various sleep parametersbefore and after introduction of regular nasalventilation were compared by the paired t test.Patients' diagnostic sleep studies were com-pared with controls by means of the Student'st test. Statistical analysis was carried out withthe Minitab computer program.20

ResultsPATIENTSThe diagnostic sleep study showed thatpatients were significantly hypoxaemic, spend-ing on average 49% (range 14-100%) of theirsleep with their oxygen saturation below 900/oand 19% (range 0-78%) of their sleep with itbelow 80%. Nasal ventilation was extremelyeffective in correcting the sleep hypoxaemiawith only one patient (6) showing any fall inthe oxygen saturation below 90% (table 1).There was no relationship between the severityof the hypoxaemia and the vital capacity, andtwo patients had a vital capacity of 50% ofexpected or more. Daytime blood gasesrepeated after the introduction of night timeventilation showed resolution of hypoxia in allpatients and hypercarbia in five.The primary respiratory abnormality was

hypoventilation (fig 1). On observation of thepolysomnographic tracings, the patients didnot usually waken during periods of shallowbreathing, despite an associated fall in theoxygen saturation.These periods of hypoventilation occurred

repeatedly, could last up to two minutes, andwere often complete apnoea. They occurredduring deep and REM sleep in all patientsexcept one. The one exception was ouryoungest patient who had the least amount ofREM sleep (patient 5) and was only hypox-aemic during deep sleep. On average 59%(range 25-80%) of the hypoxaemic time wasduring deep sleep and 33% (range 0-75%)during REM sleep, the proportions of thesetwo phases of sleep that were hypoxaemicbeing 55% and 69% respectively. Two patients(1 and 7) were continuously hypoxaemicduring all phases of sleep. At referral, twopatients (7 and 8) were on overnight oxygen bynasal cannula which did not correct thehypoventilation episodes nor the daytimehypoxaemia and hypercapnia.

Patients tolerated night time ventilationwell. The ventilator was quiet and convenientand five patients have travelled abroad with it.The mean duration of ventilation at follow uphas been 3-5 (range 2-6) years. Air swallowingfrom the nasal mask has not been a prdblem

197

on August 18, 2021 by guest. P

rotected by copyright.http://adc.bm

j.com/

Arch D

is Child: first published as 10.1136/adc.74.3.195 on 1 M

arch 1996. Dow

nloaded from

Khan, Heckmatt, Dubowitz

Table 2 Sleep parameters; values are mean (range)

Before After pventilation ventilation Valuet Controls

Sleep efficiency (O/o) 91 (68-98) 92 (80-100) NS 91 (73-99)Light sleep (O/o) 15 (2-28) 26 (10-57) NS 38 (14-76)Deep sleep (%) 65 (37-98)* 43 (14-72) 0 04 38 (3-70)*REM sleep (/) 20 (0-54) 30 (14-61) NS 18 (1-74)Awake (%/o) 8 (1-24) 6 (<1-19) NS 5 (0-12)Total wake shifts in deep sleep 5 (3-8)*** < 1 (0-2) 0 0003 1-2 (0 3)***

tComparison of diagnostic sleep study with repeat study on ventilation by paired t test.*p<0-02, ***p<0-001: on Students t test comparing patient's initial diagnostic study with controls.

with maximum airway pressure below 15 mmH20. We found that maximum airway pres-sure tended to rise during the night as thepatients no longer triggered the ventilator andit was important to adjust the controls accord-ingly. Pressure sores over the bridge ofthe nosecaused some discomfort, and were resolved byalternating the nasal mask with the Adams' cir-cuit nasal pillows. All our patients were able toput on the mask at night independently andcould operate the ventilator themselves withlittle assistance from parents.

Repeat polysomnography on ventilationshowed good respiratory excursions throughoutthe night and periods ofspontaneous respirationinterspersed with periods of complete relianceon mechanical ventilation. In two patients themask was dislodged during deep and REMsleep. This was clearly evident by immediatehypoventilation and oxygen desaturation butthe patients did not wake (the BiPAP ventilatordid not have an alarm system).

All patients reported improved sleep qualityon ventilation and resolution of symptoms,such that all would prefer to sleep with the ven-tilator. Antifailure medication and supplemen-tary oxygen were discontinued. Three patients(1, 6, and 7) have undergone a general anaes-.thetic for major scoliosis surgery without anycomplications and have been weaned backonto regular night time nasal ventilation. Theprogressive scoliosis in the boy with minicoremyopathy (3) was too advanced for surgery.

Sleep stageWake

REM1234

Oxygen saturation

100

8030

15 minutesFigure 2 Section ofhypnogram, 2 to 4 am, from diagnostic sleep study in patient 3(minicore disease). Initially he is in stage 4 sleep with oxygen saturation at 90%. This isfollowed by a period ofprofound hypoxaemia (oxygen saturation 80% to <30%o) withchanges in sleep stages without awakening. After more than 30 minutes ofhypoxaemiathere are awake epochs (arrow) from stage 3 and 4 sleep (deep sleep).

Later follow up confirmed that all patientswere continuing to use the ventilator everynight. Seven out of eight patients had startedfull time education. Symptoms of morningheadaches and poor appetite had resolved andall patients had gained weight. Four patientswere on the 50th centile for weight, two patientswere on the 25th centile, and two on the 3rdcentile. In the years after ventilation there havebeen no admissions for chest infections.

Sleep architecture was normally preservedboth before treatment and subsequently.Hypoxaemic patients had significantly moredeep sleep than controls and this fell on venti-lation. They also had significantly more wakeshifts in deep sleep, than did controls (table 2).The longest duration of these wake shifts wassix minutes. The number of wake shifts fellsignificantly on nasal ventilation but this fallwas not related to the fall in amount of deepsleep on ventilation (table 2). Neither theamount of deep sleep not the number of wakeshifts in deep sleep related to the severity of thehypoxaemia. The hypnogram in three patientsshowed deep sleep wake epochs during hypox-aemia, but other patients did not show thisrelationship (fig 2).

SLEEP CONTROLSIn the sleep control subjects the percentage oftime spent in REM sleep, sleep efficiency, andthe amount of time spent awake after the onsetof sleep were comparable with age matchednormal data.15 21 None had hypoxaemia(oxygen saturation <90%) on their poly-somnographic night.

DiscussionAlthough the occurrence of sleep hypoventila-tion in congenital muscle disorders is wellrecognised,'-5 there is probably little awarenessof the frequency of respiratory failure in mildlydisabled patients and the effectiveness of nighttime nasal ventilation.3 6 10 11 This series ofeight patients demonstrates the need for agreater awareness of abnormal sleep relatedbreathing patterns. The patients presentedwith recurrent respiratory infections, poorappetite and weight gain, and a history ofheadaches or acute-on-chronic respiratoryfailure, and it seemed likely that sleephypoventilation had been a problem for asignificant period beforehand. It is noteworthythat two referrals were based on parents' obser-vations while the patients themselves wereunaware of any breathing difficulties. Ingeneral the significance of symptoms was notfully appreciated, except retrospectively onceventilation was started.

Failure to thrive has been previously associ-ated with sleep hypoxaemia due to tonsillarobstruction and, as in our patients, resolutionof hypoxaemia was associated with weightgain.22 Ideally sleep hypoxaemia should beidentified at an early stage, that is before theoccurrence of acute decompensation duringintercurrent infection, the development ofchronic respiratory failure, or deterioration

198

on August 18, 2021 by guest. P

rotected by copyright.http://adc.bm

j.com/

Arch D

is Child: first published as 10.1136/adc.74.3.195 on 1 M

arch 1996. Dow

nloaded from

Sleep studies and supportive ventilatoty treatment in patients with congenital muscle disorders

of respiratory function associated with pro-gressive scoliosis. Scoliosis is a frequentlyassociated feature reflecting the axial distribu-tion of muscle weakness and it is essential todiagnose any respiratory problems beforescoliosis surgery.The risk of acute respiratory failure does not

correlate with the severity of night time hypox-aemia. This is not surprising as acute respiratoryfailure is usually precipitated by infection andfactors relating to infection such as change inupper airways resistance and lung compliance.Nevertheless, we have not yet seen acute res-piratory failure without some suggestion ofpriorsleep hypoventilation. There is also a poor cor-relation between the severity of the sleep hypox-aemia and the reported symptoms. Theperception ofsymptoms may relate to the age ofthe child and the chronicity of the hypoxaemia.To diagnose nocturnal hypoxaemia at an

early stage requires regular careful clinicalexamination and overnight sleep study whenindicated. Clinical assessment should includeevidence of abnormal respiratory muscleactivity such as abdominal 'squeeze' and para-doxical breathing. The former suggests a smalltidal volume3 and the latter a weak dia-phragm.23 Patients with severe weakness ofneck flexors and a rigid spine are also particu-larly liable to respiratory failure even thoughambulant.3 11 The sleep study may be byoximetry alone but more detailed informationis often useful. Oximetry with respiratorychannels, that is inductance plethysmographyand airflow, overcomes the problem of poorsignal artifacts from a dislodged oximeterprobe, is simple to perform, and does notrequire a great deal of expertise to interpretand is practical even in young children. Eightchannel polysomnography, if available, is moredefinitive, and allows good comparativesequential analysis and planned intervention.We believe that sequential sleep studies, what-ever the method, should be routinely done inambulant children with certain neuromusculardiseases - namely minicore, nemaline and'minimal change' myopathies, congenital mus-cular dystrophy, and the rigid spine syndrome.Such studies should be considered as an essen-tial part of a fully respiratory assessment, par-ticularly as standard lung function tests such asvital capacity do not seem to reliably predictsleep hypoventilation.The decision to provide regular night time

ventilation is balanced between the patient'soverall disability and the additional burden ofthe technique of ventilation.24 In our series, allpatients were ambulant and had relatively mildphysical disability and generally a good prog-nosis.

Nasal ventilation corrected the sleephypoventilation and reversed the progressionof chronic respiratory failure without imposinga major burden. Additionally, ventilationreduced the frequency of debilitating chestinfections and made spinal stabilising surgerypossible in three patients. The age of thepatient is a critical factor as the youngest thatwe and others have successfully established onnasal ventilation has been at 6 years of age.25

Therapeutic options for younger patients arelimited to cuirass or tank ventilation or atracheotomy.26 The underlying myopathiesdescribed in our patients are not generally pro-gressive but there should be some caution inadvocating ventilation when scoliosis is notamenable to surgery.

Sleep was remarkably well preserved despiteprofound hypoxaemia. There was normal pro-gression through the sleep stages, and sleepefficiency and total sleep time were within thenormal range. Most patients had REM sleep.This preservation presumably explains whypatients had so few symptoms. Nevertheless allpatients subjectively noted improved sleep andwell being on ventilation. This is consistentwith previous experience.'0 11 27 28The sleep disturbance at the diagnostic sleep

study was relatively subtle and confined toexcess deep sleep with increased 30 secondwake shifts. It would have been interesting tohave looked at microarousals - that is, twoseconds of alpha rhythm, which have beenreported to be increased in hypoxaemia.29Microarousals are likely to be increased ifwakeshifts of 30 second epochs, reflecting a stage inthe sleep/wake cycle also containing alpharhythm, were increased.'6 Increase number ofwake shifts is a previously documentedmeasure of poor sleep quality.'6 30 31Our patients showed excess deep sleep in

contrast to previous studies of hypoxaemia, inchildren and adults, which have shown deepsleep to be reduced.32 33 Other reports ofhypoxaemia resulting from hypoventilation inneuromuscular disorders show an associationwith deep sleep.35-37 In contrast obstructivesleep apnoea in Duchenne muscular dys-trophy,38 39 and in non-neuromuscular dis-orders,28 29 is associated with, and oftenconfined to, REM sleep. Hypoventilationduring deep sleep probably has a differentpathophysiology from obstructive apnoeasduring REM sleep.

In two patients, the ventilator discon-nected, causing profound hypoxaemia, butthey did not wake. Although hypoxaemia isknown to trigger awakening, the factorscausing this are not clearly understood.28 Inaddition, deep sleep may intrinsically beassociated with an increased awakeningthreshold.34 A ventilator with an alarm systemis an advantage.

In conclusion, we have shown that mildlydisabled children with certain types of muscledisease can have significantly potentially lifethreatening sleep hypoventilation. The signifi-cance of clinical symptoms may not be fullyappreciated or there may be no apparent symp-toms. In view of the chronic debility, and therisk of acute respiratory failure, cor pulmonaleand complications of spinal surgery, early diag-nosis is desirable, if not mandatory. Thereshould be clinical vigilance with a carefulrespiratory history, examination for abnormalrespiratory muscle activity, measurement ofvital capacity, and sequential sleep studies. Asnasal ventilation is an effective and non-invasive mode of treatment the decision tointervene is readily justified.

199

on August 18, 2021 by guest. P

rotected by copyright.http://adc.bm

j.com/

Arch D

is Child: first published as 10.1136/adc.74.3.195 on 1 M

arch 1996. Dow

nloaded from

200 Khan, Heckmatt, Dubowitz

This project was supported by a research grant from theMuscular Dystrophy Group of Great Britain. In addition finan-cial support was received from the Handicapped Children's AidCommittee. The authors are grateful to Dr A Sharply for hercomments and Dr M White for providing additional informa-tion on a patient. We thank Dr Hopp for the sleep study data onventilation on two of the patients, done on similar equipment.

1 Riley DJ, Danielle RP, Edelman NH. Blunted respiratorydrive in congenital myopathy. Am J Med 1977; 63:459-65.

2 Maayan CH, Springer C, Armon Y, Bar-Yishay E, ShapiraY, Godfrey S. Nemaline myopathy as a cause of sleephypoventilation. Pediatrics 1986; 77: 390-5.

3 Heckmatt JZ, Loh L, Dubowitz V. Nocturnal hypoventila-tion in children with non-progressive neuromusculardisease. Pediatrics 1989; 83: 250-5.

4 Bye P, Ellis E, Issa F, Donnelly P, Sullivan C. Respiratoryfailure and sleep in neuromuscular disease. Thorax 1990;45: 241-7.

5 Kawata A, Suga M, Miyamoto K, Hirose K, Tanabe H.Rigid spine and nocturnal alveolar hypoventilation. InternMed 1993; 32: 638-40.

6 Ellis E, Bye P, Bruderer J, Sullivan C. Treatment of respira-tory failure during sleep in patients with neuromusculardisease; positive-pressure ventilation through a nosemask. Am Rev Respir Dis 1987; 135: 148-52.

7 Bach JR, Alba A, Mosher R, Delaubier A. Intermittent posi-tive pressure ventilation via nasal access in the manage-ment of respiratory insufficiency. Chest 1987; 92: 168-70.

8 Bach J, Alba A. Management of chronic alveolar hypoventi-lation by nasal ventilation. Chest 1990; 97: 52-7.

9 Carroll N, Branthwaite M. Control of nocturnal hypoventi-lation by nasal intermittent positive pressure ventilation.Thorax 1988; 43: 349-53.

10 Kerby GR, Mayer LS, Pingleton S. Nocturnal positive pres-sure ventilation via nasal mask. Am Rev Respir Dis 1987;135: 738-40.

11 Heckmatt J, Loh L, Dubowitz V. Night-time nasal ventila-tion in neuromuscular disease. Lancet 1990; 335: 659-82.

12 Luque E. Segmental spinal instrumentation for correctionof scoliosis. Clin Orthop 1982; 163: 192-8.

13 Godfrey S, Kamburoft P, Naim J, et al. Spirometry, lungvolumes and airway resistance in normal children aged5-18 years. British Journal ofDiseases ofthe Chest 1970; 64:15-24.

14 Carskadon M, Dement W. Normal human sleep: anoverview. In: Kryger M, Roth T, Dement W, eds.Principles and practice of sleep medicine. Philadelphia: WBSaunders, 1989: 3-13.

15 Coble P, Kupfer D, Taska L, Kane J. EEG sleep of normalhealthy children. Par 1: Findings using standard measure-ment methods. Sleep 1984; 7: 289-303.

16 Carskadon M, Brown E, Dement W. Sleep fragmentation inthe elderly: relationship to daytime sleep tendency.NeurobiolAging 1982; 3: 321-7.

17 Rechtschaffen A, Kales A. A manual ofstandardised terminol-ogy, techniques, and scoring system for sleep states in humansubjects. Washington DC: US Govemment PrintingOffice, Public Health Service, 1968: 1-12.

18 Baker TL. Introduction to sleep and sleep disorders.Symposium of sleep apnoea disorders. Med Clin North Am1985; 69: 1123-52.

19 Chemiack N. Respiratory dysrhythmias during sleep.NEnglJMed 1981; 305: 325-30.

20 Minitab Inc. Minitab computer programme. Philadelphia:Minitab Inc.

21 Palm L, Persson E, Elmqvist D, Blennow G. Sleep andwakefulness in normal preadolescent children. Sleep 1989;12: 299-308.

22 Stradling JR, Thomas G, Warley ARH, Williams P,Freeland A. Effect of adenotonsillectomy on nocturnalhypoxaemia, sleep disturbance, and symptoms in snoringchildren. Lancet 1990; 335: 249-53.

23 Newsom Davis J, Goldman M, Loh L, Casson M.Diaphragm function and alveolar hypoventilation.QJfMed 1976; 45: 87-100.

24 American Thoracic Society. Home mechanical ventilationof pediatric patients. Am Rev Respir Dis 1990; 141:258-9.

25 Ellis E, Mccauley V, Mellis C, Sullivan C. Treatment ofalveolar hypoventilation in a six-year-old girl with inter-mittent positive pressure ventilation through a nose mask.Am Rev Respir Dis 1987; 136: 188-91.

26 O'Leary J, King R, Leblanc M, Moss R. Cuirass ventilationin childhood neuromuscular disease. J Pediatr 1979; 94:419-21.

27 Sullivan C, Berthon Jones M, Issa F. Reversal of obstructivesleep apnoea by positive airway pressure applied throughthe nares. Lancet 1981; i: 862-5.

28 Stradling J. Disorders of ventilatory control and sleepapnoea syndromes. In: Brewis R, Gibson G, Geddes B,eds. Respiratory medicine. London: Balliere Tindall, 1990:1352-72.

29 Fleetham F, West P, Mezon B, Conway W, Roth T, KrygerM. Sleep, arousals, and oxygen desaturation in chronicobstructive pulmonary disease. Am Rev Respir Dis 1982;126: 429-33.

30 Avital A, Steljes D, Pasterkamp H, Kryger M, Sanchez I,Chemick V. Sleep quality in children with asthma treatedwith theophylline or cromolyn sodium. JT Pediatr 1991;119: 979-84.

31 Levine B, Roehrs T, Stepanski E, Zorick F, Roth T.Fragmenting sleep diminished its recuperative value. Sleep1987; 10: 590-9.

32 Spiegel R. Sleep and sleeplessness in advanced age. In:Weitzman E, ed. Advances in sleep research. Vol 5.Lancaster: MTB Press, 1981: 43-60.

33 Guilleminault C. Obstructive sleep apnoea syndrome inchildren. In: Guilleminault C, ed. Sleep and its disorders inchildren. New York: Raven Press, 1987: 213-24.

34 Kryger M, Steljes D, Yee W-C, Mate E, Smith SA,Mahowald M. Central sleep apnoea in congenital mus-cular dystrophy. J Neurol Neurosurg Psychiatry 1991; 54:710-2.

35 Gilmartin J, Cooper B, Griffiths C, et al. Breathing duringsleep in patients with myotonic dystrophy and non-myotonic respiratory muscle weakness. QJIMed 1991; 78:21-31.

36 Coccagna G, Mantovani M, Parchi C, Mironi F, LugaresiE. Alveolar hypoventilation and hypersomnia in myotonicdystrophy. J Neurol Neurosurg Psychiatry 1975; 38:977-84.

37 Smith P, Calverley P, Edwards R. Hypoxaemia during sleepin Duchenne muscular dystrophy. Am Rev Respir Dis1988; 137: 884-8.

38 Khan Y, Heckmatt J. Obstructive apnoeas in Duchennemuscular dystrophy. Thorax 1994; 49: 157-61.

39 Dagan Y, Lavie P, Bleich A. Elevated awakening thresholdsin sleep stage 3-4 in war-related post-traumatic stress dis-order. Biol Psychiatry 1991; 30: 618-22.

on August 18, 2021 by guest. P

rotected by copyright.http://adc.bm

j.com/

Arch D

is Child: first published as 10.1136/adc.74.3.195 on 1 M

arch 1996. Dow

nloaded from