upper airway resistance syndrome. A cause of ... - Semantic Scholar

A cause of excessive daytime sleepiness. The upper airway resistance syndrome. C Guilleminault, R Stoohs, A Clerk, M Cetel and P Maistros Chest 1993;104;781-787 The online version of this article, along with updated information and services can be found online on the World Wide Web at: http://chestjournal.chestpubs.org/content/104/3/781

Chest is the official journal of the American College of Chest Physicians. It has been published monthly since 1935. Copyright1993by the American College of Chest Physicians, 3300 Dundee Road, Northbrook, IL 60062. All rights reserved. No part of this article or PDF may be reproduced or distributed without the prior written permission of the copyright holder. (http://chestjournal.chestpubs.org/site/misc/reprints.xhtml) ISSN:0012-3692

Downloaded from chestjournal.chestpubs.org by guest on May 16, 2012 © 1993 American College of Chest Physicians

A Cause

of Excessive

The Upper

Airway

Guilisminault,

Christian Mindy

Cetel,

M.D.;

Daytime

Subjects with isolated complaints of chronic mess are usually classified as “idiopathic and treated investigated

SyinptOmatically.

during

alpha short

arousals

the

sleep

daytime sleephypersomniacs”

of these subjects and daytime naps.

was In a

was fragmented by very short threughout the sleeping period. These usually ignored in sleep analyses, but

are

is significant

(in the

the

sleep

syndrome,

M.D.;

sleep

subgroup

their impact

A group

nocturnal

of them, EEC arousals

Syndrome

Resistance

M.D.; Pticcardo Stoohs IbuL Maistro M.D.

and

mean

15 subjects

identified

with

in multiple

latency

sleep

latency tests was 5.1 ± 1 mm). These arousals are directly related to an abnormal increase in respiratory efforts during sleep (the mean peak inspiratory esophageal pressure measured

in

ca.lly,

an arousal

limitation

tidal

subjects

occurs

Snaring

was

xcessive

(is,

in

arousal

associated

volume during

tance

E

onr

a transient

preceding

with

increase

The arousal

noted

in

daytime

respiratory

-33±7

daytime

in upper airway resisnormal breathing.

association

with

sleepiness

somnolence

percent

of the

tri’

In

general

many

these

is responsible

transient

for many

and constitutes socioeconomic have indicated

between

3.7

and

in Western

the

comment

daytime somnolence are usually diagnosed nia,” they

affects population

subjects,

For editorial

just

Typi-

restores

driving and industrial accidents a major health problem with a serious impact. Epidemiologic investigations that

cycle

cm 11,0).

Alex Clerk,

M.D.;

arousals

in

10 of the

neither

sufficient

clinical

syndrome.

the

affected

15 subjects;

nor

necessary

Both

gnip.

All

however,

for

sexes

were

studied

cause

of

see page

4.2

counexcessive

665

a label that purely describes the are given very limited treatment.

Obstructive sleep apnea cause of daytime somnolence with snoring. In the recent

symptoms,

and

syndrome

is a well-known is often associated research has focused

that past,

equally

of the in

represented

had

upper

airway

anatomy that was mildly abnormal. Nasal continuous posifive airway pressure, used as an experimental tool, eliminated the daytime sleepiness (multiple sleep latency mean score= 13.5 mm), the transient arousals (mean alpha EEC arousal index decreased from 31.3±12.4 to 8±2 per hour of sleep), and the abnormal upper airway resistance. Chronic daytime sleepiness is a major cause of social, economic, and medical impairment. Recognition of this syndrome

and

its cause

can be developed

is important,

to eliminate

as specific

CPAP=antinuoss positive airway sleep latency test; Pesesophageal

1993;

pressure

pattern

during

sleep

transient

arousals

noted.

A prospective

performed ferred to

the

Clinic. Idiopathic

and

causes sleep

the was

on subjects reSleep Disorders

is a diagnosis

evolution

In our study sleepiness

within 18 months have been present.

a nocturnal

for study

based

on an

ofexcessive daytime sleepiness that for more than 1 year, with insidious

progressive

of drowsiness. causes ofdaytime trauma could

period University

hypersomnia

isolated complaint has been present

responsible

a 6-month

Stanford

104:781-87)

MSLTmukiple

pressure

breathing

during

treatments

the problem.

(Chest

onset

is undetermined. These subjects as having “idiopathic hypersom-

was

snoring

the identification

subjects

within one to three breaths of flow abrupt but limited reduction in

abnormal

sleep).

the

was

Sleepiness*

polygraphic

of sleep latency test

toward

daily

periods

group, none of the known (including history of head of onset of the complaint) All subjects must have had

recording

disruption,

investigating

followed

(MSLT).9

The

latter

known

by a multiple consists of five

on chronic loud snoring as a possible indication of a health problem. Preliminary data have indicated that some snorers may be sleepy during the daytime.6’7 Their sleepiness appears to be related to very short arousals that occur during snoring.

20-min nap trials distributed throughout the day and tests for abnormally short sleep latencies and abnormal sleep pattern (le, early appearance of REM sleep).

We questioned ignored) arousals

15 S to be scored.’#{176}

subjects who niacs” might

whether the very short (and often seen in the nocturnal recordings of

have been labeled “idiopathic hypersombe responsible for their daytime somno-

lence. We also investigated

whether

there

was an abnormal

*From the Stanford University Sleep Disorders Clinic and Research Center, Stanford University School ofMedicine, StanfOrd, CaliL Supported by grant AC-07772 ofthe National Institute on Aging. Manuscript received September 4, 1992; revision accepted January 7, 1993.

Classically, 30-s epochs,

the polygraphic and arousals

tests are scored by 20- or from sleep must last at least

METHODS

initial

Screening(Nlght

1)

Men and women 18 years and older referred to the clinic for a complaint of excessive daytime sleepiness were systematically investigated during a 6-month period. Each subject was asked to fill out a standardized and validated sleep/wake questionnaire, the Sleep Questionnaire and Assessment ofWakefulness.”A complete medical evaluation was performed, including a clinical interview investigating medical, neurologic, psychiatric, and sleep-wake his-

cHEST

I 104 I 3 I SEPTEMBER,


1993

781

tory.

was reviewed, and all subjects were withdrawn of 15 days from therapy with any psychoactive or recreational drugs that might affect sleep or sleep’wake cycles. Urine drug screens were performed after the 15-day period. Schedules were structured to maximize and regulate nocturnal sleep for 8 days prior to the polygraphic record1ngs.’ Eh subject underwent a nocturnal polygraphic monitoring with lights-out time set at 10:30 PM and lights-on time at 7 AM (se, 844 hours of dark Drug

intake

for a minimum

lime).

The

monitoring

international

Included

electrode

EEC

placement

(fA2,

CNAI,

system);

O2’A1

of the

electro-oculogram

(EOC); chin and leg electromyogram (EMC); and ECC (modified V-2 lea4 Respiralion was investigated by oronasal airflow, thoracic and abdominal movements (inductive plethysmography), snoring sounds (subminiature electric microphone type MCE-2000 [MESAM-4 equipment, Conrad ElectrOnics, Hirchau, Germany] taped above the larynx), and oxygen saturation (pulse oximetry). Records were scored following the Bechtschaffen and Kalesbo international a1terla for sleep’wake determination, and the published international eriterla for scoring sleep-onset REM periods, periodic leg movements, restless legs, etc,’ were used for determinalion of speclficsleep-related syndromes. Abnormal breathingpatterns were sawed using the current criteria for identifying sleep apnea and

sleep

hypopnea.’”

The morningafterthe nocturnal polygraphic monitoring, a MSLT was performed, with five nap? SchedUled at 9:30 and 11:30 and 1:30, 3:30, and 5:30 PM. This initial screening identified subjects with emessive daytime somnolence and Indicated well-known syndromes (narcolepsy, obstructive sleep apnea, restless leg syndrome, ate) as causes ofthe somnolence in part of the population. Those patients who did not fit any o(these well-defined syndromes were the subject offurther systematic investigations:

Night

2

Ambulatory monitoringwith adigital portable recorder (MESAM 4, Madaus Inc)’” continuously recorded heart rate, body position, and snoring sounds through an electric subminiature microphone. (flils equipment performs spectum analysis ofbreathlng noises and gtvesbreath-by-breath Inforinationon snoring, with scoringon three levels of loudness.”) Pulse oximetry was also performed. This ambulatory test was performed to investigate the presence of snoring and abnormal breathing patterns in the home environment. Night

Inc).

With the data from these four nights we identified the patients with a daytime sleepiness complaint, abnormal multiple sleep latency scores, more than ten transient EEC arousals per hour of sleep, and an abnormal breathing pattern with increased respiratory eflbrts (Pbs monitoring) and increased upper airway resistance (pneumotachometer monitoringJ. Data analysis used to identify this subgroup is described in “Data Analysis and Selection of Subgroup for the Nasal CPAP Therapeutic TriaI’ In the “positive” subgroup, a therapeutic trial with nasal contintious positive airway pressure (CPAP) was performed, with titration of the necessary pressure. The methods used to calibrate nasal CPAP were as follows: as none of the subjects presented with a significant number of sleep apneas or hypopneas as classically deflned,’” nasal CPAP titration was primarily based on measuremeat of Pbs. The CPAP pressure was set at the point at which Pea nadirs were equal to or less negative than 1 SD below the mean Pea nadir measured during 30 min of quiet supine wakefulness. As is customary in our clinic, two nights of recordings were performed for CPAP titration. Patients were prescribed continuous nightly use O(CPAP for a minimum of3 weeks. They then returned for clinical evaluation and nocturnal polygraphy with nasal CPAP and a following-day multiple sleep latency test. This last polygraphic monitoring was similar to the initial one, ie, only sensors placed on the body were used, and Ps was not monitored. The schedule of the different recordings is presented in Figure 1. Before initiation of nasal CPAP, cephalometric radiographs were obtained. They were obtained following the technique reported by Riley at al, with the subject awake, seated, and at end-inspiration without

swallowing

Data Analysis Therapeutic

4

the Nasal CPAP

and SelectiOn ofSubgroupfor mid

were analyzed following the different to score sleeilwake and sleep-related abnormalities.b0,13.14 To evaluate presence/absence of snoring, the data obtained during the first three investigative nights (nights 1 and 3 in the laboratory and night 2 at home) were used. Patients were classified as “regular snorers” ifthey snored on all three nights, “irregular snorers” ifthey snored only on one or two nights. Patients international

All subjects with more than ten of these transient alpha EEC amusals per hour ofsleep in either ofthe two laboratory recordings 782

(SensorMedic,

Pblyg,aphic

3

A new nocturnal polygraphic recording was made with the same variables as before and, in addition, there was monitoring ofsnonng sounds and of respiratory efforts by measurement of esophageal pressure (),17 ‘1 esophageal catheter was placed transnasally and calibrated following the technique described by Baydur at aL’7 Once the subject was comfortable In bed, a baseline recording was obtained for 30 min during quiet supine wakefulness. The new nocturnal polygrapisic recording was scored following the Bechtschaffen and Kales’ international criteria. This recording and the baseline recording was then scored for presenc&absence of short (transient) alpha EEC arousals, defined as alpha EEC bursts hating a minimum of 3 s in the central EEC derivation.” This anal found that all subjects presented short arousals o13 to 14 s. However, some subjects presented many more than others. In a previous Investigation olseven normal subjects(mean age 27 years), none presented more than ten short, transient alpha arousals per hour of sleep in one nocturnal sleep recording.” Thus, we chose a cut-offpoint of ten short arousals per hour of sleep when selecting ts for further respiratory investigation during sleep. Night

underwent a third laboratory night recording (ie, a fourth night of investigation). During this night, quantification of airflow was performed using a tight fitting mask and a pneumotachometer. The maskcovered the noseand mouth andwas opened to the atmosphere via the heated pneumotachometer connected to a transducer (Validyne MP45). Esophageal pressure was again measured using the method described by Baydur et al,’ and all other previously described variables were again monitored. In half of the subjects, Pco1 was also measured with a transcutaneous tcPco, electrode

called

were

monitorings

criteria

used

“continuous

snorers”

night or more, “intermittent percent ofthe night.

if they

snorers”

snored

if they

75 percent snored

of the

less than

75

Recordings on the second and third laboratory nights (nights 3 and 4) were analyzed to determine the relationship between alpha EEC arousals and abnormal breathing, based on the following criteria. In night

14 s were the

arousal

3 recordings, identified. was

transient alpha EEC arousals lasting 3 to breath in the preceding 10 min before and each Pea nadir was determined. (1) the alpha EEC arousal was more negative

Each

analyzed,

Ifthe Pni nadir preceding than 1 SD below the mean Iks nadir monitored during baseline quiet supine wakefulness, (2) ifit was the most negative nadir of the snoringperiod. and(3)ifthe breath following the alpha EEC arousal was associated with an abrupt reduction in Pes nadir, the alpha EEC arousal was scored as “related to” the increased respiratory effort indicated by Pbs. In all ofthe night 4 recordings, total sleep time was reduced, as Excessive Daytime Sleepiness (Gulfiemlnauleta!)


rM

ri

ri NIGHT

I

MGHT

2

NIGHT

3

NIGHT

H

as-

.

ii

PSG

it

II

PSG

4 .....

PSG

UNATI’ENDED

BASELINE

MONiTORING MESAM

with

N

with Pes &

Pes

Pneumotachomcter

CPAP

CPAP NIGHT2

NIGHTI

H

MSLT

cPAP

4weeks

LJ

PSG with

Pes

with

follow-

Pes

& CPAP

& CPAP

titration

titration

follow-

up

up

1. Flow chart of polygraphic recordings performed on subjects with “upper airway resistance Nights 1 to 4 in the top row are the baseline nights. Continuous positive airway pressure (CPAP) nights 1 and 2 on the left side of the bottom row are the two CPAP titration nights. ‘fltration was performed based on esophageal pressure (Pes) monitoring. CPAP follow-up (bottom row, right side) was performed with nasal CPAP and with monitoring similar to that on baseline night 1, without measurement of Pes. PSG = polysomnography. FicunE

syndrome.”

the facial This

mask with

the

abnormal

EEC

pneumotachograph

was performed

recording

breathing

used

sleep.

nocturnal

the relationship

and the repetitive

pattern

It was never

arousals.

disturbs

to evaluate

for evaluation

between

alpha

transient of sleepiness.

Five

sleep, ending with a transient alpha EEC arousal, were randomly selected from each night 4 recording for breath-by-breath analysis ofPes. airflow, and tidal volume. In night 4 recordings, a short alpha EEC arousal was related to the breathing behavior preceding it (1) ifthe immediately preceding Pes nadirs were more negative than 1 SD below the mean Pes nadir periods

with

monitored

well-consolidated

during

quiet

supine

wakefulness,

Pes nadirs in a long sequence

(2) ifthe

most negative

were seen just prior to the occurred with the peak negative Pes nadir, (4) if no other polygraphic events had occurred that could cause arousal, and (5) if just after the short arousal a less negative Pes nadir was observed with increase in flow Forcomparison, transient alpha EEC arousals were also identified in the recording performed 1 month after initiation of nasal CPAP arousal,

treatment. obtained

ofbreaths

(3) if a simultaneous

This prior

Statirticd

alpha

decrease

EEC

arousal

Descriptive

index

was compared

with

that

to any treatment.

statistics

(paired

t tests,

repeated

measures

analysis

significance.

RESULTS

Initial

Population

Forty-eight

subjects,

20 men

and

28 women,

mean

age 33 ± 9 years, were diagnosed as having “idiopathic hypersomnia” (ICSD code 78O547),8 z, did not fit the criteria for a well-defined syndrome at the end of the screening evaluation. The mean multiple sleep latency test score for the group was 6.1 ± 1.6 mm. Snoring

Eighteen snored

13 of these criteria

of 48

intermittently

subjects

Thirty data

intermittent

for “upper

subjects had been

did not analyzed,

snore. only

snorers

fit the

or continuous airway

resistance

defined upper airway resistance combination of a clinical complaint ness, presence ofabnormal MSLT)

syndrome.”

We

syndrome as the (daytime sleepiwith demonstration

of flow limitation (Pes monitoring) and demonstration ofincreased respiratory efforts with arousaijust following peak negative inspiratory Pes.

(8 women

or continuously

and at home,

10 men) in the

Polygraphy

Nocturnal None

ofthe

48 subjects

syndrome as currently 13 women, 20 men)

in Pes

had obstructive

their mean monitoring

nadir

sleep

defined. Thirty-three had very few or no

(

alpha EEC arousals; was 7 ± 2 on the first change

were used to evaluate

or both. polygraphic

in flow

AnolysLc

ofvanance)

laboratory, When all

was

seen

alpha night.

during

apnea

subjects transient

arousal index No significant sleep

compared

with quiet supine wakefulness had a mean Pes nadir during

in these subjects. They sleep of 6.6 ± 2.5 cm

H20 and a mean Pes nadir during 30 mm ofquiet supine

of 5.5 ± 2. 1 cm wakeful breathing.

men Their mean

-

H2O Five

-

in this group had light, intermittent snoring. mean lowest Sa02 was 94.5 ± 1 .6 percent. Their sleep latency was 5.3 ± 2.8 mm. These 33

subjects nia.

were

Positive

Subgroup

Fifteen

believed

subjects

to have

(8 women

“idiopathic

and

7 men)

hypersom-

of the

48

presented with frequent (ie, 10/b) transient alpha EEC arousals. These 15 individuals were slim, with a mean body mass index of 23 ± 3.2 kg/rn2 and a mean age of37.5

±

7 years.

The CHEST

seven

men

had a mean

I 104 I 3 I SEPTEMBER,


1993

body 783

Table

1-Population

W’dh

Upper

Airway

Night Night 3 Night 1 Night 1 Subject Noi Age, yr/ Sex

TS1 mm

Baseline

(w/Pes)

(w/Ps)

Maxi-

MSLT,

TS1

Index

mm

min

Index

Alpha

Baseline

(w/Pes)

3

Baseline

Mpha EEC Arousal

EEC Arousal

Baseline

BMI, Snoring kg/m’

Night 3 Night 1 Baseline

Baseline

Syndrome#{176}

Resistance

CPAP Titration

mum Maximum Fos Nadir, Pes Nadir, cm H,O

IJ:XWF

+

32

481

33

2.3

492

35

-27

-8

+

28

473

31

2.8

460

30

-34

-6

3f4WF 4/3&’F 5,2NF &36/F 7/46/F &WF 915a’M 1W471M

0 0

20 24 22 21 24 23 23 23 23 19 23 22

492 468 501

19 36

5 6.1 6

23 34 23

-29 -43

17

478 490 452

497

52

8.1

450

479 486 469 485

35 24 14 20 42 16 40 42 49 31.3 12.4

6.4

495

7

479

26

-33

6.3 4

453 511

19 22

4.7

442

5 5.1 6.0 5.1 5.3 1.5

473 434 472 509 472.7

+

mt + + +

11/3WM

mt

12./1WM

+

13/44/M 14/4JJM

+

Int

13WM

+

37.5

Mean

SD7.0 *BMIbdy

499

471 465 456

26

484

23.6 3.2

480.4 12.93

mass index;

RDlrespiratory

24.7

disturbance

index;

polysomnographic nights; MSLT= positive airway pressure; Follow-up CPAP= monitoring Int= intermittent snoring; 0 no snoring. second,

third

mass

index

of 22.7

kg/m2

± 2. 1

and

a mean

age

of

The

1.)

had

Two women intermittent,

and

five

Subgroup never snored, two men and one woman light snoring, and ten subjects (five

women)

were

Night 1 Polygraphw (Without Pes and Mean mean mean

total

Monitoring Pneumotachograph

sleep

sleep latency lowest Sa02

respiratory

disturbance

alpha alpha

arousal arousals

EEG EEG

well-defined arousals nocturnal

snorers.

regular

sleep

time

was

in the

480.4

15-Subject Monitoring)

± 13 min,

and

the

in the MSLT was 5.3 ± 1 min. The was 94 ± 1 .4 percent. The mean indexwas

1.7. The mean

2.1 ±

index (calculated from that could not be related disturbances)

was

transient to other

31.3

± 12.4

per hour of sleep (‘I#{224}ble1). Investigation of sleep state and stage distribution indicated

an abnormally

low

amount

sleep (mean percentage: ofREM sleep was 17.7±

of stages

3 to 4 NREM

1.2±2 percent). 1.6 percent.

Percentage

Polygraphic Monitoring With Pes Measurement (Night 3) in the 15-Subject Subgroup Mean peak 784

total

Pes nadir

sleep related

time

was

473 ± 25 miii.

to transient

alpha

EEC

TST,

MSLT,

Index

min

min

9

398

9

393

10

-4

8

406

12.8

-5

6.1

389

14.5

-37

-4

6

14.6

46

-41

-5

0.8

38

-36 -23 -25

-4.5 -6 -4 -5

7 10 8 8

411 381 398 428 395

38

-43

-5.5

9

401

18 44 31 37 30.9 9.2

-31 -28 -24 -42 -33.1 7.1

-5 -6 -6 -5 -5.3 1.1

6 9 9 6 7.9 2.3

389 386 419 413 399.6 15

sleep time;

was

Fos=esophageal

-

33 ± 7 cm

The

mean

arousals

H20.

15.4 15.2 14.7 11.8 13.5 14.8 12.5 13.8 12.6 13.5 2.1

387

pressure;

The

17

Nights

CPAP=

1, 2, 3’first,

nasal continuous

+ =continuous

mean

alpha

EEC

snoring;

arousal

index was 31 ± 9. No increase in CO2 within the reliability range of the tcPco2 electrode was noted in the seven patients monitored with this equipment. In ten subjects (regular snorers), the behavior of the Pes curve

in This

Snoring

Arousal

Follow-up

10

ThTtotal

eight women had a mean body mass index of 24.3 ± 4.0 kg/rn2 and a mean age of 36.5 ± 7.0 years. (Individual data are presented in ‘Table

CPAP Night

Follow-up

multiple sleep latency test(on day following a recordingJ; with CPAP after nightly use of CPAP for 1 month;

nocturnal

38.6 ± 10.8 years.

CPAP Night

cm H,O

2/42/F

+

Follow-up CPAP Night Mpha EEG

changed

from

snoring occurred, end of inspiration

snoring

to nonsnoring.

the nadir of the Pes curve of each respiratory cycle)

When (at the became

more negative. Changes varied with the individual; the mean peak Pes nadir increase from the period just prior to snoring to the snoring period was 108 ± 47 percent. In the two nonsnoring subjects and the three intermittent snorers, the mean supine quiet wakefulness Pes nadir was 5. 1 ± 1 cm H20. Peak Pes nadir during sleep just prior to EEG oscillated between - 20 and - 29 cm H20 cm H20 in these five subjects, -

regardless

ofwhether

snoring

was

present.

Polygraphw Monitoring With Pneumotachograph (Night 4) in the 15-Subject Subgroup In the performed

randomly

volume 100 499 ± 96 ml.

breaths The

arousal 22±6

presented percent

during

quiet

These studied

selected segments volume calculations,

tidal

before an alpha EEC breath just prior to the

a mean compared

sleep

changes periods;

on which we the mean tidal

without

tidal with

volume breathing

increase

were not progressive they mainly involved Excesse

Daytime


Sness

arousal was alpha EEC decrease of measured

in Pes nadir. throughout the the one to three (GuiNemInaufteta

EMGFAL

mflMv\,J&

35

% Sa02

150

Ficuns 2. Monitoring of an alpha EEC arousal with quantitative evaluation of airflow. Flow is measured with a tightly fitting mask and a heated pneumotachograph (channel 7 from top). Esophageal pressure (channel 8) is at its nadir in the two breaths just preceding the arousal (indicated by arrows). The arousal begins with the second Pes nadir. Immediately following the onset ofthe transient arousal, inspiratory Pea nadir is less negative (breath just following black arrow). The “sum” signal of inductive respiratory plethysmography presents some change in shape. However, this change would be difficult to interpret if flow and Pes were not simultaneously measured. No desaturation is noted in the pulse oximetry recording. The flow (channel 7) decreased the most in the breath just preceding the arousal (black arrow) but is already decreasing in the breath marked by the white arrow.

breaths

preceding

polygraphic sient alpha

an

arousal.

recording obtained EEC arousal: the

Figure

2 presents

just prior inspiratory

at its maximum during the two breaths the arousal, and peak flow decreases with drop

in tidal

did not oximetry) Therapeutic

this

volume

for these

two

just prior to an associated

breaths.

affect the oxygen saturation in any ofour patients.

a

to the tranPes nadir is

This curve

drop (pulse

arousal

index

decreased

during indicated

CPAP therapy. a nonsignificant

to

8 ± 2 per

Nocturnal change

sleep

(18.2

in the

± 1.6

percent),

percentage

of stages

but

a significant

3 and

4 NREM

sleep (9.7± 1.9 percent, p