Sudden Infant Death Syndrome: Sleep Apnea and Respiration in Subsequent Siblings

(1)

Sudden

Infant

Death

Syndrome:

Sleep.

Apnea

and

Respiration

in Subsequent

Siblings

Toke Hoppenbrouwers, PhD, Joan E. Hodgman, MD,

Dennis McGinty, PhD, R. M. Harper, PhD, and

M.

B. Sterman, PhD

From the Newborn Division of the Los Angeles County-University of Southern California Medical Center; Department of Pediatrics, University of Southern California School of Medicine; Sepulveda Veterans Hospital; Departments of Anatomy,

Psychiatry, and the Brain Research Institute, University of California, Los Angeles

ABSTRACT. Subsequent siblings of infants who died of

the Sudden Infant Death Syndrome are at a four- to

six-times increased risk to die of this syndrome. This study

compares the respiratory development during sleep state

of this epidemiologic high risk group with that of normal

infants during the first six months of life. Subsequent

siblings exhibited higher respiratory rates in all states at 3 months of age. Quiet sleep and indeterminate respira-tory rates were elevated at 1 week of age compared to control infants. Indeterminate respiratory rates remained

higher at 6 months of age. These differences were accom-panied by a reduced incidence of total breathing pauses

of two to five seconds and six to nine seconds duration in

siblings. Study groups could not be differentiated on the basis of either breathing pauses of more than ten seconds

or central apnea of six seconds or more. Obstructive and

mixed apnea (6 seconds or more) were infrequently

ob-served in these study groups. A high degree of

intersub-ject variability characterized all data on breathing pauses. Pediatrics 66:205-214, 1980; sudden infant death

syn-drome, sleep apnea, respiration, subsequent siblings.

Sudden infant death syndrome (SIDS) is the leading cause of infant mortality beyond the

neo-natal period in the United States.’ Scientific

inves-tigation of SIDS is hampered by the difficulty in identifying appropriate populations for study.

Frog-gatt et al2 reported that subsequent sibings of SIDS were at a four- to fivefold statistically increased risk

to die of SIDS. A recent study by Peterson et al3 confirmed the increased risk for this population.

Received for publication Aug 24, 1979; accepted Nov 29, 1979.

Reprint requests to (T.H.) Director, Sudden Infant Death

Syn-drome Research Project, Room 4L40B, Women’s Hospital,

LAC-USC Medical Center, 1240 N Mission Rd, Los Angeles, CA

90033.

American Academy of Pediatrics.

Whereas, in siblings the absolute risk of dying of

SIDS is below 2% so that the risk for the individual

infant is small, this group could provide clues to the

potential mechanism of SIDS and was therefore selected for study.

The highest risk for SIDS is between 2 and 3

months of age.’

Infants

die

during normal sleeping

hours in 88% of the caseS.4 Thus, the variables of age and sleep are strongly related to SIDS. In adults, the cumulative effects of sleep apnea may

result in hypoxia leading to elevated pulmonary

arterial

pressure,

cor pulinonale, and sudden death.5

Consequently, it has been proposed that an in-creased incidence of sleep apnea may characterize

high risk infants during the first months of life.6’7

We have monitored sleep and cardiopulmonary

variables

in subsequent

siblings of SIDS. Prelimi-nary

findings

indicated that these infants exhibited a decreased number of apneic episodes

accompa-nied by increased respiratory rates.8 The objective of this study is to provide complete data on

respi-ratory rates and apnea in control infants and sub-sequent siblings of SIDS during the first half year of life.

METHODS

Material and Monitoring Procedures

Selection criteria for 25 control subjects included the absence of both maternal disease and familial

incidence of SIDS. The experimental group con-sisted of 26 infants born to women who had lost an

infant to SIDS as confirmed by autopsy reports.

The two groups were homogeneous and comparable with respect to socioeconomic status, as estimated

by level of parental education. The experimental

(2)

26.10 3.64 20 1 2 2 2.84 1.43 2.48 1.19 3,575 465 2,890-4,550 40.78 1.64 37-44 8.52 1.64 9.28 0.54 26 26.45 5.06 21 0 2 3 3.13 0.90 1.92 0.83 3,594 496 2,821-4,593 40.06 1.31 38-42 8.14 1.04 9.04 1.04 19

TABLE 1

.

Characteristics of the Study Groups

Control Subsequent Subjects Siblings Maternal age Mean SD Maternal race White Black Asian American Mexican American Gravida Mean SD Parity Mean SD

Birth weight (gm)

Mean SD

Range

Gestational age (weeks)

Mean

SD

Range

Apgar at 1 mm

Mean SD

Apgar at 5 nun Mean SD

Respiratory infections at

time of monitoring

control group had nine girLs and 16 boys. A furthet description of the study groups is provided in Table 1. All infants were monitored on six occasions,

dur-ing the first week and at 1, 2, 3, 4, and 6 months of

age. Informed consent was obtained prior to each monitoring session. Ages at time ofmonitoring were

comparable in the two study groups.

Each infant was admitted at 5:00 PM to the sleep

laboratory

for 12-hour all night monitoring sessions

during the first week of life and at 1, 2, 3, 4, and 6

months of age. The infants were fed during prepa-ration for monitoring and application of electrodes.

A demand

feeding

schedule was followed and in

several instances the infants were breast-fed. Arm

restraints

were applied before the initiation of

re-cording.

Monitoring was carried out in a darkened room adjacent to the room containing recording equipment. Room temperatures ranged between 22 C and 30 C. The infants were placed in a supine or

side-lying position and observed continuously with the use of a low ifiumination television camera and

monitor.

Behaviors

such as closing or opening of

the eyes, startles, crying, and nursing interventions

were charted on the polygraph paper.

Physiologic Recording Methods

The sleep variables recorded included two EEG derivations, a chin electromyogram (EMG), and eye

movements. Thoracic or abdominal excursions were

monitored by impedance pneumography. In addi-tion, a Beckman PCO2 monitor sampled expired gas

through a two-pronged miniature cannula taped under the infant’s nostrils. In order to adjust for the

time lag inherent to this instrument, the phase of

respiration was simultaneously detected with a

thermistor placed into one arm of the cannula. The

ECG was recorded with two disposable electrodes

placed symmetrically beneath the clavicles. A ground electrode was applied above the umbilicus. Additionally, a skin temperature probe was applied to the abdomen below the right costal margin.

Electrodes on the mattress surface under the crib sheet registered the infant’s gross body move-ments.9

Data were recorded on a 16-channel Grass model 76 polygraph and simultaneously stored on a 14-channel Honeywell analog tape recorder together

with an IRIG E time code.

Data Analysis

Each minute of the record was coded as either

active sleep (AS), quiet sleep (QS), awake (AW), or

indeterminate (IN). Scoring criteria have been re-ported elsewhere.’#{176}’2

The entire data set for each infant was digitized on a PDP-12 laboratory computer and the PCO2

signal submitted to a breath-to-breath interval

de-tection program. This signal was selected for its

reliability and resistance to movement artifacts. Feeding and sometimes crying characterized wak-ing. When the infant cried for an extended period, the PCO2 respiratory tracing occasionally

disap-peared

entirely, indicating a complete shift to

mouth breathing. The resulting long respiratory

pause would give erroneous results. To deal with this problem, long episodes of crying were deleted

from the analysis (0% to 10%). The respiratory signal was almost always preserved (although with

lower amplitude) throughout short cries and vocal-ization and could thus be calculated. The use of

median rather than mean values provided some

degree

of

protection against aberrant data as well. In all cases, the polygraphic Eecords and, in

partic-ular, the chart notations and the impedance respi-ratory signal, provided final reference for artifact evaluation. Median respiratory rates per minute

were obtained and mean values for each defined

sleep and waking state were determined.’3 The

mm-ute-by-minute

interquartile range was selected as a

measure of respiratory variability. Details of these

procedures were published previously.’4

Apnea and short breathing pauses were tabulated in three ways. First, the initial 18 recorded tracings

(3)

f .

‘1

L

R.

EYE

S0MCt

IMP

.V

.

-

10

IIIIII(II1.IIIIII(tItIIIIIlIIIIIIIIIIIIIIIIlIIfIIIIII1IIIIIIIItIIIII(fIIlIIIIIl IL.IIhIIIlItIIIIIII I I I IIIIIIIiII IIiiIIiii iiIiIIIIIiiIiiiiiiIi IIIiiIIIIIiii,IlIlitilIIii.IiII.iiIiiiiiIIiIiiI, I I iI

L. EEG

R. EEG

EYE MOV.

vg-v

SOMCt

- ..

C

...,1ltIlIhIIHt..h,.

‘“

.

\y,

.‘\y, ‘‘\\\‘l\ \\\

‘““‘

‘ s” II ‘t’

of apnea in a 3-month-old infant in active sleep. The first

apnea is a mixed apnea. This segment illustrates the

difficulty in determining the exact duration of an apneic

episode. The second pause, of central origin, seems to be

interrupted by a shallow breath. The third pause appears

to contain an obstructive component, although the

therm-istor and Pco2 tracing provide somewhat conflicting

in-formation about air flow. Finally, the fourth pause (<6

seconds) is of central oi,gin. All pauses were preceded by

a deep breath.

Fig I

.

Top, Obstructive apnea in 3-month-old infant.

Respiratory movements, evidenced in the impedance

tracing (IMP), were not accompanied by air flow, as can

be seen in the thermistor and Pco2 tracing. This sample

(100 seconds) was taken during active sleep. Motility was

limited to phasic twitches (see EMG) and did not

pre-dude identification ofwhat appeared to be an obstructive

component, following a central apnea of almost seven

seconds. Note that a heart rate below 100 beats per minute was observed twice, once following the obstructive

apnea and once not preceded by an apnea. Bottom, Series

Visually scanned for the presence of central apnea equal

to

or in excess of six seconds duration. The method has been described in detail previously.” Secondly, this study was replicated in 14 different

subsequent siblings and ten different controls. In these latter recordings, apnea with a clearly ob-structive component and mixed apnea were identi-fled as well. Occasionally an initial

cessation

of air movement as measured by expiratory CO2 will be accompanied by identifiable respiratory excursions.

We designated this

pattern

as obstructive

apnea of the unequivocal variety (Fig 1). In none of these episodes could mouth breathing be ruled out, but

annotations by the nurses such as “grunting,

breathing

through the nose, noisy respirations”

aided

in ruling

out false-positives. The actual inci-dence of obstructive episodes is

likely

to be lower than reported here. Fig 1, bottom, represents a

mixed

apnea.

Finally,

the recordings

of all subjects

(4)

identification of breathing pauses involved measurement

between points e and f. Note that this interval is

approx-imately one second longer than the two former (a-b, c-d).

The interval g-h reflects the lag time between the respi-ratory cycle from a nasal thermistor and the Pco, signal.

AS OS

60

50

40

30

20

10

60

50

40

30

20

10

IN

0.0 00

AW

#{149}#{149}%%.._

-- -- Siblings (N25) - Controls (N =25)

6mo (st)

4R ‘, , -

i

I ISgt

4 )

\prj

jl

I vsss( b

TIME IN SECONDS #{149} .. . . . . . . . . . #{149}

Fig 2. Measurement of breathing pauses. In a previous

study,” apneic episodes were identified through visual inspection of three respiratory tracings. The intervals

a-b

in the Pco tracing and c-d in the impedance tracing

had to be equal to or exceed six seconds. Computer

2mo 3rno 4rio 6mo lwk Imo 2mo 3mo

AGE

Fig 3. Median respiratory rate (ordinates) in breaths per minute as a function of age

(abscissae) in various states. Group means are based on 25 infants. At 3 months of age,

siblings breathed faster in all states.

aimed at identifying breathing pauses equal to or

longer than three seconds duration in each sleep state. Identification of the onset and termination of a breathing pause is fraught with problems,

espe-cially

when more than one respiratory signal is used

to define a pause. Fig 2 provides three respiratory tracings, an estimate of the length of a breathing

pause based upon criteria used for visual

identifi-cation

of apnea, and the computer-derived breath-ing pause based on peak-to-peak measurements of

the PCO2 signal. The latter technique, while reliable,

tends to overestimate the duration of a pause by one second, compared to visual inspection criteria.”

Duration categories will be arbitrarily adjusted to

criteria

of

visual

inspection:

short,

2, 3, 4, and 5

seconds

duration; medium, between 5 and 9 sec-onds; and long, equal to or in excess of 10 seconds. Since the Pco2 signal was used, this computer

strat-egy did not differentiate between, but included, central, mixed, and obstructive pauses.

(5)

the number of minutes spent in each state. The resulting value was multiplied by 100 to obtain a

density score in percent. An analysis of variance

with the factors of age and experimental group was used to examine developmental trends and study group differences. A Tukey test was used to assess

differences between individual means.’5

RESULTS

Respiratory Rates

Rates in both study groups declined sharply

be-tween birth and 2 months of age in each state, as previously reported in control infants.’4 Respiratory rates continued to decrease up to 3 months of age

and leveled out thereafter. Group means for each

age and state are shown in Fig 3.

The mean, standard deviation, and range of me-dian respiratory rates in both study groups at each

age are provided in Table 2. Subsequent siblings breathed significantly faster than control infants in all states at 3 months of age. During the first week

of life, the QS and IN respiratory rates were also

elevated. Finally, at 6 months of age the IN respi-ratory rates of subsequent siblings remained higher

than those of controls (Table 2).

Median respiratory rates of individual infants at

1 and 3 months of age during quiet sleep are plotted in Fig 4. This figure illustrates the shift toward

faster respiratory frequencies in the subsequent

siblings

with

less overlap in study groups at 3 months compared to 1 month of age. Consequently, a statistically significant difference in QS means

was limited to 3 months of age. The infants who contributed most to the shift in distribution at 3 months of age are not identical to the ones at 1

month of age.

A developmental decline in respiratory variabifity

paralleled

that

in respiratory

rate

(Fig

5). Breathing

was most variable during the waking state across the entire age span in both study groups. Variabifity was least during QS. The values of AS and IN were indistinguishable at all ages. Siblings exhibited a higher degree of variabifity in breathing, but

differ-ences were limited to QS variabifity at 1 week of

age (P = .034) and QS and IN variability at 3

months of age (P = .023 and .030).

TABLE 2. Mean, Standard Deviation,* and Range ofRespiratory Rates (Breaths per Minute) as a Function of Study

Group, Age, and Sleep State

Qui et Sleep Acti ye Sleep Indeterminate Awake

Control Siblings Control Siblings Control Siblings Control Siblings

Subjects Subjects Subjects Subjects

1 week

Mean 38.88 48.23 50.95 56.05 48.09 54.37 49.13 51.43

SD 7.68 13.18 10.76 11.19 13.13 13.01 9.58 10.55

Range 26.6-59.2 29.3-78.1 36.0-74.7 36.5-73.7 35.7-75.0 34.1-75.1 40.4-70.7 38.4-83.3

P value .003 .. . .039 ...

1 month

Mean 36.23 39.16 44.83 46.72 43.22 45.68 48.62 50.14

SD 6.98 7.14 7.29 8.03 7.19 9.60 5.71 7.91

Range 30.4-46.0 30.3-51.5 33.2-55.1 33.4-56.0 32.0-53.7 34.2-57.2 40.7-58.7 43.1-64.0

Pvalue ... ... ... ...

2 months

Mean 31.73 33.92 38.00 39.20 35.73 37.90 44.33 48.44

SD 5.62 5.64 6.14 5.20 6.57 6.84 8.07 10.48

Range 24.0-39.2 23.9-45.1 29.7-49.0 28.6-48.8 27.8-52.2 23.5-53.0 35.9-55.3 36.4-81.0

Pvalue ... ... ... ...

3 months

Mean 26.73 31.10 33.45 36.13 30.55 35.29 39.87 45.29

SD 4.07 4.65 5.11 5.89 5.28 7.69 6.52 8.26

Range 21.3-33.2 24.1-37.1 26.4-42.0 27.6-48.1 23.0-39.7 24.9-46.7 32.0-59.4 32.9-54.8

P value .001 .028 .001 .004

4 months

Mean 26.23 28.33 32.16 33.79 30.06 32.56 38.80 41.81

SD 3.64 5.44 4.66 6.26 4.72 7.74 6.77 10.34

Range 21.7-33.3 19.6-39.8 26.3-38.4 24.7-48.6 24.4-40.4 23.4-52.6 31.3-60.5 26.5-58.2

Pvalue ... ... ... ...

6 months

Mean 25.57 27.14 30.05 31.20 28.25 30.44 39.32 39.64

SD 4.84 3.83 5.63 4.51 5.24 4.82 9.25 8.12

Range

19.4-34.2

23.0-34.9 22.2-44.7 26.4-39.1 22.3-41.9 23.6-38.6 32.2-63.4 22.7-56.4

Pvalue ... ... .026 ...

(6)

.

g

00 - .e ?

cSCO

S

.

#{149},

I

_#{149}#{149}

.

#{149}. 0,:

.

.S

.

U)

e1

_.g

_S

-:1:

#{149}

:.

I-

:#{149}

z

..-.

0 #{149}

0

#{149} ()

#{149}

#{149}.

#{149}

#{149}#{149}

‘IIz

::

I. #{149}

#{149}:

:

#{149} g

#{149}

.

#{149}

.1

ce

#{149}:

0 c’

le

$

#{149}

.

#{149}#{149}

#{149}

_cs4

‘-‘

3

.1

#{149}

04

#{149}

;

#{149}#{149}

I’,

cJ

#{149} N

#{149}

. N

.

U)

(n.e

__J U) _J U)

0 0 0

‘I)

Z

u, tr , z

zI-. w I- w

o z

z

j4

0 If ftf 0 lIf ._“o

U,

(7)

Os As

5

25 #{149} 11 25

Li:

TTT-T-T----

-,

‘:_______

1WK. 1MO. 2M0. 3M0. 4M0. 6110. 1C. 11.10. 2MG. 31.10. 4MG. SMO

A A

Fig 5. Respiratory variability

in

breaths per minute (ordinates) as a function of age

(abscissae). Differences between siblings (broken line) and controls (solid line) were

restricted to quiet sleep (QS) at 1 week and quiet sleep and indeterminate (IN) at 3 months

of age.

Visual Identification of Apnea (Six seconds or

More)

Densities of apneic episodes for each study group are shown in Fig 6. AS densities were approximately 12% during the first week of life, indicating the

occurrence

of one apneic

episode every eight

mm-utes. A sharp reduction was seen at 1 month of age

o

to approximately 5%, indicating an apneic rate of

once every 20 minutes. No further change was seen

during subsequent ages. QS densities of apneic

ep-isodes were lower (2% to 3%) and did not decline

with increasing age.

The median and range of central breathing <

‘, ‘

‘.

pauses at each age are provided in Table 3. An .

analysis of variance for the first 18 infants with - - o o - o

factors age and experimental group revealed that . - . 5 - - - 5

siblings

exhibited less apnea than control infants 5 S

(P < .03). Data on an additional ten control and 14 subsequent siblings were then compared with the

original set. The apnea densities in the first and

,

second set of control infants could not be differen- 1 wk 1 mo 2 mo 3 mo 4 mo 6 mo

tiated.

However, a significant difference was found AGE

between mean apnea densities of the first and sec- - - - - AS OCT (N 1 9) #{149}55 (N =23) ond set of siblings, with the latter resembling both

control groups. When the total data were resubmit- Fig 6. Central apnea density (6 seconds) in

subse-ted to an analysis of variance, no significant study quent siblings and controlinfants during active sleep (AS)

(8)

.

CONTROLS

.

SIBLINGS

QS

4M0 6M0

TABLE 4. Median and Range of Obstructive and

Mixed Apnea per Study Group and Age TABLE

in

Contr

3.

ol S

Total Incidence of Central Apnea ( 6 Sec)

ubjects and Siblings as a Function of Age

Age Control Subjects Siblings

Median Range Median Range 1 wk

1 mo

2 mo 3mo 4 mo 6mo

49 2-277 33 1-276

21 4-119 15 1-102

10 1-98 13 1-68

18 0-71 6 1-38

17 0-51 5 0-34

13 1-58 13 0-51

300 IN

250

200

50

100

300 AS

i

:o(IuLJrIjL1LIL

::

I

50

IWK MO 2M0 3M0

AGE

Fig 7. Sleep state related density of breathing pauses

between two and five seconds (ordinates) as a function of

age (abscissae). Note the reduced incidence of these

pauses in subsequent siblings, particularly at 3 and 6

months of age in quiet sleep (QS).

Computer Identification of Breathing Pauses

Two to Five Seconds. In all sleep states

subse-quent siblings exhibited significantly fewer

respira-tory pauses between two and five seconds duration

(P < 0.03). Densities for each study group are plotted in Fig 7. In addition to these main effects, a significant age by study group interaction was

observed

in QS (P = .02). A Tukey test of multiple

means revealed that at 3 and 6 months of age

subsequent siblings exhibited fewer QS breathing

pauses

of this duration.

Six to Nine Seconds.

Subsequent

siblings

ex-hibited significantly fewer breathing pauses in both

AS and IN (P = .01 and .05). No additional age by study group interaction identified a specific age at

which differences were more pronounced.

Ten Seconds or More. No differences between

study groups were observed in the density of these

breathing

pauses, with the exception of a significant

age by study group interaction in IN (P = .01).

Further examination showed that subsequent sib-lings exhibited significantly fewer pauses in excess

of nine seconds during IN at 1 week of age.

Six Seconds or More. To allow for comparison

with visually identified central apnea of this length,

the density of AS and QS apneic episodes six

sec-onds or more is provided in Table 4.

Density scores determined through computer analyses were considerably higher than densities of

central apneic episodes (six seconds or more), in

particular during the first month of life (Fig 5). A

small portion of this difference can be explained by obstructive and mixed apnea, but the difference is

mostly due to movement artifacts (vide infra).

Mixed and Obstructive Apnea

The conclusion that apnea density (six seconds

or more) derived from computer analysis of the PCO2 signal minus central apnea density (six sec-onds or more) based on visual analysis would reflect the incidence of mixed and obstructive apnea is not

Age Control Subjects Siblings

Median Range Median Range

1 wk 2.0 0-29 1.0 0-10

1 mo 2.0 0-11 1.0 0-7

2 mo 1.0 0-21 0.0 0-3

3 mo 1.0 0-16 0.0 0-2

4 mo 0.0 0-8 0.0 0-2

6 mo 0.0 0-7 0.0 0-2

TABLE 5. Density of Computer Identified Apneic

Ep-isodes (6 See) as a Function of Age, State, and Study

Group

Age State

Quiet Sleep Active Sleep

Control Siblings Control Siblings

Subjects Subjects

1wk 18.4 11.4 32.6 19.1

1 mo 7.1 5.4 13.9 12.0

2 mo 6.6 5.1 10.5 8.6

3 mo 8.2 3.8 12.8 7.5

4mo 6.1 4.1 14.7 9.1

(9)

warranted. Approximately 40% of the pauses were preceded by a movement which induced artifact in

the thermistor and impedance tracings and

pre-vented identification of the origin of the breathing pause. Subtraction of central from total

breathing

pauses overestimates the incidence of obstructive

apnea. The number of obstructive and mixed

breathing

pauses

proved

infrequent

in both study

groups and the variability among infants was again large. Table 5 provides the median number of summed obstructive and mixed pauses and the range at each age for both study stroups. Siblings tended to exhibit fewer obstructive and mixed ap-neic episodes than control infants.

DISCUSSION

The data presented here demonstrate that short

breathing

pauses

were abundant throughout the age span under investigation. A sharp increase es-pecially in QS at 3 months of age reflects the

developmental slowing in respiratory rates.’4”6

Ap-nea between six and nine seconds duration were

also

common

but

declined with increasing age.

Ap-nea in excess of nine seconds duration were most

prevalent during the first week of life in all sleep states. Finally, obstructive and mixed apnea were

seen infrequently at any age.

Siblings of SIDS victims exhibited higher

respi-ratory

rates

in IN and QS at 1 week of age, in all

states at 3 months, and in IN at 6 months of age. This effect was not attributable to a few infants

who exhibited abnormal patterns; instead, as a group the siblings showed a comparable distribution with an elevated mean respiratory rate. The uneven sex distribution with fewer boys who were

subse-quent siblings was not responsible for this finding. Respiratory rates in normal male infants were found

to be higher at certain ages, and never lower than female infants.’7 As the control group consisted of more boys, this sex distribution would favor higher

respiratory rates in the control group. Respiratory rates rise

following

feeding

and are influenced by

ambient

temperature.’8”9

Feeding patterns changed with age, as expected, but could not account for the differences between the subsequent siblings and

controls.20

Seasonal

temperature

changes

and

their effect on hospital temperatures were also evenly distributed among study group infants. The rise in respiratory rate was accompanied by a significant

decrease

in short

and

intermediate breathing pauses in subsequent siblings. Apnea in excess of nine

seconds were not different in the study groups.

Upper respiratory infections, reported to increase apnea,2’ occurred with comparable frequency in

control infants and siblings and exerted no

predict-able influence on the number of apneic episodes whether central, obstructive or mixed.

It is useful to distinguish between predisposing factors that contribute to increased risk for SIDS

and precipitating events that caused death.

Al-though much attention has been paid to the final

common

pathway

accounting

for death, less atten-tion has been directed to physiologic factors that

predispose to risk. Evidence that chronic or

inter-mittent hypoxia preceded death from

sIDs,

while

indirect,

is

mounting. The increased respiratory

rates in siblings are compatible with the hypothesis that the risk infants were responding to a relative oxygen lack. Br#{252}cket al observed a similar

venti-latory

adjustment of increased rates in infants be-tween 1 and 13 weeks of age known to be

hypox-ernie. Thoman et al, in monitoring a group of normal infants, identified two infants with tachyp-flea, one of whom subsequently died of SIDS.

Fi-nally, kittens monitored while breathing 10%

oxy-gen also

exhibited

increased

respiratory rates and decreased apnea.27 Although obstructive sleep ap-nea has been proposed as the etiology for chronic

hypoxia,7 transient or persistent air flow obstruction

was infrequent in both study groups and tended to

be even less common in siblings.

The etiology of SIDS has been sought for a number of decades and is still obscure. The

elusive-ness of this disease suggests that a constellation of

minor alterations each of which alone cannot

ex-plain death interacts to produce vulnerability to

SIDS. Recent evidence indicates that risk infants may have already been challenged in utero, and that environmental pollutants, in particular carbon monoxide, may contribute to functional hypoxia.’20 Such minor aberrant stimuli in prenatal and post-natal life may trigger compensatory physiologic

re-sponses or aggravate existing minor abnormalities.

These adjustments, while initially adaptive, when prolonged may initiate a sequence of events which perpetuates rather than limits abnormal

function-ing. In this model, increased respiratory rates rep-resent such an adaptive response. The majority of

infants

would

be expected

to successfully

compen-sate with little or no clinical symptomatology. For

an occasional infant the accumulation of minor abnormalities or the occurrence of a sudden stress may present a challenge for which the infant cannot

continue to compensate. In this context, it is inter-esting that hypoxic

conditioning

of young kittens

resulted

in a separation of kittens that coped and

those that failed to compensate and died.27

Identi-fication of compensatory responses, while poten-tially useful to elucidate underlying mechanisms,

cannot be expected to identify the individual infant

(10)

SUMMARY

Subsequent siblings of infants who died of SIDS are at a four- to six-times increased statistical risk to die of SIDS.2 The objective of this study is to compare the respiratory development of normal control infants and subsequent siblings of SIDS.

Subsequent siblings exhibited higher respiratory

rates

in all states

at

3 months of age. QS and IN

respiratory

rates

were elevated at 1 week of age compared to control infants. IN respiratory rates remained higher at 6 months of age. These differ-ences were accompanied by a reduced incidence of

total

breathing

pauses

of two to five seconds and six to nine

seconds

duration

in siblings. Study groups could not be differentiated on the basis of either breathing pauses of more than ten seconds or central apnea of six seconds or more. Obstructive

and mixed

apnea

(six seconds

or more)

were

infre-quently observed in these study groups; siblings

exhibited

fewer apneic episodes of the unequivocal obstructive and mixed variety. The reduction in computer identified apneic episodes between six and nine seconds in siblings can be accounted for by a reduction in numbers of equivocal breathing

pauses

of this

length.

A

high degree of intersubject

variability characterized all data on breathing

pauses.

ACKNOWLEDGMENTS

This research was supported by National Institutes of

Child Health and Human Development contract no.

NOl-HD-2-2777

and HD 4-2810. Computing assistance was

obtained from the Health Sciences Computing Facility,

UCLA, supported by the National InstitUtes of Health

Special Resources grant RR-3.

We thank Ms B. Havens, E. Hofmann, and S. Geidel

for

the development of the monitoring program and for

their contribution to the data collection. We also

acknowl-edge the contribution of Mrs K. Arakawa, Ms Mary

Fairbanks, Ms D. Jensen, and Mr J. Mason.

REFERENCES

1. Beckwith JB: The sudden infant death syndrome. Curr Probi Pediatr 3:1, 1973

2. Froggatt P, Lynas MA, MacKenzie G: Epidemiology of

sud-den unexpected death in infants (“cot death”) in Northern

Ireland. Br J Prey Soc Med 25:119, 1971

3. Peterson D: Genetic su8ceptibility to sudden infant death.

Paper presented at the research reporting workshop for the

National Institute of Child Health and Human

Develop-ment, Sudden Infant Death Syndrome Grantees and

Con-tracts, Alexandria, VA, September 1977

4. Valdes-Dapena M: Serum proteins, viral isolation, antibodies

in milk, and epidemiologic factors in sudden death, in

Wedge-wood RJ, Benditt EP (eds): Sudden Death in Infants,

Wash-ington DC, US Government Printing Office, 1965, p75

5. Gastaut H, Tassinari CA, Duron B: Polygraphic study of the episodic diurnal and nocturnal (hypnic and respiratory)

man-ifestations of the Pickwick’s syndrome. Brain Res 2:167,

1966

6. Steinschneider A: Prolonged apnea and the sudden infant

death syndrome: Clinical and laboratory observations.

Pe-diatrics 50:646, 1972

7. Guilleminault C, Peraita R, Souquet M, et al: Apneas during

sleep in infants: Possible relationship with sudden infant

death syndrome. Science 190:677, 1975

8. Hoppenbrouwers T, Hodgman JE, Harper RM, et al:

mci-dence of apnea in infants at high and low risk for sudden

infant death syndrome (SIDS). Pediatr Res 10:425, 1976

9. Hofmann E, Havens B, Geidel 5, et al: Long-term,

continu-ous monitoring of multiple physiological parameters in

new-born and young infants. Acta Pediatr ScaM 266(Suppl):5,

1977

10. Harper RM, Hoppenbrouwers T, Sterman MB, et al:

Poly-graphic studies of normal infants during the first six months of life. I. Heart rate and variability as a function of state.

Pediatr Res 10:945, 1976

11. Hoppenbrouwers T, Hodgman JE, Harper RM, et al:

Poly-graphic studies of normal infants during the first six months

of life. III. Incidence of apnea and periodic breathing.

Pedi-atrics 60:418, 1977

12.

Sterman MB, Harper RM, Havens R, et al: Quantitative

analysis of central cortical EEG activity during quiet sleep

in infants. Electroencephalogr Clin Neurophysiol 43:371,

1977

13. Mason J, Harper RM, Pacheco R: Analysis of respiratory

data during sleep and waking. Proc Dig Equp Comput Users Soc 567, 1974

14. Hoppenbrouwers T, Harper RM, Hodgman JE, et al:

Poly-graphic studies of normal infants during the first six months

of life. II. Respiratory rate and variability as a function of

state. Pediatr Res 12:120, 1978

15.

Winer BJ: Statistical Principles in Experimental Design.

New York, McGraw-Hill, 1962

16.

Hoppenbrouwers T, Jensen D, Hodgman JE, et al:

Respi-ration during the first six months of life. II. The emergence of a circadian pattern. Neurop#{224}diatrie 10:264, 1979

17. Hoppenbrouwers T, Hodgman JE, Sterman MB, et al:

Gen-der differences in respiratory rates and apnea density in

infants between birth and six months of age. Pediatr Res 13:

359, 1979

18. Prechtl HFR, Weinmann H, Akiyama Y: Organization of

physiological parameters in normal and neurologically

ab-normal infants. Neurop#{246}4iatrie 1:101,1969

19. Hoppenbrouwers T, Harper RM, Hodgman JE, et al: Effects

of feeding on respiratory regulation. Sleep Res 7:133, 1978

20. Hoppenbrouwers T, Jensen DK, Hodgman JE, et a!: The

emergence of a circadian pattern in respiratory rates:

Corn-parison between control infants and subsequent siblings of SIDS. Pediatr Res 14:345, 1980

21. Steinschneider A: Nasopharyngitis and prolonged sleep ap-nea. Pediatrics 56:967, 1975

22. Naeye RL: Hypoxernia and the sudden infant death

syn-drorne. Science 186:837, 1974

23. Valdes-Dapena M: Sudden infant death syndrome case

anal-yses. Paper presented at the research reporting workshop

for the National Institute of Child Health and Human

De-veloprnent, Sudden Infant Death Syndrome Grantees and Contracts, Alexandria, VA, September 1977

24. Williams A, Vawter G, Reid L: Increased muscularity of the

pulmonary circulation in victims of sudden infant death

syndrome. Pediatrics 63:18, 1979

25. Br#{252}ckK, Adams FR, Br#{252}ckM: Temperature regulation in

infants with chronic hypoxernia. Pediatrics 80:350,1962 26. Thornan EB, Miano YN, Freese MB: The role of respiratory

instability in the sudden infant death syndrome. Dev Med

Child Neurol 19:729, 1977

27.

Baker T, McGinty DJ: Reversal of cardiopulmonary failure

during active sleep in hypoxic kittens: Implications for

sud-den infant death. Science 198:419, 1977

28. Hoppenbrouwers T, Hodgman JE, Sterman MB, et al: Fetal

heart rate (FHR) patterns during maternal sleep and risk

for sudden infant death syndrome (SIDS). Sleep Res

8:126,

1979

29. Hoppenbrouwers T, Calub M, Arakawa K, et al: Seasonal

relafionship ofSIDS and environmental pollutants. Clin Res

(11)

1980;66;205

Pediatrics

Sterman

Toke Hoppenbrouwers, Joan E. Hodgman, Dennis McGinty, R. M. Harper and M. B.

Sudden Infant Death Syndrome: Sleep Apnea and Respiration in Subsequent Siblings

Services

Updated Information &

http://pediatrics.aappublications.org/content/66/2/205

including high resolution figures, can be found at:

Permissions & Licensing

http://www.aappublications.org/site/misc/Permissions.xhtml

entirety can be found online at:

Information about reproducing this article in parts (figures, tables) or in its

Reprints

(12)