Identification
of a High-Risk
Group
for Sudden
Infant
Death
Syndrome
Among
Infants
Who
Were
Resuscitated
for Sleep
Apnea
Joseph
Oren,
MD, Dorothy
Kelly,
MD, and Daniel
C. Shannon,
MD
From the Pediatric Pulmonary Unit, Massachusetts General Hospital, and Harvard Medical School, Boston
ABSTRACT. Of the 1,153 infants who completed moni-toring by Aug 1, 1984, through our program at Massachu-setts General Hospital, 76 infants had an initial apnea spell during sleep which was characterized by a change in tone and color, was unresponsive to repeated vigorous stimulation, and was terminated only after mouth to mouth resuscitation. The infants were hospitalized for observation and evaluation, and no cause could be iden-tified. All were discharged on a home apnea or cardiores-piratory monitor, and subsequent episodes of apnea and/ or bradycardia were reviewed. We grouped infants based
on the intervention used to terminate subsequent
epi-sodes: Group 1, resuscitation; group 2, vigorous stimula-tion; and group 3, neither resuscitation or vigorous stim-ulation. There was no significant difference in clinical features or in the results of the initial evaluation in groups
1 and 2, compared with group 3. However, the mortality
rate was significantly higher in group 1 (4/13) and group 2 (3/12) than in group 3 (3/51) (P < .007). Siblings of
victims of sudden infant death syndrome (n = 8) were at
a significantly higher risk of an adverse outcome (two
deaths and four resuscitations) than nonsiblings (P < .02). A seizure disorder that developed during monitoring was associated with a high mortality (4/11 v 6/65, P < .02). We conclude that these relatively rare infants who have sleep-onset apnea that responded only to resusci-tation and have a subsequent similar episode or are siblings of victims of sudden infant death syndrome or develop a seizure disorder during monitoring have a very high risk of dying (31%, 25%, and 36% respectively). Physical examination, past medical history other than siblings of sudden infant death syndrome status, and
laboratory evaluation on presentation were not helpful in predicting outcome in these 76 infants. Pediatrics 1986;77:495-499; sudden infant death syndrome, resusci-tation, sleep apnea.
Sudden infant death syndrome (SIDS) remains the most common cause of death in the first year
Received for publication May 6, 1985; accepted July 12, 1985. Reprint requests to (D.K.) Pediatric Pulmonary Unit, Massa-chusetts General Hospital, Boston, MA 02114
PEDIATRICS (ISSN 0031 4005). Copyright © 1986 by the American Academy of Pediatrics.
of life following the neonatal period. Because at
present the etiology of SIDS is unknown, some
researchers have focused on physiologic studies in
subsets of infants who are at high risk for SIDS.
One such subset is composed of infants who have had a “near miss” episode.’3 The extent of risk that
is published varies”36 perhaps because of different
definitions of near miss. The purpose of this study
was to determine features associated with the
high-est risk of death in our near miss population.
PATIENT SELECTION AND METHODS
The infants in this study were selected from a
population of 1,153 patients who have been
elec-tronically monitored by parents under our
super-vision for an unexplained episode of apnea
occur-ring between Jan 1, 1973, and May 1, 1984. All
concluded monitoring by Aug 1, 1984. We identified 76 infants, who had experienced an unexplained
episode of sleep apnea accompanied by a change in
tone and color, which was unresponsive to repeated
vigorous stimulation and was terminated only after
mouth to mouth resuscitation. All infants were
hospitalized for observation and evaluation, and no
cause for the apneic episode could be identified.7 Further studies including head computed tomo-graphic scan, cranial ultrasound, esophageal pH probe, direct laryngoscopy, and bronchoscopy were carried out as indicated. A pneumogram was
con-sidered abnormal if it showed (1) apnea 15
sec-onds for infants of >37 weeks’ postconceptional age
and 20 seconds for infants of 37 weeks’
postcon-ceptional age; (2) periodic breathing during 3.5%
of sleep time from 0 to 2 months of age, 2.5%
from 2 to 3 months of age, and 2.0% for >3 months
of age; or (3) bradycardia defined as a heart rate
80 beats per minute for 10 seconds from 0 to 1
month ofage, 70 beats per minute for 10 seconds
from 1 to 3 months of age, and 60 beats per minute
were discharged on a home cardiorespiratory
mon-itor (Air Shields apnea monitor or Healthdyne in-fant monitor) and caretakers were instructed in infant cardiopulmonary resuscitation, methods of observation and intervention, and the use of the
monitoring equipment. We also taught them to respond to the alarms by initially observing the infant and to intervene immediately if there was a color change. If there was no color change, they
were instructed to observe and, if there was no
spontaneous resolution of the apnea and/or
brady-cardia within 10 seconds, to gently stimulate the infant. If this failed, they were instructed to
vigor-ously stimulate the infant while supporting the head. If this did not resolve the apnea and/or
brad-ycardia, we instructed them to begin mouth to
mouth resuscitation or full cardiopulmonary
resus-citation (CPR) if necessary.
Subsequent episodes of severe apnea and/or
bradycardia were reviewed immediately by the
pro-gram physician on call. Although these data are based on parental reporting, it is unlikely that they represent an overestimation of the number and
severity of the episodes because only those episodes
during which the physician determined that the infant received appropriate graded intervention
were included in this study. The infants were then placed into three groups based on subsequent
epi-sodes of apnea and/or bradycardia during home
monitoring. Group 1 included those infants who received resuscitation to terminate a subsequent
episode, group 2 were infants who received vigorous stimulation, and group 3 were infants who did not
require resuscitation or vigorous stimulation for
subsequent episodes of apnea and/or bradycardia. Deaths were not included as a subsequent episode of resuscitation or vigorous stimulation.
Monitor-ing was discontinued when the following criteria
were fulfilled: (1) no apnea requiring any type of
stimulation for 2 consecutive months, (2) no apnea
requiring resuscitation or vigorous stimulation for
3 consecutive months, and (3) resolution of abnor-malities on pneumogram or polygraphic recordings.
STATISTICAL
ANALYSIS
The differences in numerical variables
(gesta-tional age, birth weight, chronologic age at
presen-tation) between the groups were tested with the Wilcoxon test. For categorical variables (presence
or absence of abnormal pneumogram,
hypoventila-tion, abnormal EEG, seizure disorder, sibling of
SIDS status, and survival v death) a
x2
contingencytest for association was made between subgroups.
A relative risk with confidence interval was calcu-lated for the variables. Using the above parameters, we compared infants who died with infants who
survived. We accepted P < .05 as excluding the null
hypothesis.
RESULTS
The clinical data of the patients are presented in Table 1. The results of the initial evaluation in the
hospital revealed that 22 infants had
gastroesopha-geal reflux, six infants had hypoventilation (mean end-tidal PCo2 46 mm Hg during sleep), and 13 infants had abnormal pneumograms (prolonged ap-nea, bradycardia, and/or increased periodic
breath-ing). Two infants had both hypoventilation and
abnormal pneumograms. There was no significant difference in birth weight, gestational age, age of presenting episode, or in the incidence of pneumo-gram abnormalities, hypoventilation, or gastroe-sophageal reflux between groups 1 and 2 together when compared with group 3. Thirteen infants had a subsequent episode that resolved with resuscita-tion (group 1), and 12 had an episode that resolved with vigorous stimulation (group 2). Ten infants died between 2 and 46 weeks after the initial spell, yielding a mortality rate of 13.2%. The mortality rate was higher if subsequent spells required resus-citation or vigorous stimulation than if they did not
(P < .007) (Table 2).
Differences in past medical history including
his-tory of prematurity or in the results of diagnostic evaluations did not separate deaths from survivors. Circumstances of death are shown in Table 3 and clinical data on these infants in Table 4. All infants died suddenly and unexpectedly and autopsy find-ings, performed on six infants, were consistent with SIDS. Circumstances of death on the other four
infants were reviewed, and no other etiology for death could be identified.
The eight siblings of SIDS victims were at a significantly higher risk than the 68 infants who had no family history of SIDS (P < .02). Two infants died after an unsuccessful resuscitation with the second episode and four had successful subsequent resuscitations.
During the monitoring course, a seizure disorder evidenced by tonic clonic activity developed in 11 infants, who had a normal neurologic assessment
TABLE 1. Clinical Data of 76 Infants With Severe Ap-nea of Infancy*
Sex (M/F) 44/32
No. of preterm infants (37 wk) 15
Gestational age (wk) 40 (28-43)
Birth wt (kg) 3.28 (0.94-4.54)
Age at presenting episode (wk)
Chronologic (N = 76) 8 (0.43-36)
Postconceptional preterm 40.5 (37-69) infants (n = 15)
TABLE 2. Mortality Rates in 76 Infants With Severe Apnea of Infancy
No. of Infants
Mortality Rate (%)
All infants 76 13.2
Infants with repeat sleep apnea episodes
Requiring vigorous stimulation 12 25
Requiring resuscitation 13 30
Siblings of SIDS victims 8 25
Infants with seizures and repeat episodes 7 57.1
requiring vigorous stimulation and
re-suscitation
TABLE 3. Circumstances of Death in Ten Infants
Circumstance No. of Infants
Infants who were off monitor at the time of death
Unknown
Infants who were on monitor at
the time of death
Delay in responding to alarm Unsuccessful resuscitation 1 1 8 4 4
and a normal EEG on presentation. All were treated with anticonvulsant drugs, and six infants subse-quently had an abnormal EEG. Seven of these
infants had at least one subsequent episode requir-ing resuscitation or vigorous stimulation and four of these seven died, so that the mortality rate was
significantly higher in infants with seizures (Table 5) than in those in whom seizure activity did not develop (4/11 v 6/65, P < .02).
A relative risk, with 95% confidence interval, for death in infants with subsequent severe episodes or with seizure disorder was calculated (Fig 1). Be-cause of the small number of siblings of SIDS victims who died, we calculated a relative risk, with 95% confidence interval, for an adverse outcome (resuscitation or death) in this group (Fig 2).
Although the average time interval between
mi-tial presentation and death was 15 weeks, all infants who had repeat episodes, excluding deaths, had experienced the first repeat episode within 2 weeks of hospital discharge. In three infants, the only subsequent episode resulted in death and, therefore, they were placed in group 3. Their deaths occurred at 1.5 weeks, five days, and 12 weeks after discharge. The latter two were siblings of SIDS victims (Table
4).
The duration of home monitoring was increased
as the seriousness of repeat apneic episodes
in-creased. The average duration was 16.5 months in
group 1, 13.3 months in group 2, 4.25 months in group 3, and the average for all 76 infants was 6.9
months.
DISCUSSION
Previous reports of mortality rates in near miss SIDS infants have varied from 0% to 4#{216}%1681O
Differences in definitions of near miss may account for the variance in these rates. In this study, we report the outcome for a relatively rare group of
infants who present with an unexplained episode of
apnea during sleep that was unresponsive to
vigor-ous stimulation but resolved after resuscitation. We
found the mortality rate for these infants to be
13.2%. If these infants have a subsequent episode requiring resuscitation or vigorous stimulation dur-ing home monitoring, the mortality rate increases to 31% and 25%, respectively, which is relatively high compared with those who have only minor subsequent episodes responding to gentle stimula-tion or no subsequent episodes (5.8%).
Studies on the incidence of SIDS in a subsequent sibling of a victim of SIDS has varied from 4.6 to
22 per 1,000.11,12 In this study, siblings who
pre-sented with an episode of sleep apnea requiring
resuscitation had a high mortality rate (25%) and
a high incidence of subsequent sleep apneic
epi-sodes requiring resuscitation (50%). Although these
results must be viewed with caution because of the
small number of siblings involved, it is apparent that this subgroup of symptomatic siblings of SIDS victims has a higher risk for SIDS than was found in previous studies.
The highest mortality rate (57%) occurred in
infants who had repeated severe episodes requiring
resuscitation or vigorous stimulation who
subse-quently developed a seizure disorder. In all
in-stances, there was seizure activity unrelated to ap-neic spells. Their apneic spells were difficult to control despite adequate anticonvulsive drug treat-ment. Although the numbers are small, these data
suggest that a seizure disorder can be associated
with sudden death. The lowest mortality rates (2.1%) occurred in those infants who were not
siblings of SIDS victims and had no recurrent se-vere apneic episode (1/47). The only death in this group occurred 1 week after hospital discharge.
1.
I Episodes 0 No Episodes
. Seizures 0 No Seizures
o.8
10.6
0.4
0.2
SibIing ofSIDS O Non-Sibling
6/8
22/68
#{188}.
7/25
4/If
3/51
6/65
Fig 2. Relative risk for an adverse outcome
(resuscita-tion or death) in siblings of victims of sudden infant death syndrome (SIDS).
Fig 1. Relative risk for death in infants with subsequent severe episodes or with seizures.
TABLE 4. Clinical Data on Infants Who Died
Median Range
Age at presenting episode (wk) 5.5 0.7-32
Age at the time of death (wk) 22.5 4-48 Time interval (wk) between:
Presenting episode and death 8.25 2-46
Initial and second episode 2* 1-3.5
Hospital discharge and subsequent episode 1.25* 0.57-2
Last episode and death 1.75 0.29-17
* Values on all infants excluding 1 outlier (SIDS sibling who died 17 weeks after initial
episode without a repeat episode).
TABLE 5. Clinical Data of 11 Infants With Seizures
EEG Findings Repeat
Normal Abnormal Episodes
Total No. of infants 5 6 7
No. of infants who died 3 1 4
causes for failure of CPR in these infants include
ventricular fibrillation, asystole, complete upper airway closure in the hypopharynx or the larynx,’2 or poor CPR technique. In four cases, a delay in the caretaker’s response was identified. In one such
infant, a pneumogram recording during the episode
that resulted in her death had clusters of bradycar-dia to a low of nine beats per minute and intermit-tent prolonged apnea to a maximum of 75 seconds before the motion artifact of resuscitation was
re-corded. However, in most infants with repeat severe episodes whose intervention of resuscitation or vig-orous stimulation occurred with no delay, the in-fants survived and ultimately ceased having pro-longed sleep apnea (18/25). Thus, we believe that
timely caretaker intervention guided by electronic
monitoring will avoid death in most, but not all, infants who will have repeat severe episodes requir-ing resuscitation or vigorous stimulation.
mechanisms involved and to decrease the mortality rate.
The current methods of testing infants at high risk for SIDS are not specific or sensitive enough to prospectively identify those infants in this subset of infants with apnea who are most susceptible to SIDS. Abnormal results of pneumograms, sleep polygraphs, EEGs, and esophagograms bore no cor-relation with the severity of subsequent episodes or outcome. Because the mortality rate is extremely high, we recommend that, after a severe subsequent episode, these infants have a thorough reevaluation and be discharged only if parents have been made aware of the high mortality rate and the extraor-dinary responsibility of the home care of the infant. They should also have thorough and frequent re-views of infant CPR and home monitor training and adequate professional support. If all these cri-teria cannot be met, long-term hospitalization should be considered so that adequate medical per-sonnel and equipment will be available if subse-quent respiratory or cardiac arrests occur.
ACKNOWLEDGMENTS
The authors thank David Carley, PhD, for assistance
in statistical analysis, Lisa Knox for technical help, and
Maria Nazzaro for manuscript preparation.
REFERENCES
1. Kelly DH, Shannon DC, O’Connell K: Care of infants with near-miss sudden infant death syndrome. Pediatrics 1978;61:51 1-514
2. Steinschneider A, Weinstein SL, Diamond E: The sudden infant death syndrome and apnea/obstruction during neo-natal sleep and feeding. Pediatrics 1982;70:858-863
3. Ariagno RL, Guilleminault C, Korobkin R, et al: ‘Near-miss’ for sudden infant death syndrome infants: A clinical prob-lem. Pediatrics 1983;71:726-730
4. Keens TG, Dennies PC, Lew CD, et al: Risk of subsequent apnea in infants surviving near-miss sudden infant death syndrome, abstracted. Pediatr Res 1982;352:16
5. Hodgman JE, Hoppenbrouwers T, Geidel 5, et al: Respira-tory behavior in near-miss sudden infant death syndrome. Pediatrics 1982;69:785-792
6. Rosen CL, Frost JD Jr, Harrison GM: Infant apnea: Poly-graphic studies and follow-up monitoring. Pediatrics 1983;71:731-736
7. Guilleminault C, Korobkin R: Sudden infant death: Near miss events and sleep research: Some recommendations to improve comparability of results among investigators. Sleep 1979;1:423-433
8. Steinschneider A: Prolonged apnea and the sudden infant death syndrome: Clinical and laboratory observations. Pe-diatrics 1972;50:646-654
9. Duffty P, Bryan MH: Home apnea monitoring in ‘near-miss’ sudden infant death syndrome (SIDS) and in siblings of SIDS victims. Pediatrics 1982;70:69-74
10. Brooks JG: Apnea of infancy and sudden infant death syn-drome. Am J Dis Child 1982;136:1012-1023
11. Peterson DR, Chin NM, Fisher LD: The sudden infant death syndrome: Repetitions in families. J Pediatr 1980;97:265-267
12. Irgens LM, Skjaerven R, Peterson DR: Prospective assess-ment of recurrence risk in sudden infant death syndrome siblings. J Pediatr 1984;104:349-351