INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Keywords: sleep apnea; sudden infant death syndrome

Sudden infant death syndrome (SIDS) has been tentatively attributed to a variety of factors, including anomalies in respiratory control. A possible link was reported between SIDS and a family history of obstructive sleep apnea (OSA) (1). Anecdotal reports have also associated obstructive breathing events with SIDS. In 1979, Guilleminault reported that a 21-wk-old girl died of SIDS some 30 h after undergoing a nighttime sleep study that revealed frequent episodes of mixed and obstructive apnea (2). A 4-wk-old infant observed by Albani and coworkers died of SIDS 3 wk after frequent episodes of obstructive and mixed apnea were monitored during a nap study (3). Roberts and colleagues recorded episodes of mixed apnea in a preterm infant who later died of SIDS (4).

Planned sleep studies gathered evidence that infants who ultimately died of SIDS were characterized by frequent occurrences of OSA (2). In one study, incidents of obstructive and mixed sleep apnea were seen in 8 of 11 such infants but in only 2 of 22 matched control infants (6). Another study also found obstructed breathing events in 23 of 30 infants who eventually died of SIDS but in only 9 of 60 matched control infants (7). These studies were, however, based on a limited number of SIDS and control infants (4).

Epidemiologic studies showed that most cases of SIDS occurred after the first 8 wk of life (9), and that male infants were at greater risk for SIDS than were female infants (10). We investigated whether the distribution of obstructed breathing events correlated with the age and sex characteristics of SIDS victims. To do this, we collected sleep recordings of a group of 40 infants who subsequently died of SIDS and of 607 healthy infants who were matched for age and sex.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Infants

Between January 1977 and January 2000, 27,000 infants between the ages of 4 and 19 wk were studied polygraphically in 10 Belgian sleep laboratories. The sleep studies were done during a single night to assess various aspects of the infant's sleep characteristics or to alleviate parental anxiety about sleep apnea. All studies were conducted according to the same standard recording protocol. Forty recordings were made on infants who died within days or weeks after the sleep studies. Their deaths were sudden and unexpected, remained unexplained despite a complete postmortem examination, and were attributed to SIDS. The proportion of 1.48 deaths per 1,000 infants studied is comparable to the overall frequency of deaths from SIDS observed in Belgium during the same period, of 1.73 per 1,000 live births (11). The total group of 40 victims of SIDS was composed of six infants studied because they were siblings of SIDS victims, nine infants studied after a life-threatening event, and 25 infants admitted because of parental anxiety about sleep apnea or following various sleep studies. The group consisted of 12 girls and 28 boys. No infant was undergoing sleep monitoring at the time of death. The sleep and breathing characteristics of some of the SIDS victims had been reported previously (6, 7). The sample of 27,000 subjects included 1,023 healthy infants born at term after a normal gestation, with no family or personal history of apnea or SIDS and who survived the first year of life. Their sleep and breathing characteristics were reported separately (12). Of these infants, 607 subjects who matched the SIDS victims for sex and age at the time of recording were selected to form the control group. At the time of study, all infants were healthy and receiving no medication.

Polygraphic Recordings

The infants were admitted to the sleep laboratory for an 8-h nighttime monitoring session performed in a quiet and darkened room at an ambient temperature ranging from 20° C to 23° C. All infants slept in the supine position and without restraints. Recording began at about 9:00 P.M. The infants were observed continuously during recording. They were fed on demand, and their behavior and any nursing intervention were charted. The following recordings were made simultaneously: two scalp electroencephalograms (EEGs) with central and occipital leads (C4/A1 and C3/A2), two electrooculograms (EOGs), and an electrocardiogram (ECG). Thoracic respiratory movements were measured by impedance, and airflow was measured with thermistors taped under both nostrils and to the side of the mouth. Gross body movements were measured with an actigram placed on one arm. The data were collected with a computerized infant sleep recorder (Alice Recording System III; Healthdyne, Pittsburgh, PA).

Data Analysis

The polygraphic recordings were analyzed at 30-s intervals and classified as non-rapid-eye-movement (NREM) sleep; rapid-eye-movement (REM) sleep; indeterminate sleep; or wakefulness, according to recommended criteria (13). Episodes of apnea were observed and recorded by one of the authors (I.K.) without knowledge of the patient's identity. An apneic episode was noted when it lasted 3 s or more. Central apnea was defined as the simultaneous recording of flat tracings by both the strain gauges and the thermistors (Figure 2A). Obstructive apnea was defined as occurring when continuous deflections were shown by the strain gauges while a flat tracing was recorded by the thermistors (Figure 2B). To avoid artificially high airflow recordings due to thermistor displacement, we rejected incidents of obstructive apnea preceded by body movements, crying, or sighing. The possibility thus exists that some episodes of obstructive apnea were unduly excluded from the study. Mixed apnea was recorded when central apnea, defined as simultaneously flat tracings from both the strain gauges and the thermistors, was directly followed by obstructive apnea (Figure 2C). The duration of mixed apnea was computed by adding the durations of both types of apnea. The frequencies of obstructive and mixed apneas were measured by dividing the total number of episodes of each type of apnea by the total sleep time in minutes and multiplying this by 60. The frequencies of obstructive and mixed apneas in the SIDS and the control subjects were compared by separating the total population sample into two groups, aged respectively 4 to 8 wk and 9 to 19 wk. Statistical assessments were made with the Mann-Whitney U test and chi-square test with Yates's transforms. The level of significance was p < 0.05. The aim and methodology of the study were approved by the Ethics committee of the Free University of Brussels, and were explained to the parents of the subjects, who gave their informed consent.

View larger version (36K): [in this window] [in a new window]   View larger version (39K): [in this window] [in a new window]   View larger version (36K): [in this window] [in a new window]   Figure 2.   (A) Central apnea. (B) Obstructive apnea. (C ) Mixed apnea.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The main characteristics of the SIDS and the control infants are shown in Table 1. No differences were seen between the two groups of infants with regard to sex, gestational age, age at recording, or total sleep time. The frequency of episodes of central apnea per hour of sleep did not differentiate the SIDS infants (median of nine episodes of apnea per hour; range: 5 to 22 episodes) from the control infants (median of 8.5 episodes of apnea per hour; range: 6 to 20 episodes).

Obstructive and mixed sleep apnea occurred mainly in REM and undetermined sleep. Among the infants in the 4- to 8-wk age group who ultimately died of SIDS, 60.6% of episodes of obstructive apnea occurred in REM sleep and 26.2 % in undetermined sleep; in the 9- to 19-wk age group, 76.2% of incidences of apnea were in REM sleep and 19.4% in undetermined sleep. Among infants in the 4- to 8-wk age group in the control group, 67.8% of episodes of obstructive apnea were seen in REM sleep and 27.3 % in undetermined sleep; in the 9- to 19-wk age group, 74.8% of apneas occurred during REM sleep and 18% in undetermined sleep. Among the 4- to 8-wk group of infants who ultimately died of SIDS, 50% of episodes of mixed apnea occurred in REM sleep and 20.6% in undetermined sleep, whereas in the 9- to 19-wk age group, 76.2% of episodes of mixed apnea were observed in REM sleep and 18.7% in undetermined sleep. Among the 4- to 8-wk group of control infants, 65.7% of episodes of mixed apnea occurred in REM sleep and 26.4% in undetermined sleep, whereas in the 9- to 19-wk group, 75.2% of mixed apnea episodes occurred in REM sleep and 18.4% in undetermined sleep. No difference in sleep-state distribution was found between the infants who ultimately died of SIDS and the control group.

In comparing the subsequent SIDS victims with the control infants, we found no significant difference in the percentage decrease in heart rate (HR) associated with either OSA or mixed sleep apnea. The median decrease in HR from baseline for obstructive apnea was 14% (range: 1% to 24%) in the SIDS infants and 15% (range: 3% to 27%) in the control infants. For mixed apnea, the median decrease in HR was 21% (range: 6% to 28%) in the SIDS infants and 23% (range: 3% to 32%) in the control infants.

As shown in Table 2, the frequency of both obstructive and mixed apneas was significantly greater in those infants who subsequently died of SIDS than in the control infants. This difference was more evident in apneas that lasted 10 s or less. Infants who later died of SIDS tended to have longer obstructive and mixed apneas, but the differences did not reach statistical significance. Apneas lasting longer than 10 s were rare in both the SIDS and control infants. The longest obstructive apnea was 17 s, the longest obstructive apnea 19 s. Both were seen in 4- to 8-wk-old infants who ultimately died of SIDS.

We found a significantly greater number of infants with OSA among the SIDS than among the control infants (30 of 40 SIDS infants, or 75%; and 255 of 607 control infants, or 42%; p < 0.001). Significantly more infants who ultimately died of SIDS had mixed apnea than did infants in the control group (21 of 40 SIDS infants, or 52.5%; and 139 of 607 control infants, or 22.9%; p < 0.001).

For the ages of 4 to 8 wk, the number of infants with OSA was not significantly different between the SIDS and the control groups (Table 3). For the ages of 9 to 19 wk, a significantly greater number of infants with obstructive apnea was found among the SIDS infants (76.9%) than among the control infants (41.1%) (p < 0.001). In Figure 1, the frequencies of obstructive apnea seen in the 40 infants who ultimately died of SIDS are plotted against the 90th, 75th, and 50th percentiles of apnea frequency measured for the 607 control infants.

View larger version (14K): [in this window] [in a new window]   Figure 1.   Frequencies of episodes of OSA with age, in both the future SIDS victims (dots) and control infants (percentiles).

Between the ages of 9 and 19 wk, a significantly greater number of infants whose incidents of OSA fell above the 90th percentile of the control group was found in the SIDS than in the control group (8 of 26 SIDS infants, or 30.8% versus 50 of 540 control infants, or 9.3%; p < 0.001). The number of infants with mixed apnea was significantly greater in the group that subsequently died of SIDS, both before (p = 0.012) and after (p = 0.029) the age of 9 wk.

Among subjects younger than 9 wk, there was no difference in the number of boys with OSA in the SIDS and in the control groups (Table 3). However, for the ages of 9 to 19 wk, significantly more boys than girls with OSA were found among the SIDS than among the control infants (p = 0.006). There was no difference between the two groups of infants in the proportion of girls with apnea, either before or after the age of 9 wk.

The frequency of obstructive and mixed apneas was significantly greater among boys who went on to die from SIDS than among boys in the control group, both before and after 9 wk of age (Table 3). No significant difference was found for girls in the two study groups.

Differences in the frequency of occurrence of apnea were compared before and after the age of 9 wk. In both the SIDS and control groups of infants, the frequency of episodes of sleep apnea was lower in the 9- to 19-wk age group than in the 4- to 8-wk age group. For obstructive apnea, the decrease in frequency of apnea from before to after the age of 9 wk was significantly smaller among infants who subsequently died of SIDS than among the control subjects (p < 0.001). When analyzing the data according to gender, we found a difference in apnea frequency for boys only. No gender-related difference was found for mixed apnea.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The results of the present study are reminiscent of those of previous studies which found that episodes of obstructive and mixed apnea occurred mainly during REM sleep (2, 5, 6, 12, 14), were most frequent during the first 2 mo of life (12), and decreased with age (3, 5, 8). Apneic episodes lasted less than 15 s (3, 5, 12, 14), were rare in healthy infants (5, 15), and were found significantly more often in infants who subsequently died of SIDS than in age-matched control infants (6, 7).

The decrease in frequency of apneic episodes with age was smaller in the SIDS infants than in the control group. After the age of 9 wk, the future SIDS victims had significantly more occurrences of apnea than did the control infants. These differences could be related to differences in factors favoring obstructive apnea, such as genetic predisposition (1), anatomic characteristics of the airways or the face (16, 17), or neuronal control of the airways (13, 18). Such mechanisms were not investigated in the present study. Differences in environmental conditions known to favor apnea, such as infections (23), sleep deprivation (24), or use of sedatives (25) were, however, excluded from the study. The observation that apneic events became more frequent with age in the infants who subsequently died of SIDS than in the control group was reminiscent of reports of an increased risk of SIDS after the age of 9 wk (26).

The greater frequency of OSA that we found in boys is consistent with previous reports of a greater frequency of OSA in 2- to 3-mo-old healthy boys than in age-matched girls (9, 14, 27). These findings were tentatively attributed to sex-related differences in metabolic rates, maturation of the central nervous system (28), sex hormones (29), or airway anatomy (30). The greater frequency of apnea seen among boys who later died of SIDS was reminiscent of a higher risk of SIDS in boys (10).

Evaluation of the causes of SIDS was beyond the scope of this study. The deaths could be attributed to a variety of factors independent of the presence of OSA. The overlap of apnea frequency between the SIDS subjects and the control group, plus the observation that some infants who subsequently died of SIDS did not show more frequent obstructive apnea than did the control infants, as well as the known protective effect of the supine position during sleep, which does not affect the frequency of episodes of sleep apnea (33), raises doubts about the role of obstructive apnea as a cause of death. Sleep apnea could, however, have been a contributory factor in the death of some infants. In one study, changes in the HR monitored at the time of death from SIDS were attributed to obstructive apnea (34). Some postmortem findings in cases of SIDS have been attributed to repeated obstructions of the upper airways, such as petechiae in the pleural cavity, pulmonary edema (34), increased smooth-muscle thickness in the airway (35), or thickening of the basement membranes of the vocal cords (36).

In conclusion, we found that as compared with a large group of healthy infant controls, the future victims of SIDS were characterized by frequent airway obstructions during sleep and a smaller rate of decrease in the frequency of OSA with age. This finding was seen mainly in boys. The present report provides further indirect evidence for a sleep-related impairment or a delayed maturation of respiratory control in some infants who eventually die of SIDS.