Statistical evaluation of respiratory control in infants to assess possible risk for the sudden infant death syndrome (SIDS)

We examined the application of statistical parameters related to respiratory control in order to assess possible risk for the sudden infant death syndrome (SIDS). The suitability of these parameters was tested in 287 infants between 1 day and 6 months of age. Respiration of the infants was recorded during sleep, and duration and frequency of apneic spells was evaluated. All infants were patients of the department of pediatric surgery, and were free of pulmonary, cardiac or cerebral diseases. Of the investigated infants, 54 were assigned as an at risk group, for they had already had demonstrable serious or life threatening apneas or respiratory problems during anesthesia.We computed the frequency distribution of sleep apnea durations, and calculated the respective probabilities for the occurrence of apnea of a certain duration according to age. By analogy with Shannon's well known definition of average information content we defined an entropy term, which formally agrees with the definition of entropy in thermodynamics. This statistical procedure seems to be well suited for the identification of disorders in respiration control possibly related to a risk for SIDS. All at risk infants were treated with aminophylline, and the respiratory state improved significantly in nearly all cases.In some infants QT intervals were evaluated from the ECG recordings during sleep, but no significant differences between risk and control infants could be demonstrated.This study was supported by the Austrian Research Fund     

Sleep apnea in infants who succumb to the sudden infant death syndrome

Previous studies have shown the frequency of respiratory pauses to be altered in groups of infants at risk for the sudden infant death syndrome (SIDS). In this study, we assess the frequency of apneic pauses during quiet sleep and rapid eye movement sleep in control infants and infants who subsequently died of SIDS. Sleep states were identified in 12-hour physiological recordings of SIDS victims and matched control infants, and the number of respiratory pauses from 4 to 30 seconds in duration was computed for quiet sleep and rapid eye movement sleep. SIDS victims 40 to 65 days of age showed significantly fewer apneic pauses than did age-matched control infants across the two sleep states. Fewer short respiratory pauses accounted for most of the reduction in number of apneic events in the SIDS victims during both sleep states. During the first month of life, SIDS victims did not differ significantly from control neonates on this measure. The finding that this respiratory difference exists during the second month of life, just before the period of maximal risk for SIDS, but not earlier, may have implications for the etiology of SIDS deaths.     

Respiratory control in infants at increased risk for sudden infant death syndrome.

There is much debate relating to possible abnormalities in respiratory control mechanisms in infants considered at increased risk for sudden infant death syndrome (SIDS). The P0.1 occlusion technique was used to assess the central respiratory response to hyperoxic hypercapnia during quiet sleep in 21 normal infants, 13 siblings of SIDS victims, and 17 infants with apparent life threatening events. The slope of P0.1 plotted against carbon dioxide concentration increased exponentially with age, independent of body weight in each group. Birth weight has a significant effect on slope with a lower weight predisposing to a lower slope. Siblings as a group had a significantly lower slope at any given age than normal infants, whereas the infants who had had apparent life threatening events were not significantly different from the controls. As intragroup variation in both siblings and control groups greatly exceeded the significant intergroup differences observed, the technique cannot identify individual infants as belonging to one or other group.     

Sinus tachycardia in term infants preceding sudden infant death

The quantities of sinus tachycardia in 24-h recordings of the electrocardiogram from 16 full-term infants (37 weeks gestation) who were subsequently victims of the sudden infant death syndrome (SIDS), from 230 randomly selected age-matched full-term survivors and from 64 full-term survivors matched for age and birth weight were measured by computer and manual analysis techniques. Of 16 infants dying of SIDS, 7 had elevated levels of sinus tachycardia (>95th centile in controls) (P<0.01). Although high levels of sinus tachycardia might be of value in identifying infants at high risk of SIDS, these encouraging findings must first be validated by further prospective studies.Abbreviations IHR instantaneous heart rate - SIDS sudden infant death syndrome - ECG electrocardiogram     

Heart rate and respiratory rate differences between preterm and full-term infants during quiet sleep: possible implications for sudden infant death syndrome

The heart rate and respiratory rate of eight preterm (average gestational age, 33.3 +/- 2.2 weeks) and 13 full-term infants were determined during quiet sleep at home during the first year of life. Both heart rate and respiratory rate were greater in preterm infants throughout the first six months. The difference was maximum at age 10 weeks (21.2 beats per minute and 13.7 breaths per minute), with the difference being statistically significant, at least at the P less than .01 level at ages 10, 12 and 14 weeks, while P greater than or equal to .01 at all other ages. The 10- to 14-week period is precisely the same period during which the incidence of sudden infant death syndrome (SIDS) has been reported to be maximum. These results support the concept that SIDS is linked to a vulnerable phase of cardiorespiratory maturation.     

Care of infants with near-miss sudden infant death syndrome

The purpose of this study was to evaluate the effectiveness of home management of life-threatening apnea in infants with near-miss sudden infant death syndrome (SIDS). A total of 84 infants were monitored with apnea monitors for an average of seven months (range 1 to 27). A group of 27 infants had episodes of apnea requiring resuscitation during home monitoring, all of whom were successfully resuscitated by their parents on at least one occasion using bag and mask resuscitation, and 17 infants required more than one resuscitation. Subsequent resuscitation was unsuccessful with four infants. Infants who experienced more than ten episodes of prolonged sleep apnea (apnea longer than 20 seconds) during home monitoring had a significantly increased risk of requiring resuscitation than other infants being studied. With the use of home monitoring of these infants and parents trained in cardiopulmonary resuscitation, the survival rate was 93.4%.     

Analysis of heart rate and respiratory patterns in sudden infant death syndrome victims and control infants

Retrospective analyses of patterns of breathing and heart rate variability obtained by visual inspection and spectral analysis of ECG and respiratory activity have provided markers associated with subsequent death in a referred population of infants at high risk for sudden infant death syndrome (SIDS). Such markers include breathing patterns characterized by excessive apneic pauses and periodic breathing, heart rate spectra characterized by increased low frequency oscillations, and respiratory activity spectra characterized by a widened "bandwidth" during regular breathing. To test whether such measurements could distinguish SIDS cases and randomly selected controls from a population study the data from 10 cases and 100 age-matched control subjects were analyzed blind. The code was disclosed after completion of the analysis. We found that none of the markers served to distinguish the SIDS cases from the controls in the population at large. This observation may indicate important physiological differences between infants destined to die in the referred high risk population and infants who die of SIDS at large. The possible reasons for our inability to identify the group of SIDS in the general population, as compared to the group of deaths in the referred high risk group are: (1) different disease processes in the two groups, (2) difference responses to the same disease process in the two groups, (3) a response reflecting the psychosocial setting of the referred high risk population, (4) methodological differences between this and previous studies. We conclude that these markers are not of value in screening the population at large.     

Laryngeal chemosensitivity: a possible mechanism for sudden infant death.

In 32 anaesthetized piglets 1 to 42 days of age the distal trachea was cannulated and pressure changes were recorded. The proximal trachea was cannulated for introduction of test fluids into the laryngeal area. Arterial pressure, heart rate, and central venous pressure were continuously recorded. Arterial blood samples were obtained at intervals and analyzed for PO2, PCO2, pH, and hematocrit. Normal saline produced no, or brief, transitory alterations of the respiratory pattern and arterial pressure. In contrast, instillation of distilled water produced apnea in 29 of 30 piglets. In 20 the apnea was sustained. Eleven died within approximately 50 minutes of asphyxia. An additional nine were expected to die with continuing apnea (PO2, 10 to 15 mm Hg; PCO2 greater than 100 mm Hg; pH smaller than 6.8) but the sequence was interrupted by replacement of water with saline. Twenty-three of 29 piglets showed an apneic response to cow's milk similar to that seen with distilled water. Seven died of asphyxia and an additional three showed sustained respiratory inhibition until milk was replaced with saline. The responses were completely abolished by superior laryngeal nerve (SLN) sectioning. Electrical stimulation of the SLN produced sustained apnea in seven of eight piglets tested. Studies in two 3-day-old lambs showed similar discrimination but only transitory apnea with water or cow's milk. Both died during SLN stimulation. Two ewes showed insignificant responses. These findings suggest a lethal reflex mechanism with implications for the SIDS problem.     

Early neonatal sudden infant death and near death of fullterm infants in maternity wards

In a large Swedish 2 1/2-year population of fullterm infants the incidence of early neonatal sudden death (ENSD) was 0.12 per 1000 newborn infants considered healthy at birth but presenting with sudden cardiovascular collapse between 6 and 100 hours after birth. In the total neonatal mortality the rate of ENSD was 11%. Well-known risk infants, e.g. prematurity, perinatal asphyxia, difficult delivery or congenital malformations were not included. In an inborn population of 20 123 fullterm newborns the rate of ENSD was 0.15 per 1000. In the same population near-missed ENSD occurred in 0.35 per 1000 liveborn infants. Possible causative factors were infections in 4 and anaemia in 2 cases. But in 7 of 13 infants thorough epidemiological examinations and postmortem sterile autopsies did not reveal the cause of death or near death. These are probably cases of sudden infant death syndrome hitherto unrecognized in the first days of life.     

Pneumograms in infants who subsequently died of sudden infant death syndrome

Victims of sudden infant death syndrome (SIDS) have occasionally been reported to have had prolonged apnea or an increased frequency of short apnea prior to their deaths. To examine the extent of these abnormalities, we compared pneumograms obtained in 17 infants who subsequently died of SIDS (10 with history of apnea) with those of 34 age- and sex-matched controls. The recordings were analyzed by a computer program that avoided observer bias. SIDS infants had significantly greater mean heart rate (P less than 0.05) and periodic breathing during quiet time (P less than 0.003) than control infants had. The apnea in SIDS infants tended to be more periodic than in control infants (P less than 0.002). In addition, the incidence of bradycardia was greater in SIDS (three infants) than in controls (none) (P less than 0.03). These differences suggest a disturbance of autonomic function prior to death in some victims of SIDS.     

Ventilatory chemosensitivity in parents of infants with sudden infant death syndrome

We measured ventilatory responses to progressive isocapnic hypoxia and to hyperoxic hypercapnia (CO2) using rebreathing techniques in 16 parents of infants with autopsy-confirmed sudden infant death syndrome (SIDS) and 18 control parents matched for age, sex, and body size. Response to ventilatory loading was assessed by repeating the CO2 test with an inspiratory flow-resistive load (16 cm H2O/L/sec). During loaded and unloaded CO2 tests, respiratory effort was also assessed by measuring the pressure generated in the first 0.1 second (P0.1) of the subsequent inspiratory effort after brief manual occlusion of the inspiratory line. Ventilatory responses of the parents of victims of SIDS to chemical and mechanical stimulation were not significantly different from those of control parents. Responses in both groups were similar to previously reported normal values. There was a linear increase in ventilation (VE) in response to hypercapnia and hypoxia and in P0.1 in response to hypercapnia. We found expected increases in P0.1/PCO2 and decreases in VE/PCO2 slopes during loaded breathing in all subjects, but no difference between groups. We conclude that parents of SIDS victims have normal ventilatory chemosensitivity and respiratory drive.     

Periodic breathing in infants with near-miss sudden infant death syndrome.

Twelve-hour nocturnal home recordings of respiration and heart rates were obtained during sleep in 32 infants with near-miss sudden infant death syndrome (SIDS) and in 32 control infants, and the recordings were analyzed for periodic breathing. An episode of periodic breathing was defined as three or more apneic pauses of three or more seconds. The duration of respirations interrupting the pauses was 20 seconds or less. Analysis revealed a statistically significant difference (P less than .001) between the two groups, using criteria of percent of periodic breathing episodes, number of periodic breathing episodes/100 min of recorded sleep time, average duration of all episodes, and duration of the longest episode of periodic breathing. It is concluded that periodic breathing is present in excessive amounts during sleep in infants with near-miss sudden infant death syndrome.     

Risk of sudden infant death syndrome among infants with in utero exposure to cocaine

To determine whether the risk of sudden infant death syndrome (SIDS) among infants exposed to cocaine in utero may be elevated, we assessed the risk for SIDS in a large, well-described, prospective cohort of infants whose mothers had or had not used cocaine during pregnancy. Of 996 women consecutively enrolled while registering for prenatal care, 175 used cocaine during pregnancy. Only one infant of the mothers who used cocaine died of SIDS, a risk of 5.6 in 1000, compared with four infants among the 821 nonexposed infants, a risk of 4.9 in 1000. The relative risk for SIDS among infants whose mothers used cocaine during pregnancy compared with those whose mothers did not use cocaine was 1.17 (95% confidence interval 0.13, 10.43), suggesting that there is no increased risk of SIDS among infants exposed in utero to cocaine.     

Cardiorespiratory function in 16 full-term infants with sudden infant death syndrome

Twenty-four-hour tape recordings of ECG and breathing movements from 16 term infants (greater than or equal to 37 weeks' gestation) who subsequently died of sudden infant death syndrome (SIDS) were compared with recordings from surviving infants from the same populations. Apneic pauses of varying durations, periodic and regular breathing patterns, heart and respiratory rates during regular breathing were measured. Only one of 16 full-term infants with SIDS had findings outside the range of age-matched control infants (an excess of periodic breathing patterns and an absence of regular breathing). When the first recordings of each of infants who died of SIDS, except one who had cyanotic episodes prior to death, were compared to recordings of survivors (six for each case) closely matched for age, gestation, and weight at birth, no differences in breathing patterns or heart or respiratory rates during regular breathing could be demonstrated. These particular measurements of cardiorespiratory function were, therefore, unable to identify the majority of full-term infants at risk for SIDS.     

The QT interval in aborted sudden infant death syndrome infants.

The QT interval was measured in 12 normal and 7 aborted sudden infant death syndrome (SIDS) infants in rapid eye movement (REM) and quiet sleep at monthly intervals through the age of 4 months. An accuracy of better than 2 msec was assured by high resolution of the digitized signal and calibration of each QT measurement with an accurately generated time code. In contrast to current speculations, the QT index was significantly smaller in the infants with aborted SIDS than in the normal infants in both REM and quiet sleep (P less than 0.05). In addition, as in normal infants, the QTc was smaller in REM than in quiet sleep (P less than 0.01). Although these results offer no support for the hypothesis that SIDS results from prolongation of the QT interval, they suggest that aborted SIDS infants have a functional abnormality in the autonomic nervous system.     

Inverse relationship between beta-endorphin immunoreactivity in cerebrospinal fluid and nucleus tractus solitarius in sudden infant death

In nucleus tractus solitarius (NTS) beta-endorphin (BEND) induces bradycardia and respiratory depression which have been reported to precede death in sudden infant death (SID). Of SID victims, 50% have elevated levels of beta-endorphin immunoreactivity (BENDI) in the cerebrospinal fluid (CSF), and 50% had undetectable levels. We therefore investigated the relationship of BENDI in the CSF to BENDI levels in the NTS area. This study included SID victims (CSF fromn=47, brain stem fromn=16), borderline SID victims (CSF and brain stem fromn=2), sudden death in childhood (CSF and brain stem fromn=1), and controls (CSF fromn=32, brain stem fromn=11). BEND in CSF and NTS area, after extraction, was measured by radioimmunoasay. High performance liquid chromatography was used for closer identification of BENDI. We found that the SID victims divided into two subpopulations, one having a relatively high BENDI level in CSF and one having no detectable level (P<0.01). Furthermore, an inverse relationship was found between BENDI level in CSF and BENDI level in NTS area in the SID victims (P<0.05). We conclude that increased BENDI level in CSF is associated with low BENDI level in the NTS area in 50% of SID victims. The low BENDI level in CSF is associated with low BENDI level in the NTS area in 50% of SID victims. The low BENDI level in the NTS area may be due to increased release of BEND.