Death within the first hours of life is rare, and only little is known about its cause and possible risk factors. The incidence of sudden infant death syndrome (SIDS), one of the most common causes of death within the first year of life, has decreased tremendously in the past twenty years thanks to the establishment of risk-reduction campaigns. As the incidence of early sudden unexpected death in infancy (ESUDI) has not changed, it is essential to identify risk factors and develop guidelines for prevention. Because of its rarity, new cases can be important and help shed light on these events.
We report three cases of unexplained death within the first 4 h of life after uncomplicated pregnancies and uneventful deliveries, all of which occurred in Tyrol, Austria, between 2006 and 2011. Characteristics of patients and their mothers as well as their clinical courses were retrospectively analysed. The cases are discussed along with the current literature on ESUDI. In an effort to find all relevant studies, the authors searched MEDLINE databases and Web of Science (Thomson Reuters) using the following terms: ‘early sudden infant death syndrome’, ‘skin-to-skin contact’ and ‘early sudden unexpected death in infancy’. Reference sections of the retrieved articles were cross-searched. After extensive literature research, we now give the first summary of all previously published cases of ESUDI.
We enrolled only articles dealing with cases of unexplained death occurring early after birth, in which a collapse with unknown cause led to death within the first 10 days or within 2 months thereafter because of hypoxic sequels. No limitations were set for publication status or date. We excluded cases that were designated ‘explained’ after investigation, articles not providing information in English and articles not differentiating between early death and early apparent life-threatening events (EALTE).
Our first case is a boy infant who was born at 40 + 4 weeks of gestation with a weight of 3050 grams as the first child of a 29-year-old woman following an uncomplicated pregnancy. The child showed immediate cardio-respiratory adaptation after birth and an Apgar score of 9/10/10. At 3.5 h after birth, a nurse discovered the cyanotic baby in prone position between his mother's breasts. The mother was not asleep, but did not notice the deterioration of her child. Cardiopulmonary resuscitation, which was started immediately, was successful and therapeutic hypothermia was commenced. The boy's situation deteriorated and death occurred 21 h after birth because of multiorgan failure. The autopsy showed hypoxic encephalopathy but no malformations or any other obvious cause of death. Post-mortem screening for metabolic disorders including screening for organic amino acids in urine as well as for amino acids in blood revealed no other abnormalities than those explained by asphyxia.
In our second case, a newborn girl aged 41 + 1 weeks of gestation and weighing 3790 grams was found cyanotic and lifeless lying next to her 36-year-old mother 2 h after spontaneous birth. After an uncomplicated pregnancy, the girl was born with an Apgar score of 10/10/10. Also in this case, the mother was not deeply asleep, but only dozing in the delivery room and not aware of the baby's respiratory arrest. Resuscitation was commenced immediately upon detection, but the baby died fourteen days after the event from respiratory insufficiency due to encephalopathy. The autopsy showed signs of severe asphyxia, especially in the lung and the adrenal glands, but no cause of death could be identified.
The third case was a girl infant who was born to a 34-year-old woman at 38 + 2 weeks of gestation with a birth weight of 3480 grams following an uneventful first pregnancy. She adapted well with an Apgar of 9/10/10 and was put in a prone position on her mother's chest 1.5 h after birth. The father was lying next to the mother, and they fell asleep for approximately 15 min. When the father awoke, he noticed that the baby was not breathing. Cardiopulmonary resuscitation was started immediately, but a normal heart rate could only be detected after 40 min of resuscitation. Without improvement after therapeutic hypothermia and after careful neurologic investigations, the girl was declared brain dead and died at the age of 4 days. Autopsy showed organs of normal size and macroscopic appearance; the histological investigation revealed thrombotic microangiopathy, which can be explained by a coagulatory disorder following cardiopulmonary reanimation. Metabolic examinations for organic acids in urine and amino acids in blood showed only abnormalities due to hypoxia. In summary, neither autopsy nor post-mortem examinations were able to identify a cause of death.
All three babies were born at term after uncomplicated pregnancies, had a normal birth weight and adapted well after birth. The mothers were all primiparous, aged 29–36 years, with a normal body mass index and not suffering from any chronic diseases. All babies were placed close to their mothers, one in prone position, two alongside their mothers, maintaining skin-to-skin contact. Shortly after birth, the mothers and babies were left unattended; two mothers were awake, but did not notice their babies' deterioration. A summary of patient characteristics is given in Table 1.
Table 1. Summary of patient characteristics
| ||Case 1||Case 2||Case 3|
|Gestational age||40 + 4||41 + 1||38 + 2|
|Weight||3050 g||3790 g||3480 g|
|Apgar at 5 min||10||10||10|
|Time of collapse (hours after birth)||3.5||2||1.5|
|Mode of delivery||Vacuum-assisted vaginal delivery||Vaginal delivery||Vaginal delivery|
|Mother's body mass index||26||23||22|
In the literature, we found 132 published cases of death from unknown cause during the first hours of life. These cases are reported in 15 articles by 15 different authors, four of which were written in French but included an English abstract. Only six authors limit their presentation to early death not including life-threatening events [1-6]. In Table 2, we give an overview of all reported cases of ESUDI.
Table 2. Published cases of early sudden unexpected death in infancy (ESUDI)
|Peters C. ||4|
|Rodriguez-Alarcòn J. ||15|
|Andres V. ||2|
|Becher JC ||6|
|JY Leow ||30|
|Weber MA ||23|
|Polberger S. ||7|
|Foran A. ||7|
|Obonai T. ||10|
|Espagne S. (French article) ||2|
|Gatti H. (French article) ||5|
|Branger B. (French article) ||7|
|Dehan M. (French article) ||2|
| ||132 ESUDI|
The risk factors specified by most authors were skin-to-skin contact, prone position and primiparity. Table 3 illustrates how many and which authors reported these risk factors. The Table also gives an overview of the time of occurrence of collapse. As most authors do not specify in how many of their cases the mentioned risk factor occurred, we were not able to report a total number of patients with a certain risk factor, but only the number of publications reporting a risk factor.
Table 3. Risk factors and time of collapse published
|Risk factors for ESUDI||Number of publications reporting risk factors (Total number n = 15), n (%)||References|
|Skin-to-skin contact||6 (40)|| [5, 7, 8, 13, 14, 24] |
|Prone position||5 (33)|| [5, 8, 13, 21, 24] |
|Primiparity||5 (33)|| [8, 9, 14, 17, 21] |
|Unobserved||3 (20)|| [7, 14, 17] |
|Time (weekend, summer, morning)||3 (20)|| [1, 6, 19] |
|Time at which collapse occurred||Number of publications reporting time at which collapse occurred (Total number n = 22), n (%)||References|
|Within 2 h||4 (27)|| [8, 14, 17, 23] |
|Within 12 h||3 (20)|| [3, 9, 13] |
|In the delivery room||2 (13)|| [5, 18] |
|On the maternity ward||2 (13)|| [6, 24] |
|Within 3 days||2 (13)|| [19, 22] |
|Within 1 week to 10 days||2 (13)|| [1, 4] |
A consistent definition for unexpected death early after birth is still needed. It is termed early neonatal sudden death (ENSD), early sudden infant death syndrome (ESIDS) or early sudden unexpected death in infancy (ESUDI), but other designations and abbreviations can also be found in the literature. The period of time after birth during which this event is called ESUDI or ESIDS also lacks clear definition. Our literature search shows that most cases of ESUDI occur during the first 2–4 h of life. As there are no clear definitions, the inclusion criteria used by the various studies vary considerably. Information about EALTE should, in our opinion, be collected separately, as there are strong signs that ESIDS and EALTE do not belong to the same entity of disease .
With only 132 published cases, no reliable approximation of the incidence of ESUDI can be made. Polberger et al. examined the incidence of ESUDI occurring between six and 100 h after birth in Sweden in 1985 . The authors concluded that the incidence was about 0.12 per 1000 newborns, making early neonatal sudden deaths responsible for 11% of the total neonatal mortality rate.
The aetiology of ESUDI remains uncertain. In most of the cases reported in the literature, death could not be explained by autopsy or additional post-mortem examinations. The most common hypothesis on ESUDI is that these events are secondary to an acute upper airway obstruction . The occurrence during early skin-to-skin contact might present a higher risk for suffocation. This matches with the risk factors known for prone position and inexperienced mothers left unattended by trained staff. The positions most commonly used early after birth may be asphyxiating with direct skin-to-skin contact and might lead to suffocation. Becher et al. reported on 30 newborns that experienced sudden death or apparent life-threatening events within the first hours of life, in 24 of whom apparent accidental suffocation during breast-feeding or skin-to-skin contact was found to be the cause . In our cases, upper airway obstruction seems to be the most likely explanation for death as well, as two babies were lying prone on their mother and one baby was lying alongside her mother but very close to her body. Another discussed mechanism that could bring on a collapse is an increased vagal tone, which has been reported in cases of sudden infant death syndrome and discussed in ESIDS by Toker-Maimon et al. . However, this theory might not be applicable to ESUDI, because it does not explain the by now known risk factors.
Most autopsies revealed no explanation for death. One exception is shown by a neuropathological examination of ten children who died from ESUDI and in whom a high incidence of leucomalacia was found in the white matter and brainstem, suggesting the prenatal occurrence of ischaemic brain insults and thus an underlying disease . In some cases, this could be the reason why a patient is especially vulnerable to exogenous stressors. Knowing that the ventilatory response to hypoxic stressors changes within the first hours and days of life and a newborn's answer to hypoxia can be apnoea, this fact can explain why an obstruction can quickly lead to death without other underlying risk factors . For SIDS, a triple-risk model has been implemented. This model states that a vulnerable patient experiences exogenous stressors in a vulnerable phase of development . Some hypotheses contend that this could also be applicable to ESUDI. SIDS occurs mostly between the second and fourth month of life, when babies develop rapidly. In ESUDI, the vulnerable phase seems to be the postnatal adaptation, as most cases occur within the first hours after birth. However, it is very difficult to compare SIDS and ESUDI, because SIDS occurs during sleep and we have no reliable data on sleep status during early deaths.
The identification of additional risk factors can help detect and protect children at special risk. Primiparity and skin-to-skin contact between mother and baby not under observation are the best known risk factors described by several authors [8, 13, 14]. Our three cases also exhibited these key risk factors. Nevertheless, early skin-to-skin contact and breastfeeding should not be reconsidered, because they are known to have extensively positive effects. Moore et al. reported on a higher incidence and longer duration of breastfeeding as well as a shorter crying time, a better cardio-respiratory stability and higher blood glucose levels in newborns being skin-to-skin with their mothers soon after birth. Moreover, babies with early skin-to-skin contact were possibly more likely to have a good relationship with their mothers. No negative outcomes in association with early skin-to-skin contact were pointed out in this Cochrane review . Mori et al. reported on a higher body temperature and an increased heart rate in newborns with early skin-to-skin contact. Moreover, they found a slight decrease in saturation levels, but further studies are necessary to confirm this finding . A detailed study of ESUDI in England showed no change in incidence from 1983 to 2007. This indicates that even changed and modernized practices in delivery room care like bonding and early skin-to-skin contact did not increase the rates of ESUDI . Mothers who particularly match the risk profile are those with no expert knowledge on breastfeeding and a possible lack of skills in identifying and evaluating changes in their baby's behaviour. It was thought that this could be aggravated following an impairment of maternal vigilance due to extensive postpartal exhaustion or the use of sedatives. However, it was recently established that many deaths occurred even while the mother was awake . In two of our cases, the mothers were awake, which means that staff cannot rely on an awake mother to be alert to her baby's breathing. These facts point out the importance of close observation by skilled staff, and we thus suggest that these cases should induce healthcare workers to be more alert during the first hours of life, especially in inexperienced mothers. Some centres have established guidelines for safe skin-to-skin contact within the first hours after birth . The higher incidence on weekends and during the night might be explained by the fact that fewer trained medical staff are available at those times. We could not find enough data to be specific about a higher incidence in summer, which might be due to overheating, a tendency towards seasonal accumulation or the infant's weight. Likewise, neither the infant's gender nor the delivery mode seems to influence the occurrence of ESUDI . Ottaviani et al. suggest exposure to cigarette smoke, not only as a risk factor for SIDS, but also for ESUDI . Poets et al. found no association with maternal smoking.
A limitation of our study is the fact that most of the knowledge on ESUDI comes from case reports. Moreover, ESUDI is so rare that larger studies for the purpose of evaluating risk factors and cause are very difficult to perform. Furthermore, methodological quality is difficult to maintain because of the differences in the prevailing definitions.
In conclusion, a consistent definition is essential if we are to learn more about this rare disease. Unattended early skin-to-skin contact and primiparity combined with a lack of expertise in breast-feeding are the key risk factors. As the peak incidence is in the first 4 h of life, strategies such as close monitoring in the delivery room and improved parental instruction can be life-saving.