An official website of the United States government
NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.
Bundock EA, Corey TS, Andrew TA, et al., editors. Unexplained Pediatric Deaths: Investigation, Certification, and Family Needs [Internet]. San Diego (CA): Academic Forensic Pathology International; 2019.
HENRY
“Henry was 5 weeks old and our only child at the time of his death. Most people assume that Henry was sick or unwell when I talk about his passing. But he was a strong, beautiful, perfectly healthy little boy… until he wasn’t. Assuming a child who passed away suddenly or unexpectedly must have been weak or frail seems to suck out any motivation towards trying to figure out what really happened to this poor child. We are grateful to his medical examiner who took the time to try to answer our many questions and assist us in allowing research to learn from our precious son for a brighter future for all children.”
– Henry’s Mom and Dad
OUTLINE.
“He who saves one life, it is as though he has saved the world.”
Dr. Lester Adelson, 1991.
Dr. Lester Adelson.
For as long as humanity has suffered the deaths of their children, parents have wanted to know the reasons why. Authorities have also wanted to know the reasons, for both public health and public safety reasons. Historically, often the mother was blamed. The usual act for which she was blamed was overlying her baby during sleep in their shared bed.
In the biblical account of the judgement of Solomon, in I Kings 3:19, a case of apparent overlying is described: as two women appear before the king, one woman says of the other, “During the night this woman’s son died because she lay on him” (1). In the first century BCE, Diodorus of Sicily, the Roman historian, said that Egyptian mothers convicted of overlying were required to hug their dead child’s body to themselves for three days (2). In the first century CE, Soranus of Ephesus, the physician who wrote the oldest surviving obstetric, gynecological, and pediatric textbook, told mothers never to sleep with their babies, lest they lie on them (Image 2.2) (3).
Soranus of Ephesus
Both infanticide and overlying were recognized reasons for infant and child death in the centuries that followed. In Europe in the Middle Ages, women could be tried by ecclesiastical courts for the deaths of their children by overlying and infanticide. Although punishments for infanticide were greater, both were punished (2, 4). A German law in the 13th century made it illegal for mothers to take children less than three years old into their beds (2).
During the Renaissance, the fear of overlying gave rise to one of the most ingenious inventions in the history of the prevention of child death, the arcuccio or “arcutio” (Figure 2.1). This device, intended to allow a nursemaid and a child to share a bed safely, separated the two co-sleepers by a wooden barrier, and yet made it possible for the caretaker to breastfeed and lie on her side. There was even a place for her to comfortably position her breasts (5). But infant and child mortality throughout the centuries preceding the modern era was so great that only sparse attention was paid to the problem of overlying in sudden infant and child death.
Arcuccio.
In medicine, one must pay attention not to plausible theorizing, but to experience and reason together… I agree that theorizing is to be approved, provided that it is based on facts, and systematically makes its deductions from what is observed… but conclusions drawn from unaided reasons can hardly be serviceable; only those drawn from observed fact.
Attributed to Hippocrates (6)
By the 19th Century, civil and criminal courts had taken over from church courts, and science was beginning to prevail over tradition. Some children who died suddenly and unexpectedly underwent autopsy. This led to some early mistakes. Physicians believed incorrectly that they could distinguish signs of suffocation at autopsy. At times they ascribed pathology to findings that were not yet known to be normal variants, such as the large thymus of a healthy infant. In 1830, the physician and scientist, J.H. Kopp, developed an autopsy-based theory of infant death called “thymic asthma” that blamed the enlarged thymus for what would later be called “Sudden Infant Death Syndrome”, or SIDS (2, 7).
Kopp’s 1830 theory of thymic asthma was debunked in 1858 by A. Friedleben’s autopsy-based study. Yet researchers continued to entertain many incompatible scientific hypotheses, often with meager evidence. They hoped to find, not only methods to treat children and prevent deaths, but also an explanation for these deaths that would exonerate parents from the accusation of causing their children’s deaths through overlying (7). One early hero of forensic science was Thomas Wakley, founder of The Lancet, who endured both personal and professional reverses when calling for the office of coroner to be held by a physician (8). In his journal in 1855, he condemned the assumption of overlying, because it greatly distressed parents (4). In 1884, despite Friedleben’s 1858 paper, an updated version of thymic asthma was used in court to exonerate a serving girl who had been arrested because a child in her care was found dead in its crib (2, 7).
The defense of the serving girl marked an early phase in a 14-decade battle between authorities who relied on unnatural explanations such as overlying for infant and child death, and those who offered natural explanations. On one side of the divide, great physicians and investigators accused parents of accidentally or intentionally killing their children; on the other side, equally great physicians and philanthropists indignantly protected parents from such accusations and pointed to science as the instrument that would someday identify the cause for all these deaths.
In 1892, Dr. Charles Templeman, the Scottish police surgeon (equivalent to a forensic pathologist in that era), wrote a landmark paper in which he reported 258 deaths of infants and children caused by accidental overlying (9). Eighty years later, the pediatric pathologist who gave the name to the sudden death syndrome, Dr. Bruce Beckwith, would describe Dr. Templeman as promoting “cruel miscarriages of justice” in this paper because he blamed the parents (10). Templeman’s paper described poor, often intoxicated parents, unintentionally overlying their infants during co-sleeping: “frequently father, mother, and two, three, or even as many as five children sleeping in one bed which sometimes consists of a few jute sacks spread out on the floor” (9, 10). Templeman also identified six factors common to many of his cases that are still recognized today: greater incidence in winter months, in boys, in infants younger than six months of age, in families with crowded beds, in the poverty-stricken (Image 2.3), and in families with substance abuse problems (at that time, alcohol). Although Templeman placed blame on the parents, he emphasized in his paper that none of the deaths were prosecuted by the police for whom he worked. When he advocated “somewhat… grandmotherly legislation… which prohibits a child under the age of two years from being allowed to occupy the same bed as its parents or nurse,” he offered his community the possibility of preventing future infant and child deaths. But his advocacy of accidental asphyxia in child death would be repudiated in the next century.
Children sleeping in poverty, New York City, 1880s. Jacob Riis, iconic image entitled “Street Arabs in the Area of Mulberry Street.”
In tandem with the debate over whether parents could accidentally overlie their children, some authorities argued over whether they could ever intentionally kill them. Infanticide has been recorded since ancient times. However, the first case of child abuse to create a national focus in the United States was a child who survived. Mary Ellen Wilson (Image 2.4), ten years old, was rescued in 1874 from her abusers by a Methodist missionary with the help of the American Society for Prevention of Cruelty to Animals, which had laws at their disposal that were better than the child protection laws (11). After recognition that children could be abused, physicians such as Dr. Abraham Jacobi and Dr. Job Lewis Smith, both later considered fathers of American pediatrics, worked with private and governmental agencies in the late 19th century to address the harms befalling children from violence, exploitation, and poverty. Physicians who had always cared for children with injuries and malnutrition began to speak out for vulnerable populations in the late 1800s. However, it would not be until 1904 that a physician-social worker team would start to address these issues at a New York City hospital. Formal child protection teams would not follow until after World War II (11). The medical establishment would need more than another century to acknowledge maltreatment as a medical issue (11).
Mary Ellen Wilson upon removal to care, 1874.
The greatest number of deaths have always been among infants and children who were not abused. Fifty years after Templeman’s 1892 paper, American physician Dr. Harold Abramson (Image 2.5) identified a seventh risk factor for sudden infant death, a factor which would become all-important nearly fifty years later: the prone sleeping position for infants. His 1944 paper resounded with warnings against the dangerous combination of unsafe bedding and prone sleep. He recognized co-sleeping as well as unsafe beds and soft bedding as specific hazards for prone infants. His paper was the first to advise parents on ways to keep their children safe, rather than blame parents for having a role in their children’s deaths (12).
Dr. Harold Abramson.
On the other side of the divide, physicians who believed these deaths were due to natural causes began to attribute most to respiratory illness (13). The rush to blame bacteria began with Dr. Sidney Farber’s much-quoted, 1934 paper entitled “Fulminating Streptococcus infections in infancy as a cause of sudden death” (14). Dr. John M. Adams followed in 1943 with a paper attributing sudden death in infants to pneumonia (14, 15). Dr. Jacob Werne and Dr. Irene Garrow continued this line of investigation with papers in 1947 and 1953; the 1947 paper was provocatively titled “Sudden death of infants allegedly due to mechanical suffocation” (16, 17). As penicillin and other antibiotics were invented to fight pneumonia, physicians on the bacterial side of the divide felt themselves empowered to prevent both parental blame and sudden infant death.
The lack of satisfactory objective evidence concerning the cause of these deaths in infancy and childhood accounts for the frequency with which such speculative explanations as suffocation by bedding, lymphatism, or status thymicolymphaticus are proposed. Although the importance of an infectious process in the causation of these deaths has been emphasized in recent writings, there is no agreement as to the types of infective agents involved or the lesions produced.
Drs. Adelson and Kinney (18)
In 1956, Dr. Lester Adelson and Dr. Eleanor Roberts Kinney published the results of an investigation into the sudden deaths of 126 children between ten days and two years of age and concluded that respiratory infection was responsible for many of them (18). They disproved any possibility of detecting suffocation after death by demonstrating that autopsy findings once considered diagnostic of suffocation were just as common in children who could not have died from asphyxia as those who had. Such findings included “cyanosis, persistent post-mortem fluidity of the blood, pulmonary congestion and petechiae, diffuse subpleural, subepicardial and intrapulmonary hemorrhages, bloodstained froth in the air passages, right-sided cardiac dilatation, and focal intracerebral and subarachnoid hemorrhages” (18).
But the most striking of their findings, in retrospect, is the table in which they detail the position of each child found dead (Table 2.1). Careful scene investigation found that 65 of the children were found prone, of whom 28 were specifically “face down.” Of the 61 who were not prone, 20 were found “on back,” while 7 were described as “head covered by bedclothes,” and 22 “bedmate present” (not further specified). Adelson and Kinney did not draw any conclusion from these findings. Abramson recognized that the suffocations of prone sleeping infants on soft bedclothes were completely unintentional; his concern for the dangers of co-sleeping, unsafe bedding, and prone sleep carried no blame for parents. Still, physicians on the opposite side of the divide felt too much conviction of parents’ innocence to accept the possibility of suffocation. The guilt that the parents would feel, if they recognized they played any role in their children’s deaths, was a major point made by these advocates. Within a year of Abramson’s article, pediatrician Dr. Paul Woolley Jr. would state: “We are far from convinced that good evidence exists to support the theory—that more infants suffocate— than die from burns, trauma, foreign bodies and automobile accidents combined. One of the most distressing features to be faced upon the death of a child is the invariable presence of self-incrimination and guilt on the part of the parents. This can be dispelled only through sincere assurance by the attending physician that all possible was done to avert the catastrophe and that the outcome could not have been altered through changes in their behavior. To leave the family with a clear conscience is a duty of importance secondary only to saving the patient” (19).
Details of Death Scene in 99 Cases Where Death Was Not Witnessed.
Along with the possibility that parents might have failed to prevent accidental suffocation, the idea of infanticide became unsustainable in the mid-20th century. When Dr. John Caffey (Image 2.6) published his landmark paper in 1946, “Multiple fractures in the long bones of infants suffering from chronic subdural hematoma,” he described it as a syndrome, similar to other syndromes with natural causes (20). The medical establishment would need more than twenty-five years after World War II to acknowledge maltreatment as a medical issue (11).
Dr. John Caffey.
Due in part to the battles fought by physicians such as Caffey and Woolley, Americans gradually began to accept the idea that parents might kill their children. Not all physicians accepted infanticide; in 1967, Beckwith would call it “the most malignant theory ever proposed.” But the evidence gradually became too great to ignore. It is interesting that within ten years of the paper denigrating Abramson’s conclusions, Woolley and radiologist Dr. William Evans would publish a paper on injuries and deaths in children that they ascribed to trauma, but (because of social and legal repercussions) stopped short of labeling as child abuse (21). The reluctant recognition that parents could intentionally harm their children arrived about twenty years before the admission that parents might unintentionally harm their children by placing them prone on unsafe sleep surfaces, or overlying them in sleep (10, 22).
In 1962, Dr. C. Henry Kempe (Image 2.7) published a landmark article on the “battered child syndrome.” This brought child abuse to the forefront as a medical diagnosis and initiated the national medical acceptance of child maltreatment as a medical diagnosis. It also served as a turning point in developing national measures for child protection (23). The pattern Kempe described included injuries, fractures, malnutrition, and poor hygiene. Later, physicians and scientists described the effects of abusive head trauma. Among the first was Caffey’s 1972 article, “On the theory and practice of shaking infants,” awakening Americans to the new concept of the “shaken baby” (24).
Dr. C. Henry Kempe.
Kempe and others exposed the “hidden problem” of child sexual abuse, and Dr. Ray E. Helfer later detailed the effects of neglect (25). These leaders, building on the prior contributions of Dr. Abraham Jacobi, Dr. Amboise Tardieu, Dr. Frederic Silverman, and many others, brought forward a medical understanding of abuse and neglect and inspired a community response (26). Although communities and physicians were finally beginning to accept that parents are sometimes to blame for infanticide, it would take more than two decades to understand that parents sometimes cause the accidental deaths of babies by placing them in unsafe sleep situations.
Part of the delay in accepting the role of unsafe sleep lay in the mistreatment of parents suspected of having played a role in their children’s deaths through carelessness or neglect. William J. Curran (1972) pled for the innocence of parents accused as if they were abusers in crib deaths: “My feeling is that it is a crime to subject living parents grieving for their child who has been struck with crib death, to such treatment” (27). Curran also pointed out real defects in child death investigation: “No state, not one, requires by law that all sudden deaths be autopsied. Certain classes of sudden unexplained death are reportable… It is the decision of these officials whether the case will be investigated at all, a view taken of the body, or an autopsy ordered and performed.” He added, “The parents are…left with the inconclusive results and their own feelings of guilt forever unresolved. In many communities, crib deaths are written off in unthinking cruelty as suffocation, pneumonia, and undifferentiated accidental death. It is a serious medicolegal problem that the ‘sudden-infant-death syndrome’ is not a reportable cause of death in this country and has no code in the International Classification of Diseases” (27). Later, Dr. Milton Helpern issued a passionate rebuttal to Curran’s criticism, begging people to understand that regardless of blame, natural disease does not always provide the only answer as to why infants and children die (28).
Pediatricians, neonatal intensive care specialists, nurses, and parents feared the consequences of placing their infant supine. As late as 1988, the editors of The Lancet would write, “It would seem sensible for healthy preterm babies, including those that have recovered from respiratory distress, to be nourished in the prone position for most of the time” (29).
A tiny conference in Seattle, at the University of Washington School of Medicine in September 1963, inaugurated the era of SIDS. That September 1963 conference drew only ten panel participants and thirty guests. Pediatric pathologists and pediatricians led it; forensic pathologists were in the minority, both on the panel and among the guests (30). The children’s hospital pathologist, Dr. Benjamin Landing, started the conference by addressing the lack of consensus for the name for sudden, unexplained and unexpected infant deaths. American clinicians were using “Sudden Death Syndrome”; the British were using “Cot-death”, and the Los Angeles county coroner used “crib death.” It was agreed in 1963 that everyone would use “sudden death syndrome.” In 1967, Dr. Marie Valdes-Dapena (Image 2.8), the pediatric pathologist who pioneered the concept of SIDS with Beckwith, looked back at it in an article called “Sudden and unexpected death in infancy: a review of the world literature in 1954–1966.” She explained that the fundamental requirement was a definition of terms: “To most investigators today, sudden and unexpected death in infancy is ‘the death of a child who was thought to be in good health, whose terminal illness appeared to be so mild that the possibility of a fatal outcome was not anticipated’” (31). In this review, she mentioned a consensus decision taken at that “Conference on Causes of Sudden Death in Infants,” in which it was agreed the term “crib death” was misleading, since these deaths do not all occur in cribs, beds, or bassinets.
Dr. Marie Valdes-Dapena.
Once that debate was settled, participants reviewed putative causes that included milk allergy, hypersensitivity to a variety of other allergens, bronchospasm, adrenal insufficiency, infection, and mechanical suffocation. The attendees agreed their meeting “did not and could not provide a definitive answer.” They did feel it was a “stimulating starting point for any new program aimed at attacking this unsolved problem of sudden unexplained death which, in the United States alone mysteriously strikes at least 15,000 apparently healthy infants every year” (30).
Reviewing the four major causes of death that investigators reported from autopsies of infant deaths, Valdes-Dapena found that respiratory infections and other infectious disorders such as sepsis were certainly present; congenital anomalies and aspiration each took their toll. But a truly important reported cause was “undetermined.” Quoting another scientist, she said, “It was Diamond who pleaded that one ought not ‘yield to the temptation of assigning a lesion as the cause of death merely because it is the only abnormality present’” (31). A pathologist had to weigh carefully what explanation should be given to the parents, and indeed, her book on the histopathology of SIDS became a reference for pathologists faced with the same dilemma for the next thirty years. Should parents be told that a minor infection, one that would not kill a normal healthy child, was what had taken their child away from them? How would they prevent the deaths of future children with this information? She quoted Diamond again: “He felt that one had to begin to arrive at a solution to the total problem with a frank statement of, ‘I don’t know’” (31). This painfully honest admission marked the recognition that SIDS, as it would come to be known, was an undetermined cause of death.
In recent years the standing of the United States among nations of the world in this category has steadily declined. According to the latest figures, we rank 15th. Our infant mortality rate in 1964 was 24.8 per 1,000 live births. Had Sweden’s rate – the world’s lowest – prevailed in the United States, 42,600 fewer infant deaths would have occurred… For a nation that prides itself on its dedication to the human individual, these national figures are hard to tolerate.
Dr. William H. Stewart (32)
While pediatricians, pediatric pathologists, and research scientists continued to search for the natural cause of sudden infant death, no theory put forward prevented a single death. In the six years following the first conference, new hypotheses for the cause of sudden death syndrome in infants included electrolyte imbalance, hyperactive “dive reflex,” and nasal obstruction (33). Although some true, if rare, causes of infant death were identified and reported, most of these theories would end up in the dust. None of them would hold any more significance to history than the theory of thymic asthma, that had been so thoroughly dismissed in 1858 by Friedleben.
Researchers investigating the incidence of infant death did recognize that sudden infant death was more common in America in minority and non-white populations, and as Templeman had noticed, among the socioeconomically disadvantaged, independent of race (33). The reason for this was not elucidated.
The second Seattle conference, held in 1969, included 27 panel members. By this time, some causes of infant death had been eliminated. No one placed further credence in milk allergy or status thymicolymphaticus (33, 34). At this conference, following Beckwith’s (Image 2.9) proposal, the term Sudden Infant Death Syndrome (SIDS) was adopted and defined as “The sudden death of any infant or young child, which is unexpected by history, and in which a thorough postmortem examination fails to demonstrate a cause of death.” The exact nature of a thorough postmortem investigation would remain contentious for decades. The conference established a complete autopsy must have at least gross examination of the thorax, abdomen, brain, and larynx, and histologic examination of the heart, lungs, brain, liver, and kidneys. Despite no consensus on what caused SIDS, the three editors of the 1969 proceedings concluded “it is now possible to say that sudden infant death syndrome (SIDS) is a real disease, not a vague mysterious killer” (34).
Dr. J. Bruce Beckwith.
Of equal importance, the 1969 conference proceedings established minimal acceptable investigation requirements to make a diagnosis of SIDS. Three important requirements were added to autopsy: adequate history; ancillary studies, such as microbiology as indicated; and counseling of the family. Counseling of the family would play a larger role in the developments of the coming decades, as the medical community learned to work with families with respect for their losses, whether there was blame or not.
One reason death was not seen as a consequence of prone sleep was that the usual explanation listed on the death certificate was respiratory infection. But even at the height of the medical community’s certainty about natural explanations for SIDS, there were some who spoke out for the possibility of suffocation. In 1970, the forensic pathologist Dr. Jerry T. Francisco (Image 2.10), in his article “Smothering in infancy: its relationship to the crib death syndrome,” reviewed nearly 1700 deaths in Shelby County, Tennessee, of children older than seven days and younger than one year (35). These children died after no significant previous illness recognized by parent or physician, with no poisons suspected or identified, and no clear or significant anatomic disease at autopsy. In 49 of nearly 1700 deaths, the circumstances and autopsy findings suggested overlying as the mechanism of death. Francisco realized that the autopsy findings in these cases were the same as those in SIDS, and proposed that overlying could have been overlooked, stating: “During the years from 1944 until the present, untold numbers of man hours and millions of dollars have been spent looking for a natural disease mechanism which would explain the ‘crib’ death. During this time, the medical profession and the public have been very much concerned with the question of smothering due to bedclothes. There were no instances in this group in which smothering due to ordinary bedclothes was considered. The problem of asphyxial death due to overlying has been ignored” (35).
Dr. Jerry T. Francisco
Meanwhile, a series of editions of the Dr. Benjamin Spock’s popular and influential baby book recommended placing infants prone to sleep (36). This recommendation was enthusiastically adopted by pediatricians and parents throughout America and across the western world. The desire to spare parents guilt during their time of grief was beautifully articulated by Beckwith in 1973; he explained that he considered suffocation to be a fallacy (10).
Sudden death in older children also began receiving attention, although little and late. In 1971, Dr. Charles Hirsch (Image 2.11) published “Unexpected death in young epileptics,” the first work to discuss sudden unexpected death in epilepsy (37). Because death in older infants and children was much less common than death from SIDS, it would be decades before significant attention would be focused on preventing the deaths of these children (38).
Dr. Charles Hirsch.
From the beginning, it appeared likely that older children were dying of diseases of the heart or the brain that simply could not be detected or predicted by the medicine of the time. Despite the research concentration on infants, a few publications concerning older children were emerging. Pittsburgh pediatrician, Dr. Daniel Neuspiel, and epidemiologist, Dr. Lewis Kuller, noted two age-specific peaks in sudden “nontraumatic” deaths of children aged 1 to 21 years. The first peak involved children aged 1 to 4 years of age, most often from infections and “undetermined” causes. The second peak, in children aged 14 to 21 years, showed cardiovascular abnormalities, epilepsy, intracranial hemorrhage, and asthma as the leading causes (39). The separate significance of sudden unexplained death in childhood would not be recognized for another decade and a half (38).
As pediatric pathologists came to the fore in performing autopsies and doing research on sudden unexplained infant deaths, coroners and medical examiners were relegated to a secondary role. Science focused for the next two decades on natural causes for unexplained infant death.
There is generally agreement among pathologists as to paucity and the uniformity of gross and microscopic observations in autopsies on infected infants.
Dr. Marie Valdes-Dapena (31)
Although investigators had realized as early as the 1960s that medical history was important to the investigation of SIDS deaths, history-taking focused on the fact that these deaths occurred unobserved during sleep (10). Other historical factors that researchers considered as possible triggers of SIDS included birth order, the season of the year (just as Templeman had observed, it was noted to be more common in winter months), prior infections, immunologic mechanisms, and even variations in the weather (40).
Disregarding Francisco’s findings, the mainstream of research continued to focus on natural causes of death. The best avenue seemed to be a search for differences in autopsy findings between SIDS babies and age-matched controls. Researchers focused on a possible physiologic mechanism. Perhaps babies who would eventually die of SIDS were not getting enough oxygen, not only at the time of death, but for some lengthy period before they succumbed. Findings interpreted as signs of chronic hypoxia in SIDS included extramedullary hematopoiesis in the liver, mean increase in the medial muscle mass of pulmonary arteries and arterioles, retention of periadrenal brown fat, and increased astroglial fibers in the respiratory centers of the brainstem (40–43). Although Valdes-Dapena did some of this work, the histologist Dr. Richard L. Naeye also reported findings in multiple publications between 1973 and 1977.
This research direction was strongly influenced by the 1972 publication of Dr. Alfred Steinschneider’s paper, “Prolonged apnea and the sudden infant death syndrome: clinical and laboratory observations,” which described five babies who demonstrated episodes of apnea on apnea monitors before dying of SIDS (44). Steinschneider hypothesized that prolonged apnea during sleep could lead to SIDS. Because two of the infants were siblings, he suggested that there might be a genetic or inherited component to SIDS. The natural conclusion was that apnea monitors might empower parents to prevent SIDS in future children. As no research up to this date had prevented death, this possibility demanded attention.
A 1978 paper examining the relationship between immature pulmonary function and apnea showed that the issue had attracted the attention of Peter J. Fleming (Image 2.12), a young researcher in the United Kingdom (45). In 1982, Fleming and colleagues changed the conversation with publication of “Prolonged apnea and cardiac arrhythmia in infants discharged from neonatal intensive care units: Failure to predict an increased risk for sudden infant death syndrome”, which established that apnea episodes commonly observed in neonatal intensive care unit infants showed no ability to predict SIDS (46). Six sudden, unexpected deaths occurred in infants who had shown no significant apnea, bradycardia, or cardiac arrhythmias while being monitored in the intensive care unit. Meanwhile, parents and insurance companies were also discovering that apnea monitors had no ability to prevent SIDS.
Dr. Peter J. Fleming
After Fleming’s 1982 paper, emphasis shifted away from hypoxemia and apnea as possible causes of SIDS. Another push away from the “hypoxemia/apnea” hypothesis came in the late 1980s to early 1990s when Steinschneider’s 1972 paper supporting the “prolonged apnea” theory was discredited (44). It was discovered that the siblings discussed in Steinschneider’s paper had been murdered by their mother, Waneta Hoyt (47–50). Steinschneider never believed that Hoyt intentionally suffocated her infants and continued to defend her up to her conviction.
The medical community and public were beginning to notice maltreatment patterns like those of Waneta Hoyt’s children. These patterns had previously been attributed to medical causes (51, 52). Autopsy alone cannot detect suffocation unless there are other findings to accompany it. Without scene investigation, history, and other non-autopsy information, autopsy often could not differentiate a natural infant death from an unintentional or accidental infant death, or worse, an intentional or neglectful infant death. This was particularly true for those cases of homicidal smothering which Beckwith believed were impossible.
Hundreds of scientific papers looking for other possible methods to explain and prevent SIDS were published from the late 1970s into the early 1980s. These works concerned hypotheses as varied as central apnea, allergic response to dust mites, magnesium deprivation syndrome, effects of diphtheria, pertussis, and tetanus immunizations, infant botulism, gastroesophageal reflux, and cardiac conduction abnormalities. At the 1969 conference, the consensus had been that SIDS was a single “real disease” (34). By the mid-1980’s, major researchers were recognizing this could not be true. Valdes-Dapena wrote in 1985, “It is likely that these deaths represent a heterogeneous group of pathogenetic phenomena rather than a single entity” (53–55).
Most investigators in the 1960s through 1980s, unlike Adelson and Kinney in 1956, failed to discuss the place where infant and child death occurred. In 1977, forensic pathologist Dr. Millard Bass described 15 infants who died in their cribs in Wade County, Michigan, as a result of accidental asphyxia (smothering or strangulation) (56). “Ten cases of asphyxia were due to suffocation or strangulation as a result of entrapment of the head and neck between crib slats or crib rail and mattress. Five additional deaths resulted from strangulation when the pacifier cord, necklace or nightgown became entangled on the crib post or on a toy attached to the crib rail. It appeared to the emergency room physician that these deaths were due to the sudden-infant-death syndrome. It was only after further investigation and reconstruction of the scene that the information concerning accidental asphyxia was brought forth. One mother indicated that she had repositioned her child in the crib and withheld the information out of fear of being charged with criminal negligence. Three other cases of crib asphyxia were considered to be cases of sudden infant death syndrome by physicians even though information indicating accidental asphyxia was available.” In this prescient article, Bass emphasized that physicians and pathologists could be missing asphyxia if they did not know how the death scene looked.
Looking back at the decades when natural explanations dominated SIDS research, the “elephant in the room” was the overlooked importance of scene investigation. Possibly the most important paper in this decade was published by Dr. Bass in 1986. “Death scene investigations in sudden infant death” described 26 infant deaths initially diagnosed as SIDS, which also had scene investigation performed by a forensic pathologist (57). In six of the 26 cases the authors found “strong circumstantial evidence of accidental death.” Even more remarkable, in another 18 of those 26 cases, they discovered “various possible other causes of death other than SIDS” (57, 58). This initiated a paradigm shift in sudden infant death and young childhood death scene investigation, carried out by forensic pathologists and medicolegal investigators. Pediatric pathologists became less involved in the investigation of sudden, unexpected infant death as forensic pathologists moved to the lead.
Meanwhile, the prone sleeping movement to reduce SIDS was receiving notice all over the world. Dr. Susan Beal, a pediatrician in Adelaide, Australia, who had five children of her own, began visiting families who had lost babies to “cot-death.” Between 1973 and 1990 she visited more than 500 families. She found the rate of death was highest among babies who slept prone. Her publications in 1984 and succeeding years played an instrumental role in reversing prone sleep position in Australia, and, later, across the world (59).
In 2007, on the occasion of awarding her the Distinguished Researcher award, the International Society for the Study and Prevention of Perinatal and Infant Death wrote, “She is credited with being the first person anywhere to argue publicly against babies sleeping on their stomachs and in the countries that have heeded her advice, the incidence of SIDS has decreased dramatically” (60).
Australia was not the only country where the influence of prone position was being noticed. Norwegian researcher Dr. GussA DeJonge gave a lecture on sleeping position and the rise in sudden unexpected death in infants in the Netherlands. “This lecture attracted considerable publicity and as a result… a non-prone position for infants” was advocated in Norway (61, 62).
The attention of the indefatigable Peter Fleming was attracted by the Netherlands research. In 1990, he and his colleagues published “Interactions between bedding and sleeping position in the sudden infant death syndrome: a population-based case control study” (63). Of the 72 deaths in the Avon, UK area during the duration of his study, only five had a full explanation after autopsy (severe gastroenteritis and dehydration, septicemia, hemorrhagic disease, perforated stomach, and Down syndrome). He did show that infants who died were more heavily wrapped and therefore warmer than infants who did not die. But more of the heavily wrapped survivors had slept supine or on their sides. Sixty-two of the infants who died had been put to sleep prone. He and his colleagues began to advocate a supine sleeping position for infants and deaths in Avon dropped (64).
In 1991, British journalist and actress, Anne Diamond (Image 2.13), lost her infant child without warning on the day of her four-year-old son’s birthday party. It was devastating for her to find out after the death that there had already been a breakthrough, and that if her child had been living in Avon, he might not have died. Her efforts to publicize Fleming’s findings led to the “Back to Sleep” campaign in 1991 and she became the celebrity whose face was emblematic of the revolution. In her reminisces on the Back to Sleep campaign, she said, “It ran on TV throughout the winter of 1991 and started saving lives. Cot death numbers plummeted from 2500 per year to about 300, where they stubbornly remain, even now” (65).
Anne Diamond.
Many pediatric pathologists joined forensic pathologists investigating this aspect of SIDS, which held the new promise of prevention. In 1991, Dr. Enid Gilbert-Barness (Image 2.14) and colleagues published “Hazards of mattresses, beds and bedding in death of infants,” reporting 52 infants and young children who had died suddenly and unexpectedly (66). “In twenty of these cases, death resulted from preventable accidents rather than from sudden infant death syndrome (SIDS). The preventable infant deaths in this report fall into four categories: suffocation on waterbeds, suffocation on sheepskin rugs, accidents associated with beds and overlying” (66).
Dr. Enid Gilbert-Barness.
Her study included a single 9-year-old “severely mentally retarded child with cerebral palsy and spastic paraplegia,” who had been placed face down on a free-floating water bed and was found dead about two hours later. Autopsy disclosed “severe pulmonary edema with focal hemorrhage of up to 10 cm in diameter in both lungs” (66). It did not escape investigators that these findings were similar to those described at autopsy in SIDS.
There was growing interest in the 1980s and 1990s in sudden unexpected death in childhood as well as in infants. When the 1989 definition of SIDS was revisited in 1994 and 1995, at the Second and Third SIDS Global Strategy Meetings in Norway and Australia, respectively, it was recognized that specifying infants in the definition ignored the small but significant number of children over the age of one year who die suddenly and unexpectedly and might exclude them from important research (38, 67).
In 1985, a paper in the Journal of the American Medical Association, by Dr. Daniel R. Neuspiel and colleagues, drew attention to “sudden nontraumatic death from persons age one to twenty-one years” (39) by reporting 207 deaths in nine years (4.6 per 100 000 population per year). Dr. David J. Driscoll et al., presenting research in 1985 on “Sudden unexpected death in children and adolescents,” found a rate of 1.3 per 100 000 patient years (68). Compared with a rate of SIDS death in 1990 of 1.6 per 1000 births, or close to 160 per 100 000 population, the problem of sudden unexplained death in childhood affected only one-one-hundredth as many children as SIDS. But the problem, like SIDS, is life-altering for families who experience it, and this research raised the possibility of discovering an underlying natural cause in children whose deaths were not fully explained by scene investigation. The death of the 9-year old child described by Gilbert-Barness, who was face down in a water bed, suggested that prone position may contribute to sudden unexplained death from seizures and cardiac arrhythmias in older children. It was suggested that the episode leading to death might not have been fatal if there were no asphyxial component.
As the Back to Sleep campaign continued to decrease the rates of infant and childhood death (Figure 2.2), the advent of scene reenactment with dolls in the mid 1980s revealed many cases of actual or suspected asphyxia. The practice of doll reenactment gradually spread throughout the forensic world.
Graphic depiction of decrease in infant deaths after Back to Sleep.
The work of Adelson and Kinney, Francisco, and Bass, as well as the legacy of thousands of autopsies, had clarified that autopsy findings alone can neither show a child has died of suffocation, nor prove that it has not.
Meanwhile, Fleming and colleagues continued to identify associations between sudden unexpected infant deaths and prone position, overheating, preceding illness, and parental smoking. In addition, they demonstrated a reduction in SIDS deaths when an infant was put to sleep using a pacifier. This research was instrumental in changing the ways babies slept in Britain and in decreasing SIDS rates. Although largely unsung, Fleming and colleagues’ work contributed to the campaign that saved thousands of lives (63, 70–72).
In 1989, 20 years after the 1969 conference, the US National Institute of Child Health and Human Development (NICHD) convened an expert panel to review new scientific knowledge. New minimum requirements for postmortem investigation of a suspected SIDS case included examination of the death scene as well as the autopsy, history and ancillary studies (73). The NICHD panel recognized that a new definition of SIDS and its diagnostic criteria was required. SIDS was now only to be used for infants under one year old; Dr. Henry F. Krous would propose the addition of “and only during sleep,” which was made part of the definition.
In 1992, a letter to The Lancet on the definition of SIDS, listing authors including Valdes-Dapena, Fleming, Dr. Eric A. Mitchell, Dr. D.M. Becroft, and Dr. P.J. Berry (among others), was emblematic of the new direction in research shared by both forensic and pediatric pathologists. Two names on the author list would go on to be prolific publishers in the field: Dr. Roger W. Byard (Image 2.15), forensic pathologist, in the field of SIDS; and Henry F. Krous (Image 2.16), director of pathology at a children’s hospital, in the field of sudden unexplained death in childhood (74). Byard said in 1995, “It is also likely that the etiology of SIDS is heterogeneous and that the term SIDS is not so much a diagnosis but a term covering a variety of mechanisms which result in a common lethal outcome” (75).
Dr. Roger W. Byard.
Dr. Henry F. Krous.
As the incidence of recorded SIDS deaths decreased sharply in the early 1990s after implementation of the Back to Sleep campaign, the New Zealand cot death study, a large nationwide case control study beginning in 1997, confirmed that there were three major modifiable risk factors for SIDS. These were the prone sleeping position, maternal smoking, and lack of breastfeeding. Risk calculations indicated that about 79% of the SIDS deaths were attributable to these factors (76). At last, parents had something they could do to prevent SIDS.
It is not known how the prone position increases the risk of SIDS.
Dr. Anne-Louise Ponsonby et al. (77)
In 1994, Dr. James J. Filiano and Dr. Hannah C. Kinney introduced the Triple Risk Model (Figure 2.3) (78), specifying that three overlapping factors might put a child at risk for SIDS. These included an underlying vulnerability (an internal abnormality or genetic predisposition); a critical developmental period (within the first six months of life); and an exogenous stressor(s), such as excess bedding (79).
The Triple-Risk Model.
While the exogenous stressors were increasingly recognized in sudden unexplained infant death (SUID) investigations and doll reenactments, the underlying genetic factors that might make an infant vulnerable remained obscure. The 1991 article in the New England Journal of Medicine by Dr. Peter J. Schwartz et al., “Prolongation of the QT interval and the sudden infant death syndrome,” suggested a reason why a subset of vulnerable infants might be more endangered by soft bedding than other infants (80). This prospective investigation found that prolongation of the QT internal in the first week of life was strongly associated with sudden unexplained death in the first six months of life. Although this explained only a small fraction of SIDS, as the investigators acknowledged, it laid the foundation for a possible critical nexus between endogenous and exogenous factors. Those exogenous factors included asphyxial mechanisms.
As understanding of the potential asphyxial mechanisms improved, the US Centers for Disease Control and Prevention emphasized sudden unexpected infant death investigation. In 1993, a multidisciplinary workshop sponsored by the US Department of Health and Human Services was convened to develop a protocol for how pathologists and investigators should collect investigative and scene information on sudden unexplained infant deaths. This resulted in the 1996 publication of “Guidelines for death scene investigation of sudden unexpected infant death” (81). This critical publication included the SUIDIRF (Sudden Unexplained Infant Death Investigation Report Form), now in widespread use across the US. The form prompts investigators to check for key information that was not being uniformly collected before its advent, such as circumstances of death, prenatal history and infant’s medical history, and sleep environment-related factors (i.e., sleeping location, type of bedding, and position of the infant, including whether breathing could have been obstructed).
Infants between six weeks and four months of age are often described as obligate nose breathers. This suggests that obstruction of the nasal passages in sleeping infants may contribute to asphyxial deaths (82). This observation offers one possible explanation as to why this age group may be so vulnerable to sudden death. A team of consumer product safety commission investigators found that about 30% of infants dying of SIDS between 1992 and 1993 were found with their noses and mouth covered by soft bedding (83). Most of them had been placed prone to sleep and many were lying atop pillows and comforters. This prompted a press release from the Consumer Product Safety Commission chair person, Ann Brown, with the headline, “Don’t put your baby to sleep on top of soft bedding” (83). The wheel had come full turn since Woolley attempted to discredit Abramson for warning parents against soft bedding in 1945.
The work of Kinney et al. focused on possible brain abnormalities that might parallel long QT syndrome in creating an underlying vulnerability in SIDS; however, it did not develop predictive or preventive paradigms. Still, endogenous cardiac and central nervous system-related factors might be the natural disorders that pediatric pathologists sought without success throughout the decades of the 1960s and 1970s. These critical factors could also explain some of the deaths of children between 1 and 19 years old (68). This offered the tantalizing possibility of a common explanation for some sudden infant and child deaths.
Among other developments in the late 1990s, videos of parents in hospital rooms proved that infanticide by suffocation did occur (84, 85). The syndrome, Munchausen Syndrome by Proxy (later renamed “factitious disorder imposed on another”), was found to be a highly dangerous psychological disorder. It might result in a death that could be attributed to SIDS or SUID if video evidence were not available (86). In 2001, the American Academy of Pediatrics published a paper on differentiating SIDS from child abuse fatalities (87). Estimates of infanticide among cases designated as SIDS or sudden unexplained death in infancy now range from 1% to 10% (88).
In 2003, Beckwith affirmed that the lack of an adequate definition for SIDS was creating significant problems, and experts agreed that SIDS was a troublesome diagnosis of exclusion. Some critics even referred to it as a “wastebasket diagnosis” (89). In 2004, a panel of experts further expanded and refined the definition of SIDS, dividing it into multiple categories ranging from classical “SIDS (meeting all the scenario criteria from the 1970s)” to “unclassified sudden infant death with criteria not met” (90). Death during sleep was an important part of this diagnostic algorithm (90).
Among the risk factors garnering more attention was bed-sharing or co-sleeping. In 1991, the launching of the New Zealand Cot Death Prevention Programme had initially promoted “back or side” sleeping position, subsequently changed to “back only” (76). A fourth major modifiable risk factor, bed-sharing with another person, was added to the prevention program in 1992 (76, 91, 92). Dr. James S. Kemp et al.’s study of the deaths of 119 infants in St. Louis, Missouri, found that unsafe sleeping practices occurred in a large number of cases, and, in a large proportion, bed-sharing or co-sleeping on a couch or a non-bed surface was involved (92). In 2005, Dr. Laura D. Knight’s large retrospective meta-analysis with case controls, focused on the role of bed-sharing in the sudden unexplained infant death of 697 consecutive deaths in Kentucky. The study demonstrated that infants sharing beds with adults had a significant increased risk ratio for sudden death (91). Although Templeman had been ignored or defamed for over a century, he was right after all. Bed-sharing and overlying contribute to a significant proportion of sudden unexplained infant and childhood death.
The year 2005 also saw the publication of Krous’ important paper on sudden unexplained death in childhood, (SUDC) defining “the mystery of SUDC” (38). These 50 cases of predominantly male toddlers, mostly between 1 and 3 years of age who died unexpectedly, often with a personal and family history of seizures associated with a fever and sometimes head trauma, drew scientific attention to the possibility of brain and heart abnormalities that had not been detected before death, and might not be detectable at autopsy. If heart or brain abnormalities explained some deaths in childhood, they might also explain some deaths in infancy, especially when combined with exogenous stressors identified by a full scene investigation.
In 2007, a National Association of Medical Examiners (NAME) position paper provided a functional approach to SUID. This position paper outlined recommendations for investigation including autopsy and ancillary testing and scene investigation and specified how to certify these deaths (93). This white paper also established “a list of potential stressors or possible external causes of death that should be identified or recorded on the death certificate.”
This long list included bed-sharing, unsafe or soft sleep surfaces, excessive blanketing or swaddling, previous unexplained sibling death, prone position, parental or caretaker intoxication, prenatal exposure to tobacco smoke, injuries of unknown significance, and any abrupt change in sleep position, location, or surface. Some of these exogenous factors could be more easily established by use of dolls during the scene reenactment (Images 2.17 and 2.18), which served as a proxy to visually recreate an infant’s sleep position with special focus on the presence or absence of airway obstruction (94).
Doll reenactment of placed position.
Doll reenactment of found position.
Although the shift away from using SIDS as cause of death on certificates means current national data is difficult to compared to data from the 1960s, 1970s and 1980s, it is clear that the rate of unexplained infant death has dropped significantly; but it has not dropped to zero (95–98).
Despite many publications dictating the components of a complete autopsy, a universally agreed approach on the modern autopsy has yet to be established (99–103). The differential diagnosis must be broad, and it is often difficult for medicolegal professionals to devote resources to cover the costs of investigating all the areas of interest, including the expense of newer molecular and imaging tests and the costs of indicated consultations with neuropathologists, cardiac pathologists, child abuse pediatricians, genetic counselors, and other experts as needed.
As science and technology march on, ancillary testing has expanded far beyond the microbiology cultures and plain radiographs that were indicated in 1963 (104). Although ancillary testing still is chosen on a case-by-case basis, the technology available has advanced. Postmortem computed tomography scanning of the brain and, for research purposes, magnetic resonance imaging, are now possible. Microbiological testing may now involve immunohistochemical and polymerase chain reaction detection of viral, bacterial, and parasitic microorganisms (105). These tests may be performed on tissue as well as blood and cerebrospinal fluid. If tissue testing advances to the point that paraffin-embedded tissue can be utilized, a century of cases may become available for analysis.
Scene investigation remains at the heart of sudden unexplained infant death investigation despite all these advances. Scene investigation in combination with autopsy also identifies child abuse, which may contribute to the unexpected death of infants and children in many ways. The number of child abuse deaths is believed to be under-estimated, and communities must do more for identification and prevention; this can be addressed through collaborative scene investigation and case review (106, 107).
New technologies also exist for infant monitoring, far past the apnea monitors of Steinschneider’s day. Although most of these monitors have not yet been studied in randomized trials, the possibility exists that someday a monitor will detect and report the moment when an infant or young child progresses from a normal homeostatic milieu towards sudden death. In that day, these deaths may not only be explained, but prevented at last.
Until that day, physicians rely on modalities that are currently available and the new modalities that are becoming increasingly available. As genetic testing grows more accessible and sophisticated, the molecular autopsy has been advocated as an adjunct to the gross and histologic autopsy of infants and children. At this early stage, more variants of uncertain significance are detected than genetic causes which would offer a full explanation for sudden unexplained deaths in infancy and childhood. Still, some studies have shown that this approach increases the diagnostic yield, and it is especially useful in identifying functional defects in the anatomically normal brain and heart (108). Many publications in the last decade have focused on a potential neuropathologic basis for sudden death of infants and children, including serotoninergic brainstem abnormalities and developmental abnormalities of the hippocampus in toddlers (76, 109–114). A personal family history of febrile seizure in SUDC victims deserves more attention. Although febrile seizures are common and usually have no sequelae, a fever may unmask an innate predisposition to seizures that may cause death in sleep. It is known that apnea episodes may be the only symptom of seizures in infants.
Channelopathies, neurotransmitter dysfunction, and other genetic causes of sudden death are another promising area of research. Channelopathies and serotoninergic receptor abnormalities are difficult to demonstrate in practical autopsy procedures and diagnostics today. But if, as this body of literature seems to indicate, the brain and heart play an integral role in these deaths, then serotoninopathies or other as yet undetected abnormalities, may link sudden unexplained infant, childhood, and epilepsy deaths (115, 116).
The tradition of scientific investigation and the hope for prevention of sudden unexplained death in infancy and childhood remain strong as we conclude the second decade of the 21st century. We acknowledge that interventions that have seen a decrease in sudden unexplained infant and childhood death are those which are simple to convey and prevent accidental asphyxia from overlying, smothering, or suffocation, such as the “A-B-C” campaign for safe infant sleep (Alone, on the Back, in a Crib).
This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. For attribution, the original author(s), title, publication source (PeerJ) and either DOI or URL of the article must be cited.
Your browsing activity is empty.
Activity recording is turned off.
See more...
