In 1999, PPB staff initiated a planning process for the Institute's third, five-year strategic plan on SIDS research. A working group composed of distinguished scientists and health care professionals from around the country, in collaboration with staff from the PPB and other NICHD Branches, identified research objectives and strategies designed to achieve these objectives. The group drew on previous and ongoing planning efforts, conferences, workshops, and research findings to develop Targeting Sudden Infant Death Syndrome (SIDS): A Strategic Plan, a public health agenda that would guide the Institute's research over the next five years. Like all NICHD strategic plans, the draft plan was placed on the NICHD Web site for public comment. The final plan was published in the summer of 2001, and is available on the NICHD Web site at: http://www.nichd.nih.gov/publications/pubs/targeting_SIDS.pdf.
The plan is divided into four topics: Etiology and Pathogenesis, Prognostics and Diagnostics, Health Disparities, and Prevention/Intervention. Each topic contains a statement of the problem, background information, and specific recommendations designed to address gaps in the current knowledge and intervention activities and to correct deficiencies in basic scientific infrastructure. The following sections outline some of the Branch's accomplishments within three of these topic areas since 1999.
Understanding the pathogenic mechanisms of SIDS requires a multidisciplinary approach and involves the analysis of postmortem tissue and death recordings, as well as modeling potential environmental and physiological pathways in animals and infants. Over the past four years, research has significantly expanded knowledge about autonomic nervous system abnormalities in infants who succumb to SIDS.
In earlier work, researchers had observed decreased binding of acetylcholine and kainate to receptors in the arcuate nucleus of SIDS infants, suggesting that the number of these neurotransmitter receptors was decreased. The arcuate nucleus region, on the ventral brainstem, is believed to control chemoreception and cardiorespiratory and cardiovascular responses and is linked to other brain regions that control arousal from sleep. It has been hypothesized that SIDS infants lack the ability to respond to life-threatening hypercapnic, hypoxic, hyperthermic, and/or cardiovascular episodes that occur during sleep. Researchers have now observed deficiencies in the serotinergic receptors in the arcuate nucleus, as well as in the n. raphe obscurus, inferior olive, n. paragigantocellularis, and the n. gigantocellularis regions of the brain (J Neuropathol Expt Neurol 2000; 59:372-384). The deficits were not observed in brainstems of infants who died from other causes. These results suggest that the scope of the neurochemical abnormalities in SIDS infants is greater than previously expected and that these abnormalities involve several functionally related nuclei derived from the rhombic lip. These regions develop from the rhombic lip in early gestation. Further studies have shown that they are connected by a communicating neuronal pathway that is present by mid-gestation. A special tract-tracing dye was injected into one of the fetal brainstem regions in the rostral ventral medulla (RVM) and later appeared in all six areas that were observed to have deficits in serotonin receptors in the SIDS infants (Autonomic Neurosci 2001; 89:110-124). This information supports the theory that the development of a lifesustaining brainstem pathway may be altered during gestation in infants who later succumb to SIDS.
Another PPB-supported program project has developed a piglet model to study the role of ventral brainstem regions in cardiorespiratory and cardiovascular responses to potentially life-threatening environmental stimuli that occur during sleep. The investigative team has been focusing on arcuate nucleus homologues in the RVM. Using muscimol to inhibit the RVM and stimulate gamma-aminobutyric acid receptors, researchers found that the breathing response to elevated carbon dioxide was inhibited while the animals were awake and asleep; the effect was greatest during sleep (JAP 2001; 90:971-980). These data provide evidence that the RVM contains neurons critical to the piglet's response to elevated carbon dioxide.
In addition, the investigators obtained the first evidence that the RVM contains neurons that are involved in the regulation of sleep architecture (Sleep 2001; 24:514-527). When the RVM was exposed to muscimol, sleep cycling was abolished in some experimental animals; in others, there was a decrease in low-frequency electroencephalogram (EEG) activity and delta power, reflecting a decrease in the depth of quiet sleep. Resting respiration, blood pressure, heart rate or ventilation, and their modulation by state were unaffected. These investigators have also developed a sensitive, automated method to score sleep arousals (Sleep 2001; 24:499-513) that is being applied to analyses of infant polysomnograms collected in the Collaborative Home Infant Monitoring Evaluation (CHIME) Study (see page 28 for more details on this study).
Sleeping on the stomach increases the risk of SIDS, but the exact reason for this increase is unknown. One theory on the hazard of stomach sleep position relates to an inability to respond to elevated carbon dioxide and low oxygen levels during sleep. Laboratory studies show that infants who sleep on their stomachs bury their faces or heads in bedding, which causes them to re-breathe inspired air low in oxygen (JAP 2001; 91:2537-2545). Under these conditions, infants experience an increased frequency of oxygen desaturation in the blood. Some infant responses did not successfully get them fresh air; these infants experienced greater desaturations that required intervention (Pediatrics 2002; 111:e328-e332).
Studies of animals that were deprived of oxygen have shown that the low blood oxygen concentration results in a rapid decrease of oxidative metabolism in tissues, which leads to bradycardia, apnea, and hypoxic coma. To alleviate this decrease, the animals initiate a gasping (large breaths) response; if the gasp provides enough oxygen to the heart and lungs, the cardiovascular function rapidly improves, resulting in “autoresuscitation” and rapid recovery. Analyses of home monitor recordings from infants who died of SIDS and from other causes showed that hypoxic gasping takes place immediately before death; but, the infants who succumbed to SIDS were distinct in that they had more double and triple gasps. Most of the non-SIDS cases showed evidence of return of cardiovascular function and partial or complete autoresuscitation prior to death. Among the SIDS cases, only one infant showed a transient increase in heart rate following a gasp, and none of the infants had evidence of complete resuscitation. These results suggest that infants who die of SIDS may have a deficit in the circulatory or other components of the autoresuscitation mechanism (Pediatr Pulmon 2003; 36:113-122).
Another theory about the potential risk of the stomach sleep position relates to sleep state. Infants who sleep on their stomachs spend more time in quiet sleep (deep sleep characterized by fewer awakenings and an increased arousal threshold) and less time in active sleep compared with those who sleep on their backs. PPB-supported researchers at Columbia University have found that, after a feeding (the post-prandial period), infants enter a period of quiet sleep. In infants who sleep on their stomachs, the duration of this quiet sleep period increases with the level of carbohydrates in the diet (Pediatr Res 2002; 52:399-404); the growing infant expends a lot of energy to absorb these nutrients during the post-prandial period. The energy expenditure that results from absorption and from being in a stomach sleep position may put too much demand on thermoregulation and cardiovascular regulation, resulting in SIDS.
Although the risk for SIDS does not appear to have a large genetic component, some deaths diagnosed as SIDS may have defined causes that are genetic in origin. For instance, genetic defects in fatty acid oxidation may account for 1 percent to 2 percent of SIDS cases. A recent finding also observed that 2 percent of a prospective population of SIDS cases had an identifiable genetic defect in the cardiac sodium-channel gene SCN5A that disturbed channel function. Follow-up studies supported by the PPB are investigating all sodium- and potassium-channel genes implicated in long Q-T syndrome, a set of abnormalities that predispose a person for heart arrhythmia, and their relationship to SIDS. Analysis of one mutation in SCN5A, detected in an infant with long Q-T syndrome who died suddenly, showed that the loss of sodium-channel function was due to a trafficking defect, which prevented the channel protein from reaching the plasma membrane within the cell. The investigators also identified polymorphisms within the gene that restore the loss of function due to the mutation (Physiol Genomics 2003; 12:187-193).
Prevention/Intervention
In 1994, the NICHD formed a partnership with the AAP, the SIDS Alliance, the Association of SIDS and Infant Mortality Programs, the Maternal and Child Health Bureau at the Health Resources and Services Administration, and other organizations to launch a public health education campaign that would educate caregivers about reducing the risks of SIDS—the Back to Sleep campaign. The campaign's main message at the time was that healthy babies should be placed on their backs or sides to sleep to help reduce the risk of SIDS. In 1996, the AAP revised its sleep position statement to recommend the back sleep position as preferred over the side position. Epidemiological studies have shown that side sleeping confers about twice the risk for SIDS relative to back sleeping, in part because babies are likely to roll from their sides to their stomachs. The Back to Sleep campaign materials were revised to reflect this change.
In order to evaluate changes in infant care practices in response to the AAP recommendation and the Back to Sleep campaign, the NICHD has supported two surveys. The first, the National Infant Sleep Position (NISP) Study, initiated in 1992, is annual telephone survey of nighttime caregivers in households with infants younger than eight months of age. Since the study began, NISP has documented a decline in the number of infants placed to sleep on their stomachs that correlates with the decline in SIDS rates. Furthermore, analyses of the nighttime caregiver surveys between 1994 and 1998 (JAMA 2000; 283:2135-2142) showed that physician recommendation was the single strongest influence on caregiver choice of sleep position, independent of sociodemographic characteristics of the mother. However, the analyses also showed the strongest probability of back sleep position when the caregivers reported exposure to the recommendation from multiple sources, including the hospital nurse, the baby’s physician, magazines and newspapers, and radio and television. By the spring of 2000, 66 percent of nighttime caregivers surveyed placed babies on their backs to sleep, while 14 percent placed them on their stomachs to sleep. Correspondingly, the SIDS rate in the United States dropped from 1.2 deaths per 1,000 live births in 1992, to 0.6 deaths per 1,000 live births in 2000.
The NISP study is also providing important information about other infant care practices. For instance, if an infant sleeps on an adult bed alone or with anther person, the practice is hazardous because it can lead to entrapment, overlay, and suffocation. Bed sharing with other children is known to increase SIDS risk, but controversy remains regarding the risk associated with sharing a bed with a parent. On average, almost half of infants in the survey population spent at least some time at night on an adult bed within two weeks of being surveyed. Between 1993 and 2000, the proportion of infants usually sharing an adult bed at night increased from 5.5 percent to 12.8 percent. Factors that independently increased the probability of usual bed sharing included young maternal age, maternal race reported as African American or as Asian/other, household income of less than $20,000 annually, and infant age younger than eight weeks. NISP data also showed that bed-sharing infants were almost twice as likely to be covered by a quilt or comforter than infants who did not share an adult bed; research shows that a quilt or comforter in the bed-sharing environment is a potential hazard for SIDS if the baby's face or head gets covered (Arch Pediatr Adolesc Med 2003; 157:43-49). More research is needed to understand the range of bed-sharing practices, motivations, and potential benefits or hazards.
When the AAP recommendation about back and side sleeping was first issued in 1992, one of the concerns was that infants placed on their backs or sides to sleep may be at risk for adverse health effects other than SIDS, particularly those due to aspiration or choking. To investigate these risks, the NICHD initiated the second of its evaluation studies, the Infant Care Practices Study (ICPS), a longitudinal, prospective study of more than 15,000 mother-infant dyads enrolled at birth between 1995 and 1998 in Massachusetts and Ohio. The ICPS has provided valuable information to allay concerns about choking. When infants were one month, three months, and six months of age, the researchers questioned their mothers about sleep position and whether they had symptoms such as fever, cough, wheezing, stuffy nose, trouble breathing, trouble sleeping, and/or vomiting. Mothers of 3,733 infants reported that their infants were always placed to sleep in the same position; among these infants, researchers found that those who slept on their backs were less likely to have fevers than were infants who slept on their stomachs at one month of age. At six months old, back sleepers were less likely to develop a stuffy nose than were stomach sleepers. At three and six months old, back sleepers needed to visit the doctor less often for ear infections than did stomach sleepers. Moreover, at six months, the mothers of back sleepers reported fewer instances of infant sleeping trouble than did the mothers of stomach sleepers. None of the infants in the study were reported to have choked on their vomit (Arch Pediatr Adolesc Med 2003; 157:469-74).
In countries that have experienced successful risk reduction campaigns, there is a change in the contribution of risk factors to SIDS between the pre- and post-intervention periods. Between May 1997 and April 2000, the NICHD conducted a population-based, case-controlled study in 11 counties in California to evaluate SIDS risk since the initiation of the Back to Sleep campaign. This study provided critical evidence to support the recommendation that "back is best" for all sleep periods, and that the position should be used consistently by all caregivers. The researchers found that infants who were last placed on their sides to sleep were twice as likely to die of SIDS as infants who were last placed on their backs to sleep. In addition, the risk of SIDS was significantly increased if the infants turned from their sides to their stomachs during sleep. While the reason isn't clear, the researchers believe that the instability of the side position makes it more likely for babies to roll over onto their stomachs during sleep. When the researchers looked specifically at the position in which an infant was last placed to sleep, a pattern emerged when compared to the usual sleeping position; if an infant who was usually placed to sleep on the back was then placed to sleep on the stomach or side, his or her SIDS risk was seven to eight times greater than that of an infant who was always placed to sleep on his or her back (AJE 2003; 157:446-455).
Since the NICHD-led Back to Sleep campaign began, the rate of SIDS has declined steadily and significantly. According to preliminary figures from the National Center for Health Statistics, the SIDS rate for 2001 was about 0.5 deaths per 1,000 live births, a decline of more than 50 percent since the campaign began.
Health Disparities
Although SIDS rates have declined in all segments of the population, a significant disparity still exists between majority and minority populations. In particular, SIDS rates are two to three times higher among African Americans and American Indians compared with whites. The PPB has supported two case-controlled studies to specifically shed light on this racial disparity by examining the pattern of SIDS risk factors.
Northern Plain Indians have the highest rates of SIDS in the nation. To understand the reasons for this high SIDS rate, the PPB, in collaboration with Aberdeen Area Tribal Chairman's Health Board, the Indian Health Service, and the CDC conducted the Aberdeen Area Infant Mortality Study (AAIMS). The AAIMS identified risk factors in this population that had not previously been reported. For instance, despite reports that mothers in the study reduced their alcohol consumption significantly by their second trimester, binge drinking during the mothers' first trimester of pregnancy increased the risk for SIDS eight-fold. Any maternal alcohol use during the periconceptual period (three months before pregnancy or during the first trimester) was associated with a six-fold increase in the risk of SIDS. The study also found that infants were more likely to die of SIDS if they wore two or more layers of clothing during sleep. More positively, though, infants whose mothers were visited by public health nurses before and after giving birth were only one-fifth as likely to succumb to SIDS as those whose mothers were not visited. These data highlight the importance of health outreach to this population (JAMA 2002; 288:2717-2723).
The AAIMS was the first study to link epidemiological findings with neurochemical deficits in the developing human brain. The binding abnormalities in serotinergic receptors of the medullary regions of AAIMS infants were similar to those observed in SIDS cases from other populations. In addition, the serotinergic abnormalities in the arcuate nucleus in these infants were associated with exposure to adverse prenatal exposures, such as cigarette smoking (p=0.011) and alcohol use (p=0.075), during the periconceptual period or throughout pregnancy. These results suggest that these prenatal exposures may contribute to abnormal development of the fetal medullary serotinergic system in infants who die from SIDS (J Neuropathol Expt Neurol 2003; 62:1166-1177).
From 1993 through 1996, the NICHD and the CDC supported a case-controlled study of infant deaths in Cook County that employed standardized death scene investigation and autopsy protocols to elucidate the unique factors of SIDS deaths. In this primarily African American, urban sample, prone sleeping was found to be a significant risk factor for SIDS, after adjusting for potential confounding variables and other sleep environment factors (OR 4.0, 95% CI 1.8-8.8); approximately one-third of the SIDS deaths could be attributed to prone sleep position. Fewer case mothers of SIDS infants (46 percent) than control mothers of living infants (64 percent) reported being advised about sleep position in the hospital following delivery (P<0.001). Of those advised, a similar proportion of case mothers as control mothers were told to use the incorrect (stomach) position, but a higher proportion of African American mothers (cases and controls combined) were advised to use that position compared with non-black mothers (Pediatrics 2002; 110:772-780). Further analysis showed that, in addition to prone sleep position, a soft sleep surface, pillow use, and bed sharing in combinations other than with parent(s) alone significantly increased SIDS risk. Pacifier use significantly decreased risk (Pediatrics 2003; 111:1207-1214). Pacifier use has been found to be protective for SIDS in several studies worldwide, but the mechanism is unknown.
In 1999, the Institute's Back to Sleep campaign began a partnership with several African American organizations to develop focused, community-centered information materials and outreach to help reduce the risk of SIDS in African American communities. The NICHD, in partnership with the National Black Child Development Institute, enlisted the
In 1999, PPB staff initiated a planning process for the Institute's third, five-year strategic plan on SIDS research. A working group composed of distinguished scientists and health care professionals from around the country, in collaboration with staff from the PPB and other NICHD Branches, identified research objectives and strategies designed to achieve these objectives. The group drew on previous and ongoing planning efforts, conferences, workshops, and research findings to develop Targeting Sudden Infant Death Syndrome (SIDS): A Strategic Plan, a public health agenda that would guide the Institute's research over the next five years. Like all NICHD strategic plans, the draft plan was placed on the NICHD Web site for public comment. The final plan was published in the summer of 2001, and is available on the NICHD Web site at: 
In 1994, the NICHD formed a partnership with the 
