Sudden Infant Death Syndrome has caused heartbreak and throughout history. The aftermath of an unexpected death of a baby often leaves medics and new parents scrambling for answers, in many cases leading to unnecessary self-blame and recriminations. While public health campaigns have reduced infant mortality, the underlying causes of these tragic events and which children are most at risk remain mysterious. This month in JAMA pediatrics, epidemiologists reported that routine blood samples collected soon after birth could help identify a subset of children who are more vulnerable to SIDS1.
A Test for SIDS?
University of California, San Francisco (UCSF) researchers say that they may have discovered a set of key indicators that will help paediatricians to spot infants at risk of SIDS early enough for intervention and monitoring. In a research paper, published September 3rd 2024, epidemiologists working with paediatricians from across the United States announced a set of biomarkers that seem to predict SIDS.
In the paper, the researchers describe how they used existing blood test results collected from new born babies. They used the widely used Newborn screening results to pick out a set of biochemical indicators that cropped up again and again in children who would go on to suffer SIDS. The paper’s first author, epidemiologist Scott Oltman explained in a press release, ‘This study suggests that metabolic factors may play a crucial role in SIDS,’ he went on to say, ‘These patterns could help identify children at higher risk, potentially saving lives in the future.’
Fourteen Key Factors
The researchers identified 14 commonly tested metabolic markers. They combined the results of these biochemical indicators into a statistical model to calculate the probability that a child was at risk from SIDS. The model performed well, allowing the researchers to assign likelihood score to a given infant that could predict strongly whether that child had gone on to experience SIDS. The researchers hope that eventually paediatricians will be able to use this model to spot which infants will need special attention or screening for metabolic disorders that could cause SIDS. This work takes routine heel prick results collected from newborn babies and reanalyses the rusults to predict SIDS.
While these findings are in their early stages, Oltman and his colleagues are optimistic about the next steps. The patterns they noticed in the blood test data not only helped them to identify at-risk infants, but also provided clues pinpointing the underlying biology that makes some children more vulnerable. Oltman said, ‘This study is a critical step toward integrating metabolic markers with potential genetic markers and other risk factors to better assess the risk of SIDS in infants.’
Unexplained Tragedy
Sudden Infant Death Syndrome is the cause of death that doctors give when an infant passes away suddenly and unexpectedly before the age of twelve months with no obvious explanation2. The United States’ Centers for Disease Control and Prevention put rates of SIDS at 3.84 infants per 100,000 in 20203. While epidemiologists have identified multiple risk factors for SIDS, often times these are focused on family circumstances and parenting over biological explanations.
Parents can take action to address some of these risk factors, for example prone sleeping position (letting a baby sleep on its front, face down) or smoking or drinking alcohol during pregnancy. Others are beyond a mother or father’s control. For instance, researchers have long associated structural racism and poverty with SIDS. While it is clear that there are many different causes of SIDS, and many combinations of circumstances can affect infant health. Better understanding the physiological component will provide parents and doctors with improved tools to help children grow up healthy.
Tissue samples taken by pathologists during autopsy have yielded some clues about the underlying health problems that lead to SIDS. Tissue analysis has revealed abnormalities like fatty acid β-oxidation disorders and metabolic dysfunctions – that is, issues in how they make and use energy. Genetic tests have hinted that children born with gene variants that trigger heart problems, inflammation, high serotonin levels and other disorders are at higher risk of SIDS. Researchers hope that genetic testing could assist in identifying vulnerable children.
Looking to the Past
This huge retrospective analysis performed by UCSF biostatisticians took efforts to identify biomedical risk factors and the causes of SIDS to the next level. The team of epidemiologists dug through nearly 2.3 million health records of children born between 2005 and 2011 whose doctor had requested a full panel of metabolic data during newborn screening.
The researchers found 354 children of the 2,276,578 eligible infants born in California during the six-year interval whose deaths were attributed to SIDS. The statisticians carefully matched each case of SIDS to four healthy babies with the same gestational age at birth, weight and sex. This was to allow the researchers to exclude those factors from their analysis. They then split the infants into two groups. Group 1 was used to search for similarities between the SIDS cases, and the ways their newborn screening test results differed from healthy newborns. They set aside Group 2 to check how accurate their predictions were later.
Record Searches
The scientists meticulously looked through the health records of each of the selected newborns in Group 1 using statistical software to highlight and extract what their newborn screening results showed for each individual test. The software would identify any chemical that kept showing up as different in children who suffered SIDS. It also lit up any other factors that could correlate to SIDS, such as race or access to prenatal care.
The results of the first analysis were that 14 different metabolites (17-hydroxyprogesterone, alanine, methionine, proline, tyrosine, valine, free carnitine, acetyl-L-carnitine, malonyl carnitine, glutarylcarnitine, lauroyl-L-carnitine, dodecenoylcarnitine, 3-hydroxytetradecanoylcarnitine, and linoleoylcarnitine) presented up as potential indicators of SIDS.
Making a Model
The researchers put together the combination of test results of these metabolites to make a newborn screening test result profile that would look like the typical findings in an infant that would eventually experience SIDS. This is what statisticians call a ‘model’. Their final model included the blood test results for eight chemicals: 17-hydroxyprogesterone, alanine, glycine, free carnitine, propionyl-L-carnitine (C-3), C-5DC, C-12:1, and C-14OH, age when the blood was collected, mother’s age, infant sex and whether or not they received adequate prenatal care.
They then went to the Group 2, the medical records that they had held back, to test their predictions. They hid the SIDS status of each child so they would not know whether a child had survived past 12 months.
Next, they compared each child’s medical records to their model to see how similar that child was to their profile of a typical case of SIDS. Depending on how closely each child matched the SIDS profile, the researchers would give them a score. This score indicated how likely it was that a child had gone on to suffer SIDS.
After they assigned the probability scores, they unmasked the records to judge how well their SIDS profile worked to pick up which children had SIDS. They found that a score of 0.5 on their scale meant that a child was 14 times more likely to suffer SIDS than a child with a score below 0.1. Ninety-seven point five per cent of children identified as not at risk of SIDS were healthy and lived beyond a year.
Looking to the Future
This study not only provides a test that doctors might eventually use in neonatal units to screen for SIDS risk, but also discovered that a type of metabolite, acylcarnitines, is significant markers for SIDS. It’s too early to tell whether these chemicals are a sign that something is wrong, or whether the process that produces the acylcarnitines is triggering the underlying disorder that causes SIDS. The UCSF researchers, however, are already investigating what these molecules have to say about the condition.
Researchers need to test, validate and refine the model before it can be rolled out widely, so you might not see this on your child’s record in the next year. Given that it takes advantage of existing test results, however, if scientists can put together a consistent, validated model a test for SIDS will be easy to slip into routine newborn screening results.
References
- Oltman SP, Rogers EE, Baer RJ, et al. Early Newborn Metabolic Patterning and Sudden Infant Death Syndrome. JAMA Pediatrics. Published online September 9, 2024. doi:10.1001/jamapediatrics.2024.3033 ↩︎
- Shapiro-Mendoza CK, Parks S, Lambert AE, Camperlengo L, Cottengim C, Olson C. The Epidemiology of Sudden Infant Death Syndrome and Sudden Unexpected Infant Deaths: Diagnostic Shift and other Temporal Changes. In: Duncan JR, Byard RW, eds. SIDS Sudden Infant and Early Childhood Death: The Past, the Present and the Future. University of Adelaide Press; 2018. Accessed September 11, 2024. http://www.ncbi.nlm.nih.gov/books/NBK513373/ ↩︎
- Data and Statistics for SIDS and SUID | CDC. May 14, 2024. Accessed September 11, 2024. https://www.cdc.gov/sids/data.htm ↩︎
