Reclassification of the Etiology of Infant Mortality With Whole-Genome Sequencing

ABSTRACT Infant mortality rates are quite high even in developed countries; in the United States, infant deaths occur approximately 1 in 200 live births. The leading cause of infant mortality is congenital malformation or chromosomal abnormality, which cause roughly 20% of infant deaths, followed by other causes such as pregnancy complications, preterm birth or low birth weight, and sudden infant death syndrome. Improvements in treatments for certain conditions such as pregnancy complications or prematurity have previously resulted in a reduction in mortality due to these causes. Although the same may be true of congenital malformations or chromosomal abnormalities, the relationship between these factors and infant mortality has not been closely examined. Previous studies have largely been retrospective in nature and varied greatly among method and records used, perpetuating inaccuracies and imprecision. This study aimed to clarify and add to previous literature by examining the relationship between single-locus genetic disease (Mendelian genetic disease) and overall infant mortality. This study applied diagnostic whole-genome sequencing (WGS) to infants in intensive care to diagnose diseases of unknown etiology in a single hospital system between 2015 and 2020. The study was retrospective and included review of medical records and death certificates, as well as results from WGS. Where possible, WGS information was obtained on living infants and their parents, and where that was not possible, it was obtained postmortem. Death certificates and causes of death were also examined and compared with WGS results. In the final sample, 112 infants who died underwent WGS either premortem or postmortem. WGS identified genetic diseases in 46 (41%) of these infants. Of the genetic diseases identified, 83% had previous evidence supporting an association with childhood mortality, and evidence was gathered from previous literature that could improve outcomes for 30%. Death certificate comparisons showed that of 45 infants where WGS showed underlying genetic disease, 62% were not reported on the certificate or attributed to the cause of death. In addition to the 112 infants who died, the study population included 434 infants who survived and also had WGS performed. No significant differences were found between the groups in sex, race, or ethnicity. Types of genetic disease varied between infants who died and those who survived, with only 4% being found in both. Certain diseases were more likely to occur in infants who survived, and others were more likely to occur in infants who died. Specific etiology had some prognostic value, positively predicting either death or survival. These results indicate that first, after examination of WGS in comparison with death certificates, genetic diseases are chronically underreported. In addition, WGS showed that etiology of leading causes of infant mortality could be fundamentally different than previously thought. WGS is also not currently a widespread treatment, and although steps are being taken to make it more widely accessible, there are still many changes that need to be made. This study is limited by its small size, and may have limited generalizability to other patient populations. Future research should characterize the optimal breadth of WGS testing and implementation, as well as the potential for WGS to reduce infant mortality by identifying treatable disorders.