Evolution in Infectious Causes of Death in Liveborn Children in a Major Tertiary Children's Hospital: A Retrospective Autopsy Review of over 1200 Cases

ABSTRACT

Background: Infections have historically been a leading cause of death, particularly in children. Medical advances, including vaccines and antimicrobials, have significantly decreased infection-related deaths, but infections remain a cause of pediatric mortality, especially in premature infants. The types of infections implicated in childhood deaths have changed with these advances, for example, meningitis and meningococcal infections were leading causes in 1981 but not in the later period. The incidence and etiologies of infection- related deaths may be altered by major events that modify not only medical practices but also societal attitudes and activities. Examples of such events include the HIV/AIDS epidemic that began in the early 1980s and the more recent COVID-19 pandemic. In order to investigate changes in infection-related pediatric deaths over time, we analyzed and compared autopsy cases performed during 5-year span prior to both the HIV/AIDS epidemic and the COVID-19 pandemic in which infections contributed to death. Method(s) Review of all autopsy cases performed at our institution between 1/1/1975-1/1/1980 and between 1/1/2015-1/1/2020 was performed to identify cases in which infection directly contributed to death, comprising 1262 cases. Only liveborn children were considered, and neonatal sepsis from amniotic sac infections was excluded. Comparison of decedent characteristics and infectious etiologies between the two time periods was performed, identifying age, race, sex, gestational age (for decedents less than 3 months of age), and etiologic class of agent (bacterial, viral, fungal or parasitic). TORCH infections and vaccine-preventable illnesses were specifically assessed. Proportions were compared using 1 (assessing TORCH, vaccine-preventable, and prematurity deaths)- or 2-tailed (all others) z-tests, with significance calculated at the < 0.05 level. Result(s) In the 1970s cohort, 300 infectious autopsy cases were identified in liveborn children;73 were identified in the 2010s. Compared to the 2010s cohort, the 1970s decedents were more likely to be white (85% v 53%, p=0.012), comprise children aged 1-5 and 13+ (22% v 6.8% [p=0.003] and 16.4% v 8.3% [p=0.036]), and were less likely to be premature (66.7% v 80.4%, p=0.039). Vaccine-preventable illnesses (for example measles) accounted for 36 deaths in the 1970s cohort but only 2 in the 2010s cohort (p=0.009). Thirteen children died of TORCH infections (CMV, toxoplasmosis and HSV) versus 5 in the 2010s (CMV and HSV), which did not reach statistical significance. Conclusion(s) Pediatric mortality secondary to infections has decreased significantly compared to fifty years ago, especially in younger children and in relation to vaccine-preventable infections such as meningococcal disease. This drop is largely attributed to medical advances, including vaccines and antimicrobial medications. Additional contributing factors could include practices adopted post-HIV/AIDS, especially in the community. Further exploration of how such changes in medical and social practice impacted mortality and comparing them to changes occurring in the intra/post-COVID-19 era, is helpful. Yet, with the increased survival of premature infants, they remain at risk of devastating consequences from infections.