Epidemiology and Mortality of Invasive Staphylococcus aureus Infections in Hospitalized Infants.

Historically, Staphylococcus aureus has been a leading cause of morbidity and mortality in the neonatal intensive care unit (NICU). The current incidence and attributable mortality of late-onset invasive S aureus infection in hospitalized infants is unknown. To estimate the incidence and attributable mortality of late-onset S aureus infection among hospitalized infants in the US. This retrospective cohort study included an emulated trial among a national convenience sample of 315 NICUs within the US between 2016 and 2021. Participants were infants aged at least 4 postnatal days who were hospitalized in a participating NICU. Data were analyzed from May to August 2024. The primary exposures were birth weight and postnatal age. The outcomes were the incidence and attributable mortality of late-onset invasive S aureus infection. Methicillin-resistant and methicillin-sensitive S aureus classification was not universally available; thus, all invasive S aureus infections were pooled. From 468 201 infants (260 491 [55.6%] male; median [IQR] gestational age, 36 [33-38] weeks) eligible for analysis, 1724 infants experienced 1762 infections for an overall incidence of 37.6 (95% CI, 35.9-39.4) invasive S aureus infections per 10 000 infants. Most infants with invasive infections were 32 weeks' gestational age or younger (1394 infants [80.9%]), very low birth weight (VLBW; ie, <1500 g) (1318 infants [76.5%]), and/or had a central line during their hospital stay (1509 infants [87.5%]). Invasive infections mostly included bloodstream infections (1505 infections [85.4%]), and 1160 infections (65.8%) occurred within 4 to 28 postnatal days. Birth weight inversely correlated with incidence: infants with VLBW experienced a more than 20-fold higher incidence relative to infants born weighing at least 1500 g (227.1 [95% CI, 215.3-239.4] vs 10.1 [95% CI, 9.1-11.1] infections per 10 000 infants). Most deaths following invasive infection occurred among VLBW infants (189 of 209 deaths [90.4%]). Compared with matched infants without a late-onset invasive S aureus infection, infected infants had an absolute difference in mortality rate of 5.3% (95% CI, 3.8%-6.8%). This cohort study found late-onset invasive S aureus infection and subsequent attributable mortality disproportionally affected infants with VLBW. Targeted infection prevention and control measures are necessary to reduce morbidity and mortality from invasive S aureus infections in this vulnerable population.