Neonatal early-onset sepsis calculator associated with decreased antibiotic use and no increase in sepsis

Référence

Achten NB, Klingenberg C, Benitz WE, et al. Association of use of the neonatal early-onset sepsis calculator with reduction in antibiotic therapy and safety: a systematic review and meta-analysis. JAMA Pediatr 2019;173(11):1032-1040.

Sommaire

These authors searched several databases to identify peer-reviewed publications of any design type as long as the study validated an online neonatal early-onset sepsis calculator (https://neonatalsepsiscalculator.kaiserpermanente.org) or compared its use in newborns (34 weeks' gestation or later). Additionally, the authors used citation indices to identify any studies that cited the included studies. Two of the authors independently assessed studies for inclusion and risk of bias with disagreements resolved through discussion and third-party adjudication. Ultimately, they identified 13 studies with 175,752 newborns; none of the studies were randomized trials and most occurred in the United States, the Netherlands, and Australia. Six of the studies (with 172,385 newborns) were before-and-after studies and 7 were hypothetical modeling exercises. The studies used varying inclusion criteria: 3 assessed only well-appearing newborns, 6 included only infants born to mothers with chorioamnionitis, 2 included only infants treated with antibiotics, and the lowest gestational age varied between 34 and 36 weeks. The overall risk of bias was high for 9 studies, low for 2 studies, and unclear for 2 studies. When looking at the studies that included all infants, 5.6% of the infants managed by usual care received antibiotics compared with 3.2% of those for whom the sepsis calculator was used (RR 0.56, 95% CI 0.53 - 0.59). Among the infants whose mothers had chorioamnionitis, 94.6% of those managed by usual care received antibiotics compared with 18.7% of those managed with the calculator. Although only 3 of the studies were specifically designed to include safety outcomes, some safety-related issues—missed cases of sepsis, readmissions, treatment delay, morbidity and mortality— were discussed in all but one of the studies. In general, the studies reported no increased rate of early onset sepsis in the calculator-managed infants, but the confidence intervals were pretty wide (pooled odds ratio 0.96; 95% CI 0.26 - 3.52).