In 283 very preterm infants, biomarker-guided antibiotics did not show noninferiority for survival without major morbidity

What the research found

A randomized noninferiority trial was published on 12 November 2026. It ran in 17 Italian neonatal units from 1 June 2023 to 31 January 2025. Infants born from 25 weeks and 0 days to 31 weeks and 6 days were allocated to procalcitonin- and presepsin-guided antibiotics for early-onset sepsis, or to universal antibiotics. The sample was 283 infants. Mean gestational age was 30.2 weeks (standard deviation 2.2). 161 were girls (57 percent). Newborns are under 40, and the abstract does not limit the cohort to one sex, so the tags are all and under 40. Guided care had 143 infants and universal care 140. Survival without major morbidity was 76 of 143 (53 percent) versus 77 of 140 (55 percent). The risk difference was minus 1.8 percentage points (90 percent CI minus 11.9 to 8.3, P=0.32). The preset margin was minus 5 percentage points. The lower bound crossed that margin, so noninferiority was not shown. Gaps for late-onset sepsis, intraventricular hemorrhage, necrotizing enterocolitis, periventricular leukomalacia, retinopathy, and death were not clear. No infant developed early-onset sepsis. Registry NCT05796115. DOI 10.1001/jamanetworkopen.2026.43418. Not medical advice.

Evidence level: Clinical trial. This is among the stronger kinds of evidence, but one study rarely settles a question.

Read the original at JAMA Network Open