56/100

Moderate evidence

Cite
More
EmailLinkedInX

Sepsis warning score rose faster before severe cases in 84 premature infants, study finds

Study title: A pulse oximetry warning system predicts severity of late-onset sepsis in very low birth weight neonates

Evidence type
Observational study
Sample
n = 84
Sources
1 sourcePeer-reviewed
Published
2 min read

Study title: A pulse oximetry warning system predicts severity of late-onset sepsis in very low birth weight neonates

A multicenter retrospective cohort study analysed 89 culture-confirmed late-onset sepsis episodes in 84 very low birth weight infants. It found that a machine learning warning score built from continuous heart rate and oxygen saturation data climbed more steeply before severe episodes than before milder ones.

Read the study abstract

Late-onset sepsis (LOS) causes morbidity and mortality in very-low birthweight (VLBW) neonates. We previously developed a machine learning model, the Pulse Oximetry Warning System (POWS), to predict LOS in the next 24 h. The neonatal Sequential Organ Failure Assessment (nSOFA) score predicts sepsis mortality. We hypothesized that POWS scores would differentiate severe LOS events, defined by nSOFA >4, from milder events. We conducted a multicenter, retrospective cohort study including VLBW infants diagnosed with…

The gain

In a multicenter retrospective study of 89 late-onset sepsis episodes in 84 very low birth weight infants, the Pulse Oximetry Warning System score rose more steeply in the 24 hours before severe episodes (30% of cases) than before milder ones.

In brief

Severe sepsis, defined as a neonatal Sequential Organ Failure Assessment score above 4 within 48 hours of blood culture, occurred in 27 episodes (30%). Average scores in the 24 hours before diagnosis were higher in the severe group.

The authors say the pattern suggests the tool could prompt earlier intervention for the sickest infants. Because the study looked back at existing records, it did not test whether acting on the score improves outcomes.

Main points

  1. The study covered 89 late-onset sepsis episodes in 84 very low birth weight infants across multiple centers.
  2. Severe sepsis, defined as an nSOFA score above 4 within 48 hours of blood culture, occurred in 27 episodes (30%).
  3. Mean warning scores were higher in the 24 hours before severe episodes than before non-severe ones.
  4. The score rose at a significantly higher rate in the hours before diagnosis in severe episodes.
  5. The Pulse Oximetry Warning System uses continuous heart rate and oxygen saturation data to predict sepsis up to 24 hours before clinical diagnosis.

The full story

The researchers included only infants with culture-confirmed late-onset sepsis and continuous vital sign data. They used mixed-effects regression to examine the link between the warning score and the nSOFA score, which predicts sepsis mortality.

The warning system was developed earlier to predict sepsis in the next 24 hours. This study asked a different question: whether its score also separates severe events from milder ones.

The authors conclude the score could give earlier warning of the most severe events, potentially enabling earlier intervention in this vulnerable group. They describe this as adding to existing evidence rather than showing improved outcomes.

Sources

1 on file
  1. Pediatric Researchdoi:10.1038/s41390-026-05665-2
    Peer-reviewed2026-11-14

Talk it over

Three questions to start on, two more if you have time
The warning score rose more steeply before the 27 severe episodes (30%) than before milder ones. If you were a clinician caring for a very low birth weight infant, how would you use that signal?

Start with a question

Sepsis warning score rose faster before severe cases in 84 premature infants, study findshttps://truvace.com/item/a-pulse-oximetry-warning-system-predicts-severity-of-late-onset-sepsis-in-very-l

  1. The warning score rose more steeply before the 27 severe episodes (30%) than before milder ones. If you were a clinician caring for a very low birth weight infant, how would you use that signal?

    Answer in the thread
  2. The study looked back at 89 culture-confirmed episodes in 84 infants. Which infants are not in that group, and how could their absence change what the score shows?

    Answer in the thread
  3. The authors say the score could prompt earlier intervention. What kind of study would show whether acting on the score improves outcomes for these infants?

    Answer in the thread
If you have more time
  • Look up how the nSOFA score is used to define severe sepsis. What might change if severity were defined another way?

    Answer in the thread
  • In a neonatal unit you know of, who would decide how staff respond when the score climbs, and what would they need to see first?

    Answer in the thread

What readers say