Background The incidence of Early Onset Neonatal Sepsis (EONS) in the UK is currently estimated at 0.9/1000 live births. NICE have published guidance (CG149) for treating infants based on antenatal risk factors and clinical indicators. In the UK, 13–20% of infants are treated for suspected EONS based on CG149. This represents a significant burden, impacting on family bonding, breastfeeding, physical trauma through repeated cannulation, as well as on hospital bed status and finances. Covid-19 has compounded the issue, implementing restricted visitation disrupting the family unit. The Kaiser Permanente Sepsis Tool (KPT) was developed to determine the likelihood of infection based on multivariate analysis, has been used in the USA to successfully and safely to reduce the numbers of babies being prophylactically treated for this rare condition. Following on for their success we have now evaluated KPT for use within a UK demographic at Medway Maritime Hospital.

Objectives

1. Can KPT reduce prophylactic IV antibiotic use in well babies, when compared to NICE guideline CG149

2. Can KPT achieve this safely within our local demographic

Methods Data was collected from 62 newborn infants treated at Medway Maritime Hospital, between November 2019 and January 2020. These patients received prophylactic IV antibiotics as per NICE CG149. Inclusion criteria; ≥34 weeks gestation and infants who were clinically well enough to be managed on the postnatal ward. The following data was obtained at the point of starting treatment; gestation, clinical examination, maternal peripartum temperature, rupture of membranes, maternal group B streptococcus. Subsequently peak CRP, blood and/or CSF culture results and duration of stay were recorded. Infants were classified into a risk category based on their peak CRP (Low <5, Medium 5–10, High >10). These findings were then compared to treatment recommended by the KPT (local incidence of EONS 1/1000 live births).

Results 16 infants were classed as high risk (26%), 16 infants medium risk (27%) and 29 infants low risk (47%). The most common indication to treat was maternal pyrexia. Of infants screened, KPT recommended antibiotics for 2 infants (3%); both of these were low risk. 47% of infants would have been kept under enhanced observations, with antibiotics being started if clinical symptoms developed. All blood and CSF cultures were negative. Of the infants classified as high risk, 50% of these infants would have been allocated to routine care. In the infants classified as high risk, there were no differences in the risk factors for neonatal sepsis, when compared with other risk categories.

Conclusions KPT represents a lucrative opportunity to reduce antibiotic use in well infants on the postnatal ward. However in line with similar studies, our results have highlighted that, just as with NICE CG149, is not infallible and liable to occasionally missing an asymptomatic child. Clinical vigilance is of the utmost importance and implementation of KPT would be have to marry with robust systems of neonatal observation and workforce training. A more conservative approach to EONS will invariably be associated with risk and ultimately it will be up to individual units to agree on what is acceptable.