A search for the 'Holy Grail' in the evaluation of febrile neonates aged 28 days or less: a prospective study.

OBJECTIVE

To determine the reliability of low-risk criteria to exclude serious bacterial infection (SBI) in febrile neonates aged ≤28 days.

METHODS

All febrile neonates who were hospitalized for fever evaluation were prospectively divided into 2 groups by risk status for SBI. The following criteria were used to define low risk: (1) unremarkable medical history; (2) well-appearing; (3) no focal signs of infection; (4) white blood cell count between 5000 and 15,000/mm(3); (5) normal urinalysis; (6) no mucoid or bloody diarrhoea.

RESULTS

Of the 465 enrolled neonates, 177 (38.1%) were considered high risk for SBI and 288 (61.9%) low risk. SBIs were found in 55 (31.1%) neonates in the high-risk group compared to 10 (3.5%) in the low-risk group (p < 0.001). The sensitivity, specificity, positive predictive value and negative predictive value of the criteria for all types of SBI were 84.6% (95% confidence interval (CI) 73.9-91.4%), 69.5% (95% CI 64.8-73.8%), 31% (95% CI 27.3-35.1%) and 96.5% (95% CI 94.3-98%), respectively.

CONCLUSIONS

The defined criteria are not sufficiently reliable to exclude an SBI or an invasive SBI. We therefore suggest that all febrile neonates in this age group should be hospitalized for complete evaluation and consideration of empirical intravenous antibiotic treatment.