Reliability of measuring serum C-reactive protein or procalcitonin with Rochester criteria in febrile neonates aged...

To investigate the utility of measuring serum C-reactive protein (CRP) or procalcitonin (PCT) level in conjunction with the Rochester criteria to exclude invasive bacterial infection (IBI) and urinary tract infection (UTI) in neonates aged 8-28 days with fever without an identifiable source of infection (FWSI).

The medical records of all neonates aged 8-28 days with FWSI presenting to our pediatric emergency department between January 2018 and December 2022 were analyzed retrospectively. Only those who underwent workup based on the Rochester criteria with the serum biomarkers were included. Our findings were validated in a separate cohort in 2023.

In total, 191 neonates were included in the study, 54 (28.27 %) of whom had IBI and/or UTI. Serum biomarkers were analyzed, and higher CRP (p < 0.001) and PCT (p = 0.013) levels were found to be associated with IBI and UTI on multivariate logistic regression analysis. To exclude IBI and UTI, combining the Rochester criteria with a PCT < 0.5 ng/mL (Chang Gung criteria) had a higher negative predictive value (NPV) of 100.00 % than the Rochester criteria alone (93.06 %) and a similar positive predictive value (40.60 % and 41.18 %, respectively). Chang Gung criteria was validated in a cohort of 42 neonates aged 8-28 days with FWSI in 2023.

Using PCT at a threshold of 0.5 ng/mL alongside the Rochester criteria (Chang Gung criteria) provides a higher NPV for excluding IBI and UTI in neonates aged 8-28 days with FWSI.