Abstract
Background: Standard laboratory methods for early confirmation of late-onset sepsis (LOS) in neonates are time-consuming, flawed with poor sensitivity and specificity, and not helpful in decision-making regarding selection and initiation of empiric anti-microbial therapy (AMT).
Aim: To evaluate the predictive value of relevant clinical and laboratory parameters [complete blood count, serum C-reactive protein (CRP), procalcitonin (PCT) and Staphylococcus-specific PCR] in neonates suspected of having developed LOS.
Patients and Methods: NICU neonates were prospectively followed for suspected septic events. During the study period, 111 neonates developed 148 suspected septic events beyond 3 days of age, and comprised the study population. Clinical signs and laboratory abnormalities at onset of sepsis were recorded, including serum levels of CRP and PCT, results of Staphylococcus-specific PCR, microbiological data and the AMT instituted.
Results: Of the 148 events of suspected LOS, 26 (17.6%) had positive blood cultures (proven LOS) with gram-positive bacteria, gram-negative bacteria and Candida, accounting for 65.4%, 23.1% and 11.5% of the events, respectively. Variables that were significantly associated with subsequently confirmed LOS included hypotension [relative risk (RR)=5.6; 95% confidence interval (CI): 3.29–9.53]; mechanical ventilation (RR=2.46; 95%CI: 1.24–4.86); immature/total neutrophil ratio (I/T) > 0.2 (RR=5.13; 95%CI: 2.54–10.31); CRP >1.0 mg/dL (RR=2.85; 95%CI: 1.32–6.15); and, small for gestational age status (RR=2.13; 95%CI: 1.03–4.38). PCT was not significantly associated with LOS. For detection of Staphylococcal bacteremia, Staphylococcus-specific PCR demonstrated a sensitivity of 57.1%, specificity 94.7%, a positive predictive value of 53.3%, and a negative predictive value of 95.4%.
Conclusions: Hypotension, mechanical ventilation, I/T >0.2, CRP >1.0 mg/dL and SGA status at onset of sepsis are significant predictors of proven LOS in neonates. Staphylococcus-specific PCR is of value only in ruling out Staphylococcal sepsis. We propose an amended protocol for the approach to neonates with suspected LOS.
Article PDF
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Makhoul, I., Yacoub, A., Smolkin, T. et al. 232 Revised Approach to Suspected Late-Onset Sepsis in Neonates: Added Value of C-Reactive Protein and Staphylococcus-Specific PCR.. Pediatr Res 58, 394 (2005). https://doi.org/10.1203/00006450-200508000-00261
Issue Date:
DOI: https://doi.org/10.1203/00006450-200508000-00261