Prediction Models for Neonatal Health Care–Associated Sepsis: A Meta-analysis

Abstract

BACKGROUND AND OBJECTIVES: Blood culture is the gold standard to diagnose bloodstream infection but is usually time-consuming. Prediction models aim to facilitate early preliminary diagnosis and treatment. We systematically reviewed prediction models for health care–associated bloodstream infection (HABSI) in neonates, identified superior models, and pooled clinical predictors. Data sources: LibHub, PubMed, and Web of Science. METHODS: The studies included designed prediction models for laboratory-confirmed HABSI or sepsis. The target population was a consecutive series of neonates with suspicion of sepsis hospitalized for ≥48 hours. Clinical predictors had to be recorded at time of or before culturing. Methodologic quality of the studies was assessed. Data extracted included population characteristics, total suspected and laboratory-confirmed episodes and definition, clinical parameter definitions and odds ratios, and diagnostic accuracy parameters. RESULTS: The systematic search revealed 9 articles with 12 prediction models representing 1295 suspected and 434 laboratory-confirmed sepsis episodes. Models exhibit moderate-good methodologic quality, large pretest probability range, and insufficient diagnostic accuracy. Random effects meta-analysis showed that lethargy, pallor/mottling, total parenteral nutrition, lipid infusion, and postnatal corticosteroids were predictive for HABSI. Post hoc analysis with low-gestational-age neonates demonstrated that apnea/bradycardia, lethargy, pallor/mottling, and poor peripheral perfusion were predictive for HABSI. Limitations include clinical and statistical heterogeneity. CONCLUSIONS: Prediction models should be considered as guidance rather than an absolute indicator because they all have limited diagnostic accuracy. Lethargy and pallor and/or mottling for all neonates as well as apnea and/or bradycardia and poor peripheral perfusion for very low birth weight neonates are the most powerful clinical signs. However, the clinical context of the neonate should always be considered.