Objectives:

To evaluate the accuracy of serum amyloid A (SAA), an acute phase protein in the detection of neonatal early-onset sepsis, by means of a fast automated SAA kit.

Study Design:

Full-term infants< 72 h of age, who had risk factors and/or were suspected of having sepsis, were eligible for study. The levels of SAA were taken at 0, 24 and 48 h post sepsis evaluation. Thirty matched infants served as a control group for comparing SAA concentrations.

Results:

Of 104 infants eligible for entry to the study, 23 had sepsis and 81 had not sepsis. The SAA levels of the septic group were significantly higher than those of the nonseptic group at 0, 24 and 48 h (P< 0.01 for all time points). In comparison with C-reactive protein (CRP), SAA levels rose earlier and in a sharper manner, had higher levels and returned faster to normal values in infants with early onset sepsis. At 0 h post-sepsis evaluation, serum SAA had an overall better diagnostic accuracy for predicting early onset sepsis than CRP (sensitivity (96 vs 30%), specificity (95 vs 98%), positive predictive value (85 vs 78%), negative predictive value (99 vs 83%), positive likelihood ratio (19 vs 12), and negative likelihood ratio (0.05 vs 0.71).

Conclusions:

SSA is advocated as an inflammatory marker of neonatal early-onset sepsis.

Introduction

Neonatal early-onset sepsis (EOS) in the era of perinatal antibiotic treatment has become an infrequent event occurring in 1 to 5 per 1000 live births; nevertheless it carries a high risk of mortality and morbidity. 1 Because of the nonspecific signs, the clinical diagnosis of sepsis in neonates is difficult to determine. Furthermore, blood cultures that are the ultimate tool for its diagnosis have a sensitivity of 50 to 80% in neonatal EOS. 2 There are few adjunctive, nonspecific, diagnostic-screening tests helping the clinician to identify newborns suffering from EOS. Acute phase reactants such as C-reactive protein (CRP) and cytokines, for example, interleukin (IL)-6, IL-8, and tumor necrosis factor alpha (TNF-α), have been shown to correlate with the diagnosis of EOS. 1 However, CRP appears to be less sensitive at onset of sepsis, and cytokines owing to their short half-life are less sensitive after 12 and 24 h of onset of sepsis, thus increasing the risk of false-negative results. 3 Procalcitonin, an early marker of bacterial infection, seems to be a valuable addition to the current diagnostic workup of EOS. However, most studies investigating this early marker of neonatal EOS concluded that procalcitonin is not better than CRP in assessing EOS and that its diagnostic accuracy is complicated by a physiological increase during the first days of life. 4 Several studies, analyzing term infants suspected of sepsis, have recently shown that inflammatory markers such as whole-blood IL-8 and neutrophil CD64 are very sensitive indicators of EOS in a research setting, 5, 6 yet because of special handling requirements, they are not readily available in hospital laboratories.