Point-of-Care Serum Amyloid A as a Diagnostic Marker for Neonatal Sepsis.

Abstract

To evaluate diagnostic accuracy of point-of-care Serum Amyloid A (POC-SAA) and its comparison with procalcitonin for diagnosis of neonatal sepsis. The present diagnostic accuracy study consecutively recruited neonates with suspected sepsis. Blood samples for sepsis screen, culture, high sensitivity C-reactive protein (CRP) (hs-CRP, as a part of sepsis screen), procalcitonin and POC-SAA were collected before starting antibiotics. The optimum cut-off level of biomarkers (POC-SAA and procalcitonin) was determined by receiver-operating-characteristics curve (ROC) analysis. Sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of POC-SAA and procalcitonin were derived for 'clinical sepsis (neonates with suspected sepsis and either positive sepsis screen and/or blood culture)' and 'culture positive sepsis' (neonates with suspected sepsis and positive blood culture). Seventy-four neonates with mean±SD gestational age of 32.8±3.7 wk were evaluated for suspected sepsis, of which the proportion of 'clinical sepsis' and 'culture positive sepsis' was 37.8% had 16.2%, respectively. At a cut-off of 25.4mg/L, POC-SAA had sensitivity, specificity, PPV and NPV of 53.6%, 80.4%, 62.5% and 74.0%, respectively for diagnosis of clinical sepsis. The sensitivity, specificity, PPV and NPV of POC-SAA for detection of culture positive sepsis were 83.3%, 61.3%, 29.4% and 95.0%, respectively at a cut-off of 10.3mg/L. There was no significant difference in the diagnostic accuracy of biomarkers for detection of culture positive sepsis (area under the curve, AUC of POC-SAA vs. procalcitonin vs. hs-CRP: 0.72 vs. 0.85 vs. 0.85; p = 0.21). POC-SAA is comparable to procalcitonin and hs-CRP for diagnosis of neonatal sepsis.