Article Text

G466(P) Performance characteristics of biomarkers of bacterial infection in children in the paediatric intensive care unit following cardiothoracic surgery
  1. R Guhadasan1,
  2. S D’Souza2,
  3. R Jennings1,
  4. S Paulus1,3,
  5. C Chesters4,
  6. C Downey5,
  7. K Thorburn1,6,
  8. P Baines1,6,
  9. ED Carrol1
  1. Institute of Infection and Global Health, University of Liverpool, Liverpool, UK
  2. School of Medicine, University of Liverpool, Liverpool, UK
  3. Department of Infectious Disease, Alder Hey Children’s NHS Foundation Trust, Liverpool, UK
  4. Department of Pathology, Alder Hey Children’s NHS Foundation Trust, Liverpool, UK
  5. Department of Clinical Haematology, Royal Liverpool and Broadgreen University Hospitals, Liverpool, UK
  6. Paediatric Intensive Care, Alder Hey Children’s NHS Foundation Trust, Liverpool, UK


Aims The Systemic Inflammatory Response Syndrome (SIRS) is common after cardiac surgery and cardiopulmonary bypass (CPB), leading to difficulty in distinguishing between a bacterial infection or SIRS. Early diagnosis of post-operative infection is of fundamental importance in order to improve outcomes and reduce antibiotic over use. This study examined the predictive value of the biomarkers procalcitonin (PCT), lactate, neutrophil gelatinase associated lipocalin (NGAL) and the biphasic APTT waveform (bAPTT).

Methods Consecutive children admitted for cardiac surgery at a tertiary paediatric intensive care unit (PICU) were enrolled in the study. None of our patients had a bacterial infection (BI) at admission. Receiver operating curves (ROC) were created to determine the predictive value of the biomarkers to diagnose BI.

Results In total, 88/368 children (24%) developed BI. Median PCT in children with BI was 0.16 ng/ml (IQR 0.06–0.98) compared with 0.10 ng/ml (IQR 0.–0.36) in the non-infected group (p=0.03). Median lactate in children with BI was 1.42 mmol/L (IQR 0.93–2.37) compared with 1.20 mmol/L (0.93–1.79) in the non-infected group (p=0.03). The other biomarkers were not significantly different between BI and non-BI groups. The median inotrope score in first 12 hours was 10 (IQR 5–15) in those with BI compared with 7 (5–12) in the non-infected group (p=0.002). Prolonged hospital stay (p=0.001) and increased duration of mechanical ventilation (p=0.001) were more common in children with BI. CPB time was increased in children with BI (median 133 min, IQR 75–183) compared with those without BI (median 101 min, IQR 60–147) (p=0.005).Using ROC curves to predict BI, the AUC for PCT was 0.58, lactate 0.46, inotrope score in the first 12 hours 0.60, CPB 0.60, and circulatory arrest time 0.55 respectively.

Conclusion None of the biomarkers studied demonstrated strong predictive value for BI. The increased health utilisation of postoperative infection is highlighted by increased PICU and hospital stay. Children at higher risk of developing post-operative infection can be identified as those with prolonged bypass and crossclamp times, and those with high inotrope scores in the first 12 hours of admission.

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.