Secondary Logo

Journal Logo

ABSTRACT 950

BLOOD STREAM INFECTIONS IN PAEDIATRIC CRITICAL CARE

GETTING THE DIAGNOSIS RIGHT – A QUALITY IMPROVEMENT PROJECT

McCluskey, J.1; Cassidy, J.1; Martin, J.1; Gray, J.2

Pediatric Critical Care Medicine: May 2014 - Volume 15 - Issue 4_suppl - p 210
doi: 10.1097/01.pcc.0000449676.01625.fe
Abstracts of the 7th World Congress on Pediatric Critical Care
Free

1Department of Paediatric Intensive Care, Birmingham Children's Hospital, Birmingham, United Kingdom 2Department of Microbiology, Birmingham Children's Hospital, Birmingham, United Kingdom

Background and aims: High numbers of blood cultures (BC) are taken in PICU with low positive yields.

Aims: We aimed to identify indications for cultures, rates of true positive bloodstream infections (BSI) and quality improvement areas.

Methods: Analysis of all PICU BC results to identify true BSIs over 10 months from January-October 2013. A 3 month audit analysed culture indications, physiological parameters and timing in relation to antimicrobial therapy. BC sampling guidelines were introduced in September 2013. Institutional consent requirement was waivered.

Results: Of 1123 PICU admissions over 10 months, 348 (31%) had BCs taken. 34 patients had proven BSI. 53% were community-acquired and 47% healthcare-acquired infections (HCAI). Half of HCAIs resulted from Gram negative infections and half from either methicillin sensitive Staphylococcus aureus or coagulase negative staphylococcus (CNS). Central venous catheter-related BSIs were the commonest HCAI (64% of HCAIs).

40 patients had a false positive culture, most commonly CNS (>95%).

116 patients who were cultured were audited. The commonest indication was pyrexia followed by sepsis/systemic inflammatory response.

The majority of patients had ≤2 sites cultured. 12% of cultures had been repeated within 24 hours. 72% of cultures were taken when already on antibiotics.

Following guideline introduction, BC sampling rates fell by 25%, primarily with reductions in multi-site sampling and repeat cultures within 24 hours. Proportion of positive BCs due to contaminants remains 50%.

Conclusions: Formalised BC sampling guidelines can reduce duplicate sampling with associated cost improvements and no evidence of delayed infection diagnosis. Ongoing high contamination rates remain a problem with potential inappropriate antibiotic administration.

©2014The Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies