Article Text

Download PDFPDF
#253 Predictive value of urine WBC counts in clean-catch samples for infants aged 0–90 days
  1. Rasha D Sawaya1,
  2. Fatimah Binabdi1,
  3. Marahaini Isa1,
  4. Karl Kavanagh1,
  5. Robert Cunney2,
  6. Conor Hensey3,
  7. Hani Tamim4,5
  1. 1Department of Emergency Medicine, Children’s Health Ireland, Temple St, Dublin, Ireland
  2. 2Department of Microbiology, Children’s Health Ireland, Temple St, Dublin, Ireland
  3. 3Department of General Paediatrics, Children’s Health Ireland, Temple St, Dublin, Ireland
  4. 4Clinical Research Institute, American University of Beirut Medical Center, Beirut, Lebanon
  5. 5College of Medicine, Alfaisal University, Riyadh, Saudi Arabia

Abstract

Introduction Urinary tract infections (UTI) are a serious bacterial infection in infants presenting to emergency departments (ED). Current guidelines have not identified a White Blood Cell (WBC) cut off in urine microscopy obtained by mid-stream/clean catch to determine a UTI in infants 0-90 days of age.

Objectives Our primary objective was to determine the best WBC cut off from clean-catch urine samples in infants, presenting to the ED. Our secondary objectives were to test the diagnostic accuracy of the WBC count cut off from the literature (10/hpf from suprapubic or catheterized samples) and to assess contamination rates in clean-catch urine samples.

Methods We did a retrospective chart review of 320 randomly selected infants, 0-90 days old, who had a urine culture done between 01//01/2021 and 12/31/2022. Our primary outcome was a UTI, defined as a urine culture with a single or predominant growth of one organism ≥ 10,000 cfu with presence of WBC in microscopy or ≥ 50,000 cfu irrespective of the urine dipstick or microscopy results. We defined a contaminant as any urine culture that was considered negative but had a WBC count > 10 hpf on microscopy.

Results We analysed a total of 306 patients aged 0-90 days. 48.4% were female, 17.3% had a UTI and 29.7% were <29 days. Predictive value of WBC: we determined that a Youden score of 0.52 (our highest) was associated with a WBC count of 10.5 i.e. a WBC cut off >10/hpf had the best compromise of sensitivity and specificity in predicting a UTI. The Area under the curve was 0.83 (See figure 1). Using previously published WBC count cutoff of 10/hpf to predict a UTI, the sensitivity in our sample was 69.8%, 95% CI [55.7%-81.7%], and specificity was 82.2%, 95% CI [76.9%-86.7%]. 14.7% were false positive i.e. contaminants and 5.2% were false negative i.e. missed UTIs.

Conclusions Our data show that using the same WBC cut off (10/hpf) in urine microscopy from clean catch as from catheterized or suprapubic urine samples is most accurate. However, the predictive value is still weak. Studying this in the subgroup of febrile infants is a necessary next step.

Abstract #253 Figure 1

Receiver Operator Curve (ROC) for WBC

Statistics from Altmetric.com

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.