ESPN 50th Annual Meeting

ESPN 2017

Are white cells in children’s urine diagnostic or a distraction, and does collection and culture method matter? New data from 4910 acutely unwell children
SHOUJA M A ALAM 1 JUDITH H VAN DER VOORT 1 Mandy Wootton 2 Ben Creswell 1 Alastair D Hay 3 Chris C Butler 4

2- Specialist Antimicrobial Chemotherapy Unit, University Hospital of Wales
3- Centre for Academic Primary Care, NIHR School for Primary Care Research, University of Bristol
4- Nuffield Department of Primary Care Health Sciences, University of Oxford


To describe the relationship between urine white cell count (WCC) and the microbiological diagnosis of UTI in acutely unwell children, and explore the influence of different methods of urine collection and culture. 

Material and methods:


Prospective, diagnostic cohort study of acutely unwell children aged 3months-5years presenting to UK primary care whose urine samples were collected by clean catch or nappy pad, and cultured by both a Central Laboratory (CL) using spiral-plating, and Local NHS Laboratories (LL) using standard practice. Positive and negative predictive values (PPV and NPV) were calculated for WCC≥100/mm3 and WCC≤10/mm3 respectively, comparing laboratories and sampling methods. 



Of 4910 samples, 2630 (53.6%) were collected by clean catch and 2198 (44.8%) by nappy pad. 1.9% had UTI according to the CL and 5.3% according to the LL. 194 samples (4.0%) were positive in the LL but negative in the CL. CL samples were cultured on average 34.1hrs later. 


The table outlines PPV of WCC≥100/mm3 and NPV of WCC≤10/mm3 by laboratory culture and sampling method. 37 of the 92 positive CL cultures (40.2%) had WCC≤10/mm3.


CL culture LL culture
All Samples Clean Catch Nappy Pad All Samples Clean Catch Nappy Pad
PPV of WCC≥100/mm3 23.9% [17.5-31.6%] 33.3% [25.0-42.8%] 0

30.3% [22.7-39.1%] 38.7% [29.3-49.0%] 6.9% [1.8-23.6%]
NPV of WCC10/mm3 99.0% [98.7-99.2%] 99.3% [98.9-99.5%] 98.6% [98.3-98.9%] 95.8% [95.4-96.2%] 97.7% [97.2-98.2%] 93.6% [92.9-94.2%]




A WCC≥100/mm3 poorly predicts UTI on culture in acutely unwell young children consulting in the community. However, a WCC≤10/mm3 has good NPV regardless of collection method (though worse for LL culture, especially nappy pad samples). To discard urines with WCC≤10/mm3 before culture would miss 40.2% of CL diagnosed UTIs. Alternatively, LL culture of samples with WCC≤10/mm3 overcalls UTI by identifying 150 (3.1%) positive, which if later re-cultured by the CL’s reference standard technique would reduce to 22 (0.4%), potentially minimising unnecessary treatment or investigation.