Urine Flow Cytometry Parameter Cannot Safely Predict Contamination of Urine-A Cohort Study of a Swiss Emergency Department Using Machine Learning Techniques.
UF-4000
automated urine sediment analyser
culture growth
flow cytometry
mixed urine culture
prediction
squamous epithelial cell
urinary tract infection
urine analysis
Journal
Diagnostics (Basel, Switzerland)
ISSN: 2075-4418
Titre abrégé: Diagnostics (Basel)
Pays: Switzerland
ID NLM: 101658402
Informations de publication
Date de publication:
16 Apr 2022
16 Apr 2022
Historique:
received:
21
03
2022
revised:
10
04
2022
accepted:
13
04
2022
entrez:
23
4
2022
pubmed:
24
4
2022
medline:
24
4
2022
Statut:
epublish
Résumé
Urine flow cytometry (UFC) analyses urine samples and determines parameter counts. We aimed to predict different types of urine culture growth, including mixed growth indicating urine culture contamination. A retrospective cohort study (07/2017-09/2020) was performed on pairs of urine samples and urine cultures obtained from adult emergency department patients. The dataset was split into a training (75%) and validation set (25%). Statistical analysis was performed using a machine learning approach with extreme gradient boosting to predict urine culture growth types (i.e., negative, positive, and mixed) using UFC parameters obtained by UF-4000, sex, and age. In total, 3835 urine samples were included. Detection of squamous epithelial cells, bacteria, and leukocytes by UFC were associated with the different types of culture growth. We achieved a prediction accuracy of 80% in the three-class approach. Of the Significant bacterial growth can be safely ruled out using UFC parameters. However, positive urine culture growth (rule in) or even mixed culture growth (suggesting contamination) cannot be adequately predicted using UFC parameters alone. Squamous epithelial cells are associated with mixed culture growth.
Sections du résumé
BACKGROUND
BACKGROUND
Urine flow cytometry (UFC) analyses urine samples and determines parameter counts. We aimed to predict different types of urine culture growth, including mixed growth indicating urine culture contamination.
METHODS
METHODS
A retrospective cohort study (07/2017-09/2020) was performed on pairs of urine samples and urine cultures obtained from adult emergency department patients. The dataset was split into a training (75%) and validation set (25%). Statistical analysis was performed using a machine learning approach with extreme gradient boosting to predict urine culture growth types (i.e., negative, positive, and mixed) using UFC parameters obtained by UF-4000, sex, and age.
RESULTS
RESULTS
In total, 3835 urine samples were included. Detection of squamous epithelial cells, bacteria, and leukocytes by UFC were associated with the different types of culture growth. We achieved a prediction accuracy of 80% in the three-class approach. Of the
CONCLUSIONS
CONCLUSIONS
Significant bacterial growth can be safely ruled out using UFC parameters. However, positive urine culture growth (rule in) or even mixed culture growth (suggesting contamination) cannot be adequately predicted using UFC parameters alone. Squamous epithelial cells are associated with mixed culture growth.
Identifiants
pubmed: 35454055
pii: diagnostics12041008
doi: 10.3390/diagnostics12041008
pmc: PMC9025120
pii:
doi:
Types de publication
Journal Article
Langues
eng
Références
Arch Pathol Lab Med. 2008 Jun;132(6):913-7
pubmed: 18517272
Nat Rev Microbiol. 2015 May;13(5):269-84
pubmed: 25853778
Ther Adv Urol. 2019 May 02;11:1756287219832172
pubmed: 31105774
J Natl Med Assoc. 1986 Jan;78(1):43-8
pubmed: 3512846
J Clin Lab Anal. 2017 Sep;31(5):
pubmed: 27859671
Clin Chim Acta. 2000 Jul;297(1-2):305-11
pubmed: 10841931
Clin Chem Lab Med. 2010 Feb;48(2):289-92
pubmed: 19961394
BMC Infect Dis. 2014 Jan 09;14:13
pubmed: 24405683
Diagn Microbiol Infect Dis. 2009 Oct;65(2):103-7
pubmed: 19748419
Curr Opin Infect Dis. 2016 Feb;29(1):73-9
pubmed: 26694621
Sci Transl Med. 2013 Jun 19;5(190):190ra80
pubmed: 23785036
J Emerg Nurs. 2019 Sep;45(5):488-501
pubmed: 31445626
Clin Infect Dis. 2019 May 2;68(10):e83-e110
pubmed: 30895288
N Engl J Med. 1982 Aug 19;307(8):463-8
pubmed: 7099208
Ann Epidemiol. 2000 Nov;10(8):509-15
pubmed: 11118930
Acad Emerg Med. 2016 Mar;23(3):323-30
pubmed: 26782662
J Infect Dis. 1982 Dec;146(6):719-23
pubmed: 6815281
Aging health. 2013 Oct;9(5):
pubmed: 24391677
J Antimicrob Chemother. 2014 Jan;69(1):234-40
pubmed: 23887867
Clin Chim Acta. 2013 Sep 23;424:90-5
pubmed: 23721948
Am J Clin Pathol. 1991 Nov;96(5):582-8
pubmed: 1951183
J Emerg Med. 2015 Jun;48(6):706-11
pubmed: 25841289
Clin Chim Acta. 2010 Aug 5;411(15-16):1137-42
pubmed: 20359474
Am J Clin Pathol. 2010 Apr;133(4):577-82
pubmed: 20231611
Dtsch Arztebl Int. 2010 May;107(21):361-7
pubmed: 20539810
Pathogens. 2016 Nov 30;5(4):
pubmed: 27916925
Infect Dis Clin North Am. 2014 Mar;28(1):1-13
pubmed: 24484571
Am J Obstet Gynecol. 2018 Jul;219(1):40-51
pubmed: 29305250
Infect Dis Clin North Am. 2003 Jun;17(2):303-32
pubmed: 12848472
Dis Markers. 2019 Mar 03;2019:5853486
pubmed: 30944667
Open Forum Infect Dis. 2016 Aug 02;3(3):ofw159
pubmed: 27704014
Scand J Clin Lab Invest. 2021 Sep;81(5):379-384
pubmed: 34237238
Dis Mon. 2003 Feb;49(2):53-70
pubmed: 12601337
Pathology. 2003 Apr;35(2):161-5
pubmed: 12745465
Ann Emerg Med. 1998 Apr;31(4):455-8
pubmed: 9546013
Urol Clin North Am. 2008 Feb;35(1):1-12, v
pubmed: 18061019
Sex Transm Infect. 1998 Feb;74(1):11-9
pubmed: 9634294
Microbiol Spectr. 2016 Oct;4(5):
pubmed: 27780014
PLoS One. 2018 Feb 23;13(2):e0193255
pubmed: 29474463
JAMA. 2002 May 22-29;287(20):2701-10
pubmed: 12020306
BJU Int. 2022 Jun;129(6):668-678
pubmed: 34741796
Am J Clin Pathol. 2007 Dec;128(6):922-5
pubmed: 18024316
PLoS One. 2021 Jul 6;16(7):e0254064
pubmed: 34228764
Immunobiology. 2021 Jan;226(1):152020
pubmed: 33246308
Dtsch Arztebl Int. 2011 Jun;108(24):415-23
pubmed: 21776311