Meeting Banner
Abstract #4475

Identification of Bacterial Type in Urinary Tract Infection Using 1H NMR Spectroscopy

Ashish Gupta1, Mayank Dwivedi2, Abbas A. Mahdi3, Chunni Lal Khetrapal4, Mahendra Bhandari5

1Centre of Biomedical Magnetic Resonance, Lucknow , UP, India; 2Departments of Microbiology, Lady Hardinge Medical College, New Delhi, India; 3Department of Biochemistry, Chhatrapati Shahuji Maharaj Medical University, Lucknow, UP, India; 4Centre of Biomedical Magnetic Resonance, Lucknow, UP, India; 5Henry Ford Hospital System, Vattikuti Urology Institute, Detroit, MI, United States

Quantitative analysis of 682 urine samples from suspected UTI patients, and 50 healthy volunteers was carried out to identify the differential biomarkers between gram negative bacilli (GNB) (E. coli, P. aeruginosa, K. pneumonia, Enterobacter, Acinetobacter, Pr. mirabilis, Citrobacter frundii) and gram positive cocci (GPC) (Enterococcus faecalis, Streptococcus group B, Staphylococcus saprophyticus) uropathogenic urinary tract infection (UTI) using 1H NMR spectroscopy. Linear multivariate discriminant function analysis (DFA) reveals that 1H NMR measured metabolites can differentiate not only between healthy controls and infected urine samples but also GNB and GPC type of uropathogenic microorganism.

Abstract PDF E-Poster Video

Keywords

able accuracy acetate achieved address among bacilli bacterial band basis biochemical biomedical broad broadening carried category cell centrifuged check chemical chosen chronic classification classifications classified cocci coefficients common comprehensive comprising conditions constructed containing contaminants contamination control controls culture currently debris decanted define delay delays derived determined deuterium diagnosis differ differential differentiate differentiation discriminant dissolved divided domain estimating ethanol exhibited field ford formate forward frequency function gram healthy henry humans identification important includes indicated infected infection infections lactate lambda limitations linear materials measured medical metabolites microorganisms model multivariate nature negative overall oxide part parts patients poor positive prediction prepared pulse quantified quantitative quick rapid recurrent remaining reporting respective respectively response sample samples saturation segregate sensitivity separate served sets severe shortcomings significance similarly software specificity specimens spectra spectrometer spectroscopy spectrum statistical stepwise student subjected suppression suspected system therapeutic tract training transferred treatment type types typical univariate urinary urine variable variables vital volunteers width