Nicola Culeddu BMBCh, Abstract Co-Author: Nothing to Disclose

Vincenzo Antonio Migaleddu MD, Presenter: Nothing to Disclose

Matilde Chessa BMBCh, Abstract Co-Author: Nothing to Disclose

Pietro Fresu MD, Abstract Co-Author: Nothing to Disclose

Gian Franco Meloni MD, Abstract Co-Author: Nothing to Disclose

Giuseppe Virgilio MD, Abstract Co-Author: Nothing to Disclose

H-MRI spectroscopy of biofluids provides a wealth information on the metabolic processes in human body. Spectra are very complex and in order to focus on significant difference between a set of spectra from control and from humans with disease chemometric methods are used. NMR combined with pattern recognition has been shown to be useful in diagnosis of significant health problem like coronary heart disease, Alzheimer and Parkinson disease. Diabetes type 1( Insulin dependent ) is endemic in Sardinian people, with an incidence five greater of rest of Europe .The aim of our study is to characterize metabolic profile of urine from children affected and from controls.

NMR spectrum are obtained with 600 MHz spectrometer Bruker (Bruker BioSpin GmbH Rheinstetten, Germany. PCA (SPSS Inc. Headquarters, Chicago, Illinois) was performed on a H1 spectra dataset obtained after bucketing integration using Amix (Bruker)

37 samples of urine from healthy and diabetic childrens are anaizued: PCA doesn’t lead significative results, despite great reduction of data (from 140 variables to 5 functions) LDA leads a goood separation with only one discriminant function to well separate datasets >90% of recognizing data.

Consistent differences in the urinary metabolite profiles of children with diabete Type I were identified with a NMR-based metabonomic strategy. Furthermore, a SPSS model based on the 1H NMR-spectral profiles allowed characterization and prediction of healty and diabetic childrens. These findings highlight the potential of metabonomics as a novel approach for fundamental investigations in metabolic diseases studies.

Metabolite-targeted MR imaging technique should be useful in early stage diagnosis.

Culeddu, N, Migaleddu, V, Chessa, M, Fresu, P, Meloni, G, Virgilio, G, H-MRI Spectroscopy: Urine Profile of Children with Diabetes Type 1.  Radiological Society of North America 2007 Scientific Assembly and Annual Meeting, November 25 - November 30, 2007 ,Chicago IL. http://archive.rsna.org/2007/5013487.html