Urinary tract infection (UTI) is among the most common bacterial infections worldwide. The understanding of the physiological mechanisms affected by UTI may need modern integrative '-omics' technologies, and metabolomics in particular. Here we present the first GC-APCI-MS-based explorative metabolomics study of UTI, using MS and FID detectors simultaneously. This provides high quality mass spectral data as well as semi-quantitative information demonstrating the feasibility of the GC-APCI-MS platform for non-targeted approaches. The work is part of a bigger project aiming at providing a comprehensive overview of UTI-induced changes in urine. Taking advantage of a fully clinically characterized cohort that offers the possibility of both case-control and longitudinal modelling, we can define UTI-induced change as a list of urinary metabolites which distinguish E. coli UTI patients from the subjects with no signs of an active infection. The list of molecular descriptors includes compounds related to bacterial activity such as lactic acid and lactose while other molecules show an association with the physiological status (inositol, citric acid).