This retrospective study aimed to evaluate the applicability of the selective approach of imaging infants < 6 months old with urinary tract infection (UTI) according to the UTI guidelines of the National Institute for Health and Clinical Excellence (NICE) 2007. Infants < 6 months old with their first UTI from January 2001 to December 2006 having undergone an ultrasound examination of the urinary tract, a micturating cystourethrogram, and a late di-mercaptosuccinic acid (DMSA) scan, were included. Their condition was evaluated against a set of risk features according to the UTI guidelines. Those having any one of these were classified as atypical and those having none as typical. There were 134 infants reviewed, with a typical (98 infants) to atypical (36 infants) ratio of 2.7 to 1. Girls were found to be relatively more represented in the atypical group [male (M):female (F) = 1.3:1] than in the typical group (M:F = 4.4:1) (P < 0.004). There were significantly more infants with abnormal micturating voiding cystourethrograms (MCUGs) (P = 0.007), more refluxing ureters (P < 0.001) and more significant vesico-ureteral reflux (VUR) (>/= grade III) (P = 0.013) in the atypical group than in the typical group; while there was no significant difference in ultrasound (US) and DMSA scan findings between the two groups. In the atypical group there was no difference in imaging studies (and, thus, the results) between the conventional practice and the NICE UTI recommendation. In the typical group, if the recommendations of the guidelines had been followed (i.e. only those with abnormal US would have been further investigated), 25 refluxing ureters and 22 scarred kidneys would have been left undiagnosed. In conclusion, application of the suggested selective imaging approach would leave a significant number of VUR and renal scars undiagnosed, and it may not be an optimal practice for infants less than 6 months old with their first UTI. The best approach remains to be clarified.