Catheter-associated urinary tract infection is the most common nosocomial infection in clinical settings. For bacteria ascending to the bladder the most common route is the extraluminal, but the intraluminal route also plays a role. For this reason, we compared two urinary closed system bags (CSB), one with a double and the other with a single non-return valve (NRV), in a laboratory setting in order to establish their ability to prevent or delay the ascent of bacteria from the drainage bag to the bladder. The tests were performed in two microbiological laboratories (Copenhagen (C), Denmark and Freiburg (F), Germany). These were blinded to each other. A urinary tract model using artificial urine was set up. Two sets of ten drainage bags each with a double NRV (CSB A), and two sets of ten drainage bags each with a single NRV (CSB B) were inoculated with Escherichia coli (F: ATCC 25922; C: clinical strain). Daily samples were taken from two drainage ports on each system - one above the NRV (Port I), the other above the top of the artificial bladder (Port II). Time till E. coli was detected at the drainage ports (time to positivity) was measured. Colonization of the 'bladder' was defined as time to positivity at Port II. No significant differences in time to positivity at Port I (median 9.0, range: 6-12 for CSB B vs median 9.5 days, range: 6-13 for CSB A) were observed between the two systems. However, substantial differences were seen between the two systems in time to positivity at Port II: Port II on the bladder model using CSB B became positive after a median of 14.0 days (range: 10-22), whereas Port II of the model using CSB A only became positive after 21.5 days (range: 13-24). This amounts to a highly significant difference of 7.5 days (p = 0.0001) in the mean. Under laboratory conditions, colonization of the 'bladder' was significantly delayed when the CSB with a double NRV was used in comparison to the results obtained from the single NRV-system. Clinical trials should be conducted to investigate whether the urinary CSB with the double NRV has the ability to prevent (or to delay the onset of) catheter-associated urinary tract infection.