The aim of this study was to evaluate a commercial tube prepared with boric acid, sodium formate and sorbitol (Hemogard Vacutainer tubes, Becton-Dickinson, HG tubes). Fourteen bacterial strains were incubated in urine in HG tubes and conventional tubes. During a 24-h period, most of the microorganisms grew readily in conventional tubes at room temperature, whereas the bacterial counts were comparatively unchanged when chilled or kept in HG tubes. The bacterial counts of Alcaligenes faecalis and lactobacilli decreased by two 10 logarithms in the HG tubes, at room temperature. Experiments were also performed with the addition of various antibiotics. Ciprofloxacin caused a decrease in E. coli counts regardless of type of tube used, while the HG tubes and the conventional tubes kept chilled conserved bacterial counts upon challenge with fosfomycin, trimethoprim or mecillinam. Bedside cultures from 154 outpatients were sampled and divided into three tubes. One conventional tube was sent to our laboratory by ordinary chilled transport. Another conventional tube and one HG tube were transported to the laboratory without chilling. Cultures were performed upon arrival at the laboratory and then 24, 48 and 72 h after primary sampling. Within 24 h of sampling, no significant differences in bacterial counts were observed between chilled conventional tubes and the HG tubes at room temperature. However, in the HG tubes a significant change in enterococcal counts was noted within 48 h.