Evaluation of pooled human urine and synthetic alternatives in a dynamic bladder infection in vitro model simulating oral fosfomycin therapy.

The impact of the bladder environment on fosfomycin activity and treatment response is uncertain. Standard laboratory media does not reflect the biomatrix of urine, where limited nutritional factors are important for growth and antimicrobial kill rates. We compared fosfomycin activity against Enterobacteriaceae in laboratory media, human urine and synthetic alternatives. Sixteen clinical isolates (8-Escherichia coli, 4-Enterobacter cloacae, 4-Klebsiella pneumoniae) were studied with broth microdilution (BMD) susceptibility, static time-kill assays and dynamic testing in a bladder infection model simulating a 3 g oral fosfomycin dose. Mueller-Hinton broth (MHB) with and without 25 mg/L glucose-6-phosphate (G6P), pooled midstream urine (MSU), pooled 24 h urine collection (24 U), artificial urine medium (AUM) and synthetic human urine (SHU) were compared. BMD susceptibility, bacterial growth and response to static fosfomycin concentrations in urine were best matched with SHU and were distinctly different when tested in MHB with G6P. Fosfomycin exposure in the bladder infection model was accurately reproduced (bias 4.7 ± 6.2%). Under all media conditions, 8 isolates (2-E. coli, 2-E. cloacae, 4-K. pneumoniae) re-grew and 4 isolates (4-E. coli) were killed. The remaining isolates (2-E. coli, 2-E. cloacae) re-grew variably in urine and synthetic media. Agar dilution MIC failed to predict re-growth, whereas BMD MIC in media without G6P performed better. Emergence of resistance was restricted in synthetic media. Overall, SHU provided the best substitute for urine for in vitro modelling of antimicrobial treatment of uropathogens, and these data have broader utility for improved preclinical testing of antimicrobials for urinary tract infections.