The presence of viable but nonculturable bacteria in human clean-catch and mouse bladder-isolated urine specimens was investigated. Viable but nonculturable bacteria are alive but do not give rise to visible growth under nonselective growth conditions. Urine specimens obtained from human female volunteers with or without an active urinary tract infection were found to contain, on average, significantly more viable than culturable forms of bacteria. Additional support for the presence of viable but nonculturable cells in urine specimens considered sterile was obtained from examination of urine specimens obtained directly from the bladder of healthy mice. Because the viability assay used to study the viable but nonculturable condition is by necessity growth independent, and hence indirect, the accuracy of this assay that scores cells with intact cell membranes as being viable was studied. Greater than 95% of Escherichia coli cells exposed to lethal doses of UV irradiation were found to lose their membrane integrity within a day, a time frame similar to that used to examine urine specimens. These data suggest that viable but nonculturable cells can occur within regions of the urinary tract previously considered sterile.Urinary tract infections (UTIs) affect as many as 50% of women at least once during their lifetime (29, 32), and 25% of those who acquire a UTI will have another infection within the following 6 months (17). A UTI occurs when the urinary tract is infected with microorganisms, and uropathogenic Escherichia coli accounts for greater than 80% of all UTI cases (4, 30). One method of diagnosing a UTI is by culturing urine specimens; a threshold of 100,000 CFU/ml in clean-catch urine specimens is considered to indicate a UTI (4, 28). This threshold is not an absolute indicator, as both asymptomatic bacteriuria and patients with UTI symptoms having no culturable urine bacteria occur (29,32).Urine within the urinary tract is generally considered sterile (14). This conclusion is based upon a lack of culturable cells present in urine specimens obtained via clean-catch and catheterization methods. The presence of viable bacteria in the urine specimens of healthy patients would impact on hypotheses to explain recurrent UTIs as well as diagnostic procedures. Most recurrent UTIs result from reinfection; however, a higher percentage than would be expected by chance are caused by the index strain (6,18,26,27,41). The physical location and physiological status of index strain cells that remain after successful antibiotic therapy are unknown. Observations with a mouse model indicate that uropathogenic E. coli cells can remain in the urinary tract following antibiotic ther-