The Urinary Tract Infection Consensus Conference brought together researchers, clinicians, and consumers to arrive at consensus on the best practices for preventing and treating urinary tract infections in people with spinal cord injuries; the risk factor and diagnostic studies that should be done; indications for antibiotic use; appropriate follow-up management; and needed future research. Urinary tract infection (UTI) was defined as bacteriuria (102 bacteria/ml of urine) with tissue invasion and resultant tissue response with signs and/or symptoms. Asymptomatic bacteriuria represents colonization of the urinary tract without symptoms or signs. Risk factors include: over-distention of bladder, vesicoureteral reflux, high pressure voiding, large post-void residuals, presence of stones in urinary tract, and outlet obstruction. Possible physiologic/structural, behavioral, and demographic risk factors were identified also, Indwelling catheterization, including suprapubic, and urinary diversion and the drainage methods most likely to lead to persistent bacteriuria. Infection risk is reduced with intermittent catheterization, but more severely disabled people who require catheterization by others are at greater risk for UTIs. Clean self-intermittent catheterization does not pose a greater risk of infection than sterile self-intermittent catheterization and is much more economic. However, care must be given to proper cleansing of reusable catheters. Quantitative urine-culture criteria for the diagnosis of bacteriuria include: catheter specimens from individuals on intermittent catheterization > or = 10(2) cfu/ml; clean-void specimens from catheter-free males using condom collection devices > or = 10(4) cfu/ml; and specimens from indwelling catheters of any detectable concentration. Dip stick screening tests may offer promise as an early warning system of UTI since they can be self-administered. Symptomatic UTI should be treated with antibiotics for 7 to 14 days. Longer courses have not been beneficial. In patients with symptomatic UTIs, it is not necessary to wait for the results of cultures before starting treatment. Asymptomatic bacteriuria need not be treated with antibiotics. There is little evidence presently to support the use of antibiotics to prevent infections. Following a recent episode of febrile UTI, possible contributing prior events should be reviewed. The upper tracts should be evaluated (imaging studies) to identify possible abnormalities. A common concern among people with spinal cord injuries is that physicians will alter bladder management programs without regard to lifestyle needs. Social/vocational flexibility may be more important to them than a state-of-the-art bladder management program. Future research should focus on obtaining more representative samples and investigate psycho-social-vocational implications as well as additional clinical-medical factors.