Impact of laboratory-reported urine culture colony counts on the diagnosis and treatment of urinary tract infection for hospitalized patients.

Reducing health care-associated urinary tract infection (UTI) is a National Patient Safety Goal. The purpose of this investigation was to establish a colony count threshold to predict clinically significant UTIs that develop in hospitalized patients. A total of 185 cases were reviewed sequentially by 2 physicians. The information extracted included subjective complaints, presence of an indwelling urinary catheter, clinical signs and symptoms, WBC count, urinalysis, and urine culture results. The first reviewer recorded whether the patient was diagnosed and treated for a UTI by the clinician. The second reviewer determined if the patient met National Healthcare Safety Network guidelines for nosocomial UTI. Compared with patients with colony counts less than 100,000 colony-forming units per milliliter (CFU/mL), patients with colony counts 100,000 CFU/mL or more were 73.86 times more likely to have a clinically significant UTI (odds ratio, 73.86; 95% confidence interval, 24.23 ∼ 225.15; P < .0001; c-statistic, 0.859). Reporting positive results only for patients with 100,000 CFU/mL or more would have reduced the number of positive cultures by 38%. These data suggest that reporting colony counts less than 100,000 CFU/mL encourages treatment of non-clinically significant UTIs in hospitalized patients, causing inappropriate antibiotic use.