Reagent strip diagnosis of UTI in a high-risk population.

A study was undertaken to determine the accuracy of reagent strip diagnosis of urinary tract infection (UTI) in patients seen in the ED who were subsequently admitted to hospital with this diagnosis. The reagent strip results of 100 patients were recorded and the sensitivity and specificity of various parameters in the diagnosis of UTI were calculated. Our results indicate that even in a group of patients with high clinical probability of UTI and with symptoms severe enough to be hospitalized, reliance on positivity of any single parameter alone gives unacceptably low sensitivity: 89%, 39%, 68%, and 63% for WBC, nitrite, RBC, and protein, respectively. We conclude that dipstick analysis is not sufficiently sensitive for diagnosis of UTI in high risk patients in whom missed diagnosis would have serious consequences.