Using a urine dipstick test instead of urinalysis to detect urinary tract infection (UTI) can decrease the time needed to make a diagnosis; however, the diagnostic accuracy of urine dipsticks varies, depending on the standards used. Many experts have recommended urinalysis to confirm urine dipstick test results. Lammers and associates compared the test characteristics of urine dipstick with those of urinalysis at various test cutoff points in women with symptoms of UTI.
In a prospective, observational study, urine was obtained from 331 women using either the midstream clean-catch technique or urethral catheterization. The urine was analyzed by dipstick testing, urinalysis, and urine culture. The three factors evaluated by dipstick test included a relatively quantitative leukocyte esterase analysis, the presence of nitrates, and the presence of blood. The urinalysis included an automated dipstick reading along with an evaluation of the urine sediment. Urine culture was performed using routine techniques and was considered positive for UTI if 100,000 or more colonies of one or two species of bacteria per mL of urine were found. “Undertreatment” was defined as withholding treatment on the basis of a negative test when the culture was positive, and “overtreatment” was defined as providing treatment on the basis of a positive test when the culture was negative.
Forty-six percent of the women had positive urine cultures, with the most common organism being Escherichia coli. When the presence of leukocyte esterase, nitrites, or more than a trace of blood was considered indicative of a positive test result, the over-treatment rate was 47 percent, and the under-treatment rate was 13 percent. Using the criteria of more than three white blood cells per high-power field or more than five red blood cells per high-power field as indicators of a positive result, the overtreatment and undertreatment rates were 44 percent and 11 percent, respectively. Defining a positive culture as more than 10,000 colonies per mL lowered the overtreatment rate and raised the undertreatment rate at each test cutoff point but did not change the test characteristics for dipstick testing or urinalysis.
The authors conclude that urine dipstick testing is as accurate as urinalysis in diagnosing UTI in symptomatic women. Confirming a dipstick test with a urinalysis appears to be useless. If stricter criteria for urine dipstick testing, such as a leukocyte esterase level greater than 2 and positive for nitrites are used, the overtreatment rate declines, but the undertreatment rate increases substantially. In making treatment decisions, the limitations of both tests must be considered.