There is good evidence from rigorous studies in UK general practice that urine dipstick tests (for nitrites, leucocytes, and blood) can modestly improve diagnostic precision. However, the predictive values of negative tests were much lower than expected from previous research: even when all dipstick tests are negative, about 24% of women presenting with suspected cystitis will have urinary tract infection (UTI) confirmed by culture.

There is extensive literature on the interpretation of urine dipstick tests, including systematic reviews and meta-analyses [Hurlbut and Littenberg, 1991; Deville et al, 2004; St John et al, 2006]. Estimates of diagnostic performance of urine dipstick tests vary widely in primary studies, and this may in part be due to the variation in clinical conditions and treatment settings. However, a recent set of studies provides good evidence that is directly applicable to current UK general practice (and not out of line with the findings of previous systematic reviews) [Little et al, 2009].

• A series of studies included developing and assessing a clinical decision rule for interpreting the results of urine dipstick tests [Little et al, 2009]. The relevant results from these two studies are summarized below.
• The studies included women who sought treatment by a doctor or nurse in general practices in the south of England and who were considered on clinical grounds to have cystitis (without vaginitis or urethritis).
• The prevalence of UTI (confirmed by urine culture) was about 65% in the two studies.
• The studies reported sensitivity, specificity, positive and negative predictive values, and likelihood ratios for positive and negative tests. These results are summarized with estimates of the proportions of women who are correctly diagnosed with a positive test and incorrectly diagnosed with a negative test. Because these proportions are directly dependent on the prevalence of UTI, when applying them in practice the proportions must be adjusted for the prevalence:
• If the clinical suspicion (or prevalence) is high, the proportions of women who are correctly diagnosed with a positive test and incorrectly diagnosed with a negative test will increase.
• If the clinical suspicion (or prevalence) is low, the proportions of women who are correctly diagnosed with a positive test and incorrectly diagnosed with a negative test will decrease.
• Many of the following statistics were calculated by CKS using data in the published report.

Urine dipstick tests in isolation

• Nitrite dipstick test
• Most urinary pathogens reduce nitrate to nitrite, and a positive test is suggestive of bacteriuria.
• When cystitis was clinically suspected (prevalence of UTI about 65%):
• About 91% of women with a positive nitrite test had UTI confirmed by urine culture.
• About 57% of women with a negative nitrite test had UTI confirmed by urine culture.
• A false-negative test can occur when:
• The pathogen does not produce nitrate reductase.
• Frequent urination (which is common in cystitis) gives the enzyme less time to act.
• Prolonged exposure to air inactivates the test.
• Leucocyte esterase (LE) dipstick test
• Leucocyte esterase is a marker for pyuria (leucocytes in the urine), and a positive LE test indicates pyuria.
• When cystitis was clinically suspected (prevalence of UTI about 65%):
• About 76% of women with a positive LE test (1+ or greater) had UTI confirmed by urine culture.
• About 35% of women with a negative LE test had UTI confirmed by urine culture.
• A false-positive test can occur when leucocytes from another source (such as the urethra or vagina) contaminate the specimen.
• Blood dipstick test
• Blood is sometimes found in the urine when there is a UTI.
• When cystitis was clinically suspected (prevalence of UTI about 65%):
• About 75% of women with a positive test for blood (haemolysed trace or greater) had UTI confirmed by urine culture.
• About 48% of women with a negative test for blood had UTI confirmed by urine culture.
• Protein dipstick test
• Protein is sometimes found in the urine when there is a UTI.
• Dipstick protein tests provided no additional diagnostic information over and above that provided by the nitrite, LE, and blood dipstick tests.

Dipstick tests in combination

• Nitrite positive or both LE and blood dipstick tests positive
• Combining dipstick test results as 'nitrite positive or both LE and blood positive' increases sensitivity, but decreases specificity.
• When cystitis was clinically suspected (prevalence of UTI about 65%):
• About 80% of women with nitrite positive or both LE and blood positive had UTI confirmed by urine culture.
• Nitrite, LE, and blood dipstick tests all negative
• Combining dipstick test results as 'nitrite and LE and blood' decreases sensitivity, but increases specificity.
• When cystitis was clinically suspected (prevalence of UTI about 65%):
• About 24% of women with nitrite, LE, and blood dipstick tests all negative had UTI confirmed by urine culture.