Disappointing specificity of the leucocyte-esterase test for the diagnosis of urinary tract infection in general practice

Abstract

Objectives: To determine the sensitivity and specificity of the leucocyte-esterase test (LE test) as a diagnostic tool in general practice. Methods: In 249 women consulting their general practitioners for dysuria, both the LE test (Nephur testR + leuco/Boehringer Mannheim) and a culture (UricultRdipslide) were performed on freshly voided urine. The sensitivity and specificity of the LE test were determined with urine culture as standard. Results: Sensitivity was 96%, specificity only 15.5% (criterion for positivity of the stick: each colour change; positivity for culture: ≥ 100,000 colony-forming units (cfu)/ml or every pure culture of Staphylococcus saprophyticus. By lowering the criterion for culture from ≥ 100,000 cfu/ml to ≥ 10,000 cfu/ml, the specificity increased by only 0.3%. An increase in the specificity to 36% was possible by classifying "lightly positive" dipsticks (10-20 white blood cells/Ml) as negative; the sensitivity then dropped to 87%. Interpretation after one minute also gave a slightly higher specificity and lower sensitivity. Conclusions: In contrast with earlier hospital-based studies, we observed a disappointing specificity of the LE test in daily practice conditions, with the risk of a significant overdiagnosis of urinary tract infection. Earlier diagnostic studies in general practice have shown very diverging results. Our study confirms the most negative ones, questioning the value of the LE test in general practice. © 1998 Informa UK Ltd All rights reserved.