Stepwise strategies in analysing haematuria and leukocyturia in screening

The aim of the present work was to compare in a supposed healthy population of 680 subjects several algorithms for positive selection of urine samples requiring microscopic examination for erythrocytes and leukocytes after screening by automated test-strip measurement and particle counting on a Sysmex UF-50 flow cytometer. Four strategies have been formulated and the sensitivity, specificity, positive predictive value, negative predictive value, false positive rate, false negative rate, and microscopic review rate were measured. The strategy combining test strip analysis and automated counting on all samples, followed by microscopic examination of only discordant samples gave the best results. When the two methods of haematuria screening were in agreement (91% of samples), the false negative rate for microscopy was 1.1%, with a false positive rate of 0.8%, sensitivity of 66% and specificity of 99%, and the results are acceptable without any other examination. When the two methods of haematuria screening were discrepant, visual microscopic analysis was necessary to obtain definitive results. For leukocyturia screening, 80% of results were in agreement by test strip and automatic sediment urinalysis, with only ten results considered as false negatives (1.8%) and four as false positives (0.7%). Agreement was good and the other criteria were good (sensitivity 79%, specificity 99%). On conflicting samples, there was no agreement between methods and microscopic analysis was essential. The benefit of such an algorithm would be optimisation of the workflow without any loss of sensitivity and specificity at the expense of a two-fold increase in cost.