Résumé
BACKGROUND: The urine dipstick is widely used as an initial screening tool for the evaluation of proteinuria; however, its diagnostic accuracy has not yet been sufficiently evaluated. Therefore, we evaluated its diagnostic accuracy using spot urine albumin/creatinine ratio (ACR) and total protein/creatinine ratio (PCR) in proteinuria. METHODS: Using PCR > or = 0.2g/g or > or = 0.5g/g and ACR > or = 300mg/g or > or = 30mg/g as the reference standard, we calculated the diagnostic accuracy profile: sensitivity, specificity, positive and negative predictive value, and the area under the curve (AUC) of the receiver operating characteristic curve. RESULTS: PCR and ACR were available for 10,348 and 3,873 instances of dipstick testing. The proportions with PCR > or = 0.2g/g, > or = 0.5g/g and ACR > or = 300mg/g, > or = 30mg/g were 38.2%, 24.6% and 8.9%, 31.7%, respectively. The AUCs for PCR > or = 0.2g/g, > or = 0.5g/g, and ACR > or = 300mg/g were 0.935 (trace: closest to ideal point), 0.968 (1+), and 0.983 (1+), respectively. Both sensitivity and specificity were > 80% except for PCR > or = 0.5g/g with trace cutoff. For the reference standard of ACR > or = 30mg/g, the AUC was 0.797 (trace) and the sensitivity was 63.5%. CONCLUSION: Urine dipstick test can be used for screening in older outpatients with ACR > or = 300mg/g or PCR as the reference standard for proteinuria. However, we cannot recommend the test as a screening tool with ACR > or = 30mg/g as the reference owing to its low sensitivity.