Background: Urine dipstick tests are often used to evaluate proteinuria during health checkups. We examined the dipstick's accuracy in assessing the proteinuria levels among Japanese workers.
Methods: We assessed subjects aged ≥ 18 years who had a health checkup at the University of Tokyo in 2016 or 2017 (n = 5383). Proteinuria was stratified by urine protein-to-creatinine ratio (PCR): A1, < 150 mg/gCre; A2, 150-499 mg/gCre; and A3, ≥ 500 mg/gCre. The accuracy of a dipstick result of ± or higher to detect a PCR level of ≥ A2 was examined. We compared changes in dipstick results and PCR level in 136 subjects evaluated twice with a median interval of 119 days.
Results: The subjects' mean age was 40 years, and half were women. The dipstick results were - in 94.9%, ± in 4.1%, and ≥ 1 + in 1.0%. The PCR level was A1, A2, A3 in 98.6%, 1.2%, and 0.2% of the subjects, respectively. The sensitivity, specificity, and positive and negative predictive values of a ± or higher dipstick result to detect A2 or higher were 66.2%, 95.6%, 17.5%, and 99.5%, respectively. Among the 136 subjects examined twice, 134 (98.5%) had no change in PCR level (A1 in all cases) despite a decrease or increase in dipstick results.
Conclusion: Urine dipstick results of ± or above had a high specificity but low sensitivity and positive predictive value to detect PCR proteinuria of A2 or higher. Confirmation by quantitative protein measurement should be recommended for individuals at high risk of chronic kidney disease.
Keywords: Chronic kidney disease; Cost effectiveness; Dipstick; Health checkup; Proteinuria; Screening.