Background: Commercial testing for microalbuminuria in human urine is often performed with point-of-care semiquantitative test strips followed by quantitative testing when indicated. An ELISA that quantifies canine urine albumin concentration has been developed, but semiquantitative test strips for use in the dog are not available.
Objective: The purpose of this study was to prospectively determine the concordance of canine urine albumin concentrations measured by a commercial human test strip and by ELISA.
Methods: Urine samples were obtained from 67 dogs evaluated for a variety of clinical conditions. Dipstick urinalyses were performed on all samples; clinician discretion determined method of urine collection and performance of urine sediment examination and/or urine culture. Urine albumin concentration was determined using test strips (Clinitek Microalbumin, Bayer Corporation, Elkhart, Ind, USA), and results were compared with those obtained by ELISA.
Results: The Clinitek strips correctly determined albumin concentration in 42 of 67 (63%) urine samples tested. Concordance was lowest (48%) for dogs with microalbuminuria (10-300 microg/mL by ELISA). Clinitek strip sensitivity and specificity for correct identification of microalbuminuria were 48% and 75%, respectively. Concordance was lower in dogs with urinary tract infection or hematuria and in samples collected by catheterization. Sensitivity and specificity for correct identification of microalbuminuria after exclusion of dogs with urinary tract infection or hematuria were 59% and 83%, respectively.
Conclusions: These results suggest that the Clinitek strips lack sufficient concordance with results obtained by ELISA to be a reliable screening for test microalbuminuria in the dog. A reliable semiquantitative point-of-care test for canine urine albumin concentrations below those detected by standard urine dipsticks is still needed.