Performance of a reagent strip device for quantitation of the urine albumin: creatinine ratio in a point of care setting

Abstract

Aim: We have evaluated the performance of a reagent strip (Clinitek, Bayer plc, Newbury) incorporating a novel dip and read device for the quantitation of the albumin: creatinine ratio together with their individual concentrations in urine.

Methods: The performance was compared with that of a lateral flow device for the semi-quantitation of albumin (Micral II, Roche Diagnostics, Lewes) and also a laboratory based procedure. The device employs novel methods for both analytes, using a sulphonephthalein dye binding at pH 1.5 for albumin and the peroxidase-like activity of copper creatinine complexes. The color yields of the separate reaction pads are monitored with the Clinitek 50TM bench top urine chemistry analyzer and compared to a pre-programmed calibration algorithm.

Results: The imprecision of the device was assessed by observing the discrepancy between duplicates in a total of 144 urine samples collected from patients with diabetes and/or renal disease; there were 10 discrepancies in the case of the albumin estimation (6.9%), 12 in the case of the creatinine estimation (12.5%) and 23 in the case of the albumin: creatinine ratio (16.0%). In the case of the Micral II, where 96 of the urines were analyzed there were 12 discrepancies (12.5%). When considered as a two-class test for albumin with a cut off of 20 mg/L the Clinitek gave a sensitivity of 95.4% with specificity of 78.9% and a positive predictive value of 87.4%. When analyzing four urines (two controls, two patient pools) with replicate analysis on the Clinitek system we found 100% agreement for the albumin estimation, 95% for creatinine and 96.7% for the albumin: creatinine for 60 analysis within a day and 100, 95 and 97.5% on single analyses each day for 20 days for the two urine controls. The discrepant results were always within one color block. When considering the albumin: creatinine ratio with a cut off of < 30 mg/g the Clinitek gave a sensitivity of 76.3% with a specificity of 89.1% and a positive predictive value of 89.7%.

Conclusion: The Clinitek system provides a reliable means to screen for microalbuminuria with the opportunity of a semi-quantitative assessment when microalbuminuria is found.