Clinical performance of a urine albumin-to-creatinine ratio (ACR) strip test compared with quantitative measurement

Abstract

Background

Albuminuria is an early marker of kidney damage and an essential component of chronic kidney disease (CKD) risk stratification. Quantitative measurement of the urine albumin-to-creatinine ratio (ACR) in a spot, preferably morning, urine sample is the recommended standard. In low-throughput laboratories and point-of-care settings, semi-quantitative strip tests are sometimes used for screening, although their diagnostic performance is limited.

Objective

To evaluate the clinical utility of a commercial semi-quantitative urine strip test for albumin and creatinine by comparison with quantitative ACR measurement.

Methods

Eighty-four spot urine samples were analysed. Semi-quantitative ACR categories were obtained using an ACON Mission strip test, according to the manufacturer’s instructions. Quantitative ACR was measured on a Beckman Coulter AU480 analyser (immunochemical albumin, enzymatic creatinine). Diagnostic performance was assessed using sensitivity, specificity, positive and negative predictive values, Matthew’s correlation coefficient (MCC) and Cohen’s κ. A decision threshold of 3 mg/mmol (30 mg/g) was applied.

Results

Sensitivity for detecting albuminuria (≥3 mg/mmol, ≥30 mg/g) was 80.6% (95% CI 62.5–92.5) and specificity was 60.4% (95% CI 46.0–73.5). The negative and positive predictive values were 84.2% and 54.3%, respectively. Overall categorical agreement across A1–A3 was 63.1% (κ = 0.334; MCC = 0.398). Most discrepancies reflected overclassification of low-grade albuminuria by the strip test in samples classified as normoalbuminuric by quantitative ACR.

Conclusions

The semi-quantitative strip test shows high sensitivity with moderate specificity and is suitable for screening for albuminuria. Positive results should be confirmed by quantitative ACR measurement in accordance with current guidelines.