Diagnostic accuracy of the urinary albumin: creatinine ratio determined by the CLINITEK Microalbumin and DCA 2000+ for the rule-out of albuminuria in chronic kidney disease

Abstract

An increased urinary albumin excretion (albuminuria) is an established test for the early detection of renal disease and is also recognized as a risk factor for cardiovascular disease and mortality in a number of clinical settings. There is an established body of data which shows that a random urinary albumin:creatinine ratio (ACR) based on a random urine sample correlates well with 24-hour urinary albumin excretion measurement. However, there is little data to show whether specific point-of-care testing devices can be used to rule-in or rule-out increased urinary albumin excretion in comparison to a 24-hour urinary albumin excretion measurement.

This study evaluated the ability to rule-in or rule-out albuminuria in a cohort of patients attending a renal outpatient clinic, using the urinary ACR determined by the CLINITEK Microalbumin (Siemens Healthcare Diagnostics Inc., Deerfield, US) a semi-quantitative strip test, and by the DCA 2000+ (Siemens Healthcare Diagnostics Inc.) a quantitative cassette based test using 3 random urine samples collected within a 24-hour period compared to 24-hour urinary albumin measurement.

The CLINITEK system was shown to be a reliable test for ruling out increased urinary albumin excretion with negative likelihood ratios less than 0.05 above the 24-hour urinary albumin excretion rate of 30 mg/24 h (threshold for microalbuminuria), and less than 0.01 above the albumin excretion rate of 100 mg/24 h. The DCA 2000+ system demonstrated similar performance as a rule-out test, with likelihood ratios of less than 0.02 at 24-hour albumin excretion rates above 30 mg/24 h. Both the CLINITEK and DCA 2000+ systems could be used to rule-out increased urinary albumin excretion at the albumin excretion cut-off rate of 30 mg/24 h in this cohort of patients.

Introduction

The presence of increased albumin excretion in the urine is now recognized as an independent indicator of risk for cardiovascular disease. In addition, tests for the presence of proteinuria or albuminuria are well established tools used to aid in the diagnosis and management of chronic kidney disease (CKD) and as such are endorsed by clinical guidelines [1], [2], [3]. Furthermore, the presence of albuminuria is associated with poorer mortality and morbidity in women with pre-eclampsia [4], as well as in patients admitted to the intensive care units [5], [6], [7].

Initial observations on the value of albuminuria and proteinuria in CKD were based on quantitative laboratory techniques requiring a 24-hour urine collection. It is well documented that the complete 24-hour urine collection that is necessary for an accurate 24-hour protein or albumin measurement is fraught with difficulty as many patients are unable to comply, with the resulting measurements not truly representing 24-hour protein or albumin excretion and thus are unreliable [8]. Albumin measurement in random urine samples has been advocated, but the intra-individual variability in the albumin excretion rate throughout the day precludes this approach. Hishi et al. [9] showed that there is some variation in the urine albumin excretion in the daytime (for example between 10.00 and 22.00 h) whilst the excretion is significantly lower at night, and higher first thing in the morning. The intra-individual variation in a random urine albumin excretion can be dramatically reduced by expressing the albumin measurement as a ratio to the creatinine excretion [10], and in some guidelines it has been suggested that the 24-hour albumin excretion measurement can be replaced by the ACR determined from a random urine sample [1], [3]. Despite the limitations of the albumin excretion rate, whether as a random urine or as a 24-hour collection, it has been suggested that a urine albumin measurement using a dipstick device may be used as a first-line test with confirmation of abnormal results [11]. Much of the literature examining the utility of the ACR as a surrogate for the 24-hour urine collection has been performed as a correlation between the two methods; however this does not show whether the test is best used for rule-in or rule-out of albuminuria. Furthermore, in the case of a qualitative or semi-quantitative test, it is only possible to consider the test as being capable of such a dichotomous decision. A test that facilitates reliable rule-in can help in making the diagnosis (with or without a confirmatory test) whilst a test that facilitates reliable rule-out can direct the clinician towards an alternate diagnostic hypothesis.

The main aim of this study was to determine whether use of a simple test for the ACR could reduce the need for 24-hour urine collections used to detect albuminuria. The objectives were to assess the diagnostic accuracy of the CLINITEK and DCA 2000+ [12], [13] as rule-in or rule-out tests for increased urinary albumin excretion. This question can be posed in a range of clinical settings, from screening asymptomatic patients, where a low prevalence of albuminuria would be expected, through to a scenario where a higher prevalence is expected. This study focussed on the latter scenario with a cohort of patients attending a renal outpatient clinic.