We read with interest the letter from Comper et al. (1) that compared the performance of Bayer’s Microalbustix and CLINITEK Microalbumin urinalysis strips with a high-performance liquid chromatography (HPLC) method that allegedly detects nonimmunoreactive forms of albumin. The Comper et al. letter implies that the American Diabetes Association established the >95% detection criteria for semiquantitative microalbuminuria tests on the basis of this HPLC reference method.

This implication is misleading. The American Diabetes Association criteria were not likely based on the Comper et al. method of detecting nonimmunoreactive forms of albumin because that method is neither standardized nor commonly used. Furthermore, the suggestion of Comper et al. that Bayer product performance is deficient in this area is not supported by available evidence.

Standard immunoassay and chemical methods for measuring albumin do not detect nonimmunoreactive forms of albumin. Moreover, we are not aware of any National Bureau of Standards traceable standard containing the nonimmunoreactive form(s) of albumin, albumin fragments, and/or albumin aggregates combined with standard immunoreactive albumin in the correct proportions to standardize the HPLC method referenced in the Comper et al. letter.

To our knowledge, there is no peer-reviewed publication proving that nonimmunoreactive forms of albumin provide earlier kidney disease detection in humans compared with standard albumin. Comper et al. provided no performance data to show the false-positive rate for kidney disease detection when nonimmunoreactive forms of albumin, albumin fragments, and/or albumin aggregates are measured. To determine the clinical false-positive rate, a diagnostic work-up on all patients in the study would need to be done. This could include one or more of the following: glomerular filtration rate, creatinine clearance, urine sediment analysis, ultrasound, imaging, or persistent microalbuminuria (standard form of albumin). If no evidence of disease is found from established methods, it is unclear whether a patient with nonimmunoreactive albumin in their urine has early kidney disease.

Bayer is committed to the early detection of kidney disease and has a keen interest in new markers that are proven to advance the diagnosis of kidney disease. Bayer CLINITEK Microalbumin product sensitivity has been demonstrated at 97–98.4% when compared with standard albumin and creatinine assay methods that are accepted by health professionals managing patients with diabetes (2,3).

1.
Comper WD, Jerums G, Osicka TM: Deficiency in the detection of microalbu-minuria by urinary dipstick in diabetic patients (Letter).
Diabetes Care
26
:
3195
–3196,
2003
2.
Davidson EM, Croal BL: Introduction of an albumin-to-creatinine ratio point-of-care device: analytic, clinical, and cost-effectiveness aspects.
Point of Care
2
:
89
–95,
2003
3.
Kutter D: A chemical test strip to determine low concentrations of albumin and creatinine in urine.
Lab Med
29
:
769
–772,
1998