Microalbuminuria measurements by two in-house ELISA methods

Detection of microalbuminuria is important in the management of diabetic patients since it is predictive of development of proteinuria and nephropathy. Two sensitive and specific in-house ELISAs for microalbuminuria were established and validated. One of the ELISAs was based on antigen coating while the other employed antibody coating. Recovery and linearity experiments gave acceptable results of 100 +/- 10%, while precision results were <10% for intra-assay and <12% for inter-assay coefficients of variation (CVs). The standard curve ranged from 10-625 ug/l, equivalent to 0.2-12.5 mg/l for urine samples diluted 1:20 fold. When the antibody coated ELISA was compared to antigen coated ELISA, a correlation of r=0.996 was obtained. When compared to commercial kits, the in-house ELISAs gave good correlations of r=0.961 versus the Boehringer Mannheim Micral Test strips and r=0.940 versus Ames Microalb Turbidimetry. The normal microalbumin reference ranges determined for 12h, first morning and random urine samples were 0.7-5.3 mg, 0.1-10.2 mg/l and 0.8-26.1 mg/l respectively. The normal albumin excretion rate (AER) was 1.0-7.3 ug/min while untimed urine samples gave results of 0.1-0.9 and 0.2-1.6 mg/mmol after dividing by creatinine concentrations. The ELISAs were used to detect microalbuminuria in 338 random urine samples from diabetic patients. A high percentage 47.9% was found to be positive for microalbuminuria and 18.0% had macroalbuminuria >25 mg/mmol. Thus screening for microalbuminuria together with creatinine measurements using random urine samples can be used for management of diabetic patients.