Microbumintest and non-albumin proteinuria in diabetes

Bedside methods for the detection of microalbuminuria such as Microbumintest (TM) have the advantage of simplicity but not the specificity of radio-immunoassay. In the present study we assessed whether apparently inappropriate positive Microbumintest results in the presence of low urinary albumin concentrations could be accounted for by non-albumin proteinuria of glomerular or renal tubular origin. Urinary albumin and transferrin were considered to indicate glomerular proteinuria, and alpha-1-microglobulin and N-acetyl-beta-D-glucosaminidase to reflect tubular proteinuria. Microbumintest had a sensitivity of 100% and specificity of 67% to detect a urinary albumin concentration of 40 mg/l. Samples with albumin concentration less than 40 mg/l contained more total protein: 110 (78-155) v 60 (35-104) mg/l p less than 0.0001 (geometric means with 1 SD range), more albumin: 11.7 (5.1-26.8) v 5.4 (2.8-10.4) mg/l p less than 0.005 and more transferrin: 496 (191-1284) v 174 (78-389) micrograms/l p less than 0.001, in those testing positive with Microbumintest than in those testing negative. Microbumintest had a sensitivity of 82% and specificity of 75% to detect an albumin concentration of 20 mg/l. In samples containing less than or equal to 20 mg/l albumin, the mean albumin concentration was no greater in those testing positive compared with those testing negative. However, total protein: 108 (72-161) v 60 (34-105) mg/l p less than 0.001, and transferrin: 326 (148-715) v 157 (78-316) micrograms/l p = 0.01 both remained increased in samples testing positive compared with those testing negative.(ABSTRACT TRUNCATED AT 250 WORDS)