[Urine glucose testing in children with diabetes mellitus--obsolete as a daily control measure?].

Sixty-four children with type 1 diabetes (33 girls and 31 boys with a mean age of 11 1/4 years) were studied in order to ascertain whether urine sugar measurements performed by the patients or their parents can provide adequate monitoring of treatment. The results of two home-testing methods for urine glucose were compared with subsequent glucose dehydrogenase assays in the laboratory. Over the range from 0 to 5 g/dl the results of home testing disagreed with the laboratory checks by an average of one concentration grade, and displayed wide scatter. Only in the concentration range of 0.5 g/dl were there differences between the nonspecific reduction test (Clinitest) and the specific enzymatic assay. Glucose concentrations in urine specimens passed at 7 am, 12 noon and 6 pm (home testing) were compared with blood glucose concentrations checked at the same times. Correlation was not very close (correlation coefficients between 0.3 and 0.64). Correlation between pooled urine collections and 24 hour blood sugar profiles was equally poor (0.5-0.6). The author concludes that urine sugar testing is unsatisfactory for home monitoring of children with diabetes and the results are inadequate for diagnostic or therapeutic purposes.