| Description |
The urinary glucose determinations of diabetics are not used simple as an indicator of the glucose level in the urine but as an indicator of the blood glucose level and control of the carbohydrate abnormality. Therefore the blood glucose level predicted from semi-quantitative urine glucose tests such as Clinitest would be more useful in assessment in the control of the carbohydrate abnormality if the accuracy of the prediction could be established. The purpose of this study was to determine how accurately the blood glucose level could be predicated by measuring with Clinitest Reagent tables, the level of glucose in the second voided urine sample of diabetic children. The subjects of this study were 72 diabetic children, between the ages of nine and fourteen, attending Camp Utada for diabetic children, August 15-22, 1971. The subjects were divided into 6 groups of between 8-15 children and each group was followed for one day. A second voided urine sample was obtained from each subject before each meal and at bedtime (7:30 AM, 11:20AM, 4:30 PM, 9:00PM) and was tested for glucose using Clinitest Reagent tablets and acetone using Ketostix Reagent Strips. Immediately upon receiving the second voided urine sample, a capillary blood sample was taken and analyzed for glucose level using Dextrostix Reagent Strips and a Reflectance Meter. Simultaneous capillary blood glucose levels and urine glucose levels were thus measured on 72 diabetic children resulting in a total number of 259 samples. The results showed the negative Clinitest urine glucose level gave an accurate prediction of blood glucose levels considering the renal threshold for glucose to be approximately 180 mg/100 ml. 52 out of 55 samples with a negative urine glucose measurement demonstrated blood glucose levels under 180 mg/100 ml. The middle 4 Clinitest urine glucose measurements (trace, +, ++, +++) showed tremendous overlap making prediction of accurate blood glucose levels impossible. Over half the samples showed urine glucose measurements in the ++++ range. The ++++ range also showed wide variation making accurate prediction of blood glucose level difficult. The +++++ urine glucose level appeared to be an accurate measurement of high blood glucose levels of over 440mg/100 ml. The study suggests that the best prediction of blood glucose levels from Clinitest urine glucose values would be obtained if we combined the trace, +, ++, +++ Clinitest urine reactions and consider them all as + reaction. Negative Clinitest would include all blood glucose levels under 180 mg/100 ml, + Clinitest would include blood glucose levels ranging from 130-250 mg/100 ml and ++++ Clinitest would include ranges from 200-500 mg/100 ml. +++++ Clinitest would indicate blood glucose of over 400 mg/100ml. The Pearson Product-Moment Correlation showed r = 0.72 thus indicating a positive correlation between blood glucose levels and urine glucose levels as measured with Clinitest Reagent tables. When the samples were divided according to the time of day the test was done it was found that the Pearson Product-Moment Correlation was highest at 4:30 PM (r = 0.83) and lowest at 11:30 AM (r = 0.51). It was also found that when acetone was present in the urine there was a poor correlation (r = 0.05) of blood glucose and urine glucose measurements. Because of the small sample size of those showing acetone this was difficult to interpret. On the basis of the research findings, the conclusion may be drawn that except for the negative Clinitest reacation and +++++ Clinitest reaction we are not able to predict blood glucose level accurately from knowing the Clinitest urine glucose measurement. |
