Serial changes in blood sugar level were studied in normal, diabetic and renal glycosuric subjects after oral administration of 50 g. xylitol. No significant changes were observed in the blood sugar levels of normal and some of the renal glycosuric subjects after oral administration of xylitol. As opposed to this, in diabetic and in the majority of renal glycosuric subjects, significant elevation of blood sugar levels was observed. It is suggested that a study of the serial changes in blood sugar level after oral administration of xylitol may reveal the nature of diabetic condition.
Both oral and intravenous glucose tolerance tests were performed in 132 subjects with glycosuria, obesity, pyodermia, family history of diabetes and/or the other findings suggestive of diabetes mellitus.The standard three hour oral glucose tolerance test was performed in the ordinary way using 100 g of glucose, and the result was evaluated by the diagnostic criteria of Wilkerson et al. (1960). The intravenous test was performed 1 or 2 days later.0.66ml per kg of 50% glucose solution was injected intravenously over 3 min.Blood samples were obtained from the antecubital vene at intervals of 15 min. between 15 and 60 min. after the injection.The result of the intravenous test was expressed by the K value (glucose decay constant, glucose disappearance rate, glucose assimilation coefficient) by the way of Conard et al. (1953).The blood glucose determination was made by the o-aminobiphenyl method. According to the results of the oral test, 132 subjects were divided into 3 groups-90 diabetics, 13 possible diabetics and 29 non-diabetics. 1) Calculation of the K value was made easily in 112 cases, but failed in 20 cases (13 non-diabetics, 2 possible diabetics and 5 diabetics). More frequent sampling of the blood between 15 and 45 min.after the injection would enable us to calculate the K value in almost all cases. 2) The mean and the range of the K value was 0.67 (0.10-1.85) in 85 diabetics, 1.38 (0.85-2.37) in 13 possible diabetics and 1.59 (0.65-2.40) in 16 non-diabetics.There was considerable overlapping particularly in the region from 1.11 to 1.30 between the diabetic and the non-diabetic group. 3) If the above results were applied for diagnosis, one would say that a value below 1.10 signifies a diabetic state, while a value above 1.31 means non-diabetics.The subjects with the values between 1.11 and 1.30 would be regarded as bor-derline cases.Judging from this diagnostic criteria, the results of the intravenous test agreed completely with that of the oral one in 80% of 132 cases.Marked discrepancy occured only in 4.6%. As a whole, the oral test seemed to be a little more sensitive than the intravenous one, although there existed some cases in which the intravenous test was more sensitive for detecting impaired glucose tolerance. 4) Statistically significant inverse correlation was observed between the value of K and the blood glucose values in the oral test.The correlation seemed to be higher between the K value and 2 or 3 hour value in the oral test than between the K value and fasting or 1 hour value. 5) The value of K did not seem to be an important index to prospect the effect of the oral antidiabetic drugs. 6) Relatively low value of K was observed in the cases with postgastrectomy glycosuria.
In the standard glucose tolerance test (SGTT), 100 g of glucose is administered after the fasting time of 14-15hours. In this study the fasting time was prolonged to 20 hours by omitting meals after 2 p.m.on the day before the tolerance test. Comparative studies were carried out in 10 normal healthy subjects, 8 diabetic-borderline cases, 3 cases with renal glycosuria and 6 diabetic patients. The results are as follows: 1.There was no significant difference in blood sugar tolerance curve changes between SGTT and HGTT performed in the normal subjects and in the diabetic patients. 2.Six of the 8 borderline cases diagnosed with SGTT showed a diabetic pattern in HGTT.Mean values of blood sugar levels at 90, 120 and 180 minutes after glucose ingestion were higher in HGTT than those in SGTT (p<0.05).