Intravenous glucose tolerance test following oral tolbutamide administration. A possible model for the evaluation of a new hypoglycaemic agent.
Like any investigational drug, new oral hypoglycaemic agents must be studied in healthy subjects for the pharmacokinetic and pharmacodynamic evaluation and for tolerability profile. Unlike the biguanides, sulfonylureas do cause hypoglycaemia in non-diabetic individuals. The development of appropriate methods for the quantitation of hypoglycaemic effect of a new agent is essential for the phase I studies. A multistep design for the evaluation of a hypoglycaemic agent is desirable. Intravenous glucose tolerance test (IV GTT) is one of the tests for such evaluation. By using the intravenous route, variability in absorption of glucose is eliminated and thus the test may prove suitable for establishing the dose-response relationship of the new drug.
In order to establish a procedure for standardising IV GTT as a model for testing a new hypoglycaemic agent we performed IV GTT after tolbutamide, an established oral hypoglycaemic agent, by varying the conditions of the test such as the glucose load, the time interval between the drug and the IV GTT and the dose of the drug.
Nine healthy male volunteers weighing between 45 and 55 kg and ranging in height from 155 to 177 cm were used in this study. The age of the volunteers ranged between 25 and 40 years. Fasting blood glucose levels were in the normal range. They were hospitalized and received high (more than 250 g) carbohydrate diet for at least 3 days preceding the IV GTT. An informed consent was obtained prior to the study.
An open randomized crossover design was used. After overnight fast (about 10-12 hours) a butterfly needle was inserted into a forearm vein around 7.00 a.m. and kept in situ. After collection of fasting blood sample, the subject was given either capsules containing 200 mg of lactose or tablets of tolbutamide (1 g or 0.5 g) orally. The crossover study was carried out after an interval of 4 to 7 days.
After collection of additional blood samples at 1/2, 2 and 4 hours, the calculated dose of glucose (5 g/m2 or 15 g/m2 body surface area) was injected intravenously as 25% solution of anhydrous dextrose. The smaller dose was administered in 1 and the bigger dose was injected in 5 minutes, followed by saline flush. Since, in preliminary studies, a change in K value was not observed with tolbutamide when IV GTT was performed at 2 and 4 hours the subsequent studies were done at 1/2 hour after tolbutamide.
Blood samples were collected in heparinized glass tubes at 2, 10, 20, 30, 40, 50 and 60 minutes after the end of intravenous injection of glucose. The tubes were centrifuged, plasma was separated into plastic vials and frozen in dry ice. Plasma glucose was estimated on Technicon autoanalyser utilizing Hoffman method. Plasma tolbutamide concentrations were assayed by a modified HPLC method.
Conventional IV GTT is carried out with 0.5 g/kg glucose load. The slope of glucose disappearance curve (time against log to the base e of glucose concentration) is expressed as K value calculated for the period of 10 to 60 minutes following IV glucose. However, as our preliminary studies showed that the plasma glucose concentration reached the fasting level within 30 minutes after an IV glucose load of 5 g/m2, the K value of glucose disappearance was calculated for the period of 10 to 30 minutes after the IV injection, by the following standard formula:
K = 0.693 x 100
t/1/2 of glucose concentration disappearance (Minutes)
Continuous in vivo plasma glucose monitoring studies also suggest that the optimum period of IV GTT for calculating K value is 10 to 30 min.
Student's one tail, paired `t' test was employed for the statistical analysis.
(a) IV GTT with the glucose load of 15 g/m2 given 1/2 hour after 1 g of tolbutamide [Table 1]
At 30 minutes the plasma glucose was significantly lower with tolbutamide; the glucose disappearance rate was more with tolbutamide than without tolbutamide and significant difference in K value could be observed.
Since the dose of glucose is known to alter the K values, further studies were done with a smaller glucose loads It was also observed that plasma tolbutamide levels reach peak between 3 and 4 hours and therefore IV GTT was performed at 4 hours after tolbutamide.
(b) IV GTT with the glucose load of 5 g/m2 given at 4 hours after tolbutamide [Table 2]
[Table 2] indicates the effect of tolbutamide on the plasma glucose levels before and during IV GTT. At the end of 4 hours (i.e. just before IV glucose) the plasma glucose level was lower during tolbutamide session. Ten minutes after IV glucose injection plasma glucose was significantly lower with tolbutamide than during control session. But from 20 minutes onwards the glucose levels did not differ significantly between the two sessions except at 60 minutes, when it was lower with tolbutamide than during the control session. However, the K values did not differ significantly during the two test sessions.
Since enhancement in K value could not be demonstrated at 4 hours after tolbutamide, the time interval was reduced to 2 hours and subsequently to 1/2 hour.
(c) IV GTT with the glucose load of 5/gm2 given 2 hours after tolbutamide:
This was studied in one subject. Plasma glucose levels were lower at 40, 50 and 60 minutes during the tolbutamide session and the K value was not higher with tolbutamide compared to the control session.
(d) IV GTT with the. glucose load of 5 g/m2 given 1/2 hour after tolbutamide [Table 3]
[Table 3] shows results with 1 g dose of tolbutamide. Plasma glucose levels were significantly lower at 2 hour after tolbutamide than during the control session. Thus the hypoglycaemic effect of tolbutamide was apparent within 30 minutes after the ingestion of the tablets. During the first 10 minutes of IV GTT, the plasma glucose levels were not significantly different in two test sessions. From 20 minutes onwards the glucose concentrations were significantly lower with tolbutamide than during the control session. At 30 and 40 minutes, the mean levels were below 50 mg% during the tolbutamide session, but none of the volunteers reported hypoglycaemic symptoms. The K values were significantly higher with tolbutamide than during control sessions.
With 0.5 g dose of tolbutamide, the glucose levels were not reduced within hour [Table 4] The mean glucose levels did not differ in the two test sessions except at the 20 and 30, minutes. The K values were significantly higher with tolbutamide.
As an investigative tool IV GTT has the advantage that its results can be expressed in terms of a single number (K value) which represents the rate of disappearance of glucose. This permits easy mathematical handling and statistical evaluation of IV GTT. By contrast the results of oral GTT cannot be mathematically handled so easily.
IV GTT has been used to investigate the activity of a new hypoglycaemic agent. The present study suggests the need for optimizing the glucose load for IV GTT and the time interval between the oral administration of the drug and IV GTT during such investigation.
K value is known to increase linearly to a plateau with increasing glucose loads. We did not find such a correlation. Conventionally near-maximum glucose loads (0.5 g/kg) have been used in IV GTT for diagnosing diabetes. In our subjects this worked out to about 15 g/m of glucose load. When this load was used we found that 1 g of tolbutamide enhanced the K value. On the other hand when 5 g/m2 of glucose was used the K value was enhanced by the same dose of tolbutamide. Siegal et al also found that 1 g of tolbutamide enhanced K value only when the glucose load of less than 20 g was used. Marigo et al also have used a submaximal glucose load of 0.33 g/kg in order to demonstrate the activity of a new hypoglycaemic agent. Thus it would appear that glucose loads less than 20 g/m2 would be ideal for demonstration of hypoglycaemic activity of a new drug. The present study demonstrates that although the peak plasma levels of tolbutamide occur between 3 and 4 hours after oral administration [Fig. 1] enhanced K value is seen when IV GTT is done at 1/2 hour but not at 4 hours after ingestion of the drug. Similar results have been observed by Doshi and Kulkarni1 who noted that nadir of hypoglycaemia occurs at 11/2 hour after oral administration of tolbutamide in fasting subjects, whereas the Plasma tolbutamide levels reached peak between 3 to 4 hours.
It has been proposed that by the time plasma tolbutamide peak is reached the blood sugar has risen from the lowest hypoglycaemic levels to a higher level, probably because of the counter regulatory hormonal mechanisms.
Thus it would appear that the time interval between the oral administration of a hypoglycaemic agent and the IV GTT is critical for proper evaluation of its activity. Therefore, it is important to establish the optimum time interval for individual hypoglycaemic agents. If this is not done an effective hypoglycaemic agent may be discredited because the IV GTT was performed at inappropriate time interval.
As some basic information is required before an IV GTT is used, it cannot be employed as the primary screening test for a new hypoglycaemic agent. For this purpose lowering of the blood sugar in a fasting individual would be a more suitable parameter.