Diabetes Research Centre

Melbourne, Australia

What is Diabetes?

Clinical Trials

In the Lab

Links

Contacts

Site History

The Mixed Meal Tolerance Test

versus

The IV Glucagon Stimulation Test

for the assessment of beta cell function in type 1 diabetes.

John M Lachin (1), Paula Friedenberg (1), Carla Greenbaum (2). Presented by Peter Colman (3) on behalf of Type 1 Diabetes TrialNet.

1. The Biostatistics Centre, The George Washington University, Rockville MD, USA; 2. Benaroya Research Institute, Seattle, Washington, USA; 3. Department of Diabetes and Endocrinology, The Royal Melbourne Hospital.

Introduction

The measurement of stimulated C-peptide is widely accepted as a measure of beta cell function and is increasingly being used as an important endpoint in studies evaluating treatments to preserve endogenous insulin production at onset of type 1 diabetes. Previous trials have used either a Mixed Meal Tolerance Test (MMTT) or Glucagon Stimulation Test (GST). However, there has been no evidence based consensus regarding which test is superior.

This test will help us develop a standardised way to test for insulin production in pre-clinical and clinical type 1 diabetes. A standardised test of this kind will be invaluable both for planning intervention and prevention studies in type 1 diabetes and for measureing their outcome.

Aim

The primary objective of this study was to evaluate the reproducibility and tolerability of the two tests in the same subjects under similar test preparation conditions when performed at TrialNet clinical centers in the US, Canada, Europe and Australia.

Research Design and Methods

Study Participants

Subjects were enrolled between September 29, 2004 and December 5, 2005 at 14 North American and 3 international centers.

Subjects were eligible if they

were between 8 and 35 years of age,
weighed > 30kg,
had type 1 diabetes by ADA criteria or the judgment of a physician,
were diagnosed with diabetes 1 month - three years prior to enrollment.

Randomization

Subjects were randomly assigned to receive either two Mixed Meal Tolerance tests (MMTTs) followed by two Glucagon stimulation tests (GSTs), or two GSTs followed by two MMTTs. Subjects were instructed to withhold long acting insulin on the morning of the test. Very Short acting insulin was allowed up to 2 hours before the test and regular insulin up to 6 hours before the test. Subjects on CSII could continue with the normal basal rate.

MMTTs - 6 ml/kg of Boost up to maximum of 360ml, ingested within 5 minutes. Blood samples were obtained 10 minutes and immediately before the start of ingestion of the Boost and then at 15, 30, 60, 90 and 120 minutes.

GSTs - 1 mg of Glucagon was administered IV over 10 seconds. Blood samples were obtained 10 minutes and immediately before the bolus then at 2, 4, 6, 8 and 10 minutes after.

All samples were assayed for both glucose and C-peptide.

Results

148 subjects (mean age 16 years; T1DM duration 1.4 years) were enrolled. 123 completed all 4 visits, 15 completed 3 visits, 2 completed 2 visits, and 8 completed one visit.

Reliability

Bland-Altman plots showing the differences between the two tests are shown in Figure 1.

The mean peak C-peptide from the MMTT was significantly greater than that from the GST (Figure 2).

There was a significant difference in reliability between the MMTT and the GST for the peak value (p=0.018) and for the AUC (p=0.015). The differences in reliability in log peak and log AUC were also significantly different (p<0.0001 for both) (Table 1).

Table 1. Intraclass correlations (as estimated by the Pearson correlation coefficient) and asymmetric 95% confidence limits between the 1st and 2nd measurement of each test.

The intraclass correlations did not differ significantly across categories of age, sex, or length of time since diagnosis of diabetes.

Test Preferences 114/123 subjects expressed a test preference, with 53% (60) preferring the MMTT to GST. The preference depended strongly on age: 86% (31/36) of subjects under 13 y preferred the MMTT, compared with 45% (17/38) of subjects between 13 and 17 y and 30% (12/40) of subjects 18 y and older (p< 0.0001). Eighty five percent (51/60) of the subjects who preferred the MMTT cited the nausea experienced during the GST as a reason. Ninety four percent (51/54) of those who preferred the GST, indicated the shorter duration a reason. Figure 2 (left). Mean C-peptide in MMTT (A) and GST (B)

Conclusion

Both tests provide reproducible measures of C-peptide response, MMTT more so. MMTT peak values are higher then GST values. Incidence of adverse reactions with GST, principally nausea, was markedly higher then MMTT. 56% of subjects preferred MMTT, 44% preferred GST. These findings will guide the choice of measures of beta cell function in future TrialNet studies.

The MMTT-GST Research Study is supported by Type 1 Diabetes TrialNet (click on icon below to learn more about TrialNet)

Learn about our other TrialNet Studies: the Natural History Study and the Rituximab Trial

Return to MMTT-GST Page

Return to Clinical Trials Menu