Reply to E. Cerasi

Dr Cerasi’s questions cover two points. The first point is the assessment of insulin secretion one hour after an (exogenous) infusion with insulin that we have used in order to assess the influence of gliclazide on first phase. The second point pertains to the level of hyperglycemia.

Point 1: The insulin infusion has been given for the following reasons: 1. All subjects would start out at an identical blood glucose level; 2. Even modest hyperglycemia diminishes gastric emptying markedly, as mentioned in the text [1].

Our findings already showed an appreciable first phase with placebo, as mentioned in the results section. This, in itself, shows that the actual technique of first using an exogenous insulin infusion does not preclude the measurement of a first phase secretion. Interestingly, as Dr Cerasi points out, gliclazide already appeared to increase the “basal” level of insulin secretion (the resting C-peptide concentration) before the bolus injection with glucose, which is the actual start of the hyperglycemic glucose clamp. This is also mentioned in our article. It is open to discussion whether we should call this augmentation of first phase secretion, but we certainly agree with Dr. Cerasi that it shows the potential of gliclazide on β-cell function. However, the first phase after gliclazide was only about 20 % of the first phase that we have seen in similar experiments previously published on non-diabetic subjects, even without sulphonylureas [2]. The main message of the article is clearly that these studies failed to show a large increase in first-phase secretion. This picture also arises from the recent report of Drs Van der Wal and Heine [3]. These authors were also unable to find an appreciable first-phase secretion in hyperglycemic clamps. Interestingly, in their studies, no (exogenous) insulin infusion had been given. However, we do not deny that gliclazide is capable of augmenting first-phase secretion in suitable circumstances, especially in non-diabetic subjects. Non-diabetic subjects have a much larger first-phase secretion than diabetic subjects; the first phase represents an insulin pool, which acts as a reservoir of insulin “ready-to-go.” Probably, a type 2 diabetic does not have much of a pool of this description. Possibly, the beta-cells of type 2 diabetic subjects need a longer resting period to slowly build up such a pool again. This may have happened in the interesting studies mentioned by Dr Cerasi [4]. In his studies, subjects had been infused with CSII, which led to a superb metabolic control and, at the same time, presumably to a nice resting period for the (over-stressed?) beta cells.

Regarding the second point: The level of hyperglycemia chosen for our studies is certainly subject to discussion. As mentioned in our paper, the level of 8 mmol/l was chosen since it is around the ED50 for glucose induced first phase secretion. It was also the level chosen for our previous studies with glibenclamide [2]. It is also a blood glucose level which pertains to the level of (hyper-) glycemia generally relevant to the daily life of our patients. It should be clear that these subjects normally (that is: outside the study period) were taking SU derivatives on a daily base, and that their (fasting) blood glucose levels were indeed below 10 mmol/l.

References

• 1 Groop L C. Sulphonylureas in NIDDM.  Diabetes Care. 1992;  15 737-754
• 2 Ligtenberg J JM, Sluiter W J, Reitsma W D, van Haeften T W. Effect of glibenclamide on insulin release at moderate and high hyperglycemia in man.  Eur J Clin Invest. 1997;  27 685-689
• 3 van der Wal P S, Heine R J. Characteristics of pancreatic beta-cell secretion in type 2 diabetic patients treated with gliclazide and glibenclamide.  Diabetes Res Clin Pract. 2001;  52 103-111
• 4 Delia Casa L, Del Rio G, Glaser B, Cerasi E. Effect of 6-month gliclazide treatment on insulin release and sensitivity to endogenous insulin in NIDDM: Role of initial continuous subcutaneous insulin infusion-induced normoglycemia.  Am J Med. 1990;  90 37S-45S