Impact of the timing of metformin administration on glycaemic and glucagon-like peptide-1 responses to intraduodenal glucose infusion in type 2 diabetes: a double-blind, randomised, placebo-controlled, crossover study
Abstract
Aims/hypothesis
Metformin lowers postprandial glycaemic excursions in individuals with type 2 diabetes by modulating gastrointestinal function, including the stimulation of glucagon-like peptide-1 (GLP-1).
The impact of varying the timing of metformin administration on postprandial glucose metabolism is poorly defined. We evaluated the effects of metformin, administered at different intervals before an intraduodenal glucose infusion, on the subsequent glycaemic, insulinaemic and GLP-1 responses in metformin-treated type 2 diabetes.
Methods
Sixteen participants with type 2 diabetes that was relatively well-controlled by metformin monotherapy were studied on four separate days in a crossover design.
On each day, participants were randomised to receive a bolus infusion of metformin (1000 mg in 50 ml 0.9% saline) via a nasoduodenal catheter at t= −60, −30 or 0 min (and saline at the other timepoints) or saline at all timepoints (control), followed by an intraduodenal glucose infusion of 12.56 kJ/min (3 kcal/min) at t= 0–60 min. The treatments were blinded to both participants and investigators involved in the study procedures. Plasma glucose, insulin and total GLP-1 levels were measured every 30 min between t= −60 min and t= 120 min.
Results
There was a treatment-by-time interaction for metformin in reducing plasma glucose levels and increasing plasma GLP-1 and insulin levels (p<0.05 for each).
The reduction in plasma glucose levels was greater when metformin was administered at t= −60 or −30 min vs t= 0 min (p<0.05 for each), and the increases in plasma GLP-1 levels were evident only when metformin was administered at t= −60 or −30 min (p<0.05 for each).
Although metformin did not influence insulin sensitivity, it enhanced glucose-induced insulin secretion (p<0.05), and the increases in plasma insulin levels were comparable on the 3 days when metformin was given.
Conclusions/interpretation
In well-controlled metformin-treated type 2 diabetes, glucose-lowering by metformin is greater when it is given before, rather than with, enteral glucose, and this is associated with a greater GLP-1 response.
These observations suggest that administration of metformin before meals may optimise its effect in improving postprandial glycaemic control.
Trial registration
www.anzctr.org.auACTRN12621000878875
Funding
The study was not funded by a specific research grant.
From the article
Introduction
Metformin remains the recommended first-line pharmacotherapy for type 2 diabetes in most clinical guidelines, but its mode of action remains incompletely understood. An improved understanding of the mechanisms underlying glucose-lowering by metformin would provide the potential to refine its clinical application in the management of type 2 diabetes.
While a number of mechanistic studies have found that metformin lowers fasting blood glucose levels by suppressing hepatic glucose production [1,2,3], a substantial body of in vivo data suggest that much of the glucose-lowering action occurs at the level of the gastrointestinal tract, irrespective of its systemic bioavailability [3, 4]. In line with this concept, administration of metformin by the enteral route was found to be more effective in glucose-lowering than intravenous or intraportal administration [5]. Moreover, a delayed-release formulation of metformin with minimal systemic exposure was shown to be as effective as immediate- or extended-release formulations [6, 7]. Preclinical and clinical studies have uncovered several gastrointestinal effects of metformin, including stimulation of the incretin hormone glucagon-like peptide-1 (GLP-1) [8], slowing of gastric emptying [8], suppression of intestinal glucose absorption [9], inhibition of bile acid resorption [10, 11] and modulation of the gut microbiota [12]. The findings that blockade of GLP-1 signalling abolished metformin-induced suppression of hepatic glucose output in rodents [13], and markedly attenuated the effect of metformin in reducing the glycaemic response to a mixed meal in people with type 2 diabetes [14], attest to a major role for GLP-1 in mediating the glucose-lowering effect of metformin. GLP-1 modulates glucose homeostasis via pleiotropic actions [15, 16], including stimulation of insulin secretion, suppression of glucagon release [17], slowing of gastric emptying [18] and inhibition of appetite [19].
There is considerable evidence that strategies designed to lower postprandial glycaemia in type 2 diabetes by boosting GLP-1 secretion are preferably delivered before a meal [20,21,22]. For example, a ‘preload’ of whey protein consumed 30 min before a potato meal in individuals with type 2 diabetes was more effective in stimulating GLP-1 secretion, slowing gastric emptying and reducing the postprandial glycaemic excursion, when compared to its consumption with the meal [20]. In another study, a small amount of whey preload was shown to be sufficient to reduce postprandial glycaemia and augment glucose-lowering by the dipeptidyl peptidase-4 inhibitor vildagliptin in type 2 diabetes [21]. Routine advice is to ingest metformin with meals, with the rationale that this approach will minimise potential gastrointestinal adverse effects; however, this has not been shown to be the case [23]. The impact on postprandial glucose-lowering of altering the timing of metformin administration in relation to meals has been poorly characterised, although an open-label pilot study in five metformin-treated type 2 diabetes patients was indicative of improved glucose-lowering, enhanced GLP-1 secretion and slowed gastric emptying when metformin (1000 mg) was administered 30 min before the meal rather than with the meal [24].
We hypothesised that varying the timing of metformin administration before a standardised nutrient load would affect glucose-lowering, and, in this ‘proof-of-concept’ study, we evaluate the effects of metformin administered by intraduodenal perfusion at different time intervals (0, 30 or 60 min) before a standardised intraduodenal glucose load on plasma glucose, GLP-1 and insulin levels in metformin-treated participants with type 2 diabetes.
Discussion
This study shows that administration of metformin 30 and 60 min before an intraduodenal glucose load is more effective than its administration at the start of the glucose infusion in terms of both reducing glycaemia and augmenting GLP-1 release in participants with metformin-treated type 2 diabetes. Moreover, earlier metformin administration did not induce gastrointestinal symptoms in this group of participants. These observations suggest that the timing of metformin administration can potentially substantially influence its efficacy for postprandial glucose-lowering, and that the current recommendation (at least for the immediate-release preparation) that people with type 2 diabetes should take metformin with meals may compromise its benefits.
In the current study, the exposure of the small intestine to metformin at different time intervals before the nutrient load was precisely controlled by direct infusion via a nasoduodenal catheter. This model circumvented the confounding impact of gastric emptying, which exhibits a wide inter-individual variation in both healthy individuals and those with type 2 diabetes [29,30,31]. For the same reason, the glucose solution was infused directly into the duodenum at a standardised rate (12.56 kJ/min [3 kcal/min]) within the physiological caloric range of gastric emptying [29]. These methods allowed a highly accurate assessment of the impact of the timing of small intestinal exposure to metformin on the subsequent glycaemic, insulinaemic and GLP-1 responses. We studied participants with type 2 diabetes who were already using a stable dose of metformin in order to optimise the translational relevance of our findings.
In summary, in metformin-treated patients with well-controlled type 2 diabetes, metformin is more effective in glucose-lowering when given before, rather than with, enteral glucose, and this is associated with a greater GLP-1 response. These observations suggest that administration of metformin before meals may improve its capacity to improve postprandial glycaemic control.
Nyhetsinfo
www red DiabetologNytt
