Effect On Carbohydrate Metabolism

The Big Diabetes Lie

Home Treatment of Diabetes

Get Instant Access

When AGIs are given orally, they reduce the digestion of carbohydrates in the upper half of the small intestine, so that a larger proportion is digested in the lower part and in the colon (Fig. 2). The rise in postprandial hyperglycemia is immediately diminished when AGIs are taken with the first bites of a meal (20). The amount of carbohydrate reaching the colon, and the alpha-glucosidase activity in the lower small intestine, determines the frequency and severity of gastrointestinal side effects, such as meteorism, flatulence, and diarrhea, due to fermentation gases and short-chain fatty acids (21). The quantity of undigested carbohydrates reaching the colon can be determined by measurement of breath hydrogen. The therapeutic effects, as well as side effects, therefore strongly depend on the amount and type of carbohydrates in the diet. It has been shown that acarbose is more effective in a diet rich in starch, because it has its strongest effect on glucoamylase (22).

There is a great variety of individual and racial intestinal enzyme patterns, which may explain the striking differences in efficacy and acceptance in different areas, and among different population groups. In Asia with a nutrition rich in complex carbohydrates (rice) AGIs are frequently used as first-line oral antidiabetics with few gastrointestinal side effects. With western nutrition habits low in starch and crude fibers—only small amounts of undigested carbohydrates reach the lower part of the small intestine. Unadapted exposure of the ileum and colon to greater amounts of these undigested carbohydrates after administration of AGIs leads to the side effects listed in Figure 3. Controlled studies (23) have shown that side effects can be minimized by dose titration, starting with doses of 25 mg of acarbose or miglitol twice a day. Over 1 to 3 months, the alpha-glucosidase content of the lower part of the small intestine increases, and the frequency of gastrointestinal side effects reduces (24).

FIGURE 2 Mode of action of acarbose on postprandial glucose excursion.

Effect on Insulin and Enterohormones

There is no evidence of any direct effect of AGIs on insulin secretion and action. However, an improvement in insulin sensitivity may be achieved by control of postprandial hyperglycemia, which protects the beta-cells of the pancreas. This is supported by the results of the STOP-NIDDM trial (25) and 5-year follow-up data on clinical type 2 diabetes, which reveals an increasing efficacy in reduction of glycosylated hemoglobin (HbA1c) over time (26). Three studies (two with acarbose (27,28) and one with voglibose (29)) in subjects with IGT directly measured insulin resistance and found an increase in insulin sensitivity. However, in type 2 diabetes this effect was only marginal, and was without statistical significance in investigations with CLAMP (30,31).

A recently published paper on elderly type 2 diabetes presents data by HOMA that indicates an improvement in insulin sensitivity (32). Consistent data from all three AGIs have shown a reduction in postprandial insulin excursion lasting > 3h (33,34). So far, no data are available with respect to impact on the early insulin secretion phase. Two studies in IGT (27,28)

FIGURE 3 Pathogenesis of gastrointestinal side effects after intake of AGIs; relationship of dosage and time after start of therapy. Source: From Ref. 44.

and one with 24 h profiles in type 2 diabetes (35), show a reduction in proinsulin levels postprandium, which may be indicative of an improved beta-cell function, following protection of the beta-cells from postprandial glucose spikes.

Inhibition of carbohydrate digestion by AGIs in the upper part of the small intestine affects release of two gut hormones: gastric inhibitory polypeptide (GIP) and GLP1. GIP is produced in the duodenum and upper jejunum, dependent on transmembranal glucose transport (36); it stimulates gastric emptying. The decrease in postprandial glucose absorption by AGI intake causes a decrease in GIP release after a carbohydrate-rich meal, which leads to a slower gastric emptying supporting the action of AGIs on postprandial glucose rise (37). The impact on GLP1 release seems to be even more important (36,37). GLP1 is mainly produced in the cells of the ileum and colon. AGIs trigger a long-lasting increase in GLP1 secretion in the late postprandial phase (60 to > 240 min) when glucose excursion is already back at baseline. It is postulated that long-lasting increase in GLPa via the enteroinsular axis may support the therapeutic effects of AGIs (38,39).

Effect on Energy Balance and Components of the Metabolic Syndrome

AGIs delay the release of glucose but cause no malabsorption. In long-term studies (40,41) an average reduction in body weight of 0.7 to 0.9 kg was observed. However, no significant changes in eating pattern and energy intake were seen (22,41). In healthy volunteers, treatment with miglitol caused no significant loss of carbohydrates, proteins, and fat measured in the feces (15). The small weight loss registered in most long-term observations may therefore be a secondary phenomenon resulting from improved insulin sensitivity. AGIs have little or no effect on low-density lipoprotein (LDL)- and high-density lipoprotein (HDL)-cholesterol levels (42). However, recently published investigations in people with IGT treated with acarbose revealed a decrease in small dense LDL together with a lower lag time what is indicative of an ameliorated atherogenecity of LDLs (43).

The major effect is on fasting and postprandial triglycerides (44), with a reduction of about 15%. Furthermore, elevated LDL-cholesterol was lowered with acarbose (42) and with miglitol as adjuvant to sulfonylurea treatment (45). These weak effects on dyslipidemia may also be secondary results of improved insulin sensitivity.

In a placebo-controlled study on insulin resistance in obese hypertensive subjects with normal glucose tolerance the HOMA index in the acarbose group declined from 5.36 to 4.10 (p=.001) without a significant change in body mass index (BMI). This strong effect on insulin resistance may also be explained by reduction in postprandial glucose excursion (46).

In the STOP-NIDDM study in people with IGT the incidence of newly diagnosed hypertension was reduced by 34% versus placebo. In the MERIA meta-analysis systolic blood pressure decreased by 2.7mmHg (p=.024). The effect of acarbose was compared to long-acting insulin secretagogue glibenclamide in type 2 diabetes patients with mild hypertension using automatic 24 h blood pressure measurements. After 6 months systolic blood pressure was lowered by 5.2mmHg with acarbose versus 1.6mmHg with glibenclamide (p < .01) whereas glibenclamide was more effective in lowering diastolic blood pressure. These results indicate that acarbose has beneficial effects on elevated systolic blood pressure. There are some new results that show that acarbose may have anti-inflammatory potentials. Significant reductions of hsCRP in people with IGT treated with acarbose have recently been published. Furthermore a decrease in fibrinogen after treatment with acarbose has been reported.

Thus by extrapolation AGIs, with best evidence for acarbose exhibit beneficial effects on all components of the metabolic syndrome and low-grade inflammation in patients with prediabetes and type 2 diabetes.


AGIs have been in clinical use for 10 years and are now registered worldwide. They are among the best-studied oral antidiabetics, with data from controlled studies and long-term clinical investigations for all three clinically used compounds. AGIs are used as first-line drugs in early type 2 diabetes, as well as in combination with nearly all established oral antidiabetics and insulin. In some cases of type 1 diabetes, with rapid postprandial glucose rise, and in cases of premeal hypoglycemia, AGIs may be introduced as adjunct therapy (36).

AGIs in Type 2 Diabetes Insufficiently Treated with Lifestyle Improvement

Acarbose (47,48) and miglitol (49) have been studied in drug naive patients in multinational European- and US-trials in dosages of 25 to 600 mg three times a day (Table 3). These studies have shown a dose-response relationship for acarbose and miglitol between 25 and 200 mg three times a day, with a plateau at 50 and 100 mg, with respect to both postprandial hyperglycemia and HbA1c. However, side effects strongly increase at dosages of > 100 mg three times a day. Acarbose at a dosage of 100 mg seems to have more gastrointestinal side effects than miglitol at the same dosage (49).

Efficacy studies show that AGIs mainly act on postprandial hyperglycemia, with an effect on fasting plasma glucose occurring after 8 to 12 weeks. In clinical practice, glucose monitoring should therefore include measurements 2 h after major meals. In a meta-analysis of 13 controlled clinical trials with acarbose, the mean reduction in fasting glucose was 24 - 7.2 mg/dL, in postprandial glucose it was 54 - 15.8mg/dL and in HbA1c 0.90% -0.25%. The efficacy of therapeutic doses of miglitol is in the same range, with a somewhat higher effect on HbA1c at a dosage of 50 mg three times per day, versus the same dose of acarbose (50). Fewer data are available for voglibose. Comparative studies with other oral antidiabetics show a weaker effect on HbA1c than for metformin (34), except for one study that showed a similar efficacy (46). Except in one publication (51), a stronger effect of tolbutamide (52) and glibenclamide at 24 to 56 weeks follow-up has been consistently shown (36). It is a consistent finding that metformin and the sulfonylureas were more effective on fasting blood glucose control, whereas the AGIs were superior in the control of postprandial hyperglycemia. No reliable data have been published comparing the "prandial" insulin-secretagogues repaglinide and nateglinide with AGIs in face-to-face investigations.

A Cochrane review of 22 placebo-controlled studies with 1895 type 2 patients found a 0.78 (95% CI—0.93 - 0.63) overall reduction in HbA1c. Compared to glibenclamide the efficacy was equal. Only one study reported a face-to-face comparison with metformin (53).

Combination Therapy of AGIs With Oral Antidiabetics and Insulin in Type 2 Diabetes

AGIs are frequently used as add-on therapy in patients insufficiently treated with sulfonylurea or metformin monotherapy. Less data exist on add-on therapy with AGIs as the first-line

TABLE 3 Dose-Response of Efficacy of Acarbose and Miglitol on Plasma Glucose (PG) and HbAic after 24 Weeks Treatment

Change (%)

Fasting PG

1 h postprandial PG

2 h postprandial PG



Was this article helpful?

0 0
Weight Loss All Star

Weight Loss All Star

Are you looking to lose weight or even just tone up? What is stopping you from having the body you want and help you feel great at the same time? I created Weight Loss All-Star for all those wanting to lose weight, and keep the weight off. I know how hard it is to do diets and stick with them, and get the motivation to get up and exercise.

Get My Free Ebook

Post a comment