Sulfonylurea drugs have been used in the treatment of type 2 diabetes since the early 1950s. Extensive literature, including many review articles, is available (13,45,47). The major new information concerning the use of sulfonylureas in the treatment of type 2 diabetes has come from the UKPDS (12,52,53,54), the development programs of the newer sulfonylureas such as glimepiride (55), and recent studies investigating the effects of sulfonylureas on myocardial and vascular responses to ischemia (56,57,58,59).

The sulfonylurea drugs do not appear to correct the defect in early insulin secretion characteristic of type 2 diabetes (60,61). Their primary action is to increase the late stage of insulin secretion. This increases the likelihood of late postprandial and fasting hypoglycemia. The sulfonylurea drugs do not increase insulin biosynthesis and, indeed, seem to inhibit proinsulin biosynthesis in vitro. There is no evidence that they are either β-cytotropic or that they facilitate β-cell exhaustion (12,47). In the UKPDS, patients treated with sulfonylureas showed an increase in β-cell function for the first year but thereafter showed the same rate of loss of β-cell function as the conventionally treated group (12). Questions have been raised concerning the possibility that sulfonylureas, because of their prolonged effects on the pancreatic β-cell K_ATP channels, could lead to desensitization, with diminution of their pharmacologic effects. Several clinical studies showing that higher doses of sulfonylureas may be less effective than modest doses in lowering hyperglycemia provide some support for this hypothesis (62,63). Continuous in vitro incubation of β-cells with glyburide has been shown to lead to an increase in functionally deficient K_ATP channels (64).

Previous studies have shown that sulfonylureas can exert a variety of extrapancreatic actions in laboratory models in vivo and in vitro (47). It has been difficult to demonstrate conclusively any of these effects in humans. Some in vivo studies in humans have shown that sulfonylureas can lower plasma glucose levels under conditions in which the peripheral plasma insulin levels are unchanged. These studies have not excluded small changes in the portal vein insulin concentrations, which influence the liver but are not measurable in peripheral blood. Two effects of sulfonylureas that occur in humans that may result in extrapancreatic effects are interactions with K_ATP channels in other tissues (39) and direct effects on calcium ion-mediated exocytosis (65).

K_ATP channels in cardiac myocytes and vascular smooth muscle cells have SUR regulatory subunits, which differ somewhat from the SUR-1 subunit in β-cells (38,39,40). These different subunits change their binding characteristics such that they may have less affinity for some ligands that cause insulin secretion than for others. Brain-cell plasma membranes contain K_ATP channels that have the same regulatory subunit as pancreatic β-cells. It is therefore possible that some sulfonylureas might interact with K_ATP channels in tissues other than β-cells and cause extrapancreatic effects. Another mechanism by which specific sulfonylureas might cause extrapancreatic effects is through direct enhancement of calcium-mediated exocytosis of other hormones that are stored in secretory granules. Presently, there is little evidence that sulfonylureas exert any clinically significant extrapancreatic effects in humans. The only exception may be effects on the myocardial and vascular smooth muscle responses to ischemia and hypoxia.

Chlorpropamide is a first-generation sulfonylurea that has been prescribed extensively worldwide for almost 40 years. At dosages ranging from 100 to 500 mg once a day, it is highly effective in reducing hyperglycemia, but because of its very long plasma half-life, it has been associated with a significant incidence of severe and protracted hypoglycemia, particularly in the elderly (47,66). Other significant side effects are water retention with hyponatremia and alcohol-induced facial flushing (45). Chlorpropamide is unique among sulfonylureas in that it stimulates secretion of antidiuretic hormones and potentiates an antidiuretic hormone effect in the renal tubules. The effect on water balance is thought to explain the observation in the UKPDS of an increase in systolic and diastolic blood pressure as well as a greater need for antihypertensive therapy in the 619 patients treated with chlorpropamide than in patients treated with insulin, glyburide, or conventional therapy (2). In the UKPDS, chlorpropamide treatment reduced HbA_1c significantly more than did glyburide or insulin; however, this was not accompanied by any greater reduction in risk for progression of retinopathy (2). Many previous studies have noted that hypoglycemia is a major risk of chlorpropamide treatment. Major hypoglycemic events occurred less frequently in the UKPDS in the chlorpropamide-treated cohort than in the glyburide-treated cohort (2).

Glyburide (glibenclamide) is currently the most widely prescribed sulfonylurea—but not because it is more effective or safer than other sulfonylureas. It is marketed as a formulation with poor and variable bioavailability (dosage 1.25–20 mg daily) and as a micronized formulation with relatively good and consistent bioavailability (dosage 1.5–12 mg daily) (45,47,67). Its efficacy in reducing hyperglycemia appears to be equal to that of other sulfonylureas, but it is associated with a high rate of serious side effects, including its long duration of action, with a rate of serious and fatal hypoglycemia that is significantly higher than that with other sulfonylureas, modest weight gain, and a relative lack of specificity on the different K_ATP channels (67,68,69). Careful evaluation of the 615 patients with type 2 diabetes treated with glyburide for a mean of 11 years in the UKPDS has clearly defined the characteristics of chronic glyburide therapy (2). It is most effective in controlling hyperglycemia in patients with newly diagnosed type 2 diabetes for the first year or 2 and then begins to lose effectiveness progressively with time (secondary failure or sulfonylurea ineffectiveness) (52,53,54). Glyburide treatment alone was relatively ineffective in controlling hyperglycemia after 4 or 5 years (54). The median decrease in HbA_1c in patients treated with glyburide as compared with patients receiving conventional (diet) treatment over 10 years was 0.7% (7.2% vs. 7.9%) (2). Weight gain occurred primarily in the first 3 or 4 years of treatment and was maintained throughout the study, so that at 10 years the glyburide-treated patients had gained 1.7 kg more than the conventionally treated patients (2). Serious hypoglycemia occurred primarily during the first few years of glyburide therapy, when it had been most effective in reducing hyperglycemia. Major hypoglycemic events were noted in approximately 1.4% of patients in year 1, 1.3% in year 2, and 0.4% in year 3 (2). These rates of serious hypoglycemic events are comparable to those reported in other large series and are greater than those with other sulfonylureas (Table 41.5) (68,70,71,72).

Glipizide is a sulfonylurea that is available in both a short-acting formulation and in an extended-release formulation (glipizide- g astro i ntestinal t herapeutic s ystem, or glipizide- GITS) (45). The short-acting form is rapidly absorbed and is completely bioavailable. It is metabolized by oxidative hydroxylation and is eliminated with a half-life of 2 to 4 hours (45).
Its elimination is not affected by mild to moderate renal insufficiency (creatinine clearance ≥30 mL/min) (45). Glipizide is administered once or twice a day. Its efficacy is equal to that of glyburide in controlling hyperglycemia, but because of its metabolism and more rapid elimination, therapy with glipizide is associated with less hypoglycemia (45). The official maximum dosage of glipizide in the United States is 40 mg per day, although available clinical studies show little or no benefit in giving more than 15 or 20 mg per day. Glipizide should be administered before meals, as its absorption may be mildly delayed by food ingestion.

The extended-release formulation of glipizide provides once-a-day dosage of 5 to 20 mg (73,74). Plasma levels of glipizide 24 hours after doses of 5 mg and 20 mg of glipizide-GITS (Glucotrol XL) are 54 ng/mL and 310 ng/mL, respectively. The decreases in HbA_1c in clinical trials with glipizide-GITS were 1.50% and 1.84% compared with placebo treatment. The decrease in HbA_1c was maximal at 5 mg per day. FPG level was reduced 57 mg/dL to 74 mg/dL compared with placebo treatment. The extended-release form appears to have glucose-lowering effects similar to those of immediate-release glipizide, as measured by the improvement in HbA_1c. In a study comparing glipizide-GITS to immediate-acting glipizide, the authors concluded that the glipizide-GITS formulation at dosages of 5 mg and 20 mg have an improved metabolic profile compared with that of immediate-acting glipizide; however, the design of the study makes such a conclusion tenuous. No increase in hypoglycemia was noted with the extended-acting form.

Glimepiride is the newest of the sulfonylureas to be approved for use. It is said to be more selective for the β-cell K_ATP channel than for the cardiovascular tissue K_ATP channel (75). It appears to bind to a slightly different part of the sulfonylurea-binding site than does glyburide, although the two sulfonylureas displace each other from their respective binding sites. Glimepiride associates 2.5- to 3.0-fold faster and dissociates 8.0- to 9.0-fold faster than glyburide from its β-cell K_ATP-binding site (75,76). Insulin release is therefore more rapid and of shorter duration than that with glyburide. Hyperglycemic and euglycemic-hyperinsulinemic clamp studies in patients with type 2 diabetes demonstrated that glimepiride does not restore first-phase insulin secretion and that it increases second-phase insulin secretion, whole-body glucose uptake, and increases insulin sensitivity (61). Some studies have suggested that glimepiride reduces hyperglycemia with less insulin secretion than that required by other sulfonylureas, and this has led to the claim that glimepiride has an insulin-sparing action (57,77,78,79,80). The data are marginally significant, and additional studies are required to substantiate this claim. For glimepiride, as for all other sulfonylureas, there are claims that it has extrapancreatic effects. Some of the effects described are an increase in glucose uptake and utilization by an increase in the translocation of the GLUT4 transporter protein in adipose and muscle cells, an effect on the regulation of the intracellular routing of the insulin receptor complexes toward degradative pathways, and an increased association of the insulin receptor with protein kinase C isozymes (76,79). These extrapancreatic effects have been demonstrated in in vitro systems, and their relevance to glimepiride effects in humans is debatable.