Akhil Parulkar, M. Pendergrass, R. Granda-Ayala
Jan 2, 2001
Citations
2
Influential Citations
290
Citations
Quality indicators
Journal
Annals of Internal Medicine
Abstract
The thiazolidinediones are a new class of compounds for treatment of type 2 diabetes mellitus. The efficacy of these drugs in decreasing plasma glucose levels is well established (1-6). Troglitazone, a member of this class, became available for clinical use in the United States in 1997 but was withdrawn in March 2000 because of reports of severe hepatic injury. Rosiglitazone and pioglitazone became available in 1999 and are approved as monotherapy and in combination with other oral hypoglycemic agents; pioglitazone is also approved in combination with insulin. The glucose-lowering effects of the thiazolidinediones are mediated primarily by decreasing insulin resistance at the level of the muscle and thereby increasing glucose uptake. To a lesser extent, they decrease insulin resistance in the liver and thereby decrease hepatic glucose production (1). The mechanisms of action of the thiazolidinediones are still being investigated; however, some of their actions are mediated through binding and activation of the peroxisome proliferatoractivated receptor- (PPAR-), a nuclear receptor that has a regulatory role in differentiation of cells, particularly adipocytes (7). This receptor is also expressed in several other tissues, including vascular tissue (7). The thiazolidinediones also decrease plasma concentrations of free fatty acids and in doing so may indirectly improve insulin sensitivity (8). Thiazolidinediones may also activate other members of the PPAR family of receptors (such as PPAR- and PPAR-), which along with PPAR serve as gene transcription factors (9). These receptors are present in many tissues, and although their functions are still being elucidated, the data suggest that they have many important effects. Thus, the thiazolidinediones may affect many organ systems and disease processes (9). Substantial evidence indicates that insulin resistance, along with compensatory hyperinsulinemia, not only contributes to hyperglycemia in type 2 diabetes but also may play a pathophysiologic role in other metabolic abnormalities. These include high levels of plasma triglycerides, low levels of high-density lipoprotein (HDL) cholesterol, hypertension, abnormal fibrinolysis, and coronary heart disease (10-13). This cluster of abnormalities is called the insulin resistance syndrome, syndrome X, or the metabolic syndrome (14). Since thiazolidinediones directly improve insulin resistance (2), it has been proposed that they may correct other abnormalities of the insulin resistance syndrome, as well as improve hyperglycemia. Thus, use of these agents to treat patients with type 2 diabetes may confer benefits beyond decreases in glucose level. Because of the effects of thiazolidinediones on hyperinsulinemia and insulin resistance, their vascular effect is a subject of considerable research interest (Table 1). Table 1. Effects of the Thiazolidinediones on Cardiovascular Risk Factors We discuss the nonhypoglycemic effects of the thiazolidinediones that have been described in the literature. We emphasize their potential to improve other components of the insulin resistance syndrome, such as dyslipidemia, hypertension, impaired fibrinolysis, and atherosclerosis. We discuss their effects in other insulin-resistant states, such as the polycystic ovary syndrome; examine their effects on body weight and composition; and draw attention to other potential effects currently being investigated. Much of the data presented relates to troglitazone (the most extensively studied thiazolidinedione) but may be relevant to the other thiazolidinediones. In vitro data support the possibility of a class effect, although proof from currently ongoing clinical trials (Freed M, SmithKline Beecham; Wishner W, Takeda Pharmaceuticals. Personal communication) is needed. Long-term clinical trials are also needed to determine whether such reduction in risk factors will prevent cardiovascular disease. Cardiovascular Effects Epidemiologic studies have demonstrated that hyperinsulinemia, a marker for insulin resistance, is an independent risk factor for cardiovascular disease (47). Correction of insulin resistance may be clinically important in type 2 diabetes and may decrease risk for cardiovascular disease. In the United Kingdom Prospective Diabetes Study, treatment with metformin (another drug that decreases hyperinsulinemia and insulin resistance) was shown to produce greater reduction in cardiovascular disease events and mortality than sulfonylureas and insulin (48). The latter drugs decreased blood glucose level to a similar degree as metformin but did not decrease plasma insulin concentrations. This effect may have been mediated through a decrease in insulin resistance, although other effects of metformin, such as improvement in lipid profile, improved fibrinolysis, and prevention of weight gain, may be important (48). Further clinical trials are needed to determine whether treatment of diabetes with agents that reduce insulin resistance (such as the thiazolidinediones and metformin) is superior to use of agents that stimulate insulin secretion (such as sulfonylureas). The National Institutes of Health recently initiated such a clinical trial (49). Cardiac Output and Left Ventricular Mass Initial studies of cardiac function with thiazolidinedione therapy were performed because of reported cardiac enlargement in animals treated with drugs of this class (3). Ghazzi and colleagues (3) investigated whether patients with type 2 diabetes treated with troglitazone, 800 mg/d (a dosage higher than that used in clinical practice), or glyburide experienced an increase in cardiac mass or functional impairment. Two-dimensional echocardiography and pulsed Doppler ultrasonography demonstrated that neither troglitazone nor glyburide changed left ventricular mass index significantly over 48 weeks. However, substantial increases in stroke volume index and cardiac index and a statistically significant decrease in diastolic pressure and estimated peripheral resistance were observed in troglitazone-treated patients but not glyburide-treated patients, who experienced no change (3). These findings are reassuring and contrast with those of animal studies. Similar studies of rosiglitazone and pioglitazone have also demonstrated no adverse effect on cardiac mass or function (4, 5, 50). Nevertheless, thiazolidinediones are currently contraindicated in patients with advanced heart failure because of their effect on plasma volume (43). Lipid Metabolism and Oxidation Insulin resistance and type 2 diabetes are associated with a characteristic pattern of lipid abnormalities, including an elevated plasma triglyceride level and a low plasma high-density lipoprotein (HDL) cholesterol level. Plasma levels of low-density lipoprotein (LDL) cholesterol do not differ from those in nondiabetic persons, but qualitative changes in LDL cholesterol, with an increase in small, dense LDL cholesterol, are common (15, 51-55). In several clinical trials, troglitazone therapy significantly lowered triglyceride levels (3, 16, 17) and increased HDL cholesterol levels in persons with type 2 diabetes (3, 16). A modest increase in LDL cholesterol level was observed, but the ratio of LDL cholesterol to HDL cholesterol and apolipoprotein B levels did not change. Data published to date indicate that all of the thiazolidinediones increase HDL cholesterol levels and that troglitazone and pioglitazone decrease triglyceride levels (3-5, 16-18). A recent small study (19) demonstrated a possible difference in the lipid-lowering effects of thiazolidinediones: Compared with rosiglitazone, pioglitazone seemed to produce a greater decrease in the triglyceride level and a lesser increase in the LDL cholesterol level. However, the study was neither randomized nor double-blind. Further studies on such possible differences are therefore needed. Similar changes in lipid levels have been observed in nondiabetic persons with insulin resistance. Like troglitazone, pioglitazone significantly reduced fasting serum levels of triglycerides and increased fasting levels of HDL cholesterol in 20 patients with type 2 diabetes (56). The effects of the thiazolidinediones on LDL cholesterol are more complex. Persons with insulin resistance or type 2 diabetes are more likely than nondiabetic persons to have small, dense, triglyceride-rich LDL cholesterol particles (51). These characteristics may make LDL cholesterol susceptible to oxidation. Oxidative modification confers atherogenic properties on LDL cholesterol particles; this may be a key initial event in the progression of atherosclerosis (15) and is a measurable risk factor (52-55). Evidence suggests that PPAR- may be an important regulator of foam-cell gene expression and that oxidized LDL cholesterol regulates macrophage gene expression through activation of PPAR- (57). Furthermore, PPAR- promotes uptake of oxidized LDL cholesterol by macrophages (58). Thus, an interaction between PPAR- and oxidized LDL cholesterol may be important in the development of atherosclerosis in diabetes. Thiazolidinediones have been shown to substantially increase levels of total cholesterol and LDL cholesterol (4, 5). However, the increase is predominantly in the larger buoyant particles of LDL cholesterol, which may be less atherogenic than small, dense LDL cholesterol particles. Levels of the latter have been shown to decrease with troglitazone therapy (20). These data were confirmed in other studies demonstrating that troglitazone increased the resistance of LDL cholesterol to oxidation (20-24). Whether these effects are produced by the other thiazolidinediones or were related to vitamin E moiety in the troglitazone molecule is unclear and warrants further study. Although the effects of the thiazolidinediones on LDL cholesterol oxidation are in theory appealing, their role in preventing cardiovascular events is unclear. Of note, vitamin E, which also has antioxidant and free radicalscavenging prope