Do the Metabolic Effects of β Blockers Make Them Leading or Supporting Antihypertensive Agents in the Treatment of Hypertension?

Abstract

Reduction of blood pressure to guideline goals (i.e., <130/80 mm Hg) in persons with diabetes is crucial to optimally reduce cardiovascular events and kidney disease progression. Since many patients will be >20/10 mm Hg above this goal, most guidelines recommend using agents that block the renin–angiotensin system in concert with a thiazide‐like diuretic to achieve goal blood pressure. Meta‐analyses of clinical trials indicate that while all classes of antihypertensive agents reduce cardiovascular risk, they exert different effects on glucose utilization and lipids and, hence, may affect morbidity. Specifically, β blockers, in general, worsen insulin resistance and increase triglycerides in a dose‐dependent fashion. Moreover, they are not recommended as initial therapy for hypertension treatment in the absence of heart failure or recent myocardial infarction, especially in the elderly. Recent studies support the notion that newer β blockers with vasodilating effects have a better metabolic profile when compared with those that purely affect β receptors. Thus, vasodilating β blockers, by being neutral on glycemic and metabolic factors, are associated with less use of additional medication for lipid or glucose control and may provide a potentially greater cardiovascular risk reduction by virtue of these effects.

Hypertension and type 2 diabetes mellitus are major cardiovascular risk factors that commonly cluster in the same individual in the context of the metabolic syndrome. Achieving the recommended blood pressure guideline goal of <130/80 mm Hg in those with diabetes is possible with the use of any major antihypertensive class. ¹ In general, however, combinations of antihypertensive agents are needed if blood pressure is >20/10 mm Hg above this goal. Most guidelines recommend blockers of the renin‐angiotensin‐aldosterone system in concert with thiazide‐like diuretics as initial therapy for most people who require two‐drug therapy. ² , ³

Results of a recent meta‐analysis of clinical trials performed over the past decade indicate that while different classes of antihypertensive agents all reduce cardiovascular risk, they exert different effects on glucose utilization and, hence, may affect morbidity. Specifically, drugs that block the renin‐angiotensin‐aldosterone system, such as angiotensin‐converting enzyme (ACE) inhibitors or angiotensin receptor blockers and calcium channel blockers have neutral or beneficial effects on carbohydrate and lipid metabolism, whereas thiazide diuretics and β blockers worsen insulin resistance in a dose‐dependent fashion. ⁴ , ⁵ , ⁶ In spite of this adverse effect on glycemic and metabolic control, clinical trials of 3‐ to 5‐years' duration that used a β blocker demonstrate a reduction in cardiovascular events, especially in patients with diabetes following a myocardial infarction or heart failure. ⁷ , ⁸ , ⁹ , ¹⁰ The United Kingdom Prospective Diabetes Study (UKPDS) ¹¹ clearly demonstrates the cardiovascular benefits of a β blocker compared with an ACE inhibitor in patients with hypertension and type 2 diabetes. In this trial, as well, the metabolic profile did not favor the β blocker, yet there was no difference in the cardiovascular outcomes after a follow‐up period of 8 years. This is why there is a compelling indication for their use in those with diabetes. ² , ³ , ¹¹ , ¹² Moreover, all the older trials required two or more drugs to reduce blood pressure to <140/90 mm Hg, and thiazide diuretics were used as a second agent in all such trials. It should be noted that in these trials, there was a worsening of glucose and lipid profiles when they were measured; however, a benefit on cardiovascular outcome was still noted.

Recent studies support the notion that newer β blockers with vasodilating effects have a better metabolic profile when compared with those that purely affect β receptors. Thus, vasodilating β blockers, such as carvedilol, by being neutral on glycemic and metabolic factors, may result in lower morbidity with a lower incidence of new‐onset diabetes. Such agents are associated with less use of additional medication for lipid or glucose control and may provide a potentially greater cardiovascular risk reduction by virtue of these effects. ¹³ , ¹⁴ , ¹⁵ , ¹⁶ However, there is only one prospective comparator trial of different β blockers, and this trial had no cardiovascular end points. ¹⁶ Thus, the potential additional benefit of the vasodilating β blockers on cardiovascular outcome, in spite of their favorable metabolic effects, remains to be proven.

The cardiovascular benefit of many β blockers, i.e., propranolol, atenolol, metoprolol, and others, is associated with detrimental effects on insulin sensitivity, glycemic control, and incidence of type 2 diabetes. ⁵ , ¹⁵ , ¹⁷ Clinical evidence, however, supports the notion that β blockers with vasodilating effects are associated with far fewer adverse metabolic effects and, hence, are a valuable tool for hypertension treatment in patients with the metabolic syndrome. ¹⁵ , ¹⁶ This review summarizes the clinical evidence on β blockers and their effects on carbohydrate metabolism, i.e., glucose control, insulin sensitivity, and development of new‐onset diabetes. We discuss the comparative effects of both traditional and vasodilating β blockers on carbohydrate metabolism in the context of clinical management culminating with suggestions for use in practice.

INSULIN SENSITIVITY AND RISK OF NEW‐ONSET DIABETES

In clinical studies that have used the euglycemic hyperinsulinemic clamp technique (the most reliable available method to estimate insulin sensitivity), treatment with conventional β blockers, either non‐selective like propranolol ⁴ or β₁‐selective like atenolol ¹⁸ , ¹⁹ or metoprolol, ¹³ , ¹⁸ significantly decrease insulin sensitivity in hypertensive patients. This decrease in insulin sensitivity by these β blockers would be expected to worsen glycemic control in patients with diabetes and impaired glucose tolerance. Conversely, in those without diabetes, a decrease in insulin sensitivity would not result in elevation of blood glucose levels as long as the pancreatic β cells secrete the necessary amount of insulin.

One hypothesis is that a higher insulin demand accompanies the process of aging and increases in body weight. When the pancreatic β cells can no longer compensate for the increasing insulin resistance, impaired glucose tolerance and ultimately diabetes appear. ²⁰ This hypothesis is supported by studies that examine the incidence of new‐onset diabetes in the presence of β blockers. The Atherosclerosis Risk in Communities (ARIC) ²¹ cohort study demonstrated that after more than a 6‐year follow‐up of 3804 hypertensive subjects treated with antihypertensive therapy, those who received β blockers had a 28% higher risk of type 2 diabetes compared with those taking no medication or other antihypertensive agents, including thiazide diuretics. Many other prospective trials have new diabetes outcomes similar to those in the ARIC study and are summarized in the Table. ⁵ Note, however, that while all these trials show worsening or development of new‐onset diabetes in the short term, they all also show mortality reduction comparable to the comparator therapy.

A number of methodologic issues limit the conclusions extrapolated from these studies. The limitations include the fact that: 1) none of the studies published to date examine diabetes incidence as a primary end point ¹² , ²¹ , ²² , ²³ , ²⁴ ; 2) more than half the participants in all the trials received a second agent that could also influence glycemic control. ²³ , ²⁴ , ²⁵ (In some of these studies, an ACE inhibitor ²³ , ²⁶ or a calcium channel blocker ²⁷ was compared with diuretics or β blockers in various combinations; thus, a net effect of the latter cannot be easily assessed); and 3) detection bias could have occurred in these studies, in that some of the studies were open‐label with blinded end point evaluation. ¹¹ , ²³ Thus, the search for a diagnosis of diabetes may have been more intensive in those who received β blockers, based on an increased incidence from previous studies.

EFFECTS OF β BLOCKERS: INSULIN SENSITIVITY AND GLYCEMIC CONTROL

There are several mechanisms described that are possibly responsible for the metabolic effects of β blockers. In healthy persons, insulin yields vasodilation and increases blood flow to the skeletal muscles, an action tightly coupled with an increase of glucose disposal in the same tissue. ²⁸ In contrast, in persons with insulin resistance, endothelium‐dependent insulin‐mediated vasodilation is impaired; this impairment is considered an important cause of reduced insulin‐stimulated glucose uptake in the periphery. ²⁸ , ²⁹ Conversely, acute sympathetic nervous system stimulation in healthy individuals lowers insulin‐stimulated glucose uptake in muscles through vasoconstriction and blood flow reduction. ³⁰ , ³¹ This effect is mediated by α₁‐adrenergic pathways, as evidenced in studies with direct α‐ and β‐blockade, ³² and further supported by the vasodilating and insulin‐sensitizing effects of α‐adrenergic blockers. ³³ Thus, treatment with conventional β blockers, an unopposed α₁ activity, would cause vasoconstriction and decreased blood flow to muscles. ³⁴ , ³⁵

Treatment with β blockers also affects insulin secretion from pancreatic β cells. In particular, nonselective β blockers have been found to decrease the first phase of insulin secretion, possibly due to impairment of β₂‐mediated insulin release. ¹⁸ , ³⁴ , ³⁶ The attenuation of first‐phase insulin secretion represents a crucial step in the natural history of type 2 diabetes and has been suggested to be an important predictor of the disease. ³⁷ Hence, this action of β blockers could be a very important contributor to the development of type 2 diabetes.

Weight gain has been proposed as another contributor to decreased insulin sensitivity associated with β blockers, as these agents increase body weight. ¹¹ , ¹⁶ However, weight gain does not appear to be a major contributor to worsening of insulin resistance, as it failed to independently predict worsening of glycemic control in a number of studies, including a large long‐term study of more than 15,000 people. ¹⁸ , ²¹

Another mechanism that contributes to insulin‐resistance is the inability of insulin to suppress hepatic glycogenolysis, which leads to elevated hepatic glucose production after meals and, consequently, contributes to loss of glycemic control. ³⁷ , ³⁸ Sympathetic activation stimulates glyconeogenesis and glycogenolysis and inhibits glycogen synthesis in the liver. Although the relative importance of α‐ and β‐adrenergic receptors in mediating catecholamine‐induced hepatic glucose production in humans in vivo is unclear, in rats, α₂ receptors are involved. ³⁹ Thus, if α receptors play a major role in humans, unopposed α activity in the presence of β‐blockade could result in enhanced hepatic glucose output, increasing the risk for type 2 diabetes.

VASODILATING β BLOCKERS, INSULIN SENSITIVITY, AND GLYCEMIC CONTROL

Studies of β blockers with vasodilating effects, such as carvedilol and celiprolol, have neutral or beneficial effects on glycemic control and insulin sensitivity. Several years ago, dilevalol, a β₁‐selective blocker with β₂‐agonistic action, was found to improve insulin sensitivity by about 10% in hypertensive patients; ¹⁷ however, this compound was withdrawn from the market due to side effects. Celiprolol, a similar compound with better tolerability that is marketed in Europe, is also associated with a 35% improvement in insulin sensitivity following a year of treatment. ¹⁴

Carvedilol, a nonselective β blocker with α₁‐blocking and antioxidant effects, also improves insulin sensitivity. ¹³ In two separate studies that compared carvedilol with either metoprolol or atenolol, an improvement in insulin sensitivity was observed after a 3‐ and 6‐month time period, respectively. ¹³ , ¹⁵ The results of these small studies were corroborated by the results of the Glycemic Effects in Diabetes Mellitus: Carvedilol‐Metoprolol Comparison in Hypertensives (GEMINI) trial, ¹⁶ a multicenter trial of 1235 subjects with hypertension and type 2 diabetes. In this trial, patients were already receiving an ACE inhibitor or an angiotensin receptor blocker and were randomized to carvedilol or metoprolol twice daily. After 5 months of maintenance therapy, glycosylated hemoglobin increased in the metoprolol but not in the carvedilol group. When insulin resistance was assessed by the homeostasis model, the insulin resistance index was significantly decreased with carvedilol but not metoprolol. These differences in glycemic control could not be explained by differences in blood pressure, weight gain, or other factors. These findings support both a neutral effect on glycemic control and improved insulin sensitivity with the use of vasodilating β blockers when compared with agents that only block β receptors.

The neutral effects of newer β‐blocking agents on glycemic control and insulin sensitivity could be explained in part by their α‐blocking or β₂‐stimulating capacity, resulting in vasodilation and, hence, improved blood flow to the skeletal muscle. ⁴⁰ Therefore, the beneficial effects of carvedilol on glycemic parameters could also be explained by its hemodynamic effects, at least in part. It must be noted that carvedilol has an α‐ to β‐blockade ratio different from labetalol and has fewer side effects associated with vasodilation, such as postural hypotension and dizziness. ⁴¹ This difference in tolerability is of particular importance in patients with diabetic neuropathy and postural hypotension.

Nebivolol, a β₁‐selective blocker that modulates nitric oxide release, has also demonstrated favorable metabolic effects on lipids and glucose control in patients with hypertension. ⁴² Thirty hypertensive hyperlipidemic men and women were randomized to receive either atenolol (50 mg daily) or nebivolol (5 mg daily). Atenolol significantly increased triglyceride levels by 19%, which was not seen with nebivolol. Atenolol also increased lipoprotein(a) by 30%. Both agents, however, decreased serum high‐sensitivity C‐reactive protein levels, whereas only nebivolol reduced insulin resistance. Similar effects on glycemic control using a euglycemic clamp were also observed in a separate study with nebivolol. ⁴³

CONCLUSION

Antihypertensive treatment with traditional β blockers reduces cardiovascular risk in persons with diabetes; however, these agents are underutilized in patients with diabetes and/or the metabolic syndrome, possibly due to their detrimental effects on insulin sensitivity and the incidence of type 2 diabetes. ⁹ , ¹² , ⁴⁴ , ⁴⁵

Should these agents be used as initial therapy for the treatment of hypertension? In general, most guidelines argue against initial therapy unless compelling indications are present, such as heart failure, high sympathetic tone, or immediately following a myocardial infarction. More precisely, although these agents reduce blood pressure and, hence, cardiovascular events, there are other agents available with better tolerability and similar efficacy in reducing cardiovascular events; therefore, an agent that is better tolerated should be used. In patients with the metabolic syndrome who have impaired fasting glucose levels >100 but <126 mg/dL, a β blocker with neutral glycemic effects, if needed for initial therapy, would be preferred to reduce the risk of worsening glycemic control and hastening the development of diabetes. In persons with established diabetes already receiving oral hypoglycemic medications, β blockers, in general, are excellent third‐ and fourth‐line agents when needed for blood pressure control and should be used. Studies comparing newer vasodilating to conventional β‐blocking agents show a neutral effect on glycemic control and improved insulin sensitivity with vasodilating β blockers. This suggests that these agents could be used in subjects with hypertension with or without diabetes without fear of deterioration of these parameters. The definite answer on the possible benefits of vasodilating over conventional β blockers will come from a randomized trial of outcomes in patients with hypertension and diabetes similar to those already available in patients with heart failure. ⁴⁶