Reexamining Misconceptions About β-Blockers in Patients With Diabetes

Type 2 diabetes and hypertension are two of the most common contributors to cardiovascular disease (CVD) in the United States. Diabetes is estimated to affect 7% of the U.S. population—a total of 21 million individuals. In many patients, diabetes is asymptomatic, and as many as one-third of diabetic individuals are unaware that they have the disorder.¹  Hypertension, defined as blood pressure > 140/90 mmHg, affects one-third of Americans— an estimated 72 million people.¹  Hypertension is a common comorbid condition of diabetes, affecting ~ 20-60% of patients with diabetes, depending on ethnicity, age, and obesity.²  More than 3 million Americans have both conditions.³ 

Along with cardiovascular complications, hypertension in patients with diabetes contributes to increased risk of end-stage renal disease and diabetic retinopathy.^3,4  In patients with comorbid hypertension and diabetes, intensive pharmacological treatment to reach blood pressure goals may be even more important in reducing cardiovascular risk than blood glucose control.⁴  The high CVD risk in patients with diabetes necessitates more aggressive blood pressure targets.^5,6  The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC-7) recommends that blood pressure in diabetic patients be controlled to levels ≤ 130/80 mmHg.⁶  The U.K. Prospective Diabetes Study and the Hypertension Optimal Treatment trial both demonstrated that tight blood pressure control (< 130/85 mmHg) resulted in improved outcomes, including prevention of death and stroke and also prevention of microvascular complications.^7,8  Although tight blood glucose control decreases the frequency of microvascular complications such as retinopathy and nephropathy, it has not been shown to reduce diabetes-related mortality or the incidence of myocardial infarction (MI).⁹ 

Patients with diabetes and hypertension have a > 20% 10-year risk of developing coronary heart disease,¹⁰  the single greatest killer of American adults.¹  Both hypertension and diabetes are considered preclinical or Stage A heart failure that, if left untreated, can progress to structural heart failure (Figure 1).¹¹  The potential for complications associated with hypertension in patients with diabetes emphasizes the need for appropriate and aggressive therapy.

The JNC-7 recommends the use of angiotensin-converting enzyme (ACE) inhibitors, angiotensin receptor blockers (ARBs), low-dose thiazide diuretics, calcium channel blockers (CCBs), and β-blockers for first-line treatment of hypertension in patients with compelling indications, including diabetes (Figure 2).⁶  These recommendations are based on randomized clinical trials using a variety of antihypertensive agents that have shown that even a modest reduction in systolic blood pressure of 9-11 mmHg and diastolic blood pressure of 2-9 mmHg decreases cardiovascular events by 34-69% and the microvascular complications of retinopathy or nephropathy by 13% within 2-5 years.^7,8,12,13  National guidelines recommend β-blockers as a preferred therapy for the control of hypertension in patients with diabetes, heart failure, or high coronary heart disease risk or after MI because of the benefits and proven mortality risk reduction in these high-risk groups (Table 1).⁶ 

Based on current evidence, the American Association of Clinical Endocrinologists (AACE) has also proposed guidelines for the treatment of hypertension in patients with diabetes.⁵  Because ACE inhibitors and ARBs are associated with favorable effects on renal function and may improve insulin sensitivity, AACE recommends these agents as first-line therapy in the treatment of hypertension in diabetic patients. AACE recommends the use of diuretics in the lowest effective dosage (in conjunction with potassium replacement or the addition of a potassium-sparing agent) because thiazide diuretics can worsen blood glucose control and increase the likelihood of development of diabetes in individuals with insulin resistance.⁵ 

In recognition that β-blockers as a class may precipitate or exacerbate type 2 diabetes, these antihypertensive agents are not preferred as first-line agents for the treatment of hypertension in patients with diabetes. However, because β-blockers are effective in the management of ischemic and congestive cardiomyopathies—common cardiovascular complications of diabetes—AACE recommends the preferential use of third-generation β-blockers (e.g., nebivolol and carvedilol) as second- or third-line agents in this high-risk patient population (Table 2).⁵ 

This class of antihypertensive drugs has anti-ischemic as well as anti-atherogenic and anti-arrhythmic properties.^14,15  These actions are important because both hypertension and diabetes cause cardiac injury that can subsequently activate the reninangiotensin and sympathetic nervous systems and lead to myocardial remodeling and disease progression. β-Blockers with anti-atherogenic properties can reduce inflammation, shear stress, endothelial dysfunction, and the risk of plaque rupture; the anti-arrhythmic properties result from decreased sympathetic and heart rate activity and increased cardiac vagal tone.^14,15  Considering the detrimental effects of diabetes and hypertension on the myocardium, the beneficial effects of β-blockers beyond blood pressure lowering alone should not be overlooked.

Despite the proven benefits of β-blockers in lowering blood pressure and improving cardiovascular morbidity and mortality in clinical heart failure and post-MI trials, many physicians have been reluctant to prescribe β-blockers to patients with diabetes and hypertension. This reluctance is caused by perceived negative metabolic effects of β-blockers, including worsening of glycemic control, insulin sensitivity, and dyslipidemia and masking of hypoglycemia.^16,17 

Evidence suggests that there are differential effects of β-blockers. The first-generation β-blockers (e.g., propranolol) are nonspecific and thus block both β₁- and β₂-adrenergic receptors. The second generation β-blockers (e.g., atenolol and metoprolol) are β₁-selective. Third-generation β-blockers (e.g., carvedilol and nebivolol) offer additional benefits. Carvedilol is a nonselective β-blocker with vasodilatory activity mediated by α₁-adrenergic receptor blockade. Nebivolol is a β₁-selective blocker that also has vasodilatory properties believed to be a result of stimulation of nitric oxide release.

Many of the negative perceptions surrounding the use of β-blockers in diabetic patients involve traditional (i.e., first- and second-generation) β-blockers. Studies have shown that nonselective propranolol,¹⁸  β₁-selective atenolol,¹⁹  and metoprolol²⁰  significantly decrease insulin sensitivity in patients with hypertension.

The Antihypertensive Treatment and Lipid Profile in a North of Sweden Efficacy Evaluation study found that antihypertensive treatment with a low-dose diuretic (hydrochlorothiazide) combined with atenolol (if needed to reach blood pressure control) was associated with negative metabolic effects compared with treatment with an ARB (candesartan), combined with a CCB (felodipine) if needed.²¹  Both treatment regimens lowered blood pressure, with the majority of patients requiring two-drug therapy. Fasting levels of serum insulin and plasma glucose, as well as LDL/HDL and apolipoprotein B/apolipoprotein A-I ratios increased in the diuretic and atenolol group in contrast to no change in the ARB and CCB group. In addition, at 12 months of treatment, eight patients (4.1%) in the diuretic and atenolol group versus 1 patient (0.5%) in the ARB and CCB group were diagnosed with new-onset diabetes (P = 0.030).²¹ 

Atenolol's unfavorable metabolic effects may have a negative impact on prevention of cardiovascular events.²²  A meta-analysis of trials with atenolol in patients with hypertension revealed that there were no discernible differences between atenolol or placebo in the reduction of all-cause mortality (1.01 [95% CI 0.89-1.15]), cardiovascular mortality (0.99 [0.83-1.18]), or MI (0.99 [0.83-1.19]), despite the fact that 60% of patients were treated with an additional antihypertensive agent.¹⁹  More conclusively, the Anglo-Scandinavian Cardiac Outcomes Trial-Blood Pressure Lowering Arm showed that atenolol treatment resulted in significantly worse outcomes, including cardiovascular events and procedures (P < 0.001), cardiovascular mortality (P = 0.001), and all-cause mortality (P = 0.0247), as well as the development of diabetes (P < 0.0001) when compared with the CCB amlodipine. A total of 19,257 patients with hypertension were treated an average of 5.5 years.²³ 

The vasodilatory β-blockers (e.g., nebivolol and carvedilol) have demonstrated a more favorable metabolic profile with respect to glycemic control and lipids.^24,25  To assess the effect of nebivolol on metabolic parameters, a study randomized 30 patients with hypertension and hyperlipidemia to either atenolol or nebivolol.²⁵  After 12 weeks of either β-blocker therapy, pravastatin was added for an additional 12 weeks of treatment. Atenolol significantly increased triglyceride levels by 19% (P = 0.05) and significantly increased lipoprotein(a) by 30% (P = 0.028), whereas nebivolol did not produce significant changes in either parameter. Glucose levels remained the same in the nebivolol-treated patients, while insulin levels were reduced by 10%, and insulin resistance was reduced by 20% (P = .05).²⁵  These parameters were not significantly changed in the atenolol-treated patients. There was also no significant difference in these parameters between the atenolol and nebivolol treatment groups except for insulin-resistance reduction (0 vs. -20%, respectively; P = 0.05).²⁵ 

Carvedilol has been uniquely shown to improve the common negative metabolic effects associated with the use of first- and second-generation β-blockers.^24,26  The addition of the α₁-blocking properties of carvedilol, which interfere with vasoconstriction, are theorized to increase blood flow to skeletal muscles, thereby improving metabolic parameters.²⁷  Beneficial metabolic effects of carvedilol were demonstrated in a comparison study of metoprolol and carvedilol in the treatment of hypertension in nondiabetic patients with impaired insulin sensitivity.²⁰  Both antihypertensive agents effectively lowered blood pressure. However, after metoprolol treatment, insulin sensitivity decreased, whereas it increased after carvedilol treatment. There was also a decrease in high-density lipoprotein and an increase in triglyceride levels in patients in the metoprolol-treated group; however, these parameters remained unchanged in patients in the carvedilol-treated group. These findings suggest that β-blocker treatment, when combined with α₁-blocking activity, has advantageous effects on insulin sensitivity and lipids and could therefore be suitable for patients with impaired metabolic function.

Results from the Glycemic Effects in Diabetes Mellitus: Carvedilol-Metoprolol Comparison in Hypertensives (GEMINI) trial elucidated important treatment differences between the β-blockers carvedilol and metoprolol tartrate.²⁶  Among 1,235 patients with diabetes and hypertension, carvedilol stabilized A1C (Figure 3)²⁶  and improved insulin resistance (HOMA index) and cholesterol. In contrast, metoprolol tartrate worsened glycemic and cholesterol control. Moreover, more patients treated with metoprolol withdrew because of worsening glycemic control compared with carvedilol-treated patients.

The results of the GEMINI trial support earlier studies demonstrating that metoprolol has a negative glycemic effect.^28-30  A study of patients with essential hypertension revealed that, after 6 months of treatment, once-daily metoprolol succinate did not affect fasting plasma glucose but increased A1C levels by 5% compared to baseline levels (P = 0.04).²⁸  This effect is of importance because an A1C reduction of as little as 0.1% was associated with 12% mortality risk reduction in the Norfolk cohort of the European Prospective Investigation into Cancer and Nutrition.²⁹  Other studies have also found that each 1-percentage point decrease in A1C significantly reduced the risk of mortality, heart failure, and MI in patients with diabetes and hypertension.^30,31 

A further substudy of the GEMINI trial demonstrated that carvedilol and metoprolol tartrate treatment produced statistically significant differences in diabetes symptom scores.³²  In the Diabetes Symptom Checklist, a decrease in score indicates symptom improvement. Compared to baseline and to metoprolol tartrate, carvedilol improved overall symptom score (-0.08 [P = 0.008] and -0.08 [P = 0.02], respectively), hypoglycemia score (-0.12 [P = 0.013] and -0.12 [P = 0.02], respectively), and hyperglycemia score (-0.2 [P = 0.0001] and -0.16 [P = 0.005], respectively). Metoprolol tartrate treatment did not significantly improve these parameters and was associated with a worse psychological fatigue score compared with baseline levels (0.15 [P = 0.006]).³² 

Microalbuminuria (defined as urine albumin:creatinine ratio of 30-300 mg/g) is often the first clinical sign of renal dysfunction in patients with diabetes and is a recognized marker of cardiovascular risk and increased cardiovascular morbidity and mortality. A GEMINI substudy demonstrated that carvedilol treatment resulted in more favorable effects on microalbuminuria than metoprolol tartrate treatment.³³  In GEMINI, 25% of patients had microalbuminuria. Carvedilol treatment resulted in a 16% relative reduction in the albumin:creatinine ratio (95% CI 6-25%; P = 0.003), and significantly fewer carvedilol-treated patients with normoalbuminuria (< 30 mg/g) progressed to microalbuminuria (6.6 vs. 11.1%; odds ratio [OR] 0.53; 95% CI 0.30-0.93; P = 0.03) compared to metoprolol tartrate treatment.³³ 

Hypoglycemia is a serious condition that may lead to confusion, irrationality, and in its most severe form, coma, seizure, and even sudden death.³⁴  Theoretically, β-blockers could increase the risk of severe hypoglycemia by masking the adrenergic warning symptoms of hypoglycemia, including weakness, shakiness, sweating, pallor, and palpitations.³⁵  Clinical evidence suggests that there may be a relationship of specific antihypertensives to the development of hypoglycemia.³⁶ 

A case-control study that used 1993 Medicaid data evaluated the relative risk of hypoglycemia in a cohort of patients treated for diabetes.³⁶  The study cohort was divided into patients for whom the physician reported hypoglycemia (using ICD-9 codes) and diabetic control subjects without hypoglycemia. Exposure to specific antihypertensive drugs, including ACE inhibitors, β-blockers, and diuretics, was assessed in the two groups. A principal finding of the study was that, although use of ACE inhibitors as a class was not associated with an increased risk of hypoglycemia, a significantly increased risk was associated with the specific use of enalapril (OR 2.7; 95% CI 1.2-5.7). The lack of class effect of ACE inhibitors on hypoglycemia and the selective association of enalapril with hypoglycemia risk were consistent with earlier reports.^37-40  In contrast, β-blockers were not associated with increased risk of hypoglycemia in either insulin or sulfonylurea users.

A thorough review of the literature concluded that, although adverse effects of β₂-selective blockers on glucose metabolism are recognized, there is no evidence to withhold β₁-selective blocking agents from diabetic patients because these agents are not associated with an increased risk of severe hypoglycemia.⁴¹ 

This is especially important in light of the life-threatening consequences of hypoglycemia. Hypoglycemia produces electrocardiographic QTc interval lengthening that may play a pathogenic role in the occurrence of sudden death. Case reports have highlighted the occurrence of sudden overnight death among young patients with type 1 diabetes.^42-46 

It has been suggested that patients with type 1 diabetes with cardiac autonomic neuropathy have a greater risk of sudden death.⁴⁷  A study that used an experimental model of hypoglycemia to test this theory in 28 patients with diabetes with and without cardiac autonomic neuropathy refuted the hypothesis.⁴⁸  Participants with cardiac autonomic neuropathy tended to exhibit the smallest QTc increases, suggesting that autonomic neuropathy is not an important risk factor for sudden death from hypoglycemia. A subsequent study of eight diabetic patients who had shown QTc lengthening during experimental hypoglycemia found that atenolol, a β₁-blocking agent, significantly reduced hypoglycemic QTc lengthening.⁴⁹ 

Hypoglycemia is common in type 1 diabetes and is likely to occur more frequently in those who have tighter glycemic control. The potential effect of β-blockers on prevention of sudden death in diabetic patients warrants further investigation.

β-Blockers, in general, are associated with weight gain, which in turn reduces insulin sensitivity. However, weight gain is not a class effect of β-blockers. An analysis of the GEMINI trial showed that there was a statistically significant difference in weight gain between carvedilol- and metoprolol-treated patients.⁴⁹  Compared with baseline, patients taking metoprolol experienced a significant mean weight gain (1.2 ± 0.16 kg; P < 0.001), whereas patients taking carvedilol did not (0.17 ± 0.19 kg; P = 0.36). Compared with metoprolol-treated patients, carvedilol-treated patients were more likely to experience no weight change (44 vs. 35%; P = 0.005) and less likely to experience a weight gain of > 7% (1.1 vs. 4.5%; P = 0.006).⁴⁹ 

Current American Diabetes Association (ADA) standards of medical care for patients with diabetes use a mnemonic device (ABC) designed to remind health care providers and patients of the three clinical issues—A1C, blood pressure, and cholesterol— that must be addressed to minimize the vascular complications of diabetes, including MI, stroke, and peripheral vascular disease.⁵⁰ 

Despite these guidelines, control of these clinical issues is inadequate in the community setting. Data from the National Health and Nutrition Examination Survey (NHANES) 1999-2002 showed that 44% of diabetic patients achieved optimal glycemic control (A1C < 7%).⁵¹  Moreover, only 35% of diabetic patients achieved optimal blood pressure goals (< 130/80 mmHg) in the 2003-2004 NHANES.⁵²  Blood pressure control, in general, is poor in patients with essential hypertension. Among adults with hypertension, 76% are aware of their disease, 65% are prescribed antihypertensives, and 37% achieve blood pressure goals.⁵³ 

Polypharmacy is the natural consequence of providing guideline-recommended medical care to patients with diabetes.⁵⁴  Varying combinations of antiglycemic medications are often necessary to correct abnormally elevated levels of blood glucose in patients with diabetes, including insulin therapy and medications to increase insulin production, to decrease glucose production by the liver, and to decrease carbohydrate absorption. In addition, hypertension management guidelines acknowledge that most patients, especially those with comorbid diabetes, require a combination of antihypertensive agents from different classes to reach blood pressure targets (< 130/80 mmHg).⁶  The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT) found that > 40% of individuals required multiple drug therapy (addition of ACE inhibitors, or CCB, or diuretic) to control blood pressure.⁵⁵ 

Hyperlipidemia is a common comorbid condition among patients with diabetes and hypertension. A study of 371,221 outpatients at six Veterans Health Administration medical centers found that 30.7% had hypertension and dyslipidemia, and 66.3% had concomitant hypertension, dyslipidemia, and diabetes.⁵⁶  If optimal lipid control cannot be achieved with lifestyle modifications, statins or fibrates may be prescribed.⁵⁷  The presence of other comorbid conditions (e.g., renal disease or ischemic heart disease) compounds the polypharmacy necessary to reach optimal disease control.

In addition to complex medication regimens, side effects, inconvenience of dosing, and lack of perception of treatment benefit of an asymptomatic condition (e.g., hypertension) can negatively affect a patient's compliance with a treatment regimen. Physician reluctance to prescribe guideline-supported medications can also play a role in suboptimal disease management.

In a community study of 128 diabetic patients, in the week before the study, patients reported taking a mean of four diabetes-related medicines and a total mean of six different medications daily for diabetes and concomitant conditions: glycemic control (87%), hypertension (80%), and dyslipidemia (57%).⁵⁴  Despite good adherence to diabetes-related medicines, compliance with other medications was suboptimal in the previous week. The most frequent reasons for noncompliance were side effects (58%) and difficulty remembering to take all medications (23%). Of note, only 23% of patients reported the occurrence of side effects to their physicians. Not surprisingly, self-reported adherence rates for medications that caused side effects were significantly lower (5.4 vs. 6.9 out of 7 days; P < 0.001). Patients with negative perceptions of the immediate and future benefit of prescribed medications also had lower 7-day adherence rates (P < 0.001).

Adverse effects from antihypertensive treatment vary by drug class.⁵⁸  A general practice survey of patients who used antihypertensive medications showed that, when side effects for each drug were compared with the pooled average incidences of other antihypertensive agents, ACE inhibitors were associated with the highest incidence of dry cough (28 vs. 8%, respectively; P < 0.001); CCBs were associated with the highest incidence of peripheral edema (22 vs. 12%, respectively; P < 0.001); and β-blockers were associated with the highest incidence of sexual dysfunction (17 vs. 10%, respectively; P < 0.01).⁵⁹ 

The occurrence of adverse side effects can lead to medication nonadherence and negatively affect treatment outcomes. When multi-dosed drugs with short durations of action are taken inconsistently, blood pressure control can be compromised. Reintroduction of drugs after inconsistent use can lead to excessive side effects.⁶⁰ 

Simplifying treatment regimens by using once-daily dosing and combination drugs may improve adherence. A review of studies that measured medication compliance confirmed that the prescribed number of doses per day is inversely related to compliance. Simpler regimens involving less frequent dosing resulted in better compliance across a variety of therapeutic classes.⁶¹  Compliance is better in patients on once-daily medications (79%) than in patients on multiple-dosing regimens (twice daily, 69%; three times daily, 65%; four times daily, 51%). Not surprisingly, forgetfulness is cited as one of the most important reasons for noncompliance (30%).^62,63 

Another issue that may lead to adverse consequences is patient reluctance to tell their physician that they miss medication doses.⁶⁴  Physicians may increase dosage or add medications to the treatment regimen of patients whose inadequate therapeutic response is actually a result of nonadherence to the prescribed medications. Physicians must take a proactive approach and stress the benefits of complying with medications for both short- and long-term benefit in lieu of waiting for their patients to approach them with concerns.

The poor rate of hypertension control in both diabetic and nondiabetic patients may also reflect inadequate prescription of evidence-based medications to control comorbid conditions. In a retrospective cohort study of 3,998 diabetic patients with ischemic heart disease, > 80% of patients received two cardioprotective medications (ACE inhibitor, β-blocker, or statin) despite high levels of concomitant disease, including hypertension (90%), hyperlipidemia (> 80%), heart failure (> 34%), or post-MI (~ 50%).⁶⁵  Even fewer patients (< 40%) received all three lifesaving therapies. Not surprisingly, < 50% of these patients had control of blood pressure or A1C regardless of whether they adhered to medication.

Proper selection of treatment regimen plays a key role in optimizing patient outcomes and quality of life. Patients with diabetes who have had an MI or have hypertension, heart failure, or coronary artery disease face a real and increased risk of morbidity and mortality that should be countered with appropriate management using evidence-based lifesaving treatments. β-Blockers are indicated not only for the treatment of patients with hypertension, heart failure, or who are post-MI, but also for the treatment of hypertension in patients with diabetes. Because of the prevalence of comorbid hypertension in diabetic patients, physicians must be careful to prescribe a β-blocker that does not complicate a patient's existing medication regimen and has the most favorable side-effect profile to prevent patient noncompliance.

The once-daily formulation of extended-release carvedilol phosphate (carvedilol CR) allows for consideration of a new treatment paradigm that can help overcome barriers to adherence. The GEMINI study demonstrated that carvedilol lowers blood pressure while providing beneficial metabolic effects compared to metoprolol therapy.²⁶  Carvedilol CR has shown efficacy in significantly lowering blood pressure in a double-blind, randomized trial involving 338 patients.⁶⁶  All three doses of carvedilol CR treatment significantly decreased diastolic and systolic blood pressure by study end compared with placebo (Figure 4).⁶⁶  Adverse event findings were similar in the placebo and carvedilol CR groups for headache, fatigue, dizziness, and erectile dysfunction.^66,67  The convenience of once-daily dosing combined with a low adverse event profile is a key strategy to improve medication adherence in patients with hypertension and diabetes.

Diabetes is increasing in the United States as the population ages, becomes less active, and grows more obese, and the prevalence is expected to double in the next 25 years.⁶⁸  Diabetes and hypertension frequently coexist, affecting > 3 million adults in the United States.²  Hypertension in patients with diabetes must be treated aggressively to reduce the risk of macrovascular and microvascular morbidity and mortality.

Because of their intrinsic high CVD risk, patients with diabetes have a more stringent blood pressure target (< 130/80 mmHg) than nondiabetic patients. National guidelines recommend β-blockers among preferred therapies for control of blood pressure in patients with diabetes.^5,6  When more than one drug is necessary to reach blood pressure goals, combinations of antihypertensives of different classes (e.g., a β-blocker and an ACE inhibitor or diuretic) provide complementary actions.

Because of the need for multiple medications to effectively manage patients with diabetes and hypertension, the appropriate selection of a treatment regimen with good tolerability and simplified dosing is crucial to maximize positive outcomes in this high-risk population. Unlike traditional β-blockers, vasodilatory β-blockers have favorable tolerability and metabolic profiles, while offering effective blood pressure control.