Beta-adrenoceptor antagonists

Fatigue

Fatigue is one of the most commonly reported adverse effects of beta-adrenoceptor antagonists, with reported occurrence rates of up to 20% or more, particularly in those who exert themselves. It has to be viewed alongside the ability to produce fatigue and lethargy by a possible effect on the nervous system. The precise cause of physical fatigue is not known, but hypotheses include impaired muscle blood supply, effects on intermediary metabolism, and a direct effect on muscle contractility [ ].

Theoretically, beta ₁ -selective drugs are less likely to alter these variables, and might therefore have an advantage over non-selective drugs. However, this has not always been shown in single-dose studies in volunteers [ , ], although in two such studies atenolol produced less exercise intolerance than propranolol at comparable dosages [ , ]. For an unexplained reason, cardioselectivity impaired performance relatively less in subjects with a high proportion of slow-twitch muscle fibers than it did in those whose muscle biopsy specimens showed a high percentage of fast-twitch fibers [ ]. The muscle fibers of long-distance runners are predominantly of the slow-twitch type, and this probably explains the superiority of atenolol over propranolol when exercise performance was assessed in such subjects [ ]. The release of lactic acid from skeletal muscle cells is impaired to a greater extent by non-selective beta-blockers than by cardioselective drugs, and cardioselectivity was associated with a less marked fall in blood glucose during and after maximal and submaximal exercise [ , ]. Partial agonist activity might have been the reason for the superiority of oxprenolol over propranolol in terms of exercise duration [ ].

Lipid solubility

Lipid solubility [ ] determines the extent to which a drug partitions between an organic solvent and water. Propranolol, oxprenolol, metoprolol, and timolol are the most lipid-soluble beta-adrenoceptor antagonists, and atenolol, nadolol, and sotalol are the most water-soluble; acebutolol and pindolol are intermediate [ ].

The more lipophilic drugs are extensively metabolized in the gut wall and liver (first-pass metabolism). This first-pass clearance is variable and can result in 20-fold differences in plasma drug concentrations between patients who have taken the same dose. It also produces susceptibility to drug interactions with agents that alter hepatic drug metabolism, for example cimetidine, and can result in altered kinetics and hence drug response in patients with hepatic disease, particularly cirrhosis. Lipid-soluble drugs pass the blood–brain barrier more readily [ ] and should be more likely to cause adverse nervous system effects, such as disturbance of sleep, but the evidence for this is not very convincing.

In contrast, water-soluble drugs are cleared more slowly from the body by the kidneys. These drugs therefore tend to accumulate in patients with renal disease, do not interact with drugs that affect hepatic metabolism, and gain access to the brain less readily.

Cardioselectivity

Cardioselectivity [ ], or more properly beta ₁ -adrenoceptor selectivity, is the term used to indicate that there are at least two types of beta-adrenoceptors, and that while some drugs are non-selective (that is they are competitive antagonists at both beta ₁ - and beta ₂ -adrenoceptors), others appear to be more selective antagonists at beta ₁ -adrenoceptors, which are predominantly found in the heart. Bronchial tissue, peripheral blood vessels, the uterus, and pancreatic beta-cells contain principally beta ₂ -adrenoceptors. Thus, cardioselective beta-adrenoceptor antagonists, such as atenolol and metoprolol, might offer theoretical benefits to patients with bronchial asthma, peripheral vascular disease, and diabetes mellitus.

Cardioselective drugs may have relatively less effect on the airways, but they are in no way cardiospecific and they should be used with great care in patients with evidence of reversible obstructive airways disease.

The benefits of cardioselective drugs in patients with Raynaud’s phenomenon or intermittent claudication have been difficult to prove. Because of vascular sparing, cardioselective agents may also be preferable in stress, when adrenaline is released.

Cardioselective drugs are less likely to produce adverse effects in patients with type I diabetes than non-selective drugs. At present, hypoglycemia in patients with type I diabetes mellitus is the only clinical problem in which cardioselectivity is considered important. Even there, any potential advantages of cardioselective drugs in minimizing adverse effects apply only at low dosages, since cardioselectivity is dose-dependent.

Partial agonist activity

Partial agonist activity [ , ] is the property whereby a molecule occupying the beta-adrenoceptor exercises agonist effects of its own at the same time as it competitively inhibits the effects of other extrinsic agonists. The effects of these drugs depend on the degree on endogenous tone of the sympathetic nervous system. When there is high endogenous sympathetic tone they tend to act as beta-blockers; when endogenous sympathetic tone is low they tend to act as beta agonists. Thus, xamoterol had a beneficial effect in patients with mild heart failure (NYHA classes I and II), through a positive inotropic effect on the heart; however, in severe heart failure (NYHA classes III and IV), in which sympathetic tone is high, it acted as a beta-blocker and worsened the heart failure, through a negative inotropic effect [ ].

Partial agonists, such as acebutolol, oxprenolol, pindolol, practolol, and xamoterol, produce less resting bradycardia. It has also been claimed that such agents cause a smaller increase in airways resistance in asthmatics, less reduction in cardiac output (and consequently a lower risk of congestive heart failure), and fewer adverse effects in patients with cold hands, Raynaud’s phenomenon, or intermittent claudication. However, none of these advantages has been convincingly demonstrated in practice, and patients with bronchial asthma or incipient heart failure must be considered at risk with this type of compound.

Drugs with partial agonist activity can produce tremor [ ].

Drugs that combine beta ₁ antagonism or partial agonism with beta ₂ agonism (celiprolol, dilevalol, labetalol, pindolol) or with alpha-antagonism (carvedilol, labetalol) have been developed [ ]. Both classes have significant peripheral vasodilating effects. Drugs with significant agonist activity at beta ₁ -adrenoceptors have poor antihypertensive properties [ ].

Membrane-stabilizing activity

Drugs with membrane-stabilizing activity reduce the rate of rise of the cardiac action potential and have other electrophysiological effects. Membrane-stabilizing activity has only been shown in human cardiac muscle in vitro in concentrations 100 times greater than those produced by therapeutic doses [ ]. It is therefore likely to be of clinical relevance only if large overdoses are taken.

Acute chest pain

Worsening of angina pectoris has been attributed to beta-blocker therapy. The reports include 35 cases in a series of 296 elderly patients admitted to hospital with suspected myocardial infarction; in these 35 the pain disappeared within 7 hours of withdrawing beta-blocker therapy [ ].

Worsening of angina has been reported at very low heart rates [ ]. Propranolol resulted in vasotonic angina in six patients during a double-blind trial, with prolongation of the duration of pain and electrocardiographically assessed ischemia. It has been suggested that this reflects a reduction in coronary perfusion as a result of reduced cardiac output, and also coronary arterial spasm provoked by non-selective agents by inhibition of beta ₂ -mediated vasodilatation [ ]. The latter explanation is controversial, and the use of beta-blockers in patients with arteriographic evidence of coronary artery spasm has not consistently caused worsening of the disorder [ ].

Unstable angina has also followed treatment for hypertension with cardioselective drugs such as betaxolol [ ].

Cardiac dysrhythmias and heart block

Beta-blockade can result in sinus bradycardia, because blockade of sympathetic tone allows unopposed parasympathetic activity. Drugs with partial agonist activity may prevent bradycardia [ ]. However, heart rates under 60/minute often worry the physician more than the patient: in a retrospective study of nearly 7000 patients taking beta-adrenoceptor antagonists, apart from dizziness in patients with heart rates under 40/minute (0.4% of the total group), slow heart rates were well tolerated [ ].

All beta-blockers cause an increase in atrioventricular conduction time; this is most pronounced with drugs that have potent membrane-depressant properties and no partial agonist activity. Sotalol differs from other beta-blockers in that it increases the duration of the action potential in the cardiac Purkinje fibers and ventricular muscle at therapeutic doses. This is a class III antidysrhythmic effect, and because of this, sotalol has been used to treat ventricular [ ] and supraventricular dysrhythmias [ ]. The main serious adverse effect of sotalol is that it is prodysrhythmic in certain circumstances, and can cause torsade de pointes [ , ].

Cardiac failure

Beta-adrenoceptor antagonists reduce cardiac output through their negative inotropic and negative chronotropic effects. They can therefore cause worsening systolic heart failure or new heart failure in patients who depend on high sympathetic drive to maintain cardiac output. Plasma noradrenaline is increased in patients with heart failure, and the extent of this increase is directly related to the degree of ventricular impairment [ ]. Since the greatest effect on sympathetic activity occurs with the first (and usually the lowest) doses, heart failure associated with beta-blockade seems to be independent of dosage. Heart failure is one of the most serious adverse effects of the beta-adrenoceptor antagonists [ ], but it is usually predictable and can be attenuated by pretreatment with diuretics and angiotensin-converting enzyme inhibitors in patients who are considered to be at risk.

Beta-adrenoceptor antagonists are recommended for patients with chronic heart failure to improve outcomes and left ventricular ejection fraction. However, concerns regarding the tolerability of these drugs by elderly people have contributed to limited use in this age class. Carvedilol has been assessed in 1030 patients with heart failure aged over 70 years and reasons for drug withdrawal were recorded [ ]. The main reasons were worsening of heart failure in less than 10%, symptomatic hypotension in about 13%, and less often bradycardia and wheeze.

It has been suggested that drugs with partial agonist activity (see Table 1 ), which have a minimal depressant effect on normal resting sympathetic tone, might cause less reduction in cardiac output [ ] and thus protect against the development of cardiac decompensation [ ]. However, this has not been satisfactorily shown for drugs with high partial agonist activity, for example acebutolol, oxprenolol, and pindolol [ ], and these drugs should therefore be given with the same caution as others in compromised patients. Xamoterol, a beta-antagonist with substantial beta ₁ partial agonist activity, was hoped to be of benefit in mild congestive heart failure [ ], but its widespread use by non-specialists in more severe degrees of heart failure resulted in many reports of worsening heart failure [ ]. Heart failure has also been produced by labetalol [ ] and after the use of timolol eye-drops in the treatment of glaucoma [ ].

Identification of the possible predictors of intolerance to beta-blockade in heart failure was the object of an analysis of a series of 236 patients [ ]. A B-type natriuretic peptide (BNP) concentration of over 1000 pg/ml in the first 8 days from the start of beta-blockade was a significant predictor of worsening heart failure. This neurohormone not only has powerful prognostic value, but it can also provide useful information in the selection of patients in whom beta-blockade should be started.

Hypotension

Beta-adrenoceptor antagonists lower blood pressure, probably by a variety of mechanisms, including reduced cardiac output. More severe reductions in blood pressure can occur and can be associated with syncope [ ]. It has been suggested that this is more likely to occur in old people, but comprehensive studies have stressed the safety of beta-blockers in this age group [ ].

• Profound hypotension, resulting in renal insufficiency, has been reported in a single patient after the administration of atenolol 100 mg orally [ ]; however, large doses of furosemide and diazoxide were also given in this case, and this appears more likely to have been a consequence of a drug interaction.

Peripheral vascular effects

Cold extremities or exacerbation of Raynaud’s phenomenon are amongst the commonest adverse effects reported with beta-blockers (5.8% of nearly 800 patients taking propranolol) [ ]; Raynaud’s phenomenon occurs in 0.5–6% of patients [ ]. The mechanism may be potentiation of the effects of a cold environment on an already abnormal circulation, but whether symptoms can be produced de novo is more difficult to determine. However, a retrospective questionnaire study in 758 patients taking antihypertensive drugs showed that 40% of patients taking beta-blockers noted cold extremities, compared with 18% of those taking diuretics; there were no significant differences among patients taking alprenolol, atenolol, metoprolol, pindolol, and propranolol [ ]. Similarly, a large randomized study showed that the incidence of Raynaud’s phenomenon was the same for atenolol and pindolol [ ]. In another study, vasospastic symptoms improved when labetalol was substituted for a variety of beta-blockers [ ]. On the other hand, a small, double-blind, placebo-controlled study in patients with established Raynaud’s phenomenon showed that the prevalence of symptoms with both propranolol and labetalol was no greater than that with placebo [ ].

Intermittent claudication has also been reported to be worsened by beta-adrenoceptor antagonists, but has been difficult to document because of the difficulty of study design in patients with advanced atherosclerosis. As early as 1975 it was reported from one small placebo-controlled study that propranolol did not exacerbate symptoms in patients with intermittent claudication [ ]. This has subsequently been supported by the results of several large placebo-controlled trials of beta-blockers in mild hypertension and reports of trials of the secondary prevention of myocardial infarction, in which intermittent claudication was not mentioned as an adverse effect, even though it was not a specific contraindication to inclusion [ ]. In addition, a comprehensive study of the effects of beta-adrenoceptor antagonists in patients with intermittent claudication did not show beta-blockade to be an independent risk factor for the disease [ ]. In men with chronic stable intermittent claudication, atenolol (50 mg bd) had no effect on walking distance or foot temperature [ ]. These findings have been confirmed in a recent meta-analysis of 11 randomized, controlled trials to determine whether beta-blockers exacerbate intermittent claudication [ ].

Patchy skin necrosis has been described in hypertensive patients with small-vessel disease in the legs who were taking beta-blockers. Characteristically, pedal pulses remained palpable and the lesions occurred during cold weather and healed on withdrawal of the drugs [ ]. Three cases have been reported in which long-lasting incipient gangrene of the leg was immediately overcome when a beta-blocker was withdrawn [ , ], showing how easily these drugs are overlooked in such circumstances. In several cases of beta-blocker-induced gangrene, recovery did not follow withdrawal of therapy, and amputation was necessary [ , ]. Thus, when possible, other forms of therapy should be used in patients with critical ischemia or rest pain.

It has also been suggested that beta-blockade may compromise the splanchnic vasculature. Intravenous propranolol reduces splanchnic blood flow experimentally by 29% while reducing cardiac output by only 6% [ ].

Five patients developed mesenteric ischemia, four with ischemic colitis, and one with abdominal angina, while taking beta-adrenoceptor antagonists [ ]. Although causation was not proven, it was possible.

Airways obstruction

Since the introduction of propranolol, it has been recognized that patients with bronchial asthma treated with beta-adrenoceptor antagonists can develop severe airways obstruction [ ], which can be fatal [ ] or near fatal [ , ]; this has even followed the use of eye-drops containing timolol [ ]. Beta-blockers upset the balance of bronchial smooth muscle tone by blocking the bronchial beta ₂ -adrenoceptors responsible for bronchodilatation. They also promote degranulation of mast cells and depress central responsiveness to carbon dioxide [ , ].

Although beta ₁ -selective drugs are theoretically safer, there are reports of serious reductions in ventilatory function [ , ], even when used as eye-drops [ ]. However, it has been concluded that if beta-blockade is necessary in the treatment of glaucoma, cardioselective beta-blocking drugs should be preferred [ ]. While cardioselectivity is dose-dependent [ ], and higher dosages might therefore be expected to produce adverse effects, metoprolol and bevantolol, even in dosages that are lower than those usually required for a therapeutic effect, may be poorly tolerated by patients with asthma [ ].

Whether drugs with partial agonist activity confer any advantage is uncertain. Some of the evidence that patients with asthma tolerate beta-blockers is probably misleading, relating to patients with chronic obstructive airways disease who have irreversible changes and who do not respond to either bronchoconstricting or bronchodilating drugs [ ]. In contrast, a few patients who have never had asthma or chronic bronchitis develop severe bronchospasm when given a beta-blocker. Some, but not all, of these cases [ ] may have been allergic reactions to the dyestuffs (for example tartrazine) that are used to color some formulations. Other patients, who need not have a history of chest disease, only develop increased airways resistance with beta-blockers during respiratory infections.

It is against this background that claims that some asthmatic subjects will tolerate certain beta-blockers [ ] must be viewed. Some asthmatic patients may indeed tolerate either cardioselective beta-blockers (such as atenolol and metoprolol) or labetalol [ , ], and in patients taking atenolol beta ₂ -adrenoceptor agonists may continue to produce bronchodilatation [ ], but in most instances other therapeutic options are preferable [ ]. Celiprolol is a beta ₁ -adrenoceptor antagonist that has partial beta ₂ agonist activity. Small studies have suggested that it may be useful in patients with asthma [ ], but worsening airways obstruction has been reported [ ]; it has been concluded that celiprolol has no advantage over existing beta-blockers in the treatment of hypertension [ ].

Bronchospasm, which can be life-threatening, can be precipitated by beta-blocker eye-drops. Even beta ₁ -selective antagonists, such as betaxolol, can cause a substantial reduction in forced expiratory volume. Wheezing and dyspnea have been reported among patients using betaxolol: the symptoms resolved after withdrawal. A cross-sectional study has shown that ophthalmologists were more aware than chest physicians about the use of beta-blocker eye-drops by patients with obstructive airways disease; patient awareness was also poor [ , ].

Attention has also been drawn to the increased risks of the adverse effects of beta-blockers on respiratory function in old people [ ].

Central ventilatory suppression

Reduced sensitivity of the respiratory center to carbon dioxide has been reported [ , ]. The clinical significance of this is unknown, but lethal synergism between morphine and propranolol in suppressing ventilation in animals has been described [ ].

Pneumonitis, pulmonary fibrosis, and pleurisy

Pulmonary fibrosis [ ] and pleural fibrosis [ ] have both been described as infrequent complications associated with practolol. Pulmonary fibrosis has also occurred during treatment with pindolol [ ] and acebutolol [ , ]. Pleuritic and pneumonitic reactions to acebutolol have been reported [ ].

Car driving and other specialized skills

In view of the large numbers of people who take beta-adrenoceptor antagonists regularly for hypertension or ischemic heart disease, the question arises whether these drugs impair performance in tasks that require psychomotor coordination. The occupations under scrutiny include car-driving, the operation of industrial machinery, and the piloting of aeroplanes. The current evidence is conflicting and controversial. One report suggested that propranolol and pindolol given for 5 days impaired slalom driving in a manner comparable with the coordination defects caused by alcohol [ ]. In contrast, other studies have shown that driving skills were not impaired during long-term beta-blocker therapy and might even be improved [ , ]. There is also a suggestion that tolerance to the central effects of these drugs can develop within 3 weeks of starting therapy, provided the dosage does not change [ ]. Until more information is available from well-controlled studies, it is advisable to inform patients who are starting treatment with beta-blockers that they should exercise special care in the performance of skills requiring psychomotor coordination for the first 1 or 2 weeks.

Eyes

Keratopathy in association with the practolol syndrome is the major serious ocular effect ascribed to beta-adrenoceptor antagonists. Conjunctivitis and visual disturbances have also been reported, and a case of ocular pemphigoid has been described in a patient taking timolol eye-drops for glaucoma [ ]. Anterior uveitis has been reported in patients taking betaxolol [ ] and metipranolol [ , ]. Corneal anesthesia and epithelial sloughing with continuing use of topical beta-blockers have also been reported [ ], as have ocular myasthenia and worsened sicca syndrome. Patients who lack CYP2D6 are more likely to have higher systemic concentrations of beta-blockers after topical application, making them susceptible to adverse reactions.

• Recurrent retinal arteriolar spasm with associated visual loss has been described in a 68-year-old man with hypertension treated with atenolol [ ].

• A 60-year-old man with open-angle glaucoma developed an allergic contact conjunctivitis and dermatitis from carteolol, a topical non-cardioselective beta-blocker [ ]. He had extensive cross-reactivity to other topical beta-blockers, such as timolol and levobunolol. Cross-reactivity among different beta-blockers is possibly due to a common lateral aliphatic chain.

Disturbances of psychomotor function

Beta-adrenoceptor antagonists impair performance in psychomotor tests after single doses. These include effects of atenolol, oxprenolol, and propranolol on pursuit rotor and reaction times [ , ]. However, other studies with the same drugs have failed to show significant effects [ ], and the issue has remained controversial. A report that sotalol improved psychomotor performance in 12 healthy individuals in a dose of 320 mg/day but impaired performance at 960 mg/day [ ] has been interpreted to indicate that the water-soluble beta-adrenoceptor antagonists would be less likely than the fat-soluble drugs to produce nervous system effects. Both atenolol and propranolol alter the electroencephalogram; atenolol affects body sway and alertness and propranolol impairs short-term memory and the ability to concentrate [ , ]. These results suggest that both lipophilic and hydrophilic beta-adrenoceptor antagonists can affect the central nervous system, although the effects may be subtle and difficult to demonstrate.

In 27 hypertensive patients aged 65 years or more, randomized to continue atenolol treatment for 20 weeks or to discontinue atenolol and start cilazapril, there was a significant improvement in the choice reaction time in the patients randomized to cilazapril [ ]. This study has confirmed previous reports that chronic beta-blockade can determine adverse effects on cognition in elderly patients. Withdrawal of beta-blockers should be considered in any elderly patient who has signs of mental impairment.

In a placebo-controlled trial of propranolol in 312 patients with diastolic hypertension, 13 tests of cognitive function were assessed at baseline, 3 months, and 12 months [ ]. Propranolol had no significant effects on 11 of the 13 tests. Compared with placebo, patients taking propranolol had fewer correct responses at 3 months and made more errors of commission.

Bipolar affective disorder

Bipolar depression affects 1% of the general population, and treatment resistance is a significant problem. The addition of pindolol can lead to significant improvement in depressed patients who are resistant to antidepressant drugs, such as selective serotonin reuptake inhibitors or phenelzine. Of 17 patients with refractory bipolar depression, in whom pindolol was added to augment the effect of antidepressant drugs, eight responded favorably [ ]. However, two developed transient hypomania, and one of these became psychotic after the resolution of hypomanic symptoms. In both cases transient hypomanic symptoms resolved without any other intervention, while psychosis required pindolol withdrawal.

Anxiety and depression have been reported after the use of nadolol, which is hydrophilic [ ]. In a study of the co-prescribing of antidepressants in 3218 new users of beta-blockers [ ], 6.4% had prescriptions for antidepressant drugs within 34 days, compared with 2.8% in a control population. Propranolol had the highest rate of co-prescribing (9.5%), followed by other lipophilic beta-blockers (3.9%) and hydrophilic beta-blockers (2.5%). In propranolol users, the risk of antidepressant use was 4.8 times greater than the control group, and was highest in those aged 20–39 (RR = 17; 95% CI = 14, 22).

Organic brain syndrome

The development of a severe organic brain syndrome has been reported in several patients taking beta-adrenoceptor antagonists regularly without a previous history of psychiatric illness [ ]. A similar phenomenon was seen in a young healthy woman who took propranolol 160 mg/day [ ]. The psychosis can follow initial therapy or dosage increases during long-term therapy [ ]. The symptoms, which include agitation, confusion, disorientation, anxiety, and hallucinations, may not respond to treatment with neuroleptic drugs, but subsides rapidly when the beta-blockers are withdrawn. Symptoms are also ameliorated by changing from propranolol to atenolol [ ].

Schizophrenia

A schizophrenia-like illness has also been seen in close relation to the initiation of propranolol therapy [ ].

Prolactin

• Reversible hyperprolactinemia with galactorrhea occurred in a 38-year-old woman taking atenolol for hypertension [ ].

Thyroid

Propranolol inhibits the conversion of thyroxine (T4) to tri-iodothyronine (T3) by peripheral tissues [ ], resulting in increased formation of inactive reverse T3. There have been several reports of hyperthyroxinemia in clinically euthyroid patients taking propranolol for non-thyroid reasons in high dosages (320–480 mg/day) [ , ]. The incidence was considered to be higher than could be accounted for by the development of spontaneous hyperthyroidism, but the mechanism is unknown.

The effect of beta-adrenoceptor antagonists on thyroid hormone metabolism is unlikely to play a significant role in their use in hyperthyroidism. Since d -propranolol has similar effects on thyroxine metabolism to those seen with the racemic mixture, membrane-stabilizing activity may be involved [ ].

In one case, beta-adrenoceptor blockade masked an unexpected thyroid crisis, resulting in severe cerebral dysfunction before the diagnosis was made [ ].

Hypoglycemia and blood glucose control

Hypoglycemia, producing loss of consciousness in some cases, can occur in non-diabetic individuals who are taking beta-adrenoceptor antagonists, particularly those who undergo prolonged fasting [ ] or severe exercise [ , ]. Patients on maintenance dialysis are also at risk [ ]. It has been suggested that non-selective drugs are most likely to produce hypoglycemia and that cardioselective drugs are to be preferred in at-risk patients [ ], but the same effect has been reported with atenolol under similar circumstances [ ].

Two children in whom propranolol was used to treat attention deficit disorders and anxiety became unarousable, with low heart rates and respiratory rates, due to hypoglycemia [ ]. Hypoglycemia can be caused by reduced glucose intake (fasting), increased utilization (hyperinsulinemia), or reduced production (enzymatic defects). One or more of these mechanisms can be responsible for hypoglycemia secondary to drugs. Children treated with propranolol may be at increased risk of hypoglycemia, particularly if they are fasting. Concomitant treatment with methylphenidate can increase the risk of this metabolic disorder.

However, contrary to popular belief, beta-adrenoceptor antagonists do not by themselves increase the risk of hypoglycemic episodes in insulin-treated diabetics, in whom their use was concluded to be generally safe [ ]. Indeed, in 20 such patients treated with diet or diet plus oral hypoglycemic agents, both propranolol and metoprolol produced small but significant increases in blood glucose concentrations after 4 weeks [ ]. The rise was considered clinically important in only a few patients.

However, in insulin-treated diabetics who become hypoglycemic, non-selective beta-adrenoceptor antagonists can mask the adrenaline-mediated symptoms, such as palpitation, tachycardia, and tremor; they can cause a rise in mean and diastolic blood pressures, due to unopposed alpha-adrenoceptor stimulation from catecholamines, because the beta ₂ -adrenoceptor-mediated vasodilator response is blocked [ ]; they can also impair the rate of rise of blood glucose toward normal [ ]. In contrast, cardioselective drugs mask hypoglycemic symptoms less [ ]; because of vascular sparing, they are less likely to be associated with a diastolic pressor response in the presence of catecholamines, although this has been reported with metoprolol [ ]; and delay in recovery from hypoglycemia is either less marked or undetectable with cardioselective drugs, such as atenolol or metoprolol. Thus, if insulin-requiring diabetics need to be treated with a beta-adrenoceptor antagonist, a cardioselective agent should always be chosen for reasons of safety, while allowing that this type of beta-blocker is associated with insulin resistance and can impair insulin sensitivity by 15–30% [ ], and hence increase insulin requirements.

People with diabetes have a much worse outcome after acute myocardial infarction, with a mortality rate at least twice that in non-diabetics. However, tight control of blood glucose, with immediate intensive insulin treatment during the peri-infarct period followed by intensive subcutaneous insulin treatment, was associated with a 30% reduction in mortality at 1 year, as reported in the DIGAMI study. In addition, the use of beta-blockers in this group of patients had an independent secondary preventive effect [ ]. The use of beta-blockers in diabetics with ischemic heart disease should be encouraged [ ].

In 686 hypertensive men treated for 15 years, beta-blockers were associated with a higher incidence of diabetes than thiazide diuretics [ ]. This was an uncontrolled study, but the observation deserves further study.

In a randomized controlled comparison of the effects of beta-adrenoceptor antagonists with different pharmacological profiles, namely metoprolol and carvedilol, on glycemic and metabolic control in 1235 hypertensive patients with type 2 diabetes already taking a blocker of the renin–angiotensin–aldosterone system, blood pressure reduction was similar in the two groups but the mean glycosylated hemoglobin increased significantly from baseline to the end of the study with metoprolol (0.15%; 95% CI = 0.08, 0.22) but not carvedilol (0.02%) [ ]. Insulin sensitivity improved with carvedilol (− 9.1%) but not metoprolol (− 2.0%). The between-group difference was − 7.2%. Progression to microalbuminuria was more common with metoprol than with carvedilol. Even if both agents were effective in reducing blood pressure and well tolerated, the use of carvedilol in addition to blockers of the renin–angiotensin–aldosterone system seems to be associated with a better metabolic profile in diabetic patients.