Cleveland Clinic Journal of Medicine

Inhibition of the renin-angiotensin-aldosterone system with angiotensin-converting enzyme (ACE) inhibitors and angiotensin II receptor blockers (ARBs) is widely used in the treatment of heart failure, hypertension, chronic kidney disease, and coronary artery disease with left ventricular dysfunction.

See related article

In this issue, Momoniat et al¹ review the benefits of ACE inhibitors and ARBs and how to manage adverse effects. I would like to add some of my own observations.

ARE ACE INHIBITORS REALLY BETTER THAN ARBs?

ACE inhibitors have been the cornerstone of treatment for patients with heart failure with reduced ejection fraction (HFrEF), in whom their use is associated with reduced rates of morbidity and death.^2,3 The use of ARBs in these patients is also associated with decreased rates of morbidity and death^4,5; however, in early comparisons, ACE inhibitors were deemed more effective in decreasing the incidence of myocardial infarction, cardiovascular death, and all-cause mortality in patients with hypertension, diabetes, and increased cardiovascular risk,⁶ and all-cause mortality in patients with HFrEF.⁷

This presumed superiority of ACE inhibitors over ARBs was thought to be a result of a greater vasodilatory effect caused by inhibiting the degradation of bradykinin and leading to increased levels of nitric oxide and vasoactive prostaglandins.⁸ Another proposed explanation was that because ARBs block angiotensin II AT1 receptors but not AT2 receptors, the increased stimulation of markedly upregulated AT2 receptors in atheromatous plaques in response to elevated serum levels of angiotensin II was deleterious.⁶ Therefore, ACE inhibitors have been recommended as first-line therapy by most guidelines, whereas ARBs are recommended as second-line therapy, when patients are unable to tolerate ACE inhibitors.

Nevertheless, the much debated differences in outcomes between ACE inhibitors and ARBs do not seem to be real and may have originated from a generational gap in the trials.

The ACE inhibitor trials were performed a decade earlier than the ARB trials. Indirect comparisons of their respective placebo-controlled trials assumed that the placebo groups used for comparison in the 2 sets of trials were similar.^9,10 Actually, the rate of cardiovascular disease decreased nearly 50% between the decades of 1990 to 2000 and 2000 to 2010, the likely result of aggressive primary and secondary prevention strategies in clinical practice, including revascularization and lipid-lowering therapy.¹⁰

In fact, a meta-regression analysis showed that the differences between ACE inhibitors and ARBs compared with placebo were due to higher event rates in the placebo groups in the ACE inhibitor trials than in the ARB trials for the outcomes of death, cardiovascular death, and myocardial infarction.¹¹ Sensitivity analyses restricted to trials published after 2000 to control for this generational gap showed similar efficacy with ACE inhibitors vs placebo and with ARBs vs placebo for all clinical outcomes.¹¹ Moreover, recent studies have shown that ARBs produce a greater decrease in cardiovascular events than ACE inhibitors, especially in patients with established cardiovascular disease.^12,13

An advantage of ARBs over ACE inhibitors is fewer adverse effects: in general, ARBs are better tolerated than ACE inhibitors.¹⁴ There are also ethnic differences in the risks of adverse reactions to these medications. African Americans have a higher risk of developing angioedema with ACE inhibitors compared with the rest of the US population, and Chinese Americans have a higher risk than whites of developing cough with ACE inhibitors.^9,15

In my medical practice, I try to make sure patients with HFrEF, hypertension, chronic kidney disease, and coronary artery disease with left ventricular dysfunction receive an inhibitor of the renin-angiotensin-aldosterone system.

Which agent?

I prefer ARBs because patients tolerate them better. I continue ACE inhibitors in patients who are already taking them without adverse effects, and I change to ARBs in patients who later become unable to tolerate ACE inhibitors.

Most antihypertensive agents increase the risk of incident gout, except for calcium channel blockers and losartan.¹⁶ Losartan is the only ARB with a uricosuric effect, although a mild one,^17,18 due to inhibition of the urate transporter 1,¹⁹ and therefore I prefer to use it instead of other ARBs or ACE inhibitors in patients who have a concomitant diagnosis of gout.

Which combinations of agents?

The addition of beta-blockers and mineralocorticoid receptor blockers to ACE inhibitors or ARBs is associated with a further decrease in the mortality risk for patients with HFrEF,^20–22 but some patients cannot tolerate these combinations or optimized doses of these medications because of worsening hypotension or increased risk of developing acute kidney injury or hyperkalemia.

In most cases, I try not to combine ACE inhibitors with ARBs. This combination may be useful in nondiabetic patients with proteinuria refractory to maximum treatment with 1 class of these agents, but it is associated with an increased risk of hyperkalemia or acute kidney injury in patients with diabetic nephropathy without improving rates of the clinical outcomes of death or cardiovascular events.²³ I prefer adding a daily low dose of a mineralocorticoid receptor blocker to an ACE inhibitor or an ARB, which is more effective in controlling refractory proteinuria.²⁴ This regimen is associated with decreased rates of mortality, cardiovascular mortality, and hospitalization for heart failure in patients with HFrEF,²² although it can lead to a higher frequency of hyperkalemia,²⁵ and patients on it require frequent dietary education and monitoring of serum potassium.

I avoid combining direct renin inhibitors with ACE inhibitors or ARBs, since this combination has been contraindicated by the US Food and Drug Administration due to lack of reduction in target-organ damage and an associated increased risk of hypotension, hyperkalemia, and kidney failure, and a slight increase in the risk of stroke or death in patients with diabetic nephropathy.²⁶

Valsartan-sacubitril

Neprilysin is a membrane-bound endopeptidase that degrades vasoactive peptides, including B-type natriuretic peptide and atrial natriuretic peptide.²⁷ The combination of the ARB valsartan and the neprilysin inhibitor sacubitril is associated with a 20% further decrease in rates of cardiovascular mortality and hospitalization and a 16% decrease in total mortality for patients with HFrEF compared with an ACE inhibitor, although there can also be more hypotension and angioedema with the combination.^27,28

Very importantly, an ACE inhibitor cannot be used together with valsartan-sacubitril due to increased risk of angioedema and cough. I change ACE inhibitors or ARBs to valsartan-sacubitril in patients with HFrEF who still have symptoms of heart failure. Interestingly, a network meta-analysis showed that the combination of valsartan-sacubitril plus a mineralocorticoid receptor blocker and a beta-blocker resulted in the greatest mortality reduction in patients with HFrEF.⁷ A word of caution, though: one can also expect an increased risk of hypotension, hyperkalemia, and kidney failure.

Monitoring

It is crucial to monitor blood pressure, serum potassium, and renal function in patients receiving ACE inhibitors, ARBs, mineralocorticoid receptor blockers, valsartan-sacubitril, or combinations of these medications, particularly in elderly patients, who are more susceptible to complications. I use a multidisciplinary approach in my clinic: a patient educator, dietitian, pharmacist, and advanced practice nurse play key roles in educating and monitoring patients for the development of possible complications from this therapy or interactions with other medications.

A recent population-based cohort study found an association of ACE inhibitor use with a 14% relative increase in lung cancer incidence after 10 years of use, compared with ARBs,²⁹ but this may not represent a large absolute risk (calculated number needed to harm of 2,970 after 10 years of ACE inhibitor use) and should be balanced against the improvement in morbidity and mortality gained with use of an ACE inhibitor. Additional studies with long-term follow-up are needed to investigate this possible association.

TAKE-HOME POINTS

• Blockade of the renin-angiotensin-aldosterone system is a cornerstone in the therapy of cardiovascular disease.
• ARBs are as effective as ACE inhibitors and have a better tolerability profile.
• ACE inhibitors cause more angioedema in African Americans and more cough in Chinese Americans than in the rest of the population.
• ACE inhibitors and most ARBs (except for losartan) increase the risk of gout.
• The combination of beta-blockers and mineralocorticoid receptor blockers with ACE inhibitors or ARBs and, lately, the use of the valsartan-sacubitril combination have been increasingly beneficial for patients with HFrEF.

Inhibition of the renin-angiotensin-aldosterone system with angiotensin-converting enzyme (ACE) inhibitors and angiotensin II receptor blockers (ARBs) is widely used in the treatment of heart failure, hypertension, chronic kidney disease, and coronary artery disease with left ventricular dysfunction.

See related article

In this issue, Momoniat et al¹ review the benefits of ACE inhibitors and ARBs and how to manage adverse effects. I would like to add some of my own observations.

ARE ACE INHIBITORS REALLY BETTER THAN ARBs?

ACE inhibitors have been the cornerstone of treatment for patients with heart failure with reduced ejection fraction (HFrEF), in whom their use is associated with reduced rates of morbidity and death.^2,3 The use of ARBs in these patients is also associated with decreased rates of morbidity and death^4,5; however, in early comparisons, ACE inhibitors were deemed more effective in decreasing the incidence of myocardial infarction, cardiovascular death, and all-cause mortality in patients with hypertension, diabetes, and increased cardiovascular risk,⁶ and all-cause mortality in patients with HFrEF.⁷

This presumed superiority of ACE inhibitors over ARBs was thought to be a result of a greater vasodilatory effect caused by inhibiting the degradation of bradykinin and leading to increased levels of nitric oxide and vasoactive prostaglandins.⁸ Another proposed explanation was that because ARBs block angiotensin II AT1 receptors but not AT2 receptors, the increased stimulation of markedly upregulated AT2 receptors in atheromatous plaques in response to elevated serum levels of angiotensin II was deleterious.⁶ Therefore, ACE inhibitors have been recommended as first-line therapy by most guidelines, whereas ARBs are recommended as second-line therapy, when patients are unable to tolerate ACE inhibitors.

Nevertheless, the much debated differences in outcomes between ACE inhibitors and ARBs do not seem to be real and may have originated from a generational gap in the trials.

The ACE inhibitor trials were performed a decade earlier than the ARB trials. Indirect comparisons of their respective placebo-controlled trials assumed that the placebo groups used for comparison in the 2 sets of trials were similar.^9,10 Actually, the rate of cardiovascular disease decreased nearly 50% between the decades of 1990 to 2000 and 2000 to 2010, the likely result of aggressive primary and secondary prevention strategies in clinical practice, including revascularization and lipid-lowering therapy.¹⁰

In fact, a meta-regression analysis showed that the differences between ACE inhibitors and ARBs compared with placebo were due to higher event rates in the placebo groups in the ACE inhibitor trials than in the ARB trials for the outcomes of death, cardiovascular death, and myocardial infarction.¹¹ Sensitivity analyses restricted to trials published after 2000 to control for this generational gap showed similar efficacy with ACE inhibitors vs placebo and with ARBs vs placebo for all clinical outcomes.¹¹ Moreover, recent studies have shown that ARBs produce a greater decrease in cardiovascular events than ACE inhibitors, especially in patients with established cardiovascular disease.^12,13

An advantage of ARBs over ACE inhibitors is fewer adverse effects: in general, ARBs are better tolerated than ACE inhibitors.¹⁴ There are also ethnic differences in the risks of adverse reactions to these medications. African Americans have a higher risk of developing angioedema with ACE inhibitors compared with the rest of the US population, and Chinese Americans have a higher risk than whites of developing cough with ACE inhibitors.^9,15

In my medical practice, I try to make sure patients with HFrEF, hypertension, chronic kidney disease, and coronary artery disease with left ventricular dysfunction receive an inhibitor of the renin-angiotensin-aldosterone system.

Which agent?

I prefer ARBs because patients tolerate them better. I continue ACE inhibitors in patients who are already taking them without adverse effects, and I change to ARBs in patients who later become unable to tolerate ACE inhibitors.

Most antihypertensive agents increase the risk of incident gout, except for calcium channel blockers and losartan.¹⁶ Losartan is the only ARB with a uricosuric effect, although a mild one,^17,18 due to inhibition of the urate transporter 1,¹⁹ and therefore I prefer to use it instead of other ARBs or ACE inhibitors in patients who have a concomitant diagnosis of gout.

Which combinations of agents?

The addition of beta-blockers and mineralocorticoid receptor blockers to ACE inhibitors or ARBs is associated with a further decrease in the mortality risk for patients with HFrEF,^20–22 but some patients cannot tolerate these combinations or optimized doses of these medications because of worsening hypotension or increased risk of developing acute kidney injury or hyperkalemia.

In most cases, I try not to combine ACE inhibitors with ARBs. This combination may be useful in nondiabetic patients with proteinuria refractory to maximum treatment with 1 class of these agents, but it is associated with an increased risk of hyperkalemia or acute kidney injury in patients with diabetic nephropathy without improving rates of the clinical outcomes of death or cardiovascular events.²³ I prefer adding a daily low dose of a mineralocorticoid receptor blocker to an ACE inhibitor or an ARB, which is more effective in controlling refractory proteinuria.²⁴ This regimen is associated with decreased rates of mortality, cardiovascular mortality, and hospitalization for heart failure in patients with HFrEF,²² although it can lead to a higher frequency of hyperkalemia,²⁵ and patients on it require frequent dietary education and monitoring of serum potassium.

I avoid combining direct renin inhibitors with ACE inhibitors or ARBs, since this combination has been contraindicated by the US Food and Drug Administration due to lack of reduction in target-organ damage and an associated increased risk of hypotension, hyperkalemia, and kidney failure, and a slight increase in the risk of stroke or death in patients with diabetic nephropathy.²⁶

Valsartan-sacubitril

Neprilysin is a membrane-bound endopeptidase that degrades vasoactive peptides, including B-type natriuretic peptide and atrial natriuretic peptide.²⁷ The combination of the ARB valsartan and the neprilysin inhibitor sacubitril is associated with a 20% further decrease in rates of cardiovascular mortality and hospitalization and a 16% decrease in total mortality for patients with HFrEF compared with an ACE inhibitor, although there can also be more hypotension and angioedema with the combination.^27,28

Very importantly, an ACE inhibitor cannot be used together with valsartan-sacubitril due to increased risk of angioedema and cough. I change ACE inhibitors or ARBs to valsartan-sacubitril in patients with HFrEF who still have symptoms of heart failure. Interestingly, a network meta-analysis showed that the combination of valsartan-sacubitril plus a mineralocorticoid receptor blocker and a beta-blocker resulted in the greatest mortality reduction in patients with HFrEF.⁷ A word of caution, though: one can also expect an increased risk of hypotension, hyperkalemia, and kidney failure.

Monitoring

It is crucial to monitor blood pressure, serum potassium, and renal function in patients receiving ACE inhibitors, ARBs, mineralocorticoid receptor blockers, valsartan-sacubitril, or combinations of these medications, particularly in elderly patients, who are more susceptible to complications. I use a multidisciplinary approach in my clinic: a patient educator, dietitian, pharmacist, and advanced practice nurse play key roles in educating and monitoring patients for the development of possible complications from this therapy or interactions with other medications.

A recent population-based cohort study found an association of ACE inhibitor use with a 14% relative increase in lung cancer incidence after 10 years of use, compared with ARBs,²⁹ but this may not represent a large absolute risk (calculated number needed to harm of 2,970 after 10 years of ACE inhibitor use) and should be balanced against the improvement in morbidity and mortality gained with use of an ACE inhibitor. Additional studies with long-term follow-up are needed to investigate this possible association.

TAKE-HOME POINTS

• Blockade of the renin-angiotensin-aldosterone system is a cornerstone in the therapy of cardiovascular disease.
• ARBs are as effective as ACE inhibitors and have a better tolerability profile.
• ACE inhibitors cause more angioedema in African Americans and more cough in Chinese Americans than in the rest of the population.
• ACE inhibitors and most ARBs (except for losartan) increase the risk of gout.
• The combination of beta-blockers and mineralocorticoid receptor blockers with ACE inhibitors or ARBs and, lately, the use of the valsartan-sacubitril combination have been increasingly beneficial for patients with HFrEF.

Inhibition of the renin-angiotensin-aldosterone system with angiotensin-converting enzyme (ACE) inhibitors and angiotensin II receptor blockers (ARBs) is widely used in the treatment of heart failure, hypertension, chronic kidney disease, and coronary artery disease with left ventricular dysfunction.

See related article

In this issue, Momoniat et al¹ review the benefits of ACE inhibitors and ARBs and how to manage adverse effects. I would like to add some of my own observations.

ARE ACE INHIBITORS REALLY BETTER THAN ARBs?

ACE inhibitors have been the cornerstone of treatment for patients with heart failure with reduced ejection fraction (HFrEF), in whom their use is associated with reduced rates of morbidity and death.^2,3 The use of ARBs in these patients is also associated with decreased rates of morbidity and death^4,5; however, in early comparisons, ACE inhibitors were deemed more effective in decreasing the incidence of myocardial infarction, cardiovascular death, and all-cause mortality in patients with hypertension, diabetes, and increased cardiovascular risk,⁶ and all-cause mortality in patients with HFrEF.⁷

This presumed superiority of ACE inhibitors over ARBs was thought to be a result of a greater vasodilatory effect caused by inhibiting the degradation of bradykinin and leading to increased levels of nitric oxide and vasoactive prostaglandins.⁸ Another proposed explanation was that because ARBs block angiotensin II AT1 receptors but not AT2 receptors, the increased stimulation of markedly upregulated AT2 receptors in atheromatous plaques in response to elevated serum levels of angiotensin II was deleterious.⁶ Therefore, ACE inhibitors have been recommended as first-line therapy by most guidelines, whereas ARBs are recommended as second-line therapy, when patients are unable to tolerate ACE inhibitors.

Nevertheless, the much debated differences in outcomes between ACE inhibitors and ARBs do not seem to be real and may have originated from a generational gap in the trials.

The ACE inhibitor trials were performed a decade earlier than the ARB trials. Indirect comparisons of their respective placebo-controlled trials assumed that the placebo groups used for comparison in the 2 sets of trials were similar.^9,10 Actually, the rate of cardiovascular disease decreased nearly 50% between the decades of 1990 to 2000 and 2000 to 2010, the likely result of aggressive primary and secondary prevention strategies in clinical practice, including revascularization and lipid-lowering therapy.¹⁰

In fact, a meta-regression analysis showed that the differences between ACE inhibitors and ARBs compared with placebo were due to higher event rates in the placebo groups in the ACE inhibitor trials than in the ARB trials for the outcomes of death, cardiovascular death, and myocardial infarction.¹¹ Sensitivity analyses restricted to trials published after 2000 to control for this generational gap showed similar efficacy with ACE inhibitors vs placebo and with ARBs vs placebo for all clinical outcomes.¹¹ Moreover, recent studies have shown that ARBs produce a greater decrease in cardiovascular events than ACE inhibitors, especially in patients with established cardiovascular disease.^12,13

An advantage of ARBs over ACE inhibitors is fewer adverse effects: in general, ARBs are better tolerated than ACE inhibitors.¹⁴ There are also ethnic differences in the risks of adverse reactions to these medications. African Americans have a higher risk of developing angioedema with ACE inhibitors compared with the rest of the US population, and Chinese Americans have a higher risk than whites of developing cough with ACE inhibitors.^9,15

In my medical practice, I try to make sure patients with HFrEF, hypertension, chronic kidney disease, and coronary artery disease with left ventricular dysfunction receive an inhibitor of the renin-angiotensin-aldosterone system.

Which agent?

I prefer ARBs because patients tolerate them better. I continue ACE inhibitors in patients who are already taking them without adverse effects, and I change to ARBs in patients who later become unable to tolerate ACE inhibitors.

Most antihypertensive agents increase the risk of incident gout, except for calcium channel blockers and losartan.¹⁶ Losartan is the only ARB with a uricosuric effect, although a mild one,^17,18 due to inhibition of the urate transporter 1,¹⁹ and therefore I prefer to use it instead of other ARBs or ACE inhibitors in patients who have a concomitant diagnosis of gout.

Which combinations of agents?

The addition of beta-blockers and mineralocorticoid receptor blockers to ACE inhibitors or ARBs is associated with a further decrease in the mortality risk for patients with HFrEF,^20–22 but some patients cannot tolerate these combinations or optimized doses of these medications because of worsening hypotension or increased risk of developing acute kidney injury or hyperkalemia.

In most cases, I try not to combine ACE inhibitors with ARBs. This combination may be useful in nondiabetic patients with proteinuria refractory to maximum treatment with 1 class of these agents, but it is associated with an increased risk of hyperkalemia or acute kidney injury in patients with diabetic nephropathy without improving rates of the clinical outcomes of death or cardiovascular events.²³ I prefer adding a daily low dose of a mineralocorticoid receptor blocker to an ACE inhibitor or an ARB, which is more effective in controlling refractory proteinuria.²⁴ This regimen is associated with decreased rates of mortality, cardiovascular mortality, and hospitalization for heart failure in patients with HFrEF,²² although it can lead to a higher frequency of hyperkalemia,²⁵ and patients on it require frequent dietary education and monitoring of serum potassium.

I avoid combining direct renin inhibitors with ACE inhibitors or ARBs, since this combination has been contraindicated by the US Food and Drug Administration due to lack of reduction in target-organ damage and an associated increased risk of hypotension, hyperkalemia, and kidney failure, and a slight increase in the risk of stroke or death in patients with diabetic nephropathy.²⁶

Valsartan-sacubitril

Neprilysin is a membrane-bound endopeptidase that degrades vasoactive peptides, including B-type natriuretic peptide and atrial natriuretic peptide.²⁷ The combination of the ARB valsartan and the neprilysin inhibitor sacubitril is associated with a 20% further decrease in rates of cardiovascular mortality and hospitalization and a 16% decrease in total mortality for patients with HFrEF compared with an ACE inhibitor, although there can also be more hypotension and angioedema with the combination.^27,28

Very importantly, an ACE inhibitor cannot be used together with valsartan-sacubitril due to increased risk of angioedema and cough. I change ACE inhibitors or ARBs to valsartan-sacubitril in patients with HFrEF who still have symptoms of heart failure. Interestingly, a network meta-analysis showed that the combination of valsartan-sacubitril plus a mineralocorticoid receptor blocker and a beta-blocker resulted in the greatest mortality reduction in patients with HFrEF.⁷ A word of caution, though: one can also expect an increased risk of hypotension, hyperkalemia, and kidney failure.

Monitoring

It is crucial to monitor blood pressure, serum potassium, and renal function in patients receiving ACE inhibitors, ARBs, mineralocorticoid receptor blockers, valsartan-sacubitril, or combinations of these medications, particularly in elderly patients, who are more susceptible to complications. I use a multidisciplinary approach in my clinic: a patient educator, dietitian, pharmacist, and advanced practice nurse play key roles in educating and monitoring patients for the development of possible complications from this therapy or interactions with other medications.

A recent population-based cohort study found an association of ACE inhibitor use with a 14% relative increase in lung cancer incidence after 10 years of use, compared with ARBs,²⁹ but this may not represent a large absolute risk (calculated number needed to harm of 2,970 after 10 years of ACE inhibitor use) and should be balanced against the improvement in morbidity and mortality gained with use of an ACE inhibitor. Additional studies with long-term follow-up are needed to investigate this possible association.

TAKE-HOME POINTS

• Blockade of the renin-angiotensin-aldosterone system is a cornerstone in the therapy of cardiovascular disease.
• ARBs are as effective as ACE inhibitors and have a better tolerability profile.
• ACE inhibitors cause more angioedema in African Americans and more cough in Chinese Americans than in the rest of the population.
• ACE inhibitors and most ARBs (except for losartan) increase the risk of gout.
• The combination of beta-blockers and mineralocorticoid receptor blockers with ACE inhibitors or ARBs and, lately, the use of the valsartan-sacubitril combination have been increasingly beneficial for patients with HFrEF.

When scientists discovered the band of hemoglobin A_1c during electrophoresis in the 1950s and 1960s and discerned it was elevated in patients with diabetes, little did they know the important role it would play in the diagnosis and treatment of diabetes in the decades to come.^1–3 Despite some caveats, a hemoglobin A_1c level of 6.5% or higher is diagnostic of diabetes across most populations, and hemoglobin A_1c goals ranging from 6.5% to 7.5% have been set for different subsets of patients depending on comorbidities, complications, risk of hypoglycemia, life expectancy, disease duration, patient preferences, and available resources.⁴

With a growing number of medications for diabetes—insulin in its various formulations and 11 other classes—hemoglobin A_1c targets can now be tailored to fit individual patient profiles. Although helping patients attain their glycemic goals is paramount, other factors should be considered when prescribing or changing a drug treatment regimen, such as cardiovascular risk reduction, weight control, avoidance of hypoglycemia, and minimizing out-of-pocket drug costs (Table 1).

CARDIOVASCULAR BENEFIT

Patients with type 2 diabetes have a 2 to 3 times higher risk of clinical atherosclerotic disease, according to 20 years of surveillance data from the Framingham cohort.⁵

Mixed results with intensive treatment

Reducing cardiovascular risk remains an important goal in diabetes management, but unfortunately, data from the long-term clinical trials aimed at reducing macrovascular risk with intensive glycemic management have been conflicting.

The United Kingdom Prospective Diabetes Study (UKPDS),⁶ which enrolled more than 4,000 patients with newly diagnosed type 2 diabetes, did not initially show a statistically significant difference in the incidence of myocardial infarction with intensive control vs conventional control, although intensive treatment did reduce the incidence of microvascular disease. However, 10 years after the trial ended, the incidence was 15% lower in the intensive-treatment group than in the conventional-treatment group, and the difference was statistically significant.⁷

A 10-year follow-up analysis of the Veterans Affairs Diabetes Trial (VADT)⁸ showed that patients who had been randomly assigned to intensive glucose control for 5.6 years had 8.6 fewer major cardiovascular events per 1,000 person-years than those assigned to standard therapy, but no improvement in median overall survival. The hemoglobin A_1c levels achieved during the trial were 6.9% and 8.4%, respectively.

In 2008, the US Food and Drug Administration (FDA)⁹ mandated that all new applications for diabetes drugs must include cardiovascular outcome studies. Therefore, we now have data on the cardiovascular benefits of two antihyperglycemic drug classes—incretins and sodium-glucose cotransporter 2 (SGLT2) inhibitors, making them attractive medications to target both cardiac and glucose concerns.

Incretins

The incretin drugs comprise 2 classes, glucagon-like peptide 1 (GLP-1) receptor agonists and dipeptidyl peptidase 4 (DPP-4) inhibitors.

Liraglutide. The Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Results (LEADER) trial¹⁰ compared liraglutide (a GLP-1 receptor agonist) and placebo in 9,000 patients with diabetes who either had or were at high risk of cardiovascular disease. Patients in the liraglutide group had a lower risk of the primary composite end point of death from cardiovascular causes or the first episode of nonfatal (including silent) myocardial infarction or nonfatal stroke, and a lower risk of cardiovascular death, all-cause mortality, and microvascular events than those in the placebo group. The number of patients who would need to be treated to prevent 1 event in 3 years was 66 in the analysis of the primary outcome and 98 in the analysis of death from any cause.⁹

Lixisenatide. The Evaluation of Lixisenatide in Acute Coronary Syndrome (ELIXA) trial¹¹ studied the effect of the once-daily GLP-1 receptor agonist lixisenatide on cardiovascular outcomes in 6,000 patients with type 2 diabetes with a recent coronary event. In contrast to LEADER, ELIXA did not show a cardiovascular benefit over placebo.

Exenatide. The Exenatide Study of Cardiovascular Event Lowering (EXSCEL)¹² assessed another GLP-1 extended-release drug, exenatide, in 14,000 patients, 73% of whom had established cardiovascular disease. In those patients, the drug had a modest benefit in terms of first occurrence of any component of the composite outcome of death from cardiovascular causes, nonfatal myocardial infarction, or nonfatal stroke (3-component major adverse cardiac event [MACE] outcome) in a time-to-event analysis, but the results were not statistically significant. However, the drug did significantly reduce all-cause mortality.

Semaglutide, another GLP-1 receptor agonist recently approved by the FDA, also showed benefit in patients who had cardiovascular disease or were at high risk, with significant reduction in the primary composite end point of death from cardiovascular causes or the first occurrence of nonfatal myocardial infarction (including silent) or nonfatal stroke.¹³

Dulaglutide, a newer GLP-1 drug, was associated with significantly reduced major adverse cardiovascular events (a composite end point of cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke) in about 9,900 patients with diabetes, with a median follow-up of more than 5 years. Only 31% of the patients in the trial had established cardiovascular disease.¹⁴

Comment. GLP-1 drugs as a class are a good option for patients with diabetes who require weight loss, and liraglutide is now FDA-approved for reduction of cardiovascular events in patients with type 2 diabetes with established cardiovascular disease. However, other factors should be considered when prescribing these drugs: they have adverse gastrointestinal effects, the cardiovascular benefit was not a class effect, they are relatively expensive, and they must be injected. Also, they should not be prescribed concurrently with a DPP-4 inhibitor because they target the same pathway.

The other class of diabetes drugs that have shown cardiovascular benefit are the SGLT2 inhibitors.

Empagliflozin. The Empagliflozin Cardiovascular Outcome Event Trial in Type 2 Diabetes Mellitus Patients (EMPA-REG)¹⁵ compared the efficacy of empagliflozin vs placebo in 7,000 patients with diabetes and cardiovascular disease and showed relative risk reductions of 38% in death from cardiovascular death, 31% in sudden death, and 35% in heart failure hospitalizations. Empagliflozin also showed benefit in terms of progression of kidney disease and occurrence of clinically relevant renal events in this population.¹⁶

Canagliflozin also has cardiovascular outcome data and showed significant benefit when compared with placebo in the primary outcome of the composite of death from cardiovascular causes, nonfatal myocardial infarction, or nonfatal stroke, but no significant effects on cardiovascular death or all-cause mortality.¹⁷ Data from this trial also suggested a nonsignificant benefit of canagliflozin in decreasing progression of albuminuria and in the composite outcome of a sustained 40% reduction in the estimated glomerular filtration rate (eGFR), the need for renal replacement therapy, or death from renal causes.

The above data led to an additional indication from the FDA for empagliflozin—and recently, canagliflozin—to prevent cardiovascular death in patients with diabetes with established disease, but other factors should be considered when prescribing them. Patients taking canagliflozin showed a significantly increased risk of amputation. SGLT2 inhibitors as a class also increase the risk of genital infections in men and women; this is an important consideration since patients with diabetes complain of vaginal fungal and urinary tract infections even without the use of these drugs. A higher incidence of fractures with canagliflozin should also be considered when using these medications in elderly and osteoporosis-prone patients at high risk of falling.

Dapagliflozin, the third drug in this class, was associated with a lower rate of hospitalization for heart failure in about 17,160 patients—including 10,186 without atherosclerotic cardiovascular disease—who were followed for a median of 4.2 years.¹⁸ It did not show benefit for the primary safety outcome, a composite of major adverse cardiovascular events defined as cardiovascular death, myocardial infarction, or ischemic stroke.

WEIGHT MANAGEMENT

Weight loss can help overweight patients reach their hemoglobin A_1c target.

Metformin should be continued as other drugs are added because it does not induce weight gain and may help with weight loss of up to 2 kg as shown in the Diabetes Prevention Program Outcomes Study.¹⁹

GLP-1 receptor agonists and SGLT2 inhibitors help with weight loss and are good additions to a basal insulin regimen to minimize weight gain.

Liraglutide was associated with a mean weight loss of 2.3 kg over 36 months of treatment compared with placebo in the LEADER trial.¹⁰

In the Trial to Evaluate Cardiovascular and Other Long-term Outcomes With Semaglutide in Subjects With Type 2 Diabetes (SUSTAIN-6),²⁰ the mean body weight in the semaglutide group, compared with the placebo group, was 2.9 kg lower in the group receiving a lower dose and 4.3 kg lower in the group receiving a higher dose of the drug.

In a 24-week trial in 182 patients with type 2 diabetes inadequately controlled on metformin, dapagliflozin produced a statistically significant weight reduction of 2.08 kg (95% confidence interval 2.84–1.31; P < .0001) compared with placebo.²¹

Lifestyle changes aimed at weight management should be emphasized and discussed at every visit.

HYPOGLYCEMIA RISK

Hypoglycemia is a major consideration when tailoring hemoglobin A_1c targets. In the Action to Control Cardiovascular Risk (ACCORD) trial,²² severe, symptomatic hypoglycemia increased the risk of death in both the intensive and conventional treatment groups. In VADT, the occurrence of a recent severe hypoglycemic event was the strongest independent predictor of death within 90 days. Further analysis showed that even though serious hypoglycemia occurred more often in the intensive therapy group, it was associated with progression of coronary artery calcification in the standard therapy group.²³ Hence, it is imperative that tight glycemic control not be achieved at the cost of severe or recurrent hypoglycemia.

In terms of hypoglycemia, metformin is an excellent medication. The American Diabetes Association²⁴ recommends metformin as the first-line therapy for newly diagnosed diabetes. Long-term follow-up data from UKPDS showed that metformin decreased mortality and the incidence of myocardial infarction and lowered treatment costs as well as the overall risk of hypoglycemia.²⁵ When prescribed, it should be titrated to the highest dose.

The FDA²⁶ has changed the prescribing information for metformin in patients with renal impairment. Metformin should not be started if the eGFR is less than 45 mL/min/1.73 m², but it can be continued if the patient is already receiving it and the eGFR is between 30 and 45. Previously, creatinine levels were used to define renal impairment and suitability for metformin. This change has increased the number of patients who can benefit from this medication.

In patients who have a contraindication to metformin, DPP-4 inhibitors can be considered, as they carry a low risk of hypoglycemia as well. Sulfonylureas should be used with caution in these patients, especially if their oral intake is variable. When sulfonylureas were compared to the DPP-4 inhibitor sitagliptin as an add-on to metformin, the rate of hypoglycemia was 32% in the sulfonylurea group vs 5% in the sitagliptin group.²⁷

Of the sulfonylureas, glipizide and glimepiride are better than glyburide because of a comparatively lower risk of hypoglycemia and a higher selectivity for binding the K_ATP channel on the pancreatic beta cell.²⁸

Meglitinides can be a good option for patients who skip meals, but they are more expensive than other generic oral hypoglycemic agents and require multiple daily dosing.

GLP-1 analogues also have a low risk of hypoglycemia but are only available in injectable formulations. Patients must be willing and able to perform the injections themselves.²⁹

In 2010, among US residents age 65 and older, 10.9 million (about 27%) had diabetes,³⁰ and this number is projected to increase to 26.7 million by 2050.³¹ This population is prone to hypoglycemia when treated with insulin and sulfonylureas. An injury sustained by a fall induced by hypoglycemia can be life-altering. In addition, no randomized clinical trials show the effect of tight glycemic control on complications in older patients with diabetes because patients older than 80 are often excluded.

A reasonable goal suggested by the European Diabetes Working Party for Older People 2011³² and reiterated by the American Geriatrics Society in 2013³³ is a hemoglobin A_1c between 7% and 7.5% for relatively healthy older patients and 7.5% to 8% or 8.5% in frail elderly patients with diabetes.

Consider prescribing medications that carry a low risk of hypoglycemia, can be dose-adjusted for kidney function, and do not rely on manual dexterity for administration (ie, do not require patients to give themselves injections). These include metformin and DPP-4 inhibitors.

DRUG COMBINATIONS

Polypharmacy is a concern for all patients with diabetes, especially since it increases the risk of drug interactions and adverse effects, increases out-of-pocket costs, and decreases the likelihood that patients will remain adherent to their treatment regimen. The use of combination medications can reduce the number of pills or injections required, as well as copayments.

Due to concern for multiple drug-drug interactions (and also due to the progressive nature of diabetes), many people with type 2 diabetes are given insulin in lieu of pills to lower their blood glucose. In addition to premixed insulin combinations (such as combinations of neutral protamine Hagedorn and regular insulin or combinations of insulin analogues), long-acting basal insulins can now be prescribed with a GLP-1 drug in fixed-dose combinations such as insulin glargine plus lixisenatide and insulin degludec plus liraglutide.

COST CONSIDERATIONS

It is important to discuss medication cost with patients, because many newer diabetic drugs are expensive and add to the financial burden of patients already paying for multiple medications, such as antihypertensives and statins.

Metformin and sulfonylureas are less expensive alternatives for patients who cannot afford GLP-1 analogues or SGLT2 inhibitors. Even within the same drug class, the formulary-preferred drug may be cheaper than the nonformulary alternative. Thus, it is helpful to research formulary alternatives before discussing treatment regimens with patients.

When scientists discovered the band of hemoglobin A_1c during electrophoresis in the 1950s and 1960s and discerned it was elevated in patients with diabetes, little did they know the important role it would play in the diagnosis and treatment of diabetes in the decades to come.^1–3 Despite some caveats, a hemoglobin A_1c level of 6.5% or higher is diagnostic of diabetes across most populations, and hemoglobin A_1c goals ranging from 6.5% to 7.5% have been set for different subsets of patients depending on comorbidities, complications, risk of hypoglycemia, life expectancy, disease duration, patient preferences, and available resources.⁴

With a growing number of medications for diabetes—insulin in its various formulations and 11 other classes—hemoglobin A_1c targets can now be tailored to fit individual patient profiles. Although helping patients attain their glycemic goals is paramount, other factors should be considered when prescribing or changing a drug treatment regimen, such as cardiovascular risk reduction, weight control, avoidance of hypoglycemia, and minimizing out-of-pocket drug costs (Table 1).

CARDIOVASCULAR BENEFIT

Patients with type 2 diabetes have a 2 to 3 times higher risk of clinical atherosclerotic disease, according to 20 years of surveillance data from the Framingham cohort.⁵

Mixed results with intensive treatment

Reducing cardiovascular risk remains an important goal in diabetes management, but unfortunately, data from the long-term clinical trials aimed at reducing macrovascular risk with intensive glycemic management have been conflicting.

The United Kingdom Prospective Diabetes Study (UKPDS),⁶ which enrolled more than 4,000 patients with newly diagnosed type 2 diabetes, did not initially show a statistically significant difference in the incidence of myocardial infarction with intensive control vs conventional control, although intensive treatment did reduce the incidence of microvascular disease. However, 10 years after the trial ended, the incidence was 15% lower in the intensive-treatment group than in the conventional-treatment group, and the difference was statistically significant.⁷

A 10-year follow-up analysis of the Veterans Affairs Diabetes Trial (VADT)⁸ showed that patients who had been randomly assigned to intensive glucose control for 5.6 years had 8.6 fewer major cardiovascular events per 1,000 person-years than those assigned to standard therapy, but no improvement in median overall survival. The hemoglobin A_1c levels achieved during the trial were 6.9% and 8.4%, respectively.

In 2008, the US Food and Drug Administration (FDA)⁹ mandated that all new applications for diabetes drugs must include cardiovascular outcome studies. Therefore, we now have data on the cardiovascular benefits of two antihyperglycemic drug classes—incretins and sodium-glucose cotransporter 2 (SGLT2) inhibitors, making them attractive medications to target both cardiac and glucose concerns.

Incretins

The incretin drugs comprise 2 classes, glucagon-like peptide 1 (GLP-1) receptor agonists and dipeptidyl peptidase 4 (DPP-4) inhibitors.

Liraglutide. The Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Results (LEADER) trial¹⁰ compared liraglutide (a GLP-1 receptor agonist) and placebo in 9,000 patients with diabetes who either had or were at high risk of cardiovascular disease. Patients in the liraglutide group had a lower risk of the primary composite end point of death from cardiovascular causes or the first episode of nonfatal (including silent) myocardial infarction or nonfatal stroke, and a lower risk of cardiovascular death, all-cause mortality, and microvascular events than those in the placebo group. The number of patients who would need to be treated to prevent 1 event in 3 years was 66 in the analysis of the primary outcome and 98 in the analysis of death from any cause.⁹

Lixisenatide. The Evaluation of Lixisenatide in Acute Coronary Syndrome (ELIXA) trial¹¹ studied the effect of the once-daily GLP-1 receptor agonist lixisenatide on cardiovascular outcomes in 6,000 patients with type 2 diabetes with a recent coronary event. In contrast to LEADER, ELIXA did not show a cardiovascular benefit over placebo.

Exenatide. The Exenatide Study of Cardiovascular Event Lowering (EXSCEL)¹² assessed another GLP-1 extended-release drug, exenatide, in 14,000 patients, 73% of whom had established cardiovascular disease. In those patients, the drug had a modest benefit in terms of first occurrence of any component of the composite outcome of death from cardiovascular causes, nonfatal myocardial infarction, or nonfatal stroke (3-component major adverse cardiac event [MACE] outcome) in a time-to-event analysis, but the results were not statistically significant. However, the drug did significantly reduce all-cause mortality.

Semaglutide, another GLP-1 receptor agonist recently approved by the FDA, also showed benefit in patients who had cardiovascular disease or were at high risk, with significant reduction in the primary composite end point of death from cardiovascular causes or the first occurrence of nonfatal myocardial infarction (including silent) or nonfatal stroke.¹³

Dulaglutide, a newer GLP-1 drug, was associated with significantly reduced major adverse cardiovascular events (a composite end point of cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke) in about 9,900 patients with diabetes, with a median follow-up of more than 5 years. Only 31% of the patients in the trial had established cardiovascular disease.¹⁴

Comment. GLP-1 drugs as a class are a good option for patients with diabetes who require weight loss, and liraglutide is now FDA-approved for reduction of cardiovascular events in patients with type 2 diabetes with established cardiovascular disease. However, other factors should be considered when prescribing these drugs: they have adverse gastrointestinal effects, the cardiovascular benefit was not a class effect, they are relatively expensive, and they must be injected. Also, they should not be prescribed concurrently with a DPP-4 inhibitor because they target the same pathway.

The other class of diabetes drugs that have shown cardiovascular benefit are the SGLT2 inhibitors.

Empagliflozin. The Empagliflozin Cardiovascular Outcome Event Trial in Type 2 Diabetes Mellitus Patients (EMPA-REG)¹⁵ compared the efficacy of empagliflozin vs placebo in 7,000 patients with diabetes and cardiovascular disease and showed relative risk reductions of 38% in death from cardiovascular death, 31% in sudden death, and 35% in heart failure hospitalizations. Empagliflozin also showed benefit in terms of progression of kidney disease and occurrence of clinically relevant renal events in this population.¹⁶

Canagliflozin also has cardiovascular outcome data and showed significant benefit when compared with placebo in the primary outcome of the composite of death from cardiovascular causes, nonfatal myocardial infarction, or nonfatal stroke, but no significant effects on cardiovascular death or all-cause mortality.¹⁷ Data from this trial also suggested a nonsignificant benefit of canagliflozin in decreasing progression of albuminuria and in the composite outcome of a sustained 40% reduction in the estimated glomerular filtration rate (eGFR), the need for renal replacement therapy, or death from renal causes.

The above data led to an additional indication from the FDA for empagliflozin—and recently, canagliflozin—to prevent cardiovascular death in patients with diabetes with established disease, but other factors should be considered when prescribing them. Patients taking canagliflozin showed a significantly increased risk of amputation. SGLT2 inhibitors as a class also increase the risk of genital infections in men and women; this is an important consideration since patients with diabetes complain of vaginal fungal and urinary tract infections even without the use of these drugs. A higher incidence of fractures with canagliflozin should also be considered when using these medications in elderly and osteoporosis-prone patients at high risk of falling.

Dapagliflozin, the third drug in this class, was associated with a lower rate of hospitalization for heart failure in about 17,160 patients—including 10,186 without atherosclerotic cardiovascular disease—who were followed for a median of 4.2 years.¹⁸ It did not show benefit for the primary safety outcome, a composite of major adverse cardiovascular events defined as cardiovascular death, myocardial infarction, or ischemic stroke.

WEIGHT MANAGEMENT

Weight loss can help overweight patients reach their hemoglobin A_1c target.

Metformin should be continued as other drugs are added because it does not induce weight gain and may help with weight loss of up to 2 kg as shown in the Diabetes Prevention Program Outcomes Study.¹⁹

GLP-1 receptor agonists and SGLT2 inhibitors help with weight loss and are good additions to a basal insulin regimen to minimize weight gain.

Liraglutide was associated with a mean weight loss of 2.3 kg over 36 months of treatment compared with placebo in the LEADER trial.¹⁰

In the Trial to Evaluate Cardiovascular and Other Long-term Outcomes With Semaglutide in Subjects With Type 2 Diabetes (SUSTAIN-6),²⁰ the mean body weight in the semaglutide group, compared with the placebo group, was 2.9 kg lower in the group receiving a lower dose and 4.3 kg lower in the group receiving a higher dose of the drug.

In a 24-week trial in 182 patients with type 2 diabetes inadequately controlled on metformin, dapagliflozin produced a statistically significant weight reduction of 2.08 kg (95% confidence interval 2.84–1.31; P < .0001) compared with placebo.²¹

Lifestyle changes aimed at weight management should be emphasized and discussed at every visit.

HYPOGLYCEMIA RISK

Hypoglycemia is a major consideration when tailoring hemoglobin A_1c targets. In the Action to Control Cardiovascular Risk (ACCORD) trial,²² severe, symptomatic hypoglycemia increased the risk of death in both the intensive and conventional treatment groups. In VADT, the occurrence of a recent severe hypoglycemic event was the strongest independent predictor of death within 90 days. Further analysis showed that even though serious hypoglycemia occurred more often in the intensive therapy group, it was associated with progression of coronary artery calcification in the standard therapy group.²³ Hence, it is imperative that tight glycemic control not be achieved at the cost of severe or recurrent hypoglycemia.

In terms of hypoglycemia, metformin is an excellent medication. The American Diabetes Association²⁴ recommends metformin as the first-line therapy for newly diagnosed diabetes. Long-term follow-up data from UKPDS showed that metformin decreased mortality and the incidence of myocardial infarction and lowered treatment costs as well as the overall risk of hypoglycemia.²⁵ When prescribed, it should be titrated to the highest dose.

The FDA²⁶ has changed the prescribing information for metformin in patients with renal impairment. Metformin should not be started if the eGFR is less than 45 mL/min/1.73 m², but it can be continued if the patient is already receiving it and the eGFR is between 30 and 45. Previously, creatinine levels were used to define renal impairment and suitability for metformin. This change has increased the number of patients who can benefit from this medication.

In patients who have a contraindication to metformin, DPP-4 inhibitors can be considered, as they carry a low risk of hypoglycemia as well. Sulfonylureas should be used with caution in these patients, especially if their oral intake is variable. When sulfonylureas were compared to the DPP-4 inhibitor sitagliptin as an add-on to metformin, the rate of hypoglycemia was 32% in the sulfonylurea group vs 5% in the sitagliptin group.²⁷

Of the sulfonylureas, glipizide and glimepiride are better than glyburide because of a comparatively lower risk of hypoglycemia and a higher selectivity for binding the K_ATP channel on the pancreatic beta cell.²⁸

Meglitinides can be a good option for patients who skip meals, but they are more expensive than other generic oral hypoglycemic agents and require multiple daily dosing.

GLP-1 analogues also have a low risk of hypoglycemia but are only available in injectable formulations. Patients must be willing and able to perform the injections themselves.²⁹

In 2010, among US residents age 65 and older, 10.9 million (about 27%) had diabetes,³⁰ and this number is projected to increase to 26.7 million by 2050.³¹ This population is prone to hypoglycemia when treated with insulin and sulfonylureas. An injury sustained by a fall induced by hypoglycemia can be life-altering. In addition, no randomized clinical trials show the effect of tight glycemic control on complications in older patients with diabetes because patients older than 80 are often excluded.

A reasonable goal suggested by the European Diabetes Working Party for Older People 2011³² and reiterated by the American Geriatrics Society in 2013³³ is a hemoglobin A_1c between 7% and 7.5% for relatively healthy older patients and 7.5% to 8% or 8.5% in frail elderly patients with diabetes.

Consider prescribing medications that carry a low risk of hypoglycemia, can be dose-adjusted for kidney function, and do not rely on manual dexterity for administration (ie, do not require patients to give themselves injections). These include metformin and DPP-4 inhibitors.

DRUG COMBINATIONS

Polypharmacy is a concern for all patients with diabetes, especially since it increases the risk of drug interactions and adverse effects, increases out-of-pocket costs, and decreases the likelihood that patients will remain adherent to their treatment regimen. The use of combination medications can reduce the number of pills or injections required, as well as copayments.

Due to concern for multiple drug-drug interactions (and also due to the progressive nature of diabetes), many people with type 2 diabetes are given insulin in lieu of pills to lower their blood glucose. In addition to premixed insulin combinations (such as combinations of neutral protamine Hagedorn and regular insulin or combinations of insulin analogues), long-acting basal insulins can now be prescribed with a GLP-1 drug in fixed-dose combinations such as insulin glargine plus lixisenatide and insulin degludec plus liraglutide.

COST CONSIDERATIONS

It is important to discuss medication cost with patients, because many newer diabetic drugs are expensive and add to the financial burden of patients already paying for multiple medications, such as antihypertensives and statins.

Metformin and sulfonylureas are less expensive alternatives for patients who cannot afford GLP-1 analogues or SGLT2 inhibitors. Even within the same drug class, the formulary-preferred drug may be cheaper than the nonformulary alternative. Thus, it is helpful to research formulary alternatives before discussing treatment regimens with patients.

When scientists discovered the band of hemoglobin A_1c during electrophoresis in the 1950s and 1960s and discerned it was elevated in patients with diabetes, little did they know the important role it would play in the diagnosis and treatment of diabetes in the decades to come.^1–3 Despite some caveats, a hemoglobin A_1c level of 6.5% or higher is diagnostic of diabetes across most populations, and hemoglobin A_1c goals ranging from 6.5% to 7.5% have been set for different subsets of patients depending on comorbidities, complications, risk of hypoglycemia, life expectancy, disease duration, patient preferences, and available resources.⁴

With a growing number of medications for diabetes—insulin in its various formulations and 11 other classes—hemoglobin A_1c targets can now be tailored to fit individual patient profiles. Although helping patients attain their glycemic goals is paramount, other factors should be considered when prescribing or changing a drug treatment regimen, such as cardiovascular risk reduction, weight control, avoidance of hypoglycemia, and minimizing out-of-pocket drug costs (Table 1).

CARDIOVASCULAR BENEFIT

Patients with type 2 diabetes have a 2 to 3 times higher risk of clinical atherosclerotic disease, according to 20 years of surveillance data from the Framingham cohort.⁵

Mixed results with intensive treatment

Reducing cardiovascular risk remains an important goal in diabetes management, but unfortunately, data from the long-term clinical trials aimed at reducing macrovascular risk with intensive glycemic management have been conflicting.

The United Kingdom Prospective Diabetes Study (UKPDS),⁶ which enrolled more than 4,000 patients with newly diagnosed type 2 diabetes, did not initially show a statistically significant difference in the incidence of myocardial infarction with intensive control vs conventional control, although intensive treatment did reduce the incidence of microvascular disease. However, 10 years after the trial ended, the incidence was 15% lower in the intensive-treatment group than in the conventional-treatment group, and the difference was statistically significant.⁷

A 10-year follow-up analysis of the Veterans Affairs Diabetes Trial (VADT)⁸ showed that patients who had been randomly assigned to intensive glucose control for 5.6 years had 8.6 fewer major cardiovascular events per 1,000 person-years than those assigned to standard therapy, but no improvement in median overall survival. The hemoglobin A_1c levels achieved during the trial were 6.9% and 8.4%, respectively.

In 2008, the US Food and Drug Administration (FDA)⁹ mandated that all new applications for diabetes drugs must include cardiovascular outcome studies. Therefore, we now have data on the cardiovascular benefits of two antihyperglycemic drug classes—incretins and sodium-glucose cotransporter 2 (SGLT2) inhibitors, making them attractive medications to target both cardiac and glucose concerns.

Incretins

The incretin drugs comprise 2 classes, glucagon-like peptide 1 (GLP-1) receptor agonists and dipeptidyl peptidase 4 (DPP-4) inhibitors.

Liraglutide. The Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Results (LEADER) trial¹⁰ compared liraglutide (a GLP-1 receptor agonist) and placebo in 9,000 patients with diabetes who either had or were at high risk of cardiovascular disease. Patients in the liraglutide group had a lower risk of the primary composite end point of death from cardiovascular causes or the first episode of nonfatal (including silent) myocardial infarction or nonfatal stroke, and a lower risk of cardiovascular death, all-cause mortality, and microvascular events than those in the placebo group. The number of patients who would need to be treated to prevent 1 event in 3 years was 66 in the analysis of the primary outcome and 98 in the analysis of death from any cause.⁹

Lixisenatide. The Evaluation of Lixisenatide in Acute Coronary Syndrome (ELIXA) trial¹¹ studied the effect of the once-daily GLP-1 receptor agonist lixisenatide on cardiovascular outcomes in 6,000 patients with type 2 diabetes with a recent coronary event. In contrast to LEADER, ELIXA did not show a cardiovascular benefit over placebo.

Exenatide. The Exenatide Study of Cardiovascular Event Lowering (EXSCEL)¹² assessed another GLP-1 extended-release drug, exenatide, in 14,000 patients, 73% of whom had established cardiovascular disease. In those patients, the drug had a modest benefit in terms of first occurrence of any component of the composite outcome of death from cardiovascular causes, nonfatal myocardial infarction, or nonfatal stroke (3-component major adverse cardiac event [MACE] outcome) in a time-to-event analysis, but the results were not statistically significant. However, the drug did significantly reduce all-cause mortality.

Semaglutide, another GLP-1 receptor agonist recently approved by the FDA, also showed benefit in patients who had cardiovascular disease or were at high risk, with significant reduction in the primary composite end point of death from cardiovascular causes or the first occurrence of nonfatal myocardial infarction (including silent) or nonfatal stroke.¹³

Dulaglutide, a newer GLP-1 drug, was associated with significantly reduced major adverse cardiovascular events (a composite end point of cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke) in about 9,900 patients with diabetes, with a median follow-up of more than 5 years. Only 31% of the patients in the trial had established cardiovascular disease.¹⁴

Comment. GLP-1 drugs as a class are a good option for patients with diabetes who require weight loss, and liraglutide is now FDA-approved for reduction of cardiovascular events in patients with type 2 diabetes with established cardiovascular disease. However, other factors should be considered when prescribing these drugs: they have adverse gastrointestinal effects, the cardiovascular benefit was not a class effect, they are relatively expensive, and they must be injected. Also, they should not be prescribed concurrently with a DPP-4 inhibitor because they target the same pathway.

The other class of diabetes drugs that have shown cardiovascular benefit are the SGLT2 inhibitors.

Empagliflozin. The Empagliflozin Cardiovascular Outcome Event Trial in Type 2 Diabetes Mellitus Patients (EMPA-REG)¹⁵ compared the efficacy of empagliflozin vs placebo in 7,000 patients with diabetes and cardiovascular disease and showed relative risk reductions of 38% in death from cardiovascular death, 31% in sudden death, and 35% in heart failure hospitalizations. Empagliflozin also showed benefit in terms of progression of kidney disease and occurrence of clinically relevant renal events in this population.¹⁶

Canagliflozin also has cardiovascular outcome data and showed significant benefit when compared with placebo in the primary outcome of the composite of death from cardiovascular causes, nonfatal myocardial infarction, or nonfatal stroke, but no significant effects on cardiovascular death or all-cause mortality.¹⁷ Data from this trial also suggested a nonsignificant benefit of canagliflozin in decreasing progression of albuminuria and in the composite outcome of a sustained 40% reduction in the estimated glomerular filtration rate (eGFR), the need for renal replacement therapy, or death from renal causes.

The above data led to an additional indication from the FDA for empagliflozin—and recently, canagliflozin—to prevent cardiovascular death in patients with diabetes with established disease, but other factors should be considered when prescribing them. Patients taking canagliflozin showed a significantly increased risk of amputation. SGLT2 inhibitors as a class also increase the risk of genital infections in men and women; this is an important consideration since patients with diabetes complain of vaginal fungal and urinary tract infections even without the use of these drugs. A higher incidence of fractures with canagliflozin should also be considered when using these medications in elderly and osteoporosis-prone patients at high risk of falling.

Dapagliflozin, the third drug in this class, was associated with a lower rate of hospitalization for heart failure in about 17,160 patients—including 10,186 without atherosclerotic cardiovascular disease—who were followed for a median of 4.2 years.¹⁸ It did not show benefit for the primary safety outcome, a composite of major adverse cardiovascular events defined as cardiovascular death, myocardial infarction, or ischemic stroke.

WEIGHT MANAGEMENT

Weight loss can help overweight patients reach their hemoglobin A_1c target.

Metformin should be continued as other drugs are added because it does not induce weight gain and may help with weight loss of up to 2 kg as shown in the Diabetes Prevention Program Outcomes Study.¹⁹

GLP-1 receptor agonists and SGLT2 inhibitors help with weight loss and are good additions to a basal insulin regimen to minimize weight gain.

Liraglutide was associated with a mean weight loss of 2.3 kg over 36 months of treatment compared with placebo in the LEADER trial.¹⁰

In the Trial to Evaluate Cardiovascular and Other Long-term Outcomes With Semaglutide in Subjects With Type 2 Diabetes (SUSTAIN-6),²⁰ the mean body weight in the semaglutide group, compared with the placebo group, was 2.9 kg lower in the group receiving a lower dose and 4.3 kg lower in the group receiving a higher dose of the drug.

In a 24-week trial in 182 patients with type 2 diabetes inadequately controlled on metformin, dapagliflozin produced a statistically significant weight reduction of 2.08 kg (95% confidence interval 2.84–1.31; P < .0001) compared with placebo.²¹

Lifestyle changes aimed at weight management should be emphasized and discussed at every visit.

HYPOGLYCEMIA RISK

Hypoglycemia is a major consideration when tailoring hemoglobin A_1c targets. In the Action to Control Cardiovascular Risk (ACCORD) trial,²² severe, symptomatic hypoglycemia increased the risk of death in both the intensive and conventional treatment groups. In VADT, the occurrence of a recent severe hypoglycemic event was the strongest independent predictor of death within 90 days. Further analysis showed that even though serious hypoglycemia occurred more often in the intensive therapy group, it was associated with progression of coronary artery calcification in the standard therapy group.²³ Hence, it is imperative that tight glycemic control not be achieved at the cost of severe or recurrent hypoglycemia.

In terms of hypoglycemia, metformin is an excellent medication. The American Diabetes Association²⁴ recommends metformin as the first-line therapy for newly diagnosed diabetes. Long-term follow-up data from UKPDS showed that metformin decreased mortality and the incidence of myocardial infarction and lowered treatment costs as well as the overall risk of hypoglycemia.²⁵ When prescribed, it should be titrated to the highest dose.

The FDA²⁶ has changed the prescribing information for metformin in patients with renal impairment. Metformin should not be started if the eGFR is less than 45 mL/min/1.73 m², but it can be continued if the patient is already receiving it and the eGFR is between 30 and 45. Previously, creatinine levels were used to define renal impairment and suitability for metformin. This change has increased the number of patients who can benefit from this medication.

In patients who have a contraindication to metformin, DPP-4 inhibitors can be considered, as they carry a low risk of hypoglycemia as well. Sulfonylureas should be used with caution in these patients, especially if their oral intake is variable. When sulfonylureas were compared to the DPP-4 inhibitor sitagliptin as an add-on to metformin, the rate of hypoglycemia was 32% in the sulfonylurea group vs 5% in the sitagliptin group.²⁷

Of the sulfonylureas, glipizide and glimepiride are better than glyburide because of a comparatively lower risk of hypoglycemia and a higher selectivity for binding the K_ATP channel on the pancreatic beta cell.²⁸

Meglitinides can be a good option for patients who skip meals, but they are more expensive than other generic oral hypoglycemic agents and require multiple daily dosing.

GLP-1 analogues also have a low risk of hypoglycemia but are only available in injectable formulations. Patients must be willing and able to perform the injections themselves.²⁹

In 2010, among US residents age 65 and older, 10.9 million (about 27%) had diabetes,³⁰ and this number is projected to increase to 26.7 million by 2050.³¹ This population is prone to hypoglycemia when treated with insulin and sulfonylureas. An injury sustained by a fall induced by hypoglycemia can be life-altering. In addition, no randomized clinical trials show the effect of tight glycemic control on complications in older patients with diabetes because patients older than 80 are often excluded.

A reasonable goal suggested by the European Diabetes Working Party for Older People 2011³² and reiterated by the American Geriatrics Society in 2013³³ is a hemoglobin A_1c between 7% and 7.5% for relatively healthy older patients and 7.5% to 8% or 8.5% in frail elderly patients with diabetes.

Consider prescribing medications that carry a low risk of hypoglycemia, can be dose-adjusted for kidney function, and do not rely on manual dexterity for administration (ie, do not require patients to give themselves injections). These include metformin and DPP-4 inhibitors.

DRUG COMBINATIONS

Polypharmacy is a concern for all patients with diabetes, especially since it increases the risk of drug interactions and adverse effects, increases out-of-pocket costs, and decreases the likelihood that patients will remain adherent to their treatment regimen. The use of combination medications can reduce the number of pills or injections required, as well as copayments.

Due to concern for multiple drug-drug interactions (and also due to the progressive nature of diabetes), many people with type 2 diabetes are given insulin in lieu of pills to lower their blood glucose. In addition to premixed insulin combinations (such as combinations of neutral protamine Hagedorn and regular insulin or combinations of insulin analogues), long-acting basal insulins can now be prescribed with a GLP-1 drug in fixed-dose combinations such as insulin glargine plus lixisenatide and insulin degludec plus liraglutide.

COST CONSIDERATIONS

It is important to discuss medication cost with patients, because many newer diabetic drugs are expensive and add to the financial burden of patients already paying for multiple medications, such as antihypertensives and statins.

Metformin and sulfonylureas are less expensive alternatives for patients who cannot afford GLP-1 analogues or SGLT2 inhibitors. Even within the same drug class, the formulary-preferred drug may be cheaper than the nonformulary alternative. Thus, it is helpful to research formulary alternatives before discussing treatment regimens with patients.

A highly active, water- and alcohol-soluble, basic pressor substance is formed when renin and renin-activator interact, for which we suggest the name “angiotonin.”

—Irvine H. Page and O.M. Helmer, 1940.¹

The renin-angiotensin-aldosterone system regulates salt and, in part, water homeostasis, and therefore blood pressure and fluid balance through its actions on the heart, kidneys, and blood vessels.² Drugs that target this system—angiotensin-converting enzyme (ACE) inhibitors and angiotensin II receptor blockers (ARBs)—are used primarily to treat hypertension and also to treat chronic kidney disease and heart failure with reduced ejection fraction.

See related editorial

Controlling blood pressure is important, as hypertension increases the risk of myocardial infarction, cerebrovascular events, and progression of chronic kidney disease, which itself is a risk factor for cardiovascular disease. However, the benefit of these drugs is only partly due to their effect on blood pressure. They also reduce proteinuria, which is a graded risk factor for progression of kidney disease as well as morbidity and death from vascular events.³

Despite the benefits of ACE inhibitors and ARBs, concern about their adverse effects—especially hyperkalemia and a decline in renal function—has led to their underuse in patients likely to derive the greatest benefit.³

ACE INHIBITORS AND ARBs

ACE inhibitors, as their name indicates, inhibit conversion of angiotensin I to angiotensin II by ACE, resulting in vasodilation of the efferent arteriole and a drop in blood pressure. Inhibition of ACE, a kininase, also results in a rise in kinins. One of these, bradykinin, is associated with some of the side effects of this class of drugs such as cough, which affects 5% to 20% of patients.⁴ Elevation of bradykinin is also believed to account for ACE inhibitor-induced angioedema, an uncommon but potentially serious side effect. Kinins are also associated with desirable effects such as lowering blood pressure, increasing insulin sensitivity, and dilating blood vessels.

ARBs were developed as an alternative for patients unable to tolerate the adverse effects of ACE inhibitors. While ACE inhibitors reduce the activity of angiotensin II at both the AT1 and AT2 receptors, ARBs block only the AT1 receptors, thereby inhibiting their vasoconstricting activity on smooth muscle. ARBs also raise the levels of renin, angiotensin I, and angiotensin II as a result of feedback inhibition. Angiotensin II is associated with release of inflammatory mediators such as tumor necrosis factor alpha, cytokines, and chemokines, the consequences of which are also inhibited by ARBs, further preventing renal fibrosis and scarring from chronic inflammation.³

What is the evidence supporting the use of ACE inhibitors and ARBs?

ACE inhibitors and ARBs, used singly, reduce blood pressure and proteinuria, slow progression of kidney disease, and improve outcomes in patients who have heart failure, diabetes mellitus, or a history of myocardial infarction.^5–11 

While dual blockade with the combination of an ACE inhibitor and an ARB lowers blood pressure and proteinuria to a greater degree than monotherapy, dual blockade has been associated with higher rates of complications, including hyperkalemia.^12–17

RISK FACTORS FOR HYPERKALEMIA

ACE inhibitors and ARBs raise potassium, especially when used in combination. Other risk factors for hyperkalemia include the following—and note that some of them are also indications for ACE inhibitors and ARBs:

Renal insufficiency. The kidneys are responsible for over 90% of potassium removal in healthy individuals,^18,19 and the lower the GFR, the higher the risk of hyperkalemia.^3,20,21

Heart failure

Diabetes mellitus^6,21–23

Endogenous potassium load due to hemolysis, rhabdomyolysis, insulin deficiency, lactic acidosis, or gastrointestinal bleeding

Exogenous potassium load due to dietary consumption or blood products

Other medications, eg, sacubitril-valsartan, aldosterone antagonists, mineralocorticoid receptor antagonists, potassium-sparing diuretics, beta-adrenergic antagonists, nonsteroidal anti-inflammatory drugs, heparin, cyclosporine, trimethoprim, digoxin

Hypertension

Hypoaldosteronism (including type 4 renal tubular acidosis)

Addison disease

Advanced age

Lower body mass index.

Both hypokalemia and hyperkalemia are associated with a higher risk of death,^20,21,24  but in patients with heart failure, the survival benefit from ACE inhibitors, ARBs, and mineralocorticoid receptor antagonists outweighs the risk of hyperkalemia.^25–27 Weir and Rolfe²⁸ concluded that patients with heart failure and chronic kidney disease are at greatest risk of hyperkalemia from renin-angiotensin-aldosterone system inhibition, but the increases in potassium levels are small (about 0.1 to 0.3 mmol/L) and unlikely to be clinically significant.

Hyperkalemia tends to recur. Einhorn et al²⁰ found that nearly half of patients with chronic kidney disease who had an episode of hyperkalemia had 1 or more recurrent episodes within a year.

Another concern about using ACE inhibitors and ARBs, especially in patients with chronic kidney disease, is that the serum creatinine level tends to rise when starting these drugs,²⁹ although several studies have shown that an acute rise in creatinine may demonstrate that the drug is actually protecting the kidney.^30,31 Hirsch³² described this phenomenon as “prerenal success,” proposing that the decline in GFR is hemodynamic, secondary to a fall in intraglomerular pressure as a result of efferent vasodilation, and therefore should not be reversed.

Schmidt et al,^33,34 in a study in 122,363 patients who began ACE inhibitor or ARB therapy, found that cardiorenal outcomes were worse, with higher rates of end-stage renal disease, myocardial infarction, heart failure, and death, in those in whom creatinine rose by 30% or more since starting treatment. This trend was also seen, to a lesser degree, in those with a smaller increase in creatinine, suggesting that even this group of patients should receive close monitoring.

Whether renin-angiotensin-aldosterone system inhibitors provide a benefit in advanced progressive chronic kidney disease remains unclear.^35–37  The Angiotensin Converting Enzyme Inhibitor (ACEi)/Angiotensin Receptor Blocker (ARB) Withdrawal in Advanced Renal Disease trial (STOP-ACEi),³⁸ currently under way, will provide valuable data to help close this gap in our knowledge. This open-label randomized controlled trial is testing the hypothesis that stopping ACE inhibitor or ARB treatment, or a combination of both, compared with continuing these treatments, will improve or stabilize renal function in patients with progressive stage 4 or 5 chronic kidney disease.

NEED FOR MONITORING

Taken together, the above data suggest close and regular monitoring is required in patients receiving these drugs. However, monitoring tends to be lax.^34,37,39 A 2017 study of adherence to the guidelines for monitoring serum creatinine and potassium after starting an ACE inhibitor or ARB and subsequent discontinuation found that fewer than 10% of patients had follow-up within the recommended 2 weeks after starting these drugs.³⁴ Most patients with a creatinine rise of 30% or more or a potassium level higher than 6.0 mmol/L continued treatment. There was also no evidence of increased monitoring in those deemed at higher risk of these complications.

WHAT DO THE GUIDELINES SUGGEST?

ACE inhibitors and ARBs in chronic kidney disease and hypertension

Target blood pressures vary in guidelines from different organizations.^4,40–45 The 2017 joint guidelines of the American College of Cardiology and American Heart Association (ACC/AHA)⁴⁰ recommend a target blood pressure of 130/80 mm Hg or less in all patients irrespective of the level of proteinuria and whether they have diabetes mellitus, based on several studies.^46–48 In the elderly, other factors such as the risk of hypotension and falls must be taken into consideration in establishing the most appropriate blood pressure target.

In general, a renin-angiotensin-aldosterone system inhibitor is recommended if the patient has diabetes, stage 1, 2, or 3 chronic kidney disease, or proteinuria. For example, the guidelines recommend a renin-angiotensin-aldosterone system inhibitor in diabetic patients with albuminuria.

None of the guidelines recommend routine use of combination therapy.

ACE inhibitors and ARBs in heart failure

The 2017 ACC/AHA and Heart Failure Society of America (HFSA) guidelines for heart failure⁴⁹ recommend an ACE inhibitor or ARB for patients with stage C (symptomatic) heart failure with reduced ejection fraction, in view of the known cardiovascular morbidity and mortality benefits.

The European Society of Cardiology⁵⁰ recommends ACE inhibitors for patients with symptomatic heart failure with reduced ejection fraction, as well as those with asymptomatic left ventricular systolic dysfunction. In patients with stable coronary artery disease, an ACE inhibitor should be considered even with normal left ventricular function.

ARBs should be used as alternatives in those unable to tolerate ACE inhibitors.

Combination therapy should be avoided due to the increased risk of renal impairment and hyperkalemia but may be considered in patients with heart failure and reduced ejection fraction in whom other treatments are unsuitable. These include patients on beta-blockers who cannot tolerate mineralocorticoid receptor antagonists such as spironolactone. Combination therapy should be done only under strict supervision.⁵⁰

Close monitoring of serum potassium is recommended during ACE inhibitor or ARB use. Those at greatest risk of hyperkalemia include elderly patients, those taking other medications associated with hyperkalemia, and diabetic patients, because of their higher risk of renovascular disease.

Caution is advised when starting ACE inhibitor or ARB therapy in these high-risk groups as well as in patients with potassium levels higher than 5.0 mmol/L at baseline, at high risk of prerenal acute kidney injury, with known renal insufficiency, and with previous deterioration in renal function on these medications.^3,41,51

Before starting therapy, ensure that patients are volume-replete and measure baseline serum electrolytes and creatinine.^41,51

The ACC/AHA and HFSA recommend starting at a low dose and titrating upward slowly. If maximal doses are not tolerated, then a lower dose should be maintained.⁴⁹ The European Society of Cardiology guidelines⁵² suggest increasing the dose at no less than every 2 weeks unless in an inpatient setting. Blood testing should be done 7 to 14 days after starting therapy, after any titration in dosage, and every 4 months thereafter.⁵³

The guidelines generally agree that a rise in creatinine of up to 30% and a fall in eGFR of up to 25% is acceptable, with the need for regular monitoring, particularly in high-risk groups.^{40–42,51,52}

What if serum potassium or creatinine rises during treatment?

If hyperkalemia arises or renal function declines by a significant amount, one should first address contributing factors. If no improvement is seen, then the dose of the ACE inhibitor or ARB should be reduced by 50% and blood work repeated in 1 to 2 weeks. If the laboratory values do not return to an acceptable level, reducing the dose further or stopping the drug is advised.

Give dietary advice to all patients with chronic kidney disease being considered for a renin-angiotensin-aldosterone system inhibitor or for an increase in dose with a potassium level higher than 4.5 mmol/L. A low-potassium diet should aim for potassium intake of less than 50 or 75 mmol/day and sodium intake of less than 60 mmol/day for hypertensive patients with chronic kidney disease.

Review the patient’s medications if the baseline potassium level is higher than 5.0 mmol/L. Consider stopping potassium-sparing agents, digoxin, trimethoprim, and nonsteroidal anti-inflammatory drugs. Also think about starting a non–potassium-sparing diuretic as well as sodium bicarbonate to reduce potassium levels. Blood work should be repeated within 2 weeks after these changes.

Do not start a renin-angiotensin-aldosterone system inhibitor, or do not increase the dose, if the potassium level is elevated until measures have been taken to reduce the degree of hyperkalemia.⁵¹

In renal transplant recipients, renin-angiotensin-aldosterone system inhibitors are often preferred to manage hypertension in those who have proteinuria or cardiovascular disease. However, the risk of hyperkalemia is also greater with concomitant use of immunosuppressive drugs such as tacrolimus and cyclosporine. Management of complications should be approached according to guidelines discussed above.⁵¹

Monitor renal function, potassium. The National Institute for Health and Care Excellence guideline⁵⁴ advocates that baseline renal function testing should be followed by repeat blood testing 1 to 2 weeks after starting renin-angiotensin-aldosterone system inhibitors in patients with ischemic heart disease. The advice is similar when starting therapy in patients with chronic heart failure, emphasizing the need to monitor after each dose increment and to use clinical judgment when deciding to start treatment. The AHA advises caution in patients with renal insufficiency or a potassium level above 5.0 mmol/L.⁴⁹

Sick day rules. The National Institute for Health and Care Excellence encourages discussing “sick day rules” with patients starting renin-angiotensin-aldosterone system inhibitors. This means patients should be advised to temporarily stop taking nephrotoxic medications, including over-the-counter nonsteroidal anti-inflammatory drugs, in any potential state of illness or dehydration, such as diarrhea and vomiting. There is, however, little evidence that this advice can actually reduce the incidence of acute kidney injury.^55,56

OUR RECOMMENDATIONS

Our advice for managing patients receiving ACE inhibitors or ARBs is summarized in Table 1.

A highly active, water- and alcohol-soluble, basic pressor substance is formed when renin and renin-activator interact, for which we suggest the name “angiotonin.”

—Irvine H. Page and O.M. Helmer, 1940.¹

The renin-angiotensin-aldosterone system regulates salt and, in part, water homeostasis, and therefore blood pressure and fluid balance through its actions on the heart, kidneys, and blood vessels.² Drugs that target this system—angiotensin-converting enzyme (ACE) inhibitors and angiotensin II receptor blockers (ARBs)—are used primarily to treat hypertension and also to treat chronic kidney disease and heart failure with reduced ejection fraction.

See related editorial

Controlling blood pressure is important, as hypertension increases the risk of myocardial infarction, cerebrovascular events, and progression of chronic kidney disease, which itself is a risk factor for cardiovascular disease. However, the benefit of these drugs is only partly due to their effect on blood pressure. They also reduce proteinuria, which is a graded risk factor for progression of kidney disease as well as morbidity and death from vascular events.³

Despite the benefits of ACE inhibitors and ARBs, concern about their adverse effects—especially hyperkalemia and a decline in renal function—has led to their underuse in patients likely to derive the greatest benefit.³

ACE INHIBITORS AND ARBs

ACE inhibitors, as their name indicates, inhibit conversion of angiotensin I to angiotensin II by ACE, resulting in vasodilation of the efferent arteriole and a drop in blood pressure. Inhibition of ACE, a kininase, also results in a rise in kinins. One of these, bradykinin, is associated with some of the side effects of this class of drugs such as cough, which affects 5% to 20% of patients.⁴ Elevation of bradykinin is also believed to account for ACE inhibitor-induced angioedema, an uncommon but potentially serious side effect. Kinins are also associated with desirable effects such as lowering blood pressure, increasing insulin sensitivity, and dilating blood vessels.

ARBs were developed as an alternative for patients unable to tolerate the adverse effects of ACE inhibitors. While ACE inhibitors reduce the activity of angiotensin II at both the AT1 and AT2 receptors, ARBs block only the AT1 receptors, thereby inhibiting their vasoconstricting activity on smooth muscle. ARBs also raise the levels of renin, angiotensin I, and angiotensin II as a result of feedback inhibition. Angiotensin II is associated with release of inflammatory mediators such as tumor necrosis factor alpha, cytokines, and chemokines, the consequences of which are also inhibited by ARBs, further preventing renal fibrosis and scarring from chronic inflammation.³

What is the evidence supporting the use of ACE inhibitors and ARBs?

ACE inhibitors and ARBs, used singly, reduce blood pressure and proteinuria, slow progression of kidney disease, and improve outcomes in patients who have heart failure, diabetes mellitus, or a history of myocardial infarction.^5–11 

While dual blockade with the combination of an ACE inhibitor and an ARB lowers blood pressure and proteinuria to a greater degree than monotherapy, dual blockade has been associated with higher rates of complications, including hyperkalemia.^12–17

RISK FACTORS FOR HYPERKALEMIA

ACE inhibitors and ARBs raise potassium, especially when used in combination. Other risk factors for hyperkalemia include the following—and note that some of them are also indications for ACE inhibitors and ARBs:

Renal insufficiency. The kidneys are responsible for over 90% of potassium removal in healthy individuals,^18,19 and the lower the GFR, the higher the risk of hyperkalemia.^3,20,21

Heart failure

Diabetes mellitus^6,21–23

Endogenous potassium load due to hemolysis, rhabdomyolysis, insulin deficiency, lactic acidosis, or gastrointestinal bleeding

Exogenous potassium load due to dietary consumption or blood products

Other medications, eg, sacubitril-valsartan, aldosterone antagonists, mineralocorticoid receptor antagonists, potassium-sparing diuretics, beta-adrenergic antagonists, nonsteroidal anti-inflammatory drugs, heparin, cyclosporine, trimethoprim, digoxin

Hypertension

Hypoaldosteronism (including type 4 renal tubular acidosis)

Addison disease

Advanced age

Lower body mass index.

Both hypokalemia and hyperkalemia are associated with a higher risk of death,^20,21,24  but in patients with heart failure, the survival benefit from ACE inhibitors, ARBs, and mineralocorticoid receptor antagonists outweighs the risk of hyperkalemia.^25–27 Weir and Rolfe²⁸ concluded that patients with heart failure and chronic kidney disease are at greatest risk of hyperkalemia from renin-angiotensin-aldosterone system inhibition, but the increases in potassium levels are small (about 0.1 to 0.3 mmol/L) and unlikely to be clinically significant.

Hyperkalemia tends to recur. Einhorn et al²⁰ found that nearly half of patients with chronic kidney disease who had an episode of hyperkalemia had 1 or more recurrent episodes within a year.

Another concern about using ACE inhibitors and ARBs, especially in patients with chronic kidney disease, is that the serum creatinine level tends to rise when starting these drugs,²⁹ although several studies have shown that an acute rise in creatinine may demonstrate that the drug is actually protecting the kidney.^30,31 Hirsch³² described this phenomenon as “prerenal success,” proposing that the decline in GFR is hemodynamic, secondary to a fall in intraglomerular pressure as a result of efferent vasodilation, and therefore should not be reversed.

Schmidt et al,^33,34 in a study in 122,363 patients who began ACE inhibitor or ARB therapy, found that cardiorenal outcomes were worse, with higher rates of end-stage renal disease, myocardial infarction, heart failure, and death, in those in whom creatinine rose by 30% or more since starting treatment. This trend was also seen, to a lesser degree, in those with a smaller increase in creatinine, suggesting that even this group of patients should receive close monitoring.

Whether renin-angiotensin-aldosterone system inhibitors provide a benefit in advanced progressive chronic kidney disease remains unclear.^35–37  The Angiotensin Converting Enzyme Inhibitor (ACEi)/Angiotensin Receptor Blocker (ARB) Withdrawal in Advanced Renal Disease trial (STOP-ACEi),³⁸ currently under way, will provide valuable data to help close this gap in our knowledge. This open-label randomized controlled trial is testing the hypothesis that stopping ACE inhibitor or ARB treatment, or a combination of both, compared with continuing these treatments, will improve or stabilize renal function in patients with progressive stage 4 or 5 chronic kidney disease.

NEED FOR MONITORING

Taken together, the above data suggest close and regular monitoring is required in patients receiving these drugs. However, monitoring tends to be lax.^34,37,39 A 2017 study of adherence to the guidelines for monitoring serum creatinine and potassium after starting an ACE inhibitor or ARB and subsequent discontinuation found that fewer than 10% of patients had follow-up within the recommended 2 weeks after starting these drugs.³⁴ Most patients with a creatinine rise of 30% or more or a potassium level higher than 6.0 mmol/L continued treatment. There was also no evidence of increased monitoring in those deemed at higher risk of these complications.

WHAT DO THE GUIDELINES SUGGEST?

ACE inhibitors and ARBs in chronic kidney disease and hypertension

Target blood pressures vary in guidelines from different organizations.^4,40–45 The 2017 joint guidelines of the American College of Cardiology and American Heart Association (ACC/AHA)⁴⁰ recommend a target blood pressure of 130/80 mm Hg or less in all patients irrespective of the level of proteinuria and whether they have diabetes mellitus, based on several studies.^46–48 In the elderly, other factors such as the risk of hypotension and falls must be taken into consideration in establishing the most appropriate blood pressure target.

In general, a renin-angiotensin-aldosterone system inhibitor is recommended if the patient has diabetes, stage 1, 2, or 3 chronic kidney disease, or proteinuria. For example, the guidelines recommend a renin-angiotensin-aldosterone system inhibitor in diabetic patients with albuminuria.

None of the guidelines recommend routine use of combination therapy.

ACE inhibitors and ARBs in heart failure

The 2017 ACC/AHA and Heart Failure Society of America (HFSA) guidelines for heart failure⁴⁹ recommend an ACE inhibitor or ARB for patients with stage C (symptomatic) heart failure with reduced ejection fraction, in view of the known cardiovascular morbidity and mortality benefits.

The European Society of Cardiology⁵⁰ recommends ACE inhibitors for patients with symptomatic heart failure with reduced ejection fraction, as well as those with asymptomatic left ventricular systolic dysfunction. In patients with stable coronary artery disease, an ACE inhibitor should be considered even with normal left ventricular function.

ARBs should be used as alternatives in those unable to tolerate ACE inhibitors.

Combination therapy should be avoided due to the increased risk of renal impairment and hyperkalemia but may be considered in patients with heart failure and reduced ejection fraction in whom other treatments are unsuitable. These include patients on beta-blockers who cannot tolerate mineralocorticoid receptor antagonists such as spironolactone. Combination therapy should be done only under strict supervision.⁵⁰

Close monitoring of serum potassium is recommended during ACE inhibitor or ARB use. Those at greatest risk of hyperkalemia include elderly patients, those taking other medications associated with hyperkalemia, and diabetic patients, because of their higher risk of renovascular disease.

Caution is advised when starting ACE inhibitor or ARB therapy in these high-risk groups as well as in patients with potassium levels higher than 5.0 mmol/L at baseline, at high risk of prerenal acute kidney injury, with known renal insufficiency, and with previous deterioration in renal function on these medications.^3,41,51

Before starting therapy, ensure that patients are volume-replete and measure baseline serum electrolytes and creatinine.^41,51

The ACC/AHA and HFSA recommend starting at a low dose and titrating upward slowly. If maximal doses are not tolerated, then a lower dose should be maintained.⁴⁹ The European Society of Cardiology guidelines⁵² suggest increasing the dose at no less than every 2 weeks unless in an inpatient setting. Blood testing should be done 7 to 14 days after starting therapy, after any titration in dosage, and every 4 months thereafter.⁵³

The guidelines generally agree that a rise in creatinine of up to 30% and a fall in eGFR of up to 25% is acceptable, with the need for regular monitoring, particularly in high-risk groups.^{40–42,51,52}

What if serum potassium or creatinine rises during treatment?

If hyperkalemia arises or renal function declines by a significant amount, one should first address contributing factors. If no improvement is seen, then the dose of the ACE inhibitor or ARB should be reduced by 50% and blood work repeated in 1 to 2 weeks. If the laboratory values do not return to an acceptable level, reducing the dose further or stopping the drug is advised.

Give dietary advice to all patients with chronic kidney disease being considered for a renin-angiotensin-aldosterone system inhibitor or for an increase in dose with a potassium level higher than 4.5 mmol/L. A low-potassium diet should aim for potassium intake of less than 50 or 75 mmol/day and sodium intake of less than 60 mmol/day for hypertensive patients with chronic kidney disease.

Review the patient’s medications if the baseline potassium level is higher than 5.0 mmol/L. Consider stopping potassium-sparing agents, digoxin, trimethoprim, and nonsteroidal anti-inflammatory drugs. Also think about starting a non–potassium-sparing diuretic as well as sodium bicarbonate to reduce potassium levels. Blood work should be repeated within 2 weeks after these changes.

Do not start a renin-angiotensin-aldosterone system inhibitor, or do not increase the dose, if the potassium level is elevated until measures have been taken to reduce the degree of hyperkalemia.⁵¹

In renal transplant recipients, renin-angiotensin-aldosterone system inhibitors are often preferred to manage hypertension in those who have proteinuria or cardiovascular disease. However, the risk of hyperkalemia is also greater with concomitant use of immunosuppressive drugs such as tacrolimus and cyclosporine. Management of complications should be approached according to guidelines discussed above.⁵¹

Monitor renal function, potassium. The National Institute for Health and Care Excellence guideline⁵⁴ advocates that baseline renal function testing should be followed by repeat blood testing 1 to 2 weeks after starting renin-angiotensin-aldosterone system inhibitors in patients with ischemic heart disease. The advice is similar when starting therapy in patients with chronic heart failure, emphasizing the need to monitor after each dose increment and to use clinical judgment when deciding to start treatment. The AHA advises caution in patients with renal insufficiency or a potassium level above 5.0 mmol/L.⁴⁹

Sick day rules. The National Institute for Health and Care Excellence encourages discussing “sick day rules” with patients starting renin-angiotensin-aldosterone system inhibitors. This means patients should be advised to temporarily stop taking nephrotoxic medications, including over-the-counter nonsteroidal anti-inflammatory drugs, in any potential state of illness or dehydration, such as diarrhea and vomiting. There is, however, little evidence that this advice can actually reduce the incidence of acute kidney injury.^55,56

OUR RECOMMENDATIONS

Our advice for managing patients receiving ACE inhibitors or ARBs is summarized in Table 1.

A highly active, water- and alcohol-soluble, basic pressor substance is formed when renin and renin-activator interact, for which we suggest the name “angiotonin.”

—Irvine H. Page and O.M. Helmer, 1940.¹

The renin-angiotensin-aldosterone system regulates salt and, in part, water homeostasis, and therefore blood pressure and fluid balance through its actions on the heart, kidneys, and blood vessels.² Drugs that target this system—angiotensin-converting enzyme (ACE) inhibitors and angiotensin II receptor blockers (ARBs)—are used primarily to treat hypertension and also to treat chronic kidney disease and heart failure with reduced ejection fraction.

See related editorial

Controlling blood pressure is important, as hypertension increases the risk of myocardial infarction, cerebrovascular events, and progression of chronic kidney disease, which itself is a risk factor for cardiovascular disease. However, the benefit of these drugs is only partly due to their effect on blood pressure. They also reduce proteinuria, which is a graded risk factor for progression of kidney disease as well as morbidity and death from vascular events.³

Despite the benefits of ACE inhibitors and ARBs, concern about their adverse effects—especially hyperkalemia and a decline in renal function—has led to their underuse in patients likely to derive the greatest benefit.³

ACE INHIBITORS AND ARBs

ACE inhibitors, as their name indicates, inhibit conversion of angiotensin I to angiotensin II by ACE, resulting in vasodilation of the efferent arteriole and a drop in blood pressure. Inhibition of ACE, a kininase, also results in a rise in kinins. One of these, bradykinin, is associated with some of the side effects of this class of drugs such as cough, which affects 5% to 20% of patients.⁴ Elevation of bradykinin is also believed to account for ACE inhibitor-induced angioedema, an uncommon but potentially serious side effect. Kinins are also associated with desirable effects such as lowering blood pressure, increasing insulin sensitivity, and dilating blood vessels.

ARBs were developed as an alternative for patients unable to tolerate the adverse effects of ACE inhibitors. While ACE inhibitors reduce the activity of angiotensin II at both the AT1 and AT2 receptors, ARBs block only the AT1 receptors, thereby inhibiting their vasoconstricting activity on smooth muscle. ARBs also raise the levels of renin, angiotensin I, and angiotensin II as a result of feedback inhibition. Angiotensin II is associated with release of inflammatory mediators such as tumor necrosis factor alpha, cytokines, and chemokines, the consequences of which are also inhibited by ARBs, further preventing renal fibrosis and scarring from chronic inflammation.³

What is the evidence supporting the use of ACE inhibitors and ARBs?

ACE inhibitors and ARBs, used singly, reduce blood pressure and proteinuria, slow progression of kidney disease, and improve outcomes in patients who have heart failure, diabetes mellitus, or a history of myocardial infarction.^5–11 

While dual blockade with the combination of an ACE inhibitor and an ARB lowers blood pressure and proteinuria to a greater degree than monotherapy, dual blockade has been associated with higher rates of complications, including hyperkalemia.^12–17

RISK FACTORS FOR HYPERKALEMIA

ACE inhibitors and ARBs raise potassium, especially when used in combination. Other risk factors for hyperkalemia include the following—and note that some of them are also indications for ACE inhibitors and ARBs:

Renal insufficiency. The kidneys are responsible for over 90% of potassium removal in healthy individuals,^18,19 and the lower the GFR, the higher the risk of hyperkalemia.^3,20,21

Heart failure

Diabetes mellitus^6,21–23

Endogenous potassium load due to hemolysis, rhabdomyolysis, insulin deficiency, lactic acidosis, or gastrointestinal bleeding

Exogenous potassium load due to dietary consumption or blood products

Other medications, eg, sacubitril-valsartan, aldosterone antagonists, mineralocorticoid receptor antagonists, potassium-sparing diuretics, beta-adrenergic antagonists, nonsteroidal anti-inflammatory drugs, heparin, cyclosporine, trimethoprim, digoxin

Hypertension

Hypoaldosteronism (including type 4 renal tubular acidosis)

Addison disease

Advanced age

Lower body mass index.

Both hypokalemia and hyperkalemia are associated with a higher risk of death,^20,21,24  but in patients with heart failure, the survival benefit from ACE inhibitors, ARBs, and mineralocorticoid receptor antagonists outweighs the risk of hyperkalemia.^25–27 Weir and Rolfe²⁸ concluded that patients with heart failure and chronic kidney disease are at greatest risk of hyperkalemia from renin-angiotensin-aldosterone system inhibition, but the increases in potassium levels are small (about 0.1 to 0.3 mmol/L) and unlikely to be clinically significant.

Hyperkalemia tends to recur. Einhorn et al²⁰ found that nearly half of patients with chronic kidney disease who had an episode of hyperkalemia had 1 or more recurrent episodes within a year.

Another concern about using ACE inhibitors and ARBs, especially in patients with chronic kidney disease, is that the serum creatinine level tends to rise when starting these drugs,²⁹ although several studies have shown that an acute rise in creatinine may demonstrate that the drug is actually protecting the kidney.^30,31 Hirsch³² described this phenomenon as “prerenal success,” proposing that the decline in GFR is hemodynamic, secondary to a fall in intraglomerular pressure as a result of efferent vasodilation, and therefore should not be reversed.

Schmidt et al,^33,34 in a study in 122,363 patients who began ACE inhibitor or ARB therapy, found that cardiorenal outcomes were worse, with higher rates of end-stage renal disease, myocardial infarction, heart failure, and death, in those in whom creatinine rose by 30% or more since starting treatment. This trend was also seen, to a lesser degree, in those with a smaller increase in creatinine, suggesting that even this group of patients should receive close monitoring.

Whether renin-angiotensin-aldosterone system inhibitors provide a benefit in advanced progressive chronic kidney disease remains unclear.^35–37  The Angiotensin Converting Enzyme Inhibitor (ACEi)/Angiotensin Receptor Blocker (ARB) Withdrawal in Advanced Renal Disease trial (STOP-ACEi),³⁸ currently under way, will provide valuable data to help close this gap in our knowledge. This open-label randomized controlled trial is testing the hypothesis that stopping ACE inhibitor or ARB treatment, or a combination of both, compared with continuing these treatments, will improve or stabilize renal function in patients with progressive stage 4 or 5 chronic kidney disease.

NEED FOR MONITORING

Taken together, the above data suggest close and regular monitoring is required in patients receiving these drugs. However, monitoring tends to be lax.^34,37,39 A 2017 study of adherence to the guidelines for monitoring serum creatinine and potassium after starting an ACE inhibitor or ARB and subsequent discontinuation found that fewer than 10% of patients had follow-up within the recommended 2 weeks after starting these drugs.³⁴ Most patients with a creatinine rise of 30% or more or a potassium level higher than 6.0 mmol/L continued treatment. There was also no evidence of increased monitoring in those deemed at higher risk of these complications.

WHAT DO THE GUIDELINES SUGGEST?

ACE inhibitors and ARBs in chronic kidney disease and hypertension

Target blood pressures vary in guidelines from different organizations.^4,40–45 The 2017 joint guidelines of the American College of Cardiology and American Heart Association (ACC/AHA)⁴⁰ recommend a target blood pressure of 130/80 mm Hg or less in all patients irrespective of the level of proteinuria and whether they have diabetes mellitus, based on several studies.^46–48 In the elderly, other factors such as the risk of hypotension and falls must be taken into consideration in establishing the most appropriate blood pressure target.

In general, a renin-angiotensin-aldosterone system inhibitor is recommended if the patient has diabetes, stage 1, 2, or 3 chronic kidney disease, or proteinuria. For example, the guidelines recommend a renin-angiotensin-aldosterone system inhibitor in diabetic patients with albuminuria.

None of the guidelines recommend routine use of combination therapy.

ACE inhibitors and ARBs in heart failure

The 2017 ACC/AHA and Heart Failure Society of America (HFSA) guidelines for heart failure⁴⁹ recommend an ACE inhibitor or ARB for patients with stage C (symptomatic) heart failure with reduced ejection fraction, in view of the known cardiovascular morbidity and mortality benefits.

The European Society of Cardiology⁵⁰ recommends ACE inhibitors for patients with symptomatic heart failure with reduced ejection fraction, as well as those with asymptomatic left ventricular systolic dysfunction. In patients with stable coronary artery disease, an ACE inhibitor should be considered even with normal left ventricular function.

ARBs should be used as alternatives in those unable to tolerate ACE inhibitors.

Combination therapy should be avoided due to the increased risk of renal impairment and hyperkalemia but may be considered in patients with heart failure and reduced ejection fraction in whom other treatments are unsuitable. These include patients on beta-blockers who cannot tolerate mineralocorticoid receptor antagonists such as spironolactone. Combination therapy should be done only under strict supervision.⁵⁰

Close monitoring of serum potassium is recommended during ACE inhibitor or ARB use. Those at greatest risk of hyperkalemia include elderly patients, those taking other medications associated with hyperkalemia, and diabetic patients, because of their higher risk of renovascular disease.

Caution is advised when starting ACE inhibitor or ARB therapy in these high-risk groups as well as in patients with potassium levels higher than 5.0 mmol/L at baseline, at high risk of prerenal acute kidney injury, with known renal insufficiency, and with previous deterioration in renal function on these medications.^3,41,51

Before starting therapy, ensure that patients are volume-replete and measure baseline serum electrolytes and creatinine.^41,51

The ACC/AHA and HFSA recommend starting at a low dose and titrating upward slowly. If maximal doses are not tolerated, then a lower dose should be maintained.⁴⁹ The European Society of Cardiology guidelines⁵² suggest increasing the dose at no less than every 2 weeks unless in an inpatient setting. Blood testing should be done 7 to 14 days after starting therapy, after any titration in dosage, and every 4 months thereafter.⁵³

The guidelines generally agree that a rise in creatinine of up to 30% and a fall in eGFR of up to 25% is acceptable, with the need for regular monitoring, particularly in high-risk groups.^{40–42,51,52}

What if serum potassium or creatinine rises during treatment?

If hyperkalemia arises or renal function declines by a significant amount, one should first address contributing factors. If no improvement is seen, then the dose of the ACE inhibitor or ARB should be reduced by 50% and blood work repeated in 1 to 2 weeks. If the laboratory values do not return to an acceptable level, reducing the dose further or stopping the drug is advised.

Give dietary advice to all patients with chronic kidney disease being considered for a renin-angiotensin-aldosterone system inhibitor or for an increase in dose with a potassium level higher than 4.5 mmol/L. A low-potassium diet should aim for potassium intake of less than 50 or 75 mmol/day and sodium intake of less than 60 mmol/day for hypertensive patients with chronic kidney disease.

Review the patient’s medications if the baseline potassium level is higher than 5.0 mmol/L. Consider stopping potassium-sparing agents, digoxin, trimethoprim, and nonsteroidal anti-inflammatory drugs. Also think about starting a non–potassium-sparing diuretic as well as sodium bicarbonate to reduce potassium levels. Blood work should be repeated within 2 weeks after these changes.

Do not start a renin-angiotensin-aldosterone system inhibitor, or do not increase the dose, if the potassium level is elevated until measures have been taken to reduce the degree of hyperkalemia.⁵¹

In renal transplant recipients, renin-angiotensin-aldosterone system inhibitors are often preferred to manage hypertension in those who have proteinuria or cardiovascular disease. However, the risk of hyperkalemia is also greater with concomitant use of immunosuppressive drugs such as tacrolimus and cyclosporine. Management of complications should be approached according to guidelines discussed above.⁵¹

Monitor renal function, potassium. The National Institute for Health and Care Excellence guideline⁵⁴ advocates that baseline renal function testing should be followed by repeat blood testing 1 to 2 weeks after starting renin-angiotensin-aldosterone system inhibitors in patients with ischemic heart disease. The advice is similar when starting therapy in patients with chronic heart failure, emphasizing the need to monitor after each dose increment and to use clinical judgment when deciding to start treatment. The AHA advises caution in patients with renal insufficiency or a potassium level above 5.0 mmol/L.⁴⁹

Sick day rules. The National Institute for Health and Care Excellence encourages discussing “sick day rules” with patients starting renin-angiotensin-aldosterone system inhibitors. This means patients should be advised to temporarily stop taking nephrotoxic medications, including over-the-counter nonsteroidal anti-inflammatory drugs, in any potential state of illness or dehydration, such as diarrhea and vomiting. There is, however, little evidence that this advice can actually reduce the incidence of acute kidney injury.^55,56

OUR RECOMMENDATIONS

Our advice for managing patients receiving ACE inhibitors or ARBs is summarized in Table 1.

Prompt diagnois of infective endocarditis is critical. Potential consequences of missed or delayed diagnosis, including heart failure, stroke, intracardiac abscess, conduction delays, prosthesis dysfunction, and cerebral emboli, are often catastrophic. Echocardiography is the test used most frequently to evaluate for infective endocarditis, but it misses the diagnosis in almost one-third of cases, and even more often if the patient has a prosthetic valve.

But now, several sophisticated imaging tests are available that complement echocardiography in diagnosing and assessing infective endocarditis; these include 4-dimensional computed tomography (4D CT), fluorodeoxyglucose positron emission tomography (FDG-PET), and leukocyte scintigraphy. These tests have greatly improved our ability not only to diagnose infective endocarditis, but also to determine the extent and spread of infection, and they aid in perioperative assessment. Abnormal findings on these tests have been incorporated into the European Society of Cardiology’s 2015 modified diagnostic criteria for infective endocarditis.¹

This article details the indications, advantages, and limitations of the various imaging tests for diagnosing and evaluating infective endocarditis (Table 1).

INFECTIVE ENDOCARDITIS IS DIFFICULT TO DIAGNOSE AND TREAT

Infective endocarditis is difficult to diagnose and treat. Clinical and imaging clues can be subtle, and the diagnosis requires a high level of suspicion and visualization of cardiac structures.

Further, the incidence of infective endocarditis is on the rise in the United States, particularly in women and young adults, likely due to intravenous drug use.^2,3

ECHOCARDIOGRAPHY HAS AN IMPORTANT ROLE, BUT IS LIMITED

Echocardiography remains the most commonly performed study for diagnosing infective endocarditis, as it is fast, widely accessible, and less expensive than other imaging tests.

Transthoracic echocardiography (TTE) is often the first choice for testing. However, its sensitivity is only about 70% for detecting vegetations on native valves and 50% for detecting vegetations on prosthetic valves.¹ It is inherently constrained by the limited number of views by which a comprehensive external evaluation of the heart can be achieved. Using a 2-dimensional instrument to view a 3-dimensional object is difficult, and depending on several factors, it can be hard to see vegetations and abscesses that are associated with infective endocarditis. Further, TTE is impeded by obesity and by hyperinflated lungs from obstructive pulmonary disease or mechanical ventilation. It has poor sensitivity for detecting small vegetations and for detecting vegetations and paravalvular complications in patients who have a prosthetic valve or a cardiac implanted electronic device.

Transesophageal echocardiography (TEE) is the recommended first-line imaging test for patients with prosthetic valves and no contraindications to the test. Otherwise, it should be done after TTE if the results of TTE are negative but clinical suspicion for infective endocarditis remains high (eg, because the patient uses intravenous drugs). But although TEE has a higher sensitivity than TTE (up to 96% for vegetations on native valves and 92% for those on prosthetic valves, if performed by an experienced sonographer), it can still miss infective endocarditis. Also, TEE does not provide a significant advantage over TTE in patients who have a cardiac implanted electronic device.^1,4,5

Regardless of whether TTE or TEE is used, they are estimated to miss up to 30% of cases of infective endocarditis and its sequelae.⁴ False-negative findings are likelier in patients who have preexisting severe valvular lesions, prosthetic valves, cardiac implanted electronic devices, small vegetations, or abscesses, or if a vegetation has already broken free and embolized. Furthermore, distinguishing between vegetations and thrombi, cardiac tumors, and myxomatous changes using echocardiography is difficult.

For patients who have inconclusive results on echocardiography, contraindications to TEE, or poor sonic windows, cardiac CT can be an excellent alternative. It is especially useful in the setting of a prosthetic valve.

Synchronized (“gated”) with the patient’s heart rate and rhythm, CT machines can acquire images during diastole, reducing motion artifact, and can create 3D images of the heart. In addition, newer machines can acquire several images at different points in the heart cycle to add a fourth dimension—time. The resulting 4D images play like short video loops of the beating heart and allow noninvasive assessment of cardiac anatomy with remarkable detail and resolution.

4D CT is increasingly being used in infective endocarditis, and growing evidence indicates that its accuracy is similar to that of TEE in the preoperative evaluation of patients with aortic prosthetic valve endocarditis.⁶ In a study of 28 patients, complementary use of CT angiography led to a change in treatment strategy in 7 (25%) compared with routine clinical workup.⁷ Several studies have found no difference between 4D CT and preoperative TEE in detecting pseudoaneurysm, abscess, or valve dehiscence. TEE and 4D CT also have similar sensitivities for detecting infective endocarditis in native and prosthetic valves.^8,9

Coupled with CT angiography, 4D CT is also an excellent noninvasive way to perioperatively evaluate the coronary arteries without the risks associated with catheterization in those requiring nonemergency surgery (Figure 1A, B, and C).

4D CT performs well for detecting abscess and pseudoaneurysm but has slightly lower sensitivity for vegetations than TEE (91% vs 99%).⁹

Gated CT, PET, or both may be useful in cases of suspected prosthetic aortic valve endocarditis when TEE is negative. Pseudoaneurysms are not well visualized with TEE, and the atrial mitral curtain area is often thickened on TEE in cases of aortic prosthetic valve infective endocarditis that do not definitely involve abscesses. Gated CT and PET show this area better.⁸ This information is important in cases in which a surgeon may be unconvinced that the patient has prosthetic valve endocarditis.

Limitations of 4D cardiac CT

4D CT with or without angiography has limitations. It requires a wide-volume scanner and an experienced reader.

Patients with irregular heart rhythms or uncontrolled tachycardia pose technical problems for image acquisition. Cardiac CT is typically gated (ie, images are obtained within a defined time period) to acquire images during diastole. Ideally, images are acquired when the heart is in mid to late diastole, a time of minimal cardiac motion, so that motion artifact is minimized. To estimate the timing of image acquisition, the cardiac cycle must be predictable, and its duration should be as long as possible. Tachycardia or irregular rhythms such as frequent ectopic beats or atrial fibrillation make acquisition timing difficult, and thus make it nearly impossible to accurately obtain images when the heart is at minimum motion, limiting assessment of cardiac structures or the coronary tree.^4,10

Extensive coronary calcification can hinder assessment of the coronary tree by CT coronary angiography.

Contrast exposure may limit the use of CT in some patients (eg, those with contrast allergies or renal dysfunction). However, modern scanners allow for much smaller contrast boluses without decreasing sensitivity.

4D CT involves radiation exposure, especially when done with angiography, although modern scanners have greatly reduced exposure. The average radiation dose in CT coronary angiography is 2.9 to 5.9 mSv¹¹ compared with 7 mSv in diagnostic cardiac catheterization (without angioplasty or stenting) or 16 mSv in routine CT of the abdomen and pelvis with contrast.^12,13 In view of the morbidity and mortality risks associated with infective endocarditis, especially if the diagnosis is delayed, this small radiation exposure may be justifiable.

Bottom line for cardiac CT

4D CT is an excellent alternative to echocardiography for select patients. Clinicians should strongly consider this study in the following situations:

• Patients with a prosthetic valve
• Patients who are strongly suspected of having infective endocarditis but who have a poor sonic window on TTE or TEE, as can occur with chronic obstructive lung disease, morbid obesity, or previous thoracic or cardiovascular surgery
• Patients who meet clinical indications for TEE, such as having a prosthetic valve or a high suspicion for native valve infective endocarditis with negative TTE, but who have contraindications to TEE
• As an alternative to TEE for preoperative evaluation in patients with known infective endocarditis.

Patients with tachycardia or irregular heart rhythms are not good candidates for this test.

FDG-PET AND LEUKOCYTE SCINTIGRAPHY

FDG-PET and leukocyte scintigraphy are other options for diagnosing infective endocarditis and determining the presence and extent of intra- and extracardiac infection. They are more sensitive than echocardiography for detecting infection of cardiac implanted electronic devices such as ventricular assist devices, pacemakers, implanted cardiac defibrillators, and cardiac resynchronization therapy devices.^14–16

The utility of FDG-PET is founded on the uptake of ¹⁸F-fluorodeoxyglucose by cells, with higher uptake taking place in cells with higher metabolic activity (such as in areas of inflammation). Similarly, leukocyte scintigraphy relies on the use of radiolabeled leukocytes (ie, leukocytes previously extracted from the patient, labelled, and re-introduced into the patient) to allow for localization of inflamed tissue.

The most significant contribution of FDG-PET may be the ability to detect infective endocarditis early, when echocardiography is initially negative. When abnormal FDG uptake was included in the modified Duke criteria, it increased the sensitivity to 97% for detecting infective endocarditis on admission, leading some to propose its incorporation as a major criterion.¹⁷ In patients with prosthetic valves and suspected infective endocarditis, FDG-PET was found in one study to have a sensitivity of up to 91% and a specificity of up to 95%.¹⁸

Both FDG-PET and leukocyte scintigraphy have a high sensitivity, specificity, and negative predictive value for cardiac implanted electronic device infection, and should be strongly considered in patients in whom it is suspected but who have negative or inconclusive findings on echocardiography.^14,15

In addition, a common conundrum faced by clinicians with use of echocardiography is the difficulty of differentiating thrombus from infected vegetation on valves or device lead wires. Some evidence indicates that FDG-PET may help to discriminate between vegetation and thrombus, although more rigorous studies are needed before its use for that purpose can be recommended.¹⁹

Both FDG-PET and leukocyte scintigraphy perform poorly for detecting native-valve infective endocarditis. In a study in which 90% of the patients had native-valve infective endocarditis according to the Duke criteria, FDG-PET had a specificity of 93% but a sensitivity of only 39%.²⁰

Both studies can be cumbersome, laborious, and time-consuming for patients. FDG-PET requires a fasting or glucose-restricted diet before testing, and the test itself can be complicated by development of hyperglycemia, although this is rare.

While FDG-PET is most effective in detecting infections of prosthetic valves and cardiac implanted electronic devices, the results can be falsely positive in patients with a history of recent cardiac surgery (due to ongoing tissue healing), as well as maladies other than infective endocarditis that lead to inflammation, such as vasculitis or malignancy. Similarly, for unclear reasons, leukocyte scintigraphy can yield false-negative results in patients with enterococcal or candidal infective endocarditis.²¹

FDG-PET and leukocyte scintigraphy are more expensive than TEE and cardiac CT²² and are not widely available.

Both tests entail radiation exposure, with the average dose ranging from 7 to 14 mSv. However, this is less than the average amount acquired during percutaneous coronary intervention (16 mSv), and overlaps with the amount in chest CT with contrast when assessing for pulmonary embolism (7 to 9 mSv). Lower doses are possible with optimized protocols.^12,13,15,23

Bottom line for nuclear studies

FDG-PET and leukocyte scintigraphy are especially useful for patients with a prosthetic valve or cardiac implanted electronic device. However, limitations must be kept in mind.

A suggested algorithm for testing with nuclear imaging is shown in Figure 2.^1,4

CEREBRAL MAGNETIC RESONANCE IMAGING

Cerebral magnetic resonance imaging (MRI) is more sensitive than cerebral CT for detecting emboli in the brain. According to American Heart Association guidelines, cerebral MRI should be done in patients with known or suspected infective endocarditis and neurologic impairment, defined as headaches, meningeal symptoms, or neurologic deficits. It is also often used in neurologically asymptomatic patients with infective endocarditis who have indications for valve surgery to assess for mycotic aneurysms, which are associated with increased intracranial bleeding during surgery.

MRI use in other asymptomatic patients remains controversial.²⁴ In cases with high clinical suspicion for infective endocarditis and no findings on echocardiography, cerebral MRI can increase the sensitivity of the Duke criteria by adding a minor criterion. Some have argued that, in patients with definite infective endocarditis, detecting silent cerebral complications can lead to management changes. However, more studies are needed to determine if there is indeed a group of neurologically asymptomatic infective endocarditis patients for whom cerebral MRI leads to improved outcomes.

Limitations of cerebral MRI

Cerebral MRI cannot be used in patients with non-MRI-compatible implanted hardware.

Gadolinium, the contrast agent typically used, can cause nephrogenic systemic fibrosis in patients who have poor renal function. This rare but serious adverse effect is characterized by irreversible systemic fibrosis affecting skin, muscles, and even visceral tissue such as lungs. The American College of Radiology allows for gadolinium use in patients without acute kidney injury and patients with stable chronic kidney disease with a glomerular filtration rate of at least 30 mL/min/1.73 m². Its use should be avoided in patients with renal failure on replacement therapy, with advanced chronic kidney disease (glomerular filtration rate < 30 mL/min/1.73 m²), or with acute kidney injury, even if they do not need renal replacement therapy.²⁵

Concerns have also been raised about gadolinium retention in the brain, even in patients with normal renal function.^26–28 Thus far, no conclusive clinical adverse effects of retention have been found, although more study is warranted. Nevertheless, the US Food and Drug Administration now requires a black-box warning about this possibility and advises clinicians to counsel patients appropriately.

Bottom line on cerebral MRI

Cerebral MRI should be obtained when a patient presents with definite or possible infective endocarditis with neurologic impairment, such as new headaches, meningismus, or focal neurologic deficits. Routine brain MRI in patients with confirmed infective endocarditis without neurologic symptoms, or those without definite infective endocarditis, is discouraged.

CARDIAC MRI

Cardiac MRI, typically obtained with gadolinium contrast, allows for better 3D assessment of cardiac structures and morphology than echocardiography or CT, and can detect infiltrative cardiac disease, myopericarditis, and much more. It is increasingly used in the field of structural cardiology, but its role for evaluating infective endocarditis remains unclear.

Cardiac MRI does not appear to be better than echocardiography for diagnosing infective endocarditis. However, it may prove helpful in the evaluation of patients known to have infective endocarditis but who cannot be properly evaluated for disease extent because of poor image quality on echocardiography and contraindications to CT.^1,29 Its role is limited in patients with cardiac implanted electronic devices, as most devices are incompatible with MRI use, although newer devices obviate this concern. But even for devices that are MRI-compatible, results are diminished due to an eclipsing effect, wherein the device parts can make it hard to see structures clearly because the “brightness” basically eclipses the surrounding area.⁴

Concerns regarding use of gadolinium as described above need also be considered.

The role of cardiac MRI in diagnosing and managing infective endocarditis may evolve, but at present, the 2017 American College of Cardiology and American Heart Association appropriate-use criteria discourage its use for these purposes.¹⁶

Bottom line for cardiac MRI

Cardiac MRI to evaluate a patient for suspected infective endocarditis is not recommended due to lack of superiority compared with echocardiography or CT, and the risk of nephrogenic systemic fibrosis from gadolinium in patients with renal compromise.

Prompt diagnois of infective endocarditis is critical. Potential consequences of missed or delayed diagnosis, including heart failure, stroke, intracardiac abscess, conduction delays, prosthesis dysfunction, and cerebral emboli, are often catastrophic. Echocardiography is the test used most frequently to evaluate for infective endocarditis, but it misses the diagnosis in almost one-third of cases, and even more often if the patient has a prosthetic valve.

But now, several sophisticated imaging tests are available that complement echocardiography in diagnosing and assessing infective endocarditis; these include 4-dimensional computed tomography (4D CT), fluorodeoxyglucose positron emission tomography (FDG-PET), and leukocyte scintigraphy. These tests have greatly improved our ability not only to diagnose infective endocarditis, but also to determine the extent and spread of infection, and they aid in perioperative assessment. Abnormal findings on these tests have been incorporated into the European Society of Cardiology’s 2015 modified diagnostic criteria for infective endocarditis.¹

This article details the indications, advantages, and limitations of the various imaging tests for diagnosing and evaluating infective endocarditis (Table 1).

INFECTIVE ENDOCARDITIS IS DIFFICULT TO DIAGNOSE AND TREAT

Infective endocarditis is difficult to diagnose and treat. Clinical and imaging clues can be subtle, and the diagnosis requires a high level of suspicion and visualization of cardiac structures.

Further, the incidence of infective endocarditis is on the rise in the United States, particularly in women and young adults, likely due to intravenous drug use.^2,3

ECHOCARDIOGRAPHY HAS AN IMPORTANT ROLE, BUT IS LIMITED

Echocardiography remains the most commonly performed study for diagnosing infective endocarditis, as it is fast, widely accessible, and less expensive than other imaging tests.

Transthoracic echocardiography (TTE) is often the first choice for testing. However, its sensitivity is only about 70% for detecting vegetations on native valves and 50% for detecting vegetations on prosthetic valves.¹ It is inherently constrained by the limited number of views by which a comprehensive external evaluation of the heart can be achieved. Using a 2-dimensional instrument to view a 3-dimensional object is difficult, and depending on several factors, it can be hard to see vegetations and abscesses that are associated with infective endocarditis. Further, TTE is impeded by obesity and by hyperinflated lungs from obstructive pulmonary disease or mechanical ventilation. It has poor sensitivity for detecting small vegetations and for detecting vegetations and paravalvular complications in patients who have a prosthetic valve or a cardiac implanted electronic device.

Transesophageal echocardiography (TEE) is the recommended first-line imaging test for patients with prosthetic valves and no contraindications to the test. Otherwise, it should be done after TTE if the results of TTE are negative but clinical suspicion for infective endocarditis remains high (eg, because the patient uses intravenous drugs). But although TEE has a higher sensitivity than TTE (up to 96% for vegetations on native valves and 92% for those on prosthetic valves, if performed by an experienced sonographer), it can still miss infective endocarditis. Also, TEE does not provide a significant advantage over TTE in patients who have a cardiac implanted electronic device.^1,4,5

Regardless of whether TTE or TEE is used, they are estimated to miss up to 30% of cases of infective endocarditis and its sequelae.⁴ False-negative findings are likelier in patients who have preexisting severe valvular lesions, prosthetic valves, cardiac implanted electronic devices, small vegetations, or abscesses, or if a vegetation has already broken free and embolized. Furthermore, distinguishing between vegetations and thrombi, cardiac tumors, and myxomatous changes using echocardiography is difficult.

For patients who have inconclusive results on echocardiography, contraindications to TEE, or poor sonic windows, cardiac CT can be an excellent alternative. It is especially useful in the setting of a prosthetic valve.

Synchronized (“gated”) with the patient’s heart rate and rhythm, CT machines can acquire images during diastole, reducing motion artifact, and can create 3D images of the heart. In addition, newer machines can acquire several images at different points in the heart cycle to add a fourth dimension—time. The resulting 4D images play like short video loops of the beating heart and allow noninvasive assessment of cardiac anatomy with remarkable detail and resolution.

4D CT is increasingly being used in infective endocarditis, and growing evidence indicates that its accuracy is similar to that of TEE in the preoperative evaluation of patients with aortic prosthetic valve endocarditis.⁶ In a study of 28 patients, complementary use of CT angiography led to a change in treatment strategy in 7 (25%) compared with routine clinical workup.⁷ Several studies have found no difference between 4D CT and preoperative TEE in detecting pseudoaneurysm, abscess, or valve dehiscence. TEE and 4D CT also have similar sensitivities for detecting infective endocarditis in native and prosthetic valves.^8,9

Coupled with CT angiography, 4D CT is also an excellent noninvasive way to perioperatively evaluate the coronary arteries without the risks associated with catheterization in those requiring nonemergency surgery (Figure 1A, B, and C).

4D CT performs well for detecting abscess and pseudoaneurysm but has slightly lower sensitivity for vegetations than TEE (91% vs 99%).⁹

Gated CT, PET, or both may be useful in cases of suspected prosthetic aortic valve endocarditis when TEE is negative. Pseudoaneurysms are not well visualized with TEE, and the atrial mitral curtain area is often thickened on TEE in cases of aortic prosthetic valve infective endocarditis that do not definitely involve abscesses. Gated CT and PET show this area better.⁸ This information is important in cases in which a surgeon may be unconvinced that the patient has prosthetic valve endocarditis.

Limitations of 4D cardiac CT

4D CT with or without angiography has limitations. It requires a wide-volume scanner and an experienced reader.

Patients with irregular heart rhythms or uncontrolled tachycardia pose technical problems for image acquisition. Cardiac CT is typically gated (ie, images are obtained within a defined time period) to acquire images during diastole. Ideally, images are acquired when the heart is in mid to late diastole, a time of minimal cardiac motion, so that motion artifact is minimized. To estimate the timing of image acquisition, the cardiac cycle must be predictable, and its duration should be as long as possible. Tachycardia or irregular rhythms such as frequent ectopic beats or atrial fibrillation make acquisition timing difficult, and thus make it nearly impossible to accurately obtain images when the heart is at minimum motion, limiting assessment of cardiac structures or the coronary tree.^4,10

Extensive coronary calcification can hinder assessment of the coronary tree by CT coronary angiography.

Contrast exposure may limit the use of CT in some patients (eg, those with contrast allergies or renal dysfunction). However, modern scanners allow for much smaller contrast boluses without decreasing sensitivity.

4D CT involves radiation exposure, especially when done with angiography, although modern scanners have greatly reduced exposure. The average radiation dose in CT coronary angiography is 2.9 to 5.9 mSv¹¹ compared with 7 mSv in diagnostic cardiac catheterization (without angioplasty or stenting) or 16 mSv in routine CT of the abdomen and pelvis with contrast.^12,13 In view of the morbidity and mortality risks associated with infective endocarditis, especially if the diagnosis is delayed, this small radiation exposure may be justifiable.

Bottom line for cardiac CT

4D CT is an excellent alternative to echocardiography for select patients. Clinicians should strongly consider this study in the following situations:

• Patients with a prosthetic valve
• Patients who are strongly suspected of having infective endocarditis but who have a poor sonic window on TTE or TEE, as can occur with chronic obstructive lung disease, morbid obesity, or previous thoracic or cardiovascular surgery
• Patients who meet clinical indications for TEE, such as having a prosthetic valve or a high suspicion for native valve infective endocarditis with negative TTE, but who have contraindications to TEE
• As an alternative to TEE for preoperative evaluation in patients with known infective endocarditis.

Patients with tachycardia or irregular heart rhythms are not good candidates for this test.

FDG-PET AND LEUKOCYTE SCINTIGRAPHY

FDG-PET and leukocyte scintigraphy are other options for diagnosing infective endocarditis and determining the presence and extent of intra- and extracardiac infection. They are more sensitive than echocardiography for detecting infection of cardiac implanted electronic devices such as ventricular assist devices, pacemakers, implanted cardiac defibrillators, and cardiac resynchronization therapy devices.^14–16

The utility of FDG-PET is founded on the uptake of ¹⁸F-fluorodeoxyglucose by cells, with higher uptake taking place in cells with higher metabolic activity (such as in areas of inflammation). Similarly, leukocyte scintigraphy relies on the use of radiolabeled leukocytes (ie, leukocytes previously extracted from the patient, labelled, and re-introduced into the patient) to allow for localization of inflamed tissue.

The most significant contribution of FDG-PET may be the ability to detect infective endocarditis early, when echocardiography is initially negative. When abnormal FDG uptake was included in the modified Duke criteria, it increased the sensitivity to 97% for detecting infective endocarditis on admission, leading some to propose its incorporation as a major criterion.¹⁷ In patients with prosthetic valves and suspected infective endocarditis, FDG-PET was found in one study to have a sensitivity of up to 91% and a specificity of up to 95%.¹⁸

Both FDG-PET and leukocyte scintigraphy have a high sensitivity, specificity, and negative predictive value for cardiac implanted electronic device infection, and should be strongly considered in patients in whom it is suspected but who have negative or inconclusive findings on echocardiography.^14,15

In addition, a common conundrum faced by clinicians with use of echocardiography is the difficulty of differentiating thrombus from infected vegetation on valves or device lead wires. Some evidence indicates that FDG-PET may help to discriminate between vegetation and thrombus, although more rigorous studies are needed before its use for that purpose can be recommended.¹⁹

Both FDG-PET and leukocyte scintigraphy perform poorly for detecting native-valve infective endocarditis. In a study in which 90% of the patients had native-valve infective endocarditis according to the Duke criteria, FDG-PET had a specificity of 93% but a sensitivity of only 39%.²⁰

Both studies can be cumbersome, laborious, and time-consuming for patients. FDG-PET requires a fasting or glucose-restricted diet before testing, and the test itself can be complicated by development of hyperglycemia, although this is rare.

While FDG-PET is most effective in detecting infections of prosthetic valves and cardiac implanted electronic devices, the results can be falsely positive in patients with a history of recent cardiac surgery (due to ongoing tissue healing), as well as maladies other than infective endocarditis that lead to inflammation, such as vasculitis or malignancy. Similarly, for unclear reasons, leukocyte scintigraphy can yield false-negative results in patients with enterococcal or candidal infective endocarditis.²¹

FDG-PET and leukocyte scintigraphy are more expensive than TEE and cardiac CT²² and are not widely available.

Both tests entail radiation exposure, with the average dose ranging from 7 to 14 mSv. However, this is less than the average amount acquired during percutaneous coronary intervention (16 mSv), and overlaps with the amount in chest CT with contrast when assessing for pulmonary embolism (7 to 9 mSv). Lower doses are possible with optimized protocols.^12,13,15,23

Bottom line for nuclear studies

FDG-PET and leukocyte scintigraphy are especially useful for patients with a prosthetic valve or cardiac implanted electronic device. However, limitations must be kept in mind.

A suggested algorithm for testing with nuclear imaging is shown in Figure 2.^1,4

CEREBRAL MAGNETIC RESONANCE IMAGING

Cerebral magnetic resonance imaging (MRI) is more sensitive than cerebral CT for detecting emboli in the brain. According to American Heart Association guidelines, cerebral MRI should be done in patients with known or suspected infective endocarditis and neurologic impairment, defined as headaches, meningeal symptoms, or neurologic deficits. It is also often used in neurologically asymptomatic patients with infective endocarditis who have indications for valve surgery to assess for mycotic aneurysms, which are associated with increased intracranial bleeding during surgery.

MRI use in other asymptomatic patients remains controversial.²⁴ In cases with high clinical suspicion for infective endocarditis and no findings on echocardiography, cerebral MRI can increase the sensitivity of the Duke criteria by adding a minor criterion. Some have argued that, in patients with definite infective endocarditis, detecting silent cerebral complications can lead to management changes. However, more studies are needed to determine if there is indeed a group of neurologically asymptomatic infective endocarditis patients for whom cerebral MRI leads to improved outcomes.

Limitations of cerebral MRI

Cerebral MRI cannot be used in patients with non-MRI-compatible implanted hardware.

Gadolinium, the contrast agent typically used, can cause nephrogenic systemic fibrosis in patients who have poor renal function. This rare but serious adverse effect is characterized by irreversible systemic fibrosis affecting skin, muscles, and even visceral tissue such as lungs. The American College of Radiology allows for gadolinium use in patients without acute kidney injury and patients with stable chronic kidney disease with a glomerular filtration rate of at least 30 mL/min/1.73 m². Its use should be avoided in patients with renal failure on replacement therapy, with advanced chronic kidney disease (glomerular filtration rate < 30 mL/min/1.73 m²), or with acute kidney injury, even if they do not need renal replacement therapy.²⁵

Concerns have also been raised about gadolinium retention in the brain, even in patients with normal renal function.^26–28 Thus far, no conclusive clinical adverse effects of retention have been found, although more study is warranted. Nevertheless, the US Food and Drug Administration now requires a black-box warning about this possibility and advises clinicians to counsel patients appropriately.

Bottom line on cerebral MRI

Cerebral MRI should be obtained when a patient presents with definite or possible infective endocarditis with neurologic impairment, such as new headaches, meningismus, or focal neurologic deficits. Routine brain MRI in patients with confirmed infective endocarditis without neurologic symptoms, or those without definite infective endocarditis, is discouraged.

CARDIAC MRI

Cardiac MRI, typically obtained with gadolinium contrast, allows for better 3D assessment of cardiac structures and morphology than echocardiography or CT, and can detect infiltrative cardiac disease, myopericarditis, and much more. It is increasingly used in the field of structural cardiology, but its role for evaluating infective endocarditis remains unclear.

Cardiac MRI does not appear to be better than echocardiography for diagnosing infective endocarditis. However, it may prove helpful in the evaluation of patients known to have infective endocarditis but who cannot be properly evaluated for disease extent because of poor image quality on echocardiography and contraindications to CT.^1,29 Its role is limited in patients with cardiac implanted electronic devices, as most devices are incompatible with MRI use, although newer devices obviate this concern. But even for devices that are MRI-compatible, results are diminished due to an eclipsing effect, wherein the device parts can make it hard to see structures clearly because the “brightness” basically eclipses the surrounding area.⁴

Concerns regarding use of gadolinium as described above need also be considered.

The role of cardiac MRI in diagnosing and managing infective endocarditis may evolve, but at present, the 2017 American College of Cardiology and American Heart Association appropriate-use criteria discourage its use for these purposes.¹⁶

Bottom line for cardiac MRI

Cardiac MRI to evaluate a patient for suspected infective endocarditis is not recommended due to lack of superiority compared with echocardiography or CT, and the risk of nephrogenic systemic fibrosis from gadolinium in patients with renal compromise.

Prompt diagnois of infective endocarditis is critical. Potential consequences of missed or delayed diagnosis, including heart failure, stroke, intracardiac abscess, conduction delays, prosthesis dysfunction, and cerebral emboli, are often catastrophic. Echocardiography is the test used most frequently to evaluate for infective endocarditis, but it misses the diagnosis in almost one-third of cases, and even more often if the patient has a prosthetic valve.

But now, several sophisticated imaging tests are available that complement echocardiography in diagnosing and assessing infective endocarditis; these include 4-dimensional computed tomography (4D CT), fluorodeoxyglucose positron emission tomography (FDG-PET), and leukocyte scintigraphy. These tests have greatly improved our ability not only to diagnose infective endocarditis, but also to determine the extent and spread of infection, and they aid in perioperative assessment. Abnormal findings on these tests have been incorporated into the European Society of Cardiology’s 2015 modified diagnostic criteria for infective endocarditis.¹

This article details the indications, advantages, and limitations of the various imaging tests for diagnosing and evaluating infective endocarditis (Table 1).

INFECTIVE ENDOCARDITIS IS DIFFICULT TO DIAGNOSE AND TREAT

Infective endocarditis is difficult to diagnose and treat. Clinical and imaging clues can be subtle, and the diagnosis requires a high level of suspicion and visualization of cardiac structures.

Further, the incidence of infective endocarditis is on the rise in the United States, particularly in women and young adults, likely due to intravenous drug use.^2,3

ECHOCARDIOGRAPHY HAS AN IMPORTANT ROLE, BUT IS LIMITED

Echocardiography remains the most commonly performed study for diagnosing infective endocarditis, as it is fast, widely accessible, and less expensive than other imaging tests.

Transthoracic echocardiography (TTE) is often the first choice for testing. However, its sensitivity is only about 70% for detecting vegetations on native valves and 50% for detecting vegetations on prosthetic valves.¹ It is inherently constrained by the limited number of views by which a comprehensive external evaluation of the heart can be achieved. Using a 2-dimensional instrument to view a 3-dimensional object is difficult, and depending on several factors, it can be hard to see vegetations and abscesses that are associated with infective endocarditis. Further, TTE is impeded by obesity and by hyperinflated lungs from obstructive pulmonary disease or mechanical ventilation. It has poor sensitivity for detecting small vegetations and for detecting vegetations and paravalvular complications in patients who have a prosthetic valve or a cardiac implanted electronic device.

Transesophageal echocardiography (TEE) is the recommended first-line imaging test for patients with prosthetic valves and no contraindications to the test. Otherwise, it should be done after TTE if the results of TTE are negative but clinical suspicion for infective endocarditis remains high (eg, because the patient uses intravenous drugs). But although TEE has a higher sensitivity than TTE (up to 96% for vegetations on native valves and 92% for those on prosthetic valves, if performed by an experienced sonographer), it can still miss infective endocarditis. Also, TEE does not provide a significant advantage over TTE in patients who have a cardiac implanted electronic device.^1,4,5

Regardless of whether TTE or TEE is used, they are estimated to miss up to 30% of cases of infective endocarditis and its sequelae.⁴ False-negative findings are likelier in patients who have preexisting severe valvular lesions, prosthetic valves, cardiac implanted electronic devices, small vegetations, or abscesses, or if a vegetation has already broken free and embolized. Furthermore, distinguishing between vegetations and thrombi, cardiac tumors, and myxomatous changes using echocardiography is difficult.

For patients who have inconclusive results on echocardiography, contraindications to TEE, or poor sonic windows, cardiac CT can be an excellent alternative. It is especially useful in the setting of a prosthetic valve.

Synchronized (“gated”) with the patient’s heart rate and rhythm, CT machines can acquire images during diastole, reducing motion artifact, and can create 3D images of the heart. In addition, newer machines can acquire several images at different points in the heart cycle to add a fourth dimension—time. The resulting 4D images play like short video loops of the beating heart and allow noninvasive assessment of cardiac anatomy with remarkable detail and resolution.

4D CT is increasingly being used in infective endocarditis, and growing evidence indicates that its accuracy is similar to that of TEE in the preoperative evaluation of patients with aortic prosthetic valve endocarditis.⁶ In a study of 28 patients, complementary use of CT angiography led to a change in treatment strategy in 7 (25%) compared with routine clinical workup.⁷ Several studies have found no difference between 4D CT and preoperative TEE in detecting pseudoaneurysm, abscess, or valve dehiscence. TEE and 4D CT also have similar sensitivities for detecting infective endocarditis in native and prosthetic valves.^8,9

Coupled with CT angiography, 4D CT is also an excellent noninvasive way to perioperatively evaluate the coronary arteries without the risks associated with catheterization in those requiring nonemergency surgery (Figure 1A, B, and C).

4D CT performs well for detecting abscess and pseudoaneurysm but has slightly lower sensitivity for vegetations than TEE (91% vs 99%).⁹

Gated CT, PET, or both may be useful in cases of suspected prosthetic aortic valve endocarditis when TEE is negative. Pseudoaneurysms are not well visualized with TEE, and the atrial mitral curtain area is often thickened on TEE in cases of aortic prosthetic valve infective endocarditis that do not definitely involve abscesses. Gated CT and PET show this area better.⁸ This information is important in cases in which a surgeon may be unconvinced that the patient has prosthetic valve endocarditis.

Limitations of 4D cardiac CT

4D CT with or without angiography has limitations. It requires a wide-volume scanner and an experienced reader.

Patients with irregular heart rhythms or uncontrolled tachycardia pose technical problems for image acquisition. Cardiac CT is typically gated (ie, images are obtained within a defined time period) to acquire images during diastole. Ideally, images are acquired when the heart is in mid to late diastole, a time of minimal cardiac motion, so that motion artifact is minimized. To estimate the timing of image acquisition, the cardiac cycle must be predictable, and its duration should be as long as possible. Tachycardia or irregular rhythms such as frequent ectopic beats or atrial fibrillation make acquisition timing difficult, and thus make it nearly impossible to accurately obtain images when the heart is at minimum motion, limiting assessment of cardiac structures or the coronary tree.^4,10

Extensive coronary calcification can hinder assessment of the coronary tree by CT coronary angiography.

Contrast exposure may limit the use of CT in some patients (eg, those with contrast allergies or renal dysfunction). However, modern scanners allow for much smaller contrast boluses without decreasing sensitivity.

4D CT involves radiation exposure, especially when done with angiography, although modern scanners have greatly reduced exposure. The average radiation dose in CT coronary angiography is 2.9 to 5.9 mSv¹¹ compared with 7 mSv in diagnostic cardiac catheterization (without angioplasty or stenting) or 16 mSv in routine CT of the abdomen and pelvis with contrast.^12,13 In view of the morbidity and mortality risks associated with infective endocarditis, especially if the diagnosis is delayed, this small radiation exposure may be justifiable.

Bottom line for cardiac CT

4D CT is an excellent alternative to echocardiography for select patients. Clinicians should strongly consider this study in the following situations:

• Patients with a prosthetic valve
• Patients who are strongly suspected of having infective endocarditis but who have a poor sonic window on TTE or TEE, as can occur with chronic obstructive lung disease, morbid obesity, or previous thoracic or cardiovascular surgery
• Patients who meet clinical indications for TEE, such as having a prosthetic valve or a high suspicion for native valve infective endocarditis with negative TTE, but who have contraindications to TEE
• As an alternative to TEE for preoperative evaluation in patients with known infective endocarditis.

Patients with tachycardia or irregular heart rhythms are not good candidates for this test.

FDG-PET AND LEUKOCYTE SCINTIGRAPHY

FDG-PET and leukocyte scintigraphy are other options for diagnosing infective endocarditis and determining the presence and extent of intra- and extracardiac infection. They are more sensitive than echocardiography for detecting infection of cardiac implanted electronic devices such as ventricular assist devices, pacemakers, implanted cardiac defibrillators, and cardiac resynchronization therapy devices.^14–16

The utility of FDG-PET is founded on the uptake of ¹⁸F-fluorodeoxyglucose by cells, with higher uptake taking place in cells with higher metabolic activity (such as in areas of inflammation). Similarly, leukocyte scintigraphy relies on the use of radiolabeled leukocytes (ie, leukocytes previously extracted from the patient, labelled, and re-introduced into the patient) to allow for localization of inflamed tissue.

The most significant contribution of FDG-PET may be the ability to detect infective endocarditis early, when echocardiography is initially negative. When abnormal FDG uptake was included in the modified Duke criteria, it increased the sensitivity to 97% for detecting infective endocarditis on admission, leading some to propose its incorporation as a major criterion.¹⁷ In patients with prosthetic valves and suspected infective endocarditis, FDG-PET was found in one study to have a sensitivity of up to 91% and a specificity of up to 95%.¹⁸

Both FDG-PET and leukocyte scintigraphy have a high sensitivity, specificity, and negative predictive value for cardiac implanted electronic device infection, and should be strongly considered in patients in whom it is suspected but who have negative or inconclusive findings on echocardiography.^14,15

In addition, a common conundrum faced by clinicians with use of echocardiography is the difficulty of differentiating thrombus from infected vegetation on valves or device lead wires. Some evidence indicates that FDG-PET may help to discriminate between vegetation and thrombus, although more rigorous studies are needed before its use for that purpose can be recommended.¹⁹

Both FDG-PET and leukocyte scintigraphy perform poorly for detecting native-valve infective endocarditis. In a study in which 90% of the patients had native-valve infective endocarditis according to the Duke criteria, FDG-PET had a specificity of 93% but a sensitivity of only 39%.²⁰

Both studies can be cumbersome, laborious, and time-consuming for patients. FDG-PET requires a fasting or glucose-restricted diet before testing, and the test itself can be complicated by development of hyperglycemia, although this is rare.

While FDG-PET is most effective in detecting infections of prosthetic valves and cardiac implanted electronic devices, the results can be falsely positive in patients with a history of recent cardiac surgery (due to ongoing tissue healing), as well as maladies other than infective endocarditis that lead to inflammation, such as vasculitis or malignancy. Similarly, for unclear reasons, leukocyte scintigraphy can yield false-negative results in patients with enterococcal or candidal infective endocarditis.²¹

FDG-PET and leukocyte scintigraphy are more expensive than TEE and cardiac CT²² and are not widely available.

Both tests entail radiation exposure, with the average dose ranging from 7 to 14 mSv. However, this is less than the average amount acquired during percutaneous coronary intervention (16 mSv), and overlaps with the amount in chest CT with contrast when assessing for pulmonary embolism (7 to 9 mSv). Lower doses are possible with optimized protocols.^12,13,15,23

Bottom line for nuclear studies

FDG-PET and leukocyte scintigraphy are especially useful for patients with a prosthetic valve or cardiac implanted electronic device. However, limitations must be kept in mind.

A suggested algorithm for testing with nuclear imaging is shown in Figure 2.^1,4

CEREBRAL MAGNETIC RESONANCE IMAGING

Cerebral magnetic resonance imaging (MRI) is more sensitive than cerebral CT for detecting emboli in the brain. According to American Heart Association guidelines, cerebral MRI should be done in patients with known or suspected infective endocarditis and neurologic impairment, defined as headaches, meningeal symptoms, or neurologic deficits. It is also often used in neurologically asymptomatic patients with infective endocarditis who have indications for valve surgery to assess for mycotic aneurysms, which are associated with increased intracranial bleeding during surgery.

MRI use in other asymptomatic patients remains controversial.²⁴ In cases with high clinical suspicion for infective endocarditis and no findings on echocardiography, cerebral MRI can increase the sensitivity of the Duke criteria by adding a minor criterion. Some have argued that, in patients with definite infective endocarditis, detecting silent cerebral complications can lead to management changes. However, more studies are needed to determine if there is indeed a group of neurologically asymptomatic infective endocarditis patients for whom cerebral MRI leads to improved outcomes.

Limitations of cerebral MRI

Cerebral MRI cannot be used in patients with non-MRI-compatible implanted hardware.

Gadolinium, the contrast agent typically used, can cause nephrogenic systemic fibrosis in patients who have poor renal function. This rare but serious adverse effect is characterized by irreversible systemic fibrosis affecting skin, muscles, and even visceral tissue such as lungs. The American College of Radiology allows for gadolinium use in patients without acute kidney injury and patients with stable chronic kidney disease with a glomerular filtration rate of at least 30 mL/min/1.73 m². Its use should be avoided in patients with renal failure on replacement therapy, with advanced chronic kidney disease (glomerular filtration rate < 30 mL/min/1.73 m²), or with acute kidney injury, even if they do not need renal replacement therapy.²⁵

Concerns have also been raised about gadolinium retention in the brain, even in patients with normal renal function.^26–28 Thus far, no conclusive clinical adverse effects of retention have been found, although more study is warranted. Nevertheless, the US Food and Drug Administration now requires a black-box warning about this possibility and advises clinicians to counsel patients appropriately.

Bottom line on cerebral MRI

Cerebral MRI should be obtained when a patient presents with definite or possible infective endocarditis with neurologic impairment, such as new headaches, meningismus, or focal neurologic deficits. Routine brain MRI in patients with confirmed infective endocarditis without neurologic symptoms, or those without definite infective endocarditis, is discouraged.

CARDIAC MRI

Cardiac MRI, typically obtained with gadolinium contrast, allows for better 3D assessment of cardiac structures and morphology than echocardiography or CT, and can detect infiltrative cardiac disease, myopericarditis, and much more. It is increasingly used in the field of structural cardiology, but its role for evaluating infective endocarditis remains unclear.

Cardiac MRI does not appear to be better than echocardiography for diagnosing infective endocarditis. However, it may prove helpful in the evaluation of patients known to have infective endocarditis but who cannot be properly evaluated for disease extent because of poor image quality on echocardiography and contraindications to CT.^1,29 Its role is limited in patients with cardiac implanted electronic devices, as most devices are incompatible with MRI use, although newer devices obviate this concern. But even for devices that are MRI-compatible, results are diminished due to an eclipsing effect, wherein the device parts can make it hard to see structures clearly because the “brightness” basically eclipses the surrounding area.⁴

Concerns regarding use of gadolinium as described above need also be considered.

The role of cardiac MRI in diagnosing and managing infective endocarditis may evolve, but at present, the 2017 American College of Cardiology and American Heart Association appropriate-use criteria discourage its use for these purposes.¹⁶

Bottom line for cardiac MRI

Cardiac MRI to evaluate a patient for suspected infective endocarditis is not recommended due to lack of superiority compared with echocardiography or CT, and the risk of nephrogenic systemic fibrosis from gadolinium in patients with renal compromise.