Treatment of Heart Failure in Diabetes : Systolic Dysfunction, Diastolic Dysfunction, and Post-Acute Coronary Syndrome

Angiotensin-Converting Enzyme Inhibitors

Two key randomized controlled trials (RCTs), the Cooperative North Scandinavian Enalapril Survival Study (CONSENSUS) and the treatment arm of the Studies of Left Ventricular Dysfunction (SOLVD-Treatment) assigned approximately 2800 patients with mild to severely symptomatic heart failure to placebo or enalapril. ^, Most patients were also treated with a diuretic and digoxin, but only approximately 10% of patients in each trial were treated with a beta blocker. In CONSENSUS, which enrolled patients with severely symptomatic heart failure, 53% of patients were treated with spironolactone. Each trial showed that angiotensin-converting enzyme (ACE) inhibitor treatment reduced mortality (relative risk reduction [RRR] 27% in CONSENSUS and 16% in SOLVD-Treatment). In SOLVD-Treatment there was also an RRR of 26% in heart failure hospitalization. The absolute risk reduction (ARR) in mortality in patients with mild to moderately symptomatic heart failure (SOLVD-Treatment) was 4.5%, equating to a number needed to treat (NNT) of 22 to postpone one death (over an average of 41 months). The equivalent figures for severely symptomatic heart failure (CONSENSUS) were 14.6% for ARR and 7 for NNT (over an average of 6 months). These findings are supported by a meta-analysis of smaller, short-term, placebo-controlled trials, which showed a clear reduction in mortality within only 3 months. These trials also showed that ACE inhibitor treatment improves symptoms, quality of life, and functional capacity.

CONSENSUS had too few patients with diabetes (n = 253) to justify subgroup analysis. It is surprising to note that no diabetes subgroup analysis of SOLVD-Treatment has been published. However, the effectiveness of ACE inhibitors in patients with both diabetes and heart failure or postinfarction LV systolic dysfunction has been examined in a large meta-analysis of seven RCTs including SOLVD. Of the 12,586 patients included in that analysis, 2398 had diabetes. For the endpoint of all-cause mortality, ACE inhibitors had a similar treatment benefit in patients with and without diabetes: hazard ratio (HR) 0.84 (95% CI 0.70-1.00) and 0.85 (95% CI 0.78-0.92), respectively.

Although detailed reports of adverse events in SOLVD have been published, these do not describe the subgroup of patients with diabetes.

The only large ACE inhibitor trial in HF-REF (n = 3164 patients) to provide detailed information on patients with diabetes (n = 611) was the Assessment of Treatment with Lisinopril and Survival (ATLAS) trial, which compared low-dose (2.5 to 5.0 mg daily) with high-dose (32.5 to 35.0 mg daily) lisinopril. ^, Overall, this trial showed no difference in the primary outcome of all-cause mortality over the median follow-up of 46 months but did show a significant reduction in a number of other endpoints including the composite of all-cause mortality or hospitalization for heart failure: HR 0.85 (0.78-0.93), P < 0.001. These findings, in addition to those of the Heart Failure Endpoint Evaluation of Angiotensin II Antagonist Losartan (HEAAL) trial, discussed later, suggest that higher doses of renin-angiotensin-aldosterone system (RAAS) blockers are better than lower doses. In ATLAS, the greater relative benefit of high-dose lisinopril was similar in patients with and without diabetes, but because patients with diabetes were at greater risk, the absolute benefit of high-dose lisinopril was larger in patients with diabetes ( Figs. 26-1 and 26-2 ). Overall, hypotension, renal dysfunction, and hyperkalemia were slightly more frequent in the high-dose group than in the low-dose group, but hypokalemia and cough were less common in the high-dose group. The rate of study drug discontinuation was not greater in the high-dose group (17% versus 18% in the low-dose group). The occurrence of adverse effects with high-dose lisinopril was similar in those with and without diabetes with respect to hypotension and dizziness (35% versus 32%, respectively), renal dysfunction and hyperkalemia (29% versus 22%), and cough (12% versus 10%).

Additional support for the use of ACE inhibitors comes from a trial in patients with a low ejection fraction (EF) but no symptoms of HF (“asymptomatic LV systolic dysfunction”)—that is, the prevention arm of SOLVD (SOLVD-Prevention)—and three large (5966 patients in total) placebo-controlled, randomized, outcome trials in patients with heart failure, LV systolic dysfunction, or both after acute myocardial infarction. ^,

In the SOLVD-Prevention trial (which randomized 4228 patients with asymptomatic LV systolic dysfunction), there was a 20% RRR in death or heart failure hospitalization. Although 15% of patients in SOLVD-Prevention had diabetes, outcomes in this subgroup were not reported.

In myocardial infarction trials that evaluated captopril (Survival and Ventricular Enlargement [SAVE]), ramipril (Acute Infarction Ramipril Efficacy [AIRE]), and trandolapril (Trandolapril Cardiac Evaluation [TRACE]), there was a 26% RRR in death and a 27% RRR in death or HF hospitalization. The meta-analysis of these trials did not report a subgroup analysis according to baseline diabetes status.

In a subgroup analysis of SAVE, the RRR in all-cause mortality with captopril 50 mg three times daily in patients with diabetes was 12% (-21% to 36%) compared with 20% (2% to 35%) in those without. For the composite outcome of cardiovascular (CV) mortality, heart failure requiring either ACE inhibitor treatment or hospitalization, or the occurrence of recurrent infarction, these reductions were 17% (-6% to 36%) and 26% (12% to 37%), respectively. In a later report describing a multivariable analysis, captopril was reported to decrease all-cause mortality (HR 0.81; 95% confidence interval [CI] 0.68-0.96) as well as CV mortality or morbidity (HR 0.75; 95% CI 0.65-0.86). The benefit of captopril was similar among patients with (HR 0.83; 95% CI 0.63-0.87) and without (HR 0.80; 95% CI 0.64-0.94) diabetes (interaction P = 0.45).

Angiotensin Receptor Blockers

More patients were randomized into large angiotensin receptor blocker (ARB) outcome trials than ACE inhibitor trials.

An ARB was examined as an alternative to an ACE inhibitor in the Candesartan in Heart Failure: Assessment of Reduction in Mortality and Morbidity (CHARM-Alternative), which was a placebo-controlled RCT with candesartan in 2028 patients with a left ventricular ejection fraction (LVEF) of 40% or lower who were intolerant of an ACE inhibitor. Treatment with candesartan resulted in an RRR of CV or heart failure hospitalization of 23% (ARR 7%, NNT 14, over 34 months of follow-up). Valsartan was also beneficial in the subset of 366 patients in the Valsartan Heart Failure Trial (Val-HeFT) not treated with an ACE inhibitor. Another trial, the Evaluation of Losartan in the Elderly (ELITE II), failed to show that losartan 50 mg daily was as effective as captopril 50 mg three times daily. However, a subsequent RCT, the HEAAL trial, showed that 150 mg daily of losartan (n = 1927) was superior to 50 mg daily (n = 1919), supporting the similar findings of the ATLAS trial with the ACE inhibitor lisinopril (see earlier). In HEAAL, there was a RRR of 10% in death or HF hospitalization in the high-dose losartan group ( P = 0.027) over a median follow-up of 4.7 years.

The efficacy and safety of adding an ARB to an ACE inhibitor has been studied. Two key placebo-controlled RCTs, Val-HeFT and CHARM-Added, randomized approximately 7600 patients with mild to severely symptomatic heart failure to placebo or an ARB (valsartan and candesartan), added to an ACE inhibitor (in 93% of patients in Val-HeFT and all patients in CHARM-Added). In addition, approximately a third of patients in Val-HeFT and just over half in CHARM-Added were treated with a beta blocker, but few patients were taking a mineralocorticoid receptor antagonist (MRA). Each of these two trials showed that ARB treatment reduced the risk of heart failure hospitalization (RRR 24% in Val-HeFT and 17% in CHARM-Added). There was a 16% RRR in the risk of CV death with candesartan in CHARM-Added, but CV death was not reduced by valsartan in Val-HeFT. ^, The ARR in the primary composite mortality-morbidity endpoint in patients with mild to moderately severe symptoms was 4.4%, equating to an NNT (for an average of 41 months to postpone one event) of 23 in CHARM-Added. The equivalent figures for Val-HeFT were ARR 3.3% and NNT 30 (over an average of 23 months). The CHARM trials and Val-HeFT also showed that ARBs improve symptoms and quality of life. Other trials showed that these agents improve exercise capacity.

Pooling the two CHARM HF-REF trials (CHARM-Alternative and CHARM-Added) showed that treatment with candesartan reduced the risk of CV death or heart failure hospitalization with an HR of 0.82 (95% CI 0.74-0.90, P < 0.001) and all-cause mortality (HR 0.84; 95% CI 0.79-0.98; P = 0.018). The effect of treatment was not statistically different in patients with and without diabetes (interaction P = 0.12). The effect of candesartan in patients with HF-REF and diabetes in the CHARM program is shown in Figure 26-3 . In Val-HeFT the effect of valsartan was not statistically different in the subgroup of patients with diabetes (interaction P value not provided).

In HEAAL the treatment effect of high-dose compared with low-dose losartan was not different in the subgroup of patients with diabetes (HR 0.96; 95% CI 0.82-1.12; P = 0.35). However, in all three reports, the point estimate for the HR was less favorable in patients with diabetes than in those without.

There is little information about the tolerability of ARBs in diabetes. In the overall CHARM program, patients with diabetes had double the risk of developing hyperkalemia on candesartan compared with those without diabetes.

Additional support for the use of ARBs comes from the Valsartan in Acute Myocardial Infarction (VALIANT) trial, a trial in which 14,703 patients with heart failure, LV systolic dysfunction, or both after acute myocardial infarction were assigned to treatment with captopril 50 mg three times daily, valsartan 160 mg twice daily, or the combination. Valsartan was found to be noninferior to captopril overall, but combination therapy was not better than monotherapy. The effect of valsartan relative to captopril was not significantly different in patients with and without diabetes, either for all-cause mortality (interaction P = 0.10) or CV mortality and morbidity (interaction P = 0.12).

In the Optimal Therapy in Myocardial Infarction with the Angiotensin II Antagonist Losartan (OPTIMAAL) trial, losartan 50 mg once daily did not demonstrate noninferiority when compared with captopril 50 mg three times daily, which, considered in conjunction with the findings of ELITE II and HEAAL, suggests that 50 mg of losartan daily is a suboptimal dose.

Direct Renin Inhibitors

Recently, a third approach to blocking the renin-angiotensin system has been tested in heart failure. In the Aliskiren Trial on Acute Heart Failure Outcomes (ASTRONAUT) trial, patients with an LVEF of 40% or lower, elevated natriuretic peptides, and signs and symptoms of fluid overload who had been admitted on an emergency basis to a hospital for the treatment of heart failure were randomized, when stabilized (a median 5 days after admission), to 12 months of treatment with the direct renin inhibitor (DRI) aliskiren 150 mg daily (increased to 300 mg as tolerated) or placebo daily. The study drug was given in addition to standard therapy, which included diuretics (96%), beta blockers (83%), ACE inhibitors or ARBs (84%), and MRAs (57%). The main endpoint was the composite of CV death or heart failure rehospitalization at 6 months (the main secondary endpoint was this composite at 12 months). In total, 1639 patients were randomized and 1615 patients included in the final efficacy analysis cohort (808 aliskiren, 807 placebo). The mean age was 65 years, LVEF was 28%, and estimated glomerular filtration rate (eGFR) was 67 mL/min/1.73 m ² ; 41% of patients had diabetes mellitus. Overall, 24.9% of patients receiving aliskiren (77 CV deaths, 153 HF rehospitalizations) and 26.5% of patients receiving placebo (85 CV deaths, 166 HF rehospitalizations) experienced the primary endpoint at 6 months (HR 0.92, 95% CI 0.76-1.12; P = 0.41). At 12 months, the rates were 35.0% for aliskiren (126 CV deaths, 212 HF rehospitalizations) and 37.3% in the placebo group (137 CV deaths, 224 HF rehospitalizations); HR 0.93, 95% CI 0.79-1.09; P = 0.36. The rates of hypotension, renal dysfunction, and hyperkalemia were higher in the aliskiren group compared with placebo.

Although there was no evidence for heterogeneity of treatment effect for any subgroup with respect to the primary endpoint, there was a significant interaction between treatment and diabetes status at baseline (patients with diabetes: HR 1.16, 95% CI 0.91-1.47; no diabetes group: HR 0.80, 95% CI 0.64-0.99; P = 0.03 for interaction). There was also an interaction for all-cause mortality at 1 year—diabetes patients: HR 1.64, 95% CI 1.15-2.33; no diabetes group: HR 0.69, 95% CI 0.50-0.94; P = 0.001 for interaction). Among patients with a history of diabetes, 24.1% of patients died in the aliskiren group compared with 17.4% of patients in the placebo group, whereas the rates of death in patients without diabetes were 15.3% and 20.0% in the two treatment groups, respectively.

Whereas subgroup findings may arise by chance and are normally only considered hypothesis generating, similar findings of increased risks of hypotension, renal dysfunction, and hyperkalemia in the Aliskiren Trial in Type 2 Diabetes Using Cardiorenal Endpoints (ALTITUDE) convinced regulatory agencies to declare that aliskiren is contraindicated in patients with diabetes who are receiving an ARB or ACE inhibitor and in patients with an eGFR below 60 mL/min/1.73 m ² . In Europe this prohibition was extended to monotherapy with aliskiren in patients with diabetes or chronic kidney disease, which resulted in patients with diabetes having study drug discontinued in the ongoing Aliskiren Trial of Minimizing Outcomes for Patients with Heart Failure (ATMOSPHERE) comparing aliskiren (up to 300 mg once daily), enalapril (10 mg twice daily), and the combination of aliskiren and enalapril in over 7000 patients with chronic HF-REF. Despite this, ATMOSPHERE is expected to remain adequately powered and to run to its planned completion.

Beta Blockers

There is more evidence showing the benefit of a beta blocker in HF-REF than any other pharmacologic therapy, yet patients with diabetes are less likely to receive this type of treatment than those without diabetes.

Three key trials—the Cardiac Insufficiency Bisoprolol Study II (CIBIS II), Carvedilol Prospective Randomized Cumulative Survival trial (COPERNICUS), and Metoprolol CR/XL Randomised Intervention Trial in Congestive Heart Failure (MERIT-HF) — randomized almost 9000 patients with mild to severely symptomatic heart failure to placebo or a beta blocker (bisoprolol, carvedilol, or metoprolol succinate CR/XL). More than 90% of the patients were on background treatment with an ACE inhibitor or an ARB. Each of these three trials showed that beta blocker treatment reduced mortality (RRR approximately 34% in each trial) and heart failure hospitalization (RRR 28% to 36%) within approximately a year of starting treatment. The ARR in mortality (after 1 year of treatment) in patients with mild to moderate symptoms (CIBIS II and MERIT-HF combined) was 4.3%, equating to an NNT for 1 year to postpone one death of 23. The equivalent figures for severely symptomatic patients (COPERNICUS) were ARR 7.1% and NNT 14. These findings are supported by another placebo-controlled trial, the Study of Effects of Nebivolol Intervention on Outcomes and Rehospitalization in Seniors with Heart Failure (SENIORS) in 2128 elderly (70 years or older) patients, 36% of whom had an EF below 35%. Treatment with nebivolol resulted in an RRR of 14% in the primary composite endpoint of death or CV hospitalization but did not reduce mortality. The findings of these trials are also supported by an earlier program of studies with carvedilol (U.S. carvedilol studies).

Several subgroup analyses from these trials and meta-analyses have shown the efficacy and safety of beta blockers specifically in patients with HF-REF and diabetes. In a meta-analysis of six trials, Haas and colleagues found that beta blocker therapy reduced all-cause mortality in patients with diabetes (n = 3230; HR 0.84, 95% CI 0.73-0.96) and those without diabetes (n = 9899; HR 0.72, 95% CI 0.65-0.79). Shekelle, and colleagues analyzed data from three trials (CIBIS-II, MERIT-HF, and COPERNICUS) that gave a relative risk of 0.77 (95% CI 0.61, 0.96) in patients with diabetes (n = 1883) and 0.65 (95% CI 0.57, 0.74) in patients without diabetes (n = 7042). A third meta-analysis focused on seven trials in which carvedilol was used, including a postinfarction trial, which collectively enrolled 1411 patients with diabetes (24.5% of total patients). In all randomized patients, carvedilol reduced all-cause mortality (RRR 34%, 95% CI 23%-44%; P < 0.0001). In patients with diabetes the RRR was 28% (95% CI 3% to 46%; P = 0.029), and it was 37% (95% CI 22% to 48%; P < 0.0001) in patients without diabetes (interaction P = 0.25). The NNT for 1 year to prevent one death was 23 (95% CI 17-36) for all patients, 25 (14-118) in patients with diabetes and 23 (95% CI 17-37) in those without diabetes.

The MERIT-HF investigators published a detailed analysis of both the efficacy and the safety of metoprolol succinate compared with placebo in patients with (n = 985) and without (n = 3006) diabetes. It is important to note that this analysis also reports key nonfatal outcomes (and fatal or nonfatal composites) and shows that these too were reduced by beta blocker therapy in patients with diabetes ( Figs. 26-4 and 26-5 ). Patients with diabetes were more likely to experience adverse events (in either treatment group) compared with patients without diabetes. However, patients receiving metoprolol succinate were less likely to experience adverse events than those treated with placebo, both patients with and those without diabetes, and were less likely to discontinue the study drug. The average dose of metoprolol succinate taken during the trial was similar in patients with (162 mg at the last follow-up visit) and without (156 mg) diabetes. There was no difference between metoprolol succinate and placebo in relation to adverse events indicating impaired glycemic control.

Hypoglycemia is a particular concern in patients with diabetes treated with insulin or sulfonylureas. Theoretically, beta blockers could alter awareness of hypoglycemia by decreasing palpitations and tremor and could prolong recovery from hypoglycemia by blocking beta ₂ receptors, which partly control glucose production in the liver.

However, among patients with diabetes in MERIT-HF, only three (0.6%) in the placebo group and four (0.8%) in the metoprolol succinate group had an adverse event related to hypoglycemia (in each case in patients taking insulin).

Another trial, the Carvedilol or Metoprolol European Trial (COMET), ^, showed that carvedilol reduced mortality compared with short-acting metoprolol tartrate (different from the long-acting succinate formulation used in MERIT-HF). In COMET, both patients with and those without diabetes had similar risk reductions for mortality on carvedilol compared with metoprolol: relative risk 0.85, 95% CI 0.69-1.06, P = 0.147 for those with diabetes; and RR 0.82, 95% CI 0.71-0.94, P = 0.006 for patients without diabetes (interaction P = 0.77).

Of note, in a trial in patients with hypertension and diabetes, carvedilol had a favorable effect on glycated hemoglobin (and insulin sensitivity) compared with metoprolol tartrate.

The benefit of beta blockers in heart failure is supported by a placebo-controlled trial in 1959 patients with an LVEF of 40% or lower after acute myocardial infarction—the Carvedilol Post-Infarct Survival Control in Left Ventricular Dysfunction (CAPRICORN) trial, in which the RRR in mortality with carvedilol was 23% during a mean follow-up of 1.3 years. Among the 437 patients with diabetes in CAPRICORN, the placebo-carvedilol HR for all-cause mortality was 0.93 (95% CI 0.61-1.44) compared with HR 0.71 (95% CI 0.53-0.96) in those without diabetes (n = 1522). The corresponding figures for the composite of death or CV hospitalization were HR 0.88 (95% CI 0.80-1.13) and HR 0.95 (95% CI 0.66-1.16), respectively.

In summary, beta blockers in patients with diabetes and heart failure lead to significant improvements in morbidity and mortality associated with heart failure, benefits that far outweigh the theoretical risks related to hypoglycemia and minor changes in glycated hemoglobin and lipids.

Mineralocorticoid Receptor Antagonists

In the Randomized Aldactone Evaluation Study (RALES), 1663 patients with an EF of 35% or lower and in New York Heart Association (NYHA) functional class III (if in class IV within the past 6 months) or IV were randomized to placebo or spironolactone 25 to 50 mg once daily, added to conventional treatment. When this trial was conducted, beta blockers were not widely used to treat heart failure, and only 11% of patients were treated with a beta blocker. Treatment with spironolactone led to an RRR in all-cause mortality of 30% and an RRR in heart failure hospitalization of 35%. The ARR in mortality (after a mean of 2 years of treatment) was 11.4%, equating to an NNT (for 2 years to postpone one death) of 9. More recently, the Eplerenone in Mild Patients Hospitalization and Survival Study in Heart Failure (EMPHASIS-HF) trial enrolled 2737 patients aged 55 years or older with NYHA functional class II symptoms and an EF of 30% or lower (≤ 35% if the QRS duration was > 130 milliseconds). Patients had to have either a CV hospitalization within the previous 6 months or an elevated plasma natriuretic peptide concentration and to have been treated with an ACE inhibitor, ARB, or both, and a beta blocker. Treatment with eplerenone (up to 50 mg once daily) for an average of 21 months led to a RRR of 37% in CV death or heart failure hospitalization. There were also reductions in the risk of death from any cause (24%), CV death (24%), all-cause hospitalization (23%), and heart failure hospitalization (42%). The ARR in the primary composite mortality-morbidity endpoint was 7.7%, equating to an NNT (for an average of 21 months to postpone one event) of 13. The ARR in all-cause mortality was 3%, equating to an NNT of 33.

In RALES, the mortality benefit was similar in patients with and without diabetes, with HR of 0.70 (95% CI 0.52-0.94) and HR 0.70 (95% CI 0.60-0.82), respectively. In EMPHASIS-HF, the treatment benefit was consistent in patients with and without diabetes (interaction P = 0.10; Fig. 26-6 ).

The finding that MRAs are beneficial in patients with chronic HF-REF is supported by another trial, the Eplerenone Post–Acute Myocardial Infarction Heart Failure Efficacy and Survival Study (EPHESUS), which enrolled 6632 patients with an EF of 40% or lower and HF or diabetes 3 to 14 days after acute myocardial infarction. Patients were randomized to placebo or eplerenone 25 to 50 mg once daily added to conventional treatment including an ACE inhibitor or ARB and a beta blocker. Treatment with eplerenone reduced all-cause mortality (RR 0.85, 95% CI 0.75-0.96; P = 0.008), and reduced the rate of the other co-primary endpoint, a composite of death from CV causes or hospitalization for CV events (RR 0.87, 95% CI 0.79-0.95; P = 0.002).

A subgroup analysis for the 1483 patients with diabetes in EPHESUS has been published. The eplerenone-placebo relative risk of death from any cause in these patients was RR 0.85 (95% CI 0.68-1.05). The relative risk of death from CV causes or hospitalization for CV events was RR 0.83 (95% CI 0.71-0.98).

Hyperkalemia has been a particular concern with MRAs, and it has been suggested that this risk may be greater in patients with diabetes. There was an increase in the incidence of a potassium concentration exceeding 5.5 mmol/L with eplerenone in patients with diabetes—63 (14.1%) compared with 33 (8.5%) on placebo, P = 0.01. However, the proportion with a serum potassium level above 6.0 mmol/L was not statistically different between the groups (17 [3.8%] compared with 8 [2.1%] on placebo, P = 0.16), and there was no increase in the rate of discontinuation of eplerenone for hyperkalemia in those with diabetes compared with those without (interaction P = 0.12). Similar findings we reported in EPHESUS. Among patients with diabetes in EPHESUS, hypoglycemia was reported as an adverse event in 11 patients (1.5%) treated with eplerenone and 11 patients (2.6%) treated with placebo ( P = 0.14).

Some investigators believe that eplerenone may have more favorable metabolic effects than spironolactone in patients with diabetes.

Nitrates and Hydralazine

The African-American Heart Failure Trial (A-HeFT) examined the safety and efficacy of fixed-dose combination therapy with isosorbide dinitrate and hydralazine hydrochloride, added to an ACE inhibitor or ARB, beta blocker, and MRA in African Americans with NYHA class III or IV heart failure. The trial had an unusual composite outcome including survival, hospitalization, and quality of life but was stopped early because treatment with this medication combination led to a 43% reduction in all-cause death (HR 0.57, 95% CI 0.37-0.89; P = 0.01). There was also a 33% reduction in first hospitalization for heart failure ( P = 0.001). A very large proportion (41%) of patients in the study had diabetes. The treatment effect on mortality was similar in patients with (HR 0.56 [95% CI 0.28-1.15]) and without (HR 0.59 [95% CI 0.34-1.03]) diabetes.

Ivabradine

The Systolic Heart Failure Treatment with the / _f Inhibitor Ivabradine Trial (SHIFT) enrolled 6588 patients in NYHA functional class II to IV, sinus rhythm with a rate of 70 beats/min or greater, and an EF of 35% or lower. Patients were also required to have had an HF hospitalization in the previous 12 months. They were randomized to ivabradine (uptitrated to a maximal dosage of 7.5 mg twice daily) or placebo, added to a diuretic and an ACE inhibitor or ARB, a beta blocker, and an MRA. Only 26% of patients were, however, on full-dose beta blocker. The median follow-up was 23 months. The RRR in the primary composite outcome of CV death or HF hospitalization was 18% (HR 0.82, 95% CI 0.75-0.90; P < 0.0001); the reduction in CV death (or all-cause death) was not significant, but the RRR in HF hospitalization was 26%. Ivabradine also improved LV function and quality of life. In the 1979 patients with diabetes, the HR for the primary composite endpoint was 0.81 (95% CI 0.69-0.95), and in those without diabetes (n = 4526) it was 0.83 (95% CI 0.74-0.93).