Angiotensin II Receptor Antagonists (AIIRAs)
Objective
To provide recommendations for the appropriate use of AIIRAs (also referred to as ARBs) in patients with a diagnosis of systolic HF
Annotation
ACEIs reduce levels of angiotensin II, a potent vasoconstrictor, and inhibit the breakdown of bradykinin, a vasodilator. Production of angiotensin II also occurs through alternative pathways. The AIIRAs, on the other hand, selectively block the angiotensin II type1 receptor so that the effects of angiotensin II are blocked regardless of how it is produced. The AIIRAs do not inhibit the angiotensin II type 2 receptor, which is thought to have beneficial effects such as vasodilation and inhibition of proliferative and hypertrophic responses. The AIIRAs do not affect bradykinin, which is thought to be responsible for the cough that occurs in up to 39% of patients taking an ACEI. The incidence of cough in patients treated with an AIIRA is similar to that with placebo. The contribution of bradykinin to the favorable results of the ACEI trials in HF patients is unknown, but may be as important as suppression of angiotensin.
In the ELITE (Evaluation of Losartan in the Elderly) Study, the AIIRA losartan was compared to an ACEI, captopril, in 722 patients with NYHA class II to IV HF and LVEF <40%. Patients were randomized to losartan (up to 50 mg) once daily or captopril (up to 50 mg) three times daily for 48 weeks. Seventy-five percent of patients in the losartan group and 71% of patients in the captopril group received target doses. The majority of patients were prescribed diuretics, and 55% were taking digoxin at the time of study enrollment. The primary endpoint of the study was the effect of treatment on serum Cr (>0.3 mg/dL increase). There was no difference between treatment groups in the rise in serum creatinine during continued treatment. Death and/or hospitalization for HF occurred in 9.4% of patients on losartan and 13.2% on captopril (32% risk reduction, P = 0.075). These results were primarily due to a 46% decrease in all-cause mortality in patients on losartan compared to patients on captopril (P = 0.035), primarily due to a reduction in sudden cardiac death. The two treatment groups did not differ in the frequency of hospital admission for HF. NYHA functional class improved significantly and similarly compared to baseline for both groups. More patients in the captopril group (20.8%) withdrew from the study due to adverse events compared to patients in the losartan group (12.2%). Cough was reported in 3.8% of patients taking 28 captopril compared to 0% in losartan treated patients. The favorable mortality rate in the losartan group was not hypothesized a priori. Therefore, replication of the results was attempted in ELITE II.
ELITE II enrolled 3,152 HF patients to evaluate the effects of losartan 50 mg once daily compared to captopril 50 mg three times daily on overall mortality and cardiac events (sudden cardiac death or resuscitated cardiac arrest). There was no significant difference in all-cause mortality between the treatment groups (17.7% on losartan vs. 15.9% on captopril, P = 0.16). There was no difference between the groups in sudden death or resuscitated cardiac arrest or hospital admissions. However, this was a superiority trial not designed to detect equivalence between groups. Therefore, losartan and captopril cannot be concluded to be the same. Patients receiving captopril had significantly more adverse effects resulting in discontinuation of the drug than patients on losartan (P <0.001).
The RESOLVD Pilot Study compared candesartan, enalapril, and the combination of the two agents in 768 patients with NYHA class II to IV HF with a LVEF <40%. Patients were placed on candesartan (4, 8, or 16 mg), candesartan (4 or 8 mg) plus enalapril (20 mg), or enalapril (20 mg) for 43 weeks. The primary endpoints were exercise tolerance, ventricular function, quality of life, neurohormone levels, and tolerability. There was no significant difference between the treatment groups in results of the six-minute walk test, NYHA functional class, or quality of life. There was a trend toward an increase in ejection fraction, although not significant, in the patients treated with candesartan and enalapril compared to patients on candesartan or enalapril. End-diastolic and end-systolic volumes increased less with combination therapy compared with patients on candesartan or enalapril alone. There appeared to be a benefit of combination therapy on the patient's neurohormonal profile. Although not powered to evaluate morbidity and mortality, another analysis suggested that there might be an increase in HF hospitalizations in the patients receiving candesartan by 3-way group comparison.
More recently, the results of the Val-HeFT (Valsartan Heart Failure Treatment) study were published. The trial included 5,010 patients with NYHA class II (62%), III (36%), or IV (2%) HF on standard therapy (diuretics: 85%; ACEI: 93%; beta-adrenergic blockers: 35%; and digoxin 67%). Baseline LVEF was 27%. Patients were randomized to therapy with either valsartan (40 mg twice daily, titrated to a target of 160 mg twice daily) or placebo. Mean follow-up was 23 months. The two primary endpoints were mortality and the combined endpoint of mortality and morbidity (i.e., cardiac arrest with resuscitation, HF hospitalization, or intravenous inotropic agents or vasodilators for over 4 hours). Overall mortality was similar, occurring in 19.7% of patients in the valsartan group and 19.4% of patients on placebo (P = 0.80). The combined primary endpoint occurred in 28.8% and 32.1% of patients on valsartan and placebo, respectively (RR 0.87 CI 0.77-0.97, P = 0.009; ARR 3.3%; NNT = 30.3). This included a reduction in hospitalizations for HF (13.8% valsartan vs. 18.2% placebo; ARR 4.4%; NNT = 22.7). However, death from any cause (as first event) was higher in patients on valsartan compared to patients receiving placebo (14.2% vs. 12.6%, respectively). According to a subgroup analysis, there was an increased risk of mortality (P = 0.0009) and a trend toward an increased risk of combined morbidity and mortality (P = 0.10) in patients receiving valsartan in conjunction with an ACEI and beta-adrenergic blocker. Patients who were not on an ACEI or beta-adrenergic blocker experienced a significant reduction in mortality (P = 0.012). Patients on valsartan but not on an ACEI (with or without a beta-adrenergic blocker) had a lower risk of death (RR 0.67, CI 0.42-1.06) and a lower risk of the combined endpoint (RR 0.56, CI 0.39-0.81). A subanalysis of the 366 patients in Val-HeFT who were not on an ACEI was recently published. In these patients there was a 33% decrease in all-cause mortality (P = 0.017) and a 53% decrease in combined morbidity and mortality (P <0.001). The authors conclude that valsartan is an appropriate alternative in patients who are unable to tolerate and ACEI for the treatment of HF.
The AIIRAs have yet to be shown to be equivalent or superior to the ACEIs in patients with HF. According to a recent meta-analysis of 12,469 patients, the AIIRAs were not found to be superior to an ACEI in reducing mortality or hospitalizations. There was a trend toward improved mortality and hospitalizations with an AIIRA compared to placebo in patients not on an ACEI, and the combination of an AIIRA and ACEI significantly reduced the risk of hospitalizations compared to patients on an ACEI alone. In a previous meta-analysis of 1,896 patients, losartan contributed to a mortality benefit compared to a control group of either placebo or an ACEI, but this meta-analysis did not include the more recent outcome trials with an AIIRA in patients with HF.
An AIIRA should not be considered unless a patient is unable to tolerate an ACEI due to uncontrolled cough (or with caution in patients with history of angioedema; refer to discussion below). The benefit of an AIIRA in combination with an ACEI is still to be determined. Since the benefits of an ACEI in conjunction with a beta-adrenergic blocker is well-defined and there may be a detrimental effect in patients on an AIIRA with an ACEI and beta-adrenergic blocker, an AIIRA should not be used unless the patient is intolerant to an ACEI or unable to take a beta-adrenergic blocker. Additional information on the role of an AIIRA in patients with HF may be determined with the results of CHARM (candesartan in HF-assessment of reduction in mortality and morbidity).
The incidence of cough is estimated to be anywhere from 0 to 39% in patients treated with an ACEI. In SOLVD, cough was reported in 37% of patients treated with enalapril compared to 31% of patients randomized to placebo. In V-HeFT II, 37% of patients on enalapril complained of cough compared to 29% receiving hydralazine/isosorbide dinitrate (HYD/ISDN). The incidence of cough associated with the AIIRAs is similar to placebo (2.6 to 3.4% vs. 1.5 to 3.3%). In the ELITE Study, 3.8% of patients on an ACEI withdrew from the study due to complaints of cough compared to 0% of patients treated with an AIIRA. Use of an AIIRA can be considered in patients who are unable to tolerate treatment with an ACEI due to cough, although there is a slight chance that patients may develop a cough with an AIIRA.
The incidence of angioedema in patients taking ACEIs is approximately 0.1 to 1.2%. It has been reported that black American patients have an increased relative risk of 4.5 of angioedema associated with use of an ACEI compared to white patients. There are at least 20 published case reports of angioedema in patients treated with an AIIRA. In over one-third of these cases, the patients previously experienced angioedema with an ACEI. Almost 100 cases have been reported to the Therapeutic Goods Administration of Australia as of April 2001. Therefore, if an AIIRA is considered appropriate in a patient who has previously experienced angioedema, it should be used with caution.
The angiotensin II receptor antagonists, like the ACEIs, decrease release of aldosterone from the adrenal cortex, which can lead to potassium reabsorption. It is unclear at this time if treatment with an AIIRA would be an appropriate alternative in patients who develop hyperkalemia on an ACEI. In SOLVD, hyperkalemia with potassium levels greater than 5.5 mmol/L was reported in 6.4% of patients on enalapril compared to 2.5% of patients on placebo. In the ELITE Study, an increase in serum potassium of >0.5 mmol/L above baseline was observed in 22.7% patients receiving captopril compared to 18.8% of patients on losartan. The proportion of patients with potassium levels >5.5 mmol/L did not differ significantly among the treatment groups in the RESOLVD Pilot Study. The VAL-K Study Group reported that the change in serum potassium was not significantly different in patients on lisinopril compared to valsartan with mild renal insufficiency. In patients with moderate renal insufficiency with a glomerular filtration rate (GFR) <60 mL/min/1.73 m2, there was a significant increase of 0.28 mEq/L (P = 0.04) above baseline (4.6 mEq/L). The increase of 0.12 mEq/L seen with valsartan in this subgroup was not significant (P = 0.1). Therefore, if use of a diuretic is contraindicated or is not effective in reducing hyperkalemia, an AIIRA may be considered instead of an ACEI, under close monitoring, in patients with moderate renal insufficiency who develop hyperkalemia on an ACEI.
Patients receiving an AIIRA in conjunction with potassium supplements or potassium-sparing diuretics (including spironolactone) may result in an increased potassium level. Other clinically significant drug interactions with the AIIRAs are listed in Appendix B of the original guideline document.
See Table 7 titled "Angiotensin II Receptor Antagonists" in the original guideline document.
Pharmacologic recommendations for AIIRAs in patients with HF:
Strength of Recommendation and Evidence Rating
Grade A (always indicated and acceptable):
Grade B (may be useful/effective):
- Use an AIIRA in patients on standard therapy who cannot tolerate an ACEI due to cough and possibly angioedema. (Overall Quality: Fair; Net Effect: Moderate; ACC/AHA Recommendations: Class IIa; Evidence Level: A) (Pitt et al., 1997; Pitt et al., 2000; McKelvie et al., 1999; Greenberg, 1999; Cohn & Tognoni, 2001; Maggioni et al., 2002; Jong et al., 2002; Sharma et al., 2000; Hunt et al., 2001)
Grade C (may be considered):
- Use an AIIRA in addition to an ACEI in patients with HF, if not on a beta-adrenergic blocker. (Overall Quality: Fair; Net Effect: Moderate; ACC/AHA Recommendations: Class IIb; Evidence Level: B) (Cohn & Tognoni, 2001; Jong et al., 2002; Hunt et al., 2001; Struckman & Rivey, 2001)
Grade D (may not be useful/ effective; possibly harmful):
- Use an AIIRA instead of an ACEI in patients who are able to tolerate an ACEI. (Overall Quality: Fair; Net Effect: Negative; ACC/AHA Recommendations: Class III; Evidence Level: B) (Pitt et al., 1997; Pitt et al., 2000; McKelvie et al., 1999; Greenberg, 1999; Cohn & Tognoni, 2001; Jong et al., 2002; Sharma et al., 2000; Hunt et al., 2001)
- Use an AIIRA before a beta-adrenergic blocker in patients who are unable to tolerate an ACEI. (Overall Quality: Fair; Net Effect: Negative; ACC/AHA Recommendations: Class III; Evidence Level: A) (Cohn & Tognoni, 2001; Hunt et al., 2001)
Grade I (insufficient evidence to recommend for or against):