From Wiki Journal Club
Jump to navigation Jump to search
McMurray JJ, et al. "Effects of candesartan in patients with chronic heart failure and reduced left-ventricular systolic function taking angiotensin-converting-enzyme inhibitors: The CHARM-Added trial". The Lancet. 2003. 362(9386):767-771.
PubMedFull textPDF

Clinical Question

Among patients with symptomatic HFrEF on an ACE-inhibitor, does the addition of candesartan reduce the risk of cardiovascular death or HF hospitalization?

Bottom Line

Among patients LVEF ≤40% and NYHA class II-IV symptoms on an ACE-inhibitor, the addition of candesartan reduced CV mortality or HF hospitalization when compared to placebo.

Major Points

ACE-inhibitors became standard of care in HFrEF because of survival benefit in trials like CONSENSUS (1987) and SOLVD (1991). Likewise, beta-blockers became standard of care following publication of CIBIS-II (1999), MERIT-HF (1999), and COPERNICUS (2002). The role of ARB add-on therapy to ACE-inhibitors was first explored in Val-HeFT (2001), which demonstrated benefit for a combined CV endpoint, though it was driven primarily by a reduction in HF hospitalizations. Addition of ARBs led to higher rates of adverse events, including renal dysfunction. Of note, only 34% of patients were on beta blockers.

The industry-funded and analyzed 2003 Candesartan in Heart failure Assessment of Reduction in Mortality and Morbidity-Added (CHARM-Added) trial randomized 2,548 patients with LVEF ≤40% and NYHA class II-IV symptoms on ACE-inhibitor therapy to candesartan uptitrated to a goal dose of 32 mg PO qday or placebo. The baseline population differed from Val-HeFT in a higher rate of NYHA class III and IV symptoms and a higher use of beta-blocker therapy at baseline. With a median follow-up of 41 months, the participants of CHARM-Added had a reduction in CV mortality or HF hospitalizations when compared to placebo (37.9% vs. 42.3%; NNT 23). There was a reduction in CV mortality (23.7% vs. 27.3%; NNT 28) and HF hospitalizations (24.2% vs. 28.0%; NNT 26). There was no difference in all-cause mortality. Addition of candesartan was associated with higher rates of adverse outcomes including increased creatinine and hyperkalemia.

In the years since the publication of the RALES trial (1999), the use of aldosterone antagonists has become more widespread. It's important to note that only 17% of patients were on an aldosterone antagonist in this trial. A large trial comparing ARB therapy added on to ACE-inhibitor, beta blocker, and aldosterone antagonist therapies has not been performed.


AHA/ACCF Heart Failure Guidelines (2013, adapted)[1]

  • ACE inhibitors in patients with HFrEF with present or former symptoms unless contraindicated (class I, level A)
  • ARBs for above if ACE inhibitor intolerant unless contraindicated (class I, level A)
  • ARBs are reasonable as first-line agents for HFrEF (class IIa, level A)
  • ARBs can be added to ACE inhibitor and beta blocker for those in which an aldosterone antagonist is not indicated or is poorly tolerated (class IIb, level A)
  • Routine combined ACE inhibitor and ARB use is potentially harmful in HFrEF (class III, level C)


  • Multicenter, double-blind, randomized
  • N=2,548 patients (planned 2,300 sample size)
    • Candesartan (n=1,276)
    • Placebo (n=1,272)
  • Setting: 618 centers in 26 countries
  • Enrollment: March-November 1999
  • Median follow-up: 41 months
  • Analysis: Intention-to-treat
  • Primary outcome: CV mortality or HF hospitalization


Inclusion Criteria

  • Age ≥18 years
  • LVEF ≤40% in prior 6 months
  • NYHA functional class II-IV
    • class II needed ≥1 CV hospitalization in prior 6 months
  • Treatment with a stable ACE-I dose for ≥30 days

Exclusion Criteria

  • Creatinine >3 mg/dL
  • Potassium >5.5 mmol/L
  • Bilateral renal artery stenosis
  • Symptomatic hypertension
  • Women of childbearing potential
  • MI, stroke, or open heart surgery in the prior 4 weeks
  • Use of an ARB during the previous 2 weeks

Baseline Characteristics

From the candesartan group. Groups were similar.

  • Demographics: Age 64 years (≥75 years 17%), male 79%, European 90%, Black 5%
  • HF data: LVEF 28%, HF admission 76%
    • HF type: ICM 62%, idiopathic CM 27%, hypertensive CM 7%
    • NYHA class: II 24%, III 73%, IV 3%
  • Baseline health data: HR 73 BPM, BP 124/75, BMI 28 kg/m2
  • PMH: MI 56%, current angina 19%, stroke 8%, DM 30%, HTN 48%, AF 27%, smoker 15%
  • PSH: PCI 14%, CABG 26%, PPM 9%, ICD 4%, cancer 6%
  • Baseline medications: ACE-inhibitor 100%, diuretic 90%, beta blocker 55%, spironolactone 17%, digoxin 58%, CCB 10% (other vasodilator 35%), oral anticoagulation 38%, antiarrhythic 13%, ASA 51% (other antiplatelet 3%), lipid-lowering medication 41%
    • ACE-inhibitors used: Enalapril 16.8 mg/day, lisinopril 17.7 mg/day, captopril 82.2 mg/day, and ramipril 6.8 mg/day
      • Optimal ACE-inhibitor dosing per investigator opinion: 96%
      • Recommended ACE-inhibitor dosing per trial protocol: 50%


  • Patients randomized to a group:
    • Candesartan - Initiation of candesartan at 4 or 8 MG po qday, doubling q2weeks depending on tolerability and laboratory values to a goal dose of 32 mg po qday at week 6
    • Placebo - Titrated by the same parameters as candesartan
  • Patients seen at weeks 2, 4, 6, month 6, then q4month until the end of the trial


Presented as candesartan vs. placebo. UHR is unadjusted hazard ratio, AHR is adjusted hazard ratio.

Primary Outcomes

CV mortality or HF hospitalization
37.9% vs. 42.3% (UHR 0.85; 95% CI 0.75-0.96; P=0.011 | AHR 0.85; 95% CI 0.75-0.96; P=0.010; NNT 23)

Secondary Outcomes

CV mortality
23.7% vs. 27.3% (UHR 0.84; 95% CI 0.72-0.98; P=0.029 | AHR 0.83; 95% CI 0.71-0.97; P=0.021; NNT 28)
HF hospitalization
24.2% vs. 28.0% (UHR 0.83; 95% CI 0.71-0.96; P=0.014 | AHR 0.83; 95% CI 0.71-0.97; P=0.018; NNT 26)
CV mortality, HF hospitalization, or non-fatal MI
38.8% vs. 43.2% (UHR 0.85; 95% CI 0.76-0.96; P=0.010 | AHR 0.85; 95% CI 0.75-0.96; P=0.007; NNT 23)
CV mortality, HF hospitalization, non-fatal MI, or non-fatal stroke
40.1% vs. 43.9% (UHR 0.87; 95% CI 0.77-0.98; P=0.020 | AHR 0.86; 95% CI 0.76-0.97; P=0.015; NNT 26)
CV mortality, HF hospitalization, non-fatal MI, non-fatal stroke, or coronary revascularization
42.9% vs. 46.9% (UHR 0.87; 95% CI 0.77-0.97; P=0.015 | AHR 0.87; 95% CI 0.77-0.98; P=0.018; NNT 25)
All-cause mortality
30% vs. 32% (UHR 0.89; 95% CI 0.77-1.02; P=0.086)
All-cause mortality or HF hospitalization
42% vs. 46% (UHR 0.87; 95% CI 0.78-0.98; P=0.021)
New DM
6% vs. 6% (UHR 0.98; 95% CI 0.70-1.35; P=0.88)

Additional Analyses

Open-label ARB use at the end of the trial
2.3% vs. 5.0%
Discontinuation of trial medication in survivors at the final study visit
25% vs. 18% (no statistical analysis given)
BP change at month 6
Systolic: 4.6 mmHg greater reduction in candesartan group (P=0.007)
Diastolic: 3.0 mmHg greater reduction in candesartan group (P=0.004)

Subgroup Analysis

For the primary outcome.

Beta blocker at baseline
Yes: Candesartan better
No: No difference
P for treatment interaction=0.14
Recommended dose of ACE-inhibitor

Defined as benazepril 20 mg, captopril 150 mg, enalapril 20 mg, lisinopril 20 mg, perindopril 4 mg, quinapril 20 mg, ramipril 10 mg, trandolapril 2 mg.

Yes: Candesartan better
No: No difference
P for treatment interaction=0.26

Adverse Events

Leading to discontinuation
Hypotension: 4.5% vs. 3.1% (P=0.079)
Creatinine increase: 7.8% vs. 4.1% (P=0.0001; NNH 27)
Hyperkalemia: 3.4% vs. 0.7% (P<0.0001; NNH 37)
Any adverse event or laboratory abnormality: 24.2% vs. 18.3% (P=0.0003; NNH 17)
2 vs. 3 events


  • Few non-White participants
  • Loss of significance for those on spironolactone[2]
  • Unclear which beta blocker was used[2]
  • Unclear effect on development of malignancies with ACE+ARB therapy[2]


  • AstraZeneca (the manufacturer of Atacand, the brand version of candesartan) who also managed the data and were involved in analysis and interpretation
  • Authors with multiple financial disclosures

Further Reading