Review of the Val-HeFT trial

Valsartan in chronic systolic heart failure

N Engl J Med 2001;345:1667-1675

Background: Angiotensin II is a peptide hormone that is part of the renin–angiotensin–aldosterone system (RAAS). Angiotensin II is a potent vasoconstrictor and growth-stimulating hormone. Data suggested that it plays a role in ventricular remodeling and progression of heart failure. Although treatment with angiotensin-converting enzyme inhibitors (ACEi) reduce angiotensin II levels, physiologically active levels of angiotensin II may persist despite long-term therapy.

The Valsartan Heart Failure Trial (Val-HeFT) sough to assess whether the angiotensin-receptor blocker valsartan, could reduce mortality and morbidity when added to optimal medical therapy in patients with systolic heart failure.

Patients: Eligible patients had left ventricular ejection fraction less than 40% and left ventricular dilation, in addition to having clinical heart failure for at least 3 months with NYHA class II, III or IV symptoms. Patient also had to have been receiving a fixed-dose drug regimen for at least two weeks, that could include ACEi, diuretics, digoxin, and beta-blockers.

There were many exclusion criteria. We mention some here: Postpartum cardiomyopathy, acute myocardial infarction within 3 months, coronary artery disease likely to require intervention, serum creatinine >2.5 mg/dL and life expectancy less than 5 years.

Baseline characteristics: Patients were recruited from 302 centers in 16 countries. The trial randomized 5,010 patients – 2,511 randomized to receive valsartan and 2,499 to receive placebo.

The average age of patients was 63 years and 80% were men. The average left ventricular ejection fraction was 27%. Cardiomyopathy was ischemic in 57% of the patients. The NYHA class was II in 62% of the patients, III in 36% of the patients and IV in 2%.

Approximately 26% had diabetes and 12% had atrial fibrillation.

At the time of enrollment, 86% were taking a diuretic, 67% were taking digoxin, 35% were taking beta-blockers, and 93% were taking ACEi.

Procedures: The trial was double-blinded. The trial had an initial run-in period for 2 - 4 weeks where patients received placebo twice daily. This was performed to confirm patients’ eligibility, clinical stability and compliance.

Patients were assigned in a 1:1 ratio to receive valsartan or placebo. Randomization was stratified according to whether or not they were receiving a beta-blocker.

Valsartan was started at a dose of 40 mg twice a day, and the dose was doubled every two weeks to the target dose of 160 mg twice a day. Placebo doses were adjusted in a similar way.

Follow up occurred at 2, 4, and 6 months and every 3 months thereafter.

Endpoints: The trial had two primary end points. The first was all-cause mortality. The second was the combined end point of mortality and morbidity, which was defined as cardiac arrest with resuscitation, hospitalization for heart failure, or administration of intravenous inotropic or vasodilator drugs for four hours or more without hospitalization.

The estimated sample size was 5,000 patients. The sample size calculation assumed 20% relative risk reduction in mortality with valsartan assuming 906 patients would die during the trial. This sample size would provide the trial 90% power at 0.02 alpha. Alpha was 0.02 instead of the traditional 0.05 since the trial had two primary endpoints and to adjust for the interim analyses.

Results: The target valsartan dose of 160 mg twice a day was achieved in 84% of the patients. The reduction in systolic blood pressure was greater with valsartan vs placebo – mean of 5.2 ± 15.8 mm with valsartan compared to 1.2 ± 14.8 mm Hg with placebo, at 4 months.

All-cause mortality was not different between both groups (19.7% with valsartan vs 19.4% with placebo, RR: 1.02, 95% CI: 0.88 – 1.18; p= 0.80). The second co-primary endpoint was reduced with valsartan (28.8% vs 32.1%, RR: 0.87, 95% CI: 0.77 – 0.97; p= 0.009). This was driven by reduction in hospitalizations for heart failure (13.8% vs 18.2%). Cardiac arrest with resuscitation was 0.6% with valsartan and 1.0% with placebo. All-cause hospitalization was numerically lower with valsartan, however, this was not statistically significance (2,856 vs 3,106; p= 0.14). The mean change in ejection fraction was higher with valsartan (4.0% vs 3.2%; p= 0.001). More patients had improvement in NYHA classification with valsartan (23.1% vs 20.7%; p<0.001).

Valsartan was associated with more dizziness (1.6% vs 0.4%), more hypotension (1.3% vs 0.8%), and more renal impairment (1.1% vs 0.2%).

There were significant subgroup interactions based on background medications used. Patients were divided into 4 groups based on their use of ACEi and beta-blockers (ACEi yes/no and beta-blockers yes/no). The interaction p value was 0.009 for all-cause mortality and 0.001 for the second primary endpoint. For patients who were receiving both ACEi and beta-blockers at baseline (32% of the patients), valsartan increased all-cause mortality (p= 0.009) and had a trend toward worsening the second primary endpoint (p= 0.10). Among patients who were not receiving ACEi at baseline (7% of the patients), valsartan reduced all-cause mortality (RR: 0.67) as well as the second primary endpoint (RR: 0.56).

Conclusion: In patients with systolic heart failure, adding valsartan to standard background therapy did not reduce mortality but was associated with a reduction in heart failure hospitalizations. Although all-cause hospitalizations were numerically lower with valsartan, the difference was not statistically significant.

It’s important to note that this trial evaluated valsartan vs placebo when added to an ACEi.

Based on this trial, valsartan is not an attractive option as an add-on therapy for patients with systolic heart failure, as the observed benefit was modest and outcomes were worse in patients who were also receiving both an ACEi and a beta-blocker - both reduce mortality in patients with systolic heart failure.