Doubling Down on RAAS Blockade in HF? Aldosterone Antagonists, Not ARBs, Says Meta-analysis

December 26, 2012

An aldosterone antagonist (AA) should be the drug of choice when adding a second renin-angiotensin-aldosterone system (RAAS) blocker to ACE inhibitors in patients with symptomatic systolic HF despite standard drug therapy, say the researchers behind a 16-trial meta-analysis [1].

Alone among the RAAS blockers, the AAs spironolactone (Aldactone, Pfizer) or eplerenone (Inspra, Pfizer) as add-on therapy (compared with placebo) led to significant reductions in all-cause and CV mortality and HF hospitalization. The other RAAS blockers included angiotensin receptor blockers (ARBs) and the direct renin inhibitor aliskiren (Tekturna, Novartis).

The meta-analysis "allows us to reasonably conclude that the aldosterone antagonists are more effective than either ARBs or direct renin inhibitors," write the authors, led by Dr Sripal Bangalore (New York University School of Medicine, NY), in a report published December 12 in Congestive Heart Failure [1].

"From a safety perspective, however, any combination of dual RAAS blockade is associated with excess adverse events and these patients need to be closely monitored," they write. In the current analysis, add-on ARBs were associated with more hyperkalemia, renal failure, and hypotension vs placebo; aliskiren led to more hypotension, and AAs to more hyperkalemia.

Indeed, the safety outcomes are a key part of the analysis, Bangalore told heartwire . "When you reach for dual-RAAS blockade, regardless of what medication you add on, you're going to have side effects," he said. However, "we clearly did not find a [clinical] benefit if you put them an ARB or direct renin inhibitor along with the ACE inhibitor. But the harm was still there."

The meta-analysis closely follows an observational study of Medicare patients with systolic HF, in which AA therapy cut HF hospitalizations but didn't significantly affect all-cause mortality or CV hospitalizations, as covered by heartwire . The current data, Bangalore said, "needs to be put in that context. This is data from randomized trials showing significant benefit, but of course it is not restricted to elderly patients."

In clinical practice, he said, "a lot of physicians go by the CHARM [Candesartan in Heart Failure: Assessment of Reduction in Mortality and Morbidity] data and are putting ACE inhibitors and ARBs together to reduce the risk of hospitalization. So many still believe that there might be a benefit." But that benefit wasn't large and often went with increased potassium and creatinine, observed Bangalore, "so if anything they should be reaching out to an aldosterone antagonist." In the CHARM-Added trial, heartwire reported in 2004, addition of the ARB candesartan to ACE-inhibitors in systolic HF cut the hazard ratio for CV death or HF hospitalization by 15%.

The current analysis covers 16 trials with > 30 000 patients that included at least three months of follow-up. An AA, ARB or aliskiren were compared with placebo in systolic-HF patients who were also taking ACE inhibitors, among other standard HF meds.

Over a mean 16.2 months (range 3–41 months), add-on AAs were associated with > 20% drop in each of several clinical end points, which included all-cause and CV mortality, while there were no such significant reductions with ARBs or aliskiren.

Relative risk (RR) for clinical end points associated with AAs added to ACE inhibitors in systolic HF

End point RR (95% CrI)  
Mortality 0.79 (0.66–0.98)  
CV death 0.78 (0.65–0.93)  
HF hospitalization 0.74 (0.55–0.94)  
CV death or HF hospitalization 0.73 (0.55–0.90)  

CrI=credibility interval.

For AAs vs placebo, the relative risk (RR) for hypokalemia was 2.10; for ARBs vs placebo, the RRs for hyperkalemia, renal failure, and hypotension were 2.38, 2.26, and 1.63, respectively; and for aliskiren, the RR for hypotension was 1.98. All were significant; other safety comparisons were not.

Bangalore acknowledged that the 16 trials were completed between 1999 to 2011, and so did not include many patients with an implantable cardioverter defibrillator (ICD) or biventricular pacemaker for cardiac resynchronization therapy. Today, most patients like those in the trials would be recommended for device therapy, he observed.

"In this analysis there was a reduction in death and CV death, but you sometimes wonder if these patients had an ICD or a biventricular pacemaker--would the difference still be there?" One can only speculate, he said, "since we don't have the data."

Bangalore discloses serving on advisory boards for Daiichi Sankyo and Boehringer Ingelheim. Disclosures for the other authors are listed in the report.