Spironolactone Revisited

As a renal fellow, I have found myself reaching for spironolactone more and more. It is very helpful when treating the patient with resistant hypertension, but it continues to be under-utilized. Aside from the renal-centric effects of aldosterone, there is evidence that mineralocorticoid receptor activation is responsible for pathogenic remodeling (both structural and electrical) after myocardial infarction. In fact, aldosterone is also produced by the heart, and cardiac expression of the mineralocorticoid receptor is increased in the setting of heart failure and MI. Perhaps aldosterone antagonists can reverse the cardiac remodeling, predisposition to arrhythmia, and risk of cardiovascular death we see in our kidney disease patients?

Remember, patients with ESRD do not die from kidney failure per se, but most die from cardiovascular and cerebrovascular disease. We know that spironolactone reduces mortality and hospitalization due to heart failure in patients with reduced LVEF (RALES trial), but we don’t know if these effects carry over to the ESRD population. However, two recent papers caught my eye.

One was recently published in JACC, where 309 patients with ESRD were randomized to spironolactone 25mg once daily or a control group. The primary outcome was a composite of death or hospitalization from cardiovascular and cerebrovascular events, and the secondary outcome was death from any cause. At three years, the treatment group had significantly fewer cardiovascular and cerebrovascular events: 9 patients (5.7%) vs 19 (12.5%) in the control group (Adjusted HR 0.379, p = 0.016). There were also fewer deaths: 10 (6.4%) with spironolactone vs 30 (19.7%) in control group (Adjusted HR 0.335, p=0.003). Out of the 157 patients in the spironolactone group who stopped the drug, 7 developed gynecomastia, 9 had breast pain, and 3 patients had significant hyperkalemia (>6.5 meq/L). Of course, this was a small RCT and not blinded or placebo controlled. If these effects are true, this would yield a NNT of 14 to prevent one cardiovascular or cerebrovascular event and a NNH of 52 to cause one episode of hyperkalemia. 

A similar study published in JASN randomized 158 patients on PD and already on ACE/ARB therapy to 25mg of spironolactone or a control group.  After two years of follow up, the rate of change of LVMI was significantly lower at 6, 18, and 24 months in the spironolactone group. In a subgroup analysis of males who started the trial with LVMI ≤ 50 g/m2, the control group had a significant increase in LVMI over time while the spironolactone treated group was unchanged. Likewise, in males who started with LVMI > 50 g/m2, those treated with spironolactone had a significant reduction in LVMI over time compared to the controls. In this trial, serious hyperkalemia (>6.0) occurred infrequently: 2 in spironolactone group and 1 in control. Gynecomastia was more frequent in the spironolactone group (14.1% vs 2.5%), but some of these patients had their doses reduced or were switched to eplerenone. 

So what does this all mean? It’s exciting to think that mineralocorticoid antagonists may reduce LVH and cardiovascular events in our ESRD population. A larger, placebo controlled trial is needed to confirm these findings since these were small open-label trials. Fortunately this is underway with the ALCHEMIST trial (NCT01848639), which plans to randomize 825 hemodialysis patients to either spironolactone or placebo. In the meantime, it may be best to use spironolactone as an anti-hypertensive agent in a dialysis patient who is low risk for hyperkalemia. For instance, PD patients tend to have have low serum potassium thanks to high clearance. In the PD patients I’ve seen with hypertension and hypokalemia, I have been backing off pure anti-hypertensive agents (like hydralazine) and starting 25mg of spironolactone. These trials suggest that this dose is safe to use in dialysis patients and possibly has some effect in preventing LVH and improving long-term cardiovascular outcomes.

Posted by John Roberts