Systolic Heart Failure Trials: Fun Look at NNT

/

With the twins off at the grandparents, John sleeping in, and Pete busy running around in the backyard, I had the very rare joy of sitting on my deck reading on my iPad. One of the articles I came across had a very interesting graph (shown below) comparing the NNT for the different systolic heart failure trials. Unfortunately, I could not remember what all the different trials and was especially curious about the MADIT-CRT trial. Thankfully, Journal Club (which I spoke about in a prior blog post), came to the rescue. Just for fun, here are the different trials with NNT. 

  • Carvedilol: I'm not sure which trial the graph is refering to, so I just pulled the major carvedilol trials assuming the table is pulling the subset from the US with EF <35% (for comparison's sake)
    • COPERNICUS: Carvedilol vs placebo for patients with Class III and IV HF with EF <25%. Incredibly, the NNT for 1 year mortality was 15!! Wow, place 15 people with an EF <25% (and NYHC of III or IV) on carvedilol and you will prevent one death. That is impressive!
    • Another trial looking at Carvedilol was the COMET Trial which compared carvedilol to metoprolol tartate (the short acting, twice a day, metoprolol). There were a couple of things that were a bit unfair about the trial (metoprolol succinate was the form shown to have benefit for HF mortality (MERIT-HF) and the dosing for metoprolol tartate was a bit lower), but carvedilol did beat metoprolol tartate when it comes to mortality. The take away on this is that carvedilol really should be our drug of choice, and if we use metoprolol, we must use metoprolol succinate. 
  • SOLVD-T: Study looked at enalapril vs placebo in patients with Class II-IV HF with an EF 35% or less. "The authors esitmate that treating 1,000 CHF patients with enalapril for approimately 3 years would prevent 350 hospitalizations and 50 premature deaths". 
  • AHEFT: This was the really cool study that showed isosobide dinitrate pulus hydralazine reduced mortality and hospitalization in Black patients with a NYHA of III-IV and EF less than 35% (or <45% if certain echo findings). NNT: 25
  • MADIT-CRT: All of the MADIT-CRT trials are looking at ICD's, either with synchronization or alone following an MI and poor EF. 
    • MADIT-CRT-II: ICD following an MI and EF 30% or less. NNT: 18. By the way, there was a follow-up study which showed ICD reduced death for EF 35% or below whether post-MI or not Post-MI, but the data was not as impressive.
    • MADIT-CRT: ICD with cardiac resynchronization for patients with wide QRS (>130 msec) and EF 35% or lower. For the composit outcome of non-fatal HF events and all cause mortality, the NNT was impressive (12). Clearly, taking the non-fatal HF events out of the picture, the NNT is far less impressive (see chart). There have been follow-up studies on this one. The take away for me: morbidity certainly seems improved in this population, but I am not as convinced on the cost-effectiveness as it relates to mortality. I shoudl state though, this is not an area of medicine I have spent nearly enough time on reading and really shouldn't have a strong opinion.
  • SHIFT: This was a new one for me, requiring a bit more research. The SHIFT trial looked at ivabradine, a seletive sinus node inhibitor, in patients with an EF 35% or less and a HR of 70 or over. The trial showed an overall reduction in the combined outcome of hospitalization or death, but according to UpToDate, the reduction was more prominent in the readmissions than death. 
  • EMPHASIS: Looked at eplerenone (mineralcorticoid receptor inhibitor similar to spironlactone but without all of the enodocrine side effects such as gynecomastia breast pain, sexual dysfunction) in patients with NYHA of II or worse and EF 35% or below. In the combined outcome of death or hospitalization for HF, the NNT was 19 and for death was 51. Pretty impressive!
  • PARADIGM-HF (Prospective Comparison of ARNI [Angiotensin Receptor–Neprilysin Inhibitor] with ACEI [Angiotensin-Converting–Enzyme Inhibitor): this is the study which I was reading this morning and started it all! This is a look at new drug class, angiotensin receptor-neprilysin inhibitor, in patients with EF 35% or worse and NYHA class II-IV. Very surprisingly, know what we know about ARBs vs ACE Inh, the novel agent was actually superior to the ACE Inh! Here is a paragraph from the accompany editorial that produced the below chart explaining the pathophysiology behind this new agent:

Why did LCZ696 succeed in improving outcomes so convincingly in this representative population of patients? Drugs that inhibit the renin–angiotensin–aldosterone system (RAAS) have been foundational to cardiovascular drug therapy for almost three decades. RAAS inhibitors moderate vasoconstriction, myocyte hypertrophy, and myocardial fibrosis, an effect that has translated into clinically meaningful improvements in functional status and survival. Natriuretic peptides, which include atrial natriuretic peptide, B-type natriuretic peptide, and urodilatin, are secreted by the heart, vasculature, kidney, and central nervous system in response to increased cardiac-wall stress and other stimuli. Natriuretic peptides have potent natriuretic and vasodilatory properties, inhibit the RAAS, reduce sympathetic drive, and have antiproliferative and antihypertrophic effects as well.6 Neprilysin inhibition results in an increased concentration of natriuretic peptides. Thus, the beneficial effects of RAAS inhibition are likely to be augmented by the enhancement of natriuretic peptide activity. 

Cool stuff! 

Please note: I do not know why the below chart does not seem to exactly match the NNTs I found...