Oct 11, 2021
Please join author Milton Packer and Associate Editor Justin Ezekowitz as they discuss the Perspective "Heart Failure and a Preserved Ejection Fraction: A Side-by-Side Examination of the PARAGON-HF and EMPEROR-Preserved Trials."
Dr. Carolyn Lam:
Welcome to Circulation on the Run, your weekly podcast summary and backstage pass to the Journal and its editors. We're your co-hosts. I'm Dr. Carolyn Lam, Associate Editor from the National Heart Center.
Dr. Greg Hundley:
And I'm Dr. Greg Hundley, associate editor, director of the Pauley Heart Center at VCU Health in Richmond, Virginia.
Dr. Carolyn Lam:
Greg, it really is so great to be back with you chatting about the papers here in the Journal. Thank you for going solo and for just being the greatest partner on earth. Thank you for that. For everyone listening in, we are back with some gusto and especially with this feature discussion today. You are not going to want to miss it. We are talking to Dr. Milton Packer as well as Dr. Justin Ezekowitz. We are going to compare PARAGON and EMPEROR-Preserved trials in heart failure with preserved ejection fraction. A really interesting discussion you're not going to want to miss, but now let's start with some papers in today's issue. I'd like to start, please.
Dr. Greg Hundley:
You bet.
Dr. Carolyn Lam:
Greg, you know the optimal duration of antiplatelet therapy in patients with high bleeding risk with or without oral anticoagulation after coronary stenting? Well, that still remains a question. Today's paper is a pre-specified subgroup analysis of the MASTER DAPT trial and reports on the outcomes of patients with or without an oral anticoagulation indication in this study.
Dr. Greg Hundley:
Right, Carolyn. Remind us. What was the MASTER DAPT trial? What did it test?
Dr. Carolyn Lam:
Ah. MASTER DAPT investigated an abbreviated or one-month versus a non-abbreviated or three to 12-month dual antiplatelet therapy and a stopping of antiplatelet therapy at six months strategy after coronary stenting in an all-comer population at high bleeding risk.
Dr. Greg Hundley:
Carolyn, what did this subgroup analysis of outcomes in patients with and without oral anticoagulation show?
Dr. Carolyn Lam:
At 12 months of follow-up, ischemic and net risk did not differ with abbreviated versus non-abbreviated anti-platelet regimens in both subgroups, although significantly fewer clinically relevant bleeding events occurred in the group without an oral anticoagulation indication. Whereas only numerically fewer bleeding events occurred in the group with an oral anticoagulation indication that did not reach statistical significance. This subgroup analysis from the MASTER DAPT trial really adds additional evidence that dual antiplatelet therapy beyond one month in patients with or without an indication for oral anticoagulation really has no benefit and only increases bleeding risk.
Dr. Greg Hundley:
Oh, very important finding, Carolyn. Great research. Well, Carolyn, how the extracellular matrix microenvironment modulates the contractile phenotype of vascular smooth muscle cells and confers vascular homeostasis really remains elusive. Thus, these investigators led by Professor Wei Kong at Peking University applied protein-protein interaction network analysis to explore novel extracellular matrix proteins associated with the vascular smooth muscle cell phenotype.
Dr. Carolyn Lam:
Huh. Interesting. What did they find, Greg?
Dr. Greg Hundley:
Right, Carolyn. By combining an in-vitro and an in-vivo genetic mice vascular injury model, they identified nidogen-2, a basement membrane glycoprotein, as a key extracellular matrix protein for maintenance of vascular smooth muscle cell identity. Nidogen-2 exerted its protective function via direct interaction and modulation of Jagged1-Notch3 signaling.
Dr. Carolyn Lam:
Wow! Nidogen-2 and Jagged1-Notch3. I always learn so much. What are the clinical implications, Greg?
Dr. Greg Hundley:
Right, Carolyn. Perhaps targeting nidogen-2 to precisely modulate Jagged1-Notch3 signaling, well, that may provide novel therapeutic strategy for atherosclerosis and post-injury restenosis.
Dr. Carolyn Lam:
Very nice. Well, in the next paper, we discuss inflammation in heart failure. We know that inflammation contributes to the pathogenesis of heart failure, but there is limited understanding of inflammation's potential benefits. Interesting, huh? Well, these authors, Dr. Wollert and colleagues from Hannover Medical School in Germany, identified an adaptive crosstalk between inflammatory cells and cardiomyocytes that protects against persistent afterload stress-induced heart failure in mice. Monocytes and macrophages produced myeloid-derived growth factor in the pressure overloaded myocardium to augment SERCA2a expression in cardiomyocyte's calcium cycling and contractility. Myeloid-derived growth factor plasma concentrations were also found to be elevated in patients with aortic stenosis and to decline after aortic valve implantation indicating that pressure overload also triggers myeloid-derived growth factor release in humans.
Dr. Greg Hundley:
Carolyn, really informative preclinical science, but what are the clinical implications?
Dr. Carolyn Lam:
Ah. These observations molecularly defined a feature of the inflammatory response to hemodynamic overload that protects against heart failure development. Inflammation's beneficial trade therefore need to be considered when developing inflammation as a therapeutic target in heart failure. All of this is really discussed in a lovely editorial entitled Inflammation and Heart Failure: Friend or Foe? That's by Drs. Hajjar and Leopold.
Dr. Greg Hundley:
Great job, Carolyn. Well, my next paper focuses on resistant hypertension. Carolyn, although lifestyle modifications generally are effective in lowering blood pressure among patients with unmedicated hypertension or those treated with one to two antihypertensive agents, the value of exercise and diet for lowering blood pressure in patients with resistant hypertension is unknown. To address this, Professor James Blumenthal and co-authors at Duke University Medical Center enrolled 140 patients with resistant hypertension with an average age of 63 years, 48% women, 59% black, 31% diabetes, and 21% with chronic kidney disease and randomly assigned them to A, a four-month cardiac rehab center-based program of lifestyle modification. We're going to call that C-LIFE, consisting of dietary counseling, behavior and weight management, and exercise. Or number 2 or the B, a single counseling session providing standardized education and physician advice. We'll call that SEPA.
Dr. Greg Hundley:
The primary endpoint was clinic measured systolic blood pressure. Secondary endpoints included 24-hour ambulatory blood pressure and selective cardiovascular disease biomarkers including baroreflex sensitivity to quantify the influence of baroreflex on heart rate; high-frequency heart rate variability to assess vagally-mediated modulation of heart rate; flow-mediated dilation to evaluate endothelial function; and pulse wave velocity to assess arterial stiffness; and then finally left ventricular mass to characterize left ventricular structure and remodeling.
Dr. Carolyn Lam:
Wow! That is a very, first of all, important clinical question. Then also, just very intricate methodology in assessing this. What did they find?
Dr. Greg Hundley:
Right, Carolyn. Between-group comparisons revealed that the reduction in clinic systolic blood pressure was greater in C-LIFE compared with SEPA. Next, 24-hour ambulatory systolic blood pressure also was reduced in C-LIFE with no change in SEPA. Then next, compared with SEPA, C-LIFE resulted in greater improvements in baroreflex sensitivity, high-frequency heart rate variability, and flow-mediated dilation. There was no between-group differences in pulse wave velocity or LV mass.
Dr. Greg Hundley:
Carolyn, diet and exercise can lower blood pressure in patients with resistant hypertension. When delivered in a cardiac rehabilitation setting, a four-month program of diet and exercise as adjunctive therapy, results in a significant reduction in clinic and ambulatory blood pressure, and improvement in selected cardiovascular disease biomarkers.
Dr. Carolyn Lam:
Wow! Really nice, Greg. Okay. Well, looks like we're all going to round up already with what else there is in today's issue. Let me start. There's an exchange of letters between Drs. Fang and Vinceti regarding the article Blood Pressure Effects on Sodium Reduction: Dose-Response Meta-analysis of Experimental Studies.
Dr. Greg Hundley:
Right, Carolyn. I've got a few things in the mail bag. First, Professor Anker has a Research Letter regarding the Kidney Function After Initiation and Discontinuation of Empagliflozin in Heart Failure Patients With and Without Type 2 Diabetes: Insights From the EMPERIAL Trials. Dr. Gerstenfeld has an ECG challenge entitled Atrioventricular Block with Narrow and Wide QRS: The Pause That Refreshes. Then lastly, Dr. Donald Lloyd-Jones has an AHA update regarding the American Heart Association's focus on primordial prevention. Well, Carolyn, I can't wait to hear this fantastic feature discussion with you and Dr. Packer. How about we jump to that?
Dr. Carolyn Lam:
Great. Let's go, Greg.
Dr. Carolyn Lam:
Because side-by-side exam of PARAGON and EMPEROR is like side-by-side of...
Dr. Justin Ezekowitz:
You can compare our new and our old prime minister much like your paper did.
Dr. Milton Packer:
Yeah, yeah.
Dr. Justin Ezekowitz:
There are [crosstalk] and it could be viewed until they perform in the broader world how it goes. You don't quite know.
Dr. Milton Packer:
The only problem is you can't do a head-to-head comparison of the old prime minister and the new prime minister.
Dr. Justin Ezekowitz:
That is true except that the head-to-head comparison includes excellent care by both the new and the old. I think that comparison's going to be pretty equal. I think we can case-control that one.
Dr. Carolyn Lam:
I really liked that that was politically correct because we are recording. Everybody, welcome to the feature discussion. I am here with Dr. Milton Packer from Baylor and he really needs no introduction. We're discussing heart failure with preserved ejection fraction. As well as our associate editor, Dr. Justin Ezekowitz from University of Alberta. Hence, in case anybody's wondering, we were talking about the Canadian elections. Let's just launch straight into it, a side-by-side comparison of PARAGON and EMPEROR-Preserved. Dr. Packer... Milton, if I may, what in the world drove you to do this?
Dr. Milton Packer:
My God. Oh, my God. Yes.
Dr. Carolyn Lam:
Tell us about what drove you to do this and please, if you could just summarize the results.
Dr. Milton Packer:
Well, let me just say from the outset that this was a commentary, not an original research article.
Dr. Carolyn Lam:
Yes.
Dr. Milton Packer:
The commentary was motivated by two very straightforward observations. We had two large scale outcome trials of two different drugs in heart failure with a preserved ejection fraction. I was privileged to serve as you were, Carolyn, on the leadership committees of both trials. It's not as if we have involvement in only one trial. We have involvement in both trials and we are very proud of that involvement.
Dr. Milton Packer:
One trial came in with a effect size of about 13% on its primary endpoint with a borderline P-value. A second trial, EMPEROR-Preserved, came in with a 21% reduction and its primary endpoint with a really small and persuasive P-value. The two patient populations in the two trials were really amazingly similar. We wanted to understand why it was 21% in one trial and persuasively so and why it appeared to be smaller in the PARAGON trial with sacubitril/valsartan. We thought, well, maybe that difference was related to how endpoints were defined or maybe that difference was related to the influence of ejection fraction. The reason we got excited about that was that as almost everyone knows, PARAGON found an influence of ejection fraction on the effect of sacubitril/valsartan in patients with HFpEF. We found an influence of ejection fraction on the effect of empagliflozin in HFpEF in EMPEROR-Preserved. We wanted to understand whether that influence was similar in the two trials.
Dr. Milton Packer:
Just to make life simple, PARAGON had created certain cut points for ejection fraction. They had presented and previously published in Circulation endpoints based on those cut points of ejection fraction. All we did was we used their endpoints and their cut points, and we put the two trials side by side. We did not do a statistical comparison of the effect size. There're actually no P-values in the whole commentary. But what we wanted to see was: Was the shape of the ejection fraction influence relationship similar or different in the two trials? Well, very simple. In PARAGON, as has been reported, there was a linear relationship: as ejection fraction increased, the effect of sacubitril/valsartan got smaller. In EMPEROR-Preserved, there was also an attenuation at a highest ejection fraction, but the relationship wasn't linear. It was like a hockey stick. It was flat and then went up at an ejection fraction over 62.5, which was the cut point that PARAGON used.
Dr. Milton Packer:
When we compared patients between the low 40s and the low 60s, the effect size in empagliflozin appeared to be larger than the effect size of sacubitril/valsartan in that ejection fraction group using the same endpoints. In fact, for hospitalizations for heart failure, which is really what SGLT2 inhibitors do, it was twice as great with empagliflozin in EMPEROR-Preserved than with sacubitril/valsartan in PARAGON-HF. We thought this was really interesting. We put the pictures up side by side. We wrote a commentary and Circulation was so kind to accept it.
Dr. Carolyn Lam:
Oh, but Milton, you were very, honestly as always, very clever to have done this analysis. But if I could reiterate a few things for the audience, which is very important. First of all, as you rightly first pointed out, it's a perspective piece. It is not a head-on comparison with P-values. It could not be. Let's just also give the audience a bit of background in that PARAGON included patients with an ejection fraction of 45% and above. EMPEROR-Preserved was above 40. PARAGON looked at total heart failure hospitalizations and cardiovascular death as a primary outcome. EMPEROR looked at first cardiovascular death or heart failure hospitalization.
Dr. Carolyn Lam:
Let's just remember the designs were different. Of course in the comparison, PARAGON compared sacubitril/valsartan versus valsartan. I like the way you very carefully wrote in your study that it was more a study of neprilysin inhibition since it's sacubitril/valsartan against valsartan and it was empagliflozin versus placebo. We know that it's important to state that as a basis. Then really important to say to everybody out there, pick up our journal. You must look at this bigger. I myself have already cited it at least twice already, Milton, because people will just naturally ask that. "Are the results different because of ejection fraction or different endpoints?" What you did there in that beautiful figure is that you tried as best as you can to match it up in terms of ejection fraction bins and match it up in terms of hospitalizations. There. I just wanted to state those few things, but I'm really-
Dr. Milton Packer:
Oh, no. No. Carolyn, you're 100% right. That's why there are couple of things. I just want to underscore what you said because I think your points were spot on. First of all, we really lined up the endpoints and the ejection fraction. We tried our best to compare apples and apples. It would not have been a useful exercise for us to compare different endpoints and different ejection fraction subgroups. But I just want to make sure that everyone understands: I'm a big fan of sacubitril/valsartan and I'm a big fan of neprilysin inhibition. As you know, both PARADIGM-HF and PARAGON-HF weren't really tests of sacubitril/valsartan; they were tests of neprilysin inhibition. They were great tests at that. PARAGON in particular was a great test of that. We're comparing neprilysin inhibition and SGLT2 inhibition.
Dr. Milton Packer:
But here's my most important point: we do not want people to choose one over the other. That was not the intent. We think that there are data in patients with certain ejection fractions, let's say between 40 and 60, I'm just creating a range, where both interventions are appropriate. Now I understand there are cost considerations and I don't want to minimize that, but we are not suggesting that anyone prefer one drug over the other. All we wanted to do was we wanted to ask the question: Since the effect size in one trial seemed to be different than the effect size in the other trial, what were the ejection fraction subgroups that represented that difference? We found that the patients with ejection fractions greater than 60, 65% did not contribute to that difference. It was the patients with lower ejection fractions that contributed to the difference. I hope that's helpful.
Dr. Carolyn Lam:
Ah. That's wonderful. Justin, have you recovered from the talk about the Canadian elections?
Dr. Justin Ezekowitz:
Oh. I have indeed.
Dr. Carolyn Lam:
I'm on swinging.
Dr. Justin Ezekowitz:
I have indeed. Thanks for recognizing that Canada just had a major election we carried out in six weeks. But, Milton, I really enjoyed reading this. Maybe I can just ask you about two elements within this perspective piece, which is number 1, what's incredibly concordant is a lack of difference across cardiovascular death for both agents in both trials regardless of the trial differences and the potential differences in patient populations recruited; that's number 1. It's incredibly flat for cardiovascular death.
Dr. Justin Ezekowitz:
But number 2 is there is a danger in comparing trials even non-statistically. That's often a pitfall we get into, but we have to put some frame of reference on that. What is the one or two key things you think differ between PARAGON and EMPEROR-Preserved that you say, "You really need to look at these trials differently"? Those two questions came to mind when looking at this great figure that you produced.
Dr. Milton Packer:
Okay. The first question is so much easier and that is that these drugs don't reduce cardiovascular deaths. Full stop. It's really interesting because sacubitril/valsartan reduces cardiovascular death in people with ejection fractions of 40% or less, but not in patients with ejection fraction greater than that. The primary effect is heart failure hospitalizations. Empagliflozin didn't reduce cardiovascular death even in patients with the ejection fraction less than 40% or greater than 40%. What we're really, really talking about two drugs where the major effect is a reduction in heart failure hospitalizations. That comes out whether you do the analysis as time-to-first-event or total heart failure hospitalizations.
Dr. Milton Packer:
Of course, we're looking forward to the DELIVER trial with dapagliflozin. My own personal expectation is they're going to come out with a very striking effect on heart failure hospitalizations and not on cardiovascular deaths. Cardiovascular deaths in patients with HFpEF is really... It's a hard goal because only half of the deaths are cardiovascular. These patients have so many comorbidities that influence prognosis. The other thing, which is really important, is that heart failure hospitalizations only represented 18% of all hospitalizations in these patients; it's really small. I think of empagliflozin as being a treatment of the heart failure of HFpEF, not a treatment for HFpEF. I hope that makes sense. Justin, what was your second question?
Dr. Justin Ezekowitz:
Absolutely.
Dr. Milton Packer:
Oh, the differences between-
Dr. Justin Ezekowitz:
Yeah. Thank you, Milton.
Dr. Milton Packer:
Okay. There's always differences between two trials. As I said before, Carol and I were involved in both trials. They were done slightly at different times. They didn't overlap. Remember that the cut points in the two trials, one was 40%, one was 45%, really didn't matter to our analysis because we corrected for that in our ejection fraction subgroups. I was actually really much more impressed by the similarities than by the differences, but here's the catch. HFpEF is an incredibly heterogeneous disease. When we look at baseline characteristics, we're looking at means, medians, percentages. We're not picking up on any heterogeneity and there's a lot of heterogeneity. I actually think that HFrEF is a reasonably homogeneous disease. I think HFpEF is an incredibly diverse disease with a whole host of different disorders. What I'm amazed by is that we actually got an effect size that was greater than 20% in an all-comers HFpEF analysis.
Dr. Milton Packer:
But in all honesty, Justin, it wasn't really all-comers. We excluded people with BMIs over 45. There are a lot of patients who are obese and had BMIs greater than 45 who have HFpEF. By the way, especially in Texas. I didn't say that. We didn't enroll those patients. In all honesty, if I had to do it all over again, I would have. By the way, PARAGON didn't enroll them either.
Dr. Carolyn Lam:
Well, this is an incredible conversation. I know that we could just do a whole hour of chatting about what this implies for the higher ejection fraction, what this implies for how we should be treating heart failure. I don't even dare to ask for some last words maybe from both Justin and Milton, but recognizing that the time is short, anything else to add?
Dr. Milton Packer:
I think Justin should do last words.
Dr. Justin Ezekowitz:
Well, let me summarize by saying there is a hockey stick. We love hockey sticks in Canada. A simple and an excellent comparison. I think people should really look at that figure to understand it, but do not undertreat your patients with HFpEF and look at these with a grain of salt. Thanks for joining us, Milton. Thanks, Carolyn.
Dr. Milton Packer:
Thank you so much.
Dr. Carolyn Lam:
On behalf of Greg and I, you've been listening to Circulation on the Run. Thank you so much for joining us today and don't forget to tune in again next week.
Dr. Greg Hundley:
This program is copyright of the American Heart Association 2021. The opinions expressed by speakers in this podcast are their own and not necessarily those of the editors, or of the American Heart Association. For more, visit ahajournals.org.