Circulation on the Run

Each monthly episode will discuss recent publications in the fields of genomics and precision medicine of cardiovascular disease.
RSS Feed Subscribe in Apple Podcasts
Circulation on the Run





All Episodes
Now displaying: November, 2017
Nov 27, 2017

Dr. Carolyn Lam:               Welcome to Circulation on the Run, your weekly podcast summary and backstage pass to the journal and its editors. I'm Dr. Carolyn Lam, Associate Editor from the National Heart Center and Duke National University of Singapore.

                                                Our journal this week features novel data informing the choice between conscious sedation and general anesthesia for transcatheter aortic valve replacement. A very relevant discussion for those of us who see these patients. Stay tuned, that's coming right up after these summaries.

                                                Subclinical hyperthyroidism is known to be associated with an increased risk of atrial fibrillation, but the association with thyroid function in the normal range or subclinical hypothyroidism is unclear. That is, until today's study, which shows us that variation in thyroid function within the normal range is associated with atrial fibrillation.

                                                First author, Dr. Baumgartner, corresponding author, Dr. Rodondi and colleagues from University of Bern in Switzerland, conducted a systematic review and obtained individual participant data in more than 30,000 participants from 11 prospective cohort studies that measured thyroid function at baseline and assessed incident atrial fibrillation, which occurred in 8.6% of individuals.

                                                They found that in youth thyroid individuals, there was a significant increase in the risk of atrial fibrillation with increasing free T4 levels within the reference range. Risks did not differ significantly by age and sex.

                                                Conversely, there was no association between TSH levels within the reference range, or subclinical hypothyroidism and the risk of atrial fibrillation. Thus, free thyroxin levels might add to further assessment of atrial fibrillation risks. Further studies are needed to investigate whether these findings apply to thyroxine treated patients.

                                                The next study provides insight into how exercise promotes metabolic remodeling in the heart. First author, Dr. Gibb, corresponding author, Dr. Hill and colleagues from University of Louisville, use radiometric, immunologic, metabolomic and biochemical assays to measure changes in myocardial glucose metabolism in mice subjected to acute and chronic treadmill exercise.

                                                They found that in the heart, glucose utilization via glycolysis was reduced during exercise and in the early recovery period after exercise. Low rates of myocardial glycolysis were sufficient to activate gene programs that instigate physiologic cardiac growth. Metabolic inflexibility of the heart, such as occurs in heart failure and diabetes, was sufficient to diminish mitochondrial function.

                                                Phosphofructokinase mediated changes in metabolism appeared to regulate genes involved in processes critical for metabolic remodeling, transcription, cell division, differentiation, cell proliferation and contraction. Thus, this study provides important preclinical evidence, showing how exercise-induced changes in glucose metabolism may promote physiologic cardiac growth.

                                                The next study addresses the question of whether antiarrhythmic drugs are safe and effective when non-shockable rhythms evolved to shockable rhythms during resuscitation for out of hospital cardiac arrests. In this study from first and corresponding author, Dr. Kudenchuk of University of Washington and his colleagues, patients who initially presented with non-shockable out of hospital cardiac arrests were randomized upon subsequently developing shock refractory VF or VT to receive amiodarone, lidocaine or placebo by paramedics.

                                                The primary outcome was survival to hospital discharge, with secondary outcomes, including discharge functional status and adverse drug-related effects. The authors found that outcome from non-shockable turned shockable out of hospital cardiac arrest was poor, but not invariably fatal. Though not statistically significant, point estimates for survival showed a trend to greater survival after amiodarone or lidocaine than placebo without increased risk of adverse effects or disability. Together, these findings may signal a clinical benefit that invites further investigation.

                                                The final study provides experimental data supporting the importance of a novel Cardiokine governing the local environment in infarcted hearts and determining the fate of implanted cells. This novel Cardiokine is C1q/tumor necrosis factor-related protein-9, or CTRP9, which is a novel pro survival Cardiokine that is significantly down regulated after myocardial infarction.

                                                In today's study by co-first authors, Drs. Yan and Guo and co-corresponding authors Drs. Ma and Wang from Thomas Jefferson University in Philadelphia, mice were subjected to myocardial infarction and treated with adipose-derived mesenchymal stem cells, CTRP9 or their combination. The authors found that administration of adipose-derived mesenchymal stem cells alone failed to exert significant cardio protection.

                                                However, administration of these cells in addition to CTRP9 further enhanced the cardioprotective effect of CTRP9, suggesting a synergistic effect. CTRP9 promoted adipose-derived mesenchymal stem cell proliferation, survival, migration and attenuated cardio myocyte cell death by signaling mechanisms that included binding with N-cadherin, activation of ERK, MMP9, and ERK-Nrf2 signaling and upregulation or secretion of antioxidative proteins.

                                                In summary, these results suggest that CTRP9 is a Cardiokine critical in maintaining a healthy microenvironment facilitating stem cell engraftment in infarcted myocardial tissue. Well, that wraps it up for your summaries, now for our feature discussion.

                                                Conscious sedation is very frequently used during transcatheter aortic valve replacement, or TAVR, but with limited evidence as to the safety and efficacy of this practice. Well, that is until this week's journal and this feature paper. We're so lucky to have with us the corresponding author, Dr. Jay Giri from Hospital of University of Pennsylvania, to discuss his novel findings, as well as Dr. Dharam Kumbhani, Associate Editor from UT Southwestern.

                                                Jay, tell us your study findings and how this really helps us to characterize anesthesia choice and clinical outcomes of at least U.S. patients undergoing TAVR.

Dr. Jay Giri :                        We looked at 11,000 patients treated over a 15-month period in 2014 and 2015 with percutaneous transfemoral TAVR. Notably, this was a time period that was identified as the start of the era of conscious sedation for TAVR in the United States.

                                                Also, this five quarter period that we looked at represented a time of relative technological stability where only two valve types, the Sapien XT and original Medtronic CoreValve were being used in America.

                                                Looking at that 15-month period when conscious sedation was first being used in TAVR, we elected to compare those patients to a propensity matched group of patients who underwent TAVR by, what at that time was, the more traditional approach of general anesthesia.

                                                Our primary outcome within hospital mortality, because we had complete followup for this outcome. We also looked at 30-day outcomes for which we had about 90% followup. What we discovered was actually an associated reduction in mortality, an absolute reduction of about 1% in the patients who were treated with conscious sedation.

                                                We also noted that they had modest decreases in the hospital length of stay, as well as significant decreases in the rates of ICU length of stay and the rates of pressor or inotrope use during the procedure. Obviously, the most provocative of the findings was the fact that we seemed to discover, after propensity matching a slight improvement in in-hospital mortality that held true at 30 days, as well.

Dr. Carolyn Lam:               Thank you, Jay. What important findings ... I mean, mainly because, we really didn't have much data, did we? About conscious sedation and TAVR before this. Now, it's observational data, and I suppose the question always becomes what about bias by indication? More well patients get selected for conscious sedation versus general anesthesia, perhaps? Or even the other way around. Could you elaborate a little bit on how you think that may have impacted results and the measures you took to look at that?

Dr. Jay Giri :                        I think it was something that we were highly aware of and I also have to give credit to Dr. Kumbhani and the editorial staff at circulation for pushing us on that issue of selection bias for the two procedures. The obvious concern here, when you saw that there was a potential mortality reduction with conscious sedation patients, was that perhaps the conscious sedation patients actually represented a healthier cohort to start with, or they were perhaps treated at centers that were more highly experienced and by operators that were more highly experienced with TAVR in general.

                                                We tried to account for this in a number of different fashions. The first, as we mentioned, was with an inverse probability treatment weighted analysis that accounted for 51 co-variants that were balanced between the groups. Additionally, we did adjust for site characteristics and utilized a hierarchical method technique to take into account both the experience of sites and operators.

                                                Finally and most importantly, we performed what's called a falsification end point analysis in a postdoc fashion to verify that it looked like other outcomes outside of things, like mortality, length of stay, things we would expect to be influenced by sedation type, ended up being equal between the two groups. Falsification end point analysis represents, essentially, a negative control. You're supposed to theorize for potential outcomes that you would think would not be influenced by your intervention. In this case, those outcomes we theorized were vascular complications, major bleeding and pacemaker implantation, which we theorized would not be influenced by sedation type. In fact, we discovered that those outcomes were similar after adjustment, even though they had some differences before adjustment.

Dr. Dharam Kumbhani:  Jay, I want to congratulate you and your team on this paper. You guys really picked a very important topic to look at and then you jump ... as you outlined, you jumped through a lot of statistical hoops and try to really provide evidence for a field in which a randomized controlled trial is probably going to be just logistically probably hard to conduct, just given the sample size requirements, which also you've provided in your discussion.

                                                I think all the metrics that you looked at as far as utilization of therapies and length of stay, things like that, I think many people believe that and you were the first one to systematically evaluate and show that.

                                                As you alluded to, I think that mortality, and Carolyn mentioned that, as well. I think the mortality findings are very interesting. Again, it's always hard when you have observational data to really put a lot of stock into that and you guys, as you outline, looked at so many different ways of doing that.

                                                Again, I guess, observational data are always inherently going to have that limitation, no matter what statistical rigor we put them through. They were definitely very thought-provoking and, as you outlined, it's definitely come at the right time as the field is exploding and more and more centers are getting facile at it.

                                                The other thing that you mentioned, but which I want to make sure that people fully understand is that you also provided a very elegant analysis looking at site volumes, because traditionally the sites that are doing conscious sedation have done a number of TAVR's before and there is a very clear cumulative volume outcomes association, for TAVR.

                                                By accounting for the totality of experience, so you adjusted for the cumulative volume that sites have been doing this, so these are not just the high volume, high throughput centers, which have a lot of experience doing 150, 200 TAVR's a year, that thereby have really good outcomes by virtue of being expert, both as operators and as sites, but rather potentially something that is related to conscious sedation aspect itself. You guys really stepped up and provided a very elegant analysis to try to dissociate the two issues here.

Dr. Carolyn Lam:               Dharam, and you just provided a very elegant explanation of the thought processes that were going on with our editors about this paper. I join you in congratulating Jay. Just a question. This is the best available evidence now, what are we going to do about it? I mean, Dharam, you're an inventionist, what now?

Dr. Dharam Kumbhani:  The issues were not so much related to efficacy, initially. The initial concerns were related to safety, and Jay's paper clearly addresses that. Then, in addition to that, it says, "Well, it's not just that it's a safe procedure, but it's also effective with potential patient level and hospital level benefits from having a robust conscious sedation program."

                                                I guess the one question that I have about conscious sedation and, Jay, I would love to hear your thoughts on this, as well, is it is possible but it is usually not done, TEE's or transesophageal echos are typically not done when you're doing conscious sedation. It is possible, as I said. As you move towards lower risk patients, on the one hand, these would be ideal patients for conscious sedation because then it's almost like a day procedure, in some ways for them.

                                                But on the other hand, the fidelity of being able to look for even small paravalvular leaks, things like that, may be harder with a transthoracic echo. I don't know, as we expand towards the oldest populations, whether we'll see a greater adoption of conscious sedation, or whether there'll be some scaling back.

Dr. Jay Giri :                        Two points on that. The first is, I totally agree that it's relatively unusual for a transesophageal echo to be performed in the setting of conscious sedation. There's no question, secondly that transesophageal echo allows for the most rigorous evaluation of paravalvular leaks.

                                                It is striking, though, that the rates of paravalvular leaks, due to technological improvements to the valves, are significantly improving. Even since the time of our study two years ago, a new generation of valves is consistently coming out with leak rates in pretty well-conducted analyses that are in the low, single digit percentages for moderate leak or more.

                                                Part of I think the move towards conscious sedation, even initially and especially as we go forward, is predicated on the fact of continuing technological improvements that essentially almost solve the leak problem.

                                                I think it's true that there's always going to be a very small minority of patients that are stuck with concerns about paravalvular leak at the end of their TAVR procedure. For those who have moderate or greater leak, I think that the threshold for escalating care, even to intubation and TEE to evaluate that leak, I think should be relatively low in a lower risk population.

                                                However, I think the point that you bring up about the potential harm of trace or trivial leaks, or mild leaks, which may not be perfectly interpreted with transthoracic echo and aortograms and [inaudible 00:16:41] assessments at the time of the valve placement. It's something we're going to have to keep a close eye on.

                                                From a practical standpoint, I believe this train has left the station. Totally unscientific, but around the time they released the paper online. I just shot out a poll on Twitter and got about a couple of hundred responses from folks, what they're doing now.

                                                Now, Twitter certainly, probably doesn't represent the average transcatheter valve operator in the world, but I was surprised to see that over 70% of the respondents favored a conscious sedation approach at this point in time, which obviously is much higher than what we saw in our paper from two years ago.

Dr. Carolyn Lam:               Well, audience, I'm sure you enjoyed that. Thank you for joining us today. Don't forget to tune in again next week.

Nov 20, 2017

Dr. Carolyn Lam:               Welcome to Circulation on the Run. Your weekly podcast summary and backstage pass to the journal and its editors. I'm Dr. Carolyn Lam, Associate Editor from the National Heart Center and Duke National University of Singapore. This week's journal features novel results from the NCDR IMPACT Registry that informs us on risk prediction in patients with congenital heart disease undergoing cardiac catheterization. We'll be taking a deep dive into this right after these summaries.

                                                The first original paper provides pre-clinical data showing that delayed repolarization may underlie ventricular arrhythmias in heart failure with preserved ejection fraction or HFpEF. First author Dr. Cho, co-corresponding authors Dr. Marban, and Cingolani from Cedars-Sinai Heart Institute and their colleagues, induced HFpEF in Dahl salt-sensitive rats by feeding them a high-salt diet from seven weeks of age. They showed that susceptibility to ventricular arrhythmias was markedly increased in rats with HFpEF.

                                                Underlying abnormalities included QTc prolongation, delayed repolarization from down-regulation of potassium currents, and multiple re-entry circuits during ventricular arrhythmias. These findings are consistent with the hypothesis that potassium current down-regulation may lead to abnormal repolarization in HFpEF, which in turn predisposes to ventricular arrhythmias and sudden cardiac death.

                                                The next paper shows that genetic testing can help to identify patients with pulmonary veno-occlusive disease who were misclassified as pulmonary arterial hypertension. Now, heterozygous mutations in the gene encoding the bone morphogenetic protein receptor type II or BMPR2 are the commonest genetic cause of pulmonary arterial hypertension. Whereas biallelic mutations in the eukaryotic translation initiation factor 2 alpha kinase 4 gene or EIF2AK4 gene are described in pulmonary veno-occlusive disease and pulmonary capillary hemangiomatosis.

                                                In the current study, first author Dr. Hadinnapola, corresponding author Dr. Morrell, and colleagues from University of Cambridge performed whole genome sequencing on the DNA from 864 patients with pulmonary arterial hypertension, as well as 16 patients with pulmonary veno-occlusive disease all recruited to the NIHR BioResource – Rare Diseases study. They found that 1% of patients with a clinical diagnosis of pulmonary arterial hypertension actually carry the biallelic EIF2AK4 mutations. Patients who are diagnosed clinically with pulmonary arterial hypertension, but who had a transfer coefficient for carbon monoxide of less than 50% predicted and an age of diagnosis of less than 50 years were much more likely to carry these biallelic EIF2AK4 mutation. In fact, the diagnostic yield for genetic testing in this group was 53%.

                                                Radiological assessment alone was unable to distinguish reliably between these patients and those with idiopathic pulmonary arterial hypertension. Importantly, these patients with biallelic EIF2AK4 mutations had a worst prognosis compared to other patients with pulmonary arterial hypertension. Thus in summary, younger patients diagnosed with idiopathic pulmonary arterial hypertension but with a low transfer coefficient for carbon monoxide, have a high frequency of biallelic EIF2AK4 mutations and should be reclassified as pulmonary veno-occlusive disease or pulmonary capillary hemangiomatosis. They have a poor prognosis and genetic testing can therefore identify these misclassified patients allowing appropriate management and early referral for lung transplantation.

                                                The next study identifies a novel molecular target for the treatment of pathological cardiac hypertrophy. This target is SIRT2 [inaudible 00:04:33] poorly characterized member of the Sirtuin family of proteins, which is a family of class III NAD-dependent deacetylases that regulate metabolism and age-related diseases including diabetes and cardiovascular diseases. In the current study, first authors Dr. Tang and Chen, corresponding authors Dr. Chen and Liu from the Chinese Academy of Medical Sciences in Peking Union Medical College used wild-type and Sirt2 knockout mice, and showed that SIRT2 protein levels and activity were reduced during pathological cardiac hypertrophy.

                                                SIRT2 deficiency promoted aging and angiotensin II induced pathological cardiac hypertrophy, and blunted metformin-mediated cardioprotective effects. On the other hand, SIRT2 overexpression repressed pathological cardiac hypertrophy. The molecular pathway involved deacetylation of liver kinase B1 at lysine 48 by SIRT2 to activate AMP-activated protein kinase sickling, which prevented hypertrophy of cardiomyocytes. Thus, SIRT2 is a potential target for therapeutic interventions in aging and stress-induced cardiac hypertrophy.

                                                The next study is the largest comparison of the prognostic value of coronary artery calcium with functional stress testing in patients with stable chest pain. In this study from first and corresponding author Dr. Budoff from Los Angeles Biomedical Research Institute and colleagues, authors looked at the PROMISE trial where patients with stable chest pain or dyspnea, and intermediate pre-test probability for obstructive coronary artery disease were randomized to functional testing or anatomic testing.

                                                Their main finding was that these chest pain populations referred for testing had a low event rate and both tests had different strengths. Coronary artery calcium had a high sensitivity for future cardiovascular events whereas functional testing had a high specificity. The clinical implications are that a normal coronary artery calcium score has a very low event rate and perhaps maybe used to avoid further cardiac testing in a stable chest pain population. On the other hand, an abnormal functional test result including information on exercise and symptoms has a moderate prognostic value.

                                                Of note, coronary CT angiography provided better prognostic and discriminatory power than either coronary artery calcium or functional testing. The implications of these important results are discussed in an accompanying editorial by Dr. David Newby from Edinburgh entitled, Can I Have My Cake and Eat It? On that intriguing note, we've come to the end of today's summaries, now for our feature discussion.

                                                For today's feature discussion, we are talking about an increasingly important population that is pediatric and adult patients with congenital heart disease undergoing cardiac catheterization. A little bit out of my usual comfort zone, but then you see, I'm with two spectacular experts today, Dr. Gerard Martin from Children's National Health System in Washington DC, one of the authors of today's feature paper; and Dr. Gerald Greil, Associate Editor from UT Southwestern. Welcome gentlemen.

Dr. Gerard Martin:           Thank you Carolyn.

Dr. Gerald Greil:                Thank you Carol.

Dr. Carolyn Lam:               Gerard, no that would be Dr. Martin. Enlighten people like me who don't think about this every day, why the importance of looking at cardiac catheterization, and adverse outcomes in this particular population?

Dr. Gerard Martin:           Carolyn, that's because of the tremendous advances in medicine, and particularly medicine that's dealing with children with congenital heart defects. Cardiac catheterization was once purely a diagnostic study. Now, it's a less invasive definitive treatment option for many of our pediatric and adult patients with congenital heart defects. As you may or may not know, congenital heart defects are the most common birth defects that impact nearly one out of every hundred live births.

                                                As I mentioned, we have these tremendous advances. As a result of that, there are now over a million children living with congenital heart defects. In the USA alone, improvements in care over the past 50 years, there are now more adults than children living with congenital heart defects.

Dr. Carolyn Lam:               Wow. Now, I understand. I mean, cardiac catheterization not just meeting diagnostic but therapeutic, and such an important patient population. Tell us about your study?

Dr. Gerard Martin:           As we said, cardiac catheterization is now replacing surgery for some of our defects. For some of the more complex defects, catheterization is providing treatments that make the surgery easier. Now in surgery, we've had registries for many years. These registries provided measurement of survival that allow comparison of programs, and we didn't have that ability with cardiac catheterization. The American College of Cardiology developed the IMPACT Registry. That was to solely provide measurements of the outcomes of catheterization procedures in the children and adults with congenital heart disease.

                                                Now, one aspect of the quality of the program is your rate of adverse outcomes; but simply measuring the number of adverse outcomes does not provide enough discrimination to compare programs. I think you can probably imagine that adverse outcomes will increase based upon the complexity of the type of patients you see, or the types of procedures that you might be performing. What we wanted to do was to create a risk standardization tool for our population where we can measure variation and performance between programs. If we can do that, then we can learn from the best performers to improve all the others.

Dr. Carolyn Lam:               That's beautifully put. Could you tell us what you found?

Dr. Gerard Martin:           Sure. The IMPACT Registry began on about 2011 and has grown from 50 sites to 111 sites in 2017. That's the majority of the sites in the United States that perform cardiac catheterization on children. We have now over 115,000 procedures. What we wanted to do with this is to look at some of the early procedures that were included and to see how adverse events were occurring. When we created the registry though, we used data variables from a previous research study in Boston called the CHARM.

                                                They created a tool to risk standardized outcomes during procedures. They did it by coming up with four categories of procedures, and some four markers of hemodynamic vulnerability. We tested their methodology with IMPACT, and it didn't really performed particularly well. In this study, what we did was to increase the number of risk categories. We took the nearly 200 types of procedures we do in the cath lab and divided them into six categories. We also increased the indicators of hemodynamic vulnerability from four to six.

                                                Now, what I mean by hemodynamic vulnerability? What is the patient's oxygen level when they go into the procedure? What is their blood pressure when they're in the procedure? Do they have one ventricle, or do they have two ventricles? What is the resistance in the lung vessels? All these are critically important. Lastly, we looked at some baseline patient characteristics. In other words, was age important? Sex, genetic conditions, or other comorbid conditions like the level of mechanical support that the patients were on. Then we put all that into our model to see if we could come up with a risk score.

Dr. Carolyn Lam:               Right. The final adjustment model? Which factors that they include in the end?

Dr. Gerard Martin:           We did find that there are lot of adverse events that do occur. We found major adverse events occurring in about same 7% of our patients. Most common adverse events were bleeding, or rhythm disturbances that require some medicine, or cardioversion during the procedure, or death during the hospitalizations. We did find that these major events were more common in the youngest patients or neonates, children under a month of age, or in patients with genetic disorders, or single ventricle physiology, and also patients that went to the cath lab with their kidneys not working very well.

                                                In the end, we did create a risk adjustment model that included the type of procedure that was done, the number of hemodynamic vulnerability indicators, and whether or not the patient had renal insufficiency, or single ventricle physiology, or coagulation, and we found really good discrimination. Our discrimination had a C-stat of 0.76 in the derivation cohort, and 0.75 in the validation cohort. The slope of the curve was excellent, so we really think we have something now that we can use as a tool.

Dr. Carolyn Lam:               Gerald, you're a pediatric cardiologist. Could you give us your perspective on how important these results are?

Dr. Gerald Greil:                I think it's the largest and the first study, which kinds of give us a calibration in our field how successful interventions are. How we can make centers better without finger pointing on specific centers, and how to advance the field as a whole? From that perspective, I'm quite excited that the group offered us to publish this paper in circulation. I was kind of asking a question to Dr. Martin because obviously, all essentials are closely monitored. There's obviously data publicly available. Do you think there's a risk that this way to monitor centers within the United States or probably worldwide, that it's potentially preventing innovation or risky procedures?

Dr. Gerard Martin:           I think that, that's a good question. I think it's one thing that whenever we talk about transparency or public reporting, it's an argument against it. I think that having a model like this, actually levels the playing field. In other words, centers that are risk averse who aren't particularly innovative, you'll be able to look at those centers, see what type of patients they're doing and look at their adverse events for a low-risk population. Then, you can also look and see some other centers that are doing more complicated procedures, higher risk, and you can see what their adverse event rate is.

                                                Certainly, this is only talking about the adverse events. This has to be put together with the outcome of the procedure. In other words, if you're trying to relieve an obstruction, did you relieve it? Did you meet the intended goal of the procedure? This is only half of the story. The other part of it is, did you get the intended goal of the procedure? When you put the two of them together, perhaps some of those centers that are risk averse have lower complications, but maybe their success rate is lower. This will be able to tell the public everything they know, and they'll be able to tell their providers what they need to know to get better.

Dr. Carolyn Lam:               I have to agree. Your paper does highlight, I think. Gerard, just one other question. What do you think our next steps?

Dr. Gerard Martin:           The next step is to test the data. We have a new version of IMPACT that has rolled out, version 2 that has new procedures in it. Now, we have to test the data and we actually have to look for variability. Can we see a variation between the programs? Then, once we see if there's variation, if we see there is best performers and those performers that could improve, a question then is how do we take from what the best performers are doing to try and lift those that need to improve up. That's going to be the true hard work for this registry.

Dr. Carolyn Lam:               Thank you so much for publishing it with us. Thank you so much audience for listening with us today. Don't forget to tune in again next week.


Nov 13, 2017

Dr. Carolyn Lam:               Welcome to Circulation on the Run, your weekly podcast summary and back-stage pass to the journal and its editors. I'm Dr. Carolyn Lam, Associate Editor, from the National Heart Center and Duke National University of Singapore.

                                                What is the evidence we have for LDL-lowering therapy in primary prevention? For individuals with an LDL cholesterol above 190 mg/dL, well, you may think you know the answer, but today's featured discussion may surprise you like it did for me, and this is a must-listen in my opinion for those of us taking care of these patients. More soon right after these summaries.

                                                How can we enhance the survival and therapeutic potential of human pluripotent stem cell-derived endothelial cells? Well, the first paper in today's journal tells us how. The first author Dr. Lee, corresponding doctor Dr. Yoon, from Emory University School of Medicine in Atlanta, Georgia, developed a novel, fully-defined, cell culture system to generate endothelial cells from human pluripotent stem cells. They not only showed that these endothelial cells had pro-angiogenic activities and exerted favorable therapeutic effects in repairing limb ischemia, but also showed that encapsulation of these cells in a biocompatible peptide amphiphile nanomatrix gel improved long-term survival of these endothelial cells in an ischemic environment and improved vessel-forming properties. This novel cell culture system and gel-mediated transplantation may serve as a novel platform for cell-based therapy.

                                                The next study brings us one step closer to application of immunomodulatory therapies in pulmonary arterial hypertension. In the study, first author Dr. Saito, corresponding author Dr. Rabinovitch, and colleagues from Stanford University School of Medicine isolated lung immune complexes and pulmonary arterial hypertension target antigens from lung tissues from 16 patients with pulmonary arterial hypertension and 12 controls. SAM domain and HD1 domain-containing protein, which is an innate immune factor that suppresses HIV replication, was identified and confirmed as highly expressed in immune complexes from patients with pulmonary arterial hypertension. These immune complexes resulted from elevation in products of human endogenous retrovirus K. The human endogenous retrovirus K deoxyuridine triphosphate nucleotidohydrolase, or dUTPase, activated B cells, elevated cytokines and monocytes and pulmonary endothelial cells, and increased pulmonary arterial vulnerability to apoptosis, thus contributing to sustained inflammation, immune dysregulation, and progressive obliterative vascular remodeling. Furthermore, rats treated with the human endogenous retrovirus K dUTPase developed pulmonary hypertension. In summary, this study suggests that harnessing mechanisms that repress human endogenous retrovirus K expression and its sequelae could prevent and reverse pulmonary arterial hypertension.

                                                The next study looked at the association of timing of coronary angiography with ischemic outcomes of non-STEMI who are at high risk with a Gray score of more than 140 in the TAO Trial. In this report from first author Dr. Deharo, corresponding author Dr. Steg, and colleagues from L'Hopital Bichat from Paris, France showed that in these high risk, non-STEMI patients, a very early invasive strategy of coronary angiography within the first 12 hours was associated with a lower risk of death in MI at 180 days compared to an early strategy of between 12 to 24 hours or a delayed strategy of between 24 and 72 hours. The bleeding risk was not different between patients managed with the very early, early, or delayed strategy. These observations deserve prospective confirmation in a randomized trial.

                                                The next study provides contemporary mortality trends for STEMI and non-STEMI. In this paper from first author Dr. Puymirat, corresponding author Dr. Danchin, and colleagues from Hopital europeen Georges-Pompidou in Paris, France, the authors assess trends in the characteristics, treatments, and outcomes for EMI from five month-long registries conducted five years apart and spanning 1995 to 2015, including more than 14,000 patients admitted to cardiac intensive care units in metropolitan France. They observed major changes in the characteristics and management of both patients with STEMI and those with non-STEMI over the last 20 years. The mean age decreased in patients with STEMI and remained stable in patients with non-STEMI, whereas diabetes, obesity, and hypertension increased. At the acute stage, intended primary PCI increased from 12 to 76 percent in patients with STEMI. In patients with non-STEMI, PCI within 72 hours from admission increased from 9 to 60 percent. In parallel with these changes, six-month mortality consistently declined in patients with STEMI, whereas in patients with non-STEMI, six-month mortality reached a plateau after 2010. The authors concluded that future challenges will be to reduce pre-hospital mortality and to improve long-term survival after the acute myocardial infarction event.

                                                That wraps it up for your summaries. Now for our feature discussion!

                                                What evidence do we have from randomized trials supporting the benefit of LDL cholesterol lowering as primary prevention among patients with an LDL cholesterol above 190 mg/dL? You may be surprised to know that until today's journal, we had very little trial evidence supporting this. But I'm so pleased to have with us the corresponding author of our featured paper today, Dr. Kausik Ray from Imperial College, London, who's going tell us a bit more and discuss this very intriguing paper with our Editor for Digital Strategies, Dr. Amit Khera from UT Southwestern. Welcome, both.

Dr. Kausik Ray:                  Hi.

Dr. Amit Khera:                 Thanks for having us.

Dr. Carolyn Lam:               Kaus, you are a familiar voice and so pleased to have you here. Please tell us, is this the first evidence we have from a randomized trial for primary prevention in those with LDL above 190? Tell us about it.

Dr. Kausik Ray:                  Yeah, it is. It really came about because we were interested in familial hypercholesterolemia and we used the level of 190 to talk about either primary hypercholesterolemia, which may have a genetic basis, or not. I kept hearing that there is no trial evidence, so you're not going to be able to ethically do a trial today despite the fact there's not much evidence, because most of us think that it's a bad thing to leave people on placebo in patients above 190, so I thought the only way to do this was to go historically to the WOSCOPS Study, which is, as you remember, 6,500 people, elevated LDL cholesterol. Interestingly, you go to WOSCOPS, the median LDL in that population is very close to 190. So, that gives a good starting point, thinking that we'll have at least half the population.

                                                Now interestingly in WOSCOPS, although none of the patients had a history of myocardial infarction, a very small number of the 6,500, about 1,000 actually had evidence of some other vascular disease, so maybe a TIA, maybe angina, maybe some sort of ECG non-specific change of coronary disease. Today, you would say, well, actually, you've got to give these people a statin because there's evidence of vascular disease, PVD, et cetera. So we had to take those people out and that left us with 5,529. Once you break people down by LDLs above and below 190, you have 2,560. You could actually look at the randomized treatment effect of pravastatin, which was the statin chosen, over a five year period both above and below 190.

                                                But interestingly, this was the first study and what we showed was that in this population, even with as little as 23% reduction in LDL cholesterol, over a five year period, you saw a statistically significant 27% reduction in CHD and if you take the usual 3 point MACE of current clinical trials, there was a 25% reduction, already statistically significant. We also had the ability to link data over 20 years. Remember, after the five year randomized treatment period, it becomes observational in nature, but what it showed was that when you gave nearly 40% in each arm statins and you followed people up this legacy effect, over a 20 year period, the people with the LDL above 190, that translated into this 28% reduction in CHD death. It translated into a 25% reduction in CV death, and actually an 18% reduction in all-cause mortality, which you didn't see in the population with slightly lower LDL cholesterol.

                                                This is the best evidence we're ever going to get, really, and answer the question about what should we do in this patient population. Should we treat with lipid-lowering therapy? The answer, unequivocally, is yes, and the longer you treat, the more likely you are to see survival benefits.

Dr. Carolyn Lam:               Oh, my goodness! I just love his paper. I have to humbly admit. I mean, it's in the guidelines already that we should treat these individuals with LDL above 190, and it really made me think how I'd taken for granted that there would be a whole body of evidence behind it from randomized trials, and you are right! This is the first, and likely going to be the last we're going to get, because we can't randomize them. So, congratulations. What you said just now, I can already hear myself playing this podcast to my patients. May I just ask, are there other remaining questions to answer, and then what do you also say to those that say, well what are the harms? How do you balance that with any potential harms?

Dr. Kausik Ray:                  In this particular study, given there was overall safety data observed in the WOSCOPS Trial population and in their extended follow-up in the overall 6,500 person cohort, we didn't go on and look at that. There was no evidence of harm in the extended follow-up of 6.500 people, so we didn't see the potential added gain in specifically looking for that. The main question we wanted to answer, because people had always pulled primary and secondary prevention patients together, and in fact, your best evidence is actually from CTT, pooling of primary and secondary prevention patients where they break the data down by an upper limit of about 175. With patients above 175, they don't specifically answer that question. So, to answer your question, we didn't look at that in the overall WOSCOPS Trial population. There was no signal for harm that was noticed. Even things like glucose elevation, if you remember in WOSCOPS, tended to be a little bit lower.

Dr. Amit Khera:                 Let me comment on a few things about this paper. First, I want to congratulate Dr. Ray and his colleagues. I was a history major and I think this is a great use of a historical tool. At this point, I think we can talk about WOSCOPS. It's 22 years old. It is part of the medical history and a very seminal article. I think they got creative because, as he mentioned. We have guidelines that support this treatment, but this is almost an unanswerable question, whether you say it's from ethics, or from equipoise, it was essentially unanswerable. So, they had to go back and take this historical study where practice patterns were different, to be able to look at this question. It was pointed out, there's pretty clear evidence in here and I think if you look at that during the five-year study period of the randomized period, pretty clear evidence that treating participants with LDLs above 190 without vascular disease certainly lowers cardiovascular disease events.

                                                One of the best things about working on the editorial board is being able to work closely with authors, and I have to also thank Dr. Ray and his colleagues for being so gracious in working with us closely in some modifications as this went along. We hope, and I hope he feels this way, too, that at the end of the day, the product ends up being even better than where we started. That's our goal is to really help and work with authors in that way and they were incredibly responsive. The two things I thought they did really well that were insightful to the US guidelines and beyond. One is they also restricted to the group without diabetes, without ASCVD less than 7.5%, and some other parameters to really hone down on what we have in the current US guidelines and still the finding was consistent that the statin therapy benefited that group.

                                                The other part was just acknowledging that the legacy part, the long-term effect, is really valuable. They published heavily in this area, but at that point, it becomes an observational component. It's not part of the randomized period. The reason that adds value, if you look at our guidelines above the age of 21, an LDL above 190 can be treated with a statin, there would be less controversy if your LDL was 200 and you're 55, but if you're 22 or 23, I think there may be more angst. That's where the long-term data is important, because we're not looking necessarily always at 10 years, but we're looking at 20 or 30 or 40 or 50 years. I think this does at least shed some light. I appreciate the study population was older, but a least it helps us look at maybe some of the long-term benefits.

                                                If I may, Carolyn, I would love to ask Dr. Ray a question. Kaus, when you guys did this, the group with the LDL less than 190 had essentially similar benefit. The p-interaction was no. I think we have to acknowledge that the LDLs were higher in that group than what would seem because the lowest level was 155. Is it above 190, or should it be above 160 where we treat patients with statins?

Dr. Kausik Ray:                  Yes, and I really want to thank the editors, because there were certain things that you pushed us with analyses and I think that you could make the case that if you have a LDL cholesterol above 155, over a five-year randomized treatment period, there was a significant reduction in CHD and MACE as well. So, you could make that point that actually the cutoff should perhaps be pulled down even further to about 155. What's interesting is, these groups, when you broke them down, age was identical, BMI was identical, blood pressure, and everything else. The only thing that was different, really, was the LDL cholesterol, which impacted on total cholesterol. TGs, HDLs were absolutely identical. I think you could probably make the case.

                                                I think the one thing that we didn't see, although it's observational in those with slightly lower LDL cholesterols, is that over the 25 year period, they seem to get slightly less mortality benefits. Now, that could be a chance finding, because it's observational. We don't really know the implications of that, but I think over a five-year period, this is the best evidence you're going to get for primary prevention, right?

Dr. Amit Khera:                 Agreed. The US guidelines do say above 160, it's a point of consideration. It can be a factor to consider as we think about treatment, so perhaps this helps bolster that point as well.

Dr. Kausik Ray:                  It's not just the American guidelines. In the European guidelines, when they use score, if you look at LDL cholesterol levels, the European case fatality 10 year risk is 2.5%, which is equivalent roughly to 7.5% fatal and non-fatal MI in the pooled cohort equation. There they still have diet and lifestyle, but it says, "Consider pharmacological," and one of the things I thought was really interesting is if you did a 10 year risk calculation in this group, 67% of the population with an LDL above 190, you would have said the predicted 10-year risk was below 7.5%, but the 10-year observed risk was double that. It was 15%. If you did the same thing for the group between 155 and 190, your ten-year risk predicted would be in most of these people, you would have said about 90% actually are less than 7.5%, so you wouldn't have given them a statin. But, their observed event rates in the placebo group was about 11%.

                                                So, I think that it tells you if you have an isolated elevated cholesterol above 155, you're probably going to be underestimating risk if you're using global risk score, and perhaps a discussion with the patient about risks and benefits in the way that most of us try to do and citing data like this might encourage patients to actually start that therapy earlier, which most of us probably believe from genetic and legacy effect is probably beneficial. That's one of the other implications of this.

Dr. Amit Khera:                 This is why one has to read not just the abstract, but all the details, because there are so many kernels of interesting findings in this paper beyond just the highlights that we hit upon.

Dr. Carolyn Lam:               Thank you both for just a marvelous discussion of an incredible paper that is really, really going to be extremely clinically relevant. We're so proud to be publishing this in Circulation this week.

                                                Audience, you heard it right here. Don't forget to tune in again next week as well to Circulation on the Run for even more hot news.

Nov 6, 2017

Dr. Carolyn Lam:               Welcome to Circulation on the Run, your weekly podcast summary and backstage pass to the journal and its editors. I'm Dr. Carolyn Lam, associate editor from the National Heart Center, and Duke National University of Singapore.

                                                In just a moment, we will take a deep dive into the issue of age and its association with outcomes of primary prevention ICDs in patients with non-ischemic systolic heart failure.

                                                Yes, a long-awaited discussion from the Danish trial. That, in just a moment. First, here's your summary of this week's Journal.

                                                The first original paper provides evidence of a true association between disturbed genetic imprinting and Preeclampsia. This paper is from co-first authors, Dr. Zadora, and Dr. Singh, and co-corresponding authors, Dr. Izsvak, from the Max Delbrück Center for Molecular Medicine; Dr. Hurst, from the University of Bath; and Dr. Dechend, from the Experimental and Clinical Research Center of Berlin.

                                                These authors performed an unbiased analysis of genome-wide molecular data on raw characterized patient material, from normal controls, and patients with  Preeclampsia, and identified DLX-5 as an imprinted target gene, with novel placental function in Preeclampsia. Due to loss of imprinting, DLX5 was upregulated in 69% of placentas from Preeclampsia patients. Levels of DLX5 correlated with the classical Preeclampsia markers.

                                                DLX5 was expressed in human, but not in urine trophoblast, underlying the known human specificity of Preeclampsia. Finally, DLX5-induced overexpression if trophoblasts faithfully modeled Preeclampsia in a cell culture system. In summary, this paper shows that disturbed imprinting is common, and may play a causal role in Preeclampsia.

                                                The next study affirms that stenosis severity is better discriminated using coronary invasive physiologic indices, than using coronary angiographic assessment. First author, Dr. Lee, corresponding author Dr. Koo, colleagues of Seoul National University Hospital, studied 115 patients with left anterior descending artery stenosis, who underwent both ammonia positron emission tomography, or PET, an invasive physiologic measurement.

                                                Myocardial blood flow measured using PET, and invasively measured coronary pressures, were used to calculate microvascular resistance, and stenosis resistance. They found that both fractional flow reserve, or FFR, and instantaneous weight free ratio, or IFR, decreased as angiographic stenosis severity, resistance, and pressure gradient increased, and hyperemic myocardial blood flow decreased.

                                                When the presence of myocardial ischemia was defined by both low hyperemic myocardial blood flow, and low coronary flow reserve, the diagnostic accuracy of FFR and IFR did not differ, regardless of cutoff values for hyperemic myocardial blood flow, and CFR. However, at any given stratum of a given stenosis, physiologic classification of stenosis severity using FFR or IFR showed better discrimination of a unique relationship between absolute myocardial blood flow, and pressure gradient, than anatomic classification using angiographic percentage.

                                                In summary, by demonstrating coronary physiologic responses to coronary stenosis, these authors showed that stenosis severity is better discriminated, using invasive physiologic indices, than using angiographic assessment.

                                                The next paper identifies a previously unknown angiogenic growth factor that can be enhanced therapeutically to repair the heart after myocardial infarction. This novel growth factor is endoplasmic reticulum membrane complex, Subunit 10, or EMC10, which the authors previously identified by bioinformatic secretome analysis in bone marrow cells.

                                                In the current paper, from co-first authors Dr. [Rabel 00:04:35], and [Krof Clengobill 00:04:37], and corresponding author Dr. Wollert, from Hanover Medical Center, and colleagues, the authors investigated the angiogenic potential of EMC10, and its mouse homologue, in cultured endo fetal cells and infarcted heart explants. They found that EMC10 and its mouse homologue signal a virus, small GTAPases; p21-activated kinase; and p38 mitogen-activated protein kinase, to promote endothelial cell migration.

                                                In mice with acute myocardial infarction, bone marrow derived monocytes and macrophages produced EMC10 endogenously, to enhance infarct vascularization, tissue repair, and heart function. Furthermore, subcutaneous treatment with recombinant EMC10 for one week, after myocardial infarction, augmented infarct vascularization and repair, and led to a sustained improvement in heart function and survival.

                                                The next study is the first prospective randomized trial of screening for atrial fibrillation, with a smartphone-based, single-lead, electrocardiographic system in 1,001 patients, aged 65 years and above, with a CHA2DS2-VASc score of two and above, and without a history of atrial fibrillation.

                                                In this paper, from first and corresponding author Dr. Halcox, from Swansea University Medical School, in the United Kingdom, and colleagues, patients were randomized, either to biweekly electrocardiographic recordings with the iPhone device, or to routine over a 12-month period.

                                                The smartphone-based electrocardiographic approach was at least three times more likely to identify incident atrial fibrillation, than routine care, and at a cost of just over $10,000 per case identified, and was judged to be a highly acceptable approach in this group of patients. These results support consideration of evaluation in an appropriately-powered, event-driven randomized trial, to confirm the clinical and cost effectiveness of such an approach to stroke prevention in atrial fibrillation.

                                                Well, that wraps it up for your summaries. Now for our feature discussion. The Danish trial really created a huge splash last year, when it was reported that a primary prevention ICD in patients with non-ischemic systolic heart failure, may not actually reduce all cause mortality. Something that we had, perhaps, taken for granted, and in fact, entered our guidelines.

                                                Now, however, there was a pre-specified subgroup analysis at the time, that suggested a possible age-dependent association, between ICD and mortality, in the Danish trial. This week, we are so pleased to be discussing an in-depth analysis of the association between age and outcomes in the Danish trial.

                                                I'm so pleased to have the first author of today's featured paper, Dr. Marie Bayer Elming, of Copenhagen, Denmark, as well as Dr. Sana Al-Khatib, who's not only an associate editor of circulation, but also the author of an accompanying, and she is from Duke, Durham, North Carolina. Welcome, ladies!

Dr. Bayer Elming:              Thank you. Happy to be here.

Dr. Sana Al-Khatib:          Thank you so much.

Dr. Carolyn Lam:               Sana, could you start by framing why this paper is so important, and why we've been looking forward in anticipation to these results?

Dr. Sana Al-Khatib:          Absolutely. As you know, data on the outcomes of primary prevention ICDs in patients with non-ischemic cardiomyopathy started emerging in the early 2000s, or so. Then in 2005, the sudden cardiac deaths and heart failure trial was published, that included a large number of patients with non-ischemic cardiomyopathy, and absolutely showed survival benefits from primary prevention ICDs in those patients. Of course, there were also patients with ischemic cardiomyopathy.

                                                But really, that trial formed the basis of the guidelines, recommendations, that have informed our practice for the last 12 years, that basically tell us that we should consider implanting a primary prevention ICD in patients with non-ischemic cardiomyopathy, who have an EF of 35% or less, who have Class II or III heart failure symptoms. As long as they are on optimal care at the end, they have a reasonable life expectancy.

                                                So that's what's we've been doing for years, and then, the Danish trial was published this past year, that really called into question the prior findings, and the current practice. Because Danish, as you stated, showed no survival benefit with primary prevention ICDs, but there are many aspects about the trial that people need to pay attention to, to put the results in perspective.

                                                The fact that 58% of patients in the trial, in those arms, received cardiac resynchronization therapy ... the fact that the trial required that patients have an elevated NTproBMB level, to be considered for enrollment ... that may have biased the results toward a higher risk of non-sudden cardiac deaths, so on, so forth.

                                                I think what was really interesting, and caught people's attention, when the paper was published, was this subgroup analysis that showed that younger patients may benefit more than older patients. I think, many of us, Carolyn, were really awaiting the results of a more dedicated analysis, looking at age in Danish, and Dr. Elming and her colleagues did a great job looking at this very closely in their paper, and showed great results, and probably will let Dr. Elming share those results with us.

Dr. Carolyn Lam:               Yes, absolutely, Sana. Actually, I just wanted to echo how surprised everyone was, and the immediate thing was, "Oh, my goodness. What do we do with the guidelines?" Maybe we should get back to that later, and Marie, please share with us, what did you do, and what did you find this time?

Dr. Bayer Elming:              The reason why we did this study was that, in this main Danish trial, age was the only one of the 13 pre-specified subgroups that had a significant treatment by a subgroup interaction. This suggested that a younger patient might have a survival benefit from ICD ... the implication, even though the overall study was neutral. So we wanted to further investigate this relationship between age and effective ICD implantation.

                                                What we did was to look at the relation between age and effective ICD, and we found that there was this linear relation, for each year of younger age, that was associated with a reduction, a 3% reduction in the hazard ratio, for the benefit of ICD.

                                                Also, we did this selection impact curve, which is a bit technical, but what it does is to describe the expected survival for the population, on as a whole, for the different age cutoffs for ICD treatments.

                                                So, if we take into account, both the patients receiving an ICD, and those who did not, we could see why we would get the highest survival for the population as a whole. What we found was that, when no one in the population received an ICD, around 70% would survive.

                                                If everyone in the population received an ICD, only 72% would survive, but if we chose 70 years as the age cutoff ... so, patients younger than 70 years received an ICD, and patients older than 70 years did not receive an ICD, we got the highest survival for the population, and 75% would survive.

Dr. Carolyn Lam:               Thank you, Marie. What important results. So, maybe, still consider ICDs for primary prevention ... in our non-ischemic systolic heart failure, patients were less than 70 years old. Is it as simple as that, Sana? You wrote a beautiful editorial. Tell us, what are the clinical implications?

Dr. Sana Al-Khatib:          This is an important question. Danish was an important trial, but in my mind, it truly doesn't refute the role of primary prevention ICDs in patients with non-ischemic cardiomyopathy. As I mentioned earlier, the majority of patients enrolled in Danish received a CRT device. And so, you end up questioning, what does that actually mean, for those patients who are not eligible for cardiac resynchronization therapy?

                                                So, I actually believed that, and as you know, Carolyn, and maybe Marie knows, as well, there have been several meta analyses that have been published, combining data on patients with non-ischemic cardiomyopathy only, and excluding patients with cardiac resynchronization therapy from Danish, that have actually now shown, consistently, a significant improvement in survival, with a primary prevention ICD ... including one that was done by our group.

                                                So, no, I don't think that, based on the results, we should say, "No, we shouldn't be offering primary prevention ICDs to patients with non-ischemic cardiomyopathy," and this beautiful analysis that was done by Marie and her group actually shows that, at least for those patients who are 70 years of age and younger, I think we should absolutely continue to consider them for the therapy, and offer them the therapy, if they're appropriate candidates.

                                                Then, of course, if the patients are older than 70,, and they meet criteria for cardiac resynchronization therapy, I think it will be important for us to be talking to the patients about ... is the RTD with a defibrillator, versus a CRTP only, with a pacemaker, and talking about the pros and cons, and everything else? But in those patients who are older than 70, who don't meet criteria for CRT, I think more research is needed, to really understand the role of primary prevention ICDs in those patients. We definitely need more data there.

Dr. Bayer Elming:              I definitely agree that, of course, for the patients older than 70 years were not candidates for CRT treatment. These patients, we do not know very much about 'em, and this study that we did, do not answer that question. Based on the Danish study, and this further analysis of the age inspection, the guidelines in Denmark also state that patients younger than, we say, 68 years, because that was the age cutoff used in the '08 Danish trial, you should definitely think of giving patients with non-ischemic cardiomyopathy an ICD.

                                                But for the older patients, it depends on a variety of co-factors, such as co-morbidity, or frailty, and it should be an individual assessment of the patient. So, I agree with you, Sana.

Dr. Carolyn Lam:               That's wonderful. Hey, just one more question. Sana, I'd like you to put on your AE hat, now, and sort of think with me. In circulation, we don't ... well, we're careful about publishing subgroup analyses, so to speak, right, of results. You articulated, in your editorial, reasons why this, perhaps subgroup analysis, may be different from others. Could you elaborate on that a bit?

Dr. Bayer Elming:              Yeah, and absolutely, that's a great question. As you pointed out, I mean, you really ... the conventional wisdom in clinical research is to be careful, interpreting subgroup analyses. I think there are some strengths in this particular analysis, as Marie stated: "Here's what we specified." The other thing is, I believe that Marie and her group then came, and did their very robust statistical methods, and really, probably most importantly, if you look at their findings, they actually really align well, and support their main conclusion.

                                                For example, looking at the fact that older patients had the higher presence of co-morbidities, that they had a higher level of [Co-BMP 00:17:00], they had had a longer duration of heart failure ... I mean, all those things most likely had an impact on their mode of death, really making it more likely for those patients to succumb to non-sudden cardiac death. I think the whole story makes a lot of sense.

Dr. Bayer Elming:              If I can elaborate a bit on this, I think one of the important findings from the study is that we show that mode of death varied according to age. So, the rates of sudden cardiac death were almost similar, between the younger and the older part of the population. But the rates of non-sudden death were almost twice as high in the older part of the population. This is a really good explanation why the ICD implantations have less impact in the older patients.

Dr. Carolyn Lam:               Yeah, because ICDs would definitely not be expected to reduce non-sudden cardiac deaths. Really, really, well put. Oh, thank you so much, Marie. We're so proud to be publishing your beautiful paper, as well as your editorial, Sana, and thank you for this great conversation.

                                                Well, listeners, I'm sure you enjoyed that as much as I did. Thank you for joining us this week, and don't forget to tune in next week.