Aug 27, 2018
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.
Do we finally now have a simple, evidence-based way to make a diagnosis of heart failure with preserved ejection fraction? Well, today's feature paper certainly brings us closer to that goal and you must listen to the discussion coming right up after these summaries.
Bleeding is commonly cited as a reason for stopping oral anti-coagulants. However, what is the prognostic significance of minor bleeding events, or so called nuisance bleeding, in patients with atrial fibrillation on oral anti-coagulants?
First and corresponding author, Dr O'Brien from Duke Clinical Research Institute and her colleagues, identified 6771 patients with atrial fibrillation in the Orbit AF Prospective Outpatient Registry. They ascertained nuisance bleeding from medical records defined as minor bleeding that did not require medical attention. Overall, 20% had documented nuisance bleeding giving an incidence rate of 14.8 events per hundred person years. Nuisance bleeding was not associated with a higher risk of major bleeding, or a stroke and systemic embolism over the next six months.
These findings therefore suggest that the occurrence of nuisance bleeding or minor bleeding should not lead to changes in anti-coagulant treatment strategies in patients treated with anti-coagulants.
The next study sheds new light on mechanisms linking NLRP3 inflammasome activation to atherogenesis. Dr Westerterp from Columbia University, New York and colleagues studied mice with myeloid deficiency of ATP binding cassette transporters A1 and G1 and concomitant deficiency of the inflammasome components NLRP3 or caspase-111.
They showed that cholesterol accumulation in myeloid cells activated the NLRP3 inflammasome. NLRP3 inflammasome activation enhanced neutrophil accumulation and neutrophil extracellular trap formation in atherosclerotic plaques thus accelerating atherogenesis.
Patients with Tangier's disease, who had ATP binding at transporter A1 loss of function, had increased myeloid cholesterol content and showed markers of inflammasome activation. Thus, inflammasome activation may underline cardiovascular disease in these patients.
The next study identifies TPX20 as a novel transcription factor regulating angiogenesis. TPX20 is a crucial transcription factor for embryonic development and its deficiency is associated with congenital heart disease. However, its role in angiogenesis has been not been previously described. At least until today's paper from co-first authors Dr Meng and Dr Gu and co-corresponding authors Dr Cooke and Dr Fang from Houston Methodist Research Institute.
These authors use loss and gain function approaches to explore the role of TPX20 in angiogenesis both in vitro and in vivo. They showed that with VEGF stimulation, the transcription factor TPX20 upregulated PROK2 with is secreted from endothelial cells and gauges its receptor PROKR1 and thereby promotes angiogenesis in autocrine manner.
This novel signaling pathway appeared to be highly conserved as it functioned in zebra fish vascular development and the angiogenic response to ischemia in a mouse model of peripheral disease. The authors furthered showed the selective role of TPX20 in endothelial migration but not proliferation. Furthermore, treatment with recombinant PROK2 the critical effector of TPX20, improved blood profusion and functional recovery in the mouse peripheral artery disease model. Thus, these data highlight the therapeutic potential of PROK2 in augmenting functional angiogenesis for diseases associated with this regulated angiogenesis.
In patients with atrial fibrillation, left atrial appendage closure with the Watchman device, is known to prevent thromboembolism from the left atrial appendage. However, thrombus may still form on the left atrial face of the device, which then may potentially embolize. This next paper provides important data on the incidents, predictors, and clinical outcomes of device-related thrombus after left atrial appendage closure.
First author, Dr Dukkipati and corresponding author Dr Reddy from Icahn School of Medicine at Mount Sinai, New York and their colleagues studied the device arms of 4 prospective FDA trials of patients undergoing the Watchman implantation. These were the PROTECT AF, PREVAIL, CAP, and CAP2 trials.
They found that following percutaneous left atrial appendage closure with the watchman device, the incidence of device-related thrombus was 3.7% and this was associated with a more threefold higher risk of stroke and systemic embolism. Predictors of device-related thrombus were a history of trans- ischemic attack or stroke, permanent atrial fibrillation, vascular disease, a larger left atrial appendage diameter, and a lower left ventricular ejection fraction.
Device-related thrombus was not associated with an increased risk of cardiovascular or all-cause mortality. Nearly 75% of patients that developed device-related thrombus did not experience a stroke. And ischemic strokes occurring in patients with device-related thrombus accounted for approximately 10% of all ischemic strokes, following left atrial appendage closure. Thus, given the ramifications of device-related thrombus, a judicious surveillance strategy using periodic transesophageal echo cardiography may be considered particularly when risk factors for device-related thrombus are present.
Well, that wraps it up for our summaries. Now for our feature discussion.
Heart failure with preserved ejection fraction or HFpEF, notoriously difficult diagnosis to make, but do we finally have a validated diagnostic algorithm for HFpEF? Oh, you have to listen to our conversation today. I am so proud and pleased and thrilled frankly to have with me today the corresponding author of the feature paper, and that's Dr Barry Borlaugug from Mayo Clinic in Rochester, Minnesota as well as editorialist Dr Walter Paulusus from VU University Medical Center in Amsterdam.
Thank you so much both of you for making it here. I want to dive straight into it. So, Walter, maybe could you please paint the background to this because you wrote I think the most highly cited diagnostic guidelines of HFpEF, but that was in 2007. Tell us how does today's paper take us forward?
Dr Walter Paulus: Thank you very much, Carolyn. It's quite an honor for me to give you comments about this paper, which I think is going to be a landmark event. Over the years we have seen multiple algorithms being proposed usually by professional societies like V and C or the American Society of Echocardiography for the diagnosis of HFpEF. The major drawback of all these algorithms is that they have never been validated in clinical practice. And the reason they have never been validated was that it was extremely difficult to establish a gold standard for HFpEF.
And Barry was so clever to already invest in an establishing a gold standard for HFpEF ten years ago, and he very vigorously subjected all his patients in whom he suspected HFpEF to an invasive stress test and could establish the diagonals of HFpEF using this as a gold standard. And then he used all these consecutive patients with subsequently used to devise some form of an algorithm that was immediately validated against a gold standard. I think this has been a giant leap forwards. And again, I want to congratulate him with this unique endeavor.
Dr Carolyn Lam: Barry, I want to echo Walter's words and congratulate you. Now, has it really been ten years in the making? Tell us about this, Barry.
Dr Barry Borlaug: It has. In fact, it was 12 years ago when we started doing this, in 2006. But, yeah, these patients were examined in our laboratory between 2006 when I joined the staff at the Mayo Clinic to 2016. And really just doing this work up, we kind of started out doing it on a few patients and then we realized how powerful the methodology was. We did the invasive exercise testing with hemodynamics and a larger number of patients and just through accumulating a large number, as Walter points out, with a gold standard assessment this allowed us to then determine which less invasive attributes could be used to identify the likelihood that heart failure was the diagnosis.
Dr Carolyn Lam: That's so great. But you know beyond just that it is such a precious data set and so on, your paper is just so beautifully written and so clinically applicable. You've got this HFpEF score now for diagnosis. Everybody's going to be talking about it. So tell us about it. What does HFpEF score? What makes you think it'll work? How do you apply it clinically?
Dr Barry Borlaug: Thinking about diagnosis a lot, you really have to go back to [00:19:19] thinking, estimating the probability of disease, and when you're able to do that then you can find people where you need to perform really more invasive testing like the exercise testing. So really, we started like we need to have a better way to define who needs that more expensive and invasive evaluation. So we have this large cohort of patients, over 500 patients, 414 in the initial cohort, and then another 100 in the validation cohort. And they had all undergone this work up, they'd all undergone very detailed clinical evaluation and pheno typing. And we hypothesized which characteristics we thought would be most relevant. And then we did logistic regression to identify all the predictors.
There were many things that are associated that you would expect with HFpEF, but there were only 6 factors in the end in a multi-variable model that were all independently associated. That provided the most parsimonious sort of model or score that we could develop. We included these six different variables. So there's two for letter H- heavy and hypertensive, and by heavy we define that as a body mass index above 30. Hypertensive is defined as two or more antihypertensive medicines. The F in the H2 HFpEF score is atrial fibrillation, either paroxysmal or persistence a. Fib. The P is for pulmonary hypertension as estimated by echo with an estimated PA systolic pressure on echocardiography of 35. We wanted all of these to be noninvasive criteria for this score. E is for elder. I specifically didn't call it elderly because that can be a pejorative term and its only 60 years which is not that old. So E is elder. And F is for filling pressures, again estimated by echo doppler cardiography as an EE prime ratio greater than 9.
All of the scores are not one point each. They were arranged based on the strength of correlation in the logistic model. So being obese, having a BMI above 30 was awarded two points because it has a strong correlation. Being in atrial fibrillation or having a history of atrial fibrillation was even stronger at three points. If you tally these up, the score can range from 0-9, and based on that score you can then estimate a probability that HFpEF is present, if you're evaluating a patient that meets the entry criteria of the study, which is basically normal ejection fraction, and exertional breathlessness.
Dr Carolyn Lam: Nice. Okay, Walter, I think I can safely say that you have been thinking about this syndrome longer than either Barry or I. So I'd love your perspectives on how do you think this will be put into practice clinically perhaps, and where is the key area that it will change practice compared to perhaps the old diagnostic algorithms were like?
Dr Walter Paulus: I think this is a very important point, Carolyn. I think this score is so easy to handle and it is so well validated that we can now go to general practitioners and cause a general awareness for the disease. What vies me is that many patients are still unreported. The reason is that general practitioners and even general internal medicine people do not realize the [00:19:19] heart failure with preserved ejection fraction. Now with this score at hand, we can convince them that there needs to be an awareness when they see people that have value higher than six on the score, that they should be suspicious of heart failure being part of the symptomatology. I think this score mainly has its usefulness for general practitioners and general internal medicine.
Apart from the score, and it's more up to Barry to comment on this, but I want to highlight also, that he did not only develop the score, but he also had these very beautiful nomograms which is more of a find than a score, where he treated the variables in a continuous way. I think this is fairly useful for cardiologists and especially for people who want to have acute patients into trials because here we now have a very refined scale that goes from 0-160 and that allows you again to see what type of population you are addressing, what type of patients you are seeing that eventually what type of patients you are recruiting. I think for me the HFpEF score is of importance for general practitioners, general internal medicine, and especially I think we should also promote the nomogram. The nomogram, I think, are so refined that it would be useful tool, I think an excellent tool, for includement into trials.
Dr Carolyn Lam: Oh wonderful. Both of the simplicity and the cleverness, if I may, of this paper are precious to generalists and cardiologists. But Barry, I do have a couple of questions for you. Both you derivation and validation were in Olmsted if I'm not wrong. Now how am I supposed to apply it to my skinny HFpEF patients in Asia or elsewhere?
Dr Barry Borlaug: That's an important point, Carolyn. And it's a limitation of the paper. The people in Olmsted County, MN are not the same as they are in other parts of the United States or other parts of the world. I think that additional evaluation and other cohorts are important. We did the best we could with what we had. We did look at the patients carefully at Mayo Clinic. People think of it as quaternary referral center, but a pretty substantial number of the patients are from the local area, I think about 2/3 of them were. And when we looked in a subset in a sensitivity analysis of the people that were more local practice rather than coming from large academic medical centers, the HFpEF score, or as Walter pointed out, the continuous HFpEF model performed equally well. When we looked at people with so-called advanced HFpEF so high hemopressures at rest versus people at so called early stage HFpEF the people that have normal hemodynamics at rest but elevation during exercise. The model also worked well in that cohort.
But, like most studies that come out of where I work in southeastern Minnesota, it is mostly Caucasian people, the mean BMI was in the low 30s. So we need to look at other populations to make sure this works elsewhere as well.
Dr Carolyn Lam: Barry, let it go on record that I am your biggest fan. So thank you so much for this. I was just thinking even in other populations where the mean BMI may be lower for example here in Asia, we still definitely see an association with a higher BMI albeit at a lower cutoff with the presence of HFpEF. So it does raise this issue of do we need to maybe calibrate the score differently in different geographies or ethnicities. But that's not by any way take away from the tremendous input that you've made.
One other question is also the strength of atrial fibrillation in impacting the score. What are your thoughts on the possibility of misdiagnosis for example atrial fibrillation as HFpEF or the similar situation since they share symptomatology?
Dr Barry Borlaug: This is a great point, Carolyn. People still sort of argue about this. Somebody has breathlessness and effort intolerance and atrial fibrillation. Some doctors say they have symptomatic atrial fibrillation, but when we put catheters when we take these patients to the so-called table of truth and put catheters in and exercise them, we see hemodynamic arrangements that are diagnostic of heart failure. This led us to believe that this isn't just symptomatic a fib. It's really HFpEF. And that's why they have a fib. We published a paper earlier this year in circulation, more of a brief report, on the association between atrial fibrillation and HFpEF where we first reported this. That if you have normal EF, and especially permanent atrial fibrillation, you can pretty much take it to the bank that the patient probably does have heart failure with preserved ejection fraction, at least in the way that we have sort of defined it and the way that [00:19:19] initially defined it as an inability of the heart to pump blood adequately at normal filling pressures.
These patients almost all have that criteria for cardiac failure. I think that it is a really strong indicator and we probably are really just like in the general clinics, under recognizing HFpEF. I think probably in other clinics where people have atrial fibrillation and effort intolerance, we're again really under recognizing HFpEF in these people.
Dr Carolyn Lam: Indeed, and it's actually very consistent with Walter, your recommendations where atrial fibrillation played a big part too. Do you have any thoughts or advice?
Dr Walter Paulus: My idea is that atrial fibrillation and HFpEF are both manifestations of the same underlying process, which is systemic inflammation because of a metabolic disturbance. We used to think of atrial fibrillation as a consequence of left atrial dilatation, which itself was caused by the high filling pressures. I think that this does not hold, there is more to it. I think the atrium is as sick as the left ventricle and it undergoes similar pathological changes. That's why the presence of a fib becomes such a strong determinant of the presence of HFpEF in Barry's H of HFpEF score. All of this makes a lot of sense to me.
I just want to add something else. You spoke about the Asian population having less BMI and already having HFpEF. I think if you look at Barry's variables in uni-variant analysis, there's one which was presence of diabetes or prediabetes which did not make it in the multi-variant analysis on 0.06. It's my belief that if you got to the Asian population, that probably the BMI could be replaced with the presence of prediabetes and diabetes. Usually the insulin tolerance or insulin resistance is presence and the BMI is still low. I think there is need for some fine tuning, maybe in Asian populations, and I think this should be a challenge to go ahead with it. In fact, I'm leaving for Japan the day after tomorrow and I'm going to show the slides of Barry's paper. I'm going to try to set something up to also validate the score in Japanese populations.
Dr Carolyn Lam: We've got our work cut out for us, Barry! Let's get on to this too in southeast Asia.
Dr Barry Borlaug: I totally agree with Walter. I think that's great. And Carolyn, you, in a lot of papers, point this out, that the metabolic, cardio-metabolic associated with excess body mass, the way we define it with BMI, is shifted way down in southeast Asian population, and south Asian population, so I would agree with Walter's hypothesis that diabetes, prediabetes maybe that's the better way to go when we look at this in other patient populations.
Dr Carolyn Lam: You both absolutely made my day with this discussion today. Thank you so much. What a thrill to be on the same podcast with the people I admire most.
Listeners, I know you enjoyed this as much as I did. Don't forget to tune in again next week.