Circulation January 3, 2017 Issue

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.
Caroline Lam, associate editor from The National Heart Center and
Duke National University of Singapore. Today we will be
discussing the results of an individual level meta-analysis
regarding venous thromboembolism and its risk factors, but first,
here's your summary of this week's issue.


The first paper provides insights into paracrine signalling
pathways that regulate epicardial adipose tissue formation. That
is, referring to the adipose tissue located between the
epicardium and underlying myocardium that is known to be strongly
associated with coronary artery disease. In the current study
from Dr. Lira of Icahn School of Medicine at Mount Sinai, New
York, Dr. Pu from Boston Children's Hospital, Dr. [Chien
00:00:56] from Karolinska Institute and colleagues, the authors
used a novel modified mRNA screening approach to probe the effect
of individual paracrine factors on epicardial progenitors in the
heart. Using two independent lineage tracing strategies in murine
models, they showed that cells originating from the WT1-positive
mesothelial lineage, which includes epicardial cells,
differentiate into epicardial adipose tissue following myocardial
infarction. This differentiation process required WT1 expression
and was stimulated by insulin-like growth factor 1 receptor
activation. Insulin-like growth factor 1 receptor inhibition
significantly reduced its adipogenic differentiation and reduced
WT1 lineage cell differentiation into adipocytes following
myocardial infarction.


These results thus establish insulin-like growth factor 1
receptor signalling as a key pathway that governs epicardial
adipose tissue formation in the context of myocardial injury. And
it does this by redirecting the fate of WT1-positive lineage
cells. The study also demonstrates the utility of a modified RNA
based paracrine library screening to dissect signalling pathways
in homeostasis and disease.


The next study brings us closer to understanding the mechanisms
underlying diabetes-associated heart failure. In this study by
first author, Dr. Wang, corresponding authors, Dr. Abel and Xiang
from University of California, Davis and colleagues. High-fat
diet feeding was used to induce obesity and diabetes in wild-type
mice or mice lacking the beta-2 adrenergic receptor or
beta-arrestin 2. High-fat diet feeding was found to selectively
increase the expression of phosphodiesterase 4D in the mouse
hearts in concert with the reduced phosphokinase A
phosphorylation of phospholamban which contributed to systolic
and diastolic dysfunction. The expression of phosphodiesterase 4D
was also elevated in human hearts with diabetes. The induction of
phosphodiesterase 4D expression was mediated by an insulin
receptor and substrate as well as by beta-arrestin-2 dependent
activation of a beta adrenergic receptor, ERK signalling cascade.


Genetic deletion of beta-2 adrenergic receptor or beta-arrestin-2
or pharmacological inhibition of beta-2 adrenergic receptor with
carvedilol or G-protein receptor kinase 2 with paroxetine all
significantly attenuated insulin-induced phosphorylation of ERK
and phosphodiesterase 4D induction thus preventing
diabetes-related systolic dysfunction. Thus, targeting the
insulin beta-2 adrenergic receptor pathway may be a novel way to
prevent diabetes-associated heart failure.


The next study addresses the gap in care pertaining to
implantable cardioverter-defibrillator or ICD use among Medicare
patients with low ejection fraction following myocardial
infarction. Dr. Pokorny and colleagues from Duke University
Medical Center examined rates of post-discharge ejection fraction
assessment and ICD implantation among more than 10,280
Medicare-insured patients age 65 years above with an ejection
fraction 35% and below during an index myocardial infarction
admission in the ACTION Registry Get With the Guidelines. They
found that the cumulative incidence of ejection fraction
reassessment within one year of myocardial infarction was 66.8%.
Within the first year of post-myocardial infarction, 11% of
patients who had an ejection fraction reassessment underwent ICD
implantation which was significantly higher than patients without
an ejection fraction reassessment. After multivariable
adjustment, ejection fraction reassessment remained significantly
associated with a higher likelihood of ICD implantation within
one year in both revascularized and non-revascularized patients.
Based on these findings, the authors recommend that all patients
who are potential candidates for ICD therapy be scheduled for
follow-up outpatient ejection fraction assessment prior to
hospital discharge to bridge these currently observed gaps in
care.


The next study is the first multi-institutional study in Asia
describing current treatment strategies for total anomalous
pulmonary venous connection. This retrospective study of 768
patients with total anomalous pulmonary venous connection
operated on between 2005 and 2014 is from first authors Dr. Shi,
Zhu, and Chen, corresponding authors, Dr. Chen and Zhuang and
colleagues from the Shanghai Children's Medical Center and
Guangdong General Hospital in China. While most patients
underwent conventional repair, a sutureless patient was technique
was employed in 10% of patients. Over a median follow-up of 23
months, there were 38 intraoperative deaths and 13 late deaths. A
younger age at the time of repair, next an infracardiac total
anomalous venous connections, pre-operative pulmonary venous
obstruction, prolonged cardiopulmonary bypass time and longer
duration of ventilation were all factors associated with
increased mortality. Among these 717 survivors, recurrent
pulmonary venous obstruction was found in 15% or 111 patients.
Risk factors for recurrent pulmonary venous obstruction included
pre-operative pulmonary venous obstruction, infracardiac total
anomalous pulmonary connection, mixed venous connections and
prolonged cardiopulmonary bypass time, a sutureless technique was
associated with a lower restenosis rate compared to conventional
repair in patients with pre-operative pulmonary venous
obstruction but not in newborn patients. Thus, this study
provides an important data on the outcomes following surgical
correction and risk factors for poor prognosis in total anomalous
pulmonary venous connection in Asia.


The final study is the first systematic review and meta-analysis
on the association of genetic polymorphisms and outcome of
clopidogrel-treated patients with ischemic stroke or transient
ischemic attacks. In this paper from first author, Dr. Pan,
corresponding author, Dr. Wang and colleagues from Beijing
Tiantan Hospital, Capital Medical University in Beijing, China.
Authors looked at 15 studies of 4,762 patients with stroke or
transient ischemic attack treated with clopidogrel and this
included 3 studies from Europe and 12 studies from East Asia.
They found that carriers of the CYP2C19 loss-of-function alleles
were at increased risk of stroke compared to noncarriers.
Composite vascular events were also more frequent in carriers
compared to noncarriers while bleeding rates were similar. There
was no evidence of statistical heterogeneity among the included
studies for stroke but there was for composite vascular events
suggesting that publication bias cannot be ruled out. Genetic
variance other than CYP2C19 were not associated with clinical
outcomes. The author suggested that their findings may justify
genetic testing when clopidogrel is otherwise considered the
preferred treatment modality, especially in East Asian patient
populations in whom the prevalence of CYP2C19 loss-of-function
allele is high.


In an accompanying editorial, Dr. Simon and [inaudible 00:10:11]
suggest it maybe time to consider a prospective trial of
personalized medicine using CYP2C19 genotyping in acute ischemic
stroke and perhaps considering alternative medications in poor or
intermediate metabolizers such as in the popular ongoing genetics
trial in STEMI patients undergoing PCI. That wraps it up for the
summaries this week. Now for our feature discussion.


Today's feature paper talks about the association of traditional
cardiovascular risk factors with venous thromboembolism. And it
is the first individual level meta-analysis of prospective
studies. I am so delighted to have the first and corresponding
author here with us, Dr. Bhaktawar Khan Mahmoodie from San
Antonio's Hospital in the Netherlands. Hi Khan, thanks for being
here.


Dr. Bhaktawar Khan
Mahmoodie:            


Thank you for inviting me. Thanks a lot.


Dr. Carolyn
Lam:              


And I am particularly delighted to have associate editor, Dr.
Josh Beckman from Vanderbilt University joining us today as well.
Welcome Josh.


Dr. Josh
Beckman:          


Caroline, it is such a pleasure to be here with you. I've been
listening to these podcasts and they have been incredible. I've
been waiting to be able to jump in and today's paper is an
awesome place to start.


Dr. Carolyn
Lam:              


It certainly is. Congratulations on managing such an important
paper. Khan, maybe I could start with you. Venous thromboembolism
versus arterial thromboembolism. We're very familiar with the
latter. We know it comprises coronary heart disease, stroke,
peripheral artery disease. We're very familiar with the risk
factors such as hypertension, hyperlipidemia, diabetes, smoking.
But here you're asking, are these same risk factors applicable in
venous thromboembolism. That would include deep venous
thrombosis, pulmonary embolism, where we traditionally classify
it into provoked events that is triggered by things we know well
like immobilization, surgery and so on. And then there are the
unprovoked events that don't have any risk factors. So could you,
first of all, point out ... you were looking at venous
thromboembolism. What was your hypothesis with regards to the
traditional cardiovascular risk factors?


Dr. Bhaktawar Khan
Mahmoodie:            


Many researchers in the last 10, 15 years, they go questions
whether there is connection between venous and arterial
thromboembolism. Since then, several studies published on that
with controversial results. So our hypothesis for this paper was
to see whether there is real associations or are we looking at
some kind of associations due to confounding factors such as age
and overweight which are risk factors for both.


Dr. Carolyn
Lam:              


Yeah. And yours is actually the first individual level
meta-analyses of prospective studies dealing with this. Tell us
what you found in ... Were you surprised by your findings?


Dr. Bhaktawar Khan
Mahmoodie:            


What we found that actually traditional, modifiable,
cardiovascular risk factors like hypertension, diabetes and
hyperlipidemia were not risk factors for venous thromboembolism.
The exception was smoking, current smoking, which was
particularly associated with provoked venous thromboembolism
which is probably pro its association with cancer. And cancer
itself is a strong risk factor for venous thromboembolism. About
whether I was surprised, I was not surprised at all. We saw in
several cohort studies and well-defined cohort studies that the
association disappeared after adjustment for age and body mass
index which are important confounders in these [inaudible
00:14:06]. That's what I expected and we found it and it is
confirmed with this large individual level meta-analysis.


Dr. Carolyn
Lam:              


Great. But what did you think of the association of higher
systolic blood pressure not with higher but with lower risk of
venous thromboembolism?


Dr. Bhaktawar Khan
Mahmoodie:            


That was a bit surprising for us too but I think the best
explanation we can give at the moment is probably that we have
some kind of competing risk. And one suggestion that we gave in
the paper as well is that maybe what we already know is that
higher blood pressure is a strong risk factor for atrial
fibrillation. Most of these people they receive oral
anticoagulants. That is subsequently probably a protective factor
for venous thromboembolism. We probably deal with some kind of
competing risk from another condition like the atrial
fibrillation and use of anticoagulants which we could not
unfortunately adjust for in this analysis.


Dr. Carolyn
Lam:              


Sure. That makes sense. Josh, can I bring you into this? I mean I
remember well our multiple and long discussions at the editors
meetings. This is one of those papers that is extremely important
for its negative, neutral associations isn't it?


Dr. Josh
Beckman:          


I think this is one of the more important papers in this field in
a long time. I am one of those people who has followed this
literature and believed, based on the best previous publications,
that there was a link between many of the arterial thrombosis or
atherothrombotic risk factors and venous thromboembolism. In
fact, Circulation published one of these meta-analyses, and I'm
going to say only because this little paper is so large with only
21,000 patients demonstrating a clear association. So the first
question I would have, we published that back at 2008, the first
question I would have is can you describe for the general
readership what such a large series of patients allows you to do
that was not permitted by the other meta-analyses of say twenty
to thirty thousand patients that have been previously in the
literature.


Dr. Bhaktawar Khan
Mahmoodie:            


Thank you Dr. Beckman and thank you also for managing this paper.
This is an important question and I think what we were able to do
compared to the previous analysis in 2008, we were able to adjust
for confounding risk factors. In the course, we included were all
with validated venous thromboembolism events and also the events
are temporal character, like all the risk factors were measured
and then followed-up for event. While in that paper, there were
many case-controlled studies added and the results were not
adjusted for age and also not adjusted for body mass index. And
if we do the same with what's done there, then we have the same
results like in our [inaudible 00:17:14] associations, all of
these risk factors were indeed positively associated with risk
for venous thromboembolism.


Dr. Carolyn
Lam:              


Let me just state, I mean, there were almost 245,000 participants
in your study. With 4,910 thromboembolism events, so this is
really huge and gives you a lot of power to look at this thing
very carefully.


Dr. Josh
Beckman:          


It was a 10-fold increase from any of the major publications in
this area. It was almost geometrically larger in size which is
why, I think its conclusion will be accepted differently than all
the previous analyses. Now, let me ask one question about what
you already identified in your discussion as a possible
limitation. Is this study applicable to all populations around
the globe or do you think it is a bit more focused?


Dr. Bhaktawar Khan
Mahmoodie:            


I think it is focused at least. We don't have Asian population in
these analyses and also the proportion of African-Americans were
limited which was only limited to some U.S. cohorts so I think
that there is a limitation which is results are probably only
applicable for Caucasian population.


Dr. Josh
Beckman:          


I guess my other question is, one of the reasons that people, I
think, advance the argument that there may be overlap between the
two kinds thrombosis is that there was evidence that the
medication, statin, may ... to a much smaller degree, reduce
venous thrombosis as well as reducing arterial thrombosis. Do you
think that this is evidence of some common pathophysiology? Or is
it like smoking, it's truly working separately from arterial
disease?


Dr. Bhaktawar Khan
Mahmoodie:            


Personally, I think that this association or the finding of
statins reduce the risk of thromboembolism could be due to some
pleiotropic effects of statins. Like even for stroke, we know
that the association of cholesterol with stroke is not so
clear-cut as it is with myocardial infarction but still it
reduces risk of stroke. And also for venous thromboembolism, the
risk reduction of venous thrombosis in the JUPITER trial was like
50%, which is very high, even better than aspirin. But I think
that may not be directly related cholesterol levels but more to
another pleiotropic effects of statins. It could influence levels
of various coagulation [inaudible 00:19:56] in the endothelial
stabilization which may be also important risk factors for venous
thrombosis.


Dr. Josh
Beckman:          


One of the reasons that this paper is very important is that we
begin to look for therapies and risk factors based on what the
disease is caused by. And so the fact that you guys were able to
establish, in my opinion, quite clearly what does and what does
not contribute to venous thrombosis allows us to begin to think
about the disease differently and approach it differently. I
would like to provide congratulations. My one little ask of you
is that one of the things that I think this podcast is great for
is to explain to the readership what goes into this kind of work.
Everybody thinks that someone else's research is easier to do
than their own, which of course is a ridiculous thing. But can
you describe for us what it's like and how long it took from the
study initiation to when you completed it? How much work went
into trying to get all these studies together to create this
individual patient level data?


Dr. Bhaktawar Khan
Mahmoodie:            


Yeah. That was a great amount of work. Actually, I did a
systematic review of the only PubMed publications back in 2014
and it took almost 2 years at least. I was not always active the
whole 2 years but still I had to visit several PIs of the studies
to get them so far to share their data. Eventually, I had to
develop a code that will make it possible without sharing the
individual level data by using the same definitions and the same
categorization of variables so we call it a two-stage
meta-analysis similar to one-stage if the definitions are
similar. And eventually, I think that the real part of the
analysis and inclusion of studies took like half a year or so.
There was a lot of work.


Dr. Josh
Beckman:          


I think this is a tremendous amount of work and for those members
of our readership who do basic research, or translational work,
or practice in the clinics, it really needs to be made clear that
this is a heroic effort of hundreds and hundreds of hours. And
that getting together all of these studies is just an enormous
undertaking. And that even though, we can read the paper in 10
minutes and gleam the most important part. It is an incredible
amount of work for which you guys are to be congratulated.


Dr. Bhaktawar Khan
Mahmoodie:            


Thank you for acknowledging this. Thanks a lot.


Dr. Carolyn
Lam:              


Josh, I couldn't agree with you more and I truly couldn't have
said it any better. Thank you both of you for making this just
one of the best discussions we've had on this podcast. I'm sure
the listeners all agree what a wonderful time we've had.


You've been listening to Circulation on the Run. Please remember
to tune in next week.