Circulation September 7, 2021 Issue

This week's episode features special Guest Host Mercedes
Carnethon, as she interviews author Sung-Min Cho and Associate
Editor Marc Ruel as they discuss the article "Cerebrovascular
Events in Patients with Centrifugal-Flow Left Ventricular Assist
Devices: A Propensity Score Matched Analysis from the Intermacs
Registry."


Dr. Carolyn Lam:


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


Dr. Greg Hundley:


And I'm Dr. Greg Hundley Associate Editor, Director of the Pauley
Heart Center at VCU Health in Richmond, Virginia. Well, Carolyn,
this week's feature, we're going to look at centrifugal flow,
left ventricular assist devices and cerebrovascular events. But
before we get to the feature, how about we grab a cup of coffee
and jump into some of the other articles in the issue? And maybe
how about I go first?


Dr. Carolyn Lam:


All right. I got my coffee.


Dr. Greg Hundley:


So my first paper comes from Professor Dali Luo from Capital
Medical University. And it's pertaining to calsequestrin-1. So
calsequestrin-1, and calsequestrin-2 isoforms buffer calcium and
regulate its release from the sarcoplasmic reticulum of skeletal
and cardiac muscle. Human inherited diseases associated with
mutations of calsequestrin-1 or 2 include malignant hyperthermia
and environmental heat stroke and catecholamingergic polymorphic
ventricular tachycardia. However, patients with hypothermia,
environmental heat stroke events often suffer from an arrhythmia
for which the underlying mechanism remains unknown.


Dr. Carolyn Lam:


Wow. Okay. And what did the current paper do and find?


Dr. Greg Hundley:


Great, Carolyn. So what the authors found, calsequestrin-1, the
skeletal isoform of it is indeed expressed in cardiomyocyte
sarcoplasmic reticulum for mirroring in human hearts, mostly
presenting as a polymeric form and interacting with the ryanodine
2 receptor in ventricles. Second, calsequestrin-1 deficiency
cause sinus tachycardia in basal conditions. And this is a novel
finding which may be associated with sinus beat regulation and
ventricular arrhythmia as an independent arrhythmogenesis if a
high concentration of volatile anesthetics are used. Next, these
volatile anesthetics and heating to 41 degrees C can directly
induce calsequestrin-1 oligomerization, thereby causing
enhancement of diastolic calcium leak and premature calcium
transience through a reduced regulatory effect of calsequestrin-1
on ryanodine 2 activity. And so Carolyn, this novel mechanism
underlying the arrhythmia occurring in patients with malignant
hypothermia or environmental heatstroke episodes may provide
different strategies for heart disorders as an independent
profile in these syndromes. And finally, the finding of
calsequestrin-1 confirmational change induced by triggers in
those with malignant hyperthermia and environmental heatstroke
could lead to novel therapeutic approaches to prevent these types
of episodes. And that may also very, very useful in treatment of
heatstroke.


 


Dr. Carolyn Lam:


Wow. Thanks Greg. Well, moving from this preclinical world to a
very common clinical question of the diagnosis of acute
myocardial infarction. Now we know that in patients presenting to
the emergency department with symptoms suggestive of an MI, the
European Society of Cardiology zero and one hour algorithm is
recommended by current ESC NSTEMI guidelines with a class one
recommendation. Now, what this does is it combines a very high
safety for early rule-out and high accuracy for rule-in allowing
a definite triage of about 70 to 75% of patients using the zero
in one hour sample.


Dr. Carolyn Lam:


However, what is the most appropriate management of the 25 to 30%
of patients who remain in the gray observed zone? So this is the
question that the current paper addresses. Now to answer this, we
also need some more background that a single center pilot study
previously of patients in the observed zone had derived a cutoff
of seven nanograms per liter for a zero and three hour high
sensitivity cardiac troponin T change to identify patients also
eligible for early rule-out or rule-in of NSTEMI. So the current
study that we're talking about in today's issue from Dr.
Christian Mueller from Cardiovascular Research Institute in
Basil, Switzerland, and colleagues, really aimed to externally
validate that previously proposed seven nanogram per liter change
cutoff, and if necessary derive and internally as well as
externally validate some new criteria for these patients in the
observed zone of the ESC zero in one hour algorithm.


Dr. Greg Hundley:


Wow, Carolyn, so we're learning a lot about cutoff values and
also algorithms here with high sensitivity cardiac troponin T. So
what did they find here? Very interested to hear.


Dr. Carolyn Lam:


So in two large prospective multicenter diagnostic studies, they
found that the proposed zero and three hour high sensitivity
cardiac troponin T change of seven nanogram criteria,
unfortunately provided suboptimal safety for ruling out NSTEMI in
patients remaining in the observed zone of the ESC zero and one
hour algorithm. So this had a sensitivity of only 33% and missed
80 patients with NSTEMI. So they derived their own novel criteria
based on zero and three hour samples. And these novel criteria
combined a three hour high sensitivity cardiac troponin T
concentration of less than 15 nanograms per liter and a zero and
three hour absolute change cutoff of four nanograms per liter.


Dr. Carolyn Lam:


And that combination provided a high safety for ruling out NSTEMI
in these patients in the observed zone and with a sensitivity of
99% missing only one patient with NSTEMI. Another further thing
they found was at a zero and three hour cardiac troponin T
absolute change of greater or equal to six nanograms per liter
triage, 63 patients, or 11% towards rule-in thus resulting in a
specificity of 98%. So in summary, this novel criteria based on
zero and three hour sample seemed to balance safety and efficacy
well for the further decision making in patients who are
remaining in the observed zone after the zero and one hour
cardiac troponin T algorithm. Internal validation of these novel
criteria and external validation in an independent international
cohort showed robustness of performance metrics and further
strengthen its possible clinical use.


 


Dr. Greg Hundley:


Very nice, Carolyn. Lots of data there, and hopefully very
important clarification on both the zones as well as the cutoff
values for using cardiac troponin T. Well, Carolyn, my next paper
again comes from the preclinical science world and it's from Dr.
Anne Eichmann at Yale University School of Medicine, and it
pertains to activin receptor-like kinase 1. And we're going to
call that ALK1.


 


Dr. Greg Hundley:


Kinase 1 and we're going to call that ALK1. And it's an
endothelial transmenbrane serine threonine kinase receptor for
BMP family ligands that plays a critical role in cardiovascular
development and pathology. And loss of function mutations of the
ALK1 gene cause type 2 hereditary hemorrhagic telangiectasias, a
devastating disorder that leads to arteriovenous malformations.


Dr. Carolyn Lam:


Oh, okay. And what did the authors find?


Dr. Greg Hundley:


Dr. Carolyn Lam, ALK1 mutants displayed defective polarization
against the direction of blood flow in capillary and venous
endothelium as well as increased integran VEGF receptor 2
mediated P13K activation of YAP/TAZ signaling.


Dr Carolyn Lam:


Okay, Greg, that was super summarized but what are the clinical
implications?


Dr. Greg Hundley:


Carolyn, pharmacological integrin inhibition using cilengitide or
ATN-161, or YAP/TAZ inhibition using verteporfin, prevented AVM
malformation in ALK1 mutant mice. And therefore for this study,
the authors revealed that integrin and YAP/TAZ were novel
affectors of ALK1 signaling in AVM pathogenesis that might be
targeted for AVM treatment in patients with hemorrhagic
telangiectasias.


Dr. Carolyn Lam:


Thank you, Greg. Well, let's review what else is in today's
issue. There's an exchange of letters between Doctors Amadio and
Valentine on cell-free DNA to detect heart allograph acute
rejection. There's an AHA Update paper by Dr. Churchwell on
preemption, a threat to building healthy, equitable communities.
There's a Research Letter by Dr. Merkler on the association
between cervical artery dissection and aortic dissection.


Dr. Greg Hundley:


And Carolyn, I've got a paper from Professor Daniels regarding
the Clinical Implications of Basic Research getting inside the
engine, the myosin modulation of hypertrophic cardiomyopathy and
systolic heart failure. And then finally, there's an In Depth
piece from Dr. Viskin entitled, “Polymorphic Ventricular
Tachycardia: The Terminology, mechanism, diagnosis and Emergency
Therapy.”


 


Dr. Carolyn Lam:


Nice. Well, let's go on to our feature discussion. Can't wait.


Dr. Greg Hundley:


You bet.


 


Dr. Mercedes Carnethon:


Welcome to this episode of Circulation on the Run, our podcast
where we have an opportunity to talk with the authors of some of
the top articles within our journal for a given week. And we've
chosen today to focus on a set of articles, one of which is led
by Dr. Sung-Min Cho from the Johns Hopkins University. And I'm
really excited to have you with us today, Dr. Cho and joining us
as well as the associate editor, Dr. Marc Ruel who handled the
paper. And my name is Mercedes Carnethon from the Northwestern
University's Feinberg School of Medicine. I guess without further
ado, welcome to you both and we'll just jump right into it.


Dr. Mercedes Carnethon:


Dr. Cho, I'd love to hear a little bit more about your paper
today. What made you choose to pursue this particular topic and
what really inspired you?


Dr. Sung-Min Cho:


Thank you so much for the invitation and opportunity to talk
today. During my training as a neuro person, I'm a
neurointensivist by training and neurologist. I noticed that we
are getting a lot of consults for LVAD associated strokes. When I
took a closer look at the ENDURANCE trial, very showed really
29.7% stroke rate at two years and a few years later, we had this
MOMENTUM 3 trial, which showed HeartMate 3 device had 10% stroke
rate at two years. And we realized that a stroke is a major issue
in this population and I wanted to study the incidence respecters
and outcome of this strokes in LVAD population. However, despite
the many observational studies in the past, we were really
interested in looking at device specific stroke risk for current
continuous flow LVADs and we wanted to look at the device
specific risk and prevalence of these patients balancing
co-morbidities each cohort. And that's why we conducted this
study.


Dr. Mercedes Carnethon:


Great, well Sung-Min, it's not often that as an epidemiologist
and cardiovascular epidemiologist that I actually get to talk
with neurointensivists and get their insights on the importance
of their work. Can you tell me a little bit about what you found
and whether it surprised you?


Dr. Sung-Min Cho:


Population, we used the Intermacs registry database. This is well
established database as all cardiologists and cardiothoracic
surgeons know, and we defined a neurologic adverse event as
stroke plus TIA, transient ischemic attack. We used a propensity
score matching analysis to assess the association of HVAD with
stroke risk, to balance for pre-implant risk factors. And
basically after performing propensity score matching, we found
that hazard of stroke was higher for patients with HVAD device
compared to HeartMate 3. We kind of expected this based on the
randomized control trials in the past but there was no head to
head comparison between these two cohorts. This study really
confirmed our suspicion that HeartMate 3 actually had lower
hazard of a stroke compared to HeartMate 3.


Dr. Mercedes Carnethon:


Well, thank you so much. It's a really great explanation. And for
those who haven't had a chance to dig into the issue yet, I
really encourage you to read the piece. I found it to be very
instructive. And I'm interested as well, Mark in your take about
what excited you about this piece.


Dr. Marc Ruel:


Well, thank you very much Mercedes and Sung-Min it's really a
pleasure to have you with us today. As you know, this has been a
very impactful paper and you were very kind to share with us the
study around your idea as to why you wanted to evaluate this
question but even more than your idea and what led to the
completion of the paper are the implications of your paper. And I
think it would be great if you shared with us a little bit, what
has been the path that your paper has led to and including
amongst others, very likely a decision by the Medtronic to pull
the HVAD out of market. It's interesting that your data, to my
knowledge, correct me if I'm wrong, were presented first at the
annual meeting of the Society of Thoracic Surgeons in January,
2021. And again, I want to reiterate that Circulation's very
thankful that you chose to send your paper to our journal and we
feel that it will give it full justice, like many other journals
of would have had but we're really excited to have received your
paper and give it the fullest consideration.


Dr. Marc Ruel:


Can you tell us a little bit about the implications and for lack
of a better word, the storm that your paper has created in the
field and your take on it?


Dr. Sung-Min Cho:


Right. That's a great question. Thank you for that. Like I said,
as a neurologist, we see these patients after complication,
patients having stroke and then we see these patients and we
always wanted, cardiologists and cardiothoracic surgeons and
neurologists, we always wondered which device carried more risk
for stroke and TIA. And really our group actually worked on many
papers in the past looking at single institutional data and also
systematic review meta-analysis looking at this topic, but really
HeartMate 3 came along a couple years ago, more recent device so
we didn't have a lot of data.


Dr. Sung-Min Cho:


So intermex registry really helped since we didn't have a lot of
data. So, INTERMACS Registry really provided opportunity for us
to look at this specific question, really balancing those two
chords to look at the risk of stroke in this HeartMate 3 and
HVAD. And when we did that two years ago, we submitted a proposal
to INTERMACS, and Dr. Kirklin from UAB, he really helped us to
look at this data closely with his statistical team. And we had
really a thorough statistical method to perform a propensity
matching analysis. And we finally finished the analysis and
presented in annual STS meeting in January, and it did really
trigger a lot of attention to a lot of academic institutions and
people who are practicing LVAD, and after that, when we finally
submitted this paper to Circulation, we had to have a lot of
discussion in between FDA and the Medtronic and discussing this
implication of this paper. When it was finally published in
Circulation, we are happy that there's a lot of attention and we
made it through.


Dr. Marc Ruel:


Well, thank you, Dr. Cho, and maybe for the listener of this
podcast, I would like to reiterate some of the salient points of
your paper essentially, and correct me if I'm wrong, over 6,200
patients were included, about roughly 3,000 patients per group
comparing the HeartMate 3 versus the HVAD.


Dr. Marc Ruel:


Now, as you alluded to the HVAD is the more ancient device, if
you will. So there's a slightly longer follow-up, around 12
months on median, versus nine months with the HeartMate 3. And
there's adjustment that has been made for this. And I think to
me, really the key finding is that in the early acute phase
around implantation, there is no real difference with regards to
the risk adjusted incidents of neuro adverse events. However,
once you pass the early implantation acute phase, in the chronic
stable phase, there starts being really a signal that is
detrimental to the performance of the HVAD versus the HeartMate
3. And I think your hazard ratio, correct me if I'm wrong, it's
around 5.7 for neuro adverse events.


Dr. Marc Ruel:


So this is a very compelling hazard ratio, even coming out of an
observational study with all the careful attention that you
provided to adjust for residual confounding, et cetera.


Dr. Marc Ruel:


So obviously this is a very strong finding, but I would like you
to perhaps comment on this, the patients are not the same.
There's some indication that the HVAD patients may have been a
little sicker, more RV dysfunction, more tricuspid regurgitation,
higher INTERMACS-1 incidents more often on ECMO prior to an
implant. What are your thoughts about this?


Dr. Marc Ruel:


Obviously, you've been very careful and the reader will note in
the paper that many attempts have been made to account for those.
But please give us your take around that 5.7 hazard ratio for
neuro adverse event that you found.


Dr. Sung-Min Cho:


Right? In fact, we were really being careful adjusting those
compounders. So we did a propensity matching has a primary
analysis, but as you pointed out, as a secondary analysis, we
wanted to look at multi-variable logistic regression analysis,
looking at multi-hazard analytics. And when we did the secondary
analysis, as you said, in the beginning early hazard period, the
risk was similar, as time went on in the constant hazard period,
the hazard ratio was 5.7 for HVAD compared to HeartMate 3, which
gives a much higher risk of stroke and TIA for those patients
with HVAD compared to HeartMate 3.


Dr. Sung-Min Cho:


So, that was really convincing to us. Confirming the findings
from propensity matching analysis, showing that same findings
were consistent throughout the different analysis. As we pointed
out, HVAD patients actually were sicker, they had more ECMO, and
they had more ventilation requirement or sicker patients
INTERMACS level. Those are all carefully balanced in both
propensity matching analysis and also multi-hazard analytics. And
both of these analysis consistently showed that HVAD carried more
risk of TIA and stroke compared to patients with HeartMate 3.


Dr. Mercedes Carnethon:


Thank you so much Sung-Min. You know what excites me as I think
about choosing articles for journal clubs, when we're working
with our trainees, the propensity matched approach and comparing
it directly with what you're getting from multi-variable
regression really provides an excellent methodological strategy
to be able to generate results from these real world studies
where it's not a randomized trial of who received which device,
but we're able to yield practical conclusions that are actionable
based on these findings when we have these well done analyses.
And Marc alluded earlier to the actions that were taken in
response to the findings from your study. Can you expand on those
just a little bit more?


Sung-Min Cho:


Of course. So I guess, I don't know the real backstory, what was
going on behind the scene, but I know for sure that STS
leadership and INTERMACS leadership, they had a lot of discussion
with the company who made HVAD device and also FDA, and I know
that this study, the results of this study contributed to the
decision they made back in June, pulling up HVAD device from the
market.


Sung-Min Cho:


So I'm glad that this study could contribute to the science and
hopefully this will help the patients in the future for device
selection. So yeah.


Dr. Marc Ruel:


Sung-Min, I think it's fair to say that your study is probably,
if not the most impactful in the field of ventricular assist
devices, and I probably would personally think that it is, if not
the single most impactful, certainly one of the two or three that
are the most impactful. So congratulations to you and your team.


Dr. Marc Ruel:


If you still have a minute or two, I had a couple of more
secondary questions?


Dr. Marc Ruel


In your analysis I noted that in the early acute phase, there are
some protective predictors, such as performing the LVAD implant
by sternotomy, which essentially results in about half of the
neuro adverse events that you would otherwise observe. So I was a
little intrigued by that. And high volume centers had about 1.8
hazard ratio. I suspect that's probably reflective of baseline
risk and more acute illness in those patients coming. But if you
have a chance, I'd love to hear your thoughts around this?


Dr. Sung-Min Cho:


Yeah, that's exactly what we thought actually is, initially we
thought, hypothesized that surgical volume, the center volume
will be associated with lower risk of stroke, but it was the
other way around. But as you said, probably higher volume centers
were getting sicker patients, so that's the association probably
we were getting in the analysis. And we wanted to adjust for
surgical techniques, sternotomy versus thoracotomy, and even
after adjusting for that, HVAD remained a significant hazard per
stroke, which showed in the table two and three, I think in the
manuscript.


Dr. Sung-Min Cho:


And if I may, I want to say these couple of things. In the raw
number, in the 6.4% of patients actually had TIA and strokes,
neurological adverse events in HeartMate 3, at one year based on
our study. And the risk goes up with a longer follow-up time of
course. Moment3 trials had two-year follow-up, about 10% had
stroke. And this is still, after HVAD is taken off the market,
still there's a significant risk for stroke in these patients and
based on autopsy and MRI studies although there is a very small
studies--MRI studies, although they're a very small series,
studies looking at MRI'd brains after explantation of LVAD. And
it shows actually more than 95% of patients have cerebral micro
bleeds, which is a marker for small vessel disease in the brain.
I think this is an important issue, and although we show that one
device had a lower risk of stroke, still question remains, are
these patients have a high risk of stroke? And there is a need
for improving biomedical engineering aspect, and I'm sure
cardiologists and cardiothoracic surgeons know much better than I
do regarding hemo-compatibility, especially for stroke.


Dr. Sung-Min Cho:


There is also a dire need for early detection and intervention
for these events to improve the outcome for these patients,
because once you have a stroke, the outcome is devastating,
right? So I think there needs to be better medical management,
neuroprotective agent, as well as neuro- monitoring methods,
maybe biomarkers to predict stroke or TIA to come so we can
intervene and prevent these really devastating complications.


Dr. Marc Ruel:


Mercedes, if I'm so allowed, I do have one final comment and
question.


Dr. Mercedes Carnethon:


Most definitely. This has been delightful, so yes.


Dr. Marc Ruel:


Wonderful. So, first, Sung-Min, I want to thank you for working
with us. We at Circulation were interested in your paper. You may
recall you and I spoke on the phone offline when the decision to
revise was made, and we went carefully over what the editors were
anticipating would make your paper even better. And you were very
responsive. You and your co-author's team were tremendous. And I
think the paper that we have before us is absolutely very, very
insightful and very important. And obviously tremendously
impactful. So I want to thank you again for that.


Dr. Marc Ruel:


And my question is probably the very difficult question which is
in everybody's mind at this point and I would like your take as a
neurointensivist. You have someone who you have to care for who
has a well-functioning HVAD, two years post implant. What would
you recommend in terms of optimization for the prevention of
neural adverse events? I realize we don't have all the
information, but you are one of the few experts in the world who
can probably provide us with a very valid take on this very
difficult question.


Dr. Sung-Min Cho:


Yeah, it is indeed a difficult question. And that's what I am,
including me a lot of neurointensivists, they are very interested
in this topic. I think really, as I alluded before, only
detection is really important, but it's really tough because
either patients, they cannot get MRI. There's no way to know
who's going to have stroke or not.


 


Dr. Sung-Min Cho:


We know that a bacteremia is a huge risk factor for these
patients. Whenever they have device infection, dry valve
infection, bacteremia, their stroke risk goes up quite a bit. We
have a lot of data on that. So we can carefully monitor these
patients, follow these patients. There is some data that, within
six days from infection, their stroke risk goes quite high up for
these patients.


Dr. Sung-Min Cho:


But really, neuro-monitoring and biomarker study, there's so
little data on this, but patients who are sick like this, not
just LVAD patients but ECMO patients or ICU patients, are close
neurologic monitoring and some markers to predict occurrence of a
stroke or vascular event. I think that's something we really need
to study and look into.


Dr. Sung-Min Cho:


Of course, we have a lot of biomarkers we can pick up from the
brain, brain injury markers that we can study, and that has not
been done in this space. And there are a lot of opportunities, I
think, to look at that. And there's some signal based on
Cleveland Clinic data that Randall Starling actually looked into,
use of PDE5 inhibitor in this patient population, some protection
against the ischemic stroke, and I think that's something also we
should look into for neuroprotective agent.


Dr. Mercedes Carnethon:


Thank you so much! This has been such a delightful discussion
this morning with Sung-Min Cho, the lead author of the study and
the Associate Editor, Marc Ruel who handled it.


Dr. Mercedes Carnethon:


I really appreciate your attention. I hope the listeners enjoyed
this episode of Circulation on the Run. Please join us again next
time.


Dr. Greg Hundley:


This program is copyright of the American Heart Association,
2021. The opinions expressed by speakers in this podcast are
their own and not necessarily those of the editors or of the
American Heart Association. For more, visit ahajournals.org.