Circulation January 8, 2019 Issue

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


Greg
Hundley:                  
And I'm Greg Hundley, Professor at the Pauley Heart Center of
Virginia Commonwealth University Health Sciences in Richmond,
Virginia.


Dr Carolyn
Lam:               
In case you guys missed us last week, this is how our new podcast
is gonna work. Greg and I are going to invite you for coffee with
us, almost with a journal in hand, and we're gonna chat about the
week's issue, highlighting two original papers each, that we
thought were awesome. And don't you worry, the feature discussion
is still there, authors will join us for a feature discussion
right after our coffee.


                                               
And for this week, the feature paper speaks about the MOMENTUM 3
trial, and talks about the important analysis of stroke outcomes
in this trial. But before that, I think Greg, you've got a couple
of papers don't you?


Greg
Hundley:                  
Absolutely Carolyn. So the whole issue, I think we're gonna pick
out several stroke papers, really a stroke theme. The first paper
is Ankit Maheshwari. He looked at the utility of P-wave
morphology on the 12-lead electrocardiogram, to help predict
ischemic stroke in patients with atrial fibrillation.


                                               
Now, how did he do this? Basically, they looked at a large cohort
of individuals from the ARIC study, and these were patients that
developed atrial fibrillation. And electrocardiograms had been
recorded prior to their Afib episode.


                                               
So, what were they looking for in P-wave morphology? Well, they
were looking for changes in Lead three. They were looking for
changes in V1. They were also looking for extension of that
P-wave. So a prolonged duration. And what they observed, is that
that abnormal P-wave, could forecast abnormal atrial remodeling,
that might be an indicator of future stroke.


Dr Carolyn
Lam:               
Huh, interesting. But is it really reproducible? Did they
validate it somehow?


Greg
Hundley:                  
Yeah, so that's great Carolyn. You know, in papers like this, you
like to take a finding in one large cohort, but then you've got
to reproduce it. So they went to the MESA Study. Remember now,
Mesa are individuals without cardiovascular disease. ARIC are
patients with cardiovascular disease. And the finding was
reproducible in MESA. Also, what the authors did, is they looked
at the relevance of this EKG finding to our existing CHADS-VASc2
scoring system.


                                               
And what was really smart by these investigators, is that if you
added the information from the abnormal P-wave morphology to the
CHADS-VASc2 score, you could forecast stroke. Now you say, well
CHADS-VASc2 is already pretty reliable, but what about those
patients that have a CHADS-VASc score of one right? We're always
kind of wondering, do we anticoagulate them? Do we give them
aspirin, et cetera. Well if the P-wave morphology was abnormal
and they were at higher risk for stroke, that could sway you as a
clinician, to go ahead and prescribe anticoagulation for that
group of patients.


                                               
And something very simple, just from the 12-lead EKG before the
patients went into atrial fibrillation. You've got a paper that
also is sort of focusing on stroke. You want to tell us about
that?


Dr Carolyn
Lam:               
Yeah, one big data to another big data series. This time, it's
Get With The Guidelines Stroke series, and this paper is from Dr
Menon from University of Calgary in Canada. Where they described
the door to treatment times for endovascular therapy in acute
stroke. What is that? Well that's a time interval from when the
patient arrives in the emergency department or the door, to the
first pass of the treatment initiation and endovascular therapy.
And basically they found that the median door to first pass time
was 130 minutes. Only 3% of patients achieved a door to first
pass time of less than 60 minutes.


                                               
In multivariable analyses, older age arrival during nonregular
hours and a history of diabetes, were all associated with the
longer door to first pass time. And finally, among hospitals with
an annual endovascular therapy case volume of 40 or less, every
five unit increase in that volume was associated with a 3%
reduction in this door to first pass time.


Greg
Hundley:                  
It sounds like that could be really useful information for stroke
centers, you know, that are managing these patients acutely. How
do you think these results are going to impact that Carolyn?


Dr Carolyn
Lam:               
Great question. So first thing is, I think it provides some
benchmark times for this in hospital workflow, and it obviously
shows areas of improvement. For example, improving workflow
during nonregular hours, or increasing the experience of a
center, and basically emphasizes the point that efforts on
streamlining workflow and saving time, need to continue so that
the full potential of endovascular therapy is realized.


Greg
Hundley:                  
Oh wow, that's outstanding. I'm gonna transition sort of to a
basic science paper, also trying to help manage patients with
stroke. This one is looking at the safety of all of the
dehydrogenased right stem cells. Well, what the world is that. In
animals, what has been shown previously, is this particular cells
type, that's harvested from your bone marrow, can be infused into
the carotid artery, and those animals experience smaller
neurologic deficits after stroke. And so with that encouraging
result in animals, these investigators sought to test the
efficacy of this type of therapy, well not really the efficacy,
but the safety of this type of approach in those patients that
have sustained actually quite a large stroke.


                                               
You had to have a relatively large neurologic deficit to qualify
for this study. And just quickly, the way this works is these
cells enter up through the bloodstream and they modulate
inflammation. By modulating inflammation, that facilitates
healing in the stroke patient.


Dr Carolyn
Lam:               
Yeah, but wow. I mean bone marrow, biopsy and isolating the cells
and so on. How is the study done?


Greg
Hundley:                  
So, the key here is you've had your stroke, you're still in the
hospital with a large neurologic deficit. And so day 11 to 17,
you undergo a bone marrow biopsy. Then the cells are purified,
and they're reinfused into your carotid artery by the way.


                                               
And so, what was the study trying to do? Well, it was actually
looking at the safety off all this. And what would the concern
be? You're infusing these cells into the carotid artery. They go
into the cerebral microcirculation, and those that are working in
this field, are concerned is that going to promote more emboli?
Is that going to promote thrombus? Extend the size of the infarct
in the brain, et cetera?


                                               
So, the investigators performed MRI's and neurologic exams. And
what they found is the neurologic findings in the patients really
didn't change, so there was no benefit. But the study wasn't set
up to look for a benefit. And there were four patients that had a
little bit of an enlargement of the stroke observed on MRI. So, a
lot more to come in this basic science realm, but it's
interesting to see investigators thinking about this in a whole
different way, where we're harvesting one cell type from your
body, and then infusing it up into the brain to sort of help
rescue the situation.


Dr Carolyn
Lam:               
Well, another paper dealing with stroke. This time, a Mendelian
randomization study to explore whether genetically determined
circulating levels of cytokines and growth factors, may be
associated with stroke. And this was done in the mega stroke GWA
data set and validated in the UK biobank, and it’s by Dr Dichgans
and colleagues from the university hospital, Ludwig Maximilian
University of Munich. They basically found, that a genetic
predisposition to higher circulating levels of monocyte
chemoattractant protein one, was associated with a higher risk of
stroke. The associations also found for the etiology of the
stroke subtypes, and especially for large artery stroke and
cardioembolic stroke. In fact the genetically determined levels
of this monocyte chemoattractant protein one, was also associated
with higher risk of the related phenotypes of coronary artery
disease and myocardial infarction.


Greg
Hundley:                  
So, how do you bring this to practice in the clinic Carolyn?


Dr Carolyn
Lam:               
So, this is still some steps away, but I do think that it very
nicely supports the idea that inflammation as part of the
pathogenesis of stroke, and of course additional work is needed
to determine whether targeting the specific monocyte
chemoattractant protein one, or it's downstream effectors, may be
a meaningful strategy to lower stroke risk. So, terribly
interesting.


Greg
Hundley:                  
Yeah, you know it sounds like hitting inflammation or targeting
that, is a real theme here from the basic science group. Well
this is great Carolyn.


                                               
And now, I guess we'll transition over to our feature article.


Dr Carolyn
Lam:               
Absolutely. So, we're here to discuss the long-term results of
the MOMENTUM 3 Trial, and that was a randomized controlled trial
of the HeartMate 3 versus the Heartmate II left ventricular
assist device. And this time, with a focus on stroke. The
outcomes that's just so important to our patients. Greg and I are
incredibly pleased to have with us, the authors, Dr Mandeep Mehra
from Brigham and Women's Hospital, as well as our senior
associate editor, Dr Biykem Bozkurt, to discuss this paper.


                                               
Mandeep, perhaps just set the scene by telling us what this
secondary analysis found?


Dr Mandeep Mehra:       This
analysis is really focused on the issue of stroke, as you pointed
out. I'd like to just lace into context what this is important.
Ever since the advent of left ventricular assist device therapy
from the 80s and early 90s, to now, one of the major Achilles'
heels, whether we have used pulsatile flow devices or
non-pulsatile flow devices, has been the very constant occurrence
of a high incidence of stroke, beyond the stroke rates were
predominantly as compared to ischemic strokes. Then with the
newer devices, we actually saw a reversal, where we began to see
more ischemic strokes as opposed to hemorrhagic strokes, almost
an equal parts at this time point.


                                               
And this has been one of the critical reasons why we have not
been able to expand the therapy beyond the very, very sick
patient.


Greg
Hundley:                  
Very nice. And another particular in the results here is, you
didn't really see a difference in stroke rates, either
hemorrhagic or ischemic strokes early, but you did start to see a
difference after 180 days. Why do you think that's the case?


Dr Mandeep Mehra:       That's a
great point Greg. We really saw no difference in the first 30
days. When we analyzed this data, we divided it into a
perioperative, a first 30-day time point. Then, we looked at the
short-term time point up to 180 days or six months, and then
beyond that to the two year end point. What became very clear is
that most of the gains that we saw in the stroke rate, began to
appear after the first 30 days, did not quite reach statistical
significance at six months, but really the differences became
heavily pronounced after six months, all the way out to two
years.


                                               
So, first point that I would make Greg, is that we did see
differences beyond 30 days, it's just that they didn't reach
conventional statistical significance. The second thing is, the
more important point that you make, asking why that was the case.
We actually think that the reason behind that, is that the first
three months or so after that implant, really is a period of
chaos in these patients, where the hemocompatibility, which is
essentially the interface between the device as well as the
patient, is attempting to be established. And it's very similar
in a way as we see in heart transplantation Greg, where the real
challenge in heart transplantation is between rejection and
infection.


                                               
And in the case of left ventricular assist device is the
challenges between bleeding and thrombosis. It turns out that
three months, whether it be transplantation or whether it be left
ventricular assist devices, seems to be this period of chaos and
adjustment, during which the patient and the device are starting
to get to know each other.


                                               
And this is why we think that most of the gains occurred after
this period of chaos was overcome.


Greg
Hundley:                  
No, it's really interesting that after accounting or adjusting
for all the anticoagulant drugs, antiplatelet drugs, even the
other medical therapies that were applied, you found these
results. I mean, maybe also bring in Biykem here to answer the
question, what is this machine doing that's providing such a
benefit?


Dr Biykem
Bozkurt:         The
two-year results being quite impressive for the HeartMate 3 are
truly encouraging. Because I think we truly see a concordance
benefit beyond 180 days, especially the nondisabling strokes,
giving the hope to the providers that we can further perhaps
enhance the field by focusing on optimization of anticoagulation
strategies, prevention of atrial fibrillation, and maybe even
consider our algorithms or pathways for stroke. Because, in this
protocol, even though the stroke management was not standardized,
and I'm sure that the data will not yield that information as to
which centers were able to approach the stroke management in a
perhaps evidence based approach, the sobering facts are
regardless of the device, at two years, approximately half of the
patients died. Even the non-disabling stroke patients had
increased mortality compared to no-stroke patients.


                                               
And if you examine evidence-based approaches, only one-third of
the hemorrhagic stroke patients had reversal of anticoagulation,
and a very small percentage ... actually, none of the patients
had device intervention for the ischemic stroke. That raises the
question of yes at two years the HeartMate 3 results are very
promising. But, can we further even advance the field by doing
evidence based standardized pathway driven stroke treatment
approaches.


                                               
The other very interesting finding from this trial is, in
ENDURANCE trial, which was another trial with centrifugal device,
HVAD device, there was an association of the stroke rates with
inadequate control of blood pressure and anticoagulation, which
was not noted in this trial. Maybe Mandeep can comment on do we
truly have the adequate power to be able to infer whether blood
pressure control and/or appropriate anticoagulation management
strategies will matter?


Dr Mandeep Mehra:       Biykem
you've said it really well, and I'd like to just make some
additional points with respect to the question. So, first of all
Greg you're absolutely correct, that we tried to search for
anything that would predict this reduction in stroke with the
HeartMate 3, and it turned out that all we were left with is the
device itself. So, it really begs the question, what is it about
the device or it's interface that may have resulted in this.


                                               
And of course, some of what I'm about to tell you will be
speculation, but it may actually carry some water. So, for
example, the HeartMate 3 is very unique in one other aspect, and
that is that, even though it's a small profile device, it's
engineering principles are such that it allows for very wide
blood flow pathways. And in fact, despite its small profile, the
blood flow pathways allow for 20 times more red blood cells to
travel through the primary and secondary pathway, than other
devices.


                                               
What it means is that as blood is going through this device, it
is exposed to very low sheer stress. And in return, the benefit
that we see very clearly with this device in a very, very
important way, is the fact that we see almost no denovo pump
thrombosis developing with this device. Certainly, if the device
doesn't carry some small quad risks in it, that cause problems
with the device, it's probably also not causing the production of
smaller non-device malfunction producing thrombi, which may with
other devices, actually develop and cause strokes.


                                               
So, we think that particular engineering enhancement, may play a
very important role in reducing this stroke rate that we have
observed.


                                               
The second very important point that Biykem brought up, is this
notion about the management of ... whether it be with
anticoagulants or with blood pressure management. And for a
moment let's dwell on the blood pressure issue. One of the
striking things with the other centrifugal device, the HVAD
device, is that the ENDURANCE Trial showed a significantly higher
stroke rate with that device. And in fact, in a subsequent study,
the ENDURANCE Supplemental Trial, when blood pressure was
tightly, tightly controlled in the device, there appeared to be a
small signal in reduction in strokes, although it still did not
meet the non-inferiority endpoint, compared to the HeartMate II
in the second supplementary trial that was done with that device.


                                               
So, what's unique about this? Well, we can very clearly say maybe
we just didn't have enough ability to show a difference in this
particular trial, we didn't analyze it the right way, because we
didn't have a blood pressure intervention or low or higher
permissive blood pressures in this trial. But I would say that
there's one other issue that I think may have played a very
important role in this, and that is the HeartMate 3 is
intrinsically developed with a fixed pulse algorithm. And in
fact, the HeartMate 3 has a capacity where the magnetically
levitated rotor upregulates itself and then downregulates itself
every two seconds, and creates an internal pulsatility.


                                               
Now, engineers developed that pulsatility to really decrease
stasis, so that the pump wouldn't thrombose. But we often see
that it provides sufficient peripheral pulsatility, not to the
pulse pressures that we would normally like to see, but certainly
to some degree, where the vasculature can perceive or transduce
some degree of pulsatility. Why that may be important is, that it
may actually allow for preservation of baroreceptor function in
these patients, which tends to be lost in continuous flow pumps.


                                               
And how important that is for blood pressure regulation and its
vascular effect, may be something that needs to be looked at into
the future. But it's certainly a very, very intriguing issue for
us to examine.


Dr Biykem
Bozkurt:         Mandeep,
one final question or comment. Do want to comment on the stroke
rates of HeartMate II compared to former trials. Because that
comes as a common query as to why in MOMENTUM 3 the stroke rate
in HeartMate II, appear to be higher than the former trials.


Dr Mandeep Mehra:       So very
quickly, I'll tell you they're not. So, if you look at the 2009
randomized trials, randomized patients with a HeartMate II versus
the HeartMate XVE trial, the two-year stroke rates with the
HeartMate II in that trial were 19%, exactly what we observed at
two years in this trial.


                                               
Other trials have shown exactly that same number. The only trial
in which there appeared to be a difference in those numbers, was
in the ENDURANCE Trial, where the two-year rate of any stroke was
12%, and was a little lower in the HeartMate II than what we
observed. However, I will caution you that if someone dies before
having a stroke, then they die without a stroke. And so, stroke
can sometimes we underestimated if the population that is
enrolled, such as a transplant ineligible population at very high
risk, is dying more often than having the chance of a stroke.


                                               
So, I actually do not think at all that there was any difference
whatsoever compared to prior trials. And even when you look at
the ENDURANCE Supplement Trial, which is probably the most
contemporary comparison of HeartMate II stroke rates, with
MOMENTUM 3, the ENDURANCE Supplement Trial was only a one year
trial, and the stroke rates even at one year were right on target
with what we observed at the HeartMate II group in MOMENTUM 3.
So, frankly that criticism is probably an unfounded criticism.


Dr Biykem
Bozkurt:         Thank
you.


Dr Carolyn
Lam:               
Wow, thank you Mandeep and Biykem, for really helping us go under
the hood with this paper. I'm heart failure trained as well, but
I learned so much, I'm sure our listeners did as well, and I'm
sure you agree too Greg.


                                               
Thank you so much for joining us today. Don't forget to tune in
again next week.


                                               
This program is Copyright American Heart Association 2019.