Latest / European Society for Vascular Surgery / TEVAR vs Standard Medical Therapy in Uncomplicated Type B Aortic Dissection: A Roundtable with the IMPROVE-AD, EARNEST and SUNDAY Trial Investigators

Transcript

TEVAR vs Standard Medical Therapy in Uncomplicated Type B Aortic Dissection: A Roundtable with the IMPROVE-AD, EARNEST and SUNDAY Trial Investigators · European Society for Vascular Surgery

• 0:12
  The ESVS Podcast. I'm Safa and today's episode
• 0:14
  tackles one of the most debated questions in
• 0:17
  contemporary aortic surgery. What do we actually
• 0:20
  do with an uncomplicated type of aortic dissection?
• 0:23
  The instincts to treat feels compelling. These
• 0:26
  patients have just experienced a catastrophic
• 0:28
  event. The aorta is torn, the false lumen is
• 0:31
  open and pressurized. And we know that over the
• 0:33
  following months and years, a significant proportion
• 0:36
  will develop aneurysmal degeneration, progression
• 0:38
  or rupture. And yet, the word uncomplicated is
• 0:42
  doing a lot of the heavy lifting. These patients,
• 0:45
  by definition, have no evidence of end -organ
• 0:48
  ischemia or aortic rupture. Their immediate risk
• 0:51
  is lower and TAVAR, while far less invasive than
• 0:55
  open surgery, is not without procedural hazards,
• 0:58
  which are great type 8 isections, spinal cord
• 1:00
  ischemia, and the long -term unknowns of a stent
• 1:03
  graft left in a dissected aorta. awake concerns.
• 1:07
  This is precisely the clinical tension that three
• 1:09
  major randomized controlled trials are now trying
• 1:12
  to resolve concurrently and with complementary
• 1:15
  design and enough mythological difference to
• 1:17
  make today's conversation genuinely important.
• 1:21
  I am delighted to be joined today by three vascular
• 1:23
  surgeons who are not simply observers of this
• 1:26
  debate, they are leading it. Professor Firas
• 1:29
  Musa is a professor of vascular surgery at Baylor
• 1:31
  College of Medicine. and the principal investigator
• 1:34
  of the IMPROVE -AD trial, the largest of all
• 1:36
  three, enrolling 1 ,100 patients across 60 North
• 1:40
  American centers. Professor Colin Bicknell is
• 1:43
  a professor of vascular surgery at Imperial College
• 1:45
  London and leads the earnest trial. a UK -based
• 1:49
  study with a particular focus on cost -effectiveness
• 1:52
  and patient -reported outcomes. And Professor
• 1:55
  Jacob Budtz -Lilly is a consultant vascular surgeon
• 1:57
  in Aarhus, Denmark, and one of the principal
• 1:59
  investigators of the SUNDAY trial, a Scandinavian
• 2:02
  -led study targeting the subacute phase specifically,
• 2:05
  with access to some of the world's most complete
• 2:08
  national registry follow -up. Professor Musa,
• 2:11
  Professor Bicknell, Professor Budtz -Lilly, welcome
• 2:13
  and thank you for joining us. Thank you very
• 2:16
  much. Thank you. Good to be here. Thanks. Before
• 2:18
  we get into the individual trials, I'd like to
• 2:21
  briefly step back and ask a question that I suspect
• 2:24
  many of our listeners have wondered. We have
• 2:27
  had INSTET, we have had ATSOP, we have had INSTAD
• 2:30
  Excel. We have meta -analysis suggesting benefit
• 2:33
  from TVAR and meta -analysis suggesting no difference.
• 2:37
  We have guidelines from both sides of the Atlantic
• 2:39
  and broadly recommend... a selective approach
• 2:42
  to T -viral and uncomplicated disease, without
• 2:45
  really defining what selective means in practice.
• 2:48
  Professor Musa, you helped establish the feasibility
• 2:51
  of one in a randomized trial in uncomplicated
• 2:54
  type B aortic dissection at a time many of the
• 2:57
  field doubted it was achievable. Looking back
• 3:00
  at those early feasibility surveys, What was
• 3:03
  the strongest argument against the trial and
• 3:06
  how did you answer it? Right. So the strongest
• 3:09
  argument was that actually it's like any other
• 3:13
  trial. It's hard to convince a surgeon and hard
• 3:16
  to convince a patient to enroll. That is not
• 3:19
  to dissimilar from any other cardiovascular study.
• 3:22
  Mind you that this is not like heart disease
• 3:25
  where the disease is very prevalent. Aortic section
• 3:29
  is not that common. even in a large country like
• 3:32
  the U .S. So the main limitation was really the
• 3:36
  number required to treat and the feasibility
• 3:40
  of enrolling those patients. Professor Bicknell,
• 3:44
  earnest was designed in part because the U .K.
• 3:47
  vascular surgery community felt existing evidence
• 3:50
  was insufficient to guide practice. When you
• 3:53
  look back at the incident data, no survival benefit
• 3:56
  at two years. but a clear benefit emerging only
• 3:59
  after five years and instead excel. What does
• 4:02
  that tell us about how we have been designing
• 4:05
  and powering trials in this disease? And what
• 4:08
  gap does Ernest specifically set out to fill?
• 4:10
  So you're right. The UK vascular surgery community
• 4:14
  was not convinced by these trials and felt that
• 4:19
  further evidence was necessary. These trials,
• 4:23
  though, are... Not really different from many
• 4:27
  other situations when you have randomized controlled
• 4:29
  trials. The setup and running of a randomized
• 4:36
  controlled trial is famously difficult to fund.
• 4:40
  It's famously difficult to convince people to
• 4:42
  be with you and to join with you in the trials.
• 4:45
  And it's constrained by money. time everyone
• 4:49
  wants an answer now and doesn't want to uh doesn't
• 4:53
  want to wait five or ten years for the answer
• 4:55
  and the number of patients of course is also
• 4:57
  really important so there are a lot of issues
• 5:01
  with these trials because of these general circumstances
• 5:06
  but if you run trials that are inadequately powered
• 5:12
  or don't take into their account the longer term
• 5:16
  in this cohort of young fit often fit people
• 5:20
  which is so important you can end up making the
• 5:24
  wrong decisions about treatment of patients so
• 5:26
  i think that was all important um i think that
• 5:29
  as well as the small numbers as well as the fact
• 5:33
  that there may well have been um biases in in
• 5:37
  terms of that that it was also in a small number
• 5:40
  of centers which were expert centers it wasn't
• 5:44
  Specifically for the UK, it wasn't in a publicly
• 5:47
  funded health system and it wasn't in an area
• 5:51
  where every centre that did TVAR joined in. And
• 5:54
  so we really had a need to run a trial such as
• 5:59
  ERNEST. The binary classification that has governed
• 6:02
  management for decades was built on what happens
• 6:05
  in the first two weeks. It says very little about
• 6:08
  what the aorta will do over the next five or
• 6:11
  ten years. The absence of acute complications
• 6:13
  at presentation is not the same as long -term
• 6:16
  benignity, and the field zone attempts to refine
• 6:19
  the category through high -risk anatomical features
• 6:22
  have not yet produced a consensus strong enough
• 6:25
  to change guidelines. Professor Butz -Lilley,
• 6:28
  does the label uncomplicated still serve us,
• 6:32
  or is it time to replace it with something more
• 6:34
  granular? Yeah, that's a good question. I think
• 6:38
  uncomplicated still does serve a purpose. But
• 6:42
  I think we need to be clear about what that means.
• 6:44
  I mean, what it means to serve a purpose. It
• 6:46
  describes, I mean, it describes your patient's
• 6:49
  condition today. I mean, there's no rupture,
• 6:51
  there's no malperfusion, nothing to tell you
• 6:54
  that there's an urgent intervention required.
• 6:56
  But that's not the same as it telling us what
• 6:59
  that aorta will do over the next five or 10 years
• 7:02
  or so. At the same time, I mean, I think we do
• 7:04
  need greater granularity, but I would be cautious.
• 7:08
  And I think... If you're Austin Collin would
• 7:11
  agree that we'd be cautious about replacing one
• 7:15
  binary label with another one, such as high risk
• 7:19
  uncomplicated. Because the difference, I think
• 7:22
  people forget that between a prognostic factor
• 7:25
  and a predictive factor. So a lot of these high
• 7:29
  risk features, they might be prognostic. That
• 7:31
  might mean that they might have an adverse outcome
• 7:34
  at some point. But that's not the same thing
• 7:36
  is that they benefit more from a TAVAR. I mean,
• 7:40
  that requires evidence, which we really don't
• 7:44
  have. So I think we could still use the term
• 7:47
  uncomplicated, but obviously still do a trial,
• 7:50
  all the Sunday earnest and improve. I mean, they'll
• 7:52
  move us from these descriptive risk marker classification
• 7:56
  that actually guides treatment instead of just
• 7:59
  using these as a prognostic classification. That's
• 8:02
  an excellent point. I'd like to spend some more
• 8:05
  time now on the methodological choices in each
• 8:07
  trial. Because I think the differences between
• 8:10
  them are not just academic, they reflect genuinely
• 8:13
  different hypotheses about what matters most
• 8:16
  in this disease. For SMUSA, IMPROVE -AD is the
• 8:20
  largest of three, 1 ,100 patients, 60 centers,
• 8:24
  6 -year follow -up window, and a composite primary
• 8:27
  endpoint of all -cause death and major aortic
• 8:29
  complications. The composite is broad. aortic
• 8:33
  rupture, malperfusion, retrograde dissection,
• 8:36
  spinal cord ischemia, stroke, and aortic invention.
• 8:39
  Some might argue that such a wide NER risks diluting
• 8:43
  the signal or capturing events that are clinically
• 8:46
  very different from another. How do you defend
• 8:48
  that design choice and what does Improved AD
• 8:51
  uniquely offer that a smaller, more targeted
• 8:54
  trial cannot? Yeah, that's an excellent question.
• 8:58
  It honestly comes down to basics of trial. Pick
• 9:02
  a high event rate and a reasonable effect size.
• 9:09
  So let's pick mortality. So if you want to show
• 9:12
  that TVAR improves mortality, any treatment that
• 9:17
  you're going to pick cannot have an effect size
• 9:20
  that is huge. Let's say TVAR cannot reduce mortality
• 9:23
  by 80%, not even. 70%, not even 50%. At best,
• 9:29
  at best, would be 30%. So to power for a 30 %
• 9:34
  risk reduction with a high power, over 80 % power,
• 9:40
  this would require over 2 ,000 patients. Okay?
• 9:46
  So that becomes non -feasible. So the art of
• 9:50
  trial design is you've got to pick something
• 9:52
  that happens frequent, matters to patients, and
• 9:56
  also allow you to design within reasonable limits.
• 10:01
  That's what happened in INSTEAD. They overestimated
• 10:04
  the effect size of TIVAR. They ended up with
• 10:06
  an underpowered study. That's the premise of
• 10:09
  picking an endpoint. Now, the ones that we picked,
• 10:12
  you're right, it is a wide thing. It kind of
• 10:16
  filters out the noise of all death. Death could
• 10:20
  happen. A patient could... have a car accident
• 10:22
  and die in any trial, and that will be considered
• 10:26
  death. It is not aortic -related death. It's
• 10:29
  death or cancer or, you know, if they live enough,
• 10:32
  they die of different disease. So then you could
• 10:34
  think of it as lots of noise in a big thing like
• 10:37
  death. However, when you couple death with a
• 10:42
  few things that don't happen frequently, like
• 10:46
  malperfusion, stroke, and all that could happen
• 10:48
  after intervention. But the other thing that
• 10:50
  happens and people talk about is like, oh, back
• 10:53
  in the day, they used to get an open, big repair.
• 10:56
  People used to fear that a lot. And that mattered
• 10:59
  to patients. So we consider that to be part of
• 11:02
  the endpoint. You cannot just say open repair
• 11:04
  is bad. You can say if you have a complex endovascular
• 11:08
  repair, like a four -vessel fenestrated and branched.
• 11:12
  Then that also puts a patient at a risk of paraplegia
• 11:15
  and this is not readily available everywhere
• 11:18
  and expertise matters. So then becomes an important
• 11:21
  to patient, important from cost effectiveness
• 11:24
  standpoint. And same thing with the repeat T
• 11:27
  -VAR. I got to have to say, we do not consider
• 11:29
  patients who get medical therapy and then go
• 11:32
  on to get a T -VAR as an intervention. That is
• 11:36
  part of the standard of care. We consider those
• 11:39
  who get a big repair. whether open and endo to
• 11:42
  be an endpoint. So in a nutshell, you're going
• 11:45
  to have to be reasonable when you design within
• 11:47
  the confines of what's statistically appropriate
• 11:52
  for the funding agency in the US, that is. That's
• 11:56
  really interesting. Thank you. Professor Batslili,
• 12:00
  some days primary endpoint is all -cause mortality
• 12:03
  at five years. We must trade for the vote -free.
• 12:06
  Your trial is event -driven, powered on 80 deaths.
• 12:10
  using Scandinavian national registries to achieve
• 12:12
  new complete flow -up. That is genuinely an elegant
• 12:15
  design, but I'd like to push you on it. All -cause
• 12:18
  mortality simplifies adjudication and avoids
• 12:21
  composite dilution, but does it risk missing
• 12:24
  what TVAR may actually achieve? Because TVAR's
• 12:27
  most plausible benefit may be in preventing aortic
• 12:30
  degeneration, not in reducing early deaths. Yeah,
• 12:33
  that's an excellent question, and Firas and Colin
• 12:37
  and I have discussed this quite a bit. And I
• 12:39
  think it's a legitimate concern, but actually
• 12:41
  that's exactly why we chose all -cause mortality.
• 12:46
  We chose it not because it's the easiest endpoint,
• 12:49
  because it's actually the hardest endpoint to
• 12:53
  misinterpret at the end of the day. I mean, it's
• 12:56
  unambiguous, kind of, as you said, it's resistant
• 12:58
  to adjudication, independent of any local thresholds
• 13:04
  for re -intervention. And I think that really
• 13:06
  matters in any of these interventional trials
• 13:08
  because It's difficult. Some centers might reintervene
• 13:12
  at different thresholds or whatnot, and that
• 13:14
  can be really difficult to delineate. I mean,
• 13:18
  if this trial, if it shows that, yeah, the early
• 13:22
  TAVR, for example, improves CT scans, but doesn't
• 13:25
  really improve how long or how well a patient
• 13:28
  lives, you would kind of have to ask yourself,
• 13:31
  is that really clinically important? But I get
• 13:35
  it. I mean, mortality isn't the whole story.
• 13:37
  I mean, we include the important secondary events,
• 13:41
  reinterventions, complications, quality of life
• 13:44
  and costs and whatnot. But in a way, and you
• 13:48
  have to be clear that our trial is event -driven
• 13:51
  because we don't expect an early mortality effect.
• 13:55
  We hypothesize that the benefit is actually long
• 13:58
  -term. And I think... If TAVAR shows that there's
• 14:04
  improved survival, our secondary outcomes would
• 14:07
  obviously add further information. And particularly
• 14:11
  if TAVAR shows there's no survival information
• 14:12
  benefit, then I think that the secondary outcomes
• 14:16
  will go far to show that any of these other secondary
• 14:21
  interventions, other things I mentioned, that
• 14:22
  there's a difference there. But that decision
• 14:25
  was deliberate for some of the reasons you just
• 14:28
  mentioned. It's really great to see it from your
• 14:30
  point of view. Now, Professor Bicknell, ERNEST
• 14:33
  takes a different approach, a composite of aortic
• 14:37
  -related mortality, severe permanent neurological
• 14:39
  deficit, and severe permanent cardiorespiratory
• 14:43
  failure. More specific, and with a plan -and
• 14:46
  -train analysis assessing procedural safety and
• 14:49
  aortic remodeling at one year. You've also built
• 14:52
  health economic analysis into the primary endpoint.
• 14:56
  And, Professor Bicknell, I'll leave it to our
• 14:58
  listeners to decide whether that reflects the
• 15:00
  unique rigor of the earnest team or simply the
• 15:02
  realities of working with the NHS. Why was cost
• 15:06
  -effectiveness a priority? And do you think that
• 15:10
  kind of data would actually shift clinical decision
• 15:13
  -making? So, we went for a composite primary
• 15:17
  endpoint because our patients wanted it. After
• 15:20
  talking to all of our patients, we designed a
• 15:23
  number of secondary endpoints. One of those is
• 15:27
  cost -effectiveness. We chose to analyze cost
• 15:32
  -effectiveness because it really is the responsible
• 15:35
  thing to do in a publicly funded healthcare system
• 15:38
  with limited finances. And I think as physicians,
• 15:41
  we do have a responsibility to provide cost -effective
• 15:45
  healthcare. There are those that are involved
• 15:47
  in the institutions such as NICE, as you know
• 15:51
  from the underism discussions that we had a few
• 15:54
  years ago, that it's irresponsible to fund a
• 15:57
  highly expensive, ineffective treatment or a
• 16:02
  treatment that only provides minimal improvements
• 16:05
  in quality adjusted life years over something
• 16:08
  that's cheap and will provide many more quality
• 16:11
  adjusted life years for the population. When
• 16:16
  there is limited finances, you have to do something
• 16:19
  along those lines. So I think it's responsible.
• 16:22
  I think it's the right thing to do. And I think
• 16:24
  it will provide really interesting information
• 16:26
  for us all. But your question on whether it will
• 16:30
  actually shift clinical decision making is a
• 16:36
  very wise, level -headed question. can think
• 16:43
  of lots of examples where vascular surgeons,
• 16:47
  cardiologists and every other specialty has continued
• 16:50
  to use non -cost effective treatments. It's okay
• 16:55
  to do that as long as there is, it's a new technology
• 16:59
  which is developing that there might be other
• 17:01
  perceived advantages, but we will have to argue
• 17:04
  our case to the National Institute of Clinical
• 17:07
  Excellence, I suspect, if it doesn't come out
• 17:10
  particularly within those. costing boundaries.
• 17:13
  That's a very compelling point of view. But before
• 17:16
  we go deeper into methodology, I think our listeners
• 17:19
  deserve status reports. These trials, they're
• 17:22
  not hypothetical. They're live. They're enrolling
• 17:25
  patients right now. And the vascular surgery
• 17:27
  community is watching closely. Professor Batsley,
• 17:30
  Sunday was among the first of the three trials
• 17:33
  to open. What can you share with our listeners
• 17:37
  today? How is recruitment progressing? Yeah,
• 17:40
  actually, you know, today we just had our 200th
• 17:43
  patient recruited. I haven't called, but it was
• 17:45
  actually Nuno Diaz in Malmö recruited patient
• 17:48
  number 200. So that was great news. And I should
• 17:52
  add, I mentioned this in the last question, that
• 17:54
  it's event -driven. So even though our sample
• 17:58
  size estimates approximately 550 patients, that
• 18:02
  could be more or less. It's all driven by the
• 18:04
  number of events. And fortunately, we have a
• 18:07
  good group. And when we do this for each other,
• 18:10
  We present sort of all three trials collectively.
• 18:13
  I'd just like to add one other thought, because
• 18:15
  I think it was a loaded question you gave to
• 18:17
  Colin. You know, all these factors that go in
• 18:20
  to make the right decision. One thing I think
• 18:22
  that's important, overarching this question of
• 18:27
  what it is we'll show, I think we need to remember
• 18:30
  that even showing equivalency is still a valuable
• 18:33
  result. I don't think a trial has to prove that
• 18:36
  there's a difference. I think the knowledge we
• 18:38
  obtain... In our clinical practice, when you
• 18:41
  talk with a patient, you say, look, there's no
• 18:43
  real difference whether we go this way or this
• 18:45
  way with a strategy. That's still really valuable
• 18:48
  in a conversation with a patient, as far as I'm
• 18:51
  concerned. Professor Bechnor, ERNEST enrolled
• 18:53
  its first patient in March 2025. What have the
• 18:57
  first months of recruitment looked like in practice?
• 18:59
  Are the eligibility criteria working as intended
• 19:01
  when they meet real patients in real clinics?
• 19:04
  And has anything surprised you yet? issues with
• 19:10
  the research infrastructure that that is set
• 19:14
  up obviously around the uk the lead time into
• 19:18
  setting up centers has been really long in some
• 19:23
  in some centers so we we've we struggle with
• 19:26
  all of that more interestingly we've seen um
• 19:30
  lots and lots of screaming logs from from patients
• 19:34
  who have reported many unsuitable patients and
• 19:38
  many complex patients or many complicated patients.
• 19:43
  And we've discussed the reasons why we've seen
• 19:46
  so many complicated cases at length. I think
• 19:49
  that what surprised me, what shouldn't have surprised
• 19:54
  me, but probably... did surprise me a little
• 19:57
  bit is that we all ended up looking for that
• 20:00
  perfect dissection case that comes in has his
• 20:03
  blood pressure treatment for three days then
• 20:05
  gets better then goes to the ward and then gets
• 20:08
  randomized and then we do the stent either in
• 20:10
  the same hospital sitting at three three months
• 20:13
  but it doesn't really exist every single dissection
• 20:17
  is different in terms of its pathology and anatomy
• 20:20
  and it's and the symptoms that it sort of induces
• 20:25
  And so we have done quite a lot of work in trying
• 20:30
  to come to a consensus or at least a more rounded
• 20:34
  definition of uncomplicated. We've done some
• 20:36
  work looking at trying to increase the numbers,
• 20:39
  the sort of different patterns of dissection
• 20:42
  that could go into the trial without straying
• 20:44
  outside our exclusion criteria. And we've been
• 20:48
  very clear in... asking centers to randomize
• 20:55
  everyone that they possibly can around the UK.
• 20:58
  And with our different ways of looking at this
• 21:03
  trial, the numbers will start to increase. But
• 21:06
  I think we're right at the bottom of our acceptable
• 21:09
  level at this 18 -month period, and we need to.
• 21:13
  Professor Musa, improved ID is the most ambitious
• 21:17
  in scale. How is recruitment tracking and has
• 21:20
  coordinating a trial this size across such a
• 21:23
  heterogeneous healthcare landscape thrown any
• 21:26
  expected operational challenges? Right. So, thank
• 21:30
  you. The U .S. is a different healthcare system,
• 21:35
  obviously, than both the U .K. and Europe in
• 21:37
  general. It's fee -for -service. So, if you do
• 21:41
  a procedure, you get paid for it. If you don't
• 21:43
  do a procedure, you don't get paid for it. Not
• 21:46
  only you as a surgeon, but you and the hospital.
• 21:50
  Mind you, that if you do a procedure in a hospital
• 21:53
  that's not equipped and you have a complication,
• 21:56
  then that beats, you know, the idea that do TVAR
• 22:02
  in every hospital in the country. The healthcare
• 22:06
  system is such that you try to do as many as
• 22:09
  possible with the capacity that you have. So
• 22:12
  that's the infrastructure that we currently have.
• 22:15
  Plus what I just mentioned is the fee for service.
• 22:18
  So our challenge is really to convince vascular
• 22:21
  surgeons to give up the revenue from doing a
• 22:26
  procedure. in order to support science. And that
• 22:29
  is not intuitive in the U .S. It's not something
• 22:33
  that's done. And also, it's not unique for vascular
• 22:36
  surgeons, but vascular surgeons have different
• 22:38
  mindsets than cardiologists where one reason
• 22:41
  or the other, they do enroll thousands upon thousands
• 22:45
  of patients with medical therapy versus PCI for
• 22:48
  MI patients, but vascular surgeons don't seem
• 22:52
  to do the same for... This trial. Now, we are
• 22:58
  at 223. I think today we enrolled one, so 224.
• 23:04
  And again, in the U .S., not everybody believes
• 23:06
  in the need for the trial because I think, you
• 23:10
  know, the economic consequences of that is that
• 23:15
  you'll lose revenue. And that's really what we're
• 23:17
  up against. That's really enlightening about
• 23:20
  conditions across the Atlantic. Thank you. One
• 23:25
  of the most important questions in uncomplicated
• 23:28
  type B aortic dissection management is not just
• 23:31
  whether to treat, it's who to treat. And each
• 23:34
  of your trials had made distinct choices about
• 23:36
  inclusion and exclusion criteria. Professor Bicknell,
• 23:40
  ERNEST has taken the interesting step of including
• 23:43
  patients with uncomplicated type B aortic dissection
• 23:46
  during pregnancy, a subgroup explicitly excluded
• 23:49
  from many aortic trials. Can you explain the
• 23:53
  rationale here? And more broadly, illness excludes
• 23:56
  patients requiring carotid or innominate artery
• 24:00
  vascularization to create a landing zone, or
• 24:03
  those with planned visceral artery intervention.
• 24:06
  How does that affect the generalizability of
• 24:09
  your findings to rework practice? We're not,
• 24:12
  let's just be clear. advocating stenting during
• 24:16
  pregnancy. But if you have a type B aortic dissection
• 24:19
  when you're pregnant or during birth, and then
• 24:24
  you are still within that first three months,
• 24:27
  why are you any different to anyone else that
• 24:30
  has a type B aortic dissection over the long
• 24:33
  term? You've still got a false lumen, which is
• 24:35
  profuse. You're still at risk of dilatation.
• 24:38
  And what we don't know is whether TVAR will induce.
• 24:44
  uh, thrombosis, false lumen remodeling in a lady
• 24:48
  who had her dissection during pregnancy, just
• 24:51
  as, uh, anyone else because of hypertension or
• 24:55
  hypertension or atherosclerotic disease. So I
• 24:59
  think, I think it's reasonable. I don't think
• 25:00
  it'll be a huge number. We wanted to be as inclusive
• 25:02
  as possible, which is why we, really why we left
• 25:05
  it in. As for the carotid and the innominate
• 25:08
  and the visceral vessels, well, we brought our
• 25:10
  surgeons together in a big group. And we were
• 25:15
  very clear, or they were very clear with me,
• 25:19
  I should say, that they wanted a trial where
• 25:23
  we were tackling standard type B aortic dissection
• 25:29
  requiring standard T -bar. And they thought that
• 25:33
  if we were going to be revascularizing the carotid
• 25:36
  arteries or the anomena artery or going into
• 25:39
  the arch, that there was a... greater risk of
• 25:45
  having significant longer complications and that
• 25:49
  this would be not the trial that they wanted
• 25:52
  to run. So when we get to the end of our trial,
• 25:58
  when we get to the end of our trial, we will
• 26:01
  have results on this group of patients and we
• 26:04
  will not be able to answer whether or not non
• 26:06
  -A, non -B patients should be treated with T
• 26:11
  -bar. But we will have a sort of definite answer
• 26:14
  for that standard T -var for type B aortic dissection,
• 26:18
  I hope. Professor Musa, improved VAD stratifies
• 26:22
  patients based on high -risk anatomical features,
• 26:25
  aortic diameter of 40 mm or more, primary entry
• 26:28
  tear of 10 mm or more. or a false lumen diameter
• 26:32
  of 22 mm or more. Do you believe those markers
• 26:35
  reliably identify who will progress? And is there
• 26:39
  a risk that by including all comers and relying
• 26:42
  on subgroup analysis, you may need to wait several
• 26:45
  more years before a specific patient phenotype
• 26:47
  emerges from your data? So we do include all
• 26:51
  comers. This stratification is done after enrolling
• 26:55
  the patient. Patients sign a consent. Then...
• 27:00
  myself, let's say a side PI and the coordinator,
• 27:04
  sit in front of the computer, look on the images,
• 27:09
  and fill in what we call the imaging CRFs, okay,
• 27:14
  and decide and pick if the patient has one of
• 27:18
  these features that you mentioned, and we store
• 27:21
  those. At the end, we will have those patients
• 27:26
  who have, let's say, pick an aortic diameter
• 27:28
  of 40, versus those without. And by stratifying,
• 27:33
  we're including, we're making sure The two buckets
• 27:37
  are of equal risk. I don't want somebody to come
• 27:41
  back at the end and say, well, you just stented
• 27:43
  everybody that has a diameter of 20 millimeter.
• 27:46
  Of course, they're going to do fine. As opposed
• 27:48
  to, oh, you treated medically all those who have
• 27:52
  a diameter of 45 millimeter. Of course, they're
• 27:54
  going to need an open thoracic abdominal repair
• 27:57
  and so on. So by stratifying, we try to risk.
• 28:02
  stratification to make sure the two arms are
• 28:05
  of equal risk. Now, if you ask me about my opinion
• 28:09
  about the high -risk features, I think they,
• 28:11
  as Jacob mentioned, they ought to be studied
• 28:14
  prospectively. I think this debate is more heated
• 28:17
  in the U .S. If I come up with something tomorrow,
• 28:20
  tear size is 12 millimeter, not 10 millimeter,
• 28:22
  it will, by the end of the day, make it into
• 28:24
  the guidelines. I mean, that's how the U .S.
• 28:26
  functioned. As opposed to... But in Europe, thankfully,
• 28:32
  there are people who actually scrutinize the
• 28:34
  data and so on. So the value of those is yet
• 28:38
  to be determined. But in the trial, we just try
• 28:41
  to make sure the two arms are equal so people
• 28:43
  in the U .S. do not criticize it as completely
• 28:47
  whacked in favor of one versus the other. Sunday
• 28:52
  excludes patients with established connective
• 28:55
  tissue disorders and pre -existing thoracal abdominal
• 28:58
  aneurysms. Some will argue this creates a healthier
• 29:01
  cohort whose results may not apply to the most
• 29:04
  complex cases on our lists. How do you defend
• 29:07
  that boundary? That's a good question. I mean,
• 29:10
  we're open to that limitation. And I guess you'd
• 29:13
  have to distinguish between generalizability,
• 29:16
  which you've mentioned a few times, and then
• 29:18
  biological coherence. And I suppose when you
• 29:22
  get into biological coherence, that's also a
• 29:26
  question because, I mean, even when we were talking
• 29:28
  about the exclusions of patients that have pathology
• 29:32
  that goes up in the arch, in fact, they're getting
• 29:34
  excluded in the UK because I guess some people
• 29:38
  might feel, well, that's a bit of a different
• 29:40
  disease process there. And I think that's really
• 29:43
  getting at the limitations of our understanding
• 29:45
  of what's going on with the dissected aorta.
• 29:52
  Having said all that, though, we kind of acknowledge,
• 29:54
  though, that connective tissue disease patients
• 29:56
  and patients with pre -existing thoracodermal
• 29:59
  aortic aneurysms, that sort of the natural treatment
• 30:03
  of these patients isn't a question of covering
• 30:06
  a primary entry. It's a different process from
• 30:10
  that point on. So I guess I'd kind of challenge
• 30:12
  that we're choosing a healthier cohort. We're
• 30:17
  just trying to do our best. And we try to do
• 30:20
  as much. I mean, we're pretty liberal. on our
• 30:22
  inclusion -exclusion criteria. But we did want
• 30:25
  to maintain some form of biological coherent
• 30:29
  cohort. And I think both Colin and Firas were
• 30:33
  into that a bit. That's just where we sort of
• 30:35
  said, okay, we're going to set boundaries here
• 30:37
  with these patients. Because I think many of
• 30:41
  us agree that they are sort of a different patient.
• 30:43
  I think their treatment trajectory, as I just
• 30:46
  mentioned, is a bit different. And that was a
• 30:48
  decision there was quite deliberate. I think
• 30:51
  that's very intriguing and may deserve a whole
• 30:54
  episode on it. So I'd like to explore the intervention
• 30:58
  in more detail because TVI is not a single standardized
• 31:01
  procedure. Professor Moussa, IMPROVE -AD permits
• 31:05
  operators to select any FTI -approved stent graft
• 31:08
  with flexibility over adjunctive techniques,
• 31:11
  including additional stent grafts, side branch
• 31:13
  stenting, and other endovascular maneuvers. You've
• 31:17
  mirrored real -world heterogeneity intentionally.
• 31:20
  But does that heterogeneity compromise your ability
• 31:24
  to attribute outcomes to TVAR itself? And how
• 31:28
  will you account for the fact that operator experience
• 31:30
  and institutional volume will vary enormously
• 31:33
  across 60 centers? Right. Very good question.
• 31:37
  Very common question. See, we're not evaluating
• 31:41
  TVAR as a procedure. So you can do the TVAR.
• 31:46
  upside down, backwards, flip it over, put it
• 31:50
  in your pocket and then clean it up, spit on
• 31:53
  it, put it back into the patient. It's not really
• 31:56
  the question that we're trying to answer. We're
• 32:02
  trying to answer strategy of treatment, of upfront
• 32:06
  treatment with a stent versus delayed treatment
• 32:10
  with a stent if needed. Okay, so that stent could
• 32:15
  be one piece, could be a piece and another piece
• 32:19
  to the cilia, could be two pieces plus a bear
• 32:23
  stent or a branch stent. As in the US, as I just
• 32:28
  mentioned to you, the more we put stents in,
• 32:31
  the more we get paid. Ernest, interestingly,
• 32:43
  does not recommend Stabilize, though it doesn't
• 32:46
  explicitly exclude it. Professor Bicknell, could
• 32:50
  you explain that decision for Ernest and Professor
• 32:52
  Batsley? When you look at the emerging data on
• 32:55
  Stabilize, which appears to improve false lumen
• 32:58
  thrombosis and aortic remodeling significantly,
• 33:01
  do you think future trials will need to treat
• 33:03
  it as a separate intervention arm rather than
• 33:06
  an object? When we came to write the trial, we...
• 33:11
  had to define what was mandatory, what was recommended,
• 33:18
  and what was excluded in our intervention. So
• 33:23
  it has to be very clear about our intervention.
• 33:24
  And we ended up with very few things that we
• 33:29
  agreed should be mandatory. Coming to the conclusion
• 33:33
  this was mainly a pragmatic trial, which is kind
• 33:36
  of what Feroz is saying. This is a... trial of
• 33:40
  a strategy versus not but of course we said you
• 33:44
  need to put the t -bar in and close the primary
• 33:48
  entry tear you need to use x -ray guided imaging
• 33:55
  you need to do a completion angiogram we need
• 33:59
  a CT after the procedure within the first six
• 34:03
  weeks so very little guidance there but we decided
• 34:07
  that we would give recommendations, which would
• 34:11
  be that you could choose the length of the stent,
• 34:17
  although we would expect you to land in a straight
• 34:20
  pass of the aorta to avoid entry tears. We would
• 34:24
  expect sizing to be 10 % or less for a dissection.
• 34:27
  We'd avoid ballooning, if not, and all these
• 34:30
  sorts of things. And within that group of things
• 34:33
  that we should and shouldn't do, we discussed
• 34:34
  adjunctive techniques. Now, we were never going
• 34:37
  to... adjunctive techniques and different other
• 34:40
  things we were never going to be so restrictive
• 34:42
  that we cut off um uh certain different procedures
• 34:47
  and and parts so we said that um we're not recommending
• 34:53
  stabilize but if you want to use stabilize or
• 34:57
  petticoats especially if you need to then that
• 35:00
  that was fine we'll allow we'll allow it in um
• 35:03
  in actual fact There are very few people that
• 35:07
  are doing Stabilize within the UK, and I don't
• 35:09
  think we're going to have great enough numbers
• 35:11
  to perform any subgroup analysis unless we join
• 35:13
  with others to have a look at the results of
• 35:16
  it. And so that's how it came about. I think
• 35:23
  that really gets at the whole circularity of
• 35:28
  the discussion. And I have great respect for
• 35:31
  people doing randomized control trials, and we've
• 35:33
  met with... people with experience in doing randomized
• 35:38
  control trials. And they said, as you can see
• 35:42
  with these three trials, that they necessarily
• 35:43
  have to be pragmatic. It's a strategy because
• 35:46
  people are doing different things. And fear has
• 35:49
  hit the nail on the head. We can't dictate. We
• 35:53
  can't put out a protocol that says, you have
• 35:56
  to do it this way. This is not going to happen.
• 36:00
  And, you know, I'm really interested in the stabilized
• 36:05
  technique. We don't do it. It's not very common
• 36:09
  in the Nordic countries. But we're talking to
• 36:12
  people who do stabilize. I guess using the word
• 36:15
  zealot is a bit of an exaggeration. But they
• 36:19
  really are convinced that you get the full remodeling.
• 36:23
  And I think everyone's aware, though, that comes
• 36:25
  at a cost. And we have some patients who've been
• 36:28
  pretty well close to stabilize. I mean, they
• 36:31
  get really a lot of coverage of the aorta. extensive
• 36:35
  treatment in the iliacs and stuff like that.
• 36:37
  And they're risky procedures. And we've had some
• 36:40
  adverse events. And you have to remember that.
• 36:42
  You have to bear that in mind that, okay, if
• 36:45
  you believe that improving remodeling improves
• 36:48
  our survival, then you're exposing these patients
• 36:50
  to some big risks. And so for those people who
• 36:54
  believe that stabilize is the way to go, then
• 36:57
  they should be also part of this trial. Having
• 37:00
  said that, I kind of agree with... With Colin,
• 37:04
  I don't know if we'll have enough patients to
• 37:06
  do any really valuable subgroup analysis. But
• 37:09
  I think it's important that they're in there
• 37:10
  because there are some sites in the Sunday trial
• 37:12
  that believe in stabilize and they want to do
• 37:14
  that. So I think that should be allowed. I would
• 37:17
  add an important thing. I mean, we talked about
• 37:20
  TAVAR and, you know, the generalizability and
• 37:23
  then different types of procedures are done out
• 37:25
  there. The same could be said for the medical
• 37:27
  arm. And the same could be said about the circularity
• 37:30
  about the medical treatment, because we had a
• 37:32
  lot of input from sites saying we should have
• 37:34
  a standard procedure for the medical treatment.
• 37:36
  And we said, well, hold on. There is no evidence
• 37:38
  for which medicine is best. So why would we standardize
• 37:42
  what should be done? That's why we do the trials
• 37:44
  to find out, is there one medical treatment that's
• 37:47
  better than another? Is there one procedure in
• 37:49
  TAVR that's better than another? That's why we
• 37:51
  do the trial instead of doing it the other way
• 37:53
  around, like standardize and then randomize patients.
• 37:56
  It's just... It's putting the horse before the
• 37:59
  carriage. For those who's going to listen to
• 38:01
  this conversation and duty to petticoat or stabilize
• 38:07
  for uncomplicated tubular section, meaning no
• 38:11
  rupture, no malperfusion, let's just be clear.
• 38:14
  And even the company that makes that stent, this
• 38:17
  is an invitation from these three trials, invitation
• 38:20
  to study this inpatient. And collect like a hundred
• 38:25
  because guess what? There is no randomized or
• 38:30
  even a well -documented cohort of patient who
• 38:34
  received a bear stent that had been studied properly
• 38:37
  and followed for many years in this patient population.
• 38:42
  Okay. And in a comparative study, not in a single
• 38:46
  arm. So that's really an invitation of ask the
• 38:50
  company that makes the bear stent. They were
• 38:53
  not interested, I think. And therefore, their
• 38:57
  sales, I think, in Europe is pretty dismal. And
• 39:01
  in the U .S., I think it's also the same thing.
• 39:05
  It's very selective sites that do it routinely,
• 39:07
  and some others never will consider it. But that's
• 39:11
  one way of doing it, is team up with us, the
• 39:14
  three of us. Now to the elephant in the room,
• 39:18
  disagreement on timing. And please correct me
• 39:20
  if I'm wrong. I think the trials. have gone through
• 39:23
  some changes. InfoVD opens its enrollment window
• 39:29
  at 48 hours after symptoms onset, meaning you
• 39:33
  are prepared to randomize patients who are still
• 39:38
  in the acute phase. Sunday says wait, no enrollment
• 39:42
  before one week, and focus on the sub -acute
• 39:45
  window up to 90 days. Ernest opens at 10 days.
• 39:49
  First of all, Lily, can you explain the reasoning
• 39:52
  behind that specific choice? Yeah, sure. And
• 39:55
  that actually, that topic or issue was the most
• 39:59
  discussed in our protocol, evolvement, evolution.
• 40:04
  I think originally we, I can't remember where
• 40:08
  we started, but we were down to seven days and
• 40:11
  then five days. And then people said, well, two
• 40:13
  days, because we do see some patients. who are
• 40:16
  quite stable and doing well. And you think, okay,
• 40:19
  they're ready to go home. But then suddenly it
• 40:23
  dawned on us, we're like, well, wait a second.
• 40:24
  If we just move it down to one day or two days,
• 40:28
  then it's a different trial. Because we're kind
• 40:31
  of constrained by the, you could argue how valuable
• 40:35
  it is, the distinction between acute and a subacute
• 40:39
  dissection. So if you start including patients
• 40:42
  at day zero, then basically you're going to the
• 40:46
  adsorb trial. You're saying, let's just treat
• 40:47
  any dissection that comes in the door. Whereas,
• 40:50
  I mean, if you go back to the thinking of the
• 40:52
  trial, it's, okay, they've become stable. Again,
• 40:56
  all these terms are so debatable. It's uncomplicated.
• 41:00
  Is there value in a prophylactic treatment for
• 41:04
  these patients? What's the timeline? Well, again,
• 41:08
  we started seven days. And if you look at some
• 41:11
  of the literature, there's a Swedish study said
• 41:12
  of patients who come in that are doing well and
• 41:16
  stable and uncomplicated, the majority of those
• 41:19
  that convert to some type of complicated position,
• 41:22
  that's within one week. And I would say, I can't
• 41:26
  remember the specifics, but I would imagine I'm
• 41:28
  pretty certain that the majority of those were
• 41:30
  within the first 48 hours. So we're very wary
• 41:32
  of going too early because if you include them
• 41:35
  and then they... for example, from the medical
• 41:38
  arm and they convert to complications, I think
• 41:40
  that would just skew our results. So we wanted
• 41:42
  to get away from the acute period somewhat. And
• 41:46
  seven days was also rather practical or pragmatic.
• 41:50
  Majority of these patients are admitted usually
• 41:52
  seven to 10 days or something like that. So that
• 41:54
  worked well for most of the departments. So,
• 41:57
  but it's a good question. Thank you. Professor
• 42:00
  Musa, improve AD will enroll patients as early
• 42:02
  as 48 hours. Professor Batsley just argued it
• 42:06
  was a bit too soon. What do you think about that?
• 42:09
  EdSorb showed that you can stent people, stent
• 42:11
  them within three days, three to five days. So
• 42:16
  we know it's safe, but we know it also could
• 42:21
  be risky if done in all patients. You approach
• 42:25
  the patient within 48 hours to enroll. We do
• 42:31
  not. recommend and most sites actually do not
• 42:34
  necessarily do not really do so we pick the 48
• 42:39
  hours so we can approach them in the hospital
• 42:41
  and they don't leave the hospital before we approach
• 42:44
  them and randomize it because they once they
• 42:47
  leave the hospital it becomes much harder to
• 42:50
  enroll a patient we approach them in 48 hours
• 42:53
  we randomize them within a week That gives them
• 42:57
  a time between the two days and a week to kind
• 43:00
  of declare if they're going to have a complication.
• 43:02
  Some of them, like in the literature, is around
• 43:05
  10 % that thing happened. Develop mild perfusion
• 43:08
  or rupture or severe pain that is not letting
• 43:11
  up. So those patients, it's just a matter of
• 43:17
  logistical flow in the U .S. Because once they
• 43:21
  leave in the U .S., it's not like in Europe where
• 43:23
  they're kind of registered to their... Zip code
• 43:26
  and this and that. Here in the U .S., you can
• 43:29
  get treated in Texas and then leave and go to
• 43:32
  Louisiana and get a T -Var or go to New York
• 43:35
  and fly in California. So that kind of thing
• 43:41
  happens to capture those patients. Thank you.
• 43:43
  Professor Vicknow, where does earnest actually
• 43:47
  stand on this time in debate? Originally, when
• 43:50
  we were planning the trial, we had a lot of debate.
• 43:54
  just as Jacob did. We decided that if you stented
• 44:02
  right at the beginning, you were stenting or
• 44:05
  not stenting all those uncomplicated patients
• 44:08
  that became complicated. And it was a slightly
• 44:11
  different trial. We wanted a very clean group
• 44:13
  of patients who had proven themselves to be uncomplicated.
• 44:17
  And then we're talking about long -term differences
• 44:20
  in complications and need for major aortic surgery.
• 44:27
  So I was with Jacob, very much so. What actually
• 44:31
  happened halfway through our starting to recruit
• 44:35
  is that a lot of centers said to us, it's really
• 44:39
  impractical 10 days because what happens is that
• 44:43
  our patients go home at five, six, seven days.
• 44:47
  And we end up having to bring them back and then
• 44:51
  you give them the patient information sheet and
• 44:52
  then they go home and they have to come back
• 44:54
  again. And people got cold feet, lost interest
• 44:59
  in trying to recruit someone. And so we actually
• 45:02
  have moved to five days based. on the fact that
• 45:07
  most patients will have had their complications
• 45:10
  by five days, 48 hours, 72 hours, almost certainly.
• 45:14
  There's only a few that are going to become complicated
• 45:16
  after that. And it's a really practical thing
• 45:19
  for us to do. So I've been swayed by both Jacob
• 45:23
  and Firaz. Now, one of the most consistent findings
• 45:27
  on favorable TIVA from a TORB or from a steady
• 45:30
  cell and from multiple observational trials and
• 45:33
  series. Is superior aortic remodeling, false
• 45:37
  lumen thrombosis, reduction in false lumen diameter,
• 45:40
  and stabilization of the aorta, first of all,
• 45:43
  is that a surrogate endpoint we should trust?
• 45:46
  Or is it possible that radiographic remodeling
• 45:49
  does not reliably translate into outcomes that
• 45:51
  matter to patients' survival, freedom from re
• 45:55
  -intervention, and quality of life? No. And that
• 45:58
  would be the hope. I think that's the rationale.
• 46:00
  That's why people say that they should be stented.
• 46:03
  because you look at the remodeling, you think
• 46:05
  it's favorable, but that's obviously where both
• 46:07
  the ADZORB and the INSTEAD trial demonstrated.
• 46:12
  Better remodeling, but no difference in survival.
• 46:15
  And I think it's interesting, a lot of the retrospective
• 46:18
  studies I've looked at, like IRAD data, for example,
• 46:22
  that there's no difference in reinterventions.
• 46:26
  So you have the medical, you have patients treated
• 46:29
  medically, and you have patients treated with
• 46:30
  the TAVAR. The patients get a TAVR, they're not
• 46:32
  done. They come back and actually numerically,
• 46:36
  there was no statistical difference, but they
• 46:39
  get re -intervened upon more often than a medical
• 46:42
  arm. So it's not like you go in, fix something,
• 46:46
  and you say, okay, now you're healthy, no need
• 46:49
  for anything more, and you'll live longer. That's
• 46:51
  just not the case. And that's, I mean, that actually
• 46:55
  in essence is why we're doing the trial. And
• 46:58
  I would add too, I... That's why we touched on
• 47:01
  it a bit before, but I'm very wary of composite
• 47:05
  outcomes. I see the value and I see how, I mean,
• 47:08
  the allure is that you get more endpoints. So
• 47:12
  you're done with your trial faster, but the hypothetical
• 47:17
  there is that you look at your outcomes, one
• 47:19
  arm will have a bunch of reinterventions, you
• 47:23
  know, sort of similar quality of life and similar
• 47:28
  mortality. But because there are more interventions
• 47:31
  in one arm, you say, okay, this is a better treatment.
• 47:33
  I think the thing that matters, and it's not
• 47:37
  the only thing, but it is survival. And then
• 47:40
  you take your secondary outcomes. It's like,
• 47:42
  what about these factors? Because my fear would
• 47:46
  be to say, yeah, composite outcome, difference.
• 47:48
  Okay, any difference in survival? The study wasn't
• 47:52
  powered for that. We can't answer that. What
• 47:54
  about this? Well, we can't answer that. So you
• 47:56
  just have a composite outcome, which sometimes
• 47:58
  is good. But I don't think it's that often the
• 48:01
  case. Now I want to turn to best medical therapy
• 48:05
  because it's the comparator arm in all three
• 48:08
  of your trials. And I think sometimes it gets
• 48:11
  taken for granted. A recent analysis by Martin
• 48:14
  and Colleagues, which incidentally includes Professor
• 48:16
  Bicknell as a co -author, highlighted that antihypertensive
• 48:20
  medication adherence in chronic type aortic dissection
• 48:23
  is poor. Only 43 % of patients were highly adjuvant
• 48:28
  and up to 21 were poorly compliant. Now, if BMT
• 48:31
  is not being delivered as intended, the comparator
• 48:34
  arm is systematically weaker than the protocol
• 48:36
  assumes. Professor Bicknell, how is urinase addressing
• 48:40
  it? I wanted to go as far as testing urine to
• 48:44
  make sure that they were taking the tablet, so
• 48:46
  an enhanced medical therapy. But in actual fact,
• 48:49
  that probably isn't the right thing to do. We
• 48:52
  needed to test standard of care now. versus one
• 48:55
  intervention and so we are taking a pragmatic
• 48:58
  approach to it. We are trying to firstly understand
• 49:03
  how best medical therapy is delivered because
• 49:06
  it's not always delivered by vascular surgeons
• 49:08
  in dissection clinics in the UK. It's very variable.
• 49:11
  We are trying to produce guideline -based recommendations
• 49:18
  for GPs. you know general practitioners uh specialists
• 49:25
  and patients we are trying to um increase the
• 49:30
  understanding of type b aortic dissection which
• 49:32
  incidentally in that paper on adherence a lot
• 49:35
  the knowledge about people's the condition that
• 49:38
  people had was really poor and probably related
• 49:40
  to whether or not they took the blood pressure
• 49:42
  tablets so knowledge uh and and uh from patients
• 49:47
  and so We are doing our best to encourage people
• 49:53
  to follow the right path with best medical therapy,
• 49:56
  but we have to accept that there are some people
• 49:58
  that won't take it, presumably because it makes
• 50:00
  them feel awful. And if it makes them feel awful
• 50:03
  and they don't take it, maybe TVAR is the right
• 50:07
  thing to do if it improves their quality of life.
• 50:10
  What about Lily? Some days registry linkage means
• 50:13
  your data might ultimately extend to 10 years.
• 50:17
  If the five -year primary endpoint is neutral,
• 50:20
  but late mortality diverges beyond that, as instead
• 50:23
  Excel suggested it might, how does the feed handler
• 50:26
  result that keeps evolving? And at what point
• 50:29
  do we just call it? Yeah, I wish it could be
• 50:32
  that black and white and say, okay, we call it
• 50:34
  at this point. But that's kind of the beauty
• 50:36
  and the advantage of doing a trial. You could
• 50:38
  see. You know, the AVAR1 trials Colin mentioned,
• 50:41
  you know, we followed that avidly, you know,
• 50:47
  for years. You just, there's still value, actually,
• 50:51
  maybe even more value as you go farther out.
• 50:56
  And it sort of brings me back to the point about
• 50:58
  my concern about composite outcomes, because
• 51:01
  comps outcomes get you done faster. So you can
• 51:05
  get results out at like two or three years. But
• 51:08
  the long -term results, I think, are the ones
• 51:09
  that are really interesting. Those are the ones
• 51:13
  that really were the nuances that help you in
• 51:16
  your discussions with patients. Like the small
• 51:21
  aneurysm trial, for example, that outcome was
• 51:25
  death. And if you look at a lot of the patients
• 51:30
  who were just under surveillance, they ultimately
• 51:33
  did undergo treatment for their aneurysm. I think
• 51:36
  it was 75%. That's not a failure. You know, people
• 51:40
  might say, oh, well, 75 % of the patients crossed
• 51:43
  over. It's not. You can still say to your patient,
• 51:45
  look, there's no benefit for us to treat you
• 51:49
  early and prophylactically. We can wait. And
• 51:52
  yes, you may get treated later. That's still
• 51:54
  super valuable information. So I think long -term
• 51:57
  data is as valuable, if not more. I think that's
• 52:02
  very interesting. Right now, today, each of your
• 52:05
  teams is sitting in front of patients with a
• 52:08
  fresh dissection, asking them to consent to a
• 52:10
  trial where we genuinely do not know which arm
• 52:13
  is better. That is the honest truth of Fickley
• 52:16
  Police, and it is also for many patients a deeply
• 52:19
  unsettling thing to hear. They have just survived
• 52:22
  a catastrophic aortic event. They want you to
• 52:25
  tell them what to do. Professor Bickner, how
• 52:28
  do you have that conversation? How do you communicate
• 52:31
  genuine uncertainty to someone in that position?
• 52:34
  It's not me that doesn't know. It's the whole
• 52:36
  world that doesn't know. And I like to say that
• 52:39
  we've got this trial up and running. And the
• 52:42
  reason we've got it up and running successfully
• 52:44
  is because there is real clinical equipoise in
• 52:50
  terms of what we should be doing. I think in
• 52:55
  general, patients have been really, really receptive
• 52:58
  to listening about the trial and to working it
• 53:01
  through with us. We have got some, you know,
• 53:05
  the usual clinical patient information sheets
• 53:07
  and videos and decision aids, which help the
• 53:10
  patients through deciding. I was really worried
• 53:14
  when I started that this was asking patients
• 53:17
  to do quite something quite. significant to have
• 53:22
  the rest extent put inside them uh for this trial
• 53:27
  and that we will get a really high refusal rate
• 53:29
  but we haven't and uh i guess i'm just very proud
• 53:35
  of all of our research staff and nurses and all
• 53:38
  the consultants around the country they must
• 53:40
  be giving the right message that we really need
• 53:43
  to know the answer to this question. Professor
• 53:45
  Moussa, Professor Bicknor, Professor Baxlili,
• 53:48
  thank you. Three trials and a question that has
• 53:50
  resisted a clean answer for longer than any of
• 53:53
  us would have liked. Perhaps in a few years we
• 53:56
  will not have solved it entirely, but we will
• 53:59
  have earned the right to stop guessing. In a
• 54:02
  field where the temptation to act is always present,
• 54:05
  these trials represent something important, a
• 54:08
  commitment to finding out first. Thank you for
• 54:10
  listening to the ESVS podcast. We will include
• 54:12
  key references and links to the trial protocols
• 54:15
  in the episode notes. We hope to welcome you
• 54:18
  back when the first results begin to emerge.