Latest / European Society for Vascular Surgery / Vascular Graft and Endograft Infection – Building a Common Language with T. Wyss and K. Sörelius

Transcript

Vascular Graft and Endograft Infection – Building a Common Language with T. Wyss and K. Sörelius · European Society for Vascular Surgery

• 0:13
  ESVS Podcasts. This is Safa Riahi, one of the
• 0:17
  new voices of the ESVS Podcasts, and today we're
• 0:21
  recording an episode focused on something that
• 0:23
  may sound deceptively simple, definitions. The
• 0:28
  discussion centers on a recently published Delphi
• 0:31
  consensus document that brings together international
• 0:34
  multidisciplinary experts to agree on terminology,
• 0:38
  definitions, treatment strategies, outcomes,
• 0:42
  follow -up, and reporting standards in vascular
• 0:45
  graft and endograft infection. Because when it
• 0:49
  comes to VGEI, we all recognize the condition,
• 0:52
  but when we try to define it, classify it or
• 0:56
  even report outcomes or compare studies, we quickly
• 1:00
  realize we are not always speaking the same language.
• 1:03
  And without shared language, meaningful science
• 1:06
  becomes almost impossible. To explore that, I'm
• 1:10
  joined by Dr. Tom Wyss and Dr. Karl Sauerlis.
• 1:13
  Tom is the head of the vascular surgery department
• 1:16
  at the Canton Hospital Bündeter in Switzerland,
• 1:19
  and Karl is associate professor and vascular
• 1:22
  surgery consultant at the Lick Hospital at Copenhagen,
• 1:25
  Denmark. They are the lead authors of the consensus
• 1:29
  that set out to standardize terminology, definitions,
• 1:32
  and outcomes in vascular graft and endograft
• 1:35
  infection. Tom and Karl, welcome and thank you
• 1:39
  for joining us. Hi, Salva, and thanks for hosting
• 1:42
  this podcast. Great to be here. Hi, and thanks
• 1:45
  for having us. Before we dive into methodology,
• 1:48
  let's briefly level set for some of our trainees
• 1:51
  and early career colleagues listening in. When
• 1:54
  we talk about vascular graft and endograft infection,
• 1:57
  we are dealing with a condition that is relatively
• 1:59
  rare, but associated with extremely high morbidity
• 2:03
  and mortality. It can involve open surgical grafts,
• 2:07
  endovascular devices, aortic or peripheral reconstructions
• 2:10
  and may present with anything from sour systemic
• 2:13
  symptoms to catastrophic bleeding. And yet, course
• 2:17
  studies the terminology varies early vs late
• 2:21
  infection graft and trach erosion vs fistula
• 2:24
  cure vs remission and even the definition of
• 2:27
  treatment failure is inconsistent Karl was that
• 2:30
  variability the starting point for this project
• 2:33
  yes the variability and perhaps even more the
• 2:38
  lack of important terminology and definitions
• 2:42
  The new updated ESVS guidelines of VGEI will come
• 2:46
  out later this year with Anne Lejay as first
• 2:49
  author. And upon planning these guidelines, the
• 2:53
  working group earlier recognized a lack of standardized
• 2:56
  terminology and definitions. So the first assignment
• 3:01
  was to standardize terminology, definitions related
• 3:05
  to the disease, as well as treatment approaches,
• 3:08
  outcomes and reporting standards. Tom and I got
• 3:12
  this task, together with Robin Verhoek, Oliver
• 3:15
  Lyons, Mathilde Boucher and Xavier Berard, probably
• 3:19
  since we previously had done some similar work
• 3:21
  for infective native aortic aneurysms. One could
• 3:24
  argue that there are already SVS guidelines addressing
• 3:27
  vascular graft and endograft infection, the most
• 3:30
  recent version published in 2020, and Tommy were
• 3:34
  involved in that document. First, looking back
• 3:37
  at the 2020 guidelines, What has evolved in the
• 3:40
  field since then that makes an update necessary
• 3:42
  now? Well, obviously, you know, in six years
• 3:46
  a lot evolves in medicine and like guidelines
• 3:50
  and recommendations, they need to be updated
• 3:53
  according to the new evidence, to new literature
• 3:57
  on a regular basis. So you can definitely look
• 4:00
  forward to the publication of this new version.
• 4:05
  But like Karl stated before, upon planning the
• 4:08
  new version of the guidelines, some specific
• 4:11
  drawbacks were recognized. So we tried to solve
• 4:13
  this problem by creating suggestions for terminology,
• 4:18
  treatment methods, outcome, follow -up protocol,
• 4:22
  and even multidisciplinary team management is
• 4:26
  another cornerstone which we dealt with. why
• 4:31
  a delphi consensus needed in addition to a guideline
• 4:34
  update was the core issue not a lack of recommendations
• 4:38
  but rather a lack of standardized terminology
• 4:41
  and shared definitions and at what point did
• 4:44
  it become clear that without a common language
• 4:47
  comparing studies and even interpreting registry
• 4:49
  data becomes fundamentally unreliable well the
• 4:54
  recommendations for management and treatment
• 4:56
  will be outlined in the new guidelines but you
• 5:00
  are spot on when you point out the need for a
• 5:03
  common language. We did this to create a basis
• 5:06
  for the new guidelines, and this is the terminology
• 5:09
  that is used there. And in summary, the Delphi
• 5:13
  document offers a globally relevant groundwork
• 5:16
  of expert consensus on VGEI to make the guidelines
• 5:20
  more comprehensible and to facilitate and hopefully
• 5:24
  improve research in the field. As a practical
• 5:28
  example, What does treatment of VGEI really mean?
• 5:32
  Most VGEIs are treated with a combination of antimicrobial
• 5:37
  treatment, drainage, and surgery. But some are
• 5:41
  treated with antimicrobial medicine only, and
• 5:44
  some patients have their entire grafts explanted
• 5:47
  and others just partially. For some patients,
• 5:51
  the treating multidisciplinary team has identified
• 5:54
  the microbiological pathogen. and for others
• 5:58
  it has not. So we defined treatment approaches
• 6:02
  accordingly. We specified what a conservative
• 6:06
  treatment is and adding definitions on targeted
• 6:10
  antimicrobial treatment and empirical treatment
• 6:14
  coupled with if the treatment had the intention
• 6:17
  of possibility of being curative or only suppressive.
• 6:21
  We also defined diverse surgical approaches and
• 6:27
  put endovascular treatment into its proper context,
• 6:30
  because in order to evaluate these diverse possible
• 6:33
  treatment approaches, they needed proper definitions.
• 6:37
  For listeners who may not be familiar with the
• 6:40
  Delphi methodology, Tom, could you walk us through
• 6:43
  how this process worked? How were the initial
• 6:46
  statements generated? How many rounds were performed?
• 6:50
  How did anonymous voting help define agreement
• 6:53
  and importantly what threshold did you use to
• 6:56
  define consensus? The author group drafted propositions
• 7:01
  for statements at first, then these were sent
• 7:06
  to the guideline writing group for revision and
• 7:09
  check and were then finalized. And in parallel,
• 7:14
  we were inviting the world's leading experts
• 7:18
  in this disease so to speak vascular surgeons
• 7:21
  cardiovascular surgeons specialists in infectious
• 7:26
  disease radiology and nuclear medicine all of
• 7:33
  these experts they were not aware of who else
• 7:35
  was participating in the survey and everybody
• 7:39
  could comment and suggest revisions of the statements
• 7:43
  anonymously So the statements were then refined
• 7:48
  during the subsequent Delphi rounds according
• 7:51
  to the experts' comments in a total of four to
• 7:56
  five anonymous rounds. And a very important point
• 8:01
  here is to create like an open climate where
• 8:05
  everyone's vote is weighed equally and not just
• 8:09
  the one who shouts the loudest gets to speak.
• 8:14
  The other question concerning consensus, this
• 8:18
  was defined as 75 % or above of the expert panel
• 8:24
  rating a statement as strongly agree or agree
• 8:30
  on the Likert scale we used. Now one of the strengths
• 8:34
  of this consensus is the breadth of expertise
• 8:37
  involved. The DALFI panel consisted of 43 international
• 8:41
  experts from across Europe, North America, Oceania,
• 8:45
  Asia and Africa. We presented vascular surgery,
• 8:50
  cardiothoracic surgery, infectious disease microbiology,
• 8:53
  radiology and nuclear medicine. And notably all
• 8:57
  43 experts completed all planned rounds of the
• 8:59
  DALFI process. The 100 response rate across four
• 9:03
  rounds is quite remarkable. and suggests a strong
• 9:07
  level of engagement and commitment to the topic.
• 9:09
  So this truly reflects a global and generally
• 9:12
  multidisciplinary perspective. Now Karl, when
• 9:16
  the first round results came back, did you see
• 9:18
  clear patterns where certain specialties thought
• 9:21
  differently about key concepts like diagnosis,
• 9:24
  cure or follow -up? Were there debates that surprised
• 9:29
  you? And did surgeons and infectious disease
• 9:31
  specialists see cure differently? Well, surprisingly,
• 9:36
  there were no big differences between different
• 9:39
  medical specialties. Perhaps this can be explained
• 9:43
  by the fact that we had included experts with
• 9:46
  big experience in this disease. What we did notice,
• 9:50
  however, was that certain specialties tended
• 9:53
  to be more neutral during the first round of
• 9:57
  the statements, where the statements were on
• 10:00
  subjects outside of their area. expertise but
• 10:05
  this we only saw as a sign of upright and honest
• 10:08
  and humble participation on their behalf. One
• 10:11
  of the most interesting findings was the lack
• 10:13
  of strong agreement around traditional early
• 10:16
  versus late infection classification, something
• 10:19
  many of us were trained to use almost automatically.
• 10:22
  Tom, why do you think this distinction proved
• 10:25
  difficult to standardize, and do you think we
• 10:27
  might have been oversimplifying a biologically
• 10:30
  complex process by relying on time -based definitions
• 10:33
  alone? Well, Sava, this is truly a great question.
• 10:38
  You know, I mean, time -dependent classification
• 10:42
  into, let's say, early and late is something
• 10:46
  that generally feels right for many of us. And
• 10:51
  this is seen in many fields of medicine, we all
• 10:54
  know. But during the Delphi process, we just
• 10:58
  realized that this four months cutoff was rather
• 11:03
  arbitrarily set and does not have much backup
• 11:08
  in the literature as such. So we just decided
• 11:11
  this cutoff is not something to further focus
• 11:16
  on, I guess. you even added an extra round to
• 11:19
  clarify the distinction between secondary graft
• 11:22
  enteric erosion and graft enteric fistula that
• 11:25
  suggests this was not just semantic it had clinical
• 11:29
  implications karl could you please explain why
• 11:32
  the differentiation was so important this is
• 11:37
  extremely important in my view in fact we defined
• 11:40
  three similar but crucially different states
• 11:44
  which became secondary graft enteric partial
• 11:47
  erosion secondary graft enteric fistula and secondary
• 11:52
  aorta enteric fistula as as an example why this
• 11:57
  matters there is a wide range in the reported
• 12:02
  short -term survival after treatment of abdominal
• 12:05
  aortic VGEI between 10 to 50 percent and the most
• 12:11
  important explanation to this is whether or not
• 12:14
  VGEI and VGEI is complicated by fistulas are included
• 12:20
  in a study since they have completely different
• 12:23
  prognosis and demands on both surgical and antimicrobial
• 12:28
  treatment these are two distinct separate entities
• 12:31
  although of course being closely related and
• 12:36
  further If there is no clear distinction made
• 12:40
  between secondary orthoenteric erosions and fistulas,
• 12:45
  they might also be lumped together when evaluating
• 12:48
  treatment outcomes. And this would severely affect
• 12:52
  the survival estimates, especially when one is
• 12:55
  a bleeding catastrophe and the other one is a
• 12:59
  perioperative finding, which might not have any
• 13:02
  more implications for the patient or outcome
• 13:05
  of the surgery. Do you think this clarification
• 13:08
  will meaningfully change how cases are reported
• 13:11
  or even managed? Well, let's hope our fellow
• 13:16
  colleagues will read the study and find that
• 13:19
  they think it makes sense and follow the reporting
• 13:22
  standards which we have proposed. If they would
• 13:25
  do that, for sure, the quality of new research
• 13:28
  papers on this rare disease will increase and
• 13:32
  we will gain knowledge. At least, that is our
• 13:35
  aspiration and hope with this work. Let's move
• 13:38
  to outcomes, because I think this is one of the
• 13:40
  most impactful aspects of the paper. You introduce
• 13:43
  structures definition for cure, remission and
• 13:46
  treatment failure. Cure is defined as one year
• 13:49
  after cessation of antimicrobial therapy without
• 13:52
  signs of infection. That is a powerful statement.
• 13:56
  Tom, why was it necessary to go beyond crude
• 13:58
  endpoints such as survival? You are absolutely
• 14:02
  right. It is a powerful statement. And we wanted
• 14:06
  it to be strong. Very much so. Because I think
• 14:11
  weak, unspecific statements, they will not get
• 14:15
  us anywhere in this process. Only reporting survival
• 14:21
  is too simple, in our view. Specifically for
• 14:25
  a disease like vascular graft and dendrograft
• 14:29
  infections. You know, for some patients, there
• 14:32
  might not be a realistic treatment plan for cure,
• 14:36
  but where you aim for more for disease suppression
• 14:40
  and relief of symptoms. And for other patients,
• 14:45
  cure is the ultimate goal. For example, remission
• 14:51
  of the disease is another term that needed clarification
• 14:54
  and treatment failure as well. So we believe
• 15:00
  that definitions are essential to be able to
• 15:04
  successfully implement these variables into research.
• 15:09
  Was there hesitation when defining cure in a
• 15:13
  disease where long -term suppressive antibiotics
• 15:16
  and late relapses are realities? Absolutely.
• 15:21
  Many of these definitions are arbitrary, so they
• 15:26
  might have to be revised accordingly. to future
• 15:29
  research let's just see practically speaking
• 15:33
  does this redefine how long we should follow
• 15:35
  patients well let's see about that but but it
• 15:40
  sends a clear message that these patients they
• 15:43
  need to be followed strictly for a long time
• 15:47
  now a fair question that often arises with consensus
• 15:51
  work is how much weight should we give it this
• 15:54
  is not randomized data to cut How do you respond
• 15:58
  to clinicians who might say, this is expert opinion,
• 16:02
  why should it change my practice? That is a good
• 16:05
  question. Well, first off, they are completely
• 16:08
  right. It is expert opinion. But all experts
• 16:13
  participating in this study are the researchers
• 16:17
  who have published the most and have huge experience
• 16:20
  in this disease. So I think it makes sense to
• 16:25
  pay attention to a document. which is an aggregation
• 16:29
  of all these experts' opinions on this very challenging
• 16:32
  subject, which lacks so much robust data and
• 16:36
  evidence for optimal care. And unless we start
• 16:39
  speaking the same language when reporting on
• 16:41
  this disease and standardize research, we will
• 16:44
  never advance our knowledge. Tom, you also addressed
• 16:48
  internal consistency statistically reporting
• 16:51
  fairly high agreement levels. Well, listeners,
• 16:54
  what does that tell us about the rudeness of
• 16:56
  the process? That's an interesting point as well.
• 17:01
  I mean, we had high internal consistency of about
• 17:05
  0 .9 for all rounds, which tells the readers
• 17:10
  that most experts eventually agreed on the statements.
• 17:16
  On the other side, if there would have been a...
• 17:19
  let's say 100 % agreement on all statements from
• 17:22
  the start, could also be argued that this study
• 17:25
  hadn't even been necessary, because there already
• 17:31
  was consensus on these topics. Now let's make
• 17:34
  this very practical. If I'm a vascular surgeon
• 17:37
  performing an audit review of my own institutional
• 17:40
  data on graft infections tomorrow, How should
• 17:44
  this consensus influence how I design that study?
• 17:46
  And in the multidisciplinary team setting, does
• 17:49
  this help align discussions between surgeons
• 17:51
  and infectious disease colleagues? Well, I would
• 17:55
  recommend to read the study to better understand
• 17:58
  the statements and then try to use the terminologies
• 18:01
  and definitions proposed in order to categorize
• 18:05
  your data. And then try to follow our suggested
• 18:09
  advice for reporting standards. because probably
• 18:13
  one study alone won't change practice of this
• 18:18
  disease. But if most new studies use the same
• 18:21
  language and categorize treatment approaches
• 18:24
  the same way and report outcomes equally, we
• 18:29
  can start to perform meta -analysis and then
• 18:32
  we might become wiser in how to best take care
• 18:36
  of these patients. But if we don't do this, our
• 18:39
  knowledge horizon won't expand. Tom, do you see
• 18:44
  this as a static document or the beginning of
• 18:47
  a living framework that will evolve as imaging
• 18:50
  modalities, microbiology techniques and treatment
• 18:52
  strategies change? And do you hope this will
• 18:55
  ultimately enable more meaningful multi -centre
• 18:57
  studies and perhaps even future trials of VGEI?
• 19:02
  Static will not get us anywhere, I believe. I
• 19:07
  see it as a beginning and with new research coming
• 19:10
  up, proving some statements may be wrong or less
• 19:14
  useful, a new Delphi might be needed. So in your
• 19:19
  words, I would say, yeah, it is a living framework.
• 19:23
  Luckily, evolution does not stop. As we begin
• 19:27
  to wrap, Karl, I'd like to ask you for one practical
• 19:31
  take -home message for clinicians listening today.
• 19:34
  What is one thing they should start doing differently
• 19:37
  when describing or reporting vascular graft and
• 19:40
  endograft infections? I have two things. First,
• 19:45
  to realize that an abdominal aortic vascular
• 19:49
  graft and endograft infection and an abdominal
• 19:52
  aortic vascular graft and endograft infection
• 19:55
  complicated by a secondary aortic fistula are
• 19:59
  two completely different diseases. And this is
• 20:02
  fundamental. So these two should not, generally
• 20:06
  speaking, be mixed in cohorts when evaluating
• 20:10
  treatment outcomes. And the second is, when evaluating
• 20:14
  treatment for VGEIs, they should be separated
• 20:18
  by their anatomical location. Thank you for helping
• 20:23
  us unpack the rationale, the process, and the
• 20:25
  implications of the stealthy consensus. What
• 20:28
  this work offers is a shared language, and in
• 20:31
  a field as complex and high -risk as vascular
• 20:33
  graft and undergraft infection, that may be the
• 20:36
  foundation upon which better evidence and better
• 20:39
  care is built. For anyone involved in treating
• 20:42
  or researching VGEI, this document is likely
• 20:46
  to become an important reference point. Thank
• 20:49
  you for listening to the ESVS podcast, and we'll
• 20:51
  introduce key references in the episode show
• 20:54
  notes. Thank you, Safa, and bye. Thank you. Bye.
• 20:58
  For our listeners, we'll be back soon with more
• 21:01
  ESVS podcasts. Have a nice day. Talk to you soon.
• 21:04
  And bye for now.