Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
TTM Post-Cardiac Arrest
12 Plays4 years ago
In this episode of Critical Matters, we will discuss the results of the recently published TTM2 clinical trial and the current state of evidence for Targeted Temperature Management in the post cardiac arrest patient.
Our guest is Dr. Niklas Nielsen. Dr. Nielsen is an associate professor at the Department of Clinical Sciences at Helsingborg, Lund University, and consultant in Anesthesiology and Intensive Care at Helsingborg Hospital. He was the principal investigator for the TTM1 and TTM 2 clinical trials.
Additional Resources:
The International Cardiac Arrest Registry (INTCAR): https://bit.ly/2Xbe0yt
Hypothermia versus Normothermia after Out-of-Hospital Cardiac Arrest: https://bit.ly/3nad5Jo
Targeted Temperature Management at 33C versus 36C after Cardiac Arrest: https://bit.ly/38THRO9
Targeted Temperature Management for Cardiac Arrest with Nonshockable Rhythm: https://bit.ly/2X19aUj
2020 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care: https://bit.ly/3z0k7T3
Books Mentioned in this Episode:
Oxygen: The Molecule that Made the World by Nick Lane: https://amzn.to/2WYMmVz
Anna Karenina by Leo Tolstoy: https://amzn.to/3yVao0v
Recommended
Transcript
Introduction to Critical Care Topics
00:00:06
Speaker
Welcome to Critical Matters, a sound podcast covering a broad range of topics related to the practice of intensive care medicine.
00:00:14
Speaker
Sound provides comprehensive critical care programs to hospitals across the country.
00:00:19
Speaker
To learn more about our programs and career opportunities, visit www.soundphysicians.com.
00:00:26
Speaker
And now your host, Dr. Sergio Zanotti.
Targeted Temperature Management Overview
00:00:32
Speaker
Current guidelines recommend targeted temperature management to prevent hypoxic ischemic brain damage in comatose patients post-cardic arrest.
00:00:40
Speaker
Targeted temperature management, or therapeutic hypothermia, has become a standard practice in the ICU for patients post-cardic arrest.
TTM2 Clinical Trial and Hypothermia Research
00:00:48
Speaker
In today's episode of the podcast, we will discuss the results of the recently published TTM2 clinical trial and the current state of evidence for targeted temperature management in the post-cardic arrest patient.
00:01:00
Speaker
Our guest is Dr. Nicholas Nielsen.
00:01:02
Speaker
Dr. Nielsen is associate professor at the Department of Clinical Sciences at Helsingberg Lund University and consulted in anesthesiology and intensive care at Helsingberg Hospital.
00:01:13
Speaker
He did his PhD on induced hypothermia after cardiac arrest with the clinical focus organizing the International Hypothermia Network Registry, now renamed the International Cardiac Arrest Registry or INSCAR.
00:01:25
Speaker
Dr. Nielsen was the principal investigator for TTM1 and TTM2 clinical trials.
00:01:31
Speaker
We are truly honored to have him on the podcast to discuss this very important topic.
00:01:35
Speaker
Nicholas, welcome to Critical Matters.
00:01:39
Speaker
Thank you so much.
00:01:40
Speaker
It's an honor to be here.
00:01:43
Speaker
Excellent.
00:01:44
Speaker
Well, I think that before we dive into the TTM2 trial and what it really means for clinicians at the bedside, perhaps we could start with a high-level clinical historical context of where we started in the early 2000s with Haka and Bernard
00:01:59
Speaker
and how that led to a widespread adoption of therapeutic hypothermia or what we now know as targeted temperature management.
00:02:07
Speaker
Yeah, that sounds good.
00:02:11
Speaker
Do you want me to elaborate a little bit on that?
00:02:14
Speaker
Yeah, just give us your perspective of how we started.
00:02:16
Speaker
And this also, I think, I'm sure gave you an impetus to pursue your PhD in this area as well.
00:02:23
Speaker
Yeah, absolutely.
00:02:26
Speaker
The Bernard trial and the Haka trial they came from some observational studies that were performed in humans but mainly were results of the animal research that had been done.
00:02:44
Speaker
the decades before and I think the animal research at that time was quite compelling and it indicated that therapeutic hypothermia would be an effective intervention to mitigate hypoxic ischemic encephalopathy.
00:03:05
Speaker
And really it was done in mice and rats and most of the evidence was when the models were created to investigate ischemia and hypothermia during the ischemia and immediately after.
Impact of Early Hypothermia Trials on Practice
00:03:24
Speaker
But then we had a few studies coming out, especially Fritz Dรผrth studies on dogs, where they could also say that delayed cooling the hours after the ischemic event was effective.
00:03:40
Speaker
And that led to the observational studies and then to these randomized trials.
00:03:47
Speaker
I think they caught the intensive care society in a period where there were quite a few interventions that came out that were really
00:04:02
Speaker
moving the field forward it was the tight sugar glucose, tight sugar control, it was the Xigris studies, the Prowse studies and these early corticosteroid sepsis studies and then the hypothermia study.
00:04:22
Speaker
So it was an era when it was quite
00:04:27
Speaker
interesting to be in the intensive care environment and it was a lot of hope and I think that caught the interest of the community.
00:04:38
Speaker
And then also it was a group of patients, the cardiac arrest population that was
00:04:45
Speaker
Everyone thought it was a really, really poor prognosis and it was a kind of a nihilistic approach to these patients.
00:04:55
Speaker
I know that from my hospitals that the cardiac arrest patient was
00:05:00
Speaker
really not treated optimally.
00:05:02
Speaker
It was rather informing the relatives that this patient will die and just take care of that.
00:05:10
Speaker
And then this happened with these trials coming out and it transformed practice the years after.
00:05:18
Speaker
Absolutely.
00:05:19
Speaker
And as you mentioned, it was quite an exciting time.
00:05:22
Speaker
I recall being at the beginning stages of my career in critical care and
00:05:28
Speaker
very rapidly guidelines adopted the positive HACA trial and the Bernard results as evidence for an intervention that we could start doing.
00:05:39
Speaker
There were hypothermia conferences.
00:05:43
Speaker
There was a large consensus conference that I had an opportunity to participate as faculty.
00:05:49
Speaker
And that's where the concept of targeted temperature management emerged and really a lot of energy around this idea
00:05:58
Speaker
of an active intervention that could ameliorate or treat anoxic brain injury post-cardiac arrest.
00:06:06
Speaker
And they kind of became the standard, right?
00:06:07
Speaker
I mean, most guidelines in Europe and North America would recommend that at least for out-of-hospital V-fib or V-tac arrest that we should be doing cooling or targeted temperature management in those patients who did not wake up.
00:06:25
Speaker
But very quickly also, it got adopted into in-hospital, other rhythms, and we kind of continued.
Insights from TTM1 Trial and Challenges in Design
00:06:34
Speaker
So can you set the stage for your first trial, TTM1?
00:06:39
Speaker
And what were you thinking?
00:06:40
Speaker
What were you thinking were the holes and how that trial came about?
00:06:43
Speaker
And maybe give us just a short summary of the results and what you found.
00:06:47
Speaker
Yeah, actually it started already in 2002 because in Sweden and in some other Scandinavian countries there was an initiative to perform a randomized controlled trials
00:07:04
Speaker
And then these two trials came out in the same issue with New England Journal of Medicine and with the Hakan-Bernhard trial in press, the group behind this randomized trial said,
00:07:20
Speaker
maybe we should start to implement the intervention instead.
00:07:25
Speaker
So we had a very rapid spread and implementation in Scandinavia, in Sweden, in Finland, in Denmark, in Norway.
00:07:36
Speaker
But at the same time we had some critical voices from the health authorities that we were too fast and as an answer to that we started the registry it was called the hypothermia network registry and we collected clinical data at that time and
00:07:58
Speaker
just to see what would the effect be when we implement this in clinical practice and also as you said the indications widened from the initial vfib, vtac and cardiac cause to basically all cardiac arrest, unconscious cardiac arrest.
00:08:19
Speaker
So we started this registry and when we started to
00:08:23
Speaker
get a fair number of patients in the registry we also performed analysis and a little bit surprising we could not demonstrate or see that there was an effect of cooling longer or cooling deeper or cooling faster or initiating faster so there were actually no real signals and so when we
00:08:48
Speaker
wanted to study this further and see how could we take TTM or therapeutic hypothermia to a next level, what would the should we look at duration, should we look at timing.
00:09:03
Speaker
We realized we couldn't find any signal that was a good start for a trial and then we went back and started to reassess the
00:09:15
Speaker
the trials that were behind these recommendations, the Haka trial, the Bernard trial, the Hachimi Idrisi trial and other trials.
00:09:23
Speaker
And at that time also the GRADE concept had started to evolve and the
00:09:30
Speaker
the demands on a randomized trial had increased compared to the 90s and the early 2000s.
00:09:39
Speaker
And with that we started to realize that the evidence behind this whole therapy was a little bit too thin.
00:09:51
Speaker
and then we wanted to redo the HACA trial and that was what I had in my thesis as the last piece that was that we should we should redo the HACA trial that was the conclusion really from my thesis and at that time that was impossible
00:10:16
Speaker
and the concept was introduced everyone believed that this was something that was really good and increased survival and decreased the risk of neurological impairment so we realized we cannot do this trial so that was when we
00:10:39
Speaker
started to think of is there any other way we could do it and looking at the registry data we found that the initial temperature for almost all patients coming in was 36 degrees and then we said hey let's just keep that temperature in one group and lower the other to 33 and that was the concept behind the TTM1 trial
00:11:01
Speaker
And with that design we could attract a substantial amount of hospitals and perform that trial in a little bit more than two years between
Reactions to TTM1 Findings and Practice Changes
00:11:14
Speaker
2010 and 2013.
00:11:14
Speaker
We included almost 1 000 patients randomized to 33 and 36 degrees and as many know we could not demonstrate any difference at all between these two groups and
00:11:32
Speaker
And that shook the concept of TTM and therapeutic hypothermia in 2013 when that came out.
00:11:43
Speaker
And I think that it really, it's very interesting when you look at how the results of TTM1 were interpreted by different groups or different clinicians, right?
00:11:54
Speaker
Some people took the adoption that, well, it doesn't really show that what I was doing
00:12:00
Speaker
is inferior, so I'll keep doing what I was doing and they keep doing the 33.
00:12:04
Speaker
Other people felt this was a larger trial and that there might be advantages of not intervening to 33 and trying to keep it at 36, but that still requires active interventions.
00:12:17
Speaker
And then I think another camp basically started really having doubts of whether this is really an indication that perhaps, like you said, what we thought was a positive intervention
00:12:30
Speaker
at the end maybe perhaps had no clinical benefit from a clinical perspective.
00:12:35
Speaker
How did TTM1 impact the practice in your hospital?
00:12:40
Speaker
In my hospital, the practice changed the Monday after the presentation we presented on a Sunday and the first cardiac arrest patient that came in the Monday after
00:12:52
Speaker
was treated according to the 36th arm of the TTM1 trial and we continued with 36 for many years up until the start of the TTM2 trial.
00:13:05
Speaker
But it's actually funny because 36 is as you said a temperature that needs active intervention and it's in the sweet spot of shivering.
00:13:15
Speaker
So it's actually a temperature that's not very clinically
00:13:22
Speaker
easy to apply I think.
00:13:25
Speaker
It requires sedation and anti-shivering therapies.
00:13:33
Speaker
33 I think is easier to keep as you move out of the shivering region.
00:13:42
Speaker
So 36 was a temperature that we chose because of
00:13:47
Speaker
the possibility of doing the trial and of ethical reasons but not really from a clinical perspective.
00:13:56
Speaker
So I think that the, we'll come back to the design of the TTM2 trial but I think that is more to the point and it was really the design that we wanted to have for TTM1 but we could not at that moment.
00:14:11
Speaker
But as you said, I think it's absolutely so that some thought that yeah, TTM1 did not prove that 33 was inferior, so we can continue with 33.
00:14:23
Speaker
Some changed to 36 as we did and some maybe I can agree with that, that it is a reasonable way of assessing the TTM1 trial is that
00:14:41
Speaker
the evidence is not there for the majority of cardiac arrest patients.
00:14:49
Speaker
And with that result, it was okay to not perform any TTM or therapeutic hypothermia.
00:14:59
Speaker
Absolutely.
00:15:00
Speaker
And another, Nicholas, another important question that had remained a post-TTM1, which was tried to answer with Hiperion was,
00:15:10
Speaker
the debate or the discussion of different rhythms, right?
00:15:15
Speaker
What happens when you have a V-fib, V-tac type of arrest, which as we know is most likely originated from a cardiac source, it's most likely to be treatable or has interventions that we do know can make a difference long-term for the origin purpose and may be associated with a lower morbidity and mortality.
00:15:37
Speaker
versus what happens when people come with non-shockable rhythms, right, a systole PEA, which we usually understand as being more severe or a mechanism that might be more complicated or more prolonged.
00:15:51
Speaker
And can you comment on that piece and maybe a little bit of how you interpreted Hiperion, which came, I think, a couple of years after TTM1?
00:16:00
Speaker
Yeah, it's actually quite a few years, 2019 it was published, but it was started just after TTM1 was published.
00:16:10
Speaker
And I think it was a reasonable move forward.
00:16:14
Speaker
We had, of course, some non-shockable rhythms also in the TTM1 trial, and we had no signal that it would be of benefit of 33 or 36 in that subgroup.
00:16:28
Speaker
But based on that they started out with the Hyperion trial and one thing is that I think with the mortality that they have and they also estimated before the trial started the trial is a little bit on the small side.
00:16:54
Speaker
I think a reasonable size for that trial would have been maybe 3,000 or 4,000 or even more patients because the whole signal that comes from the Hyperion trial is based on very very few patients.
00:17:17
Speaker
I think it's about 30 patients that are
00:17:21
Speaker
alive and that were followed for neurological follow-up and the risk of random error in that group is very very high.
00:17:32
Speaker
So I think that the Hyperion trial should not be regarded as a positive trial in that sense it's more of a hypothesis generating signal at the most and that you should before you
00:17:54
Speaker
before you base your practice on a trial like that, you should demand a much, much larger trial with a more robust estimate.
00:18:08
Speaker
Absolutely, and one of the comments that has been brought up regarding Hyperion, as you said, when the mortality is very low or very high,
00:18:19
Speaker
it usually mandates a very large number of patients, which in critical care, and especially in this type of trials, obviously is quite challenging.
00:18:27
Speaker
But one of the things that I have heard often when referred to Hyperion is people have invoked the fragility index of one, which basically I understand and correct me if I'm wrong, that means that if one patient in either group went the other way, the result is a negative result.
00:18:47
Speaker
Yeah, I mean it's you end up on the other side of the 0.0, the magical 0.05 p-value.
00:18:59
Speaker
And it's actually fewer patients that need to move from one grip to the other than they were lost to follow up.
00:19:07
Speaker
And also they had
00:19:15
Speaker
they presume that all lost to follow-ups were dead and if you instead presume that they were alive the results will also be different and non-significant.
TTM2 Trial Goals and Methodology
00:19:27
Speaker
So it's balancing exactly on the middle of the scale and I think that you should if you look at the mortality it's
00:19:41
Speaker
exactly the same in both groups but then we have this very very small signal for those that were followed up until 90 days that were the time when they evaluated the neurological function and it's as we said just 30 something patients.
00:20:06
Speaker
So let's set the stage before we dive into TTM2
00:20:10
Speaker
So at this point, the guidelines at least in the practice really are suggesting that if you have an out of hospital cardiac arrest with a shockable rhythm, there's evidence to support the use of active intervention at 33 or 36.
00:20:28
Speaker
So at this point now, they're giving clinicians, let's say the option and there's emerging
00:20:36
Speaker
comments on that the results of the Hippurion trial might be added to that for non-shockable rhythms.
00:20:44
Speaker
And also people are translating that into in-hospital cardiac arrest.
00:20:49
Speaker
So now comes TTM2.
00:20:52
Speaker
So tell us about the hypothesis, how you were thinking about it.
00:20:57
Speaker
Like I mentioned, I think before we started, quite an amazing study in terms of the clinical and scientific rigor.
00:21:04
Speaker
and there's a lot of aspects that we can maybe comment on but just tell us from your perspective as a principal investigator what was the hypothesis and how things move forward yeah the we we really wanted to do ttm2 when we started ttm1 so that's so we had the plan for ttm2 for a long time we wanted to
00:21:28
Speaker
redo the HACA trial.
00:21:30
Speaker
We wanted to take the HACA trial, we considered that as a phase two trial and wanted to do a phase three trial, an adequately sized trial based on the same hypothesis that cooling 233 would be
00:21:48
Speaker
beneficial in terms of mortality and neurological function.
00:21:51
Speaker
So it's also important to realize that the intervention was 33 degrees and we compare that to what we considered
00:22:00
Speaker
standard care even if you could argue if that was standard care or not at the time when it started.
00:22:06
Speaker
But the intervention group is 33.
00:22:08
Speaker
We wanted to redo the HACA trial from 2002 and expand it to a larger group and take it to the standards of a randomized trial so that the
00:22:22
Speaker
recent years.
00:22:25
Speaker
So the hypothesis, this was rapid cooling to 33 improves mortality and neurological function for out-of-hospital cardiac arrests with a presumed cardiac or strictly unknown cause.
00:22:44
Speaker
And when we calculated the numbers
00:22:49
Speaker
we wanted to be able to detect a much smaller effect size that had been calculated in earlier trials.
00:22:59
Speaker
So we aimed at 7.5% absolute risk reduction which is 15% relative risk reduction and that gave us almost 2 000 patients to include.
00:23:12
Speaker
And just to
00:23:15
Speaker
the comment on that size, I think that might be still on the lower end of what should be acceptable.
00:23:21
Speaker
I think you have to have trials of that size or even larger to be able to be safe and sound and sure when you act on the evidence.
00:23:37
Speaker
We've been acting on small trials for so many years and I think that that leads us astray more than it leads us right.
00:23:46
Speaker
But that was the setting when we planned the TTM2 trial and the TTM1 trial had opened up the possibility and the equipoise situation that it was okay for hospitals and sites to
00:24:08
Speaker
to have a group that did not receive therapeutic hypothermia as we've known it, but only giving it a fever was detected.
00:24:23
Speaker
And I think an important aspect also that a lot of people have commented on is that when you look at the temperature curves of the HACA trial,
00:24:32
Speaker
the control group probably went even higher than just normothermia in many cases.
00:24:37
Speaker
So perhaps, I mean, that difference was really important in driving those results.
00:24:43
Speaker
But as you said, I mean, numbers are important.
00:24:45
Speaker
And I believe that a big problem that we've seen, not only in medicine, but has also been described in other scientific fields is our inability to replicate studies that have been adopted as dogma, right?
00:25:00
Speaker
Yeah.
00:25:01
Speaker
That is a problem in social science and cognitive studies, but also in clinical science.
00:25:07
Speaker
And I think that if in science, what we do know is that when a hypothesis is really, really strong, it can be replicated in different experiments.
00:25:17
Speaker
And it always proves that that's the right hypothesis, right?
00:25:20
Speaker
And that's where we're lacking a little bit in a lot of clinical and clinical arena.
00:25:27
Speaker
Yeah, I think you're absolutely right there.
00:25:29
Speaker
And I also think that's a big responsibility for guideline groups because
00:25:35
Speaker
if it's written in the guidelines, it's sometimes difficult to randomize against that guideline.
00:25:41
Speaker
And that was the reason that many countries could not participate in TGM-1, because the ethical committee said, no, this is in guidelines, we cannot allow you to do this.
00:25:53
Speaker
So I think guideline groups must acknowledge this and say,
00:25:59
Speaker
To the best of our knowledge, we recommend this, but we do it with the caveat that it might be that we don't have enough evidence and we suggest that you use it, but we also suggest that you could continue to investigate this issue.
00:26:20
Speaker
I think guidelines should be constructed that way, not to create these dogma situations as you bring up.
00:26:30
Speaker
Absolutely.
00:26:31
Speaker
And, Nicholas, so you talked about the hypothesis, you talked about how you did the power calculation.
00:26:36
Speaker
So tell us about the different groups, the 37 versus the
TTM2 Findings and Analysis Techniques
00:26:41
Speaker
33 group.
00:26:41
Speaker
I think it's important also, and you'll describe this, that the 37 group did require or did involve a lot of active intervention from clinicians.
00:26:49
Speaker
But tell us about what were the treatments that these patients received?
00:26:54
Speaker
Yeah, the 33 arm received that type of cooling that we've been
00:27:00
Speaker
giving to patients in the last 20 years based on the HACA and the Bernard trials or maybe the HACA trial, the Bernard started already in the pre-hospital scene but the HACA trial we wanted to get down to 33 as fast as possible
00:27:17
Speaker
and we wanted to get down to the lower temperatures faster than the HACA trial so we said that hospitals could use any type of equipment that they wanted to reach the target as soon as possible
00:27:35
Speaker
and then keep the target 33 stable throughout the intervention period using feedback control devices either intravascular or surface pads.
00:27:48
Speaker
And then we had a controlled rewarming of 0.3 degrees per hour up until we re-entered normal pthermia in that group.
00:27:58
Speaker
So the whole intervention period was
00:28:01
Speaker
40 hours and if you look at the graphs we came down to temperature faster than many trials and very much comparable to modern trials.
00:28:19
Speaker
And we also stretched the intervention period a little bit longer.
00:28:23
Speaker
So if you compare it with the HACA trial, we cooled a bit faster and we cooled a bit longer than that trial.
00:28:31
Speaker
And that was the 33 group.
00:28:33
Speaker
And from an academic perspective, it would have been ideal to have an arm with no TTM or no fever management or anything.
00:28:48
Speaker
at all that would have been the clean control group as they had in the HAKA trial but again it was not possible to do that and I can just go to my own hospital or hospitals around my hospital and
00:29:07
Speaker
due to the 20 years of temperature management and also the association between fever and poor outcome.
00:29:16
Speaker
So that's why we had to put in this
00:29:19
Speaker
threshold of fever and if fever started to come to the patient we treated that with TTM systems.
00:29:29
Speaker
And that's an important difference between the HACA trial and the TTM2 trial is that we in half of the patients in the normothermia group we gave a TTM therapy with intravascular or surface control devices.
00:29:47
Speaker
to clamp temperatures on normothermia, AMIA 37.5.
00:29:54
Speaker
So that is a difference from the earlier trials.
00:29:59
Speaker
And that gives us also a knowledge gap that we don't know if that therapy was beneficial, harmful, or just neutral.
00:30:10
Speaker
And what were the results, the final results of the study?
00:30:14
Speaker
Yeah, again, the results were very, very similar to TTM1,
00:30:23
Speaker
no difference and with very very neutral point estimates.
00:30:30
Speaker
So the point estimate in both TTM1 and TTM2 lean a little bit against hypothermia so it's a little bit better with the normothermia or the 36 arm but absolutely no significant signal.
00:30:44
Speaker
So very neutral in all aspects of the trial
00:30:48
Speaker
mortality, urological function, functional outcome, quality of life.
00:30:57
Speaker
And you did find that there were more arrhythmias, right, in the cooling, in the 33?
00:31:04
Speaker
Yeah.
00:31:06
Speaker
That's really the only significant finding of the TTM2 trial is that from the pre-specified adverse events,
00:31:16
Speaker
we could show that arrhythmia leading to a hemodynamic compromise was more common in the 33 group.
00:31:25
Speaker
But if you look at the hard endpoints mortality, it did not translate to a difference in mortality.
00:31:35
Speaker
So the sites could
00:31:38
Speaker
observe this arrhythmia but also cope with it and hopefully treat it in a way that didn't make it affect mortality in the end.
00:31:52
Speaker
Absolutely and I think Nicholas that an important also aspect of this discussion before we go into a little bit more of what it means at the practical end at the bedside if you could comment on some of the
00:32:10
Speaker
the methodological interventions that the trial had to really improve its rigor.
00:32:17
Speaker
One of the things that I read or heard you say in other discussions that I was quite impressed was the fact that obviously you can't, it's hard to blind what's happening to the patient because you know what you're doing, right?
00:32:32
Speaker
But that all the neurological assessment, obviously that could,
00:32:36
Speaker
traditionally is blinded, the people who do that, they don't know what intervention they got over time.
00:32:41
Speaker
But also what I understood correctly, if I'm wrong, is that you had the final data blinded and that there were two manuscripts that were written, one saying group A is 33 and group B is normothermia and another manuscript saying group A is normothermia and group B is
00:33:04
Speaker
just to really try to minimize any bias that could happen in the writing.
00:33:09
Speaker
Could you just comment on those things?
00:33:12
Speaker
Yeah, that's absolutely right.
00:33:18
Speaker
It's actually an idea from a Danish trialist called Peter Goetje who said that a lot of the bias that is introduced in clinical trials and how they are presented
00:33:33
Speaker
is introduced during the analysis phase and during the writing phase and I think that is absolutely on the spot because everyone has their own beliefs or thoughts or bias or it could be unconscious bias or a conscious bias and you want to twist and tweak it a bit and you can include one analysis but not another one and that will give
00:34:01
Speaker
Even if the results are the same, you can make the nuances in one way or the other.
00:34:09
Speaker
So I think it's critical to blind the analysis phase and also the writing phase.
00:34:16
Speaker
And I think that should be done in all trials, especially when it comes to clinical care.
00:34:20
Speaker
And it
00:34:25
Speaker
makes you extremely neutral during this phase.
00:34:32
Speaker
The statisticians didn't know which group was which and we also took out the temperature data and other data that could unmask the groups.
00:34:44
Speaker
We've separated that from the data management company, so they sent it in portions so you could not guess which group was which.
00:34:59
Speaker
And then we wrote draft manuscripts that we presented to the full author groups and these were as you said in duplicate with the groups interchanged.
00:35:11
Speaker
With this trial, with the neutral results, the wording was quite similar, but not identical.
00:35:19
Speaker
And it was the same in TTM1 that we had similar wording, but not identical in the two manuscripts.
00:35:28
Speaker
And you will always converge towards the more conservative interpretation of everything if you do with this process.
00:35:40
Speaker
Also for a neutral manuscript or a neutral trial, you must really be conservative
00:35:58
Speaker
and not lead yourself astray.
00:36:02
Speaker
And I think it's a very, very nice and elegant process to go through.
00:36:08
Speaker
And every author that's part of that, they really, really appreciate it.
00:36:14
Speaker
And I think that the end result is much less prone to bias than it would have been otherwise.
Standardizing Prognostication and Post-Trial Practices
00:36:24
Speaker
I agree.
00:36:25
Speaker
And it's important to recognize that we all have
00:36:28
Speaker
biases and that they are significantly stronger than we would like to admit.
00:36:33
Speaker
And from that perspective, I think it's a great exercise, like you said, in scientific rigor to really approach it this way because even in a neutral study, it has an impact.
00:36:45
Speaker
But I can only imagine what would have happened if there was a clear signal in one of the groups versus the other, right?
00:36:51
Speaker
How that knowing, I mean, can really drive that bias because we all have biases.
00:36:57
Speaker
and we just don't like to admit it sometimes.
00:36:59
Speaker
The other aspect that I find very interesting and quite important and worth commenting is one of the biggest challenges with these patients in real life, Nicholas, is what do we do afterwards?
00:37:12
Speaker
So it's not so much the first 48 hours at our heart.
00:37:15
Speaker
It's perhaps after hour 48, once they're rewarmed or whatever we did, and they're not waking up, and how do we prognosticate, and how do we talk with families?
00:37:25
Speaker
And you had a very, very
00:37:27
Speaker
strict protocol and standardized approach to really do neuro prognostication and kind of implement any, any further interventions or lack of interventions as you support these patients.
00:37:40
Speaker
Could you comment on that?
00:37:43
Speaker
Yeah, we have been working very closely with the neurology department and the neurophysiology department and radiology throughout the years, especially with Kronberg and Kisla Liljain.
00:37:59
Speaker
in Lund have been instrumental in really developing this neuro prognostication and also follow-up protocols for both TTM1 and TTM2.
00:38:11
Speaker
We were even more conservative in TTM1 but we also learned a lot that we could adopt into TTM2 and be sure that we
00:38:21
Speaker
had a protocol that was very strict and did define those with an ultimate poor prognosis with very very high confidence almost 100 percent and i think that's also very it's important for the
00:38:44
Speaker
for the families in all aspects of cardiac arrest, not only in trials, but in the clinical trial where we do not have blinded
00:38:55
Speaker
intervention, I think this is also something that must be done to minimize bias between groups.
00:39:03
Speaker
And that was not done in earlier trials in a systematic way.
00:39:11
Speaker
It was introduced in TTM1, but it's been part of most cardiac arrest trials afterwards.
00:39:19
Speaker
I think we've learned a lot from
00:39:22
Speaker
from these last year trials and that the neurological prognosis or prognostication guidelines have improved a lot.
00:39:32
Speaker
And I think that might have been one of the more important effects of this research and targeted temperature management because it has focused us on this critical
00:39:49
Speaker
issue in cardiac arrest patients.
00:39:51
Speaker
So that's a benefit from having started to look into TTM and therapeutic hypothermia.
00:40:00
Speaker
I think sometimes we take very simplistic views at trials as being a negative or a positive trial.
00:40:08
Speaker
Maybe you go a little bit further and we say, well, it's not necessarily negative, it's neutral.
00:40:12
Speaker
But like you said, there's a lot of byproducts of pushing science forward
00:40:17
Speaker
that ultimately really improve how we care for patients.
00:40:21
Speaker
And it's not one big discovery at a time, but it's small steps moving forward with a lot of effort from a lot of trials, right, that ultimately really mold the way we practice.
00:40:33
Speaker
But I think this is a great headway to really the question that a lot of our listeners have and those in clinical practice is, what should I do at the bedside?
00:40:43
Speaker
How do you interpret, Nicholas, as a clinician and as a scientist,
00:40:47
Speaker
the results, not only the results of TTM2, but it's really the aggregate of everything we've discussed so far and how it impacts what we should be doing at the bedside today.
00:41:00
Speaker
Yeah, I think one of the, we've seen an improvement in cardiologist care.
00:41:06
Speaker
I think
00:41:07
Speaker
most of that is probably due to better pre-hospital care all over, at least in some countries.
00:41:17
Speaker
But I think also that we have improved care in the intensive care departments of the critical care department.
00:41:23
Speaker
And that is because we are interested in the patient and we care about the patients.
00:41:29
Speaker
And it's a little bit sad to realize that
00:41:35
Speaker
you have to have this to give the patient the best possible care.
00:41:43
Speaker
It's the Hawthorne effect, it's massive.
00:41:47
Speaker
If you care, if you're there, if you stay bedside, you will improve the outcome of the patient.
00:41:56
Speaker
So I think that might be the
00:41:59
Speaker
biggest lesson that you should not give up on cardiac arrest patients.
00:42:01
Speaker
It's a patient group with a severe outcome all over, but the outcome for the individual patient might be very, very good.
00:42:11
Speaker
And that is worth recognizing.
00:42:18
Speaker
But as a clinician, I think you must be looking into different aspects of the cardiac arrest intensive care period.
00:42:29
Speaker
For the first 48 hours, I still think we do not know exactly how to best take care of that.
00:42:40
Speaker
period in terms of temperature management, in terms of sedation, in terms of mean arterial pressure, in terms of carbon dioxide and oxygen.
00:42:52
Speaker
Those trials are ongoing and should go on for years to come to really tweak that to the best.
00:43:00
Speaker
It might be that we should not do that much, just support the patient.
00:43:05
Speaker
We don't know.
00:43:06
Speaker
We need to investigate that period more.
00:43:09
Speaker
For the period after, I think that a conservative approach is key.
00:43:18
Speaker
And if we have this conservative approach and observe the patient's longer period,
00:43:27
Speaker
and also register this and study this, we will be able in a few years to pinpoint the patients with a very pessimistic prognosis, where we with almost 100% accuracy can say that this patient will not have a good outcome.
00:43:49
Speaker
We can limit futile care and burden to families.
00:43:56
Speaker
but we can also better identify those with an ultimate good prognosis that initially may look having a poor prognosis.
00:44:07
Speaker
And that might be the most important lesson from the recent trials and from the TTM1 trial and also from TTM2 when we have that data ready to
00:44:22
Speaker
with biomarkers, with radiology investigations, with neurophysiology and bedside clinical neurological examination, we will be able to define those that we should extend the observation period and extend care.
00:44:42
Speaker
to give them the chance to wake up after the arrest.
00:44:47
Speaker
I think that might be the most important lesson for the 48 hours plus of intensive care.
00:44:59
Speaker
Absolutely.
00:44:59
Speaker
And then as a parenthesis I think that the what happens after intensive care might be as important.
Adoption of Normothermia Protocols Post-TTM2
00:45:08
Speaker
I think it's a group that at least in Sweden and Scandinavia we've been poor in taking care of the patients that are sent home after cardiac arrest.
00:45:18
Speaker
I think the rehabilitation process the and the follow-up process must be much more structured and
00:45:27
Speaker
like in a stroke patient, for instance.
00:45:32
Speaker
No, I agree.
00:45:32
Speaker
And that's an area that in general in critical care we have just come to really realize, and with COVID we're seeing it as well, which is the post-intensive care syndrome or care of these patients needs to be better, needs to be studied, needs to be more systematic.
00:45:48
Speaker
But in particular for this population, which like you said, has very difficult outcomes for those survivors,
00:45:55
Speaker
having a more systematic and organized standardized approach and studying that is going to be critical.
00:46:01
Speaker
So you had mentioned, Nicholas, that after TTM1, the week after practice had changed at your hospital, how has this TTM2 results impacted practice now?
00:46:14
Speaker
What are you guys doing these days in the first 48 hours?
00:46:18
Speaker
Yeah we at the moment we use the normothermia arm of the TTM2 trial so we basically have not changed our practice since the trial was ongoing
00:46:38
Speaker
after the trial when it was finished we went back to the 36 protocol but after the release of the ttm2 results we adopted the normothermia arm of the trial protocol so full supportive care and sedation according to to the product
00:46:56
Speaker
and TTM management if we hit the target of fever 37.8 and then very strict neurological prognostication and recommendations for continued care or withdrawal of life-sustaining therapies in a conservative manner.
00:47:20
Speaker
So right now we are
00:47:23
Speaker
as if we were in the trial, but just using one arm and not randomizing.
00:47:28
Speaker
But of course, we are planning for a next trial.
00:47:33
Speaker
So this will be an interim period.
00:47:36
Speaker
Interim period, of course.
00:47:38
Speaker
And it seems that when you look at the aggregate, right, and it's important to now mention that when you add TTM1, TTM2,
00:47:45
Speaker
The number of patients that have been studied in these trials really outnumber significantly all the other trials done before, even some people who have argued about non-shockable rhythms.
00:47:57
Speaker
The number of non-shockable rhythms that you have treated in these studies outweighs, I mean, the heparion and
00:48:04
Speaker
and maybe we'll see a meta-analysis coming soon.
00:48:06
Speaker
But really, both of you.
00:48:09
Speaker
Actually, there is a meta-analysis where they've excluded the in-hospital portion of the Hyperion trial.
00:48:20
Speaker
So just if you look at the out-of-hospital portion of the Hyperion trial, it's very similar to the non-shockable group of TTM2.
00:48:28
Speaker
So for out-of-hospital cardiac arrest, if you aggregate the data from TTM1 to
00:48:34
Speaker
Hyperion, it's quite clear-cut that there is no inconsistency between these trials.
00:48:42
Speaker
They point the same direction.
00:48:47
Speaker
The signal from
00:48:48
Speaker
from the Hyperion trial was in the in-hospital cardiac arrest group where we have very limited data all over.
00:48:55
Speaker
There's just a few hundred patients randomized in hospital.
00:49:02
Speaker
So there we do not know at all.
00:49:04
Speaker
So based on what we know, I think it's reasonable to keep normothermia
00:49:12
Speaker
based on TTM1, TTM2 and Hyperion.
00:49:15
Speaker
For out-of-hospital cardiac arrest, for in-hospital cardiac arrest, we still await trials.
00:49:21
Speaker
Yeah, absolutely.
00:49:22
Speaker
What do you think is the next big question?
00:49:24
Speaker
Should we, I guess one of the things that you mentioned earlier is the active intervention on fever.
00:49:30
Speaker
Is that really something that helps, doesn't help or is neutral, right?
00:49:35
Speaker
Yeah, I think it's very interesting to see what people say about fever.
00:49:45
Speaker
We have these data that are observational.
00:49:49
Speaker
There's an association between fever and poor outcome, but if that's just an epiphenomena or if it's causative, no one knows.
00:49:59
Speaker
If we look at systematic reviews of fever management in general, there is no evidence of benefit of fever management in any type of condition really.
00:50:13
Speaker
And if you look specifically on cardiac arrest patients, we have no data at all to rely on.
00:50:22
Speaker
So I think we need to,
00:50:26
Speaker
investigate whether fever management is of benefit for the patients or not.
00:50:32
Speaker
And that should be done before we start a new period of dogma behavior in the critical care community.
00:50:41
Speaker
I think we should just realize and be humble and say we don't know we need to study this in a large cardiac arrest trial.
00:50:51
Speaker
Absolutely.
00:50:52
Speaker
Well, I think that we can keep talking about these unknowns for a long time, but want to be respectful of your time and really appreciate, Nicholas, all the work your group has done, really important, important questions, looking with very rigorous science.
00:51:10
Speaker
And we look forward to TTM3, and we'll talk more about that.
00:51:14
Speaker
But before we end the conversation today, we'd like to ask our guest,
00:51:19
Speaker
some questions unrelated to the clinical topic to close the podcast.
00:51:23
Speaker
Would that be okay?
00:51:24
Speaker
Yeah, absolutely.
Recommended Readings and Reflections on Scientific Values
00:51:26
Speaker
So the first question, Nicholas, relates to books.
00:51:29
Speaker
Are there any books that have influenced you the most or books that you have gifted most often to others?
00:51:38
Speaker
Gifted to Others, that would be a book by a British biologist, I think he is, Nick Lane.
00:51:51
Speaker
It's called Oxygen, the Molecule that Made the World.
00:51:55
Speaker
I think that's an incredible story about our origin
00:52:02
Speaker
It has a few years now.
00:52:05
Speaker
It's from the late 90s or maybe early 2000s.
00:52:10
Speaker
But the history is a million years.
00:52:14
Speaker
It's a wonderful book.
00:52:15
Speaker
If you should read one scientific book, I would definitely recommend Oxygen by Nick Lane.
00:52:24
Speaker
I have not read it, but definitely we'll look it up and we'll put it in the show notes and
00:52:28
Speaker
the end of the day everything we talk about in critical care is about oxygen right so that's the only thing that matters yeah it's a wonderful book but when it comes to fiction i read a lot of swedish danish finnish but but that may be not applicable to the us or the world community so if i think if i would
00:52:55
Speaker
recommend something in fiction I would recommend to dive into the classics.
00:53:01
Speaker
I've been doing that quite a lot and I realized that there were very very good books written in the early 1900s, the late 19th century
00:53:14
Speaker
So I just finished Anna Karenina by Leo Tolstoy.
00:53:20
Speaker
It's a magnificent book and it's very much, it tells a lot about moral and behavior and relationships that's applicable to everyday life today as well.
00:53:35
Speaker
So that's my tip.
00:53:37
Speaker
Yeah, good book.
00:53:38
Speaker
And I think that it's interesting because I am
00:53:41
Speaker
I was told from a young age, encouraged to read.
00:53:46
Speaker
And I remember that my grandfather would always comment that you should always read old books because only the great books get to get old, right?
00:53:55
Speaker
There's a reason why people are still reading them.
00:53:59
Speaker
So we'll definitely add those too.
00:54:02
Speaker
Excellent.
00:54:03
Speaker
Is there something you believe to be true, Nicholas, in medicine or in life that most other people don't believe or don't behave like they believe?
00:54:13
Speaker
Yeah, I think I've been thinking about that quite a lot.
00:54:18
Speaker
And I think that we should be much more humble towards science and what we know in medicine.
00:54:32
Speaker
I think that is the key for us to give better care is that we
00:54:42
Speaker
just tone down our argumentation and start to just encompass that we don't know and we have to find out and to do that we need to collaborate and so I think the whole system, the academic system, the whole merit system
00:55:10
Speaker
this bibliographic chase that we're all in must change for us to move the field forward.
00:55:19
Speaker
I think we're a little bit stuck.
00:55:21
Speaker
I don't know if that makes sense but that's my feeling that we're a little bit astray in how we perform science, how we perform medicine based on science.
00:55:36
Speaker
I
Advancing Medicine Through Quality Studies
00:55:37
Speaker
agree I think that really speaks to
00:55:39
Speaker
to the need that we have to always have answers as opposed to having better questions, right?
00:55:47
Speaker
And just being humble and recognizing the more you learn, the more you realize that you don't really know anything, right?
00:55:55
Speaker
And just recognizing that and being more humble about that.
00:55:59
Speaker
I agree and I think that it's played out quite dramatically in the last two years with the pandemic.
00:56:05
Speaker
but I think it just speaks to the state of science or medicine in general.
00:56:10
Speaker
So that's a great point.
00:56:11
Speaker
Absolutely.
00:56:13
Speaker
And the last question, Nicholas, is what would you want every one of our listeners to know could be a quote, a fact, or just a thought on what we discussed today?
00:56:24
Speaker
Oh, wow.
00:56:28
Speaker
What comes to my mind is that's, I think, yeah,
00:56:34
Speaker
size and quality matters.
00:56:40
Speaker
I think that is really something that I learned from John Ioannidis, California Stanford professor
00:56:51
Speaker
Many have read him, but I think what he said about we cannot advance medicine with observational papers and small randomized trials.
00:57:04
Speaker
We need to
00:57:06
Speaker
We need to take this to a much larger perspective and it's a little bit relates to what I said in relation to your last question but I think that is something that we must realize, all of us, that
00:57:31
Speaker
To be able to give the best care to the individual patient, we need to observe much larger populations and work on the group estimates, because I think that is what gives us
00:57:50
Speaker
the best possible therapy for the individual.
00:57:55
Speaker
It's a lot of talk about tailored therapy.
00:57:57
Speaker
And I think that is very, very dangerous because we will never know unless we have a genetic signal or something like that.
00:58:05
Speaker
But for the main
00:58:08
Speaker
diseases and symptoms or groups of symptoms, I think we need to rely on large groups and the robust estimates based on adequate sample sizes.
00:58:25
Speaker
I agree and I think that's a perfect place to stop.
00:58:29
Speaker
Niklas, thank you so much for all the hard work that your team has done for sharing that knowledge with us today in the podcast.
00:58:36
Speaker
We look forward to seeing you in person soon, hopefully in a conference down the road, not too distant, and also to having you back to discuss these and other topics.
00:58:46
Speaker
Thank you very much.
00:58:49
Speaker
Thank you so much for having me.
00:58:50
Speaker
It was a pleasure, really.
00:58:54
Speaker
Thank you for listening to Critical Matters, a sound podcast.
00:58:58
Speaker
Make sure to subscribe to Critical Matters on Apple or Google Podcasts and share with your network.
00:59:04
Speaker
Sound's transforming the way critical care is provided in hospitals across the country.
00:59:08
Speaker
To learn more, visit www.soundphysicians.com.