Live 1057

Free Internet Radio Stations

December 2015 Radiology Podcast (3 discussions)


[music] Hi. This is Herb Kressel
and welcome to the December 2015 Radiology podcast. We have a very full lineup for this
month’s podcast. First, I’ll be speaking with the authors of a controversies in radiology
pairing that we’re featuring on contrast-induced nephropathy. There’s been a lot of controversy
about whether or not the entity exists and what the risks are. I’ll be speaking with
Drs. Ulf Nyman and Jonas Björk from Malmö, Sweden and Lund, Sweden; as well as Dr. Robert
McDonald and Matthew Davenport. Dr. McDonald is from the Mayo Clinic and Dr. Matthew Davenport
is from the University of Michigan. I think you’ll find this discussion quite illuminating.
Next, my colleague Dave Kallmes, our Deputy Editor for Neuroradiology, will be speaking
with Jody Tanabe on a very provocative study on Sex Differences in Gray Matter Changes
and Brain Behavior Relationships in Substance Dependence. I think you’ll find this a very
stimulating discussion. And finally, Dr. Alex Bankier, our Deputy Editor for Thoracic Imaging,
will be speaking with Drs. Miranda Kirby and Grace Parraga on their manuscript, Do Imaging
Measurements of Emphysema and Airways Disease Explain Symptoms and Exercise Capacity in
Mild to Moderate Chronic Obstructive Pulmonary Disease? This is a rather long podcast for
the end of the year. So take it in small bits. Thank you very much and have a Happy New Year. [music] Herbert Y. Kressel, MD Hi. This is Herb Kressel
and welcome to the December Radiology podcast. This month we have a fascinating pair of articles
as part of our controversies in radiology series. And the controversy that we’re discussing
is the existence of and the risk of contrast induced nephropathy. And joining us for the
discussion are Drs. Ulf Nyman and Jonas Björk. Dr. Nyman is Associate Professor of Radiology
and Translational Medicine at Skåne University Hospital in Sweden. And Dr. Jonas Björk is
a Professor of Epidemiology at Lund University. Welcome Drs. Nyman and Björk. Jonas Björk, MD Thank you. Ulf Nyman, MD, PhD Thank you. H.Y.K. In the U.S. we have Dr. Matthew Davenport
who is Assistant Professor of Radiology and Urology at the University of Michigan, and
Dr. Robert McDonald who is joining us by phone only. Dr. McDonald is a fellow in the Department
of Radiology at the Mayo Clinic in Rochester, Minnesota. Welcome Drs. Davenport and McDonald. Matthew Davenport, MD Thank you.
Robert McDonald, MD Thank you. It’s a pleasure to be here. H.Y.K. Nice to have you with us. Just to sort
of set the discussion, our readers may recall several years ago a provocative article authored
by Dr. Jeff Newhouse, that questioned the existence of contrast induced nephropathy.
The core observation, as I remember it, that Dr. Newhouse made was that although we have
kind of an entire literature about contrast induced nephropathy, most of the studies that
drove the concept of this entity were not controlled retrospective studies, and in the
absence of a control in in-patients who are sick and tend to have alterations in renal
function while they are in the hospital, he questioned whether or not contrast induced
nephropathy exists. Then a couple of years ago, the group at Mayo and at the University
of Michigan, Drs. Davenport and McDonald and your colleagues, reported on very large, retrospective,
propensity matched cohorts of in-patients undergoing CT with and without contrast; and
both of these studies, which certainly got a lot of attention in the radiology world,
showed that the risk of contrast induced nephropathy using the newer low osmolar and isoosmolar
agents was lower than previously thought and perhaps in the case of the Mayo study non-existent.
So that’s the world as we knew it, and then Drs. Nyman, Aspelin and colleagues wrote a
very thought provoking essay which we are publishing this month and they sort of questioned
whether the conclusions based on the two studies may be overstated, and in fact that contrast
induced nephropathy is a real item or certainly we haven’t excluded it. It is something
that we ought to be concerned about. And they identified a number of key areas and we’ll
be discussing them individually. The first area of concern was the use of relative versus
absolute glomerular filtration rate as a way of stratifying the patient population and
also in general the use of serum creatinine as the measure of damage in these patients.
Dr. Nyman, do you want to comment on this? U.N. Yes. Briefly, the relevant GFR is generally
used in cardiology and radiology papers when estimating GFR when you’re doing contrast
examination. Fundamentally this is wrong because when you are evaluating toxicity of drugs
that are excreted through the kidneys, then you should use the individual’s GFR. These
are called absolute GFR and not the one that is adjusted to the certain or the surface
area. The counterpoint that Davenport/McDonald did, we are not sure if they really, if they
misunderstood a bit of what we were writing. What we want to make clear is that we didn’t
mean to use any measured absolute GFR, we meant to use estimated GFR and when you’re
using (inaudible) or the (inaudible) equation to estimate GFR, then you have to recalculate
the absolute values using body weight and height and the certain body surface equation.
What we tried to explain in the article is that if we have a certain relative GFR entered
for example between 30 and 45, if you take those patients and calculate their absolute
GFR, I mean up to one third of the men will end up in the higher GFR interval and about
10 percent of the women will end up in a lower GFR interval. So by using relative GFR, you
saw the mix. It’s more women with poor renal function with large men with better renal
function which may dilute the results of said. H.Y.K. So you’re basically saying that the
populations may have been somewhat skewed with individuals that were actually functioning
at a higher level than the way the estimate was done? Dr. McDonald you’re sort of from
a place with a lot of big, beefy guys, do you want to respond to that concern? R.M. I’m from a place with what sort of
guys? H.Y.K. Beefy guys. R.M. Here at Mayo we still, although we’ll
probably eventually adopt and EGFR model and we certainly calculate an estimated EGFR based
upon the MDRD equation and obviously we can use whatever other equations we want, we still
sort of derive our estimates from certain creatinine measurements in terms of that’s
how here how we’re sort of risk stratifying our patients for acute kidney injury following
CT contrast administration. Our lab uses the NIST standard, the traceable isotope diluted
mass speck. So I’d argue our, at least our serum creatinine results, even though we can
certainly talk about how serum creatinine isn’t the best biomarker in the world, at
least we’re using the most rigorous example where the inter and intra laboratory variability
compared to another lab that uses IDMS that’s going to be very small. Going back to this
idea of absolute, I think what could happen there certainly, if we looked at our data
and compared to estimated to absolute, what might happen even though our results didn’t
suggest there was any, I think it will just blur the lines where that cut off between
when you look at Dr. Davenport’s paper where that cut off would be in terms of where there’s
an increased risk of acute kidney injury. H.Y.K. So Dr. Davenport, do you think the
difference in your studies might be due to the fact that there are thinner people in
Michigan? M.D. Having lived here for ten years, I would
say that’s not possible. But I think they make a good point. I mean their point is that
there are better ways to accurately evaluate someone’s GFR. I actually have no problem
with that. They’re probably right. You use individualized based GFR which was actually
modeled based on the patient’s individual weight and individual height; you probably
would have a more accurate measure of renal function. They are probably right about that.
The reason why we chose to use GFR on our paper was because that’s how risk stratification
for this is basically always done. So to be more clinically relevant, that’s why we
chose that. H.Y.K. Thank you. I think that’s very helpful
and Dr. Nyman I was also confused about what you meant about absolute. I thought you had
to have people in a clinical research center to calculate the absolute GFR so that’s
very, very helpful. Now the next thing was this issue of the use of the non-contrast
enhanced CT as the comparison group and how there’s maybe a misallocation of confounders
as a result of that. Dr. Björk is that something you want to talk about? J.B. Yes I can do that. I think it’s a bit
problematic from a methodological point of view, but although you’re using the propensity
score of stratification which I think is a very good technique. Still you might ask the
question whether you can sufficiently account for the selection of differences between the
controlled group and the treatment group. So I would suggest that you could do propensities
for matching within the CM group, looking at different doses and match against each
other; a different approach. H.Y.K. I think that’s a very interesting
idea. My own thought about that is it obviously makes a lot of sense but if we go back to
where we started from where we had this whole literature on this entity that was based on
uncontrolled studies, only of patients who got ill, the question that I would have is
sort of where does this kind of high level of concern derive from? The core observation
where at least is flawed as what we’re dealing with now with much more refinement. I don’t
know if you’d like to sort of respond to that. J.B. From my perspective, it’s not either
or, you can do both because I mean there is still this risk for a severe selection bias
in the results, and one way of looking at that from a different angle was this suggestion
I would say. H.Y.K. I think it’s kind of a, in my own
mind, sort of this discussion is very, very helpful because it’s kind of more the where
do we go from here I think in doing the best for our patients. Dr. Davenport do you have
any thoughts about sort of the comparison group and the problems of using the non-contrast
enhanced CT? M.D. Yeah I mean I actually share his concerns.
When we did this analysis I thought to myself repeatedly throughout the process, is this
good enough. Of course at the time, this is now 2013; it feels like it’s a long time
ago. H.K.K. You’re a very young man if you think
that’s a long time ago. M.D. At the time, none of these studies were
controlling for anything, and if they were using controls, they were simple retrospective
controls with no adjustment and bias. And so to appropriately select a group that would
be a suitable control, this seemed to us to be the best group at the time and then adjust
for all the possible reasons renally that might predict someone would or would not get
contrast. But when I read your guys’ paper suggesting to dose within a group that was
a great idea I thought. I think that’s a great way to take this to the next level.
Of course, at our hospital everyone was getting the same dose. So we would be unable to do
that analysis. Maybe somebody else would have the data available to do it, but at the time
we were not basing our dose based on that. Of course it would probably need to be in
a prospective fashion because those people who are varying their dose often times are
doing it on the basis of somebody’s weight and height which brings in the other issues
we discussed a moment ago. H.Y.K. Dr. McDonald any further thoughts on
this issue? R.M. Sure. I agree it is a great point on
multiple levels. The first is you know is it fair to compare someone in the contrast
exposed group who had a PE to someone in the naïve group who didn’t get contrast because
they had some other far less severe condition but they have other co-morbidities that the
propensity score matches those patients up. Fortunately when we actually drilled down
to the data, it doesn’t happen that often, but it does raise a good point. There is still
room for some bias in that. To sort of touch on the point in terms of varying contrast
dose, our practice here does use a weight based nomogram so our dose does vary a bit
with weight. Obviously we try to basically shoot for relatively similar concentrations
based on body weight in adults. We actually, although we didn’t in our 2013 papers because
I think I share Dr. Davenport’s sort of viewpoint, at that point we looked and we
said well nobody has done anything yet so this is the first start and also I think for
us it was a lot of work just to get there, but subsequently we have published something
that didn’t make it into the journal Radiology but it’s in Mayo Clinic proceedings where
we did actually control for contrast dose in our model. And so I think you know it certainly
is a concern, but I think what really our, hopefully I’d like to think the initial
2013 papers did, is it put sort of some bookends on the problem and from one end our findings
said look we can’t identify contrast induced nephropathy at all to Davenport’s and his
colleagues finding that it might exist, but it’s really up to the patients who have
compromised renal function. I think as studies continue to come out where we’ll be able
to refine the window in which if this entity exists, we’ll have a narrow window to gauge
what patients are at risk. H.Y.K. Dr. McDonald what was the result? You’re
leaving us hanging here when you… R.M. We didn’t find that that was a – it
didn’t seem to be an independent risk fact for, I apologize, for contrast nephropathy,
but again this is a narrow window of doses. It’s not like anyone is getting triple or
quadruple dose of contrast. All we’re basically doing is keeping the intravascular concentration
roughly similar between someone who weighs 50 kg and someone who weights 150 kg for example. H.Y.K. Good. Now the next item of concern
that was raised, and I think this relates to the McDonald study, was the limited attention
to the results stratified on non renal risk factors. J.B. Yes I could probably comment on that
because I think one strength of the propensity score technique is that you can look at the
subgroups and for example you can look at the group that has received contrast media
despite the fact that the propensity score is saying that you in a way shouldn’t because
of the risk factors that you have at the individual level. If you look at that in the original
publication by McDonald and co-workers, you can actually see that with the medium of this
group, there is an elevation in this sub-group. It’s only a sub-group and it’s only observation,
but I think it’s still important to look at these data more carefully. H.Y.K. Right. Dr. McDonald do you want to
comment on that? R.M. Can we just back up? What were you seeing
in our medium risk sub-group again? I’m sorry I missed that part. J.B. That was in your supplementary material.
For the medium risk sub-group you had stratum one, so that is the group with the lowest
propensity score right? R.M. Yeah. J.B. You had a clear elevation there in the
same risk? R.M. Yeah, I see what you’re saying. Right
so one of the propensity score matching techniques we used was a stratification method which
we showed the strata within each, so it’s sort of like stratification of the strata,
so we stratified our patients in those risk groups based upon our sort of clinical experience
of how we sort of mentally triaged patients. The low risk group were patients with creatinine
less than 1.5 who we really didn’t think were our great risk. Although we didn’t
know, but those were the people we were least concerned about. The medium risk were between
1.5 and 2 in terms of a baseline serum creatinine and the high risk was anything about 2 and
so those were the patients we were the most concerned with. And frankly for that 2013
paper, we were very interested in that medium risk group because that comprises a large
percentage of our patients. The high risk group patients, even though they seem very
ominous, they are a very low fraction of patients you actually scan every day. And so I remember
exactly what you’re talking about, that in the stratification, it doesn’t clearly
march as you normally would expect it. I remember seeing that and being a little bewildered
by it as well. As to why that is, I guess for that one thing I guess I was more interested
in being intellectually honest and putting our results than trying to redo it, but when
we subsequently re-analyzed those data again using like a boot strapping model and this
ended up being in the supplement to the paper that looked at dialysis and mortality. So
we did a boot strapping model where we compared multiple different propensity score matching
method and we had the computer rerun the propensity score matching 100 times for each of I believe
the five different methods, we had very similar results every time and we were able to sort
of narrow of confidence interval. I think the results overall still argue that, at least
from a statistical point of view, we can’t discern an affect that you can call contrast
induced nephropathy based upon our data, but no I acknowledge exactly what you’re talking
about that the stratification was not this clear ramp up from the lowest to the highest
risk based upon their propensity score. H.Y.K. So I know that a lot of this discussion
will seem very heavily in the weeds to a lot of people who are listening to this, but I
must say I think it’s important because in the end we have these very complex, multi-factorial
situations and we’re trying to make decisions about giving an agent to people that will
affect their lives potentially. It’s not trivial and I think it highlights the importance
of having good quality data and a good quality analysis and that I’m pleased to see that
both Drs. Davenport and McDonald the questions aren’t over, that we’re trying to build
from there. I think that was really the point from Drs. Nyman and Aspelin and Björk why
you wrote that. That kind of we don’t have the final answer, there are issues that remain
to be explored. On the other hand, in the practical world of clinical radiology, people
need to make decisions and practice. I think one of the points of the Davenport and McDonald
papers was that perhaps you’re withholding contrast from people who wouldn’t have adverse
events on the basis of it and who would benefit from the added information available. So that
brings me to sort of knowing what we know now, where are we? Dr. Davenport you’ve
been involved I think with ACR Committee looking at contrast guidelines and perhaps you can
tell us where we are, what the guidelines are, and what the changes have been. M.D. Sure I’d be happy to. Recently the
chapter on contrast induced acute kidney injury was rewritten. I think there are two major
changes. One is definitional and one is based on the threshold. The definition difference
now is that what was previously termed CIN, which is you get contrast and then 48 to 72
hours later you a have a bump in your serum creatinine and the old definition CIN is no
longer. That’s now called post contrast acute kidney injury. In other words we’re
taking away the causative statement and placing it within the name and that’s now post contrast
kidney injury. And CIN is now restricted to only that which can be confirmed to be directly
causative from the contrast material which of course is extremely difficult to disentangle
on a per patient basis and requires a randomized controlled trial or some kind of advanced
retrospective statistics. The second change is based on the threshold. This caused a lot
of controversy, frankly speaking, in that room as to what number should we put if anything?
There was some people that said we shouldn’t even put a threshold down because we don’t
know yet and you might imagine who that would have been who would have made those comments.
And then there were other people who said no, we need to guide people so they know what
to do and after a lot of discussion we decided that the level which has the most level of
evidence in the literature based on controlled studies was a number of less than 30. And
we make a bunch of statements in there which is the following, one this pertains to IV
media only and secondly is that anything involves a risk benefit decision. It’s more complicated
than just a simple number. I can go into a long winded discussion about why we chose
less than 30, but that’s what we chose. H.Y.K. Good. Dr. Nyman how do things stand
in Europe? Has there been any changes? U.N. No there has not been any change in standard.
We had a discussion in Sweden at the radiology meeting and there were people from the European
Society of Radiology taking part too, and the general recommendation was to sit on your
hands and wait, don’t do any changes. Basically the guidelines we have in several countries
in the European society, one risk group is those with GFR below 45. Then you have the
other patient group with multiple risk factors and basically independent of what the GFR
is, if you have a patient with cardiac decomposition, diabetes and so on, unstable renal function,
unstable hemodynamics and so on, be careful then. These are the general guidelines. But
then it came into practice from my personal point of view, I seldom see any problems because
you can solve a problem in a lot of different ways and I what I use very frequently in the
risk categories is doing ADCT and wrapping up the mAs, the patients are pretty old generally.
You don’t have a concern about radiation and you can – basically half the contrast
we can dose. H.Y.K. So just reduce the dose if you’re
concerned? U.N. Sorry? H.Y.K. The idea of using a lower dose and
upping the mAs if you’re sort of in that borderline zone? U.N. Exactly so you don’t get too much noise
in the images but then you can almost half the contrast medium dose. H.Y.K. Okay, going back to the US, Dr. Davenport
what’s happened to practice at the University of Michigan? Are you using the current ACR
guidelines or are you doing something a little different? M.D. So we started using something very similar
to the ACR guidelines prior to ACR releasing those guidelines. So what we use is, if a
patient has acute kidney injury or chronic kidney disease with EGFR of less than 30,
it triggers a provider to provider conversation. That’s our guideline. So we have a discussion
about what to do. When the EGFR is 30 or higher than we don’t have any specific requirements. H.Y.K. Experientially, have you had these
conversations? What’s the usual outcome? M.D. Because our policy tends to be more liberal
than what the referring services believe about the nephrotoxic potential of contrast, we
have these conversations infrequently because there is not many people who are asking for
this. And if they’re asking for it, they know there’s a risk probably too and it’s
on everybody’s radar. H.Y.K. Good. I’ve actually kind of consulted
with physicians and sort of had this discussion and we have kind of just on limited numbers,
but we have been a little bit more liberal about using it if it’s really an indicated
situation. Dr. McDonald, where are we at Mayo with this issue? R.M. Right so you know our internal use we
still, like I said, even though we calculate the EGFR we still are using serum creatinine
right now, but it’s very similar, we have a similar policy to the ACR and to Michigan
of that below a creatinine of 2 we feel it’s safe to administer contrast based upon the
results from our study and Michigan study. We thought that was the most conservative
approach and again it aligns nicely with ACR’s guidelines because a serum creatinine of 2
is relatively close to 30. I anticipate at some point we’ll transition to EGFR based
measurements and adopt a 30 threshold; and in terms of phone calls, well certainly we
still get them because what our thoughts are in the department versus what providers think,
clearly we certainly get phone calls where people are worried about giving contrast to
someone with serum creatinine of 1.4. We certainly try to counsel them, but our policy here is
to sort of ultimately the decision is with the ordering provider, so we’ll try to let
them know what the best evidence is and let them make the best clinical decision from
there. H.Y.K. Well thank you. I want to thank all
the participants today. I think this has been a very enlightening discussion. I think people
perhaps will have a better sense of the complexity and thought process that goes into kind of
thinking about these guidelines. Also from my perspective, the importance of good quality
studies, we kind of throw out what we need is a prospective, randomized controlled trial,
but with something like this where the incidence is relatively low, it’s very, very challenging
to put these together and quite frankly in the environment that we’re in, I don’t
know that we’re going to be able to see this and so we’re stuck in a situation where
we kind of lurch forward using the best information and processing what we have, and we just have
to be aware of the limitations of that. So I want to thank you all for participating.
Thanks to our colleagues in Sweden for your contribution in the essay and also in the
podcast and thanks to doctors Davenport and McDonald once again. So thank you. [music] David F. Kallmes, MD Hello and welcome to
this video podcast. My name is David Kallmes, I am Deputy Editor for Neuroradiology at the
journal Radiology. I’m joined today by Jody Tanabe who is Professor of Radiology, Chief
of Neuroradiology, and Vice Chair of Research at the University of Colorado. Welcome Dr.
Tanabe. Jody Tanabe, MD Thank you for having me. D.F.K. Sure we’re here to discuss your really
exciting paper entitled “Sex Differences in Gray Matter Changes and Brain Behavior
Relationships in Substance Dependence.” First of all can you just share with us what
are the known sex differences in the natural history of substance dependence? J.T. So there have been several research studies
showing several differences at every level of drug use. So for example women have been
shown to escalate their use of drugs much more quickly than men. While it’s true that
men compared to women tend to engage in more risky behavior such as experimenting with
drugs, it turns out that women are much more likely than men for example to use drugs to
reduce depression or alleviate stress and that is one of the thoughts behind the quicker
downward spiral for women than men. There are also some behavioral differences for example
with psycho-stimulants, women report different levels of euphoria and that is thought to
be related to the menstrual phase. The good news is that women also tend to seek treatment
at an earlier time point than men. A lot of the sex differences have also been substantiated
by animal studies. There have been several animal studies that show that female rodents
acquire self administration paradigms much more quickly than male rodents. D.F.K. Okay so with that background of the
differences between men and women with substance dependence, what did you do in this study? J.T. In this study, it was a cross-sectional
study where we recruited about 127 individuals roughly divided between healthy controls and
abstinent substance dependent individuals and we compared the gray matter volume at
a whole brain level and at a regional level between patients and controls and we determined
and investigated the effects of sex on those differences. There have been a number of studies
that have shown that gray matter volumes differ between drug users and controls, and we also
know there is sexual dimorphism just being female or male. But there have been very few
studies that have looked at the interaction of these two factors. So we conducted that
analysis on T1 weighted structural images and then we determined whether there were
any relationships between the gray matter volume and behavioral measures that may be
important in drug dependence as well as measures of drug severity itself. D.F.K. What did you find? J.T. So our main finding was that there was
an interaction between sex and drug dependence diagnosis on gray matter volumes involving
multiple areas. Primarily frontal, there’s a lot of pre-frontal ventrimea pre-frontal
and the lymphics system as well as the temporal lobe. It turned out that these differences
were really driven by the women. Drug dependent women had much lower gray matter volume than
control women, and we did not find these substantial differences in the men. Those were our volume
metric findings and then we took it further and found that there were correlations between
certain behavioral metrics. So, the one that is most interesting is that there was a negative
correlation between the gray matter volume in the nuclei accumbens and drug severity.
We measure drug severity by summing the dependence and abuse symptoms across all drugs for each
individual. D.F.K. Was that an expected finding or was
that surprising? J.T. So the finding that the differences would
be driven more by the women than the men, that was a surprising finding. We were not
expecting to find such vast differences. D.F.K. What are the potential clinical ramifications
of these findings? J.T. I think one of the interesting clinical
ramifications is our finding of correlations between the nuclei accumbens gray matter volume
and drug severity. All drugs of abuse, for lack of a better term, hijack the main reward
system. And that system consists of cell bodies of dopaminergic neurons that line the ventral
tegmental areas with axons projecting forward into the nucleus accumbens. So the fact that
we found this relationship between drug severity and accumbens gray matter volume suggests
that there is a neuron anatomical link between true behavior. D.F.K. And does that point to any specific
therapy or is it more just understanding the pathophysiology better? J.T. Well it certainly helps us understand
the pathophysiology better. It does confirm the current notions about the role of the
reward system. Eventually we hope that this will improve therapies and management. I think
one way is that because we know that there are differences in the natural history of
drug addiction in women compared to men and if we have some biological evidence of such
a difference, it can spur more research into reasons for those differences. For example,
there may be hormonal differences that may contribute to vulnerabilities, and those hormonal
differences have varying effects upon different regions of the brain. That could direct therapy.
It could also help direct management. D.F.K. Okay and is there anything from this
study that the practicing neuroradiologist should take away or look for in everyday clinical
work? J.T. That’s a good question. I think the
most important message for the practicing neuroradiologist is that we read MRIs all
the time. They’re the bread and butter, but I think that this study should make radiologists
realize that there may be a wealth of data in those MRI images which we are not necessarily
tapping into. I think that it demonstrates that if you can come up with the right question
then radiologists can make substantial contributions to understanding the biology of disease not
just making a diagnosis. D.F.K. So speaking of looking at brain volumes,
you mentioned at the beginning of the podcast this was a cross-sectional study. Does that
provide adequate data rather than a longitudinal study to understand the changes with abstinence
and difference between the sexes? J.T. Well ideally you would want to do a longitudinal
study and look at how brain volume changes over the dynamic range of short term versus
long term abstinence. We did not do that, but that is of course the ideal study. What
we can conclude from this study is that these changes are sustained without the effect of
acute drugs. On average our patients were off of drugs for about a year. That’s important
because, for example, we know that alcohol in the short term can have significant changes
in brain volume and metabolism with acute sobriety over the course of weeks. So I think
what this study demonstrates is that we can detect sustained changes. Whether it came
before or after the drug, I cannot say. D.F.K. Sure. So I note that you have a current
NIH grant to study this, can you share with us what you’re doing in that grant? Are
you looking at other features of the disorder with other imaging techniques? J.T. So we’re currently funded to look at
decision making in drug users. Specifically, we look at risky decision making and we’re
looking at the striatal system as well as the frontal networks in looking for neuro
correlates that may underlie bad decisions made by drug users. Interestingly, our RO1
was not originally funded to look at sex effects, but this information came out afterwards. D.F.K. Okay. Certainly drug addiction is a
public health problem and congratulations on this fantastic line of research. Is there
anything else you want to share with the listeners about your project or future projects? J.T. I’m appreciative of the chance to share
this information and I hope it inspires people in the radiology community to do similar sorts
of work. D.F.K. Congratulations on your paper. We greatly
appreciate your support of our journal and we look forward to future great papers from
your group. Thank you. J.T. Thank you for having me. [music] Hello. My name is
Alex Bankier and I’m Deputy Editor of the journal Radiology in charge of Thoracic Imaging.
Today with us is Dr. Miranda Kirby from the James Hogg Research Center at the University
of British Columbia, and on the phone is Dr. Grace Parraga from the Imaging Research Laboratories
at the Roberts Research Institute. Our podcast today is to discuss the recent article from
this group. The title of this article is the question, “Do Imaging Measurements of Emphysema
and Airways Disease Explain Symptoms and Exercise Capacity in Mild to Moderate Chronic Obstructive
Pulmonary Disease?” Dr. Kirby, what is the answer to this question? Miranda Kirby, MD The short answer to that
question is yes. We know that exercise, capacity limitation, and symptoms are well described
in all COPD subjects including mild COPD subjects, but what’s not as well known is what are
the underlying determinants of the limitation and symptoms in these patients? So what we
investigated, a relatively large group of these patients using hyper-polarized helium
magnetic resonance imaging, and we found that it was emphysema that was actually the dominant
contributor to both exercise limitation and symptoms. A.A.B. I see. Why does helium MRI work so
well in early COPD while pulmonary function tests which are considered the reference standard
if you want for this disease do in fact not? M.K. Well I think that the hyper-polarized
helium MRI measurements are actually giving us functional information on a regional basis.
So we’re able to visualize all of the places in the lung essentially where air are able
to access, so if there are parts of the lung that there’s obstruction of the smaller
airways or medium sized airways, we don’t see the gas going there. As well when we have
a lot of emphysematous tissue destruction, we’re able, and as the gas is able to access
those areas; we’re also able to measure those enlarged airspaces. We think that these
more global pulmonary function measurements like FEB1 are maybe not as sensitive at measuring
the contributions of both the smaller airways and more mild emphysema. A.A.B. I see. Dr. Parraga, why is it so important
to diagnose COPD early in the course of the disease? Grace Parraga, MD Well I think there are two
ways to look at this from our perspective, and I think mainly what we wanted to explain
was why in milder disease are there such differences and why is there so much heterogeneity amongst
patients. Patients tend not to be diagnosed until they have symptoms and they’re diagnosed
using the standard clinical tools not imaging. So they’re diagnosed using spirometry. Even
yet, in the mildest forms, we see a large heterogeneity in symptoms and exercise limitation.
I think early on I think it would be important for patients to get a full workup so that
the underlying contributions to their disease are well understood at the earliest time point
so that intervention can be thought through based on that information. I think the second
point is that folks who are smokers or ex-smokers need to understand that these kinds of tools
are out there and that the sooner they understand what’s going on in their lungs the better.
And I point to other kinds of tests for heart disease, and Alzheimer’s where the paradigm
for early understanding helps with treatment decisions. A.A.B. I see. Dr. Kirby, there is a lot of
talk about CT definable phenotypes of COPD. Is helium MRI looking at the same phenotypes
and what is the added value that helium MR can bring to the feel of imaging in diagnosing
COPD? M.K. That’s a really great question. With
CT, we’re able to quantify emphysema in all regions on the lung, where with hyper-polarized
helium gas we’re only accessing those areas of the lung where gas had access. We’re
also able to directly quantify the dimensions of the airways. Where with the hyper-polarized
gas MRI we’re getting more of a functional measurement of which airways may be obstructed
or narrowed with the gas can access. So we are really getting very complimentary information.
I think that our studies and our previous studies have shown that particularly with
our hyper-polarized diffusion weighted imaging we may be able to identify and measure emphysema
more at the earlier stages. We’ve shown that there’s an elevated ADC values as well
as normal diffusion capacity at level of carbon monoxides of DLCO values in subjects that
have a normal spirometry and normal CT measurements. So it may be more suited for looking at disease
in early emphysema in these more mild subjects. A.A.B. I see. The question for Dr. Parraga,
given the limitations to the availability of hyper-polarized helium and helium in general,
what are the practice of clinical patients of using the method you described in COPD
patients? G.P. Well there are practical limitations
for certain and this study should be taken in that context. There is likely very little
translational potential for helium MRI, but I think what we’re doing now is opening
up a conversation about using imaging in general in COPD patients; and I also point to the
fact that while we have been using helium MRI, and this is a very large longitudinal
study that started back in 2009, we’ve transitioned to helium xenon MRI, excuse me, and that’s
the 129 xenon isotope and that technology is now on the verge of FDA approval and more
widespread clinical use. So I think our group and most groups worldwide now have transitioned
to xenon and there’s some interesting properties of the xenon gas itself that actually gives
us great optimism because there is more sensitivity, xenon is more sensitive to some of these airway
and parenchymal abnormalities in COPD and other obstructive lung diseases. A.A.B. I see. So just to make it clear for
our listeners, you believe that most, or a substantial part at least, of the results
that were previously obtained with helium can be extrapolated into the xenon framework? G.P. Absolutely. I think we’ve shown that
and other groups have shown that and in particular in COPD and in asthma and in fact we’re
very optimistic about xenon because of its sensitivity to airway obstruction. The gas
is thicker and more dense and it is a better approximate or estimate of the constituents
of air, the air that we breathe. So we’re very optimistic about the potential for xenon
to be used clinically especially in cases where you can’t explain why the patient
can’t exercise, why they’re feeling so bad, and FEB1 and other sort of clinical measurements
don’t explain what’s going on with the patient. A.A.B. Interesting. A question to both of
you, in the COPD imaging with MR is always in competition with CT, with pulmonary function
test, so what methods either low cost methods or very widely available methods while MR
is none of the two. Where do you extrapolate a little bit into the future? Where do you
see the future role of MR in the work up of COPD and let’s say in five or ten years
from now, where do you see the role of MR in the clinical management of patients with
COPD. M.K. I guess I look at it as not as MR versus
CT versus FEB1, I look at it as imaging versus these other cheaper more global measurements
that aren’t as sensitive. So I think that first I really believe in more multi modality
imaging approach because they do provide very complementary information and I think long
term I see imaging being used for more clinical phenotyping and if these COPD patients can
be phenotyped early on, then treatment can be targeted towards those specific phenotypes
and that can in the long term improve outcomes. A.A.B. I see. Dr. Parraga do you want to answer
that? G.P. Yeah. I agree with Dr. Kirby. I also
want to point out that I acknowledge that we’re living in a constrained health care
economic system and that in some cities and some sectors, access to MR and CT may be limited,
but again there is an economic argument for phenotyping patients and having a better look
at what’s going on with patients, because COPD costs a lot of money to the healthcare
systems independent of how you’re paying for it and what we’ve tried to do is show
the utility of imaging to save costs. For example, to save costs if you understand what’s
going on better, you might obviate the necessity of hospitalization for exacerbation. You might
be able to treat earlier and better so that the patient is more functional and feels better,
and then they can more readily differentiate when their symptoms are just day-to-day problems
or they’re in fact experiencing an exacerbation and need immediate care. We’ve looked at
the health economic framework for our health care system which is a little bit different
than the U.S. and I think that on the whole if you can make a commitment to scanning specific
patient populations as quickly as possible, an MR and CT for a patient can be completed
in about ten minutes, then there’s a good argument for using it on the economic and
on the patient and the patient treatment basis. A.A.B. If I understand your message right,
you’re making a point for potentially investing more in the diagnosis in early stages of the
disease in order to make economy to spend less in later stages of the disease, correct? G.P. That’s correct and perhaps there’s
an argument for even every two years doing follow-up on the patients to watch the pathologies
as they progress. The cost of a scan every two years far outweighed by the cost of all
the other types of interventions that happen in these patients. On that basis alone I think
it’s justified that we continue to develop these tools. A.A.B. I see. I think this is a very interesting
perspective and a very interesting note and look out into the future to end our conversation.
Dr. Kirby, Dr. Parraga, thank you very much for being with us. [music]

Leave a Reply

Your email address will not be published. Required fields are marked *