Non-alcoholic fatty liver disease and Alcoholic liver disease

By Brian Lemay 6 comments


– So I’m gonna speak to you
today a little bit about a few abstracts coming
from the liver meeting about non-alcoholic fatty liver and alcoholic liver diseases. Dr. Lock said alcoholic liver diseases, my area of research interest
and what I work in mostly, but we tend to group
these two things together. Obviously the causative
factor of fat in the liver in these two conditions
is quite different, but pathophysiologically
they look the same under the microscope. And importantly for all of us to realize is that these are behaviorally
mediated liver diseases. Obesity with non-alcoholic fatty liver, in addition to diabetes and
some of the metabolic syndrome and then obviously alcohol use and sever alcohol use disorders are the behaviors that mediate this. So important for us to understand because as Dr. Fontana pointed out, we are kind of getting rid of Hepatitis C. I’m not gonna say we’ve got it completely cured, or completely treated, but all of the projections indicate that in the coming years we’re not
going to see as much of this because the direct acting antivirals are so effective at getting rid of it. And so what is rising to take its place? It’s these two diseases. It’s non-alcoholic fatty liver
and alcoholic liver disease. And as we’ll talk about, I
tried to select some abstracts from a very large group of
abstracts, over 500 abstracts in these two areas that
might have the most clinical relevance for us
right now in clinical practice and ways to think about how
we care for these patients, how we diagnose and what
implications that has for us as we take care of them. So starting off with non-alcoholic
fatty liver or NAFLD. As we all know, NAFLD
is really a spectrum. It’s beginning with just fat in the liver and then progressing on to
those who have inflammation or non-alcoholic steatohepatitis and in some of those patients, a percentage of them will
go on to develop cirrhosis. This was an article
from Dr. Allen at Mayo, looking at the cardiovascular risk, as it relates to gender in
patients who have NAFLD. What we all know and
are quite familiar with is that in general, prior to menopause, women are generally
protected compared to men. They tend to have lower risk of hard cardiovascular outcomes, like heart attacks,
stroke, AFib, et cetera. But what Dr. Allen’s abstract looked at was whether or not NAFLD
influences this risk separately. What we also know is in general, in epidemiologic studies and studies of cardiovascular outcomes, NAFLD gives you an excess
risk above and beyond diabetes, hypertension,
hyperlipidemia, smoking, those types of things, not only for hard cardiac outcomes, but also for some of the
intermediate measures that we have of cardiac outcomes like
coronary artery calcium, cardiac intima-media thickness or carotid intima-media thickness. In all ways, it seems to be a separate and independent risk factor. And what Dr. Allen wanted to do was explore sex related differences in that cardiovascular
risk in NAFLD to see if women versus men had differential risk and if it tended to
give you an excess risk. So this was a long term study. Again, retrospective from the Rochester Epidemiologic Project looking back at NAFLD patients in the Mayo Clinic area in Olmsted County. And they looked at incident
cardiovascular disease events after the NAFLD diagnosis. So your time point was the NAFLD diagnosis and their heart outcomes were as they typically are in these studies. Angina, myocardial
infarction, heart failure, AFIB and stroke. They matched them four to
one with cases and controls. And they matched them in all the things you would want to match on
for cardiovascular risks. Their age, gender and then
cardiovascular risk factors, as well as any baseline
known cardiovascular disease. And they had just over 4,000 subjects, with the median age of about
52, and about half were women. So they had a good, well
characterized cohort. And what they found is that
the risk of these incident cardiovascular events
all total was about 1,600 over the course of this study. And I can tell you, having
done one of these studies in the Framingham Health System, it can be a little bit
difficult to get that number of incident cardiovascular outcomes. You really have to follow
people for a long time to be able to get that level of incident cardiovascular disease. They had enough and enough
power to be able to tell this. And what they found here was that in men, men with NAFLD, did not
appear to increase the risk of incident cardiovascular events, but for women it did. And it increased it, and not greatly, but definitely statistically
significantly up in all events to about 1.21 here and then when they looked
at individual events, angina, MI and heart
failure, you can see here, that they had some slightly
differential effects, but all significant. For AFIB and stroke, that actually wasn’t
significant for gender. And they also found that these events tended to occur more
frequently and earlier compared to the general population. So what you’re seeing here is, this doesn’t project as
well as I would like, but, the solid lines are the people with NAFLD and the dashed lines are their controls. And the controls, you can see, it’s the progression
that we typically see. It’s related to age, it progresses in almost a
linear fashion across age and men greater than women as you age. But with NAFLD, it tended
to here be earlier. So you had more
cardiovascular events earlier and women were greater than men here. That crossed over as they got older. So men, again, had more
cardiovascular events with NAFLD compared to women with NAFLD, but both of these were higher
than any of the controls, up until you hit about the age of 65, when those lines began to converge. So really the takeaway is here, again, this is retrospective, so it does have retrospective
study limitations, but there do appear to be an attenuation of those cardiovascular protective affects of gender. And what this means for us clinically is to be aware of this
and to really focus on the cardiovascular risk
protection in people, maybe even at a younger age than we might normally think about it. Checking lipid profiles
and being aggressive about diet and exercise,
controlling blood pressure, diabetes and hyperlipidemia to try until we get effective
treatments for NAFLD and NASH to help with the cardiovascular
risks of these patients. So the next abstract that was presented in one of the parallel sessions was about okay, so how do we find these people? There’s multiple different
ways of diagnosing fatty liver and obviously we wanna not just diagnose fatty liver on its own, but we want to try to diagnose the more advanced forms as well. So NASH and obviously cirrhosis. We want to find these people. And one of the ways that has
been looked at to do this is with Fibroscan. This has been used quite
extensively throughout and this study looked
at how feasible was it to do this in a primary care clinic? So in people who are just walking in and don’t have pre-existing liver disease and you’re looking at whether
or not they have NAFLD. Can you do this and what
are some of the issues with screening in a primary
care clinic for this? They looked three months prospectively. All patients who came in to a primary care clinic in California, so this was a population
in Coronado, California, they consented just over 800 patients and everybody underwent
Fibroscan in this clinic. Nobody had liver disease. They excluded people who had pre-existing or known liver disease,
HEP B, HEP C, et cetera, and they used what most
frequently is called Fibroscanner or Vibration Controlled
Transient Elastography. This is a method of determining, essentially, liver stiffness. You’re looking for how stiff is the liver as a shear wave is passed
through the liver tissue. And there’s a calculation
based on Hook’s Law. we can dig that out of our physics classes from undergraduate. I knew that that would
come in handy some day. I didn’t know it at the
time, but now it has. There’s a buncha math that gets done here. And they calculate how stiff is the liver. And in the proper context, acknowledging some of the
limitations of the study, of the use of this,
which we’ll talk about, you can measure how stiff that liver is, how much fibrosis or scar
tissue is in the liver. There also is now a
relatively newer technology within the last couple of years called Controlled Attenuation Parameter, which again, looks at the attenuation of the tissue under ultrasound. The takeaway from that is it
gives you an estimate of fat. You can look at both fat and fibrosis in the liver with Fibroscan. In general, the multiple
studies that have been done on this technology show that
it’s actually pretty good for distinguishing the extremes. You can distinguish, you
can say cirrhosis or not. You can say a lot of
fat or very little fat. When you get into the
gradations of fibrosis, we generally measure
it f zero, no fibrosis or f four is cirrhosis, when you can kind of get
into those gradations of f one, f two, it
doesn’t do quite as well. So we use this a lot of times
to say do ya have cirrhosis or do ya not have cirrhosis? And that seems to be where
it functions the best. Some caveats to it, are that
if you’re going to use it you want to make sure that
there’s not gonna be anything going on with the patient that’s going to falsely
elevate the stiffness or is going to have too much tissue or too much distance in between the probe, which sits right here between the ribs and that liver. The more distance you
have, between the probe tip which sits right at
the skin and the liver, the greater the degree of error that you’re going to have with that study. People who have ascites. We don’t use it in
people who have ascites. If you have heart failure, and you have passive
congestion in the liver, it’s going to look
stiffer when it may not be just because of the backflow. If your ALT is too elevated, over 100, that can indicate more
inflammation in the liver and inflammation, as you might imagine, might give you a greater
degree of stiffness, not truly reflective of scar tissue. We want people to be fasting, ’cause if you eat, it’s
gonna increase blood flow. Again, you might have more stiffness that’s not necessarily
reflective of scar tissue. Very severe steatosis
actually can be a confounder. So having the CAP method
with this can be helpful. And then obesity, BMIs 30 or
above, can be confounding, again ’cause we have this distance in between the liver tissue, so there is an XL probe that comes along with this, is
being used increasingly now. The technology of which
is a design to help us be able to more accurately
sense true stiffness across a greater distance in between if you have more
subcutaneous tissue there. What they found in their study was that about 2/3 of the patients were female, they’re fairly
young, only about 45. Mostly Hispanic and they
had quite a few people who had BMIs over 30, but they didn’t really comment extensively about if they used the M or the Xl probe, so have that in your mind. About 13% had diabetes and
13% had hyperlipidemia. And they found that
about a third of people had significant fatty liver. They’re kind of above 290 on their CAP and then for the fibrosis score, they also found that they
had about 4% of people had a score over 15, which
is very clearly cirrhosis in this population. About 12% had a score from
anywhere from seven to 15, which is consistent with advanced fibrosis or kind of the f three to f four scale. The takeaways from this were
really that it’s feasible to do this in a PCP clinic and you might actually be
able to identify some people who have more advanced
fibrosis and cirrhosis and then get them into
appropriate screening with HCC screening protocols et cetera. It may help your patients be
more motivated to lose weight. Sometimes when patients
have a knowledge of negative consequence of something, so you have liver disease, you
have more advanced fibrosis, those types of things can be
motivational for some people, to help them lose weight
and with NAFLD and NASH, weight loss is bar none the best way to help this get better. There have been studies
that have been presented a couple of years ago where
10% of the body weight or greater in patients lost and kept off in paired biopsy studies led to resolution of fatty liver and NASH in as
many as 80% of those patients. Generally if they’re in
the cirrhosis category that’s not going to go away,
but it’s pretty powerful to be able to tell your patients that if they could lose 10%
of their body weight or more and keep it off, they
might actually see healing of their fatty liver and some
of their fibrosis over time as they keep that off. They didn’t really do alcohol
use questioning in this and that’s actually quite important because alcohol use can,
as you might imagine, cause inflammation and
falsely elevate that score. If people are using alcohol, you may want to not do a
Fibroscan right away on them but give them some time
in abstinence first before you go back and Fibroscan them to see how much scar
tissue is in the liver. This is just a nice graphic
from our own Dr. Tapper, who’s gonna be speaking later on the things to know about Fibroscan and how to use it. The first, most important
thing, is that you need to know what that underlying disease
is because your cutoffs for what is and isn’t high
grade fibrosis or cirrhosis change based on your disease process. So as you can see here, some of these kilopascal scores
for cirrhosis or fibrosis, they’re different depending upon what the validation studies have shown in these various different liver diseases. Knowing that allows you
to interpret that result. And again, you wanna obviously
pair this with your history and physical pair it with your labs. If, for whatever reason, you
feel you can’t do a Fibroscan, then you wanna be considering other ways to potentially assess
for advanced fibrosis, both with blood testing, some of the cirrhologic scoring systems. We offer an MR Elastography at the University of Michigan now that can be helpful for obese patients for whom Fibroscan is
not going to be accurate. Functions in a similar
way with elastography, meaning vibration through the liver, but it’s not a probe
that’s gonna be impacted by that distance in between
your liver and your skin. So that’s another option if
you’re dealing with people whose BMI is too great and you feel it’s not
going to be accurate. Moving on to alcoholic liver disease. The first abstract that we’ll talk about is actually one that
came out of our center looking at the burden
of alcoholic cirrhosis in the United States. As many of you are aware of, we really have a huge problem
with alcohol use disorders and there was recently a large article in the JAMA Psychiatry that
looked at a very big study of the epidemiology of
alcohol use disorders. It’s called the NESARC and
it’s a large survey study, very, very good survey
study that the NIAAA, the Alcohol Research arm of the NIH runs about every 10 years or so. And what they found was that the rate of alcohol use disorders in the U.S., the alcohol use disorder
being the more severe form of alcohol use problems had gone up by 50% over the course of the past 10 years. And in women, it was 80%. We’re seeing more alcohol use problems, but because of the nature of the American Health Care System, and our datasets, it’s hard to really estimate
how big of a problem is alcoholic cirrhosis in the U.S. Anecdotally, we see this a
lot and I’m sure you do too. Certainly it’s the second leading
cause for liver transplant and that’s going up, not
down, across the nation. There’ve been studies about that as well. And so our aim with this
was to try to estimate the health care burden overall in privately insured
alcoholic cirrhosis patients. Why privately insured? The Center for Medicare and Medicaid, where many of these patients
are probably located in terms of insurance, had eliminated substance
abuse related claims from their dataset because
of concerns about privacy, so that makes that dataset
tough to use, obviously, if you’re going to be studying a very substance abuse related disorder. We looked at a large
dataset called MarketScan, which is tens of millions of patients with employer sponsored insurance across the nation. And it comes with a weighting strategy that allows you to make estimates to 50% of the U.S. population,
so 150 million people roughly, who have insurance bought and
paid for by their employer. It’s a big dataset, it’s
very well characterized and it gets us this larger
sense of what’s happening with these patients across the nation. We looked at this from 2009 to 2015. We defined it by ICD nine codes, which is how most administrative
claims datasets are done, where you look at the ICD nine code and define it in that sense. At baseline, these patients’
ages were roughly similar. We had about 66,000 patients who had alcohol related cirrhosis and about 100,000 that had
non alcohol related cirrhosis. The women were about a third of the alcoholic cirrhosis patients as compared to about 50% of those patients who did not have it. They had long term coverage,
so we had a lot of data in terms of their coverage and Elixhauser is a co-morbidity score to detect how sick someone
is in another sense, how many other co-morbidities they have. And alcoholic cirrhosis patients tended to be slightly
sicker than non alcoholic. When we looked at the prevalence
across this time frame, so looking at how big of a problem is this and is it growing, what we found was that it actually is growing
by about just over 40% across this relatively short time frame and that alcoholic
cirrhosis patients, again, made up just over a third
of that total burden. When we did some subgroup analysis to see was this going up in men versus women, it was slightly higher in
terms of rate for women, but we found that in the young population, people under the age of 45, it had actually tripled
across that time frame, even though the numbers were quite small. Anecdotally it kind of matched some of what we were
seeing in our own hospital, where very, very young patients are coming in in their early 30s, sometimes even their late 20s, with advanced alcohol
related liver disease. Alcoholic hepatitis and
alcoholic cirrhosis, even at that young age. We looked at event rates, meaning how many decompensating events did these patients have? Trying to see, are these
patients sicker at presentation? And effectively, the
takeaway from this slide is that they are. When they show up with
their first diagnosis, they are more likely to have ascites, hepatic encephalopathy, a
variceal bleed and kidney injury and when we project this out
to two years, that holds true. They tend to have more
decompensating events and to just be sicker
when they’re showing up for the first time and that
stays that way over time. Obviously the data doesn’t tell us this, but it makes one wonder, are
we missing these patients or are we missing these patients? Was there an opportunity for
us to help them stop drinking, which might help them
not decompensate as much. They’re admitted more
frequently certainly, for all reasons and also for cirrhosis, alcohol and readmissions as well. So not only are they sicker, but they’re using up more or being admitted and readmitted more. Just overall sicker patients. And in terms of the cost, when
we look at per person costs, the highlighted line is for
one alcoholic cirrhosis patient compared to a non
alcoholic cirrhosis patient and it’s almost double. Single patient to single
patient, almost double. What you see here in
the subsequent lines is an isolation of ascites by itself, so comparing any person with ascites, to any person with cirrhosis without it. And what this tells us
is that as expected, these complications, regardless
of where they’re occurring, are costing more in these patients. The cost of alcoholic cirrhosis patients is really coming from
their decompensation. It’s coming from the fact
that they have more ascites, they have more bleeding, they have more encephalopathy in HCC. And overall, they cost about
over half of the total cost to the health care system in these privately insured patients. It’s about $9.5 billion in 2015 and just over half of that was coming from our alcoholic cirrhosis patients. The takeaway of this
was that we really have a defined now population, even in the privately insured cohort, where we might expect
them to be healthier. Again, remember, this is
in people who are employed, or have employer sponsored insurance or their dependents. This isn’t in Medicaid, Medicare patients or in the VA system where
often these patients might move into those systems once they become too sick to work. We expect that this is actually
the floor of the problem and that if we were look
and see a Messer, Vijay date it would actually get worse. We’re gonna see more
costs and more illness with a consequence of this. Once of the other areas of debate that comes along with advanced
alcoholic liver disease is should we or should we
not be transplanting patients who have alcoholic hepatitis? Alcoholic hepatitis often co-occurs on top alcoholic cirrhosis. This is a very severe
inflammation of the liver that comes about from
heavy, heavy alcohol use and then results in a
systemic inflammation and extremely high mortality. As high as 50% at three months in patients who have alcoholic hepatitis. And as you might imagine,
it’s prompted people to ask should we be transplanting these patients? And by way of background,
one of the seminal papers that was published about
five, six years ago in the New England Journal
came out of the French study that looked at transplanting
highly selected patients with alcoholic hepatitis
and looking at how they did. They did actually fairly well. When you look at the survival curves here, looking out one to two
years, about 3/4 were alive compared to those alc hep patients who didn’t get transplanted,
who actually died. Of course as you might imagine, this is not without controversy. Not every center does this. We at Michigan do not transplant
acute alcoholic hepatitis, but other centers are beginning to look at larger scale studies of how people do. What this initial study looked at was how do these patients do
compared to status 1A patients. Your status 1A patients are your acute liver failure patients who come in suddenly very sick, they get listed, they go
to the head of the class right away because they are so sick and they get allocated organs first before anyone else on the
list because they are so sick and their mortality is so high. So they compared these alc hep with status 1A patients
to see how they did. And this group looked
at five years of UNOS or United Network for Organ Sharing data, which collects data on
all liver transplants that occur in the United States. They compared waitlisted alc hep patients with status 1A patients and they divided those into effectively Tylenol and non-Tylenol
related acute liver failure because Tylenol is the number one cause of acute liver failure in the U.S. And they looked at 90
day waitlist mortality, a 90 day transplant rate and then one and five year
post transplant survival, which are really our
standard outcome metrics for transplant. And what they found was that here, if you look at the acute alc hep, the takeaways are that
their transplant rate was comparable to the non DILI group, but greater than those for Tylenol. This may be a consequence of
potentially Tylenol patients who were listed, then
recovering, it’s hard to know ’cause they didn’t present that data. But their one and three year
post liver transplant survival were actually quite
good and in this study, better than those who were transplanted for ALF of whichever cause. The takeaway from that
is that the outcomes actually look like they
may be just as good. In this study, which is called
the Accelerate-AH Study, they provided us with a
little more granular detail about a rather large group of patients who had been transplanted
across multiple centers with acute alcoholic hepatitis. Again, the aim was roughly the same, to determine how do these people do when we transplant them. Should we continue doing that or not? They looked at 12 transplant centers across eight UNOS regions. Again, transplant is by region, by geographic region across the U.S., so they had a good coverage of the U.S. And they looked at a time
frame of just over 10 years. Anybody who’d had alcohol use within six months of transplant and met the clinical definition
of alcoholic hepatitis was included in their cohort. Note that they also had
to have no other prior liver disease diagnosis and no prior history of
alcoholic hepatitis episodes. This gets at the choice of these patients. What you see in these studies and in transplant centers that are transplanting alcoholic hepatitis is it is a very select group
of alcoholic hepatitis patients that meet the criteria
that they’ve assessed and typically if they’ve had any other alcohol related liver disease in the past and drank over it, people typically won’t
transplant those individuals. If they had that before, they
did not transplant them here and they looked at survival. They also, importantly, really importantly for these studies, they looked at alcohol use post transplant and they did a nice job of mirroring the addiction medicine literature and looking at slips, where you go back to some alcohol use, but you return to abstinence quickly, versus relapse, where you
go back to heavy alcohol use and there is not a return
to abstinence quickly. One of the challenges with this literature is that frequently we
conflate those things. And in the hepatology literature we put it all together. Any alcohol use is considered the same, but that isn’t true and
it isn’t evidence based with the addiction medicine literature. They did a nice job in
making it evidence based with our addiction medicine colleagues and distinguishing
between slip and relapse. They had 147 recipients with alc hep. 147 over 12 centers over 10 years. That gives you an idea of how rare it is for an alc hep patient
to qualify for this. Most were male and white,
2/3 were privately insured, some of which may relate
to some of the requirements for substance use that are attached to other forms of insurance. The median pre-transplant
abstinence was 55 days. Much less from the typical six months that you’ve heard about
that, we often hear about. Over half had received steroids before they were transplanted and we’ll talk about
this a little bit later in the subsequent talk, but Prednisolone, as many of you know, is used often in acute alc hep in certain scenarios to see if people improve and if they do, they go on to have potentially a month’s worth of Prednisolone to try to help them recover from alc hep. If they don’t, these were often the people who got in to this study
to be transplanted. Their median meld was really
high at transplant, so 39. The meld caps out at 40. These patients were very, very sick and they followed them for
just about a year and a half. What they found was that
there were approximately about 25% used alcohol
afterward in some form. At one year, about a
quarter would’ve used it, but that increased at three years to about a third of the
patients who had used it. And the median time to that first drink was about 160 days after
they left the hospital. Factors that were predictive of this was a lack of self admission at the hospitalization
predictive of drinking. If they’re not saying yep,
yep, yep, this is alcohol and not having that level of insight that was predictive of
going back to drinking, but the survival at one and
three years was 94% and 84%, even with that level of slip
versus slip and relapse. I should also distinguish here, the slips versus the relapses again. Relapses were slightly
less common than slips. If we miss a slip, it
can turn into a relapse, so it’s important for us to
catch those types of things in our patients who have
this kind of an issue. Predictors of death then afterward were if you had more
than 10 drinks per day prior to coming in to the hospital, so the heavy, heavy drinkers,
more than 10 drinks. And if you had any alcohol
use at all post transplant, that was slip or relapse, that
was also highly predictive and this was really one of the only things that was predictive on multi-vari analysis of post transplant death. They had 18 deaths overall. Nine of them actually occurred within three months of transplant and most of those patients,
eight out of the nine, had received steroids and
five of them died of sepsis. This gets at the infectious
complications of alc hep and of the steroids that we sometimes use as being an issue in that
early post transplant period. When you get out to one
year after transplant, it’s alcohol, seven out
of nine of the patients who died after a year was
because of alcohol use and because they had
gone back to drinking. The take home from this, for me, is that it’s possible to do this, and it is possible to do it with relatively good outcomes long term. However, you do have to have very robust alcohol cessation
efforts after transplant. You have to follow people very carefully and be very selective about doing this. Again, we do not transplant these patients at University of Michigan currently, but centers that are choosing to do so really need to be aware of these issues of post transplant alcohol use and that it isn’t insignificant. It’s one of the biggest reasons why these patients die long term. It really points out the need for us to be able to better manage that alcohol use disorder in the long run and come up with some better
management strategies, medications, and behaviorally
to help these patients stay off alcohol. Thank you.

6 Comments

👉 CIick Here: Reverse Fatty Liver Disease Naturally!

Nov 11, 2018, 2:08 am Reply

FATTY LIVER REMEDY

Johnny Gross

Jan 1, 2019, 12:52 am Reply

after a C-scan, i was diagnosed with Hepatic steatosis., 6 months i've lost 30 pounds, changed diet, excercised, how do I know if I'm cured? blood Test?

Martin Irving

Mar 3, 2019, 10:56 pm Reply

I know alcohol is bad when used to excess. But, it just seems very misplaced to be so concerned about alcohol liver disease when something like 50% and more like 75% of the entire US population has NAFLD. What gives with this? Sugar just keeps getting a free pass? Ignore the elephant in the room and focus on minutiae? Sort of like insulin vs cholesterol?

Afia Ama

May 5, 2019, 7:59 pm Reply

Be sure to read honest and real reviews of The Fatty Liver Solution on my blog before you buy. Go to gohonestreviews. com/the-fatty-liver-solution-review/ Thanks, Berke.

indo

Jul 7, 2019, 8:23 am Reply

Is alcohol fatty liver cured with keto just like nafld does?

Torok Akos

Aug 8, 2019, 9:33 am Reply

You're absolutely right. However, I do not understand the biological difference between alcoholic and non-alcoholic fatty liver? How do I know which one we have? Because today almost everyone drinks alcohol and many of them a lot. …. ??? Thx.

Leave a Reply