Applying Nutritional Ketosis to Treat Type 2 Diabetes by Dr. James McCarter, August 29, 2019

Applying Nutritional Ketosis to Treat Type 2 Diabetes by Dr. James McCarter, August 29, 2019

November 5, 2019 1 By William Morgan


today we’ll be talking about applying
nutritional ketosis to the treatment of
type 2 diabetes this is a talk that I
gave earlier this month at the American
Association of Clinical Endocrinology in
Kansas City my name is James McCarter
I’m an MD and a PhD I’m an adjunct
professor of genetics at Washington
University School of Medicine and the
former head of research at Virta Health
the continuing medical education
objectives for this talk are the five
following points and during the talk
you’ll see on the various slides a gold
asterisk that marks where some of these
objectives are covered we’ll talk about
describing the evidence for carbohydrate
restriction in the management of type 2
diabetes distinguishing between the
terms nutritional ketosis fasting
ketosis and ketoacidosis we’ll talk
about the role of ketones as both energy
sources and signaling molecules we’ll
discuss the rationale for the reduction
of glycemic control prescriptions during
nutritional ketosis and the treatment of
type 2 diabetes and hypothesized factors
that allow for sustained behavior change
in diabetes reversal so before diving
into this let me give you just an
example of one of the thousands of
patients that is being treated with this
protocol for type 2 diabetes Nate has
had type 2 diabetes for 24 years and
when he began working with us he was on
290 units of insulin and with the
treatment he’s been able to bring down
his hemoglobin a1c while simultaneously
getting off of insulin he’s been able to
bring down his blood pressure while
getting off of blood pressure
medications and he’s also seen weight
loss and restoration of metabolic health
and what Nate said about the experience
was the following he said when I first
stopped my medicines I was scared but I
trusted my coach and personal doctor’s
guidance first I cut back the insulin
and then my blood pressure meds by half
then I got the news not to take insulin
injections or the blood pressure meds at
all as I loaded my needle the morning
after getting the news it dawned on me
that I didn’t have to do that anymore
after so long using insulin as a crutch
to
control my glucose now I just looked at
the bottles and needles sitting there
and they reminded me that I’m conquering
diabetes I have hope
I have the following financial
disclosures I’m a shareholder of Virta
health it’s a company founded in 2014
and headquartered in San Francisco it’s
a nationwide telemedicine provider and
full-stack technology company focused on
the treatment of type 2 diabetes and
metabolic disease I’m also the chair of
the Scientific Advisory Board of Readout
it’s a company founded in 2018 and
headquartered in st. Louis it’s a
digital health company that helps
customers manage their health through
real-time biomarkers so here’s the
outline for the talk today and we’ll try
to get through this in under an hour if
I can we’ll start with some definitions
we’ll talk about changes to the American
Diabetes Association guidelines origins
and mechanisms of the process of using
nutritional ketosis for the treatment of
type 2 diabetes including prior clinical
evidence the outcomes of the trial
that’s been conducted by Virta health
and Indiana University Health talked a
little bit about how care is provided
how it’s sustainable and then some
conclusions and for those of you who
stick around I have got some appendix
slides as well
so let’s begin with definitions what is
diabetes reversal and diabetes remission
so the cost of diabetes is staggering
over 10% of US adults have diabetes
about 95 percent of that is type 2
diabetes about 5% is type 1 diabetes and
1/3 of US adults amazingly have
pre-diabetes and the cost of caring for
diabetes as well as the economic impact
of that disease is well over 300 billion
annually so in terms of the idea that
you can actually reverse type 2 diabetes
somewhat novel concept but there
actually are three clinically validated
methods that patients may choose from to
reverse type 2 diabetes bariatric
surgery
very low-calorie diets and low
carbohydrate including ketogenic
nutrition patterns and I like this quote
from a recent review that our team did
as a society we can no longer tolerate
the continued rising rates of diabetes
some patients would surely choose
reversal if they understood that it was
a choice the choice can only be offered
if providers are not only aware that
reversal is possible but have the
education needed to review these options
in a patient centric discussion so I
like the emphasis on putting the patient
at the center and emphasizing patient
choice so definitions are as follows
there’s a 2009 American Diabetes
Association consensus panel definition
of remission that cause calls remission
glycemia below the diabetic range of
hemoglobin a1c of 6.5 without using any
pharmacological or surgical therapy in
2013 bariatric surgeons came up with a
similar recommendation for a remission
definition but of course allowing for
surgery there’s other definitions that
have made a carve-out for metformin as a
drug that can be continued to use be
used even in in the case of remission or
reversal and so that’s kind of the
direction that we’ve taken with Virta
health and Indiana University Health in
our clinical trial we define reversal as
a glycaemia below the diabetic range of
six point five in the absence of
pharmacological or surgical therapy but
with the exception of metformin and
that’s because metformin is increasingly
being used in pre-diabetes to prevent
progression to diabetes and we’re
continuing to prescribe metformin in 64
percent of our patients and we do
prescribe it only in the case of side
effects or at patient request and
importantly we should point out that
reversal or omission does not mean a
cure for type 2 diabetes if patients
return to their prior behavior patterns
and nutrition they can see a
reoccurrence of the disease it’s amazing
actually how rare remission is under
usual care in 2014 in diabetes care a
group from Kaiser Permanente published a
medical records review of one hundred
and twenty-two thousand adults that
showed a seven-year cumulative incidence
of remission of only one point four
seven percent in terms of had shown any
reversal over that entire seven-year
period even doing a carve-out for
metformin as we do with the definition
of reversal cases of reversal are
extremely low with usual care so let’s
talk about the three methods of diabetes
reversal there’s a large evidence base
from bariatric surgery and meta-analysis
shows
63% remission at 18 months and that
long-term reversal can be achieved it’s
lower than 63% but long-term complete
reversal of 24% and that that is
associated with a reduction in
cardiovascular events and death but of
course bariatric surgery has the
disadvantage of that it is a major major
surgical procedure with costs and risks
that are associated with that surgery as
well as ongoing nutritional challenges
that come from the alteration of the
gastrointestinal tract so can you
achieve the same thing without requiring
a surgical procedure one approach is
calorie restriction using a very low
calorie diet also called a protein
sparing modified fast requiring less
than 900 kilocalories per day for three
months or more and that can be done with
meal replacement shakes and followed by
the reintroduction of whole foods and
was first demonstrated actually by
Bistrian and colleagues published in
diabetes in 1976 in hospitalized
patients and they show that they were
able to rapidly remove exogenously after
restricting calories but continuing to
provide enough enough protein to prevent
the loss of lean body mass and then
that’s been kind of taken up again more
recently
in Spain and in the United Kingdom including the DIRECT study published in Lancet
that showed a 46% one-year
remission beginning with patients that
were not on exogenous insulin and the
question that’s one of the challenges of
this method is the reintroduction of
whole foods so if people have become
used to getting calories, a very low amount
of calories from meal replacement
how well do they do in terms of
following the reintroduction of whole
foods in terms of sustained remission
and then the third method of type 2
diabetes reversal is a very low
carbohydrate nutrition pattern including
the use of nutritional ketosis and
that’s the approach that we’ve taken and
here I’m recapping some of the results
of the trial that I’ll get to later in
the talk but we see a diabetes reversal
rate following two years of treatment of
over half so fifty four point seven
percent of our patients that complete
two years of the trial have diabetes
reversal nearly 20% meet the criteria
for diabetes remission and we see much
lower rates of diabetes reversal and
remission in the usual care arm of the
trial so continuing with definitions
what actually is low carbohydrate
nutrition and this is a definition that
can be quite confusing and sometimes
people will allow quite high levels of
carbohydrate in the definition of a low
carbohydrate diet but really what we’re
looking for is that for a very low
carbohydrate or ketogenic diet that is
going to be less than 50 grams per day
of total carbohydrate less than 10
percent of daily kilocalories it often
requires less than 30 grams per day to
achieve nutritional ketosis defined by a
blood beta-hydroxybutyrate level of 0.5
millimolar or higher very low
carbohydrate diets are not high protein
diets they are moderate protein diets
where fat either from body fat stores or
from the diet is the primary energy
source and those dietary fats can
include saturated monounsaturated and
polyunsaturated fatty acids
now moving on from very low to just low
carbohydrate that allows somewhat higher
levels of carbohydrates generally 50 to
100 but not enough to achieve ketosis
unless you’re looking at times of
exercise or fasting let’s dive a little
bit more into fueling on ketosis and
where is the energy coming from if
somebody for instance has type 2
diabetes and and obesity and is in a
weight loss phase of the use of a
ketogenic diet they’re going to be
drawing upon fat both from the diet in
this example 820 kilocalories coming
from the diet but then even more so
coming from body fat stores during the
weight loss phase and you can see that
here that the amount of carbohydrate and
protein remains relatively constant over
the course of the dietary phases but
that what happens is that as the patient
moves from the state of obesity to more
of an ideal body weight that the fat
loss diminishes over time and eventually
all of the energy is coming from dietary
fat and in that case the kilocalories
from fat make up about 80 percent of the
energy so what do these diets look like
nutritional ketosis can be highly
individualized and that’s one of the
things that we’ve emphasized at Virta
health we’re looking at carbohydrate
restriction of generally less than 30
grams initially in order to achieve
nutritional ketosis highly personalized
to budget culture religion dietary
preferences for instance an omnivorous
diet versus a vegetarian diet we
emphasize education and problem-solving
not meal replacement or or delivery we
emphasize eating delicious Whole Foods
until satisfied and not emphasizing
calorie counting so the idea is to eat
to satiety and importantly it’s not a
zero carbohydrate diet the dietary
recommendations include recommended five
servings daily of non starchy vegetables
plus calories
is coming, if the patient prefers use of
nuts and berries and other sources
that they can incorporate
into their diets so it’s a very
satisfying diet so one of the exciting
things that’s happened in 2019 are
changes to the American Diabetes
Association guidelines in 2019 ADA
changed both its standards of care and
it’s nutrition consensus statement to
incorporate very low carbohydrate
nutrition as an established eating
pattern for the treatment of type 2
diabetes and this is the first time that
they have made that statement quoting
from their April nutrition consensus
statement low carbohydrate eating
patterns especially very low
carbohydrate eating patterns have been
shown to reduce hemoglobin a1c and the
need for anti hypertensive medications
these eating patterns are among the most
studied eating patterns for type 2
diabetes this eating pattern does not
appear to increase overall
cardiovascular risk with long term
studies with clinical event outcomes are
needed the other organization that also
changed their consensus statement around
nutrition as the European Association
for the study of diabetes in a joint
paper that came out in late 2018 with
ADA in diabetes care and interestingly
the Veterans Administration and the
Department of Defense was actually a
little bit ahead of the American
Diabetes Association in terms of changes
to their guidelines back in 2017 they
incorporated low carbohydrate nutrition
into their recommendations for the
treatment of diabetes so let’s talk a
little bit about origins and mechanisms
low carbohydrate eating patterns
actually date back thousands of years to
Paleolithic times if you look at
hunter-gatherers their amount of
carbohydrate in their diet actually
varies based on the environmental
availability and high fat ancestral
eating patterns were seen in Native
Americans of the Great Plains and
Pacific Northwest in the inuit the Masai
Mongol nomads and then fasting has been
used as a religious practice and for the
treatment of disease since ancient times
thousands of years
putting the body into a state of ketosis
is beneficial for a number of different
medical conditions in terms of the
published medical literature the first
use of caloric restriction and ketogenic
diets for the treatment of diabetes can
be seen in the literature in England and
Italy in Germany and the 1800s and then
in the early 1900s in the United States
often times when people talk about use of
ketogenic diets they’ll reference
epilepsy that was first published in
1924 by a group from the Mayo Clinic
they had actually adopted a diet from
the University of Michigan that was
being used in the treatment of type 1
diabetes and was published in 1920 and
then these diets generally fell out of
favor with the availability availability
of insulin for the treatment of type 1
diabetes as well as the demonization of
fat in the diet with the rise of the
diet heart hypothesis beginning in the
1950s and then growing in its fervor
through the 1970s and 80s here’s an
example of what one of those early diets
looked like emphasizing sources of fat
on the Left butter olive oil cream as
well as meat and poultry fish and eggs
and cheese and then avoiding foods that
are shown on the right such as sugars
and starches so why is this effective
why does carbohydrate restriction help
in the treatment of type 2 diabetes so
one sort of simplistic piece of this is
that it gets patients off of a glucose
insulin rollercoaster if you look at
plasma glucose response two different
diets and you provide as this 2002 paper
did either glucose or fat you see a
spike in plasma glucose following a high
carbohydrate meal whereas you don’t see
that you see almost no response
following a high fat meal and putting
this in a little more of a cartoon form
as follows here we’re showing both the
rise in blood sugar over time following
a meal of carbohydrate protein and fat
and then in the dotted line the insulin
change so the insulin basically is
following the carbohydrate to
move glucose into cells to be used
as energy or to be stored and so when
you shift from a high carbohydrate diet
to a low carbohydrate diet you get off
of this glucose insulin rollercoaster as
a 2014 statement from ADA said total
amount of carbohydrate eaten is the
primary predictor of glycemic response
so thinking about this over time high
carbohydrate diets raise insulin levels
as you have that ongoing insulin glucose
insulin rollercoaster and increase
insulin resistance carbohydrate
restriction reverses this trend so look
at this as follows on the right
you’re eating carbohydrates over your
body’s tolerance that results in the
need for more insulin more insulin
release or hyperinsulinemia and then the
cells become habituated or less
sensitive to insulin over time so
there’s a lot more to this though as
well so it’s not just this glucose
insulin rollercoaster it’s also that
there’s an underlying biochemistry to
ketosis nutritional ketosis is a
metabolic state in which the body is
primarily fueled by either dietary fat
or body fat so a lot of the cells are
fueling on on fatty acids but then the
liver is converting fatty acids into
ketones beta-hydroxybutyrate acetoacetate
and acetone and that generally is
occurring when carbohydrates are limited
to 30 grams or less per day and during
fasting in any of us if you fast for two
or three days ketones will actually be
providing 60% of your brain’s energy so
brain the brain fuel is not just on
glucose but on ketones as well and the
benefits of utilizing ketones for fuel
as an alternative to glucose are several
fold one is this issue of insulin
resistance but secondly it’s over
whelmingly lowering insulin levels and
restoring insulin sensitivity and then
ketones are not just a fuel source
they’re actually they have hormonal
properties as well and they signal for
reductions in both oxidative stress and
inflammation
so a couple of definitions nutritional
ketosis is not diabetic ketoacidosis so
the levels that are seen in nutritional
ketosis are generally between 0.5
millimolar up to 3 or slightly more
level of beta-hydroxybutyrate there’s no
insulin insufficiency and there’s no
metabolic acidosis whereas diabetic
ketoacidosis results with high ketone levels of 10 to 20
millimolar beta-hydroxybutyrate
in the situation of insulin and
sufficiency and results in a metabolic
acidosis so very different states and we
don’t see that there’s any
predisposition of one to another so
being in nutritional ketosis does not
predispose a patient to having
diabetic ketoacidosis so as you look at
the fuel utilization it actually is
fascinating studies of keto adapted
athletes demonstrate that nutritional
ketosis can actually double the rates of
fat oxidation and this was published in
2016 in a study comparing 20 male
ultrarunners that were consuming either
a low carbohydrate or high carbohydrate
diet of their own choosing for at least
six months and in this case studying
these folks in a clinical setting peak
fat oxidation was 230 percent higher in
the low carbohydrate group as they were
utilizing fat for fuel and mean fat
oxidation was 59 percent higher the
glycogen levels did not differ between
the two groups either before or after
exercise so the body actually revs up to
utilize fat for fuel in the case of
nutritional ketosis so this is a really
important point there’s a real paradox
here people often will confuse dietary
fat with fat in the body either in body
fat stores or being deposited in the
liver or pancreas or in the blood and
but it’s important to remember that we
are not what we eat and as you shift
from a high carbohydrate diet to a low
carbohydrate dietary pattern you ramp
up this use of
fat for fuel and this 2019 JCI insight
paper showed that despite daily dietary
saturated fat increasing in these
patients from 40 grams to 100 grams in
this crossover study that their actual
total plasma saturated fatty acid
declined both in the phospholipid
fraction as well as in the triglyceride
fraction it actually was the high
carbohydrate situation in which there
was higher levels of plasma saturated
fatty acid and so it’s really quite a
paradox and then beta-hydroxybutyrate
there’s a number of interesting studies
over the last few years that have shown
signaling properties of this molecule so
as it accumulates in in the blood and
travels throughout the body it first is
suppressing oxidative stress by
inhibition of histone deacetylases and
secondly it’s inhibiting inflammation
through blocking the NLRP3 inflammasome
and this has been shown not just in
animal studies or in cell culture but
also in clinical studies this was a 2008
paper that looked at a very low
carbohydrate ketogenic diet versus a low
fat diet looking at 14 biomarkers of
inflammation and showed that with the
ketogenic diet nine of those 14 markers
showed significant declines and six of
those the decline was greater than was
seen in the low-fat diet okay so let’s
move on from origins and mechanisms to
looking at clinical evidence so even
before Virta
and Indiana University began their work
on type 2 diabetes there was a
substantial evidence base for the use of
low carbohydrate and ketogenic diets in
the treatment of type 2 diabetes in fact
more diabetes trials have examined
carbohydrate restriction than any other
dietary pattern here we’re showing
randomized and controlled clinical
trials in light blue and other clinical
trials in dark blue and we’re comparing
– Mediterranean plant-based and low
carbohydrate eating patterns
and almost all of those low-carbohydrate
trials have observed reduction in
hemoglobin a1c reduction in medication
use there are three recent meta analyses
looking at these trials that have
occurred mostly over the last 15 years
that have been published in 2016 2017
and 2018 and there’s an excellent review
of the literature published by Hallberg
and colleagues improving the scientific
rigor of nutritional recommendations for
adults with type 2 diabetes so what I’ll
do in the next few slides is not take
you through all 30-plus studies but just
pick out a few that are interesting to
highlight certain points the first is
from 2005 this Boden et al. paper in the
annals of internal medicine effect of
low carbohydrate diet on appetite blood
glucose levels and insulin resistance in
10 obese patients with type 2 diabetes
so this was a small study of 10 patients
just over three weeks the first week
they were on their usual diet and then
the second and third week they shifted
over to a low carbohydrate diet of 21
grams of total carbohydrate per day and
they saw this substantial decrease in
fasting glucose they saw medication
reduction in quite a few of the patients
and the insulin sensitivity improved 75%
as measured by you glycemic
hyperinsulinemic clamp this 2007 study
looked at subjects with both diabetes
and pre-diabetes and studied them over
56 weeks in a single arm outpatient
study restricting carbohydrates to about
20 grams per day for the first 12 weeks
than 40 grams for the remainder of the
study and saw decreases in fasting
glucose triglycerides and weight this
2008 randomized two-year trial looked at
three hundred and twenty-two moderately
obese subjects some of whom had type 2
diabetes the trial was notable for high
retention of 85 percent at two years the
low carbohydrate arm began at 20
per day but then that increased to well
over a hundred grams per day and the
type-2 diabetes patients showed a1c
reduction that a1c reduction was
greatest and significant for the low
carbohydrate group as opposed to the
Mediterranean diet group or the low fat
diet group for all subjects low
carbohydrate and Mediterranean dietary
patterns were significantly better than
the low-fat dietary pattern this also
was published in 2008 the effect of
low-carbohydrate ketogenic diet versus
low glycemic index diet on the treatment
of type 2 diabetes it was a 24 week
randomized trial of 84 patients lower
retention in this case 58% the low
carbohydrate arm began at 30 20 grams
per day with no restriction on
kilocalories whereas the high
carbohydrate arm was eating a low
glycemic index diet that was 55%
carbohydrate and they showed a
significant reduction in hemoglobin a1c
in the low carbohydrate but not the high
carbohydrate group as well as
significantly more medication reduction
in the low carbohydrate group and
greater weight loss this more recent
trial out of UCSF in 2014 and 2017 was a
12-month randomized trial looking at a
moderate carbohydrate versus a very low
carbohydrate diet in the treatment of
type 2 diabetes or pre-diabetes it was a
randomized trial with 34 patients the
very low carbohydrate dietary pattern
has significantly greater reductions in
hemoglobin a1c weight loss more
reduction in medication than those
following model moderate carbohydrate
calorie restricted diet this study from
2018 was a two-year study of 115 adults
with type 2 diabetes had fairly low
retention of 53% they tried to energy
match the diet and it was a hypo
caloric diet with exercise the low
carbohydrate group was 14% carbohydrate
the high carbohydrate group was 53%
carbohydrate both of these diets
actually reduced weight and hemoglobin
a1c at two years there was no
statistical significance
between the groups but there was
greater medication reduction in the low
carbohydrate arm
with less glucose variability and a
greater reduction in triglyceride I’ve
also included one very interesting
recent study that was not specifically
on type-2 diabetes but was looking at
metabolic syndrome and it’s an
interesting study because it addressed
the question of weight loss so one of
the things that’s often said is that the
success of these diets is entirely
dependent upon the weight loss so in
this case they held weight study by
either increasing or decreasing the
daily calorie consumption based on
changes in weight and in this crossover
study looking at a high carbohydrate
moderate carbohydrate and low
carbohydrate diet for four weeks each
they looked at the resolution of
metabolic syndrome in these 16
individuals from baseline to change over
each of these dietary periods and they
showed that even holding weight study
that following a low carbohydrate diet
the majority of the patients reversed
their metabolic syndrome whereas the
same thing was not seen in the high
carbohydrate or moderate moderate
carbohydrate eating pattern let’s move
on then from the prior clinical evidence
to looking at the trial that’s been
conducted by Virta health and Indiana
and University Health so this is a trial
that has been conducted over the last
several years and the results have been
published so far in six peer-reviewed
publications looking at diabetes
outcomes at one year and two years
looking at cardiovascular outcomes at
one year looking at the very early
changes in metabolic patterns at the
first seventy days as well as looking at
non-alcoholic fatty liver disease at one
year and patient reported sleep outcomes
at one year their papers in progress
looking at a separate cohort with
pre-diabetes as well as further
characterizing metabolic syndrome system
utilization depression joint function
and inflammation so Virta
health and Indiana University Health
began recruiting for this trial in 2015
it began as a two
study that was then extended out to five
years there are 465 participants that
were recruited from central Indiana
between August of 2015 and March of 2016
it’s a non randomized trial in which
patients self-select into the
intervention the intervention is
referred to as the continuous care
intervention it includes individualized
nutritional ketosis
there were 262 individuals with type 2
diabetes that undertook that
intervention along with another 116 with
pre-diabetes there is also a usual care
group that had 87 individuals with type
2 diabetes following the 2015-16 ad a
guidelines under the care of an under
chronologist or primary care
practitioner working with a certified
diabetes educator the baseline
characteristics of the type 2 diabetes
population in the intervention group
were a mean age in the mid-50s a BMI of
40 average weight of 257 pounds average
duration of diabetes of 8 year so this
included some folks that were newly
diagnosed but it also included people
that had had diabetes for 20 years or
more the cohort was 67% female retention
at one year was 83 percent and at two
years with 74 percent which is very good
for a trial that requires this degree of
behavioral intervention so here are the
highlights of that trial it demonstrated
that the continuous care intervention
had reversal of type 2 diabetes in one
year in 60% of patients who completed
that first year and that again as we
talked about in terms of definitions
reversal is defined as glycemic control
without the use of diabetes specific
medications
the average a1c reduction was 1.3
percent 70 percent of participants
achieved in a1c below 6.5
who was substantial including 94% of
insulin users reduced or eliminated
their use of insulin average weight loss
was 12 percent or 30 pounds and
cardiovascular risk factors improved
including a 12 percent improvement in
the 10-year a/s CVD risk score and
improvement in 22 of 26 risks factors
for cardiovascular disease showed
statistically significant improvement so
those are some of the highlights now
let’s dig into some more of the details
so it’s important to point out that the
hemoglobin a1c was being reduced
simultaneous with the removal of
medications so you may see for instance
that if you increased medication use you
could get a drop in hemoglobin a1c but
here we’re seeing a 1.3 percent drop in
hemoglobin a1c while simultaneously
removing many of the medications if you
look at medications by cost there’s a 46
percent cost reduction in the first year
resulting in savings of over two
thousand dollars on average per patient
and that reduction occurs very early the
most most of the reduction is occurring
in the first seventy days so as you
might suspect if folks start fairly well
controlled there’s a less of an
opportunity to improve that whereas for
people that begin uncontrolled with a
higher hemoglobin a1c you see a greater
reduction here we’re showing the
reduction in the intervention group in
blue and the change in the usual care
group in black and gray you can see for
people who begin better controlled with
an a1c of less than nine that they’re
seeing about a 1% reduction in a1c in
the intervention a slight rise in the
usual care group whereas for folks who
begin with a hemoglobin a1c greater than
nine you actually see a three point four
five percent reduction in hemoglobin a1c
and in that first year it’s interesting
to actually look at this by every single
individual patient as opposed to just
the means and the scatterplot that were
showing here if you look at the x-axis
as being their baseline hemoglobin a1c
and the y-axis as being their one-year
hemoglobin a1c no change would mean that
everyone would be along that dotted line
diagonal line and essentially that’s
what we see with usual care group you
see folks along the dotted line as well
as scatter on either side of the line
you can see that with the intervention
shown on the Left almost everyone is
below that line showing a reduction in
their hemoglobin a1c and that the higher
their starting a1c the greater the
duction and that hemoglobin a1c is an
improvement is sustained out at two
years so we see the reduction that
occurs in the first 70 days out of the
year and then a little bit of
an increase in a1c out of two years
but most of the gain is maintained so
at two years we’re seeing a 0.9 percent
reduction in hemoglobin a1c also a 32%
reduction in insulin resistance by HOMA-IR as well as a 55 percent diabetes
reversal rate of completers at two years
digging in then to the medications I
mentioned earlier that we were reducing
the diabetes medications as high as
glycemia improved and the greatest
reductions are in sulfonylurea and
insulin to avoid hypoglycemic events
during the dietary changes we see 67% of
all diabetes prescriptions are
eliminated 100% of sulfonylurea
prescriptions are eliminated 91% of
insulin prescriptions are reduced or
eliminated and there’s an 81% mean
reduction in insulin dose and so putting
this in all the details we break this
out by medication class looking at
eliminations and in dark blue decreases
in light blue no changes in gray and
then increases and new medication use in
in yellow and pink note the sample size
on the different medication classes so
for instance in the TZD’s there’s only
eight people actually using those drugs
whereas most people are using metformin
and many people began on sulfonylurea
and insulin and then if we look at
that as change over time you can see the
percent change in starting users by a
drug class with the rapid reduction
early in SGLT-2 inhibitors in sulfonylureas the reduction in the use of
insulin in DPP-4 we actually continue to
use GLP-1 in a lot of patients in the
first year of the trial and then that
decreased more substantially
in the second year of the trial that was
generally done as a way of transitioning
patients off of insulin weight-loss was
also substantial 12 percent at two years
and we were using a home scale that
was cell-enabled so we were able to get
daily weights on the patients and this
is showing all of those patients daily
weights over time 75% of patients lost
5% of their weight or more 49 percent of
patients lost 10% of their weight or
more now as you look at change in weight
and change in A1c between one and two
years it’s important to point out that
most dietary interventions display a
return of weight and a1c toward baseline
over time and that the changes we see in
the Virta intervention that continuous
care intervention are actually far less
than is seen in most trials of either
low carbohydrate intensive lifestyle
interventions like look-ahead
or very low calorie diets like direct so
they see a greater bounce back in weight
and hemoglobin a1c and the same thing is
seen in plant-based trials as well a
bounce-back in A1c toward baseline so
moving on then from diabetes status
let’s talk about the cardiovascular risk
score and here we’re looking at the
atherosclerosis
cardiovascular disease ASCVD risk score
that was developed by the American
College of Cardiology and the American
Heart Association and the following
couple of diagrams will be showing
improvement positive improvement to the
right and a negative to the to the left
and at one year the continuous care
intervention shows a 12 percent
improvement in ASCVD risk score whereas
the usual care group saw a decline in
their score and then if you break that
out by all of the cardiovascular risk
markers 26 that we were captured in our
trial and those include markers for
hypertension atherogenic dyslipidema
chronic inflammation fatty liver disease
we actually saw it that 22 out of 26
improved significantly in the continuous
care intervention whereas 0 of 26
showed statistically significant
improvement in the usual care group
so you can see all of those blue bars
moving toward the right whereas the gray
bars are generally not moving or moving
to the left so digging into greater
detail in those various categories we
can look at glycemic control measures
not just hemoglobin a1c but also looking
at glucose looking at serum insulin
looking at HOMA-IR the model of insulin
resistance and all of those are showing
improvement in the intervention group
hypertension also improves we see
substantial reductions in both diastolic
and systolic blood pressure and we see
those simultaneous with reduction in the
use of antihypertensive medicine
including a 25% reduction in the use of
diuretic medications looking at
atherogenic dyslipidemia we see an
improvement in all of these markers with
the intervention group but not the usual
care group we see reduction in
triglycerides arise in HDL and
improvement in the triglyceride HDL
ratio and improvement in the Apo-B /
Apo-A1 ratio looking at low-density
lipoprotein we see that while
the calculated LDL rises that if you look at the actual particles as
measured by either Apo-B or NMR lipoprofile for LDL-P we see a non-
significant decline in the actual number
of particles which are a better measure
of cardiovascular risk than the
calculated LDL and then looking at
inflammation both high sensitivity C
reactive protein and white blood cell
count show reductions and those
reductions are substantial at one year
we see a 39 percent reduction in high
sensitivity C reactive protein
increasing a indicating a decrease in
chronic inflammation and then we’ve
published papers on both fatty liver
disease as well as sleep and so looking
at non-alcoholic fatty liver disease we see
statistically significant improvement in
the intervention group on the left but
not the usual care group on the right
for both abnormal ALT as well as
steatosis and fibrosis all of those
scores improve in the continuous care
intervention group but not in the usual
care group we also did a number of
surveys with our patients and one of
those was on sleep and so for patient
reported sleep outcome we used the
Pittsburgh Sleep Quality Index and
showed improvement in both the type 2
diabetes population as well as here we
report the pre diabetes population as
well we see improvement in the global PSQI score in the subjective sleep quality
score in the sleep disturbance score and
the daytime dysfunction score whereas
there was no improvement in the usual
care group one of the questions we get
is that if 60% of patients are reversing
their diabetes at 1 year what happened
to the other 40% and actually they do
very well as well they just did not meet
our threshold for diabetes reversal they
show a 1.2 percent reduction in
hemoglobin a1c 45 percent medication
elimination 27 percent triglyceride
reduction 23 percent loss of weight 81
percent reduce or eliminate insulin and
a 17 percent improvement in the ASCVD
risk score so the vast majority of
patients are doing very well so far
we’ve talked a lot about outcomes let’s
talk about how this care is actually
delivered in the case of the Virta
Indiana University Health intervention
so for the continuous care intervention
in clinical practice there are two key
pillars to the way in which this care is
delivered the first is what we call
continuous remote care Virta health
built a telemedicine platform for a
personalized physician and health
coaching for our patients and that
includes five components interaction
with a health coach through asynchronous
text messaging as well as
occasional phone and video calls
interaction with a physician by
telemedicine an online patient community
for peer support online educational
resources for nutrition and behavior
change education and then biomarker
tracking to provide real-time feedback
to patients that’s also shared with the
health coach and physician the work with
patients begins with a call with an
intake specialist a telemedicine visit
with the physician onboarding with
health coach and then a starter kit with
durable medical equipment that’s
delivered to the patient for
individualized continuous remote care by
licensed medical providers and so this
is kind of what that looks like in terms
of the use of the app in a cell phone
patients it’s like having a doctor and a
health coach in your pocket the second
pillar of the clinical practice is
proprietary care it’s a medication
management and patient lifestyle
guidance as we’ve talked about it
includes the use of nutritional ketosis
on the back end
what providers are seeing is a
custom-built electronic medical record
and data tools that help to provide
continuous remote care with the
physician and the health coach and this
is a little bit about how the data flow
looks the patient is reporting on a
daily basis their mood energy hunger and
cravings as well as biomarkers for
ketones glucose and weight they are
receiving in return clinical guidance
from the health coach and physician that
is tailored for their own individual
circumstance and that results in the
vast majority of cases cases in diabetes
reversal and improved metabolic health
so a little bit about that biofeedback
there is a daily measure of blood
ketones looking at beta-hydroxybutyrate
and glucose through a finger stick there
is weight data from a cell-enabled scale
and then cell phone based reporting of
mood energy hunger and cravings as well
as user activity with the health coach
and looking at beta-hydroxybutyrate
over time you can see the steep increase
in the first 50 days in the patient
level of home reported
beta-hydroxybutyrate confirmed in gray
with the laboratory values that are seen
at baseline seventy days and one year
and part of what you’re seeing here is
that the use of nutritional ketosis is
especially strict in the first couple of
months and then as patients regain their
metabolic health they can have more
dietary flexibility over time it’s
important to point out that electrolyte
management is very important for success
during keto adaptation sodium needs to
be generally 3 to 5 grams per day for
most patients and this is because
carbohydrate restriction causes
increased naturesis so there’s
greater urine sodium output actually
insulin is signaling to the kidneys for
fluid and sodium retention and as
insulin levels decline with nutritional
ketosis the kidneys are not receiving
that signal and so therefore they are
dumping water and sodium that sodium
depletion if not replenished can lead to
hypovolemia headache fatigue and
constipation some of those so-called
side effects of a so-called “keto flu” but
those can be largely gotten rid of by
replenishing sodium patients with
hypertension are monitored closely in
terms of their sodium intake and their
use of blood pressure medications
it’s important also to pay attention to
dietary or supplementing with magnesium
to avoid muscle cramps
so what about sustainability? this
treatment aims for a lifelong change in
nutrition and behavior it’s not a
temporary diet in which patients return
to their prior behavior and prior
dietary patterns to look at longer term
effects the Virta Indiana University
clinical trial has been extended out to
five years for the intervention subjects
and in terms of what those patterns look
like over time some subjects enjoy
nutritional ketosis and plan to stay
with that long-term other patients use
their improved insulin sensitivity to
enjoy greater dietary flexibility while
staying generally toward the low
carbohydrate end of the spectrum so
generally less than a hundred grams per
day and this treatment is by no means a
cure for diabetes and we do not advise a
return to high levels of carbohydrate
consumption which could result in in a
return of the disease so some
conclusions so what can we conclude from
the work that’s been done in
particularly the clinical trials that
have been done looking at carbohydrate
restriction for the treatment of type 2
diabetes and especially the Virta Indiana University Health trial first
carbohydrate restriction is a viable
patient choice for type 2 diabetes
reversal nutritional ketosis supports
diabetes reversal by reducing insulin
resistance while providing an
alternative fuel to glucose that has
favorable signaling properties low
carbohydrate nutrition patterns
including ketosis have extensive
clinical trial evidence for type 2
diabetes improvement including both one
and two your outcomes from the Virta-IUH
trial the American Diabetes Association
and other organizations have updated
their guidelines to include low
carbohydrate eating patterns for the
treatment of type 2 diabetes and
continuous remote care using a physician
health coach education biomarker
tracking in an online community may be
helpful in supporting long-term
behavioral change there are a number of
points of agreement between the use of
ketogenic diets and plant-based
approaches to diabetes management both
advocate reduced dietary sugar and
avoiding high glycemic index
foods both approaches advocate cooking
and eating whole foods whenever possible
both include non-starchy vegetables both
include sources that are high in
monounsaturated fatty acids like olive
oil
both advocate consuming moderate amounts
of protein both support behavior change
using a coach and a community and both
support individualization and patient
choice and ketogenic diets have been
adapted for vegetarian and vegan
lifestyles so it’s not necessarily a
requirement to choose between one and
the other I’d like to thank the
co-authors and collaborators that have
made this work possible
both folks at Indiana University
Health as well as our team at Virta
Health and collaborators from other
academic institutions that have joined
us in this important work and then
especially to thank our clinical trial
participants without whom none of this
work could have been done so thank you
very much for your attention
I would love feedback I can be reached
at jamespmcarter at gmail.com also
you can follow me on twitter @JPMcCarter
and on Medium and on LinkedIn as
well so for those of you who are still
watching all the way to the end
I do have an appendix that goes over two
more topics in a little bit more detail
and the first is sort of a deep dive
into some of the lipids one of the main
questions that we get is around the LDL-C
response and a number of trials have
looked at low carbohydrate and ketogenic
diets for obesity and diabetes and don’t
show a mean increase in LDL particle
number and this is true both in the
absence of statin therapy as well as
with statin therapy so I’ve already
shown the data for the Virta-IUH trial
which was published in Cardiovascular
Diabetes in 2018 but let me take you
through some of these other papers the
most direct way of measuring LDL
particles is to look at the ApoB
protein that’s associated with each
particle
this 2006 paper looking at very low
carbohydrate diets showed a significant
decrease in ApoB following the
dietary intervention this 2006 paper
also showed a decrease in ApoB that
was non-significant this 2008 paper also
showed a decrease in ApoB that was
non-significant so none of these papers
are showing an increase in LDL particle
number here’s another one from 2009
showing a non significant change in ApoB
and then for patients who are already
taking statins this paper from 2003
showed a non-statistically significant
decrease in LDL particle number and this
paper from 2013 with patients taking
statins looking at LDL-C an indirect
measure the calculated LDL showed a non
significant increase so then moving on
from lipoproteins
to just a little bit more about dietary
patterns this is a what a food pyramid
might look like if you’re if you’re on a
ketogenic diet where you’re getting the
majority of your calories from fat from
various sources that can be both from
animal sources as well as plant sources
of fat you’re getting protein from meat
poultry fish eggs tofu and nuts and this
can be tailored to either an omnivorous
or vegetarian lifestyle importantly
you’re getting five servings a day of
non starchy vegetables you can use
dairy and then berry fruits as well
although not in overwhelming amounts and
here’s a snapshot of a diet that’s
having about 30 grams per day of
carbohydrates showing a breakfast with
bacon and eggs a snack with pecans lunch
with salmon and a salad a snack of
cucumbers and ranch dressing and a
dinner with meat as well as mushrooms
and cauliflower
and then you can formulate this as a
vegetarian diet as well here’s a
breakfast of eggs and spinach a snack of
nuts a Caesar salad without the
meat for lunch a snack of celery
with dressing and then a mac and cheese
dinner so again thank you for your
attention and I look forward to
receiving any feedback