Dr. Sarah Hallberg – Type 2 Diabetes Reversal

November 8, 2019 0 By Bertrand Dibbert

– So, I’m here to talk
about my favorite subject, besides my kids, and that is diabetes reversal. First, my disclosures. And the problem. I know in this audience here, people are well aware of this, but we are in the midst
of a terrible epidemic. So the last time this was
published in a major journal was in GEMA, looking at 2012 data. This was published in 2015. And over 50% of adults in this country have diabetes or prediabetes. I mean, it’s really striking. And the consequence of this is we can’t afford this anymore, okay? Of course, the primary consequence is people’s quality of life, their length of life is
being really impacted. But when we look at what is
really going to drive change, unfortunately, we all know money talks. And take a look at what has
happened just over six years. I mean, we cannot afford
to allow this disease to continue to spiral out of control. And here is the bottom line, type 2 diabetes is reversible. And, you know, for
everyone in this audience, this fact that type 2
diabetes is reversible is everyone’s responsibility. So for those people who
are health care providers, it is all of your responsibility to let your patients know this. For people who are not
health care providers in the audience, it’s your responsibility
to let your friends, your coworker, your
family members understand that they are not trapped in
an irreversible condition. They have their own options on how to take care of their disease. And they can back out of where they start. All right. Okay, so, there are, actually,
three clinically proven ways to reverse diabetes. Bariatric surgery, the
literature is quite robust. I mean, we do get reversal, and we get prolonged
reversal in many cases. A very low calorie diet
has also been shown to reverse type 2 diabetes. And a low-carbohydrate diet. Now, I bet no one will
be really surprised, especially at a conference
that has low-carbohydrate in its name that I’m gonna
focus on the third one. But it’s really important to understand that there’s not just one option. And that is because
patients have a choice. All right, so, what does not have evidence of diabetes reversal? Sorry, this looks like
it smudged together. And that is the standard of care. So a study at Kaiser Permanente,
a really large study, looks at what happens. What is the amount of
diabetes remission that occurs following the standard of care? And it doesn’t, okay? So what doesn’t work? We have three methods that work, one doesn’t, yet we’re constantly talking and practicing the one that has failed us. It doesn’t make any sense. So, focusing now on low carbohydrate. So, just really quickly here, why does carbohydrate restriction work? Because we know it does, but what’s the physiology behind it? And, again, it’s that
different macronutrients produce different glucose
and insulin responses. And when we remember, what is the problem with type 2 diabetes? The problem with type 2 diabetes is elevated blood sugars, right? But even before the blood
sugars became elevated, the pre-problem, if you will,
is elevated insulin levels. And it’s just so important
and really so simplistic to understand that our
three food macronutrients create very different elevations in both glucose and insulin. Carbohydrates cause them both to go up. And it is really important, really important for
everyone to understand, not just the people in this room, that fat does not cause a
glucose and insulin response. And so if we want to instruct our patients to eat something that,
actually, will control the root cause of their disease, and not just Band-Aid it as we do with more and
more and more medications, we have to instruct them to eat what scientifically makes sense. Everyone is an individual, and personalization is
key to sustainability. But although everyone
has individual quirks in their own physiology, what we see here as far as
their responses to macronutrient are just generalized to humans, to, really, mammals. Carbohydrates will cause the
glucose and insulin to go up, and fat is flat. So fat must be a part of
science-based recommendations for anyone who has type 2 diabetes. So, let’s talk about healthy carbs because our guidelines,
again, that have failed us are full of recommendations
for healthy carbs, right? Well, it’s okay, it’s a healthy carb. You know, I’m not eating white rice, I’m eating a really healthy carb. It’s brown rice. So a cup of brown rice with 45
grams of carbohydrates in it is gonna cause two really
different responses in individuals depending
on if they have a high or a low carbohydrate tolerance. And, clearly, when we’re
talking about type 2 diabetes, we’re talking about the person in red, the person who has a very
low carbohydrate tolerance. So ingesting these healthy carbohydrates are going to cause an
excessive insulin response. And, of course, we’ll go back to, why are patients with type 2
diabetes generally overweight and often morbidly obese? Because these high levels of insulin, insulin being our fat storage
hormone, cause problems. And what do we do? What do we all do? We blame the patient, right? They’re not following our instruction. But we gotta go back and look
at this simple physiology. Wait, we really set this
person up for failure. So, again, this is from the American Diabetes
Association guidelines. The total amount of carbohydrate eaten is the primary predictor
of glycemic response. Okay, so let’s take a
look at one of patients that I saw very recently. I pulled this CGM data just this weekend. So his starting insulin dose
was 210 units a day, which, if anyone who is seeing
patients with type 2 diabetes, that’s what? Pretty typical, that’s pretty typical. That is not like a patient who
comes in once in a blue moon. That’s, you know, the patient
at nine o’clock on Wednesday. And his starting A1C was 11.3. So this is three years into a low-carbohydrate, ketogenic diet. And this is his CGM reading. Insulin dose today, zero units. He’s still on 2,000 units of metformin. And his last A1C was 6.1. And I’m gonna predict
from this glucose curve, again, that I just pulled. Oh, actually, no, this
is an old one, I’m sorry. This is from April. His last A1C was 6.1, which
was about four months ago. So pretty soon after this glucose curve. But you can see what happens when you remove the carbohydrates. So this guy had consistent ketones and a flat curve now. And I wanna stay on this for just a minute because I really feel that the ability for us to give CGMs, the continuous glucose monitors, to our patients with type 2 diabetes is going to be a game changer. Because when they can
actually look at this and see this happening, and go, “Oh, my goodness, “I can’t eat this,” you know, they’re gonna
wind up pleasing ourselves. And what we’re gonna get there is another facet of
support for these people, which is really exciting. Because when I throw a
CGM monitor on someone, I am just super excited as
to watch what’s gonna happen. Because I’ll tell you,
even in the patients who we’ve been working with
for a long period of time, once they have access to
that data in their hand, like the real time, and
they can see the curves, not just these simple points, I mean, it makes all the
difference in the world. Okay, so, many of you may have heard the study that came out recently, right? And for people in this room, I bet a lot of you had the
same response to it as me. Duh. Okay, so, this was picked
up by the media though and flashed everywhere. Oh, my goodness, even people who are well, who don’t have insulin resistance are having these daytime surges when we put CGMs on them, right? So this is gonna, again, really help push carbohydrate restriction further. Because what we’re gonna be seeing, which we have not been
talking about in past, is this is a problem with everyone. So, since I have a CGM and
I wear it all the time, I decided to do a little experiment. This is me deciding to sit down and have watermelon with
my kids in the morning. This is a healthy carb, right? Now, I have no diabetes, I
have no insulin resistance. My triglyceride:HDL
ratio is well below one. And my goal, the blue bar there,
is set to 70 to 100, okay? This spike, it was 170. 170. And I get asked a lot, like, “Well, how much were you actually eating?” I’m like, “Like not much. “Like, I sat down and had a bowl “of watermelon with my kids.” This is me. This is a problem. So this was just last week. I was so bothered by this that my new plan is I’m gonna come home and I’m gonna slap in a CGM on
each of my kids for 10 days. (audience laughs) My kids are just gonna
roll their eyes and go, “Oh, my God, my crazy mother again.” But I’m like, I wanna
know what’s happening. I mean, we have low-carb kids,
but I don’t restrict fruit. And now I’m like, man, do we
need to restrict some fruit? I don’t know, but this is
really bothersome to me. Okay, so, how about this? When we talk in the general public, away from this type of audience, one of the questions we always have is there’s not enough evidence, right? That, you know, we have
to go with the guidelines because the guidelines are
where all the evidence is. They’re evidence-based. So if you wanna come
up with this new idea, you just have to have much evidence to support us changing our ways. And quite frankly right
now, it’s not there. Anybody heard that before? Anybody had that argument? There’s not enough evidence. Okay, so let’s take a look at this. So looking at low carbohydrate
intervention as a treatment for type 2 diabetes. There are 20 randomized control trials, five meta-analysis, and
10 other published trials, all supporting carbohydrate restriction for diabetes treatment. Now, we all know that as a general rule, low carbohydrate intervention
is not recommended by the American Diabetes
Association guidelines. I mean, the last iteration they said, “Well, maybe for three months or so.” Like, maybe we could do it short-term, but you can’t do it for long-term. So what do they recommend? The eating patterns that they recommend are the DASH diet, Mediterranean
diet, and plant-based. So let’s see how low carbohydrate compares to all the evidence of the recommended evidence-based eating patterns. There it is. So, here, the white blue, is randomized control
trials plus meta-analysis. And then when we add in
other clinical trials, here’s how low-carb stacks up. And I would love, take a picture
of this. (audience laughs) And show it to anybody who tells you that there’s no evidence. And the thing is, I’m gonna be working on more and more graphs of this because the other thing was, “Well, the studies weren’t long enough.” Actually, overall, taken together, the studies on low carbohydrate had more people and were longer. And there’s just more of them in general. So it’s just not true. Okay, so now I wanna go back
to talking about evidence and spend some time talking
about our ongoing study at Indiana University Health. So this was initially two year, and we have increased the time
of the study to five years. Non-randomized prospective
controlled study. And I’m gonna come back in a little while to talk a bit more about study design. But we recruited 465 patients. And in our active intervention
arm, there are 378 of them, of which 262 had type 2 diabetes, and the remaining 116 have prediabetes. The usual care arm were patient recruited from the same community,
same, all at IU Health, but they were being given
nutrition instruction by the diabetes educators. You can see here that the mean (mumbles) of our patients was quite high. So these were not cherry-picked patients. And the other really important
thing about this trial at baseline is we were
recruiting all commerce, meaning, we weren’t putting
restrictions on insulin or length of time with the diabetes. In fact, if you see here the mean years that people had type 2
diabetes was really long. These were really sick people. If you compare that mean
years with type 2 diabetes to other studies, including
other reversal studies, you see that our length
of years with diabetes is significantly longer. Once again, not cherry-picked people. So our primary outcome was body weight, metabolic syndrome criteria,
and type 2 diabetes status. Secondary outcomes, we
looked at a number of things, and we also still have banked samples that we’re gonna be doing
a lot of fun stuff with. So what happened at a year? A1C reduced, and it reduced significantly. We went from a starting average A1C of 7.5 to 6.2 at a year. And a couple really important
points on this graph, which is just how fast
this improvement occurs. And the reason that that is so important is because it motivates patients. We have to remember at every single graph that any of us put up, there’s all these patients
that make up those graphs. And they are all important. And for each of those
people as an individual, they wanna see results that are
actually meaningful quickly. They wanna be given
instruction that works, and they wanna be able to see
the improvement very quickly, including, you know, one of my favorite things to
hear from a patient is when they start at the clinic and then a month later they say, “Oh, my gosh, I can’t believe it. “The very next time I
went to the pharmacy, “my cost was reduced already.” So this quick drop is important for each of those individual data points. But, again, we don’t see the
drop and then a rebound, right? We see a continual decline
out to a full year. And if we look at the gray bar above, this is the usual care. So remember what type 2
diabetes has been called by many organizations, including the American
Diabetes Association. It is a chronic and progressive disease. And I agree with that statement if we’re using standard of care. That’s what happens over time. We slowly get worse. So, again, it’s not just not
the A1C that was improving, it was the medications that
were also being reduced. 57% of prescriptions for
diabetes were discontinued. And if we look here, like sulfonylureas, which I think for any of you
who work with these patients, like this is the first one
to get people off of, right? I mean, it’s taxing the pancreas. They’re just, in my opinion, nightmares. But, unfortunately, in general practices we turn to them all too
often because they’re cheap. We were to able to
remove all sulfonylureas. And for insulin, 94% of the patients who
began the trial on insulin had the insulin decreased
or totally eliminated. And if you can see, the total elimination was really significant. But, you know, this goes across to other classes of medications, too. And the one you see not
that much of a change in it, the N there is metformin because metformin has indications outside of type 2 diabetes. And no matter how wonderful we
make their glycemic control, these are people who had type 2 diabetes, and they are at risk. And so I really feel like metformin is one that I have a risk-benefit discussion with the patients about. And I let them choose. I say, “We have reasons to believe “that this may be
helpful in the long run.” And some patients are like,
“Great, I can stay on it. “it’s not bothering me, it’s super cheap.” And other patients are
absolutely, “I want off of it.” So I let the patients choose. And then you see that we
did add some GLP-1 agonist. And the reason, clinically, that I do this is it’s a great bridge to
get people off of insulin. Because we know those
high levels of insulin are gonna impede their weight loss, and it also cause risks
of hypoglycemic events. We really wanna get people off the insulin as quickly as possible. And that’s kinda my favorite way to do it, is we switch them just
to another injectable, and then we can get them
off of those over time, too. But, really, we’ve reduced
their hypoglycemic risk quickly. All right, so not only do
patients get off of them, but, of course, this
doesn’t surprise anybody, with that much reduction in medications, we’re saving money as well. So medication reduction at a year was 46%. Again, remember, this is
just medication reduction, this doesn’t yet take into account all the other cost associated that make up the over
$300 billion a year now, such as reduced productivity,
days off of work, hospital costs, all those things. And do patients stay engaged, right? Because in addition to the argument that there’s no evidence for it, the second biggest argument
I get is people can’t do it. And so 83% at a year. Like, when we look at prescriptions that we write for our patients, there’s no way at a year 83% of them are still taking those prescriptions. So this people adhered
to better than a pill. Don’t tell me that’s not sustainable. Okay, so, our goal was not weight loss. Straight up, we told patients that’s not. If you do it, great,
that’s a wonderful benefit. But despite the fact that our
focus was not on weight loss, people lost a lot of weight. And as you see, we do not
have the six-month uptick that you see in most
weight loss trials, right? People lose weight for
three to six months, and then we go up again. But, here, we are all
the way out to a year and we aren’t continuing
to lose on average, but we’re flattening out
at a much lower rate. So in addition to publishing our one year diabetes-related
results, we also published a one year cardiovascular
risk paper as well. And this is my favorite graph probably from the whole trial so far. And it looks at all these
markers of cardiovascular risk, comparing our intervention to the control patients which are in gray. So we can see here, I mean, really, our intervention
patients rocked it (chuckles) as far as cardiovascular risk goes. What is the one, one exception where it got a little bit worse? If you notice here, LDL-C
did go up a little bit. It went up by about 10%, which many would consider very significant as far as increasing cardiovascular risk until you take a look at
their apo B and LDL-P. And Ron already talked
very nicely about this. And that in these patients,
they can have a normal LDL-C or a slightly elevated LDL-C, yet their LDL-P and apo B can be sky-high. But that’s not what we were seeing here. We were seeing the
increase only in the LDL-C. Their apo B and their LDL-P,
actually, went down slightly. Wasn’t statistically significant. But the point is, we weren’t
increasing this risk. So, really, we had all these improvements. Yes, you can reverse your type 2 diabetes and improve all these other risk factors, and not be making other things worse. So that’s really exciting. And so how about their 10-year risk score? We took a look at that, too. And, again, what happened
with our intervention patients compared to usual care? It’s more of the same thing. And so liver function, you know, another big problem and another big cost
associated with type 2 diabetes is all the problems people get because of fatty liver disease. And so here we see a
significant improvement in liver function as well. So kinda to take a look,
big picture overall, really, 60% of the patients in the intervention arm at a year reversed their type 2 diabetes, which means they had a glycemic control under the diabetes threshold and we’re off all
diabetes-related medications with the possible exception of metformin for the reasons I discussed before. And, again, medication reduced
while the improvement in A1C, weight loss, and improvement
in cardiovascular risk factors. I mean, what’s not to
love from this, right? So here’s a question. And I know I’m running short on time, but I wanna make sure that I cover this. Did they actually eat
what we told them to? Because in any nutrition intervention, we’re giving them instructions
that’s all well and good. But the question is, did they do it? And how have we tracked this? In all prior nutrition interventions, we rely on food records. What do we all know about food records? They stink. They stink, they’re not good,
they’re fraught with error. And it really makes, I believe this is actually
one of the reasons that nutrition science is so contentious. Because no matter what someone publishes, someone could come up and say, “Oh, they didn’t really do it. “Your food record method stunk.” We can’t say things
definitively until now. So did they actually eat a ketogenic diet? They did. Because we had a biomarker to follow. We followed beta-hydroxybutyrate. So our study is really unusual. We didn’t use food journals. Our patients didn’t have to write down everything
that they were eating. We used this as a marker. And, yes, they were doing it. So I think not only is this, in and of itself, important
for this particular study, it also brings us to a
question of study design. Because we did not do a
randomized control trial. And, you know, that’s often a criticism. Oh, my goodness, everything needs to be randomized control trial. That’s the gold standard. And it’s the gold standard, and makes really good
sense for drug studies. But the problem with nutrition studies is that the people have
to be invested in it, especially if you’re planning
on doing a long-term study. So if you all of a sudden
take a group of people who let’s just say have been following a low-carb, high-fat diet,
and you tell them all, “Well, we’re gonna put you in a study, “and half of you are gonna have
to be vegan for five years.” Or, you know, or anything. That really significant
change to their diet. And they’re not invested in it. They don’t believe in that. It’s not going to work because you’re not gonna
get the compliance. So you want people who are invested in it. And this goes back to my
argument of patient choice. Patients need to choose the intervention that they want to follow. And I think that that’s really important. So the idea of randomizing
that is a little tricky. So how do you support
sustainable behavior change? And support from many angles is critical. So the patients in our study, actually, had a ton of different
ways of getting support. They tracked their biomarkers
and we’re able to follow them. They all had a health coach, and they all had a
physician following them. In the trial, it was me. And they were able to coach
based on those biomarkers to help people and really
make it personalized for them. They had a patient community, and they also had a ton of resources that they had access to. So the other thing you need to really have sustainable changes, and you have to give
people advice that works. And, whoops. And, again, we know from
the graph how quickly not only this works and works
for a long period of time, but, you know, people are excited to eat this kind of food, too. The idea that people can’t stick with it because it’s boring
and doesn’t tastes good is just not the case. Oops, sorry. So I just wanna end with this. Reminding everyone that there are three clinically proven ways to
reverse type 2 diabetes. We need to be talking about this. We need to let patients know that they once again have control. Because telling a patient who is diagnosed with type 2 diabetes that they are stuck, that it
is progressive and reversible, and there’s nothing to do about it takes the control away from that patient. And when you can tell them, “There’s something you can do about it, “you can reverse out of this, “and here are the ways.” And the most important thing is then to give patients the choice. And I clearly believe that most patients, when given the choice, are going to choose a
carbohydrate-restricted plan, but if they choose one of
the other ones, that’s fine. It’s not gonna be ever a
one-size-fits-all for each patient. We have to respect that always. But the problem is that patients are not given the choice right now, right? They’re not given the choice because this is not being discussed. So I’ll go back to what I
said at the very beginning. It’s everybody’s
responsibility in this room to make sure that all people
with type 2 diabetes know that it does not have to be a chronic and progressive disease. Give them the power back. Give them the choice. Thank you. (audience applauds)