Introducing Dr. Eric Kossoff
Dr. Eric Westman: From time to time, I get to ask people who I think are the movers and shakers in the low-carb, keto, or medical world onto my channel. Today, it’s my great pleasure to have Dr. Eric Kossoff. How are you doing, Eric?
Dr. Eric Kossoff: Good, good. Thanks for inviting me, Eric.
Dr. Eric Westman: My pleasure. For those who don’t know who you are, who are you? What do you do? How did you get into this area? How could I possibly be interviewing you?
Dr. Eric Kossoff: Sure, yeah. I am a pediatric neurologist. I take care of children with various neurologic diseases, although most of my time – about 95% – is focused on epilepsy. I’ve been doing it here at Johns Hopkins in Baltimore for about 20 to 25 years now. I started down in Virginia, did my training here in Baltimore, and then never left.
Early in my training, I met Dr. John Freeman and Dr. Patty Vining. They were running the Ketogenic Diet Center for Children here at Hopkins for probably about 15-20 years when I arrived. It just blew me away; I thought it was incredibly interesting. I had really never heard of it before in my pediatric training or in medical school. So, I followed them around, did a little bit of research as a resident and epilepsy fellow, and then decided that I wanted to make it my career. Around the time I became faculty, Dr. Freeman retired, so I took over the pediatric program, which has continued to grow.
I’m very involved with children with epilepsy. I’m the associate director of our Adult Epilepsy Diet Center, which is very exciting. Dr. McKenzie Svena runs that, but I help and assist. It’s exciting to see that, and occasionally, we collaborate with others like you and others working in fields other than epilepsy who ask for our advice and help. It’s an interesting career, and I love the ketogenic diet. It’s a lot of work but also a lot of rewarding fun.
Dr. Eric Westman: What year was it when you came through training and went to Johns Hopkins to learn about this?
Dr. Eric Kossoff: I started my training in 1998, and it was probably about 2000 – 2001 that I finished my fellowship and came onto the faculty – almost 25 years ago.
Dr. Kossoff’s Patient Profiles
Dr. Eric Westman: Who is your typical patient?
Dr. Eric Kossoff: I see a lot of children from around the mid-Atlantic area, and sometimes from other countries and parts of the U.S., with typically refractory epilepsy, meaning they have seizures that haven’t responded to two or three, or sometimes many more, medications. Many times, they’ve been evaluated to see if they could have epilepsy surgery, which in many cases can be a cure, but it’s not always appropriate for every child, and sometimes the families don’t desire that.
So, they get referred to me to see if there are other options. Families have heard about the ketogenic diet quite a bit now; the last 20 years have been exciting just for public awareness through a lot of your work and others. Families are familiar with it, and some have even tried sort of dabbling in it on their own. So, we meet with them. My dietitian meets with them. We explain how we do it: for a medical ketogenic diet for children, we typically admit them and gradually start them on it, explaining what follow-up will look like.
We tell families that, for most of them, it’s a combination of medications and the diet – about 80% of our kids are on both. But if they do well, we can lower or reduce medication, which is ideal if we can pull that off. A lot of them see better cognitive improvement as well. We go through all this with the family to ensure they understand the ketogenic diet program at Hopkins. Then, shortly after, we get them started.
We see babies, toddlers, and definitely a lot of adolescents. We tend to use more of a modified Atkins diet for most teenagers, but for the most part, it’s a discussion in the clinic about what it’s going to be like and making sure the family has good goals and expectations.
Positive Outcomes
Dr. Eric Westman: When you counsel someone about the diet change for epilepsy, what percent of children have total resolution even without meds? Does that ever happen?
Dr. Eric Kossoff: Oh, for sure! I mean, that’s what makes it so exciting. If you look at the statistics before we even talk about the diet, medications certainly work, but if you try two or three or four medicines, the chance of seizure freedom starts to drop pretty quickly. The first drug you pick could have a 50% seizure-free rate, but by drug number three or four, it gets close to about 5%. Not zero, which is why we still keep trying new drugs, but there’s probably about a 1 in 20 chance that they’re going to become seizure-free with a new medication.
Dr. Eric Westman: That’s not great.
Dr. Eric Kossoff: No, no, it’s not, for sure. And that’s why families look for other options and say, “What else do you have?” The diet, no matter the age, gender, or where you are in the world, has about a 15% chance of seizure freedom. So, three times as high with a dietary therapy, and that’s why we often talk to families about it. In fact, there’s a lot of research now looking at doing it sooner, not waiting until so many medications have been tried, because there’s no reason it wouldn’t be successful if used first or second.
Dr. Eric Westman: So, that’s 15% will have total remission of epilepsy by the ketogenic diet?
Dr. Eric Kossoff: Exactly. That’s all.
Dr. Eric Westman: It’s amazing. I wonder, would it be even higher if you didn’t have the referral selection bias of probably getting the worst cases now?
Dr. Eric Kossoff: It’s interesting, Eric. I think that was probably true maybe 10-15 years ago. We were the therapy of last resort. Families were sent to us or found out about us because they were desperate and had nothing else being offered by their center. And even those kids – you look at the data, it’s like 15%, which is remarkable.
What I think has started to change is, as a community of neurologists, we’re starting to realize there are specific indications that are more responsive to diet therapy than others. Epilepsy is complicated. There are multiple kinds of epilepsy, multiple reasons why children have seizures. The genetics of epilepsy is exploding year by year, with new genes and new syndromes emerging. That’s sort of the big push of our community is to find those indications where it may not even be 15%.
For instance, there’s a condition called GLUT1 deficiency – about 80% seizure-free. I mean, why would you even think of trying anything else? And so we’re really trying to establish those indications to get these children on the diet sooner.
Dr. Eric Westman: I’ll never forget that paper where they were trying to find the mechanism of why there was total remission. I remember one case, overnight all the seizures stopped – like multiple seizures per day. With GLUT1, the receptor that allows glucose to go into the brain was deficient, right? GLUT1 deficiency was an explanation for why some kids would respond that way. Because when you shift them over to ketone fuel for the brain, you can bypass this glucose receptor that no longer works, or never did work.
GLUT1 Deficiency
Dr. Eric Westman: Let’s say 100 people with epilepsy, how many have GLUT1 deficiency?
Dr. Eric Kossoff: It’s relatively rare. That is one of the many indications; there are probably about 15 to 20 that we know about, but GLUT1 is the classic one because it’s the treatment of choice for GLUT1 deficiency. I’d say maybe in an average year, two or three of our population that we put on the diet have GLUT1 out of 50 to 60 children. But obviously, we’re a referral center, so if they’re diagnosed with GLUT1, they get on the diet as soon as possible. But it’s relatively rare in the scheme of epilepsy, I think.
Dr. Eric Westman: How does someone get diagnosed with that? Do they need a brain scan, or is it a genetic test?
Dr. Eric Kossoff: Yes, that’s also changed really just in the last 10 years. There’s now a genetic test – it’s called the SLC2A1 gene, and that’s in a lot of epilepsy panels. We send these panels off, looking at 300 to 400 epilepsy genes. It used to be a lumbar puncture, which obviously is more cumbersome. Now, with genetic testing, we sometimes find it by coincidence. The neurologist might send the panel looking to see if anything sticks, like throwing darts at a dartboard, and they get a result back showing the child has GLUT1 deficiency. Then they’re like, “Oh my gosh, we need to get them to a keto center immediately.” It’s wonderful having that genetic test available because, years ago, some of these children likely were never diagnosed.
Dr. Eric Westman: Let’s say one of my patients was watching, and they have a family member with a child with epilepsy. How likely is it that their neurologist will think about it or even send the genetic testing? What’s the penetration of the knowledge of this in the field?
Dr. Eric Kossoff: It’s starting to get better. I think as the genetic test has become more widely available, people are learning about all of the disorders, not just GLUT1 deficiency. We have more epilepsy neurogeneticists now who specialize in this. The GLUT1 Deficiency Foundation, which is a wonderful foundation, has been trying to get the word out through advocacy, newsletters, websites – to families not only who are diagnosed but also those who may be undiagnosed, to at least consider getting checked.
There are other constellations of physical features these children may have, such as movement disorders, early-onset absence seizures, or cognitive difficulties. Sometimes the clues add up, pointing to GLUT1 deficiency.
Dr. Eric Westman: It sounds like a family member may need to push a little to get a neurologist to do this kind of testing?
Dr. Eric Kossoff: Sometimes. It’s becoming more standard of care. It’s amazing how these panels are done. It can be a cheek swab – not even a blood test nowadays. There are many neurologists who just do it in the office and send it off to the companies for analysis. It’s getting better.
Dr. Eric Westman: I know you’ve always been very careful not to talk down to other doctors or other practices. I’ve been around longer and I tend to be a little more open. After 25 years in practice, I’ll say that some doctors just don’t know, so you might need to push toward asking about a keto diet.
There are so many similarities here to the treatment of diabetes, for example. In our population, doctors are often told that drugs are the way to go and that diet doesn’t do anything for diabetes – even though it’s a condition of elevated blood glucose. You might put two and two together. But it’s only when someone’s on one or two medicines, or insulin, and nothing works, that they might wake up to the idea that diet could help.
In our field, the Virta Health study has the best data so far. In five years and in one year, 70% of people on insulin came off with type 2 diabetes, which is pretty much an issue of helping some to stay on now that I see it.
The Prognosis of a Keto Diet For Children With Epilepsy
Dr. Eric Westman: If you put people on the diet and you know they’re following it, you must get better results than sending someone home and life happens. Assuming someone is following the diet, if you have 100 children with epilepsy, what’s the percentage today of children who get better, can get off meds – how do you explain the prognosis?
Dr. Eric Kossoff: It’s a great question. For the families, that’s certainly one of the things they want to know. Every child is different, and every situation can be different. And, again, knowing what the cause of the epilepsy is certainly impacts the likelihood of benefit. But the data is pretty clear that it’s about 50% that will respond. So, about 50% of children – I tell families we do our admissions in groups, usually three or four in a group – and I tell the families sitting around the table, if you play the odds, two of you will respond, two of your children probably won’t respond.
One nice thing about the diet, besides how well it works, is that usually you know pretty quickly. If you’re having seizures frequently enough, within about four to eight weeks, you have a sense of if the diet’s going to be beneficial. Obviously, if it doesn’t look like it is, we’re not going to keep children on it long term. But it’s one of the quicker-responding epilepsy treatments.
Again, about 50% respond, and probably about 30% of children will have a greater than 90% response. So, instead of 100 (seizures) a day, maybe like one or two a week – not seizure-free but dramatically better. And then, as I mentioned, about 15% are seizure-free. And again, many of them, we can reduce their medication; about 20%, we can get them off all medications.
One nice thing about pediatric epilepsy is that even in some of the toughest cases, there can be some resolution, and we think the diet may accelerate that resolution. But medications can do that too; sometimes, you know, children will “outgrow” their seizures, as we like to call it. We think that the diet may help us accelerate that so that they’re getting better quicker. It still remains to be proven scientifically, but we’ve also seen it.
The Charlie Foundation
Dr. Eric Westman: You know, one example of that was Charlie Abrahams with the The Charlie Foundation, which is an example of philanthropy. Your family member is affected by it, and you happen to be the son of a Hollywood producer – you have access to things. Would you explain the Charlie Foundation, please?
Dr. Eric Kossoff: Oh my gosh, yeah. The Charlie Foundation is absolutely responsible for the renaissance in our field. The ketogenic diet for epilepsy has been around since 1921; we actually just had our 100-year anniversary. It was exciting for all of us in the field. Lots of memorabilia, lots of history lectures. For, boy, I’d say 70 years, it was relegated to a treatment of last resort, children only, labeling it as being really difficult to adhere to, and a difficult, unpalatable diet. You’d see that written in the literature.
What changed was really the Charlie Foundation. In 1994, Charlie was, I think, about 18-19 months old. He came from California with his dad and mom to Johns Hopkins and saw my predecessor, Dr. John Freeman who put him on the ketogenic diet with his dietitian, Millie Kelly. Very quickly, his seizures improved, and I think by about two to three weeks, they stopped. He was able to come off medications and he’s now an adult, he’s a teacher, doing great – he’s a great example of what the diet can do.
Dr. Eric Westman: I understand he’s no longer on the diet?
Dr. Eric Kossoff: No. That’s part of getting into the essence of it. Part of what’s so interesting about it is there may be some children, and maybe even some adults, who don’t need to be on the diet long term. There may be something about this acute metabolic change that happens, that maybe even only a month or two can sort of reset your brain in order to stop the seizures, and then you can come off the diet.
There’s some research looking into that, but generally, we keep children on for about two to three years, and then we gradually see what happens as the diet is weaned. Many of them remain seizure-free, like Charlie, and he’s off the diet now.
It’s just a great example of what a parent organization can do. They not only got the word out; they made movies – a movie with Meryl Streep that was well known. But they also funded research, did training programs for teaching children’s hospitals how to do the diet. They’re still incredibly active today, and their website is wonderful. It has lots of good information, recipes, and links to other ketogenic diet centers. It’s absolutely responsible for the renaissance of our neurologic ketogenic diet world.
The Evolution of the Keto Diet For Treating Epilepsy
Dr. Eric Westman: Natt Baszucki happened to be a child of a wealthy family, and now the Baszucki Group and family are putting an effort into studying a different brain disorder – the psychiatric problems, major psychiatric problems – and hopefully that’s going to bear similar fruit, not just advocacy, but the science that needs to be done.
It surprised me when you said that the improvement happens so fast, because we don’t see some keto adaptation changes until about six months – full muscle and fat adaptation. In the movie Cereal Killers 2, they keto-adapt a couple, and it took them six months to fully fat-adapt. This is the muscle, not the brain. Then, they went into a rowboat and rowed from San Francisco to Hawaii and broke the world record by 15 days. It’s a heroic effort, but we’re learning that every organ is under a different silo in our medical world.
The brain is an organ, but an internal medicine doctor has very little training in the brain. I know more about muscles and the heart. That’s very interesting about how often or how fast it would work. I wonder, how have you evolved the thinking in the implementation of the program?
Dr. Eric Kossoff: One of the things I often tell my trainees and medical students is the only thing that doesn’t change is change. What I learned, what Dr. Freeman was doing even before me, was really questioning a lot of traditions and anecdotes, again going back to the 1920s, about how to do the diet that had some scientific basis but when you really push hard, didn’t clearly show evidence for why that was necessary to help a child’s seizures.
It’s amazing. You look back in the 1920s, they would fluid-restrict children, they would calorie-restrict children – probably to artificially raise their ketones or what it looked like in their urine – but maybe it really didn’t make a difference for their epilepsy. We started to see in the 1990s some articles, a lot of them by dietitians, looking at aspects of how we started the diet. Do we need to admit them? Do we need to do a fast? Do we need to be so strict with our ketogenic ratio? Maybe we can be a little bit less restrictive.
It’s absolutely wonderful for the field – not only to help maybe reduce the side effects by being on such a super high-fat diet (again, our children are typically like 92% calories made up of fat with the classic ketogenic diet), but also, can you maybe do it in different ways for different cultures, different religions to make it more accessible? That’s one of the things that was a problem in our field. People would say, “Well, you can’t do it if you’re a vegetarian, you can’t do it if you’re vegan, you can’t do it if you can’t be in the hospital. You have to be in the hospital.” I’d say, well, wait a second, maybe there are alternative ways, or as we call them now, alternative diets, that can be adapted to each individual child and get them on the diet where they would have been discouraged before. It’s definitely expanded our population and helped children that we couldn’t have helped before.
The modified Atkins diet was created by us at Hopkins as a change versus the classic Atkins diet but keeping the carbohydrates at 20 indefinitely, not going up periodically and really encouraging the fat. That was a big aspect of our counseling to the families. It’s not that fat is okay, protein is okay; it’s really that more fat is critical, and protein you have to keep an eye on because that’s anti-ketogenic. It’s done by families not coming into the hospital; it’s done a lot again for teenagers and adults nowadays, and we teach them how to read the food labels, not how to weigh and measure on a gram scale.
It’s a different way of getting to the same goal. There’s a low-glycemic index treatment out of Boston. There’s a diet called the MCT diet (medium-chain triglyceride), which actually goes back to the 1970s, looking at using more MCT oil versus long-chain triglycerides that maybe allows more carbohydrates and seems to work equally well. People are really creative out there in coming up with various ways to question what we’ve been doing for years, and it’s just wonderful for our families.
Dr. Eric Westman: Is a modified Atkins is beneficial? Does it work as well, or almost as well?
Dr. Eric Kossoff: If you look at studies, there have been some comparing the modified Atkins diet to the classic ketogenic diet. There have been some studies out of India looking at classic ketogenic to modified Atkins to low glycemic. They’re all generally similar. It seems like with maybe a little less side effects with modified Atkins, which makes a bit more sense. But for every child it can be different. We’ve had children, and we tell families, you can switch from one to the other if it’s clinically deemed necessary.
We’ve had children start with a modified Atkins diet; they’re 95% better. We’ll see them and say, maybe let’s switch you to a classic ketogenic diet and see what happens. There are some of those children who became seizure-free when we did that. Most commonly, it’s the other way around, Eric, where they’re on a strict classic ketogenic diet, they’re having compliance issues, some of them are reaching puberty, and we’re like, “Alright, maybe let’s back off a little bit just to keep you on a dietary therapy,” and we’ll go to a modified Atkins diet and try to see if that works. Usually, it does.
Ketone Measurement Technology
Dr. Eric Westman: That’s just so interesting. How has your field addressed the new technology of ketone measurement? I get people coming to me with reams of ketone data now, glucose and ketone, the blood, the breath, the urine, and it doesn’t always correlate with the disease modification that we’re trying to modify. Have you incorporated the ketone measurements, and does that matter?
Dr. Eric Kossoff: Your clinic sounds like my clinic, Eric. Our families come with, like, three-ring binders of ketone measurements. It’s amazing what we can do now and the meters are getting smaller and easier. There are breath meters now that are really accurate. It depends on the situation. Obviously, for really young children, we will lean more towards urine ketones. It’s a little less invasive. You can check blood ketones in the clinic. There are some families who really find that for their child, ketones really matter.
We keep track on a calendar and try to correlate that with seizure control. In those children, I may recommend getting a blood meter so we can be more precise. The GLUT-1 families, which we were talking about earlier, it’s in fact recommended that they do blood ketones because it seems to be more correlated with their outcomes.
There’s a large, large – I’d say majority – of these families that we treat with epilepsy that it’s sort of a marker that they’ve made the metabolic change. Most of our children, they’re in large ketosis, so large in the urine, maybe 3 to 4 mmol/L in their blood, and it may vary day by day, but usually not a lot. Not a lot where I would say you need to be checking your blood ketones on an everyday basis. It’s not really going to change what we’re doing. Periodically, sure, it’s definitely more accurate. Some of our families will, in fact, do it more regularly, but most of our families are still checking urine ketones, at least at Hopkins.
Dr. Eric Westman: Do you require that people do it or is it optional?
Dr. Eric Kossoff: We tell families that the first couple of months we really do want to get some data, so on their calendars they should check it, write down what their ketones are, keep track of their weight, keep track of their seizures. As time goes on, we can loosen up on that as well and tell them, “Okay, maybe once a week or a few times a month, check your urine ketones.” If there’s a problem and seizures increase for some reason, we don’t know why, check your child’s ketones. Make sure they didn’t get a hold of something that knocked them out of ketosis. For most children, yeah, we can start backing off over time.
Again, whatever we can do to keep them on the diet longer and make it less onerous for the family, the easier – less labs is where our field is starting to head.
The Correlation Between Ketones and Seizure Reduction
Dr. Eric Westman: There’s always that balance of what you ask people to do, whether it tips over the boat. Has there been a formal investigation of the ketone level and extent of seizure reduction? Does the level of ketones matter?
Dr. Eric Kossoff: You would be amazed that after over a hundred years, there’s been very little studies that have looked at that. Part of it is how hard it can be. It would be a very difficult study to do because seizures fluctuate and ketones fluctuate throughout the day. Then, you can get into, what are the families witnessing versus what actually could be happening in the brain, and should we be monitoring EEGs to do a study like that, electrograms, and see if those change. It would certainly be a complicated study.
It’s been done. There have been a few studies that have shown a slight correlation, some that have shown that if the ketones are stable, that’s better than if they fluctuate. Hard to replicate that in other studies. It seems to be that if you are at a ketogenic level – so again, 3 -4 mmol/L on blood ketones, or large ketones – you’re more likely to at least show compliance, and then that could be a surrogate for doing better on the diet.
It’s really hard to say what number is correlated with control. In many ways, we tell families it’s like drug levels. If you’re on Dilantin or you’re on Keppra, for some patients, they don’t need to have a high level; for some parents and for some children, they do. It’s the same thing, I think, with ketones.
Dr. Eric Westman: So it sounds like as long as you’re in ketosis, that’s the main thing?
Dr. Eric Kossoff: I think so, I think so. Although, if you look, the low glycemic index treatment out of Boston, if you try to check urine, they don’t seem to be in ketosis; it’s very scant in blood, which suggests maybe there’s another mechanism at work beyond ketones, which is super exciting too.
Dr. Eric Westman: That confounder is maybe it’s just the absence of the carbs?
Dr. Eric Kossoff: Could be, yeah, could be. That is their theory, that maybe keeping your blood glucose stable for some epilepsies may be more important than being in ketosis. Fascinating work.
Babies in Ketosis
Dr. Eric Westman: I have been around the country talking to people at meetings, you have too, and people come in and tell you things. I had a few people, mothers, change their diets, and they send their breast milk off to labs. There’s, I think, one in UCLA that’ll do the breast milk, and they come in and they tell me that the breast milk changes if they are in ketosis. I traveled to Jakarta, Indonesia, and the pediatrician next to me says the children develop even better if they’re in ketosis.
Mary Newport, who’s a neonatologist, retired now, makes the claim that babies are born in ketosis, and that we basically turn off ketosis by giving carbs. A mother who checked the ketones in the child, saw that the child was in ketosis when she was breastfeeding on a keto diet. The gentleman at the meeting who was a speaker walked by at the same time and said, “Oh, that makes sense to me. There are cultures around the world that ferment lactose that doesn’t get absorbed as a sugar.”
So, what if we have it all wrong and we should always be in ketosis, and the babies should be in ketosis, and we give them the seizure disorder by feeding them carbohydrates? I know this seems ridiculous, turning it upside down.
Dr. Eric Kossoff: It’s an interesting question, in two parts. You’re right; I think there was a recent case report of a mother who was on a ketogenic diet and they analyzed her milk. The child had epilepsy, and she was able to keep the child in ketosis by breastfeeding. That’s the first time I’d ever heard of that.
In general, even though you can do it, most of the times that we’ve heard about this from pregnant women or women who are breastfeeding and they’re in ketosis, it’s not enough of the ketones that cross over, and that the breast milk itself still is high in carbohydrates. It may have more fat, but it’s not like the 90% fat ketogenic formulas that exist. I think they’d have to be on a very strict diet. I’ll have to read that case report again. They’d have to be on such a strict diet.
Dr. Eric Wesmtan: The thing that never occurred to me, and it was just because this evolutionary biologist was the keynote speaker at the meeting, and he gets up with a rock and goes, “This is what changed human history,” because it allowed us to cut – that’s how the theory goes. But he’s walking by, and he says, “No, the lactose in the milk’s not an issue because it will be fermented and the baby will be in ketosis.” The mother was shocked because the only couple of times she checked, he was in ketosis.
Then of course, the question is how long was the mother keto-adapted into the breast milk? I learned a while back that the “normal” range and what “normal” breast milk is, is entirely dependent on the environmental situation of normality, which for most papers and textbooks will be people eating carbohydrates. The normal range for breast milk – that’s based on women who eat carbs. That’s fascinating. I didn’t know, and yet this had an effect on the child. Was the child in ketosis then?
Dr. Eric Kossoff: I don’t think so, but there were at least some trace levels, maybe. It wasn’t like nutritional ketosis, but I guess enough ketones had crossed over into the breast milk that the child, without any other obvious reason, had some seizure benefit. It’s very interesting.
I think it’s an interesting question. People are certainly looking at using the diet for babies. People are looking at using the diet as a preventative therapy. Right now, we do it more reactionary – the child has epilepsy from various reasons, and then we’ll put them on the diet and alter the course of their epilepsy. But what if we knew (which isn’t always possible, but sometimes) that someone, not even necessarily a child, was likely to develop epilepsy, could you put them on a ketogenic diet in advance of whatever insult or genetic change is likely to happen, and then prevent them from maybe having epilepsy in the first place? Or, if they do, it’ll be a milder course of their epilepsy.
People are looking at traumatic brain injury in that way. I think the military is interested in it for preventing some of the brain injury and preventing epilepsy from traumatic injury. There is a very famous case of a child with GLUT1 deficiency who was born to a mother with GLUT1 deficiency. The mother was on a ketogenic diet, and they’re like a celebrity in the GLUT1 community. She became pregnant, stayed on the diet, and as soon as the baby was born (it’s autosomal dominant), they suspected the baby was going to have GLUT1 deficiency. They put her immediately on a ketogenic diet, checked her genetically, and she, in fact, did have GLUT1 deficiency. I think she’s now like eight or nine years old, completely normal, 100% normal – never has had a seizure, no cognitive decline. It’s a proof-in-principle with one patient of what if you knew? She has GLUT1 deficiency. It’s not like it went away.
Dr. Eric Westman: You’re assuming everyone with GLUT1 deficiency gets seizures. Do we know for sure?
Dr. Eric Kossoff: That’s true; some of them don’t. I think some of them can; the majority will have cognitive decline by a certain age. A lot of them will have movement disorder abnormalities. To be completely normal at eight or nine is pretty unheard of in the GLUT1 community, even for those who don’t have seizures.
Dr. Eric Westman: In some areas, you might screen a newborn for different hereditary things, including a seizure disorder.
Dr. Eric Kossoff: Potentially for some that have genetic causes. Getting back to GLUT1, actively working to have it on some of the newborn screening panels because we know there is a treatment that is remarkable for changing the course of their disease. Early identification seems to be really important, at least from what we’ve seen so far. And this is, again, a case of one, but it’s quite dramatic.
Carbohydrates and Epilepsy
Dr. Eric Westman: Does the history of epilepsy give any insight into this? Did the prevalence go up as carbs went up? You used a term that I hadn’t heard in a long time – the anti-keto effect of protein? As you increase the protein, you can elicit an insulin response, and that turns off ketosis. I think that’s the mechanism still.
Dr. Eric Kossoff: I think that’s probably right. But, again, we try to get them into this high ratio of fat to carbs and protein combined. People think of carbs, of course, but protein also can be anti-ketogenic. For some patients, they do fine, but for some, we do have to cut back a bit on their protein if they’re losing their ketones, and it seems to be relevant for their seizures.
It’s a common question that’s asked. At least from what we’ve seen epidemiologically, it doesn’t appear that there’s been an increase in epilepsy, which again you’d expect with the increase in carbohydrate consumption over the last 100 years. It doesn’t seem like that. In parts of the world where they maybe have a higher fat intake, it doesn’t seem like there’s less epilepsy. Again, epilepsy is mostly genetic, sometimes due to structural abnormalities, and rarely metabolic. There are some metabolic conditions that can cause epilepsy, and perhaps a ketogenic diet, if used in many children, could reduce that risk. But it’s a small proportion of the epilepsy group. Many of them, again, are genetic causes that are not due to metabolism changes.
What Is Epilepsy?
Dr. Eric Westman: Could you describe what epilepsy is? To an internist (I’m not a neurologist) or to the general public – it has a lot of different presentations. It could be grand mal or generalized seizure, the classic tonic-clonic motions, and losing consciousness. But there are also the less severe forms. Do we know what’s going on, or is it a lot of different things that could be creating a seizure?
Dr. Eric Kossoff: It’s very complicated, and for sure there’s a lot we still need to learn. We know that many genetic causes are usually due to a channel disorder, sodium channel, or GABA; some of the neurotransmitters are off. A lot of times, as I mentioned, it could be structural. Children are born with brain malformations, gray matter, or white matter not in the right places, and that can, of course, set them up for seizures. Sadly, trauma – trauma to children can predispose them, just like adults, to having seizures and epilepsy. So, there are many different potential causes. There are some cases where we just don’t know. They have seizures, and all the tests have come back normal. Those are cases where, often in children, they can grow out of them, but it’s still a mystery in many cases.
Even knowing those causes, we don’t know why seizures start at one age, go away at another age, or on any given day, at any given minute, why a seizure happens. And then, we also don’t know why it stops. Most seizures are 30 seconds. Why doesn’t it just go on indefinitely? But seizures stop after 30 to 60 seconds in most patients. There’s a lot we’re still trying to learn as an epilepsy community, just as we’re trying to learn why the diet works. We also don’t know a lot about why some of the drugs work. We know they affect certain channels or certain neurotransmitters in many of our drugs, but why they work specifically for one epilepsy and not another, and how they prevent those channels from causing a problem on any given day, is still a lot of work we have to do.
But epilepsy is common. It’s one out of 26 people. It’s incredibly common. It’s one of those disorders where, if you look at someone, you wouldn’t know that they have epilepsy. And that’s a good thing, but it also becomes one of those things where it’s a disorder in the shadows a lot because it’s not so obvious to people that somebody may have epilepsy or that it’s as common as it is. We’re trying to get the word out as an epilepsy community, not only about the diet but about epilepsy in general, getting aggressive treatment, and getting to an epilepsy center as quickly as possible.
The Metabolic Health Summit
Dr. Eric Westman: The Metabolic Health Summit meeting is a way for those involved with ketogenic diets for epilepsy, diabetes, and of course, what was called Atkins, Protein Power, and the newer scientists all get together once a year. The Metabolic Health Summit is a great place and I’ve learned so much. I remember asking if someone with epilepsy has a child and they’re on the keto diet and sensitive to carbs, if they had a little bite of a candy bar, they might have a seizure. I was blown away. In my world, you have a bite, your blood sugar goes up a little, you don’t lose weight for a day. But the consequence of having that with epilepsy might create a seizure, is that right?
Dr. Eric Kossoff: Yeah, it’s thankfully not universal. Because children are children, right? And even adults who are on the ketogenic diet for epilepsy, it’s hard. It’s hard to kind of stick to it 24/7, I’m sure you know that. Especially in the adult clinic, compliance is probably the biggest issue that they’re working with, whereas in the pediatric world, it’s more side effects and trying to improve seizure control. Not that they’re not doing that on the adult side, but it’s less of a compliance issue for young children. The parents can, for the most part, make them do it.
That was something I was certainly taught. We were even taught that the carbohydrates in liquid medicines can sometimes throw children out of ketosis. I was taught as a fellow that some sunscreens can, you can absorb carbohydrates from the carbohydrates in sunscreen, and there have been anecdotes of those children being knocked out of ketosis. There’s always a case like that. We, of course, at least to start, try to be very strict and discourage that. So, when families ask, “Well, you know, can they cheat once in a while?” I usually say, “No, try to be strict.” If they get a hold of brownies, it could certainly knock them out of ketosis. But as time goes on, we learn about each individual child, and some have done that and had no consequences, while some, obviously, have a little bit of carbs, and they’ve had a seizure.
It’s incredibly humbling to know that for each child, it could be so different, and one rule doesn’t fit all. So we usually start pretty strict, and then we may loosen up a bit after a few months in these children if we notice that maybe some of those things aren’t causing a problem.
Dr. Eric Westman: I’ve seen that some dietitians who’ve been trained only in the ketogenic diet for epilepsy will think that adults have to be that strict. What we’re talking about with adults, and even the modified Atkins, is not as strict as what you’ve been taught about the children needing a very strict version of the – was it really 95% fat? Is it?
Dr. Eric Kossoff: Yeah, 4:1, you’re getting pretty close to that. (A ratio of 4 parts fat to 1 part combined protein and carbs.) Some centers will even go higher than that.
Risks of Other Diseases
Dr. Eric Westman: I’ll never forget, it’s one of these articles I just don’t forget, that they looked at the lipid values of children on the ketogenic diet, and then the media generalized, “Well, therefore, adults on an Atkins or a keto diet are going to have this terrible lipid phenomenon.” It was entirely in the old “cholesterol is bad” paradigm. What have you learned since that time? I mean, now I’m much more comfortable talking about triglyceride and HDL as the main markers (for cardiovascular disease). Do children on ketogenic diets get a higher risk of other diseases because of the diet they follow?
Dr. Eric Kossoff: I think as time has gone on, we’ve at least seen, as we’ve followed these children and done some long-term studies, that we don’t seem to see cardiovascular risk years down the road, which is obviously reassuring. We’ve also seen that in many of these children, they do have an increase in their cholesterol, they have an increase in their LDLs, but it kind of levels out at around six months, nine months, and then it comes back down, sometimes below normal, which is pretty impressive.
Most of the time, intervention when we check the labs and if we see there’s a problem, is usually to wait. Usually, I’m like, “Let’s just check again in three months,” and it usually will come back down. If not, my dietitian may make some changes to the kinds of fats they’re on or maybe lower their ratio. Sometimes if we do see that there’s a surprising change, we may actually suggest the families get checked – the parents – and see if there is some familial hypercholesterolemia that runs in the family. We’ve identified a few parents where that’s been the case, and good for the families, too, not just the child.
We don’t know a lot about the impact of this on a child. Our adult center, as you’d imagine, is very interested in this and the ramifications for adults, so they’re looking at lipoproteins, they’re looking at the sub-particles, because, again, it’s much more complicated than just total cholesterol, LDL, and HDL, and to really see what impact that has. There have been some studies in children looking at other outcomes, like carotid distensibility, looking at heart function, and for the most part, they seem to be fine. It’s something we definitely need more information on.
And again, the other part of it, too, as I mentioned at least on the pediatric side, is it’s usually a couple of years. I’ve had some children where their lipid values were definitely abnormal, and I’m like, “Well, let’s see if we need to go beyond two or three years, okay, we can make some more aggressive interventions.” But in most of them, I’m like, “You know, two years – we’re not sure or we don’t think it’s going to have any impact on the child.” It doesn’t seem so far.
Dr. Eric Westman: That’s reassuring. There’s an ongoing study right now in the adult world we’re following very carefully. The Citizen Science Foundation funded a study looking at people who have high LDLs, self-describing a low-carb keto diet on average for four and a half years. And they are all sent to a lab or a testing center in LA to get a CT angiogram, not just a coronary calcium score. Half of these people had no coronary calcium or even no blockages on CT angiogram despite a four-and-a-half-year duration of even up to 580 LDL. This was a comparative study. It’s at least double the LDL you see in Miami with a similar matched population, and the baseline data were presented.
We’re now anxiously awaiting the one-year data. They know they’re doing keto. One of the limitations is it was self-described, but in a year we’ll see. I’m told the data is imminent, coming out any day now. I’ve heard that for a while, but again, it says relax about this LDL thing. The average HDL was 90 which is really high. So, is it possible the good cancels out the bad if you use the old paradigm? Or maybe it is just triglyceride and HDL, and maybe LDL wasn’t a big deal ever. But that puts you into this quackery, especially in a place like Duke, where the cardiology belief system has that in their mainstream treatments. There is some value to it; it’s just not the only thing one can do.
The diet change is an intervention. If a doctor just says you need a pill, you’re actually doing an intervention. Conversely, if that LDL elevation always had a problem, you would have seen an epidemic of heart disease and stroke even at two years. And so that’s reassuring that you haven’t seen that.
Dr. Eric Kossoff: This is great. This is why I love the Metabolic Health Summit meeting. I learn from you guys, you learn from us. It’s a great way to share information. I tell families, “I went to this meeting and they’re not so worried about this number or that, and this is what the data is saying, and this is important for your child with epilepsy.” So that’s great information.
Dr. Eric Westman: How can people find you, and what are you into now that you want people to know about?
Dr. Eric Kossoff: As we mentioned, there’s a lot of great information on the internet. The Charlie Foundation is a great website. Matthew’s Friends is a similar charity out of England. We have a pretty robust website at Johns Hopkins about the history of the diet in our center, which is a great way to reach out to us. And also the Adult Epilepsy Diet Center has information on all of our websites.
Watch the full video here.