Babies In Ketosis

April 2018 Update:

Here in the year 2018, it's fair to say that #Paleo is pretty much over. Nothing tells us this more succinctly than a quick trip over to Mark's Daily Apple, a website I rarely frequent these days. I'd heard rumors that Mark had "gone keto" ... given the seemingly meteoric rise of #Keto of late, is anyone really surprised?  This is the same man who developed the "more addictive than crack" P90X recovery drink and other things, 

Ever the businessman (can't fault him for that), Mark seems to have transitioned from using his blog to sell paleo/primal-related books (did he ever find a proper ghostwriter for his wife's book?) to selling over-priced avocado oil products.   The whole "primal" thing is about as downplayed as possible for someone whose signature product is The Primal Blueprint, can be.    Poor Grok.  Now it's Keto Reset time!  Can these marketeers PLEASE make up their minds?  

When someone doesn't "take" to a so-called keto diet, it's always "they probably weren't fat adapted yet".  This mythical and magical fat-adaptation has been purported to take 6 months to a year to reach full nirvana-level.  But now?  Why you can reboot your metabolism (whatever that means!) in just three weeks, and be a fat burning beast evermore!  Kewl!!

What brought this post update about?  I came across a link to a recent post by Sisson addressing What Happens “After Keto”?
"Google searches for this question have shot up in recent weeks. I’m not surprised. An unprecedented number of people went keto in January purely as a quick weight loss hack, and now they’re looking to transition off of “this weird diet.” ...
Well duh.  The best indicator of the fad-factor of a diet may well be the January bump it receives in Google trends. 

Keto is a reset (a Keto Reset (if you're going to buy this, why not use my affiliate link?), even). It’s a return to the ancestral metabolic state, the metabolic state we were born into (newborns are filthy with ketones, even on their relatively sweet mother’s milk diet). Keto happens really easily in humans compared to other animals. You go 12 hours without eating and then wake up in the morning? You’ll have detectable ketones on your breath. It’s almost like we’re made for it. "endquote
So I tweeted out a tongue in cheek response.  Basically, if you even believe in some sort of rebooting/resetting of your metabolism to its infantile state, why stop with diet and metabolism?  The more one thinks about this, the more absurd the "babies are in ketosis" argument becomes.  Read on.  Don't fall for this!!

As to Mark's contention that "it's like we're almost made for it" ... there's a table at his own link that shows that as a species outside of infancy and pregnancy, compared to rats, we're actually "not so much".   Mark knows this.  Staying in therapeutic ketosis as an adult (not this nutritional gimmick nonsense) is rather difficult for most people.  I've been reminded recently that for some, even a leaf of spinach or the taste of artificial sweeteners can be enough to banish one from planet Keton (coined by @LouSchuler on Twitter).  I won't even quibble (again from conversation on Twitter) with evidence that we aren't "born in ketosis".  

As the recent Gardner-led, partially NuSI funded study demonstrated, modern American humans -- for various reasons -- have difficulty adhering to extreme diets.  This is not evidence against the *potential* for various diets to be effective, nor is it evidence that following such diets is harmful for the long term.  But it is evidence that as a dietary prescription of sorts, neither is going to garner much success in the general population.  But one last thing before I go on this point.  I've mentioned this before, and I'll mention it again.  We have BILLIONS of humans who have thrived on very low fat diets throughout their lives, for generations.  We just don't have anything near this for low carb diets, let alone keto versions.  In his article, Sisson says the following:
Studies as long as five years (link to full text) show no negative side effects of long-term ketogenic dieting, so, provided you do it right (no salami and cream cheese keto diets), I see no reason not to continue.
Look at the full text I provided.  This was a case study on THREE young adults with GLUT-1 deficiency.  Two who were normal weight at baseline lost a little bit of weight in the form of fat.  One who was underweight at baseline gained weight:  this individual grew 1.5 cm indicating she wan't fully grown at baseline, and gained more fat than weight (gained 1.5 kg overall, 2kg fat, thus lost 0.5 kg FFM).  The authors mention that the results here are in contrast to those seen in children treated for epilepsy because of full bone mineralization at baseline.  Perhaps, at the very least, there should be extreme caution for growing children or teens to embark on a keto diet!   I see no reason to trust Mark Sisson on the long-term appropriateness of any diet.  

ORIGINAL POST DATE:  June 23, 2015

This post topic was inspired by the following article:  Ketosis - key to human babies’ big brains?  It is hosted on Tim Noakes' website and written by one of his associates in nutritional disinformation, Tamzyn Murphy Campbell, RD.  I'm going to address this misdirection and the disturbing parts of this article vis a vis Campbell in a future BabyGate Files, but for now I want to discuss the role of ketones in metabolism.  In doing so, I'd also like to explain my somewhat cryptic recent post on heating my kitchen.  (I've C&P'd that entire post to the end further down in this one, so if you don't wish to go to another page, you can just scroll down to The Kitchen Heating Analogy).     I'm going to structure this post a little differently than most and get to my point, then provide the back up information.  Let's see how this goes.  

The major source (6 of 12 numbered citations, 1 of 6 unique sources) for her article is:   Survival of the fattest: fat babies were the key to evolution of the large human brain (2003) Stephen C. Cunnane, Michael A. Crawford (I'll call this C&C)

In this paper, Cunnane & Crawford make the following points:
  • Human babies have higher body fat than other mammals and this may be as a source of ketones in early infancy.
  • Infants have slightly elevated ketones (mild ketonemia) regardless of feeding status which differs from adults who only have elevated ketones when food is restricted (no details here).
  • Glucose is the primary fuel for the human brain
  • Ketones are an alternate source of fuel when glucose is less available
  • Ketones "appear to be" an essential fuel for the midterm fetus, and may provide as much as 30% of the developing brain's needs
  • Ketones are both an energy source and carbon source for fat and cholesterol synthesis in the brain.  
Campbell relates these somewhat inaccurately in my opinion.  She fails to mention that glucose is still the primary energy source, a key point to overlook.  Also missing are some qualifiers, though to be fair, C&C don't put numbers on what they mean by "slightly elevated". Ketone levels are often barely high enough to reach Volek & Phinney's arbitrary 0.5 mmolar for "nutritional ketosis", let alone the 1.0 mmolar Campbell sets as the lower limit of her range.   Furthermore, the inclusion of fetal ketone usage adds some confusing information.  The progression of infant metabolism, from the equivalent of intravenous feeding via the umbilical cord in utero through early infancy, is not nearly as cut and dried as Campbell portrays, and is laid out in far greater detail in one of Campbell's other sources.  Furthermore, although C&C do provide some data on the energy needs of the brain during various stages of development, they do not address these stages specifically vis a vis ketones, nor do they seek to put a number to the contribution ketones make in terms of energy usage of the infant at any age.

I'm going to stipulate to some assumptions here for the sake of my analogy.  These are:
  • The newborn's brain metabolism can be extended to the 6 month old (solid foods introduced), and beyond up to two years.  
  • The statement in C&C that ketones *appear to be* an essential fuel to the fetus in mid-gestation can be translated into "ketones are an essential fuel for the infant".  
  • Per the previous bullet point, this need for ketones may be as high as to meet 30% of the brain's energy needs.
My kitchen analogy in brief:
  • Baseboard (throughout the house) is insuffient to provide the heat energy to adequately heat the room kitchen.
  • I need the auxilliary source, the kickplate heater, to provide some heat to do so.
  • When it's cold outside and/or the rest of the home is kept cooler, the kickplate provides more heat, up to ~30% at maximum usage.
Thus, although imperfect (as many  analogies tend to be), this models the brain with:
  • Glucose, the primary energy source, being represented by the baseboard (and residual heat from other rooms heated by this source)
  • Ketones being represented by a "backup" kickplate heater that is, nonetheless, required.
This analogy drives home an important part of the equation that Tim "not one gram of dietary carb is required" Noakes and Campbell ignore.  My kitchen will be a dang cold room if the baseboard heat is not functioning.  Even if it ran 24/7, my kickplate heater could never heat that kitchen, even on a mild day.  Simple math dictates that if the heater can supply up to 30% (maximum!) of the heat, that means the baseboard provides the large majority of the heat:  100 - 30 = 70% minimum.

This analogy is quite good in this regard.  That 30% figure (Adam,1975) was derived at elevated (4.5 mmolar) ketone levels in the perfused brains from non-viable fetuses.  The contribution in a mildly ketotic living infant is likely to be far less.

But the ketones are likely produced because they are required by the rapidly growing brain of the fetus and infant -- e.g. my kitchen on a very cold day.  Even if I have plenty of heat from the baseboard (glucose from lactose) I still need more from the kickplate (ketones) in that room.  It's probably worth noting that if I turned the baseboard thermostat up a bit more and perhaps used some fans, I could possibly achieve a comfy kitchen without the kickplate -- oh, like more frequent feedings for the baby.    In this scenario, the ketones are a valuable, but not an absolutely necessary supplementary fuel provided that glucose can be kept "sufficient".  I need my kickplate, but I don't need it in the "my kitchen is totally unlivable without it" sense.  My kitchen may not be optimally warm, but I can live in it.  No amount of ketones, however, can replace my need for glucose.  I freeze if the baseboard system fails.

Low Carb Logic ~ It Goes Something Like This:

  • If ketones might be beneficial, therefore more would always be better.
  • Ketones can be a significant source of energy therefore let's extrapolate that to major source
  • If ketones can be described as a major source of energy, it's only a small stretch to call them the primary source.
  • Breast milk appears to be ketogenic*
  • Breast milk is like 55% fat (a bit of an overestimate of most averages) so that must be why it's ketogenic.   Pay no attention to the carbohydrate.
  • Low carb diets are ketogenic in adults
  • Therefore:  Wean babies onto LCHF to keep the keto magic going!
*Breast Milk is Ketogenic(?)
This is going to be the topic for a BabyGate File or two.  Tamzyn Murphy Campbell is utterly irresponsible in her presentation here.  It does appear that breastfeeding initiates the ketogenic adaptation best, but there are several caveats to this that would caution against the macro composition being a key factor in this adaptive phenomenon, if even a factor at all.  

The Problem With Ketogenic Diets for Infants

To quote Campbell:  "Nature seldom does something without a reason"

  • There's a reason that "nature" put lactose in milk, and quite a lot of it in human breast milk.  
  • Furthermore, there's probably a reason why human milk is quite varied but the lactose content varies the least regardless of nutritional status of the mother (while fat and protein vary far more widely).  
  • Carbohydrate is the most conserved macronutrient in breast milk, where the 2.5th percentile contains 6.4 grams/dL vs. the 97.5th at 7.65 grams/dL.  

Let's stop pretending that the glucose source part is some after thought here.  Even most will acknowledge that it's there to support the enormous (proportionally) energy needs of the growing infant brain.   Yet somehow these needs go away vis a vis glucose but remain for ketones?

If Campbell and others had actually read the papers they are citing -- perhaps at all, or certainly with a less biased preconception -- they would notice a running theme about ketones.  They are indeed the kickplate heater of the late-term fetus and neonate.  Their production is largely to counter hypoglycemia between feedings, although very early on ketogenesis is not even as tightly associated with blood glucose levels as one might expect.

The term adaptation is everywhere, and there are copious references to "fasting" or between feeding levels ... ketones to replace or supplement glucose in times of high demand. High demand being the most likely reason that a carbohydrate intake that would otherwise prevent ketosis, doesn't prevent it entirely.  And high demand being why the infant (especially the newborn) falls so easily into a deeper state of ketosis if feedings are too widely spaced.

But even at that, we are not talking dramatic levels of ketosis, even in the neonate, as shown below.

According to Cunnane & Crawford, total energy needs of the infant almost double at 4-6 months, and more than triple at 1-2 years  (161, 300, 590 kcal/day respectively).  Meanwhile, though energy needs of the brain decline as a percentage of the total, (74, 64, 53% respectively), the absolute energy needs of the brain increase dramatically as well, from 118 kcal/day for newborns, to 192 kcal/day at 4-6 months, to 311 kcal/day at 1-2 years.

So as the absolute glucose needs of the infant continue to rise along with its total caloric needs, Tim Noakes, Pete Evans, Tamzyn Campbell, Caryn Zinn, Nora Gedgaudas and a whole host of others would have you begin to replace some breastmilk with low carb foods and eventually transition your child off of the breast entirely onto LCHF.

Yes, let's see.  This is the equivalent of it getting colder outside, and turning down the thermostat on the baseboard heat to increase the output of the kickplate heater.  That makes total sense.

And if you're thinking about ketogenic diets in very young children for epilepsy, realize these are ALWAYS undertaken under close supervision and guidance by doctors and a dietitian.  They are not without their negative side effects, but these are considered acceptable trade-offs for improvement in other areas.  There's some special kind of nuttery going on out there to even consider applying anything resembling this to a child with no such therapeutic need.

Suggested Reading:

Just some articles you may like to have a read for yourself.  To be discussed in the next post(s).

  1. Patterns of metabolic adaptation for preterm and term infants in the first neonatal week  (source of my ketone graphic)
  2. Maternal Ketosis and its Effects on the Fetus (some early infant info as well)
  3. Medium-Chain Triglycerides in Infant Formulas and their Relation to Plasma Ketone Body Concentrations  (ref 5 in Campbell article)
  4. Cerebral metabolic adaptation and ketone metabolism afterbrain injury  (ref 11&12 in Campbell article)

The Kitchen Heating Analogy

How NOT To Heat Your Kitchen

In my former home, we had a large "country kitchen" and baseboard heating.  For those not familiar with baseboard heating, hot water is passed through a pipe and at various locations these pipes go through elements like the one shown.  Eventually the water goes back to the furnace and round and round it goes.   There's always some loss of energy for the parts of the pipes that don't go through elements, so the shorter the loop the more efficient.  In most floor plans, this means a single loop around the outside walls of the house.  

Well in that home, our kitchen was a large 14' x 28' room with roughly 8' of outside wall on one end (the breakfast nook).  The rest of the outside wall had cabinets and appliances, as did the only other wall that would have made sense to put a heating unit on.   There were also three large openings to the room -- one to the vaulted ceiling entryway, large arch to the dining room, and a floor to ceiling passthrough to a family room -- and two doors to other rooms.  All of this to describe the fact that the kitchen was somewhat inadequately heated by the baseboard system when the outside temperature dropped below around 35 °F (~ 2°C).  Any residual heat that might have come in from adjacent rooms was somewhat negated by the cooler entryway, especially on particularly cold days. 

And so, under the center island cabinet, the builders installed a kick-plate heater.  The kick-plate heater had its own thermostat.  If it wasn't too cold outside, we were using the stove, or somesuch, the heater may have come on a few times when we turned down the heat at night, or perhaps not even at all.  If it was a very cold day, that thing came on periodically pretty much constantly.  I would guesstimate it provided as much as 30% of the heat to the kitchen.  When the central baseboard system provided less heat, that heater provided more of it.  

So I had this brilliant idea!  The kick plate heater does such a nice job of helping to heat my kitchen.  It works more when the baseboard heat is low.  So I decided I should turn the baseboard thermostat down to 55 °F (~ 13 °C) so the kickplate heater can really heat that kitchen!

What?  That doesn't make sense?  

Excerpt from Cunnane & Crawford

6. Body fat and ketogenesis.  
The link between infant fat stores and human brain expansion during evolution involves more than providing fatty acids for oxidation to meet energy needs. It also involves three breakdown products of fat oxidation collectively called ketone bodies (ketones; b-hydroxybutyrate, acetoacetate and acetone). There are two reasons why ketones became important to human brain evolution.  
First, the brain can oxidize ketones but it does not oxidize the fatty acids they come from. In adults, glucose is the main fuel for the brain. If food is restricted, body glucose stores (glycogen) last less than 24h. Without ketones, brain function would be rapidly compromised or muscle protein would need to be degraded to release amino acids that can be converted to glucose.  Hence, ketones are an essential alternative fuel to glucose for the brain. 
Healthy human infants have a large store of fat that is available to make ketones. In infants, slightly elevated blood ketones are present all the time (mild ketonemia) regardless of feeding status. This is not the case with fed adults. In human fetuses at mid-gestation, ketones are not just an alternative fuel but appear to be an essential fuel because they supply as much as 30% of the energy requirement of the brain at that age (Adam et al., 1975).
Second, ketones are a key source of carbon for the brain to synthesize the cholesterol and fatty acids that it needs in the membranes of the billions of developing nerve connections. ...
Hence, in all mammals studied, ketones have two important roles in the brain—they provide a reliable source of brain energy in between feeds, and they provide a major proportion of the lipid building blocks for developing brain cells. 
The uniqueness of the human situation is that babies are endowed with proportionally by far the largest ketone reserve (body fat) of any mammalian infants (Widdowson, 1974), a reserve which is suitably matched to the high energy and structural demands of the developing infant brain.

Excerpt from Campbell
Did you know that human newborns and exclusively breastfed babies are in ketosis?[1] I am a dietitian, with two years of intensive postgraduate training in nutrition, and I didn’t realise this until just over a month ago. The fact that human babies are naturally in ketosis is an inconvenient truth because it implies that ketosis (which also occurs when fasting or eating a low-carbohydrate, high-fat diet) is not only a natural metabolic state for human infants, but that it’s probably beneficial too. Nature seldom does something without a reason, so it’s likely that ketosis may confer some kind of evolutionary advantage to human infants. Research suggests that it may be one of the main factors behind the development of the large human brain.[2] ...
...In the second half of pregnancy ketones supply as much as 30% of the energy required by the foetal brain, implying that ketones are essential for foetal brain development.[3]
... Babies convert the fat from their stores and their diet (ideally breastmilk) into ketones.[6] These ketones provide them with the energy they need between feeds and act as key building blocks for essential brain structures.[7] Human babies are thought to be fatter than other mammals so that they have a reserve to support their higher demands for energy and brain-structure development.[8]
I've bolded and put in [ ] her citation numbers.  These ALL reference the same Cunnane & Crawford paper.