Paradigm shift

From PWS Notes
Jump to: navigation, search

Contents

Overview

The basic paradigm with PWS is that it is a disease of obesity and uncontrollable cravings. These cravings come from a messed up hypothalamus which does not receive a satiety signal. That, in combination with the low movement and a naturally slow metabolism and body compartmentalization that favors fat over muscle, results in the characteristic PWS obesity. Neurological issues such as mental retardation, skin picking, sensory problems, autistic behaviors, learning disabilities, are considered to be the downstream effect of a different (non-hypothalamic) primary defect.

Since this is the basic paradigm, the basic treatment addresses body compartmentalization through growth hormone. The thinking is that this allows the body to develop more muscle, which makes movement easier and consequently raises metabolism. Growth hormone has the added benefit of being a neuronal growth factor, which may help with mental retardation issues.

Growth hormone is then combined with a calorie-restricted diet as need. Calorie restriction often comes in the form of locking cabinets to prevent foraging. There is a heavy emphasis on limited portions and low-calorie food substitutes. People with PWS can lose and keep off weight if they adhere to these quantity-focused diets.

Buried in the PWS literature is a minority opinion. That opinion holds that PWS is a disease of starvation masquerading as a disease of obesity.

This interesting article argues that PWS is more a model of starvation than it is a model of obesity. The paradox of Prader-Willi syndrome: a genetic model of starvation Holland A, Whittington J, Hinton E. Section of Developmental Psychiatry, Douglas House, CB2 2AH, Cambridge, UK. ajh1008@cam.ac.uk The neurodevelopmental disorder, Prader-Willi syndrome, is generally regarded as a genetic model of obesity. Although the values of some hypothalamic neuropeptides are as expected in obesity, and should result in satiety, we propose that abnormal hypothalamic pathways mean that these are ineffective. We postulate that the body incorrectly interprets the absence of satiation as starvation, and therefore, paradoxically, this syndrome should be redefined as one of starvation that manifests as obesity in a food-rich environment. Also, this syndrome is generally believed to be a contiguous gene disorder, which results from the absence of expression of the paternally derived alleles of maternally imprinted genes on chromosome 15 (15q11-13). We argue, however, that the whole phenotype can be explained by one mechanism and, by implication, the failure of expression of the paternal allele of a single maternally imprinted gene that controls energy balance. We suggest clinical and laboratory approaches to test our hypotheses.


I think that this alternative viewpoint deserves more study. I will attempt to flesh out this theory based upon personal observations with my son as well as reports from other parents and analysis and brilliance from Oneida.

I/we propose that the primary deficiency is actually one of carbohydrate metabolism. Individuals with PWS cannot use carbohydrates as energy. Instead, they shunt them immediately to fat storage and begin to look for more food. Most humans, when very hungry, look for quick energy (carbohydrates). When a person with PWS does this, they again store the energy as fat and yet they remain starving. Ingestion of a carbohydrate-heavy diet can also lead to cravings of more carbohydrates. Thus the individual with PWS feels profoundly that they are starving and that they crave carbohydrates. The result is the food-seeking behavior and obesity that characterizes PWS.

But, the result goes beyond that. A child that is starving does not develop the mental capacity of a normal child. A child that is starving does not develop the height of a normal child. A starved child does not have the mental stability of a normal child. A starved and starving child is more likely to have learning disabilities and mood swings and bad behavior at school. And a starving child is likely to be obsessed and obsessive about food to the detriment of all other development.

So, how could this be tested? A primary deficit of metabolism can be addressed through some of the supplements currently available (coenzyme Q10 and carnitine fumarate). Although these supplements may not help a normal person, they would be expected to help a child with a primary metabolic deficiency. There is plenty of anecdotal evidence in the PWS community that they help. In my case, my son jumped 10 points on his gross motor Peabody within a week of beginning carnitine fumarate. This leap was unprecedented and unbelievable in the eyes of Kian’s PT.

A primary deficit of carbohydrate metabolism could be treated by providing an alternative energy substrate. While the typical American diet is very high in carbohydrates, I have always been compelled and advised to provide my son with a high protein diet. He is considered do be above average intelligence and at three years old he has no developmental delay.

Weight gain and hunger can be curbed by decreasing/removing carbohydrates. Kian began crossing weight growth curves when he turned 2 ½. At age three, we were told that his weight acceleration had gone on too long and that we needed to stop it. Instead of limiting calories and portions, we opted to remove rice, bread, crackers, and cereal and bananas from his diet. We replaced them with nuts (almonds and walnuts) and hard boiled eggs and yogurt. We still allowed some fruit and as much vegetables as we could persuade him to eat.

He lost weight and developed normal-looking musculature. He became a chatterbox overnight. He comfortably walks wherever we go (no stroller or carrying). He is considered normal by his physician and teachers and therapists. He always leaves food on his plate. He never asks for food other than at appropriate mealtimes or when triggered by food smells or other appropriate things. On the few occasions when we have let him have a heavy carb load (a bowl of rice for example), he spends the next 24 hours begging for more food.


Metabolic Problem at Birth

(2005) Stefan et al, Hormonal and Metabolic Defects in a Prader-Willi Syndrome Mouse Model with Neonatal Failure to Thrive (http://endo.endojournals.org/cgi/content/full/146/10/4377). For those not familiar with the study, basically what the Stefan team did was develop transgenic (TgPWS) mice in which the PWS region is deleted and then monitored the energy metabolism of fetal and newborn pups. This is what they found (adapted and extended from a post I made on the Holistic list):

1. The neonatal PWS mice pups had insulin and glucagon levels below the level of detection, as well as lower overall body fat. (Like insulin, glucagon is an important hormone in carbohydrate metabolism that is synthesized and secreted from special cells in the pancreas. Normally, when glucose levels start to drop, glucagon signals the liver to create new glucose (gluconeogenesis) and release it into the bloodstream in order to prevent the development of hypoglycemia (low blood sugar).)

2. As is usual, at birth there was an abrupt loss of the maternal supply of the glucose the neonatal pups relied on for energy.

3. The newborn pups then rapidly used up their own serum glucose supply (as is normal) and so had to turn to the breakdown of glycogen (stored in the liver) into the glucose needed for energy.

4. However, liver glycogen supplies were also soon depleted (as is normal) and the pups had to turn to gluconeogenesis and ketogenesis for energy between nursing. This is where the PWS pups (and probably babies with PWS) started running into serious trouble. Normally the rise in plasma glucagon, fall in plasma insulin and resulting increase in liver cAMP that occur immediately after birth triggers the activation of phosphoenolpyruvate carboxykinase (PEPCK), the rate- limiting enzyme for the gluconeogenesis pathway. However, glucagon levels remained nil in the TgPWS mice, which means that liver PEPCK gene transcription was not activated and therefore gluconeogenesis was blocked. The result was that the pups' brain, muscles, etc. didn't have the energy they needed to function properly, so they became severely hypotonic and lethargic with poor suck.

5. As would be expected, when confronted with increasingly severe hypoglycemia, the pups' bodies then tried to switch to fatty acid oxidation (burning) for energy production. But it is glucagon that causes the release of lipids (fats) so they can be burned for energy and since glucagon was still nil, that release of lipids didn't occur, therefore blocking the use of fatty acids for energy production.

6. Like gluconeogenesis, the change in the ratio of insulin and glucagon that occurs at birth causes ketogenesis in the liver to turn on, so that pathway was also impaired by the very low levels of insulin and glucagon in the neonatal TgPWS mice. The result was that although the ketone level in the TgPWS mice was higher than in their wild-type littermates, it was not enough to make up for the overall severe energy deficiency they were experiencing. The pups were therefore caught in a vicious downward spiral in which their energy deficit continued to worsen because of their poor ability to suckle due to their hypotonia and lethargy.

7. Finally, the Stefan team found that "[a]n increase in plasma ghrelin levels occurs in postnatal TgPWS mice and appears to begin at the onset of severe hypoglycemia but is not directly coincident with hypoinsulinemia [low insulin levels]. These findings are consistent with known regulators of ghrelin expression and secretion because both glucose and insulin have been shown to suppress ghrelin levels. By [day 5], ghrelin levels in TgPWS mice are approximately 3-fold higher than in [wild type] littermates, suggesting that high ghrelin levels in TgPWS might be a physiological adaptive mechanism in an attempt to increase feeding via its actions on the [hypothalamus] to ameliorate the rapidly worsening failure to thrive."

The TgPWS mice typically die within a week of birth, but that doesn't happen with babies with PWS because tube feeding is typically started when it's finally realized after a day or two that there's a serious feeding problem. Later, when they're finally strong enough to suck, we're always sticking a bottle in their mouth and repeatedly arousing them to keep them sucking so that their nutritional intake will be adequate. However, the profound hypersomnolence and lethargy and slow growth during the failure to thrive stage strongly suggest that both the use of glucose and fatty acids for energy production is significantly impaired, which is what would be expected if insulin and glucagon secretion continue to be reduced in the neonatal period.

One of the Stefan findings of particular interest is that although ketogenesis occurred in the TgPWS mice, it was not high enough to make up for the overall deficit in energy production. Normally the change in the insulin:glucagon ratio that occurs at birth results in the activation of hydroxymethylglutaryl-CoA synthase (3-hydroxy-3- methylglutaryl-CoA synthase). The final steps in the ketogenesis pathway are then:

1. Hydroxymethylglutaryl-CoA synthase catalyzes the reaction of acetyl-CoA + H2O + acetoacetyl-CoA to form (S)-3-hydroxy-3- methylglutaryl-CoA + CoA.

2. 3-hydroxy-3-methylglutaryl-CoA (HMG-CoA) lyase (located mostly in mitochondria but also found in peroxisomes) then cleaves 3-hydroxy-3- methylglutaryl-CoA, producing acetyl-CoA and acetoacetate, the ketone body that is then easily reduced to the primary ketone body, beta- hydroxybutyrate. Some acetoacetate also spontaneously decarboxylates to yield acetone, the distinctive smell of which is a key diagnostic sign of severe ketosis.

Interestingly, an acylcarnitine profile for an 11-month-old baby boy with PWS shows elevated 3-OH-isovalerylcarnitine (C5-OH) and adipoylcarnitine/methylglutarylcarnitine (C6-DC), both of which are elevated in 3-hydroxy-3-methylglutaryl-CoA (HMG-CoA) lyase deficiency (indeed, according to the Metagene knowledge base of inborn errors of metabolism (IEM), it's the only known IEM in which C6-DC is increased - http://www.metagene.de/programm/o.prg? pos=5&lan=de&diseases=4). Of course, HMG-CoA lysase deficiency as a distinct genetic syndrome is not present in PWS, but the increased C5- OH and C6-DC acylcarnitines do suggest a down-regulation of HMG-CoA lyase and/or synthase that results in a significant impairment of ketogenesis, especially in infants but I suspect also in older children and adults with PWS. In that regard, it is interesting to note that carnitine supplementation is key part of the treatment of HMG-CoA lysase deficiency and is also proving to be of significant benefit for many of those with PWS.

To stop for a moment and recap, the overall picture we have here is of a neonate in which glucose and fatty acid oxidation, gluconeogenesis and ketogenesis are probably all severely impaired, most likely due to one or more missing or misleading metabolic feedback signals caused by the loss of non-coding DNA binding sites in the PWS region on chromosome 15. As a result, aside from an insufficient level of ketogenesis, the infant essentially becomes dependent on the catabolizing of protein for amino acid oxidation in order to meet his/her energy needs and the limited lab tests from the NICU period and early infancy I've been able to review suggest that is exactly what is happening. A serum amino acid screen from one baby's third day of life showed "severe generalized hypoaminoacidemia" with the added notation "? malnutrition." However, a urinary amino acid screen taken the same day was normal. In other words, the level of amino acids circulating in her blood was extremely low but there was no excessive urinary loss of amino acids. So where were the amino acids going since, as we all know, neonates with PWS aren't packing on large amounts of either muscle or fat? I suspect the answer is that protein was being catabolized for amino acids to oxidize (burn) for energy to keep her body running, albeit in a minimal state that results in the profound hypotonia and comatose-like lethargy so characteristic of early infancy in PWS.

Central to the Stefan energy metabolism model of PWS is the hypothesis that hyperphagia in PWS "results from a failure in compensatory mechanisms" that were resorted to in order to cope with the severe energy deficit - that is, functional starvation - characteristic of the neonatal stage and early infancy. Indeed, there are many aspects of PWS that parallel what occurs in energy and protein starvation, including -

  • hypotonia
  • low muscle mass
  • lethargy
  • failure to thrive
  • developmental delays
  • cognitive impairment
  • behavioral disturbances
  • brain imaging abnormalities
  • anorexia/lack of appetite
  • increased tendency to hypothermia
  • reduced fever during infectious illness
  • increased tendency to hypoglycemia
  • angular stomatitis
  • hypopigmentation
  • hypoaminoacidemia
  • hypoproteinemia (low serum protein)
  • low serum albumin
  • low BUN (blood urea nitrogen)
  • hyperammonemia
  • increased urinary 4-hydroxyphenylacetate
  • low creatinine
  • low alkaline phosphatase (ALP)
  • anisocytosis (unequally sized RBCs)
  • elevated ghrelin
  • low serum leptin
  • low serum IGF-1
  • low serum IGFBP-3
  • slow growth
  • low fasting insulin
  • impaired glucose tolerance
  • low serum carnitine

The above is only a sketchy outline of what the Stefan study and limited lab test data suggest about metabolic impairment in PWS and does not include other important aspects such as impaired mitochondrial respiratory chain transport and the development of a secondary carnitine deficiency due to impaired fatty acid metabolism. Those who would like to examine the issue of significantly impaired metabolism and concomitant functional starvation in PWS in more detail may be interested in the pages below at my web site as well as the Stefan study (http://endo.endojournals.org/cgi/content/full/146/10/4377) and a 2003 abstract (found by Teresa :-), Dual pathology in two hypotonic children with Prader-Willi syndrome and muscle mitochondrial Complex I deficiency at http://adc.bmj.com/cgi/content/full/archdischild% 3b88/suppl_1/A70.

http://www.pwsdots.org/ResearchNotes/ProteinStarvation http://www.pwsdots.org/ResearchNotes/AcylcarnitineProfiles http://www.pwsdots.org/ResearchNotes/LabResults http://www.pwsdots.org/ResearchNotes/MuscleBiopsy http://www.pwsdots.org/ResearchNotes/NeonatalMetabolism

I apologize for the length of this post, but I think the question of metabolic impairment in PWS has been ignored for far too long and I am interested in what others think about the issue. Personally, I think the Stefan study was landmark research and I would really like to see FPWR continue to support that team, as I understand from Teresa that they are continuing their research with the TgPWS mice. I believe such research is critically important because a thorough understanding of metabolic function in PWS could allow the development of nutritional and other treatment protocols with the potential to greatly ameliorate many of the severe impacts of PWS such as failure to thrive in infancy, hyperphagia, hypotonia, low energy and sleepiness, cognitive impairment, emotional and behavioral disturbances, growth hormone deficiency, etc. In other words, it seems very possible that at least some aspects of PWS that have long been assumed to be due to a central hypothalamic dysfunction may actually be secondary to impaired energy metabolism.

ALA

I have spent quite a bit of time looking into alpha-lipoic acid (ALA), not just regarding the appetite issue but also for its neurocognitive and mitochondrial function benefits. (You can see the abstracts I've collected about ALA at http://www.tcmnotes.org/ala-abstracts/.) I do think it might have benefits for PWS, perhaps in concert with acetyl-l-carnitine. However, there are some concerns I need to resolve, for example, its marked hypoglycemic effect during fasting states[1] and its inhibition of insulin[2] and gluconeogenesis[3] (the making of new glucose in the liver), as well as the question of ALA-induced reductions in pyruvate carboxylase and pyruvate dehydrogenase complex activity[4] (and perhaps phosphoenolpyruvate carboxykinase, which is the rate-limiting enzyme of gluconeogenesis), all of which are vital parts of glucose metabolism. So far I have not been able to resolve those questions to my satisfaction, so I really can't say at this point whether or not it is appropriate for PWS. If anyone has used it for PWS, I would very much like to hear about the results.

The reason why I'm concerned about ALA's possible impacts on such things as blood sugar and pyruvate dehydrogenase levels is important, so I'm going to seize the opportunity to discuss that a bit.

I agree with Lara that there is likely to be an impairment of fat metabolism in PWS, but I think there also might be a significant impairment in glucose metabolism, especially gluconeogenesis (the making of new glucose by the liver from fats and amino acids). That suspicion initially arose because of the neonatal presentation of profound hypotonia and mental and physical lethargy which, in the absence of any of the known inborn errors of metabolism, suggests either severe hypoglycemia or hyperammonemia. I haven't been able to find any suggestion of neonatal hyperammonemia in the PWS literature, but have found a few scattered hints at hypoglycemia[5]. What really strengthened my suspicion about impaired glucose metabolism in PWS, though, was an Oct. 2005 article in the journal Endocrinology by Stefan et al, _Hormonal and metabolic defects in a Prader-Willi syndrome mouse model with neonatal failure to thrive_.[6] If folks haven't read it yet, I really recommend doing so, as my sense is that it may represent the most important research into PWS to date. Basically what they did was develop transgenic mice in which the PWS region was deleted and then monitored the energy metabolism of fetal and newborn pups. This is what they found:

1. Both the fetal and postnatal PWS mice pups had insulin and glucagon levels below the level of detection, as well as lower overall body fat, all of which strongly suggests a primary pancreatic defect. (Like insulin, glucagon is an important hormone in carbohydrate metabolism that is synthesized and secreted from special cells in the pancreas. Normally, when glucose levels start to drop, glucagon signals the liver to create new glucose (gluconeogenesis) and release it into the bloodstream in order to prevent hypoglycemia (low blood sugar).)

2. As is usual, at birth there was an abrupt loss of the maternal supply of the glucose the pups needed for energy.

3. The newborn pups rapidly used up their serum glucose supply (as is normal) and so had to turn to the breakdown of glycogen (stored in the liver) into the glucose needed for energy.

4. However, liver glycogen supplies were also soon depleted (as is normal) and the pups had to turn to gluconeogenesis. This is where the PWS pups (and probably babies with PWS) started running into serious trouble. Even though a severe metabolic crisis was developing and insulin levels were below the limits of detection, the pups' glucagon levels also remained below the level of detection, which means the liver never got the signal to start up the production of new glucose, with the result that the brain, muscles, etc. didn't have the energy they needed to function properly, resulting in severely hypotonic, lethargic pups with poor suck.

5. As would be expected, when confronted with increasingly severe hypoglycemia and energy deficiency, the pups switched to fatty acid oxidation (burning) for energy production. But it is glucagon that causes the release of lipids (fats) from peripheral tissues so they can be burned for energy and since glucagon was basically nonexistent, that release of lipids didn't occur. As a result, the only source of energy the pups had available was the fats they got directly from nursing. But since the pups were severely hypoglycemic and weak, they were unable to nurse well and so lipid levels remained low (a situation that could be exacerbated by impaired lipid uptake due to down-regulated CD36/FAT expression). The result was the development of a vicious cycle with the pups remaining in a state of severe energy deficiency with resulting hypotonia, lethargy, poor suckling and failure to thrive.

6. Finally, the study authors found that "[a]n increase in plasma ghrelin levels occurs in postnatal TgPWS mice and appears to begin at the onset of severe hypoglycemia but is not directly coincident with hypoinsulinemia [low insulin levels]. These findings are consistent with known regulators of ghrelin expression and secretion because both glucose and insulin have been shown to suppress ghrelin levels. By [day 5], ghrelin levels in TgPWS mice are approximately 3-fold higher than in [wild type] littermates, suggesting that high ghrelin levels in TgPWS might be a physiological adaptive mechanism in an attempt to increase feeding via its actions on the [hypothalamus] to ameliorate the rapidly worsening failure to thrive. However, either this signal is unrecognized due to an unknown mechanism, or it may be too late to elicit a physiological response."

Care must of course be used when attempting to translate the findings from mouse studies to humans, but I think Stefan et al have potentially moved PWS research light years beyond the usual "central hypothalamic dysfunction" mantra of current PWS "experts," which has never been able to provide a comprehensive explanatory mechanism of any depth for what is going on with PWS and instead seems to just feed into an attitude of, "Well, what can you do, it's a central hypothalamic dysfunction. <shrug>" Indeed, the essential weakness of the "central hypothalamic dysfunction" (CHD) dogma becomes readily apparent when it is realized that the _only_ clinical guidance it has been able to provide for the last 20+ years is the recommendation of growth hormone treatment. Even that recommendation, though, is purely empirical, as the CHD theory has never been able to demonstrate exactly how the presumed CHD leads to growth hormone deficiency. The Stefan energy metabolism model, however, easily explains why there is growth hormone deficiency in PWS - growth takes energy, which is exactly what the PWS infant has a severe shortage of, so the physiological processes of a newborn frantically trying to cope with a severe energy deficit become focused almost entirely on survival, not growth.

Plus, see muscles.

The phases of PWS

In addition to growth hormone deficiency, the Stefan energy metabolism model also helps to explain other notable aspects of PWS, including the biphasic presentation, severe neonatal hypotonia and lethargy (and their resolution at about four to six months of age), hyperphagia, hypogonadism, low muscle mass, low resting energy expenditure, cognitive deficits, and certain behavioral aspects (e.g., temper tantrums). (Note that I am not saying there is no central hypothalamic dysfunction in PWS, as there may very well be. It remains to seen, though, how much of that presumed CHD is truly central in nature and how much is actually the effects of real (in the neonatal phase) and, later on, perceived energy starvation.)

PWS mice typically die within a week of birth, but that doesn't happen with PWS babies because NG tube feeding is implemented when it's finally realized after a day or two that there's a serious feeding problem. Later, when they're finally strong enough to suck, we're always sticking a bottle in their mouth in order to maintain adequate nutrient intake. Eventually solids are introduced, which usually means such first foods as rice cereal, banana, applesauce, etc., all of which just happen to be wonderful sources of glucose. It has been a puzzle to me why PWS babies typically emerge from the severe hypotonia and lethargy of early infancy around the age of four to six months, but now I suspect it may be at least in part due to the start of feeding glucose-rich foods (it (also) may be that is when the pancreas is finally able to synthesize enough glucagon to prompt the liver into making glucose or some related metabolic process has matured enough to the point that the infant finally has enough energy available for such things as movement and alertness). However, by that time their whole metabolic regulatory system is seriously out of whack due to their body's desperate attempts to compensate for the severe energy starvation during the neonatal period and early infancy, including chronically elevated ghrelin levels, and so the body is primed for hyperphagia to develop when an as yet undefined trigger occurs. That trigger could be something like a spurt in the maturation of the pancreas' endocrine activity or some other metabolic process.

If the Stefan energy metabolism model of PWS is correct, what does that mean in practical terms? Unfortunately, there is more research that needs to be done in order to say for sure. (For example, the Stefan authors note that "[i]t will be important in future studies to examine the expression levels of genes encoding critical gluconeogenic enzymes or the levels and/or activity of phosphoenolpyruvate carboxykinase[7] or pyruvate carboxylase, for example, that could be altered in the TgPWS mouse.") However, the Stefan findings as well as what is known about energy metabolism in general and experience from the clinical treatment of inborn errors of glucose metabolism (such as pyruvate carboxylase deficiency[8] and pyruvate dehydrogenase complex deficiency[9]) do suggest that complete avoidance of fasting states (which require gluconeogenesis to avoid the development of hypoglycemia) is critical at least during the neonatal period and early infancy.

Recommendations for the child with PWS

Central to the Stefan energy metabolism model is the hypothesis that hyperphagia in PWS "results from a failure in compensatory mechanisms" that were resorted to in order to cope with the severe energy deficit characteristic of the neonatal stage and early infancy. Another way to conceptualize that is that the body continues to think it is still facing a survival-threatening state of energy starvation and so is determined to grab every bit of food in sight to alleviate that while continuing to minimize energy-consuming processes such as growth. If the Stefan hypothesis is correct, the question then is how to convince the body that the threat of energy starvation is past. Answering that could be complicated by the fact that fat absorption might be impaired (due to CD36/FAT down-regulation and perhaps other factors) and that gluconeogenesis and/or other aspects of carbohydrate metabolism may also be impaired. What I'd really like to see is something like a day-long conference of 10 or so clinical nutritionists who have experience working with inborn errors of fatty acid and carbohydrate metabolism during which they review the Stefan article and other pertinent materials with a goal of providing recommendations as to how to persuade the PWS body that it doesn't have to worry about energy starvation any longer. Short of that, though, the following might be helpful:

1. Avoidance of fasting states, that is, any period over about 3 hours without food.

2. As Lara and Peter have found, the judicious addition of healthy fats to the diet via flax seed meal and oil, fish and fish oil, avocados, omega-3 eggs, nuts like walnuts, pecans, macadamia, etc. is likely to be helpful. Indeed, I suspect the benefits of flax and fish oil supplements for PWS stem at least in part from the increase in the amount of dietary fat suitable for burning for energy. Peter's point about scrupulous avoidance of trans fats (hydrogenated oils, margarine, etc.) is very well-taken, and I would also avoid poor quality generic oils such as vegetable or cooking oils.

3. Since CD36/FAT down-regulation suggests an impairment in the absorption and metabolism of palmitic acid, it might be a good idea to avoid coconut and palm oil, given that both are high in palmitic acid. (Note that we currently think that coconut oil is quite useful for our kids. 11/2013)

4. Because fish oil is also high in palmitic acid, if fish oil supplements are used it might be advisable to use products that have high amounts of EPA and DHA relative to the total amount of fish oil. For example, the Nordic Naturals Ultra Omega mentioned by Lara has 1280 mg of omega-3 oils per 2000 mg of total fish oil and Coromega provides 1230 per 2000 mg, while the supplement that my friend's baby reacted so badly to requires 4,295 mg of total fish oil to provide roughly the same amount of EPA and DHA as the Coromega and Ultra Omega products. (Note that high quality fish oil seems to work well for most of our kids. 11/2013)

5. Carbs should be low-glycemic so that their glucose slowly trickles into the bloodstream, not hit it all at once. For the same reason, carbs should not be eaten by themselves but always in conjunction with protein and fat. As many PWS parents already do, fruit juices should be minimized, that is, by diluting them or restricting them to small amounts. In general, most carbs should be in the form of veggies, not fruits and grains.

6. Protein and fats should probably constitute the bulk of the diet due to their high satiety effect and ability to support growth (including lean muscle mass) while providing a steady stream of the substrates needed for energy production via fatty acid metabolism.

7. A high-quality, complete and balanced vitamin and mineral supplement, particularly one that provides at least several times the RDA for B vitamins (which are required to make the enzymes necessary for both glucose and fatty acid metabolism), seems advisable. Regular monitoring of iron levels is also advisable with supplementation implemented if they drop below the mid-normal range. (Ferrochel is my preferred iron supplement due to its high uptake efficiency and safety profile.)

8. Because those with PWS may be more dependent than others on fatty acid metabolism to provide the energy needed for proper physiological functioning and carnitine is necessary for the transport of fatty acids into the mitochondria for burning for energy, supplementing with L-carnitine may also help convince the body that it can stop worrying about a shortage of energy.

Taken together, points 1-6 above suggest something like six mini-meals spaced about three hours apart throughout the day, each comprised of roughly 40% protein, 40% high quality fats and 30% low glycemic carbs (primarily veggies). The various Zone Diet books by Barry Sears, PhD, have menu examples, recipes and snack examples that meet the 40-40-30 protein-fat-carbs ratio. My experience (with both myself and clients) is that the 40-40-30 macronutrient ratio typically provides sustained high levels of mental and physical energy and satiety while promoting lean muscle mass and optimal body fat levels, especially when used in conjunction with regular exercise (which for most toddlers and young children means their usual busy activity).

Note that the above suggestions are merely my best guess as to what might help re-program the PWS body from perceived starvation mode to more normal metabolic functioning and could be totally wrong. Also, there are likely to be other things such as other supplements, herbs, etc. that could help with the re-setting of the body's perception of available energy supplies. However, that's a whole other huge topic.


1. Apr. 1999. Lipoic acid acutely induces hypoglycemia in fasting nondiabetic and diabetic rats. </pws-abstracts/99/4/1/apr-1999-lipoic-acid-acutely-induces-hypoglycemia-in-fasting-nondiabetic-and-diabetic-rats.html> http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=pubmed&cmd=Retrieve&dopt=AbstractPlus&list_uids=10206446&query_hl=10&itool=pubmed_DocSum

2. July 2006. Alpha-lipoic acid regulates AMP-activated protein kinase and inhibits insulin secretion from beta cells. </pws-abstracts/2006/7/1/july-2006-alpha-lipoic-acid-regulates-amp-activated-protein-kinase-and-inhibits-insulin-secretion-from-beta-cells.html> http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=pubmed&cmd=Retrieve&dopt=AbstractPlus&list_uids=16752177&query_hl=1&itool=pubmed_DocSum

3. May 1984. Inhibition of gluconeogenesis in rat liver by lipoic acid. Evidence for more than one site of action. http://www.pubmedcentral.nih.gov/picrender.fcgi?artid=1153543&blobtype=pdf

4. Oct. 2004. R-lipoic acid inhibits mammalian pyruvate dehydrogenase kinase. </pws-abstracts/2004/10/1/oct-2004-r-lipoic-acid-inhibits-mammalian-pyruvate-dehydrogenase-kinase.html> http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=pubmed&cmd=Retrieve&dopt=AbstractPlus&list_uids=15512796&query_hl=10&itool=pubmed_DocSum

5. E.g, Oct. 1996. General anesthesia in Prader-Willi syndrome. http://www.tcmnotes.org/pws-abstracts/96/10/1/oct-1996-general-anesthesia-in-prader-willi-syndrome.html "Our data support the hypothesis that hypoglycemia in the Prader-Willi syndrome originates from inadequate lipolysis during starvation." [Note that by starvation they are referring to a fasting state, as typically occurs before, during and after surgery.]

6. Oct. 2005. Hormonal and metabolic defects in a Prader-Willi syndrome mouse model with neonatal failure to thrive. http://www.tcmnotes.org/hormone-metabolic-aspects-pws-mouse-with-ftt/

7. Phosphoenolpyruvate carboxykinase - http://www.nutritionandmetabolism.com/content/2/1/33

8. Pyruvate carboxylase deficiency - http://www.emedicine.com/PED/topic1967.htm

9. Pyruvate dehydrogenase complex deficiency - http://www.emedicine.com/PED/topic1969.htm

Fat metabolism

CD36/fatty acid translocase (FAT) on chromosome 7 is significantly down-regulated in PWS. (Clin Genet. 2006 Jan: CD36 expression and its relationship with obesity in blood cells from people with and without Prader-Willi syndrome) 9-cis-retinoic acid (a vitamin A metabolite) up-regulates CD36/FAT. (Biochim Biophys Acta. 2005 May 30: Gene expression profiling identifies retinoids as potent inducers of macrophage lipid efflux). So your suspicion that vitamin A might play a role in the regulation of the expression of at least some genes in PWS seems well-founded.

Btw, I came across the CD36-PWS-vitamin A connection a few days ago because a friend's baby with PWS (UPD), who was severely hypotonic and "lethargic" until acetyl-l-carnitine (ALC) supplementation was started about 1-1/2 months ago, suddenly redeveloped severe hypotonia and lethargy for a day after being given a one tsp dose of fish oil (not on my recommendation, btw). On a per weight basis, that would be about 3 tbs for an adult, which might make me somewhat nauseous and sluggish, but I doubt it would plunge me into severe hypotonia. Her response to the fish oil, plus the fact that she responded so well to the ALC and there are anecdotal reports of other children with PWS who have responded well to L-carnitine, were further confirmation of my suspicion for a while now that there is an impairment in fatty acid uptake and/or energy metabolism in at least some with PWS. Unfortunately, energy metabolism in PWS is very poorly characterized despite the fact that hypotonia, hyperphagia and other characteristics of PWS fairly scream out impaired energy metabolism (at least to me :-). Anyway, I ran across the Clinical Genetics abstract while trying to figure out her response to the fish oil and decided to investigate CD36 more closely. So far two things really stand out about CD36/FAT with regard to PWS - (1) it plays a primary role (along with carnitine palmitoyltransferase I (CPT I)) in the transport of long-chain fatty acids into muscle mitochondria for oxidation, and (2) it also plays a primary role in dietary long-chain fatty acid processing for absorption in the intestines. So if dietary long-chain fatty acids aren't being absorbed well in PWS and their transport into mitochondria are impaired in PWS, it shouldn't be a big surprise that hypotonia and mental and physical lethargy are part of the classic presentation of PWS. Maybe you should start a page on mitochondrial dysfunction and energy metabolism? :-)

Should children be fed throughout the night?

Due to lack of sleep after the arrival of a new baby, we parents are usually greatly relieved when they reach that wonderful milestone of finally sleeping through the night. Given their hypotonia, lethargy and hypersomnolence, PWS babies tend to achieve that milestone very easily, some practically on the first day (as with my friend's baby). But if the Stefan model is correct and PWS babies are in fact in a state of functional energy starvation, that early sleeping through the night is not necessarily a good thing. Instead, it is likely to be a sign of a serious problem in energy metabolism. As a result, we can't use guidelines based on babies who don't have PWS to make a decision about when to stop the middle of the night feeding for a PWS baby.

When my friend's baby was about 5 months old, I suggested that she re-institute a 3 or 4 a.m. feeding (she was still doing a midnight feeding). I did so in part because, despite not finding anything in the literature about hypoglycemia or some other disturbance in energy metabolism in PWS babies, I couldn't shake my gut feeling that there had to be some problem there. (The other reason for my suggestion was to increase the baby's nutritional intake, as she was an extremely poor feeder and it was necessary to repeatedly arouse her during feeding to prompt her to take a few more sucks on the bottle. She was not growing well despite my friend's diligent efforts to ensure adequate intake and it was my thought at the time that an additional feeding would help improve that.)

Although the baby's growth did improve somewhat, re-instituting the middle of the night feeding did not seem to make much of a difference in her hypotonia, lethargy, alertness or activity levels. It could be, though, that the small incremental increase in nutrition that she received through the additional feeding was not enough to make an observable difference in what was still a deep state of functional energy deficiency. However, given the potentially profound impacts that prolonged energy starvation can have on the developing brain, it could well be that the additional feeding still had something of a protective effect on neurological development that made it worthwhile. The fact that she made such rapid progress after the acetyl-l-carnitine was begun and was essentially developmentally on track a month after it was started hints that such a protective effect might have occurred.

The neonatal PWS mice in the Stefan study had essentially no insulin or glucagon secretion. I don't know if that is also true in PWS infants but it is known that insulin levels are unexpectedly low in older PWS children. A search at PubMed for "prader glucagon" turns up only five studies (none of which were particularly interested in glucagon levels) and I cannot determine from the abstracts what typical glucagon levels are in PWS. However, given that the profound hypotonia and lethargy of PWS babies begins to spontaneously resolve at around the age of 4-6 months and that insulin secretion is taking place in older PWS children, although typically at a much lower level than normal, suggests that even if insulin and glucagon levels are nil in PWS newborns, at some point the pancreas does develop the ability to secrete at least low levels of insulin and probably glucagon, which would mean that at some point in infancy the liver starts getting the signal via glucagon to make new glucose (gluconeogenesis) for burning for energy when serum glucose levels fall during a fasting state such as that which occurs during night-time sleeping. However, many aspects of PWS lead me to suspect that gluconeogenesis might still be low. If so, the development of some level of hypoglycemia during the night-time "fast" could occur. The brain always needs around 30-40% of its energy to come from glucose because the ketones produced by fatty acid metabolism during starvation, fasting states or very low carb diets can only supply about 60-70% of its need for energy, so there will be a problem if gluconeogenesis is not sufficient to supply that 30-40% during the overnight fasting state.

There is a blood test for glucagon levels (see, e.g., http://www.mdadvice.com/library/test/medtest69.html) and if I had a infant or child with PWS, I would seriously consider requesting that test as a first step towards determining if there is an impairment in gluconeogenesis or some other aspect of energy metabolism.

In the meantime, given that gluconeogenesis might be impaired and that impairment could possibly lead to the development of hypoglycemia during overnight fasting, for infants who still have significant hypotonia, lethargy and hypersomnolence, I would give them a bed-time feeding, as well as two feeding during night-time sleep, e.g., around 11 p.m. to midnight and 3-4 a.m.

For infants who have substantially emerged from the early hypotonia and lethargy, I would (1) continue to give them a bed-time feeding (breast milk or formula), and (2) consider one or two feedings during night-time sleep (e.g., 11-midnight and 3-4 a.m.), especially if they are groggy, low-energy and/or fussy or cranky when they wake up in the morning and those symptoms improve after the first feeding of the day. (Note, though, that those morning symptoms can also be the result of untreated sleep-related breathing disturbances such as obstructive sleep apnea.)

For toddlers and older children, especially if there is over-sleeping, morning grogginess, low energy and/or crankiness (irritability) upon natural awakening (e.g., on the weekends) not due to sleep-related breathing problems and those symptoms improve after breakfast, I would consider (1) a bed-time mini-meal/snack, small protein bar (or portion thereof) or 4-8 oz. of a meal replacement drink, all of which should contain roughly equal amounts (as a percentage of calories) of protein, fats and carbs, and (2) possibly a small protein bar or 4-8 oz. of a meal replacement drink in the middle of the night.

As noted above, these suggestions are based on my best guess about what could be going on with PWS and could prove to be totally off the mark. Hopefully further research into gluconeogenesis and other aspects of energy metabolism in PWS, particularly during infancy and early childhood, will help provide more definitive guidance.

Ketogenic diet

Please see Ketogenic Diet page for latest (2013) thinking on this subject.