I emailed Ray Peat yesterday to get his opinion about my “Does-saturated-fat-cause-high-blood-sugar?” experiment. I asked him if it was indeed the case that too much saturated fat in one’s diet could cause impaired cellular glucose uptake. I’ve never written to him before, and felt nervous about doing so. I felt like a kid writing to John Lennon or something. He wrote back in about 20 minutes and said:
Butter, cream, and coconut oil are the only common food fats that are mostly saturated, and because coconut oil is oxidized more quickly than most fats, it’s the least likely to block sugar oxidation. Any long chain fat can interfere with sugar oxidation, but the polyunsaturated fats, such as in poultry, fish, and pork, are more water soluble, and slower to be oxidized than saturated fats, and they affect hormones and other regulatory systems differently, so they interfere a little more strongly.
He also sent me the abstracts for a few relevant studies (cited at the bottom of this post).
I’ve been thinking today about my experiment. I did a statistical analysis on the data I collected (fat grams eaten vs. fasting blood sugar) and the data showed a moderate correlation between the two variables (r=0.42). If you’re familiar with statistics, you know that 1.0 is a perfect correlation – when X increases, Y always increases, and when X decreases, Y always decreases. A correlation of r=0.0 means there is no meaningful relationship between the two variables. So a correlation of r=0.42 is a reasonable correlation – but because I didn’t have enough data points (days) it wasn’t statistically significant – meaning, it could have been just chance that made my graph look the way it did. Another week and it would have been statistically significant.
So pondering today, I’m thinking, this is my health here, and my hyperglycemia is one of my biggest health concerns. I really just need to get over my whininess about eating too many sweet things and give this a good solid trial. No confounders – just eating low fat and monitoring my blood sugar.
So I’m going to do that, starting tomorrow, and I’m going to do it for at least 3 weeks. Hopefully longer. I’m not going to make any conscious effort to eat low-calorie. Just low fat. By low-fat, I mean I’ll keep fat under 20% of my total calories per day. If things are going well, I may go lower, but I’ll consider each day a success if I can do that. Fat calories will be coming largely from coconut oil. My diet will be centered around fat-free dairy, fruit, juice, lean meat, broth, honey, coffee, eggs, salt, liver, shellfish, and gelatin. I may need to eliminate cheese for a while.
It occurs to me that if I can manage to do this – prove at a level of statistical significance that a low-fat high-sugar diet fixes diabetes – well, that would really be something. I hear stories of people having done this, but I don’t know of anyone who has made their story public so others could learn from it and try it themselves.
So tomorrow begins….The Great High Sugar Low Fat Diabetes N=1.
Citations provided by Ray Peat:
Proc Natl Acad Sci U S A. 1988 Aug;85(16):6137-41.
Essential fatty acid deficiency prevents multiple low-dose streptozotocin-induced
diabetes in CD-1 mice.
Wright JR Jr, Lefkowith JB, Schreiner G, Lacy PE.
Department of Pathology, Washington University School of Medicine, Saint Louis,
Multiple i.p. injections of low-dose streptozotocin (40 mg/kg) produce insulitis,
beta cell destruction, and diabetes in male CD-1 mice. Recent data also suggest
that macrophages figure in the low-dose streptozotocin model. Because other
recent studies have shown that essential fatty acid deficiency prevents
autoimmune nephritis in mice, decreases the number of resident Ia-positive
glomerular macrophages, and decreases the elicitation of macrophages into the
glomerulus in inflammation, we examined the effect of essential fatty acid
deficiency on the incidence and severity of insulitis and diabetes in CD-1 mice
treated with low-dose streptozotocin. Streptozotocin-treated mice on the control
diet uniformly developed diabetes (19/19). Essential fatty acid-deficient mice
treated with streptozotocin did not develop diabetes (1/13). Mean plasma glucose
levels for the control and essential fatty acid-deficient mice were 384.5 ±
23.6 and 129.1 ± 15.5 mg/dl, respectively, at the end of 1 month. To discern
whether essential fatty acid deficiency prevented the streptozotocin-induced beta
cell injury or the inflammatory response to injured beta cells, mice were
repleted with daily injections of 99% pure methyl linoleate beginning 3 days
after the last streptozotocin injection. These mice also quickly developed severe
(3/4) or mild (1/4) diabetes. Histologic examination of the pancreata of control
mice or repleted mice showed marked insulitis and beta cell destruction; in
contrast, the pancreata of essential fatty acid-deficient mice showed
preservation of beta cells and only focal mild peri-insulitis. Essential fatty
acid deficiency thus prevents the insulitis and resultant diabetes in low-dose
streptozotocin-treated CD-1 mice, suggesting a central role for macrophages and
lipid mediators in this autoimmunity model.
Acta Diabetol. 1995 Jun;32(2):125-30.
Essential fatty acid deficiency prevents multiple low-dose streptozotocin-induced
diabetes in naive and cyclosporin-treated low-responder murine strains.
Wright JR Jr, Fraser RB, Kapoor S, Cook HW.
Department of Pathology and Surgery, Izaak Walton Killam Children’s Hospital,
Halifax, Nova Scotia, Canada.
We have previously shown that essential fatty acid (EFA) deficiency prevents
diabetes and ameliorates insulitis in low-dose streptozotocin (LDS)-treated male
CD-1 mice. The effects of EFA deficiency on the incidence of diabetes after LDS
treatment has not been examined in other strains. In contrast to highly
susceptible CD-1 mice, several other strains of mice are only partially
susceptible to LDS treatment and do not develop appreciable insulitis; however,
the susceptibility of these strains can be markedly increased by cyclosporin A
(CsA) pretreatment to reduce suppressor cell function. Weanling male BALB/cByJ,
DBA/2J, and C57BL/6J mice were placed on EFA-deficient (EFAD) or control diets
for 2 months and then divided into experimental and control groups. Ten EFAD and
10 control mice from each strain received LDS treatment (40 mg/kg/d 5 d); an
additional 10 EFAD BALB/cByJ and another 10 control BALB/cByJ mice received
subcutaneous CsA injections (20 mg/kg/d) for 14 days prior to and for 5 days
simultaneous with LDS treatment (40 mg/kg/d 5 d). Plasma glucose levels for all
mice were determined 3 times per week for 3 weeks after LDS treatment. Mean
plasma glucose levels (+/- SEM) at the end of the experiment were significantly
lower in the EFAD groups vs control groups in BALB/cByJ (P < 0.001), DBA/2J (P <
0.00001), and C57BL/6J (P = 0.012) mice. CsA supplementation increased the
severity of diabetes in LDS-treated BALB/cByJ mice (P < 0.0005); however, EFA
deficiency also prevented diabetes in CsA-supplemented BALB/cByJ mice.(ABSTRACT
TRUNCATED AT 250 WORDS)
Pancreas. 1995 Jul;11(1):26-37.
Essential fatty acid deficiency prevents autoimmune diabetes in nonobese diabetic
mice through a positive impact on antigen-presenting cells and Th2 lymphocytes.
Benhamou PY, Mullen Y, Clare-Salzler M, Sangkharat A, Benhamou C, Shevlin L, Go
Diabetes Research Center, UCLA School of Medicine 90024-7036, USA.
Protective effects of essential fatty acid deficiency (EFAD) on autoimmunity were
shown in rodents. Our goal was to investigate the mechanisms of EFAD effects on
autoimmune diabetes in nonobese diabetic (NOD) mice. Weanling female mice were
randomized between a control diet group and an EFAD diet group, and the
development of diabetes and immune response was determined over a 6-month period.
The cumulative incidence of diabetes was significantly reduced in the EFAD group
(20 vs 68.75% in the control group; p < 0.01), without affecting the insulitis
process. Splenocyte reactivity to phytohemagglutinin and anti-CD3 antibody was
significantly increased in EFAD-fed mice (p < 0.01). The EFAD group also
exhibited a dramatic increase in baseline (29-fold) and antigen-presenting cell
(APC)-stimulated (10-fold) T cell responses in syngeneic mixed leukocyte
reaction. These responses were associated with a marked increase in splenocyte
interleukin-4 (IL-4) production, a reduction in interferon-gamma production, and
a down-regulation of CD45RB isoform expression. Macrophages in the EFAD group
exerted a reduced suppressive effect on concanavalin A-induced splenocyte
proliferation and were found to release increased amounts of tumor necrosis
factor-alpha and IL-1 and reduced amounts of prostaglandin E2. These results
clearly demonstrate that EFAD prevents diabetes in NOD mice. The data suggest an
enhanced activity of Th2-like cells, as well as an effect on APC activity linked
to alteration in eicosanoid metabolism.
Prostaglandins Leukot Med. 1986 Aug;23(2-3):123-7.
Essential fatty acid deficiency: a new look at an old problem.
Lefkowith JB, Evers AS, Elliott WJ, Needleman P.
Essential fatty acid (EFA) deficiency is a useful tool to study the role of
arachidonate and its metabolites in various physiologic and pathologic states.
Recent studies have clarified the effects of EFA deficiency on membrane
arachidonate and its metabolites, and have demonstrated that 20:3(n-9) (which
accumulates in EFA deficiency) can be metabolized to a variety of eicosanoids.
EFA deficiency has been shown to exert an anti-inflammatory effect. The mechanism
of this effect may in part be mediated through a decrease in leukocyte
leukotriene formation. In contrast, studies using the novel fatty acid,
columbinic acid, have shown that the epidermal dysfunction seen in EFA deficiency
may be a function of linoleate and its lipoxygenase metabolites rather than of
arachidonate and the prostaglandins. Finally, it has recently been shown that EFA
deficiency potentiates the effects of volatile anesthetics. EFA deficiency may
thus provide a useful tool to investigate the molecular mechanism of these drugs.
J Lab Clin Med. 1981 Nov;98(5):764-75.
Effects of experimental diabetes on the essential fatty acid-deficient rat.
Riisom T, Johnson S, Hill EG, Holman RT.
An evaluation of the EFAD syndrome in rats rendered diabetic with either alloxan
or streptozotocin was performed. Diabetic rats fed an EFA-deficient diet for 7 or
13 weeks were less severely EFA-deficient than were nondiabetic rats fed
EFA-deficient diet, as judged by dermal symptoms or by biochemical parameters
such as the ratio of 20:3 omega 9/20:4 omega 6 (T/T ratio) and total fatty acids
derived from linoleic acid. The T/T ratios of liver PL of diabetic EFA-deficient
rats were lower than those of deficient control rats, and the ratios varied
inversely with the blood glucose concentrations. The product/precursor ratios,
arachidonic acid/linoleic acid, in liver PL were higher in diabetic deficient
rats than in deficient control rats. Analysis of liver and heart PLs revealed
higher arachidonic acid levels in the diabetic deficient rats than in the
EFA-deficient controls, perhaps because of different growth rates. The activities
of the delta 5, delta 6, and delta 9 desaturases were evaluated in liver
microsomal systems. The delta 9 desaturase was depressed in diabetic rats in
agreement with literature reports. The delta 6 desaturase, however, was slightly
increased. The relative levels of delta 5, delta 6 and delta 9 desaturation
products in liver and heart PLs did not parallel the measured desaturase
activities of liver microsomes.