Gretchen's postprandial diet experiment II

I previously posted Gretchen's postprandial diet experiment, in which she consumed a low-fat diet for a day, followed by a low-carbohydrate diet for a day. Grethen monitored blood glucose and triglycerides with fingerstick checks. (Blood glucose can be checked on any widely available glucose monitor; triglycerides can be monitored with the Cardiochek device.)

Let's now discuss what happened.

On the low-carb, high-fat day, there was an initial surge in triglycerides to 250 mg/dl late morning, followed by a secondary peak several hours following dinner. Because fat is mostly triglycerides, Gretchen's high-fat (sausage, bacon, butter, whole-fat yogurt) breakfast provided a large quantity of triglycerides that needed to be absorbed. This generally occurs over approximately 6 hours, varying depending on body weight, how accustomed you are to fat, activity level during the day, the kind of fat in the meal. The high content of saturated fat in Gretchen's high-fat breakfast likely caused the somewhat slower drop in triglycerides over approximately 7 1/2 hours.

As Gretchen herself had noted, triglycerides the following day were lower, a typical low-carb response. Blood sugar throughout showed only minor variation, with only small postprandial increases.

Thus, Gretchen experienced what we'd expect with a low-carb, high-fat diet: an initial high surge in triglycerides, followed by a decline in fasting levels, while blood sugar shows a normal contour.

Now, the more confusing low-fat experience:

Blood glucose makes a striking peak at 200 mg/dl after the low-fat breakfast of pasta and rice, in contrast to the low-carb breakfast. Triglycerides behaved very differently from the low-carb experiment: While there was no initial postprandial surge, there was a late surge developing 6-24 hours later. The late surge continued into the next day, with fasting levels the following morning (210 mg/dl) exceeding the starting triglyceride level (60 mg/dl).

The one potentially confusing aspect of all this is Gretchen's late rise in triglycerides on the low-fat diet. This phenomenon is due to something called de novo lipogenesis, or the liver's conversion of carbohydrates to triglycerides that occurs when an excessive carbohydrate load comes through diet. Because the human body cannot store anything beyond a minor quantity of carbohydrates (as glucose and glycogen), carbohydrates are converted to fats.

Another factor causing the late triglyceride increase is insulin resistance, given the high blood sugar response. When insulin resistance is present, the activity of the enzyme, lipoprotein lipase, is reduced. Less lipoprotein lipase activity allows slower VLDL degradation, allowing VLDL (and thereby triglycerides contained in VLDL) to "stack up" in the blood. Thus, the higher triglycerides late after eating and into the next morning.

One issue to be aware of: Acute responses can differ from chronic responses. In other words, had Gretchen had the luxury (and time and money) to conduct the experiment over, say, 4 weeks, rather than a single day, there would be somewhat different responses. The best data on this come from Dr. Jeff Volek of the University of Connecticut, in which 4 weeks of low-carbohydrate eating modify fasting and postprandial responses over time.

Several conclusions can be made from Gretchen's experience:

1) Low-carb, high-fat acutely generates extravagant postprandial triglyceride responses.
2) Low-fat causes a late triglyceride surge and higher fasting triglycerides.
3) Low-fat leads to high blood sugars and, by implication, diabetes.

Both the low-carb and the low-fat responses are undesirable, both leading to increased risk for heart disease. Which is worse? I believe that low-fat is more destructive, since it leads over time to both high triglycerides and diabetes, while low-carb/high-fat only leads to postprandial triglyceride surges, at least acutely.

How to best balance the responses to reduce risk for heart disease? That's a discussion for future.

Again, my thanks to Gretchen and the substantial amount of effort that went into generating these numbers. More of Gretchens' own writing can be found on her blogs:

Comments (37) -

  • Pythonic Avocado

    1/3/2010 3:53:05 PM |

    Why do you say the low-carb response  is leading to increased risk for heart disease?

  • Stan (Heretic)

    1/3/2010 5:34:56 PM |

    Re: Both the low-carb and the low-fat responses are undesirable, both leading to increased risk for heart disease.

    With due respect, I would disagree that BOTH are undesirable, but I agree that the one caused by the low fat high carb diet (only) is indeed dangerous!  

    It was never proven that high triglycerides alone cause heart disease.  The available trials such as Framingham (1)  may have shown some correlation with the total cholesterol but correlation is not causuation, especially that the statistics do not distinguish between various diets. Furthermore, Framingham's correlation is restricted to men only (not women) and only for the age group 30-55.  For older men and for women of all ages the correlation becomes insignificant or reversed.

    Given the above data, I think the most plausible interpretation leads me to a conclusion that the most likely direct cause of atherosclerotic cardiovascular disease is excessive blood glucose with hyperinsulinemia  (see the following papers, and Stout's papers(2) ).

    Hyperglycemia and hyperinsulinemia also happen to coincide with elevated TG and LDL but those are coincidental markers of metabolic syndrome induced by the common high carb (high sugar) diets rather than causing heart disease.  That I believe has nothing to do with dietary fat.

    Stan (Heretic)


    1) JAMA. 2004 May 12;291(18):2243-52. Drug treatment of hyperlipidemia in women. Walsh JM, Pignone M.

    R.W. STOUT, Lancet Sept 28, 1968, p702.


    R.W. STOUT, Lancet Aug 30, 1969, p467.

  • Nigel Kinbrum BSc(Hons)Eng

    1/3/2010 5:41:31 PM |

    As serum TG's with the HF meal are lower at the end of the day than at the beginning, does this suggest that successive days of HF meals produce progressively lower & lower TG's?

  • Anonymous

    1/3/2010 6:02:56 PM |

    I would love to hear about how to balance the risk between these extremes!  Wow, nice post.

  • Dr. William Davis

    1/3/2010 10:11:35 PM |

    Pythonic and Stan--

    It's not that triglycerides per se are atherogenic, but the POSTPRANDIAL PARTICLES that triglycerides represent are atherogenic.

    In other words, high triglycerides signals extravagant chylomicron remnants and VLDL, both of which are atherogenic.

    In the ongoing debate over what constitutes a healthy or unhealthy diet, the entire issue of postprandial patterns has been ignored. Yet much of heart disease develops IN THE POSTPRANDIAL PERIOD.

  • shel

    1/3/2010 10:55:38 PM |

    i eat copious amounts of fat with positive effects. i'm a layman, but "postprandial" or not, i'm having a hard time accepting this, as the evidence i've read seems to contradict what you're saying regarding overall benefits.

    why did no one pick up on the "postprandial particle" issue until now?

    will this be a new controversy sweeping through the paleo/low carb blogging community now? ;)

  • Dr. William Davis

    1/4/2010 2:15:45 AM |

    Anyone desiring a full accounting of the hundreds of studies documenting this effect will need to refer to the Track Your Plaque Special Report, Postprandial Lipoproteins: The Storm After the Quiet.

    This literature is, unfortunately, relatively difficult to understand. But just because nobody else has incorporated these findings into diet advice doesn't mean it isn't important.

    Keep in mind that most dietary advice is NOT based on observation of postprandial phenomena.

  • Eric

    1/4/2010 5:53:50 AM |

    I had a stroke last month. (Very minor, I went immediately back to work.)  As a 35 year old non-smoker/non-drinker who generally ate low carb and avoided sweets before the stroke (and had been exercising regularly for 6 weeks), I'm baffled as to why my triglycerides are between 600 and 900.  I'm now on Lovaza.  Against my doctor's advice, I decide to do a paleo type diet.  We'll see how the lipid panel does at the end of this month, but so far avoiding wheat and grains has generally made me feel better.  I've lost some weight as a bonus.

    The doctor is blaming my triglycerides for the stroke, and just calling it hereditary.  It'd be nice to know just what gene I have that causes strokes!  No blood clotting disorders, no diabetes, no pancreaitis, and no hole in my heart either.

  • Alen Kcatic

    1/4/2010 10:58:09 AM |

    First let's take a brief overall look at heart disease because you need to combine two other major lifestyle habits with diet to truly make a difference in reversing or preventing heart disease.

    But here's the good news. You can prevent or reverse heart disease by following care

    Avoid tobacco
    Be more active and walk 30 minutes
    Choose healthier food, including more fiber, less saturated fat, and less salt.

  • Peter

    1/4/2010 12:38:21 PM |

    If it were as simple as low fat leads to diabetes, the Japanese who ate the traditional low fat high rice diet would have had extremely high levels of diabetes, but they hardly had any. I would be more inclined to wonder if our diabetes epidemic is due specifically to flour and sugar rather than low fat in general.

  • shel

    1/4/2010 4:27:36 PM |

    ~Dr Davis, i wonder if this finding is going to point toward advocation of "grazing", rather than two or three meals per day (in a low carb context).

  • kris

    1/4/2010 6:04:12 PM |

    Dr. Davis,
    Happy new year.
    here is the link to videos about FDA and drug giants. An eye opener. just in case you have not looked at it. total of 8 videos.

  • StephenB

    1/4/2010 7:16:32 PM |

    Dr. Davis, the abstract of the study you linked read: "Ten men consumed a low-carbohydrate diet rich in monounsaturated fat (MUFA) and supplemented with n-3 fatty acids for eight weeks."

    As you know, there are multiple low carbohydrate diets. This particular study did not examine a high saturated fat diet. I would love to see the result of chronic dieting featuring quality saturated fats like tallow, lard, butter, and coconut milk and avoiding hydrolyzed fats.

  • mark

    1/4/2010 7:19:28 PM |

    Dr. Davis wrote:
    "The one potentially confusing aspect of all this is Gretchen's late rise in triglycerides on the low-fat diet. This phenomenon is due to something called de novo lipogenesis, or the liver's conversion of carbohydrates to triglycerides that occurs when an excessive carbohydrate load comes through diet. Because the human body cannot store anything beyond a minor quantity of carbohydrates (as glucose and glycogen), carbohydrates are converted to fats."

    We don't see triglycerides being converted to glucose. It's a one way street.

    Would having increased carbohydrayte storage space in the body be preferable to storing triglyceride? We have an example of that in hyperglycemia. That's storage in the bloodstream. But the body works quickly to CORRECT that.

    So it's not that the body lacks glucose storage and triglyceride is the bad alternative. Not that at all. The body works hard to push glucose out of blood storage and into triglyceride in fat cells. That's a good thing.

    Postprandial high triglycerides from a high fat diet is a marker of fat intake. Postprandial low triglyceride on a high carb diet is a marker of carbohydrate metabolism. The later increase in triglyceride is the corrective process.

    It's hard to make the case that triglyceride is itself bad when it's one of the body's innate responses to the bad hyperglycemia. If triglyceride is bad, then the body is stuck between a rock and a hard place. It's win-lose, so the low fat diet is worse than the low carb one.

    Given my experience with dieting, I would favour low carb over anything balanced in the way of fat/carb. From dietary intervention trials, I'm unconvinced that high fat is worse (or much better) than a mixed blend of carb and fat from a mortality perspective. But from experience, I favour low carb for general sense of well-being.

    Stan wrote:
    "Hyperglycemia and hyperinsulinemia also happen to coincide with elevated TG and LDL but those are coincidental markers of metabolic syndrome induced by the common high carb (high sugar) diets rather than causing heart disease. That I believe has nothing to do with dietary fat."

    It's tough to isolate lipids as causal as opposed to effects of diet. Smoking and fructose lead to increased LDL. So in this case, high LDL is really a symptom smoking and fructose intake. The latter two likely being causal for anything related to your health. 8 egg yolks a day on a high fat diet and your LDL is a symptom of that. And some people have normal LDL on that even.


  • Dr. William Davis

    1/4/2010 11:33:43 PM |

    Hi, Shel--

    No, absolutely not.

    Quite the opposite: Given what happens after eating, grazing is a destructive practice that likely increases risk for heart disease.

    See the previous Heart Scan Blog post:

  • shel

    1/4/2010 11:52:20 PM |

    ~hi Dr Davis.

    i agree. and i do much better when i eat two meals within an eight hour window.

    my focus has always been on keeping blood glucose low (i'm not diabetic, but use a glucose meter for my own curiosity), so was a bit floored by the thought that i have to watch every postprandial spike!, what's left? huge salads and skinless chicken breasts? ;)

  • vin

    1/5/2010 9:33:36 AM |

    It is pure and simple observation of two parameters after eating a low carb and low fat meal. Nothing more that that.

    After all most of us eat more than just carbs and fat: there are vitamins, enzymes, minerals, fiber and hundreds of other nutrients. I am certain that they more than compensate any damage that glucose or triglycerides can cause to your arteries.

  • Dr. William Davis

    1/5/2010 4:39:02 PM |

    Sorry, Vin.

    You're kidding yourself if you belief that.

    It reminds me of the people I meet who take a list of supplements 30 items long (though lacking the most crucial like vitamin D) prior to their bypass or heart attack. That's called magical thinking.

  • Kurt

    1/5/2010 9:27:14 PM |

    Both diets seem to be extremes, whereas many of us are trying to eat a balanced diet of vegetables, lean meats, nuts, and some legumes and whole grains - call it moderate fat and moderate carbohydrate - but focusing on heart-healthy foods. I'd like to see postprandial data on that.

  • O

    1/5/2010 10:39:05 PM |

    I have been eating a primal low-carb diet for almost 2 years and feel great.  My fasting blood work is : trig = 40, HDL=88, LDL=114 (calculated), total chol=214, testosterone=606.  My heart scan score is 0.  I am physically active muscular male with 4 intense weights + some cardio workouts 2 hours each.  My bodyfat % is about 8-10% (I can see a clear 6-pack), age=43, height=5'7", weight=160 lbs.

    I have made an analysis of my daily intake in a spreadsheet.  My diet on workout days is 3000 kcal, of which 50% fat (167g, out of which 60g saturated), 20% carbs (150g), and 30% protein (200g).  Half the carbs are timed post-workout (workout shake, followed by dinner of meat + sweet potato).  On a non-workout days, I do not have a shake nor sweet potatoes, so the carbs drop to 10% (70g).

    Given the amount of fats I take every day, I am rather alarmed by the postprandial triglycerides.  My breakfast, in particular, has 57g of fat which will cause probably a substantial postprandial triglycerides.  Breaking up my food intake into many smaller meals doesn't seem to be a good thing.   We don't want increasing carbs or protein at expense of fats either.  Therefore, what is the solution here?

  • Dr. William Davis

    1/5/2010 11:21:32 PM |


    I am afraid there's no quick answer. That question is answered in an exhaustive report on the Track Your Plaque website.

    Alternatively, you could conduct your own do-it-yourself postprandial triglyceride test.

  • Anonymous

    1/6/2010 3:17:23 PM |

    This doesn't detract from any of the points you're making about postprandial triglycerides--but it looks like you're reading the chart from the wrong side here for triglycerides, from the left instead of from the right.

  • Catherine

    1/6/2010 4:16:40 PM |

    I have been experimenting with a gluten-free, low carb, low sugar diet for 5 months and my LDL just shot UP from 220 to 230 and my HDL went DOWN a little from 66 to 61. (tryglicerides and CRP are excellent). This is opposite what's supposed to happen. Serum D level is good at 54.

    Can someone please tell me the name of the test to request from my doctor to tell if I have the small evil-type LDL or the big fluffy okay-type LDL?

    Thanks for your help,
    Warmly, Catherine

  • Lucy

    1/6/2010 4:33:54 PM |

    These results seem completely contradictory to the way Dr. Eades described the breakdown of saturated fat in his blog "The blood samples were taken two hours after the meal.  Dietary carbohydrate is absorbed directly into the blood and makes a pass through the liver where it stimulates the production of triglycerides, the fat you see in the blood.  Fat, especially long-chain saturated fat digests very slowly, and doesn’t reach the blood until much later than the two hour mark.  While carbs go directly into the blood, fats take a different route."

    Why was there a triglyceride spike after a high fat meal, but not a high carb one?  Were the fats Gretchen consumed not saturated?  It can't be both ways, which metabolic pathway is correct?

  • Catherine

    1/6/2010 9:08:44 PM |

    Oops, sorry i made a mistake.---My LDL went up from 120 to 130 (not 220 to 230)

  • vin

    1/7/2010 11:16:29 AM |

    Thanks for your comment Dr. Davis but for the first time you sound just like my cardiologist. He does not believe in reversal and thinks it is all down to one's genes and there is nothing one can do to change that. Well I think differently having postponed bypass surgery seven years ago. I will continue with 'magical thinking'. Its nice, you should try it sometime.

  • Anonymous

    1/7/2010 4:29:08 PM |

    To eat one extreme one day (low fat) and then eat another extreme (low carb, high fat) the next is a sure-fire way to get whacky blood/lipid results. Also, lipid levels can fluctuate more than 10% within a given day under normal circumstances. Alternating eating radically different extremes in terms of diet is anything but normal in my view.  Stress, exercise, or even one's sex can cause also shifts.  Trying to prevent every potential/alleged problem with postprandial lipids seems like a sure-fire way to increase stress which is also damaging to the heart and body.  I doubt paleo manwoman worried about the spike in his/her triglycerides from gorging on the fat from fatty meat meals.  I think one can micromanage one's self or rather labs, lipids etc to death...

    PS: Forgot to mention lab error.  Before drawing any conclusions from any lab results, have them repeated multiple times and at different labs.  My husband has to communicate with lab techs for his job and was horrified when one kept referring to "esterified" as "stir-fried"! And this was the one of the largest labs in the US...

  • Jim Purdy

    1/18/2010 4:38:41 AM |

    Doctor Davis, have you seen the new PR campaign from manufactured "food" giant Unilever to "Ban butter to save thousands of lives."

    Unilever is the company behind many fake foods, including fake butters Country Crock and  I Can't Believe It's Not Butter!

    I've posted a little about this ban-butter campaign on my blog at

  • shoby

    1/28/2010 3:25:55 PM |

    I also have a blog about the diet we can share experiences and exchange links.
    This blog

  • Fran

    2/1/2010 5:51:52 AM |

    "How to best balance the responses to reduce risk for heart disease? That's a discussion for future."

    Please tackle this discussion soon.

  • ET

    2/26/2010 6:52:28 PM |

    I also keep a spreadsheet that details all the different fatty acids and such for every meal.  I've also had three lipid tests this year which were not fasting.  Two were in the afternoon, eight hours after a five-egg omelet with coconut oil, cheese and bacon; 2 hours later I had eight oz of full-fat greek strained yogurt which adds up to over 120 g of fat 6 to 8 hours before the test.  Add another 40 g of fat 2-hours pre-test and according to your theory, my triglycerides should be high.  On each occasion they were 91.  This is on a low-carb (<70g/day) diet.

    There's more to postprandial triglyceride metabolism than is covered here.  My next test will be a non-fasting NMR lipoprotein analysis.  Should be interesting how that stacks up against a fasting NMR test.

  • A. Lanine Pro

    6/21/2010 8:55:09 AM |

    Well I think differently having postponed bypass surgery seven years ago. I will continue with 'magical thinking'. Its nice, you should try it sometime.

  • sharon

    6/30/2010 4:28:54 AM |

    Thanks for the information and looking forward to reading more.

  • javieth

    8/24/2010 5:44:47 AM |

    I liked this blog specially because i love to know about diet or exercises because i am trying to be in shape. I wish to complete my exercises program and get the better result, lose weight and be beautiful for conquer a boy.

    buy viagra

  • brenda

    9/14/2010 10:12:09 AM |

    Excellent stuff with wonderful information! I'm new here and loving the post! Thanks for sharing this great info!


  • buy jeans

    11/2/2010 7:06:16 PM |

    Buy jeans from our clothing store online!