When is a calorie not a calorie?

One ounce of raw almonds (about 23 nuts) contains:


6 grams protein

14 grams fat

6 grams carbohydrate

3.5 grams fiber

For a total of 163 calories per ounce.


(From the USDA Nutrient Database)


Calorie content of foods is determined by summing up the calories from each constituent: 1 gram of fat = 9 calories; 1 gram protein = 4 calories; 1 gram carbohydrate = 4 calories. Calorie content can also be directly measured using a device called a burn calorimeter, in which the amount of energy released from a specific food is measured by literally burning it and gauging precisely how much energy is released.


The problem with both of these methods is that it is assumed that all foods are digested with equal efficiency. That is, it assumes that a potato chip is as readily digested and absorbed as energy from table sugar, a pretzel, oatmeal, a piece of steak, or a handful of nuts. In real life, of course this is not true. Different foods are absorbed with varying efficiency.

For a long time I've suspected that some foods are very inefficiently absorbed. I've particularly suspected that raw nuts are relatively poorly absorbed and thus yield only a fraction of the calories ingested.

Among the studies recently reported at the Federation of the Association of Societies for Experimental Biology (FASEB) meetings I attended in San Diego this past week were several devoted to almonds.

One study, to my surprise, documented this phenomenon. In Manipulation of lipid bioaccessibility of almonds influences postprandial lipemia in healthy human subjects, it was determined that, of 100 calories ingested from the fat fraction of almonds, only about half was actually absorbed. The remaining half passed out in the stool. (They did this by collecting stool samples and comparing the fat composition after eating the different almonds prepartions. This is not discussed in the limited text of the abstract.) In addition, postprandial (after-eating) surges in triglycerides were much less with whole almonds compared to the oil separated from the nut (i.e., broken down into almond oil + defatted almond flour). The researchers attributed the difference to the inhibitory effects of the almond nut's "food matrix," or the structural properties of chewed foods.

Add to this the fact that, of 6 grams of carbohydrate per ounce of whole almonds, 3.5 grams are indigestible fibers. This means that 6 - 3.5 = 2.5 grams of digestible carbohydrates are present per ounce (assuming 100% release).

If we follow the reasoning that only about half the fat fraction of almonds are absorbed, and assume that the protein and carbohydrate (minus the indigestible fibers) are absorbed efficiently (100%), then we would re-calculate the calorie content of almonds to be 97 calories per ounce, or 40% less than calories calculated by composition or measured with a calorimeter.

If we were to assume that protein and carbohydrates were, like fats, inefficiently absorbed because of the effects of the food matrix, then one ounce of almonds yields 88 calories per ounce, or 46% less. This is, in fact, a likely scenario, since the food matrix is largely created by the cell wall and should impede digestive access to fat, protein, and carbohydrate equally.

My point? Almonds and other nuts at first appear to be calorically dense due to fat composition. However, this simplistic view of nuts is misleading because of the confounding effects of the food matrix. Stated differently: Whole foods yield less calories. And, judging by the postprandial triglyceride effects: Whole foods yield less undesirable effects, such as postprandial rises in triglycerides.

Some other observations with almonds included:

The effect of almonds on plasma lipids in persons with prediabetes This study confirmed the LDL-reducing and modest HDL-raising effects of almonds.

Almonds (Amygdalus communis L.) as a possible source of prebiotic functional food This curious observation suggests that almonds modify the bacterial flora of the intestinal tract in a positive way (like the cultures in yogurts).



Copyright 2008 William Davis, MD

Comments (8) -

  • Anna

    4/11/2008 1:46:00 AM |

    I often soak raw almonds (and other nuts) about 24 hours in filtered water with sea salt, then dry them a day or two at about 150°F in the oven.  I doubt the studies take this kind of "processing" into account, but any idea how that might change the absorption scenario?

  • Anonymous

    4/11/2008 4:27:00 PM |

    I'm glad you posted this.  A few women I know have been wary of eating many nuts for fear that fatty nuts would cause them to gain weight.  

    Really like the little bit on healthy gut flora caused by almonds too.  A healthy gut is important to me.

  • Peter

    4/11/2008 8:19:00 PM |

    I would certainly agree that it is possible to eat enormous quantities of nuts without absorbing all of their calories. I noticed I could remain weight stable while sedentary and eating 3,500 calories back when I used to eat large quantities of nuts. Even when well chewed a proportion of them end up you-know-where! Just flush...

    Peter

  • brian

    4/12/2008 4:22:00 PM |

    Dr. Davis, this is a great example. Thanks for posting it. I have a question on the 88 calories. I tried to work through the calculations to demonstrate this to one of my clients. I couldn't come up with 88 calories per ounce.

    I kept coming up with 80 and here's how. 50% calories from fat is 63. I used 50% for protein and digestible carbs - based on the info provided. For protein, that leaves 3 gms or 12 calories. For carbs, half of 2.5 gms is 1.25 gms, which equals 5 cals. These add up to 80 (63 + 12 + 5).

    I’m probably making a silly mistake but I’d like to come up with the same numbers when demonstrating this to clients – makes me feel kind of silly.

    Thanks again for the blog, I greatly appreciate the information you post.

    Brian

  • Anne

    4/13/2008 9:57:00 PM |

    Does pasteurization affect the positive effect almonds have on gut bacteria? Last year it became manditory that all California almonds must be pastuerized by one of these methods: fumigation with propylene oxide, blanching and oil roasting. This is to prevent salmonella. http://www.almondboard.com/Programs/content.cfm?ItemNumber=890&snItemNumber=450

    I eat a handful of almonds just about every day. I have wondered if there benefits have been compromised.

  • Anonymous

    4/14/2008 12:25:00 PM |

    The study was sponsored by The Californian Almond Association

  • Katherine

    10/21/2008 6:30:00 AM |

    In the era of the 64-oz. soda, the 1,200-calorie burger, food companies now produce enough each day for every American to consume 3,800 calories per day as compared to the 2,350 needed for survival. Not only adults but kids are also consuming far more calories than they can possibly use. http://www.phentermine-effects.com

  • buy jeans

    11/3/2010 6:53:22 PM |

    For a long time I've suspected that some foods are very inefficiently absorbed. I've particularly suspected that raw nuts are relatively poorly absorbed and thus yield only a fraction of the calories ingested.

Loading
More on ASTEROID

More on ASTEROID

Since we are on the topic of the ASTEROID trial and rosuvastatin, I'd make one more point before I start to sound like I'm plugging this drug (which I definitely am not).

In an informative Roundtable Discussion (open to subscribers to the American Journal of Cardiology; sorry) amongst Dr. Steve Nissen, principal investigator behind ASTEROID; and Drs. Vincent Friedewald, Christie Ballantyne, P. Shah, and William Roberts, Dr. Nissen made some interesting comments:


Dr. Shah: In ASTEROID, was the magnitude of atheroma volume change seen across different levels of LDL-C and HDL-C?

Dr. Nissen: No. There was no plaque regression seen in the 17 persons with LDL-Cs >/= 100 mg/dl, and there was little change in persons with LDL-Cs of 70 to 100 mg/dl. Only in persons with LDLs less than or equal to 70 mg/dl was there significant regression. The study was not powered to look for an HDL-C(which increased by 14.7%)-raising effect.



Interesting. In other words, ASTEROID, in a fairly internally consistent way, suggests that the lower the LDL is reduced, the more likely plaque regression is obtained. This is consistent with the Track Your Plaque experience, in which we've advocated reducing (calculated) LDL cholesterol to 60 mg/dl for the past several years.

Unfortunately, the message that the ASTEROID Trial sponsors, AstraZeneca, as well as the roundtable discussion panel (later in the discussion) try to make is that there is something magical about Crestor, that it yields benefits superior to other statin agents or other means of reducing LDL.

I disagree with this message. In the Track Your Plaque experience, we do aim for a similar LDL target. But we also employ a number of other strategies. We have also succeeded in regressing plaque without use of any statin drugs (though, admittedly, many people do require statin drugs to obtain LDLs in this range). We also witness magnitudes of reversal that often far exceed that seen in ASTEROID.

The Rountable Discussion is unfortunately tainted, as is the ASTEROID Trial itself, with deep drug industry financial involvement of the Roundtable participants. In fact, the discussion begins with a listing of the financial disclosures of the participants, a listing that occupies a full column of a two-column page. The potential biases of the participants doesn't necessarily invalidate the arguments, but to me suggests that participants are more likely to argue in favor of the sponsor's drug, or that participants were chosen because of these biases.

Why bother to even mention the ASTEROID Trial in a venue (the Heart Scan Blog, that is) that purports to seek unvarnished, unbiased truth in coronary plaque reversal? Because useful information can sometimes be found in unlikely places. Just like the four-year old child who blurts out an unexpected pearl of wisdom, so it can happen with the gobbledy-gook that emerges from the drug industry.

Every once in a while, they are worth paying attention to.

Comments (3) -

  • Naruwan

    4/19/2008 11:54:00 AM |

    Dr. Davis, your readers may be interested to listen to a recent BBC Radio 4 program which takes a close look at some impressive-sounding figures obtained from statin drug trials and shows how they are in fact not in the least bit impressive (e.g. a reduction in mortality from 3 in 600 people to 2 in 600 being hyped by statin manufacturers as being an over 33% reduction! Brings to mind Disraeli's adage about damned lies and statistics.

    The take home message is that statins appear to be hugely over-subscribed - the UK is gaining on the US in this regard - and statins are of no benefit for 99% of people taking them for primary prevention of heart disease.

    You can listen to the program online using the BBC Real Player at this link. Click Listen Live at the top right of the screen.

  • Naruwan

    4/19/2008 12:11:00 PM |

    My apologies, it appears that the BBC radio 4 program (The Investigation) about statins is no longer available. I think the show's transcript will be made available in due course.

    If anyone would like me to send them the mp3 file of the show (shhh, don't tell the BBC!), please leave a comment here. It's a worthwhile listen.

  • Anonymous

    8/22/2008 8:41:00 PM |

    I would like to know how long one can expect to lengthen their life by taking a statin How many actual years does it add to your life and what about the risk of cancer from taking statins over a 20-30 year period?

Loading