Tuesday, February 4, 2014

16 Weeks on High Fat Diet - Weight Loss, Fat Loss, Lowered Inflammation. But is 50% Fat, 30% Carbs Really "High Fat"? Plus: SFA, MUFA & PUFA (n-6) - Doesn't Make a Difference

This is one of the studies, where you have to look at the full-text before you celebrate or curse. Otherwise you run the risk of missing the actual research interest of the scientists and the fact that this nutritional intervention worked, although it was not particularly low carb.
It has been a while since the last SuppVersity article on high fat diets - specifically if we are talking about benefits, by the way. That being said, you may already have seen quotes from the slightly misleading abstract of a recent paper in the scientific journal Metabolism which could have led you to in the "magic" of low carb dieting more than ever (Silver. 2014). The fact that the author's name is "Silver" and not gold, however, does already suggest that the questionable long-term effects of the "balanced high fat diet" the 245 pre-menopausal obese women, who met the inclusion criteria of being 21-50 years old, and weight stable at a BMI of 30-34kg/m² for over three months, were fed is not the only relevant information you can't know if you don't have access to the full-text of the study.

Beneficial, but not magical 

As I hope to be able to show in a closer analysis of the study, he provision of a diet with a high fat content is beneficial for premenpausal obese women. It is yet not the magic solution to all their health and physique problems.
  • Figure 1: Macronutrient composition of the "high fat diet" (Silver. 2014)
    Caveat #1: This high fat diet, is not low in carbs -- When you're reading the words "high fat diet", you will either think of the "fake" high fat diets (HFD) in rodent trials with carbohydrate contents of up to 40% or an (almost) Atkins-type diet with a high fat, a moderate protein and a low / very-low carbohydrate content.

    It goes without saying that you'd expect the former to be unhealthy - it does after all have a pretty similar macro-nutrient composition as the previously cited high fat diets in rodent studies.

    Consequently, you would probably not believe that the latter, i.e. a diet with max. 10% of the total energy from fat would produce results as they are listed in the "results"-section of the abstract to the paper at hand:
    "Significant improvements occurred in fat oxidation rate (↑6%), body composition (%fat: ↓2.5 ± 2.1%; %lean: ↑2.5 ± 2.1%), inflammation (↓ IL-1α, IL-1β, 1L-12, Il-17, IFNγ, TNFα, TNFβ) and vascular function (↓BP, ↓PAI-1, ↑tPA activity)."
    If I am now telling you that the exact macro composition of the diet at hand had more resemblance with the classic HFD rodent chow (see Figure 1) , than with what you'd probably expect to see, when you hear about "a balanced high fat diet" in the title of a human study, this is hopefully not going to shatter your worldview altogether. 
  • Caveat #2: The high fat intake led to a significant decrease in energy intake -- Despite the presence of a pretty high amount of dietary carbs in the diet, the satiety effect of the diet (or the absence of the foods that made the women obese in the first place) led to a significantly reduced energy intake of ~10%, or 200kcal.
Fat intake or energy restriction? If we make the likely assumption that the women under-reported their baseline energy intake by only 5% (cf. figure on the left - Heitmann. 1995), the following calculation would yield a predicted fat loss of exactly 4.8kg. This would mean that the macronutrient composition had - on average - no direct influence on the weight loss success.
  • It goes without saying that this is not a reliable calculation, but if we still do the math, this would amount to a total energy deficit of 22,400kcal over the whole study period and should result in a fat loss of 3.2kg. In practice, the ladies lost 33% more body fat (4.8kg on average), however - an unrealiable, but at least significant indicator that the reduction in energy intake is not the only, but still a significant contributor to both the fat loss and health effects of the 16-week study intervention.
In spite of these caveats, the study at hand provides very intriguing data and compelling evidence that it is not fat in general that's making us fat.

Something you may not have read in the short abstract-based newsposts in the blogopshere...

...is that the important information that the actual aim of the present study was to test the hypothesis that body composition, inflammation and vascular function in obese premenopausal women would improve with a high fat diet (HFD) when - and this is where the magic is - "the type of fat is balanced as 1/3 SFA, 1/3 MUFA and 1/3 PUFA." (Silver. 2014).

In that, Silva and her colleagues from Vanderbilt University, the Northwestern University and the Tennessee Valley Healthcare System hypothesized that...
"[...]while consuming the balanced HFD, improvements in body composition, inflammation and vascular function would be greater when supplemented with 18C fatty acids - in proportion to degree of 18C fatty acid unsaturation." (Silva. 2014)
In other words, the scientists expected to see the exact opposite of what the average low carber would propagate, i.e. greater improvements in body composition, inflammation and vascular function with a higher PUFA (omega-6, of all!) content of the diet.

To standardize the diets, the fat intake was modulated by the provision of  9 g/d of encapsulated pure stearate (18:0), oleate (18:1), linoleate (18:2) or a identically looking placebo supplement that were ingested on top of the high fat diet with the previously mentioned ratio of 50% fat,  30% carbohydrate and 20% protein. To further standardize fat  intakes, all sources of fats including oils, spreads, nuts and seeds were provided weekly in pre-portioned containers.
Figure 2: Fat intake (in % of total fat) in the four groups of this 16-week dietary intervention (Silver. 2014)
As the data in Figure 2 confirms, the scientists managed to single out the ratio of saturated to mono- and poly-unsaturated fats as their main experimental parameter, successfully. The following observations, do thus inform us about the advantages and disadvantages of three different ratios of dietary fat intake - yet not a bout the effects of a "high fat", let alone a "high fat low carb" diet, as you may have read it elsewhere on the Internet, as of late.
Figure 3: Changes in body composition (% of baseline) after 16 weeks on different "high fat" diets (Silver, 2014)
As you can see in Figure 3 the changes in body composition were not significantly different, if anything there was a minor advantage for the placebo diet.

The only inter-group differences that did achieve statistical significance were the reduction in interferon γ (↓74%) in the HFD+placebo, and the HFD+stearate diets and the significant effect of the 
HFD+linoleate diet on the PAI-1 (↓31%) levels. The additional changes in vascular function as evidenced by an 8 mmHg reduction in systolic and a 6 mmHg reduction in diastolic blood pressure, as well as the minimal changes in the "suprisingly low" changes of the primary arterial compliance parameters, ie. the small and large artery elasticity index (SAEI/LAEI ml/mm Hg x 100) and the similarly disappointing improvements in flow-mediated dilatation (FMD) didn't show significant inter-group differences either.
Fatty acid ratio as a modulator of inflammation: The fact that the provision of supplemental stereate, much contrary to that of palmitic acid, another saturated fatty acid led to a significant reduction in IFNƴ, a driver of macrophage activity and endothelial cell inflammation (Kofler. 2005), is as notable as the the "curious" (Silver. 2014) fact that the additional oleate did not produce the reductions in IL-1α and TNFβ, the scientists had expected based on previous studies, where oleate replaced some of the palmitic acid in the diet (Coll. 2008).
So what's the verdict then? Eating more fat, if it is done in a controlled manner and from non-self-selected foods, can help obese pre-menopausal women lose weight - that's not debatable.

What is debatable, though, is whether sticking to a prudent low(er) fat diet wouldn't have elicited the same, in terms of the disappointing improvements in the main parameters of arterial compliance which have previously been reported to be compromised on otherwise "healthy" high fat diets (Keogh. 2005).

To answer this question is beyond the scope of the study at hand, which focused on the main effect of allegedly mild differences in the fatty acid composition of the diets, which affected only two of two dozen of health parameters, i.e. IFNƴ and PAI-1, which contributes to the development of  atherosclerosis (Festa. 1999).
  • Coll, Teresa, et al. "Oleate reverses palmitate-induced insulin resistance and inflammation in skeletal muscle cells." Journal of Biological Chemistry 283.17 (2008): 11107-11116. 
  • Festa, Andreas, et al. "Relative contribution of insulin and its precursors to fibrinogen and PAI-1 in a large population with different states of glucose tolerance the insulin resistance atherosclerosis study (IRAS)." Arteriosclerosis, thrombosis, and vascular biology 19.3 (1999): 562-568.
  • Heitmann, Berit Lilienthal, and Lauren Lissner. "Dietary underreporting by obese individuals--is it specific or non-specific?." Bmj 311.7011 (1995): 986-989. 
  • Keogh, Jennifer B., et al. "Flow-Mediated Dilatation Is Impaired by a High–Saturated Fat Diet but Not by a High-Carbohydrate Diet." Arteriosclerosis, thrombosis, and vascular biology 25.6 (2005): 1274-1279.
  • Silver, Heidi J., et al. "Consuming a balanced high fat diet for 16 weeks improves body composition, inflammation and vascular function parameters in obese premenopausal women." Metabolism (2014).