That New Volek (& Phinney) Study: Part III ~ Science or Advocacy?

LC Biomarker Bingo

OK ... before I begin on this, let me say this has languished in the draft hopper probably past it's usable publish date :-)  So first, I'll give a belated shout out to Kevin Klatt over at Nutrevolve blog for taking a close look at this study and "getting it out there" in rapid time!  Carbs, SFAs, and Circulating Fatty Acids.





Effects of Step-Wise Increases in Dietary Carbohydrate on Circulating Saturated Fatty Acids and Palmitoleic Acid in Adults with Metabolic Syndrome

I confess to skimming it b/c the last time he scooped me Kevin wrote so much of what was in my head I wasn't sure how to formulate the parts I still wanted to discuss without sounding repetitious.  For me, this study and its media blitz has hit a nerve and awoken my interest in a topic I had gotten a bit side tracked from.  It's almost a perfect ending to this year of what can only be described as utter insanity in the nutrition world.  

So, with that said, I won't be doing any further analysis of this study ... this will be the last post on this paper per se.  It's not worth the attention it received.  Instead it will provide a foil for discussing the rank dishonesty permeating the low carb "scientific" community.




For reference:  Part I The Journal Article & The Headlines , Part II ~ More on Journal Choice, Funding and Thoughts on Press Quotes

Previous Studies?  What Previous Studies?

When you read the first two sentences of the abstract of this study -- the part that would usually be under "Background" when formatted in parts -- you get the impression that this is somehow novel research.

Recent meta-analyses have found no association between heart disease and dietary saturated fat; however, higher proportions of plasma saturated fatty acids (SFA) predict greater risk for developing type-2 diabetes and heart disease. These observations suggest a disconnect between dietary saturated fat and plasma SFA, but few controlled feeding studies have specifically examined how varying saturated fat intake across a broad range affects circulating SFA levels.

When a research group has done prior related work, it is usually mentioned here, often  in the context of how the current study builds upon that work and/or seeks to clarify issues those prior studies left unresolved.    In an academic publication -- such as a thesis, or even a research review for a class term paper -- it would be obligatory, as well as a thorough review of what other researchers have done.   As it turns out, Volek's group had conducted not just one, but two prior studies looking at the fatty acid profiles of triglycerides (and cholesterol esters and phospholipids) while varying the saturated fat and/or carb content of the diet.  I outlined these studies in Part II.  Not only did they neglect to mention these in the abstract, but one comes up empty after perusing the 36 references cited in the publication!!   Meanwhile, one study from Volek's research was cited, this from 2005:   Modification of lipoproteins by very low-carbohydrate diets.   One would think that more recent studies (2008, 2010) looking at the same parameter -- fatty acid profiles in triglycerides etc. -- would have made the cut!  It would be difficult to believe that none of the various overlapping co-authors thought to include a mention of their prior related work (those authors in italics)?  After all, the results were rather "pro" low carb in general, and on point with the dietary SFA doesn't "accumulate" in the selected circulating lipids mantra.  

Study One, 2008:  (Authors: Cassandra E. Forsythe, Stephen D. Phinney, Maria Luz Fernandez, Erin E. Quann, Richard J. Wood, Doug M. Bibus, William J. Kraemer, Richard D. Feinman, Jeff S. Volek.) 

40 overweight men & women with atherogenic dyslipidemia, consuming roughly 2100-2300 cal/day at baseline were randomized to either a VLCKD or an LFD of ~1500 cal/day for 12 weeks.

Results pertinent to comparison to current study, values are weight % of total:

Study Two, 2010:  8 mildly obese men (BMI ~30 - two of whom were following a low carb diet at baseline) were assigned to 6 weeks each, in randomized crossover fashion, of weight stable low carb diets.   Each intervention phase was preceded by a 3-week run-in "standard LC diet" of 10% carb, 65% fat, 25% protein.  Of note, self report baseline calories were considerably below weight stable amounts during the intervention.

Results pertinent to comparison to current study, values are millimolar.

Unfortunately, differing units in the reporting make direct comparisons next to impossible.  Still, the question they are supposedly asking in the current study had already been answered.  Saturated fat in fasting triglycerides is not a function of the saturated fat content of the diet.

Let's Play What If ...

What IF the other studies were acknowledged?  Something to the tune of:

[hypothetical]  Our group has previously demonstrated that in the context of caloric deficit and weight loss, the saturated fat content of fasting triglycerides nominally decreased on both a low fat and low carb diet.  Meanwhile, palmitoleic acid concentrations were reduced markedly with the low carb diet but remained unchanged on the low fat diet.  In addition, our group studied two low carb high fat interventions in the weight stable state and demonstrated reductions in both triglyceride saturated fats and palmitoleic acid.  

Such a backgrounder could have been followed by:

In this study we sought to ...

But how to complete this?

... determine if the same effect is seen in lean individuals on LCHF vs. HCLF diets in calorie balance.

Nope ...

... repeat the experiment in obese who were not dyslipidemic ... in both calorie deficit and majntenance 

Nope ...

... seek to explain why palmitic acid and its desaturate -- palmitoleic acid -- levels do not seem to track proportionately?

Nope ... 

I could go on, but I think you get the point.  The fact that there was no mention of previous observations, and thus any desire expressed to add to the collective knowledge, opens the door for speculation.  This is supposed to be scientific exploration after all.  

The PLOS One Study Methodology

Methods:  Sixteen adults with metabolic syndrome (age 44.9±9.9 yr, BMI 37.9±6.3 kg/m2) were fed six 3-wk diets that progressively increased carbohydrate (from 47 to 346 g/day) with concomitant decreases in total and saturated fat.

Results:  Despite a distinct increase in saturated fat intake from baseline to the low-carbohydrate diet (46 to 84 g/day), and then a gradual decrease in saturated fat to 32 g/day at the highest carbohydrate phase, there were no significant changes in the proportion of total SFA in any plasma lipid fractions. Whereas plasma saturated fat remained relatively stable, the proportion of palmitoleic acid in plasma triglyceride and cholesteryl ester was significantly and uniformly reduced as carbohydrate intake decreased, and then gradually increased as dietary carbohydrate was re-introduced.

They took more obese individuals, and put them on a low carb diet BEFORE the study for a three week run in.  They then did the first phase of the diet for another three weeks.  Why?  Because they thought they knew what would happen.  In their second prior study they wrote:

Each dietary feeding period was 6 weeks in duration, based on previous research showing that the fatty acid composition of plasma PL stabilizes within 4–6 weeks of dietary change.

So let's establish baseline in the middle of that period, and hope for some further "improvements" in three more weeks.  Let's do this in weight loss so the LDL doesn't go up like it did in 2010, but in that study sat fat in the trigs did go down.  So we can show that increasing carbs will "cause" it to rise.   It is not a stretch to imagine that they hoped that by establishing a low baseline at what is essentially three weeks into the intervention,  come a few phases in, levels might have climbed above this artificially set value.   Folks, a "run in" phase in any nutritional study is supposed to be a period of stabilizing "usual" intake levels and/or stabilizing all subjects on a similar "normal" dietary intake pattern.  It is not supposed to be a period of adaptation.   This should have been a 21 week study, with the first 6 weeks on low carb.  Baseline should have been at ... well ... actual baseline ... before any dramatic changes in intake.

But let's take some more overweight people, so intake levels can be higher as the study progresses, so that we can see changes in a meaningless biomarker.  This is what they did.  What is also problematic is that they did not attempt to control for weight loss.  The weight loss trajectory is exactly what one might expect for imposing a mild calorie deficit.   Of course the usual suspects would presume this is the effect of adding carbs in on weight loss, when, indeed, it is simply the leveling off of weight at a reduced state.  Couple that with a strange progression of carbohydrate amounts ... I would note that carbs were essentially doubled in the last six weeks with continued, albeit slowed, fat loss (and net weight loss).

And yet, Volek and company struck out on the intended outcome, which was to show that carbs increased saturated fat in the triglycerides.    In the end, they were forced to conclude:

Conclusion:  The results show that dietary and plasma saturated fat are not related, and that increasing dietary carbohydrate across a range of intakes promotes incremental increases in plasma palmitoleic acid, a biomarker consistently associated with adverse health outcomes.

Already Advocating ... Part II Revisited

In May 2012, a paper authored by Jeff Volek, Brittanie Volk (lead author on current paper), and Stephen Phinney was published:  The twisted tale of saturated fat.  Now THIS paper, more of an editorial in Lipid Technology journal than any sort of review, did reference the two prior works of Volek's research group.   But it also wasn't mentioned in current 2014 paper.

Given that this current study was only 18 weeks (under 5 months, I know you can do the math but it never hurts to do it anyway) ... it seems unlikely it was started before the above salespitch for carbohydrate restriction and the Volek & Phinney "Nutritional Ketosis" (hereafter NuttyK).  Point being, it was designed to achieve a particular result that could be used to splash just the sort of headlines we've seen across news media and the internet.

This dovetails nicely into recent study in the British Medical Journal, The association between exaggeration in health related science news and academic press releases: retrospective observational study , that concluded:

Exaggeration in news is strongly associated with exaggeration in press releases. Improving the accuracy of academic press releases could represent a key opportunity for reducing misleading health related news.

The gist of this study as related in popular media, is that quite often, erroneous science reporting can be traced back to the scientists themselves.   It is quite ironic, that this is reported in a journal that has commissioned articles by the evermore imaginative Gary Taubes, and seemingly collaborates with authors on when and how to best release their "groundbreaking work" in their journal.  For the latter, I speak of Aseem Malhotra's article, for which the pre-publication exchanges are available here.   Most recently they published this: Are some diets “mass murder”?   A spoof?   The journals letting the journalists off the hook for their part in all of this.  It's sickening.  Is this really what "peer review" has come to?    

And yet, exaggerated claims, such as those being made in this latest Volek study, can be traced to the press release.  Most of the reports quoted liberally from it verbatim.   The press release quoted Volek.  Nobody bothered to even read the study, it would appear, and some reports had palmitoleic acid (a MUFA) designated as saturated fat.  

Biomarker Bingo ...

I just had this vision of Dr. Mary Vernon in prison playing LC Biomarker Bingo with her Fred Hahn troll doll and prison "Minister" Jimmy Moore doing the calling ...  a la this episode of Rosanne.   

I don't recall who it was in comments here, I want to say Kade Storm, who said something to the effect of how LC is unique as a dietary philosophy in that it is the only one that requires explaining away or ignoring risk factors.  You see it all the time.

Let's just keep calling out random things and hope to get a win?  Large fluffy LDL?  Sorry, particle number, not size is what is important (despite the Attia circus side show on the Dr. Oz show).   HDL?  Sorry, Dr. Lipid himself has said repeatedly that raising HDL (past restoring normal levels) has never been shown to improve outcomes.   Triglycerides?  De novo lipogenesis is simply NOT responsible for this phenomenon per se, nor does it contribute much to the fatty acid content of triglycerides according to radiolabel tracer studies.  Furthermore, high triglycerides implicated as a biomarker, are truly high:  in the few to several hundreds, not between 100 and 150 mg/dL.  Levels in that range indicate things may be going south,  but the Inuit!  Oh wait, I mean the Kitivan!!  Remember them?  HCLF poster culture.

So Volek is putting out VERY bad science here.  For one, free fatty acids (FFA or NEFA) are ignored entirely.  Palmitoleic acid is somewhat of an enigma, but Volek makes no effort to identify the source of this fatty acid.    Even if it could be attributed to DNL, the numbers just don't add up in terms of significant amounts.  Isn't it about time to at least hold scientists accountable for basic calculations?   He is solidly in the employ of various entities with financial interests in promoting low carb diets.  In  addition, he has significant personal financial (not to mention notoriety) interests.  Yes, yes, others do it, no doubt.  Show me where the scientists or data expose the bias and I'm onboard criticizing that too.  But explain THIS.  This cannot be explained away by "everyone does it".

I'm going to leave you with this flashback -- I have at least one or two more on the topic -- Partial inhibition of adipose tissue lipolysis improves glucose metabolism and insulin sensitivity without alteration of fat mass.  This post was titled after the study discussed therein.  Full disclosure:  Big Pharma contributed some funding, it was published in PLOS Biology (more than twice the publication fee, but fewer than 1% of the number of articles published vs. PLOS One).

... reduction in WAT lipolysis reshapes FA fluxes without increase of fat mass and improves glucose metabolism through cell-autonomous induction of fat cell de novo lipogenesis, which contributes to improved insulin sensitivity.

 I hope to revisit palmitoleic acid in the near future.