Eric Helms of 3dmj:
Hey folks! So I've been in New Zealand since August 2012, a year prior to that I started an extensive lit review on protein. Primarily on hypocaloric, resistance training individuals, who are not obese, with resistance training experience. There's ~8-9 total studies out there so far that look at these conditions (depending on inclusion criteria) and 6 if we're talking studies that report LBM and bodyfat (some just give bodyweight and NBAL, etc.).
First part of my masters (which I submit this week literally) was this systematic review. Alan was actually one of the few I trusted to give it a look over in it's very initial stages, and now it's in the final stages of review at IJSNEM, it's been through 2 rounds of revisions, updates etc., and will be in press soon. The review is as systematic and non-narrative as you can get considering the limited data available and I was able to present recommendations relative to LBM vs weight which is important. A huge issue with available data is not only the methodology of nitrogen balance as Alan pointed out but also that 90% of the studies looking at protein intake give us intakes as g/unit of TOTAL bodyweight and were done on overweight and obese populations. This drives the perception of the "right" amount of protein down as the optimal intake is likely relative to LBM.
So the review is enlightening, puts a bit more structure and hard evidence to the discussion, when it comes out on pubmed you can believe I'll be spamming it via social media
Also I finished my own study on July 29th of this year, I recently (today I **** you not) finished analyzing the data, I couldn't report what was going on during because I was still blinded.
Before I get into it let me put my research, and truly all research in perspective: my study at best will only carve out a small piece of the puzzle, add context and create more questions than we began with. It's not the end to the discussion.
Let me start with the good stuff about my study:
Double Blinded (diet plans were modified with supplement powder that researchers and participants were not aware of the content of)
Cross Over
Analyzed using magnitude based inferences vs P values
Participants were taught to track, weigh, and measure food accurately and had communication with a registered dietician throughout the study to make sure any mishaps/slip ups/food exchanges etc kept the diets in line with the experimental design
We measured strength, anthropometrics and athlete specific psychological changes
So, my biases won't effect the results, we can be very sure of the compliance of the participants, individual effects can be analyzed and discussed (mean changes can mask individual results), and small effects won't be "no effect" and results won't be inappropriately labeled in a binary way (two flaws of p values). These statistical aspects are critical when discerning small changes that may matter over time. And finally, we have data on a number of relevant variables
Next, let me point out the the downsides/limitations:
MRI, DEXA, hydrostatic weighing and ultra sound ALL fell through due to broken equipment, the inability to fix equipment, equipment not arriving, and having a grant turned down. The realities of research hit hard. So that left with me with only anthropometry to measure changes in fat mass and lean mass. Even though some of the best anthropometrists (ISAK level 4) are here at AUT, you just can't get a reliable measure of LBM from it. You can get a reliable measure of bodyweight change and skinfold (fat mass) change though, highly reliable in fact, but not muscle or LBM.
The final limitation is the length. Crossovers need wash outs at least twice the length of the intervention, of course the interventions are separate so they take twice as long as parallel group designs. So although the time length intervention was only 2 weeks on each diet, that was the most I could manage as I was getting ethics approval from Sep-Oct, recruiting from October all the way through June, and collecting data from Feburary to July, and my Masters thesis is due...well...now pretty much lol.
So all that said, what did I do? I compared an isocaloric, 40% caloric deficit (same as Walberg, Pasiakos, and Mettler), matched carbohydrate (walberg found performance changes when protein was modified by reducing carbs, Mettler used fat and avoided this) diet, of 2.8g/kg protein with a low fat intake to a 1.6g/kg protein with a moderate fat intake for 2 weeks in lean (13-14% bf average), resistance trained (1 year min), adult males. We tested full body maximal strength before and after, anthropometry, and athlete specific psychological stress.
To put it simply, changes in anthropometry were almost exactly the same. Changes in strength were as well. However, the group on the lower protein intake reported higher levels of symptoms and signs for athlete related stress, high number of sources of athlete related stress, greater total mood disturbance, greater fatigue and greater dissatisfaction with the diet.
I specifically analyzed the data with carb sources as a covariate to make sure this was not related to the large amount maltodextrin powder that comprised the 1.6g/kg groups carb intake (the 2.8g/kg group had protein powder), and the fact that not only total mood disturbance, but also specifically fatigue (which is unrelated to satiety or hunger) increased indicates this was not just due to proteins satiating effect.
When people report stress it typically precedes or accompanies measurable physiological changes. Had the study been longer small differences in LBM changes would have become more and more detectable, if this had been a 2 month vs 2 week study would they have showed greater LBM loss in the lower protein group which might have caused the stress? Or, if we had a reliable measure of LBM would that have discerned these hard to measure changes?
Don't know
But what we can say is that in the context of a 40% caloric deficit, the 2.8g/kg protein, low fat diet was less stressful, less fatiguing, caused less diet stress and is therefore likely more sustainable. That said, I think probably an approach where you use a 20-30% deficit would be even better, this would allow you to not have such a low fat intake and might improve performance measures.
I'll be posting my Masters after the external reviewers approve it and ask for revisions, it will have the systematic review, double blind cross over study, and also a chapter on what it all means for bodybuilders during contest prep (which is from an excerpt of another review I hope to publish). The masters will eventually be open access on scholarly commons. I'm also planning on trying to publish the study.