If you’re going to claim to improve hypertrophy, measure hypertrophy. (P.S. Your experiment has to be replicable)

There’s been lots of attention to one of the latest studies out of McMaster University on low-load high volume resistance exercise and protein synthesis. I, for one, am not beneath jumping on bandwagons of any kind. However, let’s strike to the core of the matter, as opposed to dancing around all the peripheral (and also somewhat inconsequential) criticisms of the study.

Burd NA, West DWD, Staples AW et al. Low-load high volume resistance exercise stimulates muscle protein synthesis more than high-load low volume resistance exercise in young men. Public Library of Science 5(8): e12033, 2010.


The authors highlight research in their introduction that suggest that lifting heavy may not be necessary to increase muscle protein synthesis. Their theory is that the number of repetitions is the determining factor to maximal motor unit recruitment (as per the size principle of motor unit recruitment) and therefore, load is actually a secondary consideration if muscle protein synthesis is the goal.

Unfortunately, we don’t understand muscle growth very well still. There are a whole host of signals that we are barely beginning to understand. Pathways such as the Akt-mTOR pathway, as well as Pax7, which is a marker for satellite cell activation have been identified, but how they play a role in making muscles bigger is poorly understood.

So, these researcher set out to answer the question of whether lifting to failure with a protocol involving 90% 1RM loads would change things like, “anabolic signalling” compared to a protocol involving lifting to failure at 30% 1RM.

Methods (and some results):

The authors studied young men. There doesn’t seem to be any indication that there were any pre-planned inclusion or exclusion criteria. They _happened_ to get 15 guys who _happened_ to train legs/lower body with weights or with weights and cycling more than 3 times a week for at least 6 months. It looks like they were looking for guys who were familiar with lower body exercise to minimize the contribution of neuromuscular strength gains.

All subjects underwent 1RM testing for leg extension. This 1RM was repeated to determine if it was a true 1RM. Once the 1RM had been established, the subjects were randomly assigned to 2 of three experiments:

1) 90% 1RM to failure
2) 30% 1RM in which the “work” (as measured by reps x weight) was matched to the 90% 1RM to failure (e.g. if you pushed 90 pounds for 5 reps as your 90% 1RM, “work” would be 90 x 5 = 450. You would then have to do 450/30 = 15 reps at 30 pounds to match the work)
3) 30% 1RM to failure

The subjects were randomly assigned in a “counterbalanced” fashion(i.e. not actually random) to match for leg strength and body weight. There were 10 subjects in each training experiment.

When subjects arrived in the lab, the load was set according to the previous 1RM test. The 30%1RM with “work matching”, was performed with the number of reps required to match the work of the 90%1RM experiment. Subjects performed 4 sets with 3 minutes of rest between sets. The tempo was 1-0-1.

[So, in case you’re just as confused as I am, this is not a training study. It’s a study to look at the effect of a single session of a training condition on “anabolic pathways”. The authors had a pool of 15 guys, but for reasons not stated, couldn’t run all 15 guys through all three conditions. So, each guy got to do two of the three conditions. For all intents and purposes, there would be an overlap of 5 guys between condition 1 and 2, 2 and 3, and 3 and 1. However, in subsequent text, it seems that the subjects would have gotten either the 1/2 combination or the 1/3 combination, with no 2/3 combos because performing condition 2 was based on the “work” performed in condition 1 (you can’t match the work for 90%1RM to failure if you never do the 90%1RM protocol).

So, to summarize this gong-show of a methods section so far:

1) 15 guys.

2) 10 guys perform the 90%1RM to failure protocol and then the 30%1RM with “work matching” protocol. 10 guys perform the 90%1RM to failure protocol and the 30%1RM to failure protocol. By the reporting, no one does the 30%1RM to failure AND the 30%1RM with “work matching” because they HAVE to have done the 90% to failure condition in order to match the work.

3) Each guy did two and only two protocols.

There is no math/logic that I’m aware of where all of these conditions can be true.

This is, in short: What. The. F&#K.]

Diet was controlled using a protocol that I’m not going to go into here because I’m already losing interest in this study.

On testing days, subjects had baseline blood samples drawn. After that, they received an infusion of radio-labelled phenylalanine. The subjects then had one muscle biopsy taken from the vastus lateralis muscle of one leg to determine their resting muscle protein synthesis. Four hours post-exericse, and 24 hours post-exercise, a biopsy was taken from each vastus lateralis.

This muscle was analyzed for protein synthesis as well as the markers for gene expression of 6 different genes linked to muscle protein synthesis.


Muscle protein synthesis: Mixed protein synthesis was elevated for all conditions at 4 hours post-exercise. The increase was about 3.5x more than resting in the 90%to failure group, 3.2 fold in the 30% to failure group and 2.1x more in the 30% to failure group. This increase was not sustained at 24 hours post-exercise, but remained higher than resting at 24 hours (less than 2x for all conditions).

Myofibrillar protein synthesis was elevated in all three conditions 4 hours post-exericse, but only remained statistically elevated in the 30%1RM to failure group at 24 hours.

Sarcoplasmic protein synthesis was elevated in only the 90% and 30%1RM to failure groups (statistically) at 4 hours post-exercise. Only the 30%1RM to failure group continued to experience statistically elevated levels at 24 hours post-exercise.

“Anabolic” pathways: I’m sorry, I just can’t read this anymore.


In their discussion, the authors have the audacity to claim that using a 30%1RM to failure bout of exercise is more effective at increasing muscle hypertrophy than a 90%1RM to failure bout. Their reasoning is that the 30%1RM to failure group had similar myofibrillar protein synthesis as the 90%1RM to failure group at four hours and continued to exhibit higher level of myofibrillar protein synthesis at 24 hours (where the 90%1RM to failure group’s level of synthesis dropped back to close to baseline at 24 hours).

This is, in fact, the crux of the entire paper.

I have three points to make:

1) There are massive issues related to inadequate reporting in this paper. The experiment, as reported, is completely not-replicable, and in some places, make no sense at all. While this might not invalidate the results of this study, it does point to sloppy work, which may or may not reflect on the rigor of the actual study quality/protocol. Garbage in, garbage out.

2) Ignoring the reporting atrocities, there are no measures of muscle hypertrophy in this paper. Hypertrophy implies muscle size is getting bigger. Muscle size was never measured. All we can say is that protein synthesis is elevated more in the low-load/high-rep scheme. Just because you can make more money, does not necessarily make you a better dresser. You’re definitely richer, but you might not be very good lookin’.

If we accept that muscle size is determined by mixed, myofibrillar and sarcoplasmic protein content (and I’m not convinced that we actually know this to be true), hypertrophy can only occur if there is a NET increase in these constituents. This study, as with most protein synthesis studies, does not take protein breakdown into consideration. If a certain type of apple tree can produce 100 apples a month while another type of apple tree can produce 200 apples a month, this difference is pointless if the second apple tree drops 150 apples to the ground while the first apple tree drops only 25.

3) This is NOT a training study. This is a study about muscle protein synthesis after a single bout of training one way or another. We cannot make any conclusions about how protein synthesis might be different by using the high-rep/low-weight lifting approach as a training strategy. For all we know, protein synthesis falls off after two of these workouts. Or it might not. We just don’t know–at least from this study.

The bottom line: If I ignore the horrible reporting (and that’s REALLY hard to do), this study STILL isn’t good enough for anyone to justify changing their training program to a high-rep, low-load scheme. If what you’re doing is working, keep going. You’re not missing anything–at least based on this study. In short, from a “How should I train for maximum muscle hypertrophy?” perspective, this study contributes almost nothing.

Click Here to view the Full Version of our Website