Muscles are the body’s largest glycogen (stored glucose) tank because despite the liver storing more glycogen per gram, we have a lot more muscle than we do liver (4).
Both muscle and liver represent finite tanks, unlike our fat stores, the muscle and liver don’t grow to accommodate demand for space. There is some increase in the volume of the cells: for example bodybuilders ‘carbohydrate load’ to make their muscles look bigger for a show; and we can develop enlarged fatty livers, but to my knowledge we don’t lay down new cells to act as tanks.
Problems begin when we exceed the capacity of our glucose tanks as this leads to glucose being converted to fat in a process called de novo lipogenesis – the making of new lipids (7-9).
However, not all of us have full use of the available tanks. Exercise and insulin are the signals for our muscles to take on as much glucose as they can, but in type 2 diabetes neither of these signals work as well (10,11).
Many hypothesise that the path to type 2 diabetes starts with muscles becoming insulin resistant and therefore taking up only minimal glucose (10). Certainly, by the time type 2 diabetes is diagnosed our muscles no longer act as good glucose stores (12).
Despite people with type 2 diabetes being on average somewhat resistant to the effects of exercise in terms of improving glucose control, in the majority of cases exercise helps (13,14). When looking at measures beyond glucose control, for example blood pressure and blood lipids, the benefits of exercise are even more compelling (14,15).
It’s these insights that led the team I was part of to examine a particularly potent and time efficient form of exercise – high-intensity intermittent training or ‘HIIT’. We wanted to know if this form of exercise (see my series on HIIT here) would benefit people with non-alcoholic fatty liver and type 2 diabetes in terms of glucose control, and heart health and function.
We assumed HIIT would be effective at using up glycogen stores because the higher the intensity of an exercise the more the muscle cells involved rely on using glucose, and the stronger the muscle contraction.
We did two studies that used exactly the same exercise programme (16,17). The studies differed in that the first contained only volunteers with a liver fat of 5% or more (16), and the second contained only volunteers who had been diagnosed with type 2 diabetes (17).
You can download and read the full articles by clicking: study 1; and study 2
The question we most wanted to answer is what effect exercise, specifically HIIT, would have on liver fat and glucose control if we took weight reduction out of the equation. To achieve that, we asked people to weigh themselves regularly and eat a little more if their weight was going down.
If you’re thinking that this seems a little odd, I’d agree, but in research we often try teasing things apart. Sometimes there is no good way of doing that, as was the case here.
But why would we want to know what exercise without weight reduction can achieve? Because for many people it’s the reality. See the article Exercise Won’t Make You Thin? for more on that.
Back to what we did. We randomised our two sets of volunteers to either a HIIT programme 3 times per week or no exercise for 12 weeks. We only included people who were doing less than an hour of exercise per week, so we didn’t ask anyone to stop what they were doing. We also gave the non-exercising volunteers our programme and a 12-week gym membership at the end.
I’m going to let you in on something that gets lost in the clear concise descriptions that make up published research – real life isn’t so clean. In real life some people get sick, snowed in, or go on holiday over a 12 week period, so some people had a break in their exercise routine, and had to make up the remaining time.
That’s good to know, because it means you too can pick things back up if you’ve had voluntary or involuntary break.
The exercise sessions were always a challenge to describe, so I recorded instructions for each session. We gave our volunteers an mp3 player so that they only had to come to the gym, select the recording for whatever week they were in, and follow the instructions.
Basically everyone did a 5 minute warmup on a stationary bike (normal or recumbent) before going into their first interval of high-intensity pedalling (more on how to gauge intensity here). In the first week each interval was 2 minutes long, then the length of the interval went up by 10 seconds per week.
I wouldn’t recommend doing this indefinitely, 4 minutes intervals are about as long as you want them, so if you were to follow this programme past 13 weeks, think about reducing the active recovery period or some other option described here.
Everyone did 5 intervals in total per session. They also got a 3 minute active recovery between intervals. That gave them time to come off the bike, do a minute of light resistance training with an elastic band, and get back on the bike for the next interval.
One thing you learn when doing research upfront and personal, as opposed to reading published research, is that reporting average results for a group obscures as much as it reveals.
As researchers we hope that the average changes because at least then we can say something like do this because people get better or don’t do this because people get worse.
More often though the reality is that some people get much better and others hardly change at all, so the message should be – this is worth trying, you might really benefit.
Marketers are seldom this honest, and we scientists get criticised for equivocating, but the reality is that results will vary. I’m guessing you care mostly about your results anyway.
So what did we find? Both studies showed that average liver fat went down, also, and despite our efforts to prevent it, average bodyweight went down (about 1 kg, or 2 lbs) mostly due to a reduction in fat mass. If anything lean body mass actually went up a little with HIIT.
Some people think lean mass just refers to muscle, but it actually refers to anything that isn't fat. Maybe people did gain a bit of muscle, or maybe they stored more glycogen, which is stored alongside extra water.
Both studies also showed improvements in some measures of heart form and function. This was a first for HIIT research, so we were very pleased to show some reversal of what we and others had previously shown to be changes associated with fatty liver and type 2 diabetes (18,19).
These were small studies in terms of the number of volunteers (16,17). That's largely because these tests are expensive to do, and the people who profit from exercise seldom support these kinds of studies, so the bill is often footed by charities and tax payers when governments pay.
The small number of volunteers means some of the other interesting findings aren’t as consistent between the two studies. Still, if we look at them both together, we see that on average HIIT improved liver health as indicated by a reduction in the liver enzyme ALT and AST.
In people with diabetes, there was also a small reduction in HbA1c after HIIT, and an improvement in 2-hour blood glucose – that’s blood glucose 2 hours after a 75 g glucose drink. Overall though, the improvements in glucose control were not as definitive as we were expecting.
We don’t have the data to tell us why with certainty, but there are two likely reasons:
- We asked people specifically not to change their diets, other than to eat a little more if they were loosing weight; and
- We took our post study measurements, two or more days after the last exercise session.
Most people would be guided by hunger unless they are eating to a specific plan. The people taking part in these two studies would have preferred to reduce their weight, and reap the benefits, so we may have been hampering better results by being rigorous in our requirements.
The other factor is that many of the positive effects of exercise wear off quickly, and this includes glucose control. There are exceptions, but mostly we see the main benefit in a 24-48 hour window (more on that here).
We’ve also known, at least since the early 1980s that if you replace the used glucose with a carbohydrate rich meal shortly after exercise, the insulin sensitising effect of exercise mostly disappears (1). Once the glycogen stores are full, no more is going in.
For me the take home lesson is that diet and exercise go together when it comes to improving health, but that exercise deserves to be more widely recognised as important. This also means that if you can’t see yourself changing your eating habits, you can still benefit from changing your physical activity habits.
References: click here for a full set of references cited