Exercising for Diabetes – Insuline Resistance Protocol

I have been recently asked for exercise advice by someone who has been diagnosed with pre-diabetes, ie with significantly – albeit not yet dangerously – elevated blood sugar levels due to insulin resistance. I am not a medical expert, but I could piece a few things together out of what I have read over the years. In the following are my recommendations – I would welcome any comments, privately or publicly, as to whether this advice is indeed optimal.


Insulin resistance is a condition where some of the body’s glucose-absorbing tissues (notably the muscle cells, the liver, and the fat cells) need more and more insulin to remove the glucose from the bloodstream. As a consequence, the pancreas needs to compensate by producing more and more insulin. Initially this works well, so there are no effects other than elevated insulin levels. At one point however the pancreas can no longer keep up, and blood sugar levels remain elevated. Finally the pancreas might collapse under the overload, in which case Type I diabetes ensues.

It is a well known fact that exercise can to some extent reverse the insulin resistance. I am not too sure about the exact mechanism, but my hypothesis is that this is related to the exhaustion and rebuilding of glycogen stores in the muscle and also in the liver (it is probably also important to reduced overall stress on the liver, in particular by severely limiting daily fructose intake, even if this means limiting fruits).

Insuline Resistance Protocol

At this point it is important to point out that the person in question is almost 70 years old, and whilst he is reasonably fit, he is overweight, and he has not been doing other sports than some recreational skiing and some back- and mobility focused exercises in the past. So the programming of the protocol needs to optimise the following parameters:

  • maximum glycogen exhaustion in the muscles and in the liver
  • minimum impact on joints and tendons
  • appropriate (ie limited) cardio-vascular stress levels

In my view, the best way to deal with those is to design the program as a HIIT circuit. This means I have to define the nature of exercises to be used, their load/intensity, the time under load / repetitions, and the rest periods. I will visit those areas in turn

Nature of exercises

The two key parameters here are that (a) glycogen stores should be exhausted, and (b) that they are safe to execute for a relatively untrained person. It is well known that glycogen that is stored in a muscle can only be used in this particular muscle, as there is no mechanism to return the glucose into the bloodstream. As a consequence, the exercise chosen must be such that muscle glycogen is exhausted individually, with a focus on the largest muscle groups first, simply because this is the biggest bang-for-the-buck.
As for the exercise type (ie composite vs compound vs isolation) I am rather agnostic, or rather it depends on the circumstances. Composite movements (eg a lunge combined with an overhead press) might be the fastest to use up glycogen (in grams / minute) but they are complex, and if the cardio-vascular system cant handle them (which is likely in this case) one might simply revert to doing easier movements with shorter breaks instead. Compound movements (multi-muscle, multi-joint) are arguably generally preferrable to isolation movements (single-muscle, single-joint) but ultimately this is a matter of personal preference, assuming the exercises are appropriately rotated, and rest periods are carefully chosen.

Load/intensity of exercises

As I have mentioned before, the main purpose of the exercises is to exhaust the muscle’s glycogen stores. In principle this can be achieved at various load/intensity levels, but a number of considerations have to be balanced.

  • Risk of injury to muscles, tendons, and joints – this risk is biggest at very high loads (especially if form is bad) and at very low loads (overuse injury due to excessive number of reps)
  • Activation of maximum amount of muscle fibers – to reach the glycogen stores in all fibers, i would assume that those fibers need to be activated (is this true? They might be able to share locally). Assuming this is the case, either weights need to be heavy, or the speed must be high
  • Energy-efficiency of the movement – the less efficient the movement in energetic terms the better in order to maximise glycogen usage. There are two kinds of movements that fit this bill: firstly anaerobic movements (where one glucose only yields two ATP as opposed to many more under aerobic movements) and secondly slow / isometric-style movements where muscles expend energy holding, or are even working against each other.

So there are two major options emerging – either (1) moderately heavily weighted (maybe 60-80% of 1RM) movements executed at a slow pace, or (2) lightly weighted movements (maybe 30-50% of 1RM) executed at a dynamic pace. Note that loading does not necessarily imply using weights: one might base the routines around bodyweight exercises, and loading or deloading might be achieved by positioning of the body (eg raising / lowering of the legs for push-ups).

Time under load and RTR

This one is the last variable to be adjusted once the other one’s (exercise type, intensity/load) are fixed. The goal here is to work to complete muscular exhaustion, caused by a depletion of glycogen stores. Note that this is different from the type of exhaustion that is caused by a built-up of lactate and that is reversed much more quickly. It is also different from cardio-vascular exhaustion.

Total glycogen stores in the muscles are about 350g for an average male, and an additional 80g in the liver that can be used to replenish muscle glycogen stores. This corresponds to 1400 / 320 cals respectively, which is a considerable amount (as a reference, recreational exercise tends to use up between 300-600cals per hour). So the first recommendation would be to exercise fasted (or at least without having consumed carbs) for 10hours, so that liver glycogen stores are empty and there is no ready glucose supply from the blood. It is nevertheless not realistic to assume that all muscle glycogen stores can be exhausted in one session.

I have no specific data with respect to this, but my hypothesis is that it is better to completely exhaust the stores in a limited number of muscles than to partially exhaust the stores in all muscles. At least this is consistent with the supercompensation principle where optimal adaptation is generally caused by near-maximal stresses together with an adequate recovery period.

On the most basic level, in order to exhaust the glycogen stores in a muscle, a sufficient number of reps under a sufficient load must be performed, and the other constraints (cardio-vascular, lactate must be overcome). The way to do this is to allow for sufficient recovery between the sets so that the cardio-respiratory system can be recover, and that the lactate can be removed. For different people (and/or different muscle groups) this will necessitate a different protocol: if cardio-vascular exhaustion is the binding constraint, then it is enough to perform a set until the heartrate is above a certain maximum threshold level (say 90% of maximum heartrate) and wait until it is back in a reasonable range (say below 100-110bpm) and then to restart. More often than not however, the build-up of lactate will be the binding constraint, especially on the smaller muscles. In this case it is not optimal to wait for recovery of this particular muscle, and it is better to engage in a rotation of exercises (“super-setting”), with appropriate rest-periods thrown in as dictated by the requirements of one’s cardio-vascular system.

A strawman exercise plan

This is just a very simplistic first draft of a protocol drawn up under the aforementioned principles. Please note that throughout the exercises, provision of water (and electrolytes if necessary) is OK, but provision of any kind of food (including sports drinks and protein shakes) is not.

Day 1 focuses on the body’s biggest muscle groups, the lower body. Day 2 focuses on the second biggest muscle groups, the upper body. Day 3 involves smaller muscles, and after one rest day the cycle repeats.

Day 1

  • Unweighted squats, executed dynamically in the upwards movement (some rest in the top or bottom position is acceptable; if stability is an issue it is possible to hold on to for example a rope or a sling)
  • Optional: plank 1-3x for maximum hold period

The assumption is that for squats the lactate recovery will be in line with (or slower than) the cardio-vascular recovery so there is no need for additional exercises. The squats are executed until the heartrate reaches 80-90% of the max, followed by a recovery (walking, not standing/sitting) until heart-rate reaches 100-110. This is repeated until there is no strength left in the legs.

To ensure that indeed glycogen, and not lactate, was the binding constraint a 5-10min rest is taken (sitting down is OK) and the exercise is started anew. It is expected that failure ensues after very few reps. If this is not the case, lactate was the binding constraint, which means that (a) the whole protocol should be repeated untile complete failure, and (b) a rotation of exercises (firstly including the plank, if this is not enough also with for example push-ups, pull-ups, rows, back extensions) should be considered

Day 2

  • Push-ups, appropriately deloaded (or loaded for that matter) so that 10-20 reps are achievable in the first set (dynamic in the upward movement; hold on the top is acceptable). Can be sub’d with dips.
  • Bodyweight rows, same (de-)loading and movement pattern. Can be sub’d with pull-ups.

The assumption here is that for push-ups and rows – which involve smaller muscle groups – lactate is the constraining factor rather than cardio-vascular performance. Therefore those two exercises are rotated (“super-setted”), with a small rest period in between. Each set is ended upon muscular failure (or slightly before that), and the rest-period is timed so that the heart-rate can go down to 100-110 area. The sets are alternated until complete failure on both.

Like on Day 1, a 5-10min rest is taken and another attempt at both exercises is made. If a meaningful number of reps can be obtained, another exercise should be added to the rota (eg dips or pull-ups if possible, otherwise something like kettlebell swings, or back extensions to get a different lower body movement from Day 1).

Day 3

  • Overhead presses (dumbbell, barbell, kettlebell), appropriately loaded so that 10-20 reps are achievable in the first set (dynamic in the upward movement; hold on top or bottom is acceptable)
  • Biceps curls, same loading and movement pattern
  • Side flies, same loading and movement pattern
  • Overhead triceps curls, same loading and movement pattern
  • Front flies, same loading and movement pattern

This is a selection of exercises hitting smaller muscle groups that are appropriately rotated to allow for lactate recovery until a muscle is hit the next time in the rotation. Appropriate rest periods to allow for the cardio-vascular system to recover should be added between sets and rounds.

Like on Day 1, after a 5-10min break no meaningful reps should be possible, otherwise more exercise should be added to the rota.


6 thoughts on “Exercising for Diabetes – Insuline Resistance Protocol

  1. Pingback: Exercise plan to counter type II diabetes / insulin resistance | Mark's Daily Apple Health and Fitness Forum page

  2. Your first paragraph under “Background” should refer to type 2 diabetes instead of type 1, I think.

    The Pennington Biomedical Research Center recently reviewed which type of exercise – aerobic or strenght training – works better to lower hemoglobin A1c in type 2 diabetics. With their protocols, a combination of the two works best. I blogged about it here:

    Interesting stuff. Keep it up.


    • Thanks Steve – interesting post. A shame that they don’t specify more precisely what strength training protocol they used. They did specify the aerobic load though: it was 12kcal/kg per week. That’s quite a bit I believe – more towards the 80% than the 50% side of things given the relatively short time they had.

  3. Pingback: Anaerobic training / HIIT | Thor Falk

  4. Pingback: Research Review: HIIT & Diabetes « Thor's Reads

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s