Since getting my hands on my Freestyle Libre, I’ve been able to gain a whole new insight into how my diabetes works, and how my body functions, playing with inputs and seeing how the variations occur in my outputs. Some of these have lead to me undertaking some further research to determine what is going on with what I’m seeing and what might be causing some of the effects. This has then led me to a number of hypotheses and some further thought experiments.
The areas that I’ve found to be most interesting appear, at least at the offset, to be only loosely linked, however, the more deeply you look into it, the more complex the system seems to become and the more holistic therapy should perhaps be. This stems from food intake, glucose generation in the body, artificial pancreases, exogenous insulin and how the body really operates.
But where to start? The impact of Protein is a good place.
Protein and Glucagon
Having spent the best part of the last nine months eating a higher than average amount of protein in my diet and a lower than average amount of carbohydrate, I had come to notice two things. One is that protein causes a noticeable quick increase in bg levels within 1hr post eating and the second is that protein causes a longer term (2-4 hour) steady increase in BG levels post eating.
Now all the NHS guidelines say that for MDI, carb counting is required, yet there is no mention of counting protein or what might happen with it. In fact, there is little discussion amongst healthcare professionals (HCPs) with the average T1 patient about this at all. My experience of the past few months is that this is wrong.
If body builders take a Whey protein shake to induce an insulin reaction (and bear in mind that in a non-diabetic, this would also induce a glucagon reaction), why wouldn’t this nugget of information be used more carefully?
Further research determined that protein, or more specifically certain amino acids, drive an insulin reaction in order to get these little blighters into use in the muscle. To manage this, there is also a glucagon release to avoid a low blood sugar caused by an excess of insulin with no carbs present.
The diabetic body being what it is, the amino acid ingestion seems to trigger the glucagon release, regardless of the fact that the body doesn’t have insulin. There is therefore a sudden increase in glucose in circulation, in spite of no carbohydrate being ingested.
What also occurs in the diabetic is the presence of excess protein being converted to glucose via gluconeogenesis. This is a slower, longer process, but is clearly visible via the libre. This causes a requirement to inject delayed boluses at 1.5 to 2 hour intervals post protein ingestion.
These protein effects can be bolused for, indeed there are many others who recommend it including Dr Richard Bernstein and also the work being done by Marty Kendall.
There is also evidence to suggest that any food source causes a release of glucagon, including carbohydrate (here), and that in a diabetic, this is not fully managed due to the lack of beta cells. I call this the glucogenic multiplier effect of certain food types. We see this happening both on carbohydrate and protein.
But why is this significant? Both of these side effects are driven mainly by glucagon and in a diabetic there is nothing to stop glucagon’s action.Of course, that isn’t strictly true, there is exogenous insulin, but that’s where the second part of the story starts.
Where does Exogenous Insulin come into play and why is it related to Glucagon? In a non-diabetic, insulin is released by the beta cells into the hepatic portal vein. Its first point of call is the liver. In the non-diabetic, the liver is then responsible for removing the majority of glucose from the blood. In addition, the action of insulin in hitting the liver first is to limit the impact of glucagon and reduce the propensity for gluconeogenesis and glucose production as a result of glucagon action.
In the diabetic, it’s a little different. The insulin is injected subcutaneously, and as a result, it reaches muscles typically ahead of the liver. This means that the majority of glucose is removed from the blood by the muscles, but also the off switch for the production of glucose by the liver is much less strong. As a result, more insulin is required to reduce the overall impact of eating either carbohydrate or protein containing foods.
But the other key thing here is that there are no beta cells, and therefore a further hormone is not released, Amylin. Amylin is a partner hormone with Insulin, that amongst other things, improves satiety and reduces the impact of glucagon. Instead of diabetes being a two factor condition (insulin and glucose) it is looking more like a four factor one (lack of insulin and amylin, unregulated glucagon and unmanageable glucose).
What does this mean? I strongly suspect that it means that, again, the lack of this hormone results in overuse of glucagon, generating additional glucose within the diabetic system. There is also some consideration that the loss of the beta cells removes a key regulatory function of the alpha cells which again allows unregulated glucagon production (see this paper from Roger Unger and Alan Cherrington).
The pancreas is therefore doing a much more complex job than has been the doctorine for many years and the insulin closed loop feedback system would appear to be poorly understood by many HCPs. Indeed, I would argue that the endocrine closed loop feedback system is considerably more complex than the analogues that we currently employ to approximate and replicate it.
In addition, the lack of access directly to blood means that controlling the liver with insulin alone is much more difficult. Recent introduction of the inhaled insulin “Afrezza” has highlighted a very important point. Due to the level of access to the circulatory system, Afrezza seems to reach the liver as fast as any other part of the body. In doing so, the hypothesis is that it is inhibiting the action of glucagon on the liver and may very well reduce the overall glucose spikes that the majority of diabetics see with carbohydrate heavy foods. It also suggests that an insulin delivery system directly into the bloodstream would also have a similarly useful effect, if one could be arranged without damage.
With regard to the glucagon metabolism, this is a strong development that warrants further observation.
The artificial pancreas and glucagon inhibitors
Given the complexities outlined, the driver to have a closed loop external pancreas is clearly beneficial. Research for this appears to fall into two camps. One that includes glucagon as part of the closed loop and one that doesn’t.
For me, I believe that the glucagon inclusive closed loop is the more beneficial, and if combined with a glucagon inhibitor, so that the only source was the artificial pancreas, I think that managing control would be even better. What inhibitors are there? Symlin is an analogue of Amylin and has been around for some time. Leptin is also being researched as a glucagon inhibitor. Use of both these and a monitoring and dosing system would potentially enable incredibly tight management of diabetic BG levels, We are a long way from this point, but it raises some interesting questions with regard to how diabetes and blood glucose might be managed.
Conclusions and many, many questions
Since writing this, a paper has been brought to my attention that draws the conclusion that blocking glucagon will lead to better Glucose control. I will be first in the queue to experiment on myself… http://www.pnas.org/content/112/8/2503.abstract