• Guest, the forum is undergoing some upgrades and so the usual themes will be unavailable for a few days. In the meantime, you can use the forum like normal. We'd love to know what you think about the forum! Take the 2025 Survey »

When eating a low carb diet, should we change the way that the MDI model is used?

tim2000s

Expert
Retired Moderator
The guidance for the use of MDI is that we should use a basal insulin to provide a background insulin level and boluses to counter mealtime carb caused BG increases.

Typically Basal insulin should provide a bg level that doesn't vary more than 1.7 mmol/l over a >5hour fasted state. It should effectively be used to counter the variation caused by the liver alone.

Bolus insulin has a much shorter life and peaks much more quickly.

When eating a low carb diet, we know that gluconeogenesis from protein generates a slow increase in blood glucose that typically doesn't fit the profile of the rapid acting insulin and that there is a glucagon related increase linked to the immediate ingestion of any protein. This has been readily discussed in this topic: http://www.diabetes.co.uk/forum/threads/insulin-load-index-most-ketogenic-foods.75704/

Taking this a step further, the existing MDI model doesn't work terribly well for the low carb diet. If we stick to the rules above, post the use of the mealtime bolus, there is a steady increase in BG level that is not countered by anything. In my view there are three ways to treat this:
  1. Sugar Surfing: Using additional boluses to counter the steady protein induced rise - this requires multiple additional injections and comes with the down side of increased damage to tissue due to more injections.
  2. Provide multiple bolus insulins with different durations to be taken at slightly different times, e.g. Apidra for the meal and Actrapid for the protein increase post meal. This also requires additional shots, but not as many as multi-bolusing.
  3. Increase the basal insulin level so that is counters the protein caused rise. This reduces the number of injections required but runs the risk of basal hypo if the amounts used are too high.
I'm not sure which of these is the best approach to take. I have been using option one as I have next to continuous BGM, which makes it relatively easy to see where I am and sugar surf, but not everyone does have it.

What are your thoughts on dealing with this Low Carb linked phenomenon?
 
Although I don't technically low-carb now myself (180g a day) experience has taught me like many other type 1's that the traditional methods of carb counting and bolusing goes out of the window when you follow a LC diet.

You've already highlighted the reasons why Tim and to me its just a case of trial & error when trying to adjust your insulin for a LC diet, there's talk about TAG and other approaches but the one that works best is T & E and learning from the results, pretty much as you are doing now.

Just one thing, could you not leave a bigger gap between injecting and eating and just use the Actrapid insulin rather than using the Apidra as well, the longer profile of this insulin should match the slow breakdown of the protein and fats better IMHO........ but it does takes a while to kick-in.
 
Just one thing, could you not leave a bigger gap between injecting and eating and just use the Actrapid insulin rather than using the Apidra as well, the longer profile of this insulin should match the slow breakdown of the protein and fats better IMHO........ but it does takes a while to kick-in.
It's a thought but as I do eat around 15g -25g of carbs per meal, and I also see a glucagon response to some forms of protein, I do need to use the fast acting!
 
It's a thought but as I do eat around 15g -25g of carbs per meal, and I also see a glucagon response to some forms of protein, I do need to use the fast acting!


TBH its many years since I used Actrapid so forget how long it takes to start lowering bg levels, but maybe worth trying a longer gap and see if its deals with the small amount of carbs in your food, by increasing the amount of injections your only going to get injection problems further down the road.

Maybe enquire about a pump, this would definitely work better for low-carb meals as you can use features like Extended Boluses and TBR's, I'll tag @Spiker as he low-carbs and uses a pump.
 
That's high on my list of requests @noblehead. At my next clinic appointment (end of June) I will be making this case. The consultant was already concerned about injection sites, so it adds to my arguments.

What I am doing is effectively extended bolusing already. It would just be nice to have it down the same pipe.
 
What I am doing is effectively extended bolusing already. It would just be nice to have it down the same pipe.

That's it, on MDI when I had a meal that was high in fat I'd have to inject twice and sometimes three times so that my bg wouldn't end up in double figures, its not great or ideal when you already have site issues :(
 
There are a couple of injection site "cannulas" that don't seem to be properly on the market that are supposed to make it an "inject once" model on MDI. Would love to get hold of one as it would make this a whole lot less of an issue.
 
There are a couple of injection site "cannulas" that don't seem to be properly on the market that are supposed to make it an "inject once" model on MDI. Would love to get hold of one as it would make this a whole lot less of an issue.

Not heard of that, sounds interesting.
 
There are a couple of injection site "cannulas" that don't seem to be properly on the market that are supposed to make it an "inject once" model on MDI. Would love to get hold of one as it would make this a whole lot less of an issue.
ask @iHs to PM you...
 
Back to the OP...

Tim as you probably know I used to use Novorapid or Actrapid (*) for protein, post meal, vs Humalog pre-meal for carbs. This worked reasonably well, and I used it during my first foray on CGM so I was able to see it working (sorry, not screen shots). My second and subsequent uses of CGM I was already on a pump so I was no longer using these techniques. I won't say it was perfect but it wasn't terrible. It was better than just using Humalog post-meal for protein, for example. In those days I was on more of a Bernstein diet rather than an LCHF. In fact I noticed that in Bernstein's most recent podcast he is asked this specific question, using a slower insulin for protein, and he dismisses the idea. His own advice on protein (which is almost incomprehensible, alas) is to use the same insulin, and presumably the same shot at the same time, and factor the protein as carbs. I defy anyone to confirm what Bernstein's formula for protein dosing is, however!

When I was thinking about this more, pre pump, I thought that the insulin manufacturers should try to produce an insulin that had an action profile better matched to protein. Maybe even a range of insulins matched to different food groups. Of course, if they even thought about that, they would see there is far more market, far more money, in designing an insulin with an action profile matched to pizza. :banghead:

* = for me anyway, Novorapid has less total action and a slower, longer action than Humalog. Only rarely did I get my hands on any Actrapid (weaker and slower still than Novorapid). That was pre CGM and when I was doing a very pure Bernstein diet, very high protein. I would have liked to see how Actrapid (regular human insulin, non-analogue) worked by watching it on the CGM.
 
Last edited by a moderator:
Having seen a lot of GNG going on over the past couple of weeks and having enough of giving myself a shedload of injections every day (typically 12-15 Rapid Acting jabs), I took the decision today to go with option 3 and increased my basal insulin by 30% to counter the constant upwards pressure I've been seeing on my BG levels.

I'm keeping a close eye on my bg levels, but without a pump, frankly, this seems like the best option as it reduces my need to interact with my diabetes all the time. The first 6 hours seem to have been effective in that I've seen spikes that are only related to food ingestion and the constant desire of my body to push my BG up seems to have gone away. I therefore feel much more comfortable this way.

Given that I am comfortable with the way that Levemir works, what will be more interesting is seeing what the duration of the increased amount is and how I need to vary my overnight dose.

I'll keep an eye on it and report back, regardless as to whether there are any interesting findings...
 
I-ports arrived to test today. Easy to insert, I have a 6mm one in now. Insuflons should turn up tomorrow.

0502c8a82effc0c463c6c97419816660.jpg
 
Been an interesting couple of weeks with the increased basal. I've seen a marked drop in average bg level and far less upward pressure on my bg level. All in all a worthwhile change that has had an almost immediate impact.
 
I think it based from the amount of protein and the physiological effects of that @Spiker

Basal testing shows a flat result at the levels I was formerly using. During the day, when eating my normal diet, I have been tending towards a higher level, typically requiring multiple boluses at regular intervals post meals to restrain my bg. The only reason I can see for this is gluconeogenesis, so I increased basal to try and suppress this somewhat. It seems to have worked.
 
Back
Top