Changing Basal Insulin - Observing what happens and hopefully providing some "lessons learned"

As I'm changing from Lantus to Levemir for reasons documented elsewhere, I thought it may be helpful to others to keep a record of what I've been doing, steps taken and a variety of things that have gone on so that if others do it, they've got some idea of what to expect.
I'm not going to suggest that this is the 100% correct way to do this, but it will be my experience.

This isn't the first time I've changed basal insulin. I moved from Insulatard to Lantus some 10 odd years ago, however this is the first time I've moved to one with greater hypo indication.

So let's get started. First things first. Getting set up.

Having agreed to move to levemir, I was told that to start with, use the same doses as I used for my split lantus. This is 10u at 9am and 10u at 9pm. Many people who move in the same direction as I am find that Levemir requires larger doses, so we will see what happens.

Having taken Lantus at 10u the night before, I took the decision to not do the second split on the am following so that it would be flushed out before starting on the Levemir. I also have the libre so that I can keep tabs on what's going on with my bg level, which helps a lot.

With these decisions made, I took my Levemir dose at 9pm yesterday, into my bum (first time in 26 years), and decided that as my blood sugar was in the 6 range, I still had a couple of hours of novorapid on board and I wasn't sure what my reaction would be, I would take my bg level up to ~8 for safety reasons.

Then off to bed I went. Now I know I should probably have set an alarm to get me up in the night and test, but I figured that a hypo would be as effective at that, and with the dose I had taken, this felt unnecessary.

On waking this morning, a quick scan with the Libre revealed the following:



It would appear that the combination of 10u of Levemir and my bum has induced a noticeable decline in bg overnight. The drop is roughly 4mmol/l. I'm very pleased I had that extra carb the night before. Now clearly one day is not a pattern so I'll be doing the same process this evening and following results tomorrow morning, I suspect it is likely that my evening Levemir dose will drop by two units.

Once we've got that one stabilised, it's on to the daytime checks, with the associated fasting, which I am much less looking forward to!

The observation I would make is that the Libre continuous data is invaluable in this process and makes the whole thing massively easier (and means that I get to sleep properly as well!).

A quick follow up to this one - I know when Basal testing on a pump that one is supposed to have three repetitions to observe any patterns, but I'm thinking that just the one with Levemir will be enough. As my eating patterns won't be particularly different tonight I will be changing the Levemir dose after one night. The low (even though it's not that low) this morning has felt more prolonged and persistent than those with Lantus.

Of course it may just be that I was far more used to Lantus, so this is my body's low reaction to Levemir, which is different to that of Lantus (and has reset/modified my hypo awareness symptoms again).

Day number 2 of the change over to Levemir and it appears that my body sees Levemir as a more efficient insulin. Let's start with the overnight scan!



What does this tell us, and what is the background?

I went to the gym yesterday night and given previous experiences with Lantus, decided that my decision to reduce Levemir to 8u would go ahead. As I had eaten and had a positive bg trajectory, this made sense.

As you can see from the graph, my body seems to be finding Levemir easier to live with than Lantus. At around 2am, there is an increase in the rate at which the blood sugar drops, which feels a little odd, and I don't have an explanation for. I woke up around 2.30 and treated the hypo, but clearly went back to sleep before it was properly checked. This meant it continued until I got up. Not the greatest thing to happen.

What is fascinating for me is that everything I was told about Levemir was that I would probably need more of it as a basal than the Lantus, however I seem to be seeing the opposite.

Anyway, the lessons learned this time are that post gym, I'll be taking the dose down by a further 2u (i,e, tonight) and seeing what happens. So a total basal dose at the moment of 16-18u. That's possibly the lowest I've had it since I went on to MDI!

And yet another great advert for the benefits of having a Libre.

End of day number 5, and I've not eaten anything carby since about 3pm, as can be seen in the picture. Dinner involved pork belly and salad and there has been no spike from the pork belly protein, which I sometimes see. Instead I've had a slow climb, for which I bolused 1u, but hit a blood vessel. I wonder whether that hit at all as I see no blip in the chart to suggest it did. The daytime levemir was a 10 and this seems to work okay, although my hypo correction glucose doses are slightly different to previously and I'm still coming to terms with what I've been used to for ten years.

Here's the graph.

Thanks smidge. I certainly notice the difference when exercising in that I need less of the Levemir.

I've not yet seen a need to increase Novorapid, but then I had to do a variety of injection timings with that even on Lantus due to the way I seem to process protein. I also need to do a Novorapid profile test, where I run around 8 and take one unit, then see how it looks in the absence of carbs. Again, something that the Libre is invaluable for.

The other thing I discovered, which was slightly unexpected, was that the amount of glucose tabs required to adjust a low seems to be fewer. As always with these things, it ends up being trial and error, but the knowledge of what is going on in the background is very useful.

Here's the intriguing one for me. The graph shows a few interesting things happening.

1. At 2am there is a sudden drop from about 9mmol/l to about 5mmol/l
2. At around 3am, my blood sugar decides to climb (almost as though there is a dawn phenomenon going on). Did I wake up I wonder?
3. It then declines in a fairly flat way, before giving a another noticeable jump between 4.30 and 5am, when I really did get up.
4. From 5am to 6am, there is a steady decline until the trace becomes much flatter. This could be due to me getting up and moving about, then walking to the station. At around 6am I sat on the train for half an hour and the blood sugar flattened out. It's worth noting that throughout this eight hour period I ate and drank nothing and injected no insulin. It's certainly not the flat trace I was expecting!

Observing what's going on intraday with the Basal insulin has again raised some interesting questions. It's currently seven hours since I ate and I'm heading into lunch time territory. Breakfast consisted of two scrambled eggs with 100g of 97% meat sausages. So, essentially no carbs. I've been keeping tabs on the glucose levels, via the libre as always, and seeing an interesting trend. It looks as thought the gluceneogenesis that converts protein to glucose is a slower process than I thought.

I bolused 1u for protein roughly 20 mins after eating, as I often do and applied the Levemir at 9.30, 12 hours after the last dose. Five hours after eating the glucose level started to increase again, but not rapidly and did so for two hours until it levelled out again. This appears to be as a result of the earlier protein rather than a basal mismatch, although tomorrow I'll skip breakfast to confirm. Initially I had thought it may be down to too little basal. If this is the case it will be difficult to catch with Novorapid without introducing a bit of a drop. Maybe this is a better approach to take to keep overall increases lower. I wonder whether Actrapid is a better option when low carbing due to the profile of the insulin.

Well to blow it all out of the water with regard to protein, here's the graph from last night.
Meal at 8 was chicken burgers from the local farmer's market, which are 95% chicken. The binding agent is rice flour. This was taken with a Novorapid bolus including a correction to bring me back in line for overnight. The Levemir went in at 9.30 as usual and I couldn't be more happy with this graph. It shows a completely flat line all night (although the bg reading was at 5.4, not what is shown by the libre). More noticeable is no protein "hill".

I wonder if the hills from the previous day were simply my body reacting to too little basal or whether I should really be bolusing for protein at all, given some of the papers I've read in the last two days.

Anyway, time for the daytime adjustment to the Levemir, and based on yesterday I'll be going in at 12u today.

Right then, here's the Levemir trace from last night through this morning. 12u of Levemir was taken on board at 9.27pm . What's of particular concern to me is that the blood sugar starts rising from about 7am, and just continues. It is a slow climb rather than a sharp spike, so I'm interpreting it as being the Levemir wearing off from about that point. I'm taking it this way rather than DP, as I get up around 4.45 in the morning, so would expect a DP to occur around then. Even with the injection at 9.27, I should probably have corrected to bring everything back in line rather than waiting to see whether the Levemir on its own would.



What is slightly concerning is that the overnight amount seems great to manage a flat line through the night, but doesn't last long enough to keep me going through the first part of the morning. Should I have expected Levemir to have only lasted for 9.5 hours instead of 12? This is slightly frustrating.

What was your experience with Levemir @smidge? Did it last a full 12 hours for you?

Based on what I've seen from the Libre, both Rapid and Long acting insulins seem to become ineffective in my system in a shorter period of time than the prevailing knowledge suggests. Novorapid seems to be about 3.5 hours, and this one seems to be 9.5.

noblehead said:

@tim2000s, if I'm not mistaken but I'm sure I read that you are having trouble with your injection sites at the moment (Lipoatrophy), if this is the case then you can experience unexplained highs and lows as the insulin isn't always absorbed properly or is delayed, thus predicting postprandial bg levels becomes less predictable.

Click to expand...

Maybe @noblehead , but the Levemir is going in somewhere that has never been injected into, so I would expect a pretty consistent absorption profile. Hence the question about how long it lasts for others!

I've asked the DSN what she can tell me about Levemir. The issue with the higher dose is that it causes too much of a drop overnight. Looking at the graphs from the earlier days I can also see that my blood glucose was increasing at roughly the same time as seen in this graph, however I had put it down to protein in the breakfast, which seems only to have accelerated it, looking at my plots this morning.

Thanks @smidge. According to the graph below, at 0.2u/kg, one sees a significant drop off at 10 hours. As I'm taking 0.09 it's perhaps no surprise that I'm seeing the lack of useful insulin when I am. The issue for me is that I'd have to eat a huge lump of carbs if I was to use the minimum they show, which I don't want to do. More conversations with the DSN I fear as this seems like a crazy approach! Do you have any links to the trials?