I use levemir at 8:00pm with Novorupid at meals for my 6 year old son , but I notice that he gets high glucose in the afternoon specially from 4:00 pm till dinner, so does spliiting the dose of levemir is better or just giving him an extra novorapid at the afternoon? If i splitted it ; it should in equal doses? And is there a chance of hypos during the day?
Hi Sokar
do you visit an endocrinologist with your son? from what i have learned, you take levemir twice a day. i even had struggles to make it last for even 12 hours.
we have a few kids in camp who take it only at bedtime, but that is extremely rare and usually those kids are still in a honeymoon phase where they produce some of their own inusulin. i would definitely talk with your doctor about splitting the dose.
for me personally i had equal shots of levemir, 12 at day and 12 at night, but that may vary from person to person.
4 Likes
Thanks swisschocolate, now I give him 6 units of levemir at night, but today i decided to split it to 3 units morning and 3 evening. His endocrinologist told me to increase the dose from 5 to 6 units, but it didn’ work, still he get hyperglycemia at the afternoon and sometimes hypo at night. Thats why i decided to split the dose.
1 Like
ask the Dr, but splitting background insulin can be recommended
I don’t know if this will be of any help. it’s in USA numbers
2 Likes
Actually there are things they don’t tell you. Long acting insulin isn’t always long acting. The larger the dose, the longer the action. A dose of any basal insulin that is small (like 6 units) just may not last all day. And overnight more insulin can be “used up” due to more active insulin clearance process.
You can also split the dose into unequal amounts (like 2 and 4 units) and shift the time of doses to further refine things. It is not uncommon to have different basal needs at different times during the day and with kids who have highly variable activity levels it can be even more challenging.
5 Likes
This discussion is like a zombie. It will come again and again and again. The problem is that Levemir and Lantus are designed differently and they approach the problem differently. This is all fine if you take these instruments as they are: they offer different solutions to have the best coverage with basal insulin. One might work better than the other. Which one this is depends on the properties of the individual alone. You can even split Lantus if this will work better for you.
The real problem is convenience. In the US often the most convenient medications will be favoured by patients. In contrast in Germany the insulin with the best coverage will be the prefered choice. The best coverage for Levemir is with two injections per day. Most German Doctors and Dieticians would never recommend to use only one injection per day.
Why? As you can see in the following chart one injection of Levemir can not cover 24 hours equally. It simply can’t and it won’t. Normal dosages will cover 12 hours and then they fade out. The more you inject the longer the phase of fading out. So technically Levemir can even cover 24 hours - but not equally, not even close. This “only one injection per day” is just pure marketing to compete with the “convenient” Lantus. Look at the chart and imagine how this would change with two injections every 12 hours combined. When the first shot starts to fade out the second fades in. This will result in a much more even distribution of the activity of the basal insulin. This of course will also influence the insulin per carb ratios in a positive way.
For good coverage you will have to pay a price. I gladly pay it every day with two injections of Levemir. Sufficient basal coverage is the key to good glucose control. Try different basal insulins. Try different splittings and stick to the best solution. I tried Lantus once and twice per day but it can not beat the quality of control with 2 x Levemir for me. For others it might be Lantus that beats Levemir.
It puzzles me that it is such a big deal for Nordisk to write in the usage description of Levemir: 2 injections per day are recommended. For marketing they should better focus on the advantages of Levemir. For example that it binds to albumin and not fatty tissue like Lantus. Thus is makes no difference to accidentally hit a blood vessel while injecting Levemir. Do that with Lantus and you most likely will go low very quickly. Instead of discussing these issues more openly Nordisk developed Treshiba a very long acting insulin - sadly at a much higher price than Levemir. In fact the pricing is that absurd that Treshiba is not covered by the German Health Care System anymore.
I inject 8 units at 7am and 5 units at 9pm. This way I can prevent to go low at night. I also eat a slice of Wasa bread before going to bed.
7 Likes
This just gives me one more reason to be happy that I switched endocrinologists 
My ex-endo incorrectly told me the exact opposite – that for smaller doses, there should be no reason to split the dose for 24-hour coverage, whereas longer doses often wear off faster. Clearly that was NOT my experience and not born out by experience or documentation. I split my doses when they were small - and continued to do so when I took a larger dose - and coverage was much more consistent. I also did an uneven split, but, due to problems with dawn phenomenon, I took the higher dose (late) at night and the smaller one in the morning.
2 Likes
This is one of those (many) things for which there is no formulaic answer. You just have to determine empirically what gives the best results.
When I began using Levemir, I observed a definite peak at a certain time. I split the dose (morning and evening, more heavily weighted toward evening) and now it’s pretty flat. But everyone’s results are their own. You just need to try it and see. But definitely seek medical advice.
3 Likes
Can you please give the reason not to split Lantus into two doses or is that something that would work in a similar way to Levemir?
@drsoosie I tried Lantus with one injection first. For me the majority of the Lantus did not last 24 hours but more like 22 hours. Thus my BG climbed significantly after the 22 hour mark. I refrained from adjusting my I:C to compensate that because it would have forced me to eat.
Instead I tried to split the Lantus in two equal dosages. To my surprise this caused great turbulences for me. One day I was high and the next day I was lowish at exactly the same time. My endo thought of it as an oscillation pattern - assuming that some residue of the two Lantus dosages are acting far longer than 24 hours for me. Here I stopped the experiment without testing other dosage adjustments and gladly switched back to Levemir.
The experience itself is of value I would say. But be prepared for trouble and do more testing. Others might have better experiences with the Lantus split. But keep in mind that Lantus is intentionally designed to be equally distributed over 24 hours. The Levemir chart clearly shows that this was never intended for Levemir.
I tried splitting Lantus in the exact same way that I split Levemir. For me, the results - after a 3-day period to let it even out - were exactly the same with Lantus as Levemir: I was going low at around the same times and any reduction in the doses caused me to go quite high at other times, without fixing the lows significantly. In the end, I went to a pump to have greater control over basal rates. My endo did not think it would be worthwhile for me to try a single Lantus dose instead, so never tried that.
I’ve never personally even tried splitting Lantus doses, but I know of several regulars on another diabetic message board who have done it and found it preferable. Even though I take a very small dose of Lantus, currently 8u but it has been as low as 5u, I’ve never found the need to split. However, I’ve always taken my Lantus right after dinner, about 6 pm. That way, if my Lantus did run out a couple of hours earlier, I’d be taking my dinner bolus about the time the basal ran out, and any extra need for basal would be calculated right into my dinner I:C.
Note, though, that my basal is only about 25% of my TDD and my dinner hour is very consistent. (About the only aspect of my schedule that is consistent.) Someone with a much higher proportion of their TDD as basal or with a less consistent dinner schedule might not find my system to work at all.
I love your technical background and support for explanations and comments. Not sure what “Clamp” is on the graph, but I assume this is something like “Activity of Insulin” or equivalent, and the message is clear (also based on the units). Is there a publicly-available repository for data like this for various long- and short-acting insulins? I’m not impressed with what comes in the micro-type inserts with the pens.
FWIW, I liked the sample of Tresiba I used recently and apparently my (US) insurer agrees with Germany about inflated costs so I am still also a 2x/day Levemir user. I used to split doses equally, now I bias to a higher morning dose.
@truenorth Clamp stands for “Glucose clamp technique”. In most cases this is the Hyperglycemic clamp technique: Healthy individuals will get an injection of insulin. The blood glucose is measured continously. If the BG goes down this is compensated by minimal glucose infusions. The more glucose was infused the more active the insulin was at that time.
My graph is from the web. It was created by someone with a technical background that sadly passed away several years ago. From the information I can recall he gathered the data from official sources and the manufacturer itself. I think these should be available to the public for all insulins since they are needed for the FDA clearance. But from my understanding it was a lot of work to actually create a 3 dimensional graph to represent multiple dosages at once - showing the true character of the insulin.
I’ve seen a “clamp” referred to in many studies that I’ve read. I never understood it before. So what you’re saying is that by holding the BG steady using a variable glucose infusion, it could infer the relative corresponding action of the insulin by measuring the amount of glucose infused. So the glucose infusion is proportional to the insulin action. I learned something new today. It’s kind of like measuring c-peptide as a proxy for insulin.
1 Like
@Terry4 Comparing it with c-peptide is a bit misleading because this is just one result for one individual. The clamp studies are conducted with a number of healthy individuals - usually a cohort that is representative for the population. Thus the infered activity pattern is likely the median and with less likelyhood the average of that cohort. This should also compensate for the very individual capability of clearing excessive insulin in the kidneys. But this might not even be of importance because there is never a state of excess if the experiment can establish a true steady state between glucose consumption and infusion. Very likely sensitive lab equipment is used to measure the blood glucose in real time to counteract accordingly with micro dosages of glucose. It would be quite nerve wrecking to conduct this experiment with insulin dependent diabetics like us. Good that healthy people have no experience with severe lows.