Back to Ask the Diabetes Team Ask the Diabetes Team

From Bristol, England:

Last summer, after a severe hypoglycemic reaction, my 15 year old son switched from mixed insulin injections twice a day to a basal dose of Levemir and injections of NovoRapid at meal times. He subsequently lost quite a lot of weight, which pleased him but caused concern until it recently stopped.

Our main problem is that his blood sugar levels consistently drop overnight usually by at least 5 mmol/L [90 mg/dl] from 8 to 3 mmol/L [144 to 54 mg/dl] or 10 to 4 mmol/L [180 to 72 mg/dl] between 3:30 a.m. to 8 a.m. after having a snack of 40 to 50 grams of cereal before he goes to bed at about 10.30 p.m. He has 30 units of Levemir before he goes to bed. If we reduce the Levemir, his control during the day is poor since his glucose levels are consistently high.

Therefore, we soldier on with constant tension between him having enough Levemir to get reasonable daytime control, but so much that the risk of nighttime lows is ever present. Levemir was meant to reduce the risk of nighttime hypoglycemia, but it does not feel like that.

Also, I do not like the idea of stocking him up with cereal before he goes to bed, which leads to high glucose levels during the night, high enough that he is not low in the morning, often over 12 mmol/L [216 mg/dl] and sometimes over 15 mmol/L [270 mg/dl].

I have also now read that the effect of Levemir is different (shorter and therefore I presume more intense) if it is given at night. This has never been mentioned to us before. Our health care professionals said, at first, that he should not need a snack at bedtime. However, without it he would go severely low, unless he had so little Levemir that his day time control was terrible.

I am not sure I understand what we're being asked to do. For example, we were suggested to carbohydrate count with variable meals (what the ADA calls advanced counting) with no guidance or training or follow up about how to work out the carbohydrate to insulin ration or his sensitivity factor. We were just told to try 1:15, which was far too little, and go from there.

While my son is at school, we currently give nine units in the morning for breakfast for 105 grams of carbohydrates and 10 units for a lunch of 120 grams of carbohydrates, and for tea we use a ratio of 1:10 plus three units, as it seems the Levemir is running out in the evening. When he is not at school and less active, we add two units at breakfast and lunch. My son weighs 60 kg (132 pounds). His last A1c was 7.4.

Lastly, we have worked out some guidelines that work at least half the time, but control is not as good as it should be. Do you have any ideas?


We do not have Levemir in the U.S., so I have no personal experience with it, but I have used Lantus (Glargine) a lot, which is similar. It sounds like your son may need to decrease the Levemir to prevent nighttime lows, but then increase the daytime insulin to prevent highs during the day. Fixed insulin/carbohydrate ratios do not always work. Some patients need more insulin for the same amount of carbohydrates at different times of the day. He may need different insulin to carbohydrate ratios at different meals. If the Levemir isn't lasting the entire 24 hours, you could try splitting it into two doses, morning and evening, or, alternatively, you could just give more fast acting insulin later in the day as "supplemental basal" in addition to the fast acting insulin he takes to match his food. Sometimes it is easier to give this "supplemental" daytime basal as Regular (I believe this is Actrapid in the UK) mixed with the NovoRapid (aspart) given to cover the meals or, you could add a fixed amount of NovoRapid for supplemental basal to the NovoRapid he takes for meals.

Although it is nice to try and figure out a formula to calculate how much insulin you need for different amounts of food (insulin to carbohydrate ratio), sometimes it is easier to go back temporarily to a fixed meal plan, eating the same amount of food at the same time each day and figure out how much insulin you need for this fixed meal plan. Then, you can experiment with varying the amounts you eat and see how much you need to change the insulin. If his blood sugars aren't dangerously high, you also might want to hold off temporarily giving extra insulin for high blood sugars while you figure out the basal and insulin/carbohydrate ratios. After you establish the best baseline basal and mealtime insulins, you can determine how much insulin it takes to correct a high blood sugar.


Additional comments from Dr. Stuart Brink:

It is possible that you are using the wrong snack at bedtime. For instance, why not try some high fat ice cream (not sugar-free)? The fat will make the sugars last for several hours and, perhaps, provide just the balance you need for overnight insulin peaks. We also have some youngsters who do significantly better with morning rather than night time long acting analogs like glargine or Levemir. The only way to find out would be to experiment and do so with lots of night time monitoring of blood glucose levels, i.e. every two hours, so that there is no risk of unexpected hypoglycemia while asleep.

You should feel comfortable enough to go back to your diabetes team and request more information so that you do not have to do all the learning on your own. Initiating carbohydrate counting with a 1:15 ratio of insulin units to grams of carbohydrates is quite reasonable for most teenagers, but everyone is an individual and must have these insulin to carbohydrate ratios adjusted for themselves. Once again, it is not so uncommon to have 1:10 or 1:12 rather than 1:20. In younger children, sometimes the ratio is 1:30 or even less. All is done not with dogma, but with individual blood glucose monitoring pre and post-prandially to learn.

You may also want to try glargine rather than Levemir to see if this provides any better basal coverage. Glargine is a bit longer lasting than Levemir in many people, but often also does not last a full 24 hours.


Additional comments from Dr. Linda DiMeglio:

We do not have Levemir available yet in the U.S., but our experience here with Lantus (insulin glargine) suggests that some children will have hypoglycemia when it is given as a single bedtime dose (perhaps because of overlapping of doses that last more than 24 hours, perhaps because of lower insulin requirements at night. In these cases, I have moved the glargine to breakfast. If the glargine at breakfast appears to wear off with overnight highs, I then have split the glargine 50/50 between morning and evening. Sometimes this works. I would venture that the same things might help with detemir.


Additional comments from Jane Seley, diabetes nurse specialist:

I recommend you look into insulin pump therapy where you can truly regulate the basal (background) insulin and get better results. It's a shame to work so hard and not have tighter control.


[Editor's comment: Information on possibly splitting the dose between the morning and evening is also discussed briefly on the Diabetes UK web site. BH]

Original posting 27 Feb 2005
Posted to Hypoglycemia and Insulin Analogs


  Home Return to Top

Last Updated: Tuesday April 06, 2010 15:10:00
This Internet site provides information of a general nature and is designed for educational purposes only. If you have any concerns about your own health or the health of your child, you should always consult with a physician or other health care professional.

This site is published by T-1 Today, Inc. (d/b/a Children with Diabetes), a 501c3 not-for-profit organization, which is responsible for its contents.
By using this site, you agree to our Terms of Use, Legal Notice, and Privacy Policy.
© Children with Diabetes, Inc. 1995-2015. Comments and Feedback.