Insulin doses

This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.

quickfeet

Smooth Operator
10+ Year Member
Joined
Dec 2, 2010
Messages
154
Reaction score
90
When calculating insulin carb ratios, do you do
500 ÷ TDD = units of aspart (Novolog®) or lispro (Humalog®) per 1g of carbohydrate

or is it:

500 ÷ (0.5 x TDD)


Reason I am confused is because I thought you only give 50% of the TDD as bolus, the other 50% is the basal dose (e.g. Lantus).

Would be grateful if @Raryn could provide help

Members don't see this ad.
 
When calculating insulin carb ratios, do you do
500 ÷ TDD = units of aspart (Novolog®) or lispro (Humalog®) per 1g of carbohydrate

or is it:

500 ÷ (0.5 x TDD)


Reason I am confused is because I thought you only give 50% of the TDD as bolus, the other 50% is the basal dose (e.g. Lantus).

Would be grateful if @Raryn could provide help
The total daily dose.

So the carb ratio rule of thumb would be your magic number (people use either 450 or 500, I typically use 450) divided by the total amount of insulin the patient requires per day between basal and bolus. There's all kinds of variations on this formula, including things like multiplying the weight by some number to get your magic number instead of using 450-500 all the way up to weird formulas with various correction factors I don't understand.

The insulin sensitivity factor rule of thumb would be your magic number (people use anything from 1500-2000, I typically use 1800) divided by the same total daily dose. There's similar variations here too.

So lets say someone is 100kg and you anticipate a total daily dose of 50 units of insulin to start. You can say something like use 25 glargine (or levemir or degludec) daily and 8 aspart (or lispro or glulisine) with each meal. But they're totally gung ho and ready to carbohydrate count. Then you can say use a carb ratio of 450/50= 1:9. If they also want to use a sensitivity factor, this same patient would be 1:36.

The assumptions here are manyfold:

1. 50% basal/50% bolus is an easy rule of thumb to remember but is often quite wrong for any individual patient. A young, active patient may need 60% bolus insulin. Most type 1s who are actually adherent to all your recommendations use a majority bolus insulin. A sedentary, insulin resistant patient may need 60% basal. It's when you get much outside the 40-60% range for a given indication that it gives you a clue that something is truly off.
2. The starting dose (0.2-0.5 unit/kg total daily dose) is almost always an underestimate of true needs, especially in a type 2. A typical type 2 might end up needing closer to 1 unit/kg total daily dose. The problem is that while that may be an average, if you started everyone on that you'd cause half your patients to be hypoglycemic, so we always prefer to undershoot than overshoot.
3. The ISF and ICR calculations above (as seen by the umpteen variations available) are guesstimates at best. They give you a general starting point, but everyone is unique (and often patients have varying needs even over the course of the same day!). You can assume all the rules of thumb are right, but really, after you pick your initial doses for someone, you should be adjusting them based on real results.

For example, if you extrapolate enough you'll see that the assumptions also boil down to our patients eating ~70g of carbs with each meal. Otherwise either the carb ratio rule of thumb will be off or the 50/50 rule will be wrong. Well, many patients eat significantly less or more than 70g of carbohydrates with each meal. That will throw a wrench and make at least some of our rules of thumb wrong for each patient.

It gets even more complicated when you consider that with a fixed dose, the patient should eat a consistent amount of carbs with each meal, and with a carb ratio the patient should accurately count carbs. Neither of which happen consistently in real life. Mistakes with carb counting and meal size probably lead to more screwy blood sugars than doses being slightly off for the vast majority of our patients.
 
  • Like
Reactions: 4 users
The total daily dose.

So the carb ratio rule of thumb would be your magic number (people use either 450 or 500, I typically use 450) divided by the total amount of insulin the patient requires per day between basal and bolus. There's all kinds of variations on this formula, including things like multiplying the weight by some number to get your magic number instead of using 450-500 all the way up to weird formulas with various correction factors I don't understand.

The insulin sensitivity factor rule of thumb would be your magic number (people use anything from 1500-2000, I typically use 1800) divided by the same total daily dose. There's similar variations here too.

So lets say someone is 100kg and you anticipate a total daily dose of 50 units of insulin to start. You can say something like use 25 glargine (or levemir or degludec) daily and 8 aspart (or lispro or glulisine) with each meal. But they're totally gung ho and ready to carbohydrate count. Then you can say use a carb ratio of 450/50= 1:9. If they also want to use a sensitivity factor, this same patient would be 1:36.

The assumptions here are manyfold:

1. 50% basal/50% bolus is an easy rule of thumb to remember but is often quite wrong for any individual patient. A young, active patient may need 60% bolus insulin. Most type 1s who are actually adherent to all your recommendations use a majority bolus insulin. A sedentary, insulin resistant patient may need 60% basal. It's when you get much outside the 40-60% range for a given indication that it gives you a clue that something is truly off.
2. The starting dose (0.2-0.5 unit/kg total daily dose) is almost always an underestimate of true needs, especially in a type 2. A typical type 2 might end up needing closer to 1 unit/kg total daily dose. The problem is that while that may be an average, if you started everyone on that you'd cause half your patients to be hypoglycemic, so we always prefer to undershoot than overshoot.
3. The ISF and ICR calculations above (as seen by the umpteen variations available) are guesstimates at best. They give you a general starting point, but everyone is unique (and often patients have varying needs even over the course of the same day!). You can assume all the rules of thumb are right, but really, after you pick your initial doses for someone, you should be adjusting them based on real results.

For example, if you extrapolate enough you'll see that the assumptions also boil down to our patients eating ~70g of carbs with each meal. Otherwise either the carb ratio rule of thumb will be off or the 50/50 rule will be wrong. Well, many patients eat significantly less or more than 70g of carbohydrates with each meal. That will throw a wrench and make at least some of our rules of thumb wrong for each patient.

It gets even more complicated when you consider that with a fixed dose, the patient should eat a consistent amount of carbs with each meal, and with a carb ratio the patient should accurately count carbs. Neither of which happen consistently in real life. Mistakes with carb counting and meal size probably lead to more screwy blood sugars than doses being slightly off for the vast majority of our patients.
Thanks for this information.

I have a question about patients on continuous tube feeds. Has there ever been shown one way or other that just doing 100% basal TDD is inferior to splitting into q4h rapid acting or regular insulin doses with small percent of basal?

I also have a question about diabetics on short courses of glucocorticoids. This lecturer here suggests giving patients NPH with their morning dose of prednisone, at like 0.40 units per KG of NPH for 40mg of Prednisone. First question related to this is - have you ever heard of such a suggestion? And second question - provided one did follow this strategy, would you give these people this additional NPH along with their already prescribed long-acting insulin and bolus insulin? In other words, you'd be giving them 3 different kinds of insulin while they were on the steroid course...
 
Members don't see this ad :)
The total daily dose.

So the carb ratio rule of thumb would be your magic number (people use either 450 or 500, I typically use 450) divided by the total amount of insulin the patient requires per day between basal and bolus. There's all kinds of variations on this formula, including things like multiplying the weight by some number to get your magic number instead of using 450-500 all the way up to weird formulas with various correction factors I don't understand.

The insulin sensitivity factor rule of thumb would be your magic number (people use anything from 1500-2000, I typically use 1800) divided by the same total daily dose. There's similar variations here too.

So lets say someone is 100kg and you anticipate a total daily dose of 50 units of insulin to start. You can say something like use 25 glargine (or levemir or degludec) daily and 8 aspart (or lispro or glulisine) with each meal. But they're totally gung ho and ready to carbohydrate count. Then you can say use a carb ratio of 450/50= 1:9. If they also want to use a sensitivity factor, this same patient would be 1:36.

The assumptions here are manyfold:

1. 50% basal/50% bolus is an easy rule of thumb to remember but is often quite wrong for any individual patient. A young, active patient may need 60% bolus insulin. Most type 1s who are actually adherent to all your recommendations use a majority bolus insulin. A sedentary, insulin resistant patient may need 60% basal. It's when you get much outside the 40-60% range for a given indication that it gives you a clue that something is truly off.
2. The starting dose (0.2-0.5 unit/kg total daily dose) is almost always an underestimate of true needs, especially in a type 2. A typical type 2 might end up needing closer to 1 unit/kg total daily dose. The problem is that while that may be an average, if you started everyone on that you'd cause half your patients to be hypoglycemic, so we always prefer to undershoot than overshoot.
3. The ISF and ICR calculations above (as seen by the umpteen variations available) are guesstimates at best. They give you a general starting point, but everyone is unique (and often patients have varying needs even over the course of the same day!). You can assume all the rules of thumb are right, but really, after you pick your initial doses for someone, you should be adjusting them based on real results.

For example, if you extrapolate enough you'll see that the assumptions also boil down to our patients eating ~70g of carbs with each meal. Otherwise either the carb ratio rule of thumb will be off or the 50/50 rule will be wrong. Well, many patients eat significantly less or more than 70g of carbohydrates with each meal. That will throw a wrench and make at least some of our rules of thumb wrong for each patient.

It gets even more complicated when you consider that with a fixed dose, the patient should eat a consistent amount of carbs with each meal, and with a carb ratio the patient should accurately count carbs. Neither of which happen consistently in real life. Mistakes with carb counting and meal size probably lead to more screwy blood sugars than doses being slightly off for the vast majority of our patients.
Thank you for clearing that up for me.

Not to pile on with the questions, but could you answer a related question? What is the rational for some people dosing Lantus in the AM and the PM? E.g. splitting the dose.

Additionally - is the glargine to degludec conversion 1:1? I've read lately that degludec is actually associated with less hypoglycemia than Lantus and want to see about prescirbing this more often.
 
Thank you for clearing that up for me.

Not to pile on with the questions, but could you answer a related question? What is the rational for some people dosing Lantus in the AM and the PM? E.g. splitting the dose.

Additionally - is the glargine to degludec conversion 1:1? I've read lately that degludec is actually associated with less hypoglycemia than Lantus and want to see about prescirbing this more often.
A) Two reasons:

1) Lantus absorption gets more inconsistent at higher doses. The way that it works is that Lantus after injection microprecipitates and gets absorbed slowly from the subq. A bigger bolus leads to issues where there's a lower surface area, and absorption becomes kinda hinky. Thus, at some point above 50 units/day (ask 10 endocrinologists and you'll get 10 answers between 50 and 80), it's better to split the Lantus into two separate injections. At that point, you either split the Lantus into two injections given at the same time or to one injection given twice a day. Which leads to reason #2:
2) Kinetics of all of our insulins are based on averages. Lantus lasts 24 hours in theory, but there exist people for whom it might only last 20 hours. Or 26 hours. 26 hours is no big deal (a tiny bit of stacking won't kill you), but if you're sensitive, 4 hours without basal can lead to issues. In that case, splitting the dose definitely gives you at least some coverage all day long.

So most people who have it split have it split because of a high dose. A few on a low dose might have it split because they just want to make sure they have consistent coverage. On inpatients, it might also be split because that gives you two opportunities each day to modify the next dose (thus letting you stay more on top of things).

Problem #2 is actually much more pronounced with detemir btw. At low doses, detemir will almost never last a full day and has a bit of a peak besides. I'll almost always use it bid if I'm stuck using the drug.

B) The manufacturer says glargine to degludec conversion is exactly 1:1. In studies, to achieve the same fasting plasma glucose the degludec dose actually tends to be a little bit higher on average. Both of those said though, if the patient is well-controlled previously, whenever I switch from one insulin to another I decrease the dose by 10-20%. Why? Quite simply, people can react differently to different medications. I'd rather underdose them initially and then go up later rather than risk hypoglycemia. I've had plenty of patients who insist they need radically different doses with novolog as compared to humalog, which makes zero sense to me based on any actual evidence I know of, but I just go with it as long as they're doing otherwise OK.

Degludec in SWITCH-1 and SWITCH-2 clearly has less hypoglycemias than glargine. It also definitely has kinetics that last well over a day (leading to no issues with the "might only last 20 hours" problem), can be given at a different time every day without negative effects (thus being better for our patients who have variable schedules), and is available in a U200 formulation that lets you go up to 160 units per dose (as opposed to even U300 glargine which still has a maximum dose of 80 units).

That said, it can't be adjusted as often (takes something like a week to reach steady state), leads to issues if the patient is admitted (never on formulary=awkward period of overlap during the initial admission), tends to have worse coverage (for governmental insurances, commercially insured patients can just get a discount card), and hasn't been around as long.

I've been using more and more degludec recently myself, but it's not for every patient.
 
  • Like
Reactions: 2 users
This is the last answer tonight because I gotta get to sleep

Thanks for this information.

I have a question about patients on continuous tube feeds. Has there ever been shown one way or other that just doing 100% basal TDD is inferior to splitting into q4h rapid acting or regular insulin doses with small percent of basal?
No. There has never been a conclusive answer to this. You ask 10 endocrinologists and you'll get 12 answers. I hate continuous tube feeding and TPN. Both are unphysiologic and just a pain in my ass to deal with.

The reasoning in general is that if the tube feeding (or TPN) is ever interrupted and you're dosing 100% basal, you're screwed. Lets say the patient pulls their NG tube. You can't take the insulin out of them. You either then need to put the patient on high dose D10 infusion or something similar until you can get the tube feedings going again.

OTOH, if you have a shorter acting insulin dosed more often, you'll often be able to get away with just holding or halving the next dose of insulin to better avoid getting yourself in trouble. So most will advocate that at least some of your insulin should be dosed as rapid acting, regular, or NPH just to give you that flexibility in case of interruption of the tube feeding.

I've seen so many different wacky strategies from board certified endocrinologists though:
A) Daily glargine plus q6h regular
B) q8h NPH
C) daily glargine plus q8h NPH
D) Twice daily glargine
E) q6h regular plus q6h NPH (!@#$ if I knew why)
F) daily glargine plus q6h rapid acting
etc.

All of the above include a correction scale of course. My personal preference at this point is probably half the insulin should be basal (such as glargine) and the other half something like q6h regular (with instructions to hold the regular if the tube feeding is interrupted), but I just go with whatever the boss wants at this point. For this particular question, I've stopped asking why.

The job I accepted is outpatient only, and leaving continuous tube feeding behind is one of the things I look forward to the most.

I also have a question about diabetics on short courses of glucocorticoids. This lecturer here suggests giving patients NPH with their morning dose of prednisone, at like 0.40 units per KG of NPH for 40mg of Prednisone. First question related to this is - have you ever heard of such a suggestion? And second question - provided one did follow this strategy, would you give these people this additional NPH along with their already prescribed long-acting insulin and bolus insulin? In other words, you'd be giving them 3 different kinds of insulin while they were on the steroid course...

That strategy (give NPH with prednisone in addition to regular basal/bolus insulin) has been studied and seems to work. Just going up on their basal/bolus (particularly their bolus) has also been studied and seems to work. Neither is particularly better than the other. The last time I remember looking this up, there were a handful of small papers, and the highest quality one (that still only had ~30 people in each group) had slightly better sugar control in the NPH group but more hypoglycemias. I don't use it as a strategy, I just go up a bit on their basal and more on their bolus (~70% bolus insulin isn't unusual for someone on high dose steroids) until I get the numbers under control.

The other thing to keep in mind is the NPH kinetics only match once daily prednisone. If it's any other steroid or any other dosing schedule, the NPH doesn't make any sense.
 
  • Like
Reactions: 2 users
This is the last answer tonight because I gotta get to sleep


No. There has never been a conclusive answer to this. You ask 10 endocrinologists and you'll get 12 answers. I hate continuous tube feeding and TPN. Both are unphysiologic and just a pain in my ass to deal with.

The reasoning in general is that if the tube feeding (or TPN) is ever interrupted and you're dosing 100% basal, you're screwed. Lets say the patient pulls their NG tube. You can't take the insulin out of them. You either then need to put the patient on high dose D10 infusion or something similar until you can get the tube feedings going again.

OTOH, if you have a shorter acting insulin dosed more often, you'll often be able to get away with just holding or halving the next dose of insulin to better avoid getting yourself in trouble. So most will advocate that at least some of your insulin should be dosed as rapid acting, regular, or NPH just to give you that flexibility in case of interruption of the tube feeding.

I've seen so many different wacky strategies from board certified endocrinologists though:
A) Daily glargine plus q6h regular
B) q8h NPH
C) daily glargine plus q8h NPH
D) Twice daily glargine
E) q6h regular plus q6h NPH (!@#$ if I knew why)
F) daily glargine plus q6h rapid acting
etc.

All of the above include a correction scale of course. My personal preference at this point is probably half the insulin should be basal (such as glargine) and the other half something like q6h regular (with instructions to hold the regular if the tube feeding is interrupted), but I just go with whatever the boss wants at this point. For this particular question, I've stopped asking why.

The job I accepted is outpatient only, and leaving continuous tube feeding behind is one of the things I look forward to the most.



That strategy (give NPH with prednisone in addition to regular basal/bolus insulin) has been studied and seems to work. Just going up on their basal/bolus (particularly their bolus) has also been studied and seems to work. Neither is particularly better than the other. The last time I remember looking this up, there were a handful of small papers, and the highest quality one (that still only had ~30 people in each group) had slightly better sugar control in the NPH group but more hypoglycemias. I don't use it as a strategy, I just go up a bit on their basal and more on their bolus (~70% bolus insulin isn't unusual for someone on high dose steroids) until I get the numbers under control.

The other thing to keep in mind is the NPH kinetics only match once daily prednisone. If it's any other steroid or any other dosing schedule, the NPH doesn't make any sense.
Should sliding scales on inpatients begin at 150 mg/dL or 180 mg/dL ?
Our VA and regular hospital defaults to 150 but I always change it, and people make fun of me. But that's what the NICE SUGAR trial was about, right?
 
Should sliding scales on inpatients begin at 150 mg/dL or 180 mg/dL ?
Our VA and regular hospital defaults to 150 but I always change it, and people make fun of me. But that's what the NICE SUGAR trial was about, right?
No evidence.

NICE SUGAR randomized people to 81-108 vs 140-180 and said 140-180 was better. The older trials (particularly van der berghe medical and surgical trials) compared 80-110 with no insulin unless >215 and said 80-110 was better (sort of, the evidence was mixed). No one has ever compared the in between numbers (say 100-140 with 140-180), and it's perfectly reasonable to guess there's a J shaped curve with the best number being somewhere in the mid 100s. These studies are expensive to do though, so I doubt we'll get a definitive answer other than "avoid going too high or too low". The sepsis guidelines of 140-180 or the ACP guideline of 140-200 is fine.

Regardless, the question of how to formulate the correction inpatient correction scale is probably minimally important. Our main hospital starts scales at 150, our VA starts it at 200. If you're picking the correct scale, it will be safe either way. Or with 180 (which I haven't personally used, but also makes some sense). The problem is that lots of doctors just reflexively go up on the correction scale if patient isn't correcting, when it is far more likely that the scheduled insulin is simply inadequate.

The strength of the scale is more important than the target IMO. A general rule of thumb is this: If the patient is on a total daily dose of insulin <40 units/day, use the low dose (1:50) sliding scale. If they're on 40-100 units/day, use the moderate dose (2:50) correction scale. And if they're on >100 units/day, use the high dose (3:50) correction scale. Never correct more often than every 4 hours, and don't do a correction within 4 hours of a mealtime insulin dose.
 
Last edited:
  • Like
Reactions: 2 users
No evidence.

NICE SUGAR randomized people to 81-108 vs 140-180 and said 140-180 was better. The older trials (particularly van der berghe medical and surgical trials) compared 80-110 with no insulin unless >215 and said 80-110 was better (sort of, the evidence was mixed). No one has ever compared the in between numbers (say 100-140 with 140-180), and it's perfectly reasonable to guess there's a J shaped curve with the best number being somewhere in the mid 100s. These studies are expensive to do though, so I doubt we'll get a definitive answer other than "avoid going too high or too low". The sepsis guidelines of 140-180 or the ACP guideline of 140-200 is fine.

Regardless, the question of how to formulate the correction inpatient correction scale is probably minimally important. Our main hospital starts scales at 150, our VA starts it at 200. If you're picking the correct scale, it will be safe either way. Or with 180 (which I haven't personally used, but also makes some sense). The problem is that lots of doctors just reflexively go up on the correction scale if patient isn't correcting, when it is far more likely that the scheduled insulin is simply inadequate.

The strength of the scale is more important than the target IMO. A general rule of thumb is this: If the patient is on a total daily dose of insulin <40 units/day, use the low dose (1:50) sliding scale. If they're on 40-100 units/day, use the moderate dose (2:50) correction scale. And if they're on >100 units/day, use the high dose (3:50) correction scale. Never correct more often than every 4 hours, and don't do a correction within 4 hours of a mealtime insulin dose.
Never do a correction within 4 hours of a bolus insulin dose? When are you supposed to give correction insulin?
 
Never do a correction within 4 hours of a bolus insulin dose? When are you supposed to give correction insulin?
Sorry if that wasn't clear.

At the same time as the bolus is fine. Or before the bolus. What you shouldn't do is say, give the mealtime insulin with dinner at 5pm, then check their sugar at 7pm and give a correction then. Because at 7pm, some of the 5pm insulin still hasn't been absorbed, so you're not taking into account the insulin already on board.

Thus, if dinner insulin is planned at 5pm, you need to give the correction either with it at 5pm or wait until 9pm or you risk stacking. Most of the time at the hospital they're given together, but sometimes trays come at wacky times and you end up with a bunch of shots overlapping.
 
  • Like
Reactions: 3 users
The strength of the scale is more important than the target IMO. A general rule of thumb is this: If the patient is on a total daily dose of insulin <40 units/day, use the low dose (1:50) sliding scale. If they're on 40-100 units/day, use the moderate dose (2:50) correction scale. And if they're on >100 units/day, use the high dose (3:50) correction scale. Never correct more often than every 4 hours, and don't do a correction within 4 hours of a mealtime insulin dose.

Just so I have this clear... if they are on a TDD of 60 (30 Lantus with 10 aspart mealtimes x 3), then the correction scale you'd use is:

1800 ÷ TDD = 1800 ÷ 60 = 30
So the correction scale would NOT be 1 unit for every 30 mg/dL above 150 mg/dL, but 2 units for every 30... in other words:

150-180 = 2 units
181-211 = 4 units
212-242 = 6 units

etc....
 
Just so I have this clear... if they are on a TDD of 60 (30 Lantus with 10 aspart mealtimes x 3), then the correction scale you'd use is:

1800 ÷ TDD = 1800 ÷ 60 = 30
So the correction scale would NOT be 1 unit for every 30 mg/dL above 150 mg/dL, but 2 units for every 30... in other words:

150-180 = 2 units
181-211 = 4 units
212-242 = 6 units

etc....
No. The correction scale would be 1:30 as per your calculation. So ideally, 151-180 give 1 unit, 181-210 give 2 units, etc.

But...

Every individual patient on the wards having a customized scale would be confusing as hell for the nurses, and difficult to order for the doctors. So hospitals tend to have some preset scales, the most common example of which would be a "low" "moderate" and "high" scale... which correspond to a 1:50, 2:50 (or 1:25), and 3:50 (or 1:17).

Ex:
151-200 give 1/2/3 units for a L/M/H scale respectively
201-250 give 2/4/6 units
etc

These scales aren't perfect, but they're typically "good enough".
 
  • Like
Reactions: 2 users
Hey @Raryn , what is the correct way to transition a patient from insulin drip to SUBQ? This is what I have:

1. Give patient dose of basal insulin while the insulin infusion is running

2. Two hours later, stop the insulin infusion

3. Right after the infusion is stopped, allow the patient to eat

4. Check POC glucose after first bite of the meal and administer the first dose of bolus (prandial) insulin


Do you have any other tips or is there anything wrong with the above?
 
No. The correction scale would be 1:30 as per your calculation. So ideally, 151-180 give 1 unit, 181-210 give 2 units, etc.

But...

Every individual patient on the wards having a customized scale would be confusing as hell for the nurses, and difficult to order for the doctors. So hospitals tend to have some preset scales, the most common example of which would be a "low" "moderate" and "high" scale... which correspond to a 1:50, 2:50 (or 1:25), and 3:50 (or 1:17).

Ex:
151-200 give 1/2/3 units for a L/M/H scale respectively
201-250 give 2/4/6 units
etc

These scales aren't perfect, but they're typically "good enough".
Ah... so the Low/Medium/High Scale is just an alternative to the 1800 Rule thing?
 
Hey @Raryn , what is the correct way to transition a patient from insulin drip to SUBQ? This is what I have:

1. Give patient dose of basal insulin while the insulin infusion is running

2. Two hours later, stop the insulin infusion

3. Right after the infusion is stopped, allow the patient to eat

4. Check POC glucose after first bite of the meal and administer the first dose of bolus (prandial) insulin


Do you have any other tips or is there anything wrong with the above?
That works just fine. You don't have to quite so obsessive regarding the timing. One or the other overlap strategy works.

You need to keep the infusion after the first dose of sc insulin going to allow for some overlap. So your options are

1) Give SC basal. It takes up to a couple hours to start getting absorbed, so turn off the drip two hours later.
2) Feed them a meal, give them SC rapid acting. It will start getting absorbed in about 10-15 minutes and last for 4 hours, so you can turn the drip off 30 minutes after you give them the rapid acting and it will be fine. You should give them the basal at the same time here (or close to it), but no need to wait the full 2 hours.

Most people do the basal transition strategy, but the latter makes physiologic sense and has been described in the literature. Regardless, you don't always have to do both. They must have basal on board once they're off the drip, but for the actual moment of turning it off, they just need some kind of sc insulin in their system.

The whole keep them NPO while on IV insulin and transition them when they start eating (but don't allow them to eat before!) thing is also kinda silly. You can eat on an insulin drip, it just makes adjusting the rates a bit more of a pain for the nurse (unless you also schedule SC mealtime insulin, which is also confusing for the nurse). And you can be NPO on SC insulin, so you can still transition someone who isn't eating. Half the time this is just hospital policy.

Ah... so the Low/Medium/High Scale is just an alternative to the 1800 Rule thing?

Yes. The rule of 1800 is typically used for insulin pumps and when you're making the patient a custom scale for their own use (often with the help of an app or smart meter). The 1:50, 2:50, 3:50 preset scales are a simpler alternative we give patients who just stick a table on their fridge and that we use in the hospital.
 
  • Like
Reactions: 1 users
That works just fine. You don't have to quite so obsessive regarding the timing. One or the other overlap strategy works.

You need to keep the infusion after the first dose of sc insulin going to allow for some overlap. So your options are

1) Give SC basal. It takes up to a couple hours to start getting absorbed, so turn off the drip two hours later.
2) Feed them a meal, give them SC rapid acting. It will start getting absorbed in about 10-15 minutes and last for 4 hours, so you can turn the drip off 30 minutes after you give them the rapid acting and it will be fine. You should give them the basal at the same time here (or close to it), but no need to wait the full 2 hours.

This is what I've told the nurses to do, I dont know if its correct:

1. Allow patient to eat

2. Give patient the first dose of basal insulin

3. Two hours after the first dose of basal insulin, turn off the insulin infusion

4. At the next meal, give the first dose of bolus (prandial) insulin
 
Top