Inpatient and Outpatient Insulin Therapy in Treating Diabetes Mellitus

CARLOS R. HAMILTON, JR., MD, FACE: Hello, I'm Dr. Carlos Hamilton and welcome to the AACE Clinical Conversations. In this program, we will discuss insulin therapy for patients with diabetes mellitus in the inpatient and outpatient settings.

Joining me today is Dr. Alan Garber, Professor of Medicine and Professor of Biochemistry and Molecular Biology, Professor of Molecular and Cellular Biology at the Baylor College of Medicine and our colleague, Dr. Glenn Cunningham, a Professor of Medicine, Professor of Molecular and Cellular Biology at Baylor and the Medical Director of the St. Luke's Hospital, Baylor College of Medicine diabetes program. Thanks to both of you for joining us here today.

We all know that the management of diabetics in the inpatient setting whether they are type 1 or type 2 diabetics, differs in many ways from the management in the outpatient setting.

We have come to understand that the very careful control of the blood sugar in many inpatient circumstances in the hospital is of real benefit to the patients. Alan, what has been your experience, and what is the experience in the literature with the outcomes related to the degree of diabetic control?

ALAN GARBER, MD, PhD, FACE: In the last 10 years Carlos, there has been an amazing amount of literature that has indicated reduced morbidity and mortality for diabetic patients who are post-cardiac surgery, who are in the ICU regardless of the reason, in medical ICU patients and in patients who have illnesses such as sepsis, who clearly benefit from close control of the diabetes while in the intensive care unit.

And there is good inferential evidence that those benefits in terms of reduced morbidity and mortality are also bulwarked by good diabetes control in the non-ICU inpatient setting as well.

CARLOS R. HAMILTON, JR., MD, FACE: One of the problems that has always occurred in the inpatient setting is the tendency of many of our colleagues to write sliding scale insulin orders. We all know that this is almost universally infective in terms of really achieving any degree of control. Glenn, I'm familiar with your program at St. Luke's Hospital in instituting an intensive insulin control program throughout the hospital. Tell us a little bit about it and how you were able to achieve this.

GLENN CUNNINGHAM, MD: Well, let me start by saying I agree 100% with you that sliding scale insulin protocols are not what we need for inpatient management of diabetes or hyperglycemia. That really only leads to increased either severe hyperglycemia or hypoglycemia, so we really need to discourage that, although lots of physicians still use it. So I think that we would like to see some additional data supporting the use of inpatient, and particularly ICU as well as floor management of hyperglycemia and diabetes. But there is enough data at the present time, I think, to warrant trying to bring hospitals into a mode where we do try to intensify glucose management.

CARLOS R. HAMILTON, JR., MD, FACE: And this includes primarily continuous intravenous insulin treatment. Is that correct?

GLENN CUNNINGHAM, MD: Yes, I think that's true. For patients who are admitted to intensive care units, that is definitely true, that the use of insulin infusion protocols is really the way to go to achieve target glucoses in the 80-110 or 80-120 range, which is where the data is showing efficacy of treatment.

CARLOS R. HAMILTON, JR., MD, FACE: Yes.

GLENN CUNNINGHAM, MD: Now one might consider that there is increased risk of hypoglycemia, and indeed there would be if one did not monitor these patients at sufficient, frequent intervals. But I think if you use any one of a number of infusion protocols, and if nurses are educated to follow the protocol sufficiently, then it's a relatively safe effort.

It does require, first of all, convincing a hospital administration that it's worthwhile. So I think you have to first bring them into the fold. And you can do this by pointing to the studies that Dr. Garber mentioned about reducing morbidity and mortality.

You also could bring them into the fold by talking about lowering, reducing hospital stay and lowering hospital costs. Those are big issues, and programs now are beginning to show that this kind of effort does bear fruit.

CARLOS R. HAMILTON, JR., MD, FACE: There really has been some very clearcut evidence, has there not, that the intensive control of the blood sugar in the intensive care unit after surgery, especially after major cardiovascular surgery and so forth, really does shorten the duration of hospitalization and it also prevents complications, such as infections and so forth that sometimes occur. Is that not correct?

GLENN CUNNINGHAM, MD: That is correct, so there is data, particularly...

CARLOS R. HAMILTON, JR., MD, FACE: There is a real economic benefit to the hospital, as well as to the patient

GLENN CUNNINGHAM, MD: Right. Right. I think the data in the medical ICU area needs to be amplified, and so there are ongoing large clinical trials now trying to resolve that, but...

CARLOS R. HAMILTON, JR., MD, FACE: So how do you go about instituting this in a hospital, other than talking to the...

GLENN CUNNINGHAM, MD: I think you get the administration support, and then you really need to have an organizational effort in which you involve all the disciplines in the hospital that are involved with diabetic patient care. And so once you bring those people into an organizational effort and get their buy-in into both an infusion protocol -- we developed a transition protocol to convert patients from insulin infusion to sub-Q insulin.

And then I think it's important to have kind sort of a diabetes order sheet, which helps to minimize the use of sliding scale insulin and helps to get people in the frame of thinking about all the things that a patient with diabetes needs to have addressed.

CARLOS R. HAMILTON, JR., MD, FACE: When I have used your order sheet at St. Luke's Hospital, the one thing the nurses always express concern and anxiety about is low blood sugar...

GLENN CUNNINGHAM, MD: Right.

CARLOS R. HAMILTON, JR., MD, FACE: ...because you're continuing to give them some insulin even though their blood sugars may be in a very acceptable range. Have you actually seen problems with severe hypoglycemia in using this protocol?

GLENN CUNNINGHAM, MD: So with the order sheet, we're talking about using a basal and a bolus kind of insulin regimen. And in general, I think that it is appropriate for use on the floors and in the units after patients have been transitioned. It really depends on sort of what kind of nutritional support they're receiving as to how you organize this. But in general, you need both basal insulin kind of rapid-acting or short-acting insulin onboard.

CARLOS R. HAMILTON, JR., MD, FACE: Well, I think that's a good transition to talk a little bit about the outpatient management of our patients with insulin, both type 1s and type 2s.

Alan, obviously type 1s have an absolute requirement for insulin. But what is your general approach to these patients in terms of the way you intensively manage their diabetes?

ALAN GARBER, MD, PhD, FACE: Well, I think type 1s who are absolutely insulin deficient require both basal or inter-meal insulin and mealtime insulin. It can be administered in terms of all injectable protocols, a combination of inhaled and injectable insulin, or by continuous subcutaneous insulin infusions. And those are really the only three approaches to be used.

I think any other way to deliver insulin to patients with type 1 diabetes, the use of only one form of insulin, such as basal, or the use of oral agents really should be condemned. These are just not things that are useful, even in the early stages of type 1 diabetes.

CARLOS R. HAMILTON, JR., MD, FACE: So for the most part, you use a long-acting insulin given once-daily generally.

ALAN GARBER, MD, PhD, FACE: Or twice-daily.

CARLOS R. HAMILTON, JR., MD, FACE: And then a short-acting insulin given before meals depending upon what the blood sugar is prior to the meal. Is that generally the protocol you follow?

ALAN GARBER, MD, PhD, FACE: I like to use a rapid-acting insulin analog before meals, either by injection or through a pump. Now, of course, we have inhaled insulin, which it does for the most part get in the bloodstream quickly, so that might be an alternative.

For the inter-meal or basal insulin, which is about half the total daily dose or a little less, the doses in type 1s who are thin is frequently so small that you do not necessarily get full 24 hour coverage from a single dose. So oftentimes, I wind up having to use two small doses.

CARLOS R. HAMILTON, JR., MD, FACE: So using the long-acting insulin, for example either the Lantus or the detemir twice-daily instead of just once-a-day.

ALAN GARBER, MD, PhD, FACE: That's my most common experience. And the kinetics of both insulin is about the same.

CARLOS R. HAMILTON, JR., MD, FACE: And is that...

GLENN CUNNINGHAM, MD: Yes, I would agree with that.

CARLOS R. HAMILTON, JR., MD, FACE: So you like to use the long-acting insulins twice-a-day.

GLENN CUNNINGHAM, MD: In patients who have larger doses, who require larger doses of basal insulin, then you may be able to get by with once-a-day a dosing, but in the people who only require small doses, then I think it's preferable to use twice-a-day.

CARLOS R. HAMILTON, JR., MD, FACE: And the insulin that you give before meals to the type 1 diabetics is the very short-acting...

GLENN CUNNINGHAM, MD: Rapid-acting insulin.

CARLOS R. HAMILTON, JR., MD, FACE: NovoLog, Humalog type insulin.

GLENN CUNNINGHAM, MD: Correct.

CARLOS R. HAMILTON, JR., MD, FACE: Rapid-acting programs.

GLENN CUNNINGHAM, MD: Drugs.

CARLOS R. HAMILTON, JR., MD, FACE: What about the type 2 diabetics? We've already discussed, I think, in previous sessions, the fact that many of our type 2 diabetics simply do not succeed with oral therapy over the long haul and that some sort of insulin is appropriate. Alan, what is your first type of insulin that you like to give them?

ALAN GARBER, MD, PhD, FACE: Well, as I generally point out, either a basal insulin at bedtime; usually one of the long-acting analogs, such as glargine or detemir insulin, or a premixed analog before the evening meal, such as an analog 70/30 or 75/25 are about equally effective in controlling fasting glucoses the next morning and A1cs to under 7%.

The real issue becomes what do when that first injection fails because up until recently, if that basal analog given at bedtime has failed to control the blood sugar in terms of A1c reductions, then the next step is to add three additional mealtime insulin injections to a total of four injections, which most type 2 patients are not particularly fond of.

On the other hand, if you use a premix and you use an analog premix before the evening meal, you can find you can pick up another substantial fraction of the uncontrolled patients by adding a second injection before breakfast.

Adding inhaled insulin to the paradigm changes things, and we haven't fully studied this yet, but clearly it may be more acceptable to patients to go with a basal by injection and inhaled insulin for mealtime supplementation.

CARLOS R. HAMILTON, JR., MD, FACE: But again, for type 2 diabetics in the outpatient environment, a long-acting insulin is an integral part of the program. Is that what I'm hearing you say?

ALAN GARBER, MD, PhD, FACE: Well, a patient who has a weakened pancreatic insulin response needs insulin...

CARLOS R. HAMILTON, JR., MD, FACE: So treating them just with...

ALAN GARBER, MD, PhD, FACE: ...at meals and between meals.

CARLOS R. HAMILTON, JR., MD, FACE: ...just with short-acting insulin would not be the way you would go. Is that what you're saying, Glenn?

GLENN CUNNINGHAM, MD: Yes, I agree. And in terms of the consideration whether to use two shots of premixed insulin versus long-acting insulin and three shots of rapid-acting insulin, a lot of times I make that decision based upon sort of social issues.

I'm interested in knowing if the patient really will monitor their glucose with frequency. I'm interested in knowing if they eat scheduled times of the day or if they need more flexibility in that.

If they need more flexibility, I personally find that using a long-acting insulin plus multiple injections gives them that flexibility, whereas the premix doesn't. They really need to eat pretty much on schedule if they're using a premix.

CARLOS R. HAMILTON, JR., MD, FACE: I have seen patients that have been referred to me that have type 2 diabetes that need insulin, and their dose of long-acting insulin has been increased to really rather remarkably high levels without them having been giving the shorter-acting insulin or the premix insulin. How high would you go with the long-acting insulin before you would add in the shorter-acting, more frequent injections?

ALAN GARBER, MD, PhD, FACE: There was a study that was presented at the ADA meeting in June from a group of German investigators which showed that around 36 to 40 units of basal insulin in large type 2s was about as far as they went, and they were then using mealtime insulin to actually larger amounts, maybe 50-60 units per day.

CARLOS R. HAMILTON, JR., MD, FACE: Well, I'm glad to hear that because that's basically what I do. But, you know, once they get over say 30 to 40 units of long-acting insulin, you know, I decide that we're going to have to do something else.

And I agree with Glenn that the premixed insulin has a lot of usage in these patients.

GLENN CUNNINGHAM, MD: Right.

CARLOS R. HAMILTON, JR., MD, FACE: Is that right?

GLENN CUNNINGHAM, MD: The one thing though I think that we have to consider in the patient who is taking an evening injection of the long-acting insulin, if they also couple that with oral hypoglycemic agents, they may be able to sustain good A1c levels for some period of time.

CARLOS R. HAMILTON, JR., MD, FACE: Yes. And most use of oral agents in these patients, as we've mentioned previously, are mainly the insulin sensitizers, such as metformin and the glitazones.

GLENN CUNNINGHAM, MD: Primarily, yes.

CARLOS R. HAMILTON, JR., MD, FACE: That's primarily the case.

ALAN GARBER, MD, PhD, FACE: Correct. Correct.

CARLOS R. HAMILTON, JR., MD, FACE: Well certainly the new availability of the inhaled insulin may give some options to individuals that may be very attractive to them, but our experience with that is still fairly limited. But as we gain more, the role of this modality may become more obvious to all of us.

Again, thank you very much for joining us here today. We appreciate that. Alan, did you have something to say?

ALAN GARBER, MD, PhD, FACE: I just wanted to point out to our listeners that AACE will be putting online in the spring a resource center for inpatient diabetes management along the lines that Dr. Cunningham has advised in terms of tools to how to persuade various collaborators and participants in the hospital along the lines that we think they should behave, and different pieces...

CARLOS R. HAMILTON, JR., MD, FACE: I'll tell you what Dr. Cunningham has done at the St. Luke's Hospital has been remarkable because the anesthesiologists, the cardiologists, the surgeons, the nurses, everybody has a different take on this whole issue, and to get them all onboard, all paddling in the same direction was really a piece of work.

GLENN CUNNINGHAM, MD: It takes a lot of effort and a lot of support. I think it also is more difficult to do an open hospital as compared with a closed hospital. And so those of us who are working that setting have to go through more hoops than usual.

CARLOS R. HAMILTON, JR., MD, FACE: Well, there are plenty of hoops, but you've gotten through them very well.

GLENN CUNNINGHAM, MD: Thank you Carlos.

CARLOS R. HAMILTON, JR., MD, FACE: Well thank you very much for being with us today. This has been very helpful. Again, thank you for watching. I'm Dr. Carlos Hamilton.