Home / Resources / Exclusive Interviews / Transcript: Dr. Joseph Aloi on Treating Diabetic Ketoacidosis with Glucommander Software

Transcript: Dr. Joseph Aloi on Treating Diabetic Ketoacidosis with Glucommander Software

Chair; Endocrinology & MetabolismThis interview is also available as a video: click here to view.

Dr. Joseph Aloi, Wake Forest Baptist Medical Center, Winston-Salem, NC.


Steve: This is Steve Freed with Diabetes in Control and we’re here at AACE 2016 and we have with us a special guest who’s been working on some unique projects that we’d like to share with you. Maybe you can start off and tell us a little bit about your background, where you’re at and what kind of practice you have.

Dr. Aloi: I’ve been in academic medicine for the last 25 years. I worked in three hospital systems. I started at the University of Virginia and then migrated to Eastern Virginia Medical School, which a lot of the data we’re presenting came through that relationship. And most recently I’m now at Wake Forest where I’m chief of the endocrine section there. But I’ve been working in the area of in-patient glycemic management for about the last 10 years.

Steve: Because I noticed on one of your posters that you’re using this new software, fairly new software, to help treat diabetic ketoacidosis.

Dr. Aloi: Correct.

Steve: How does that work?

Dr. Aloi: Patients coming in with diabetic ketoacidosis generally are treated with IV insulin. That’s standard of care. There are some other models for that. And what we were doing at Eastern Virginia Medical School was looking at treating patients coming in with Glucommander, which is a computer dosing algorithm for IV insulin and we had good experience with using Glucommander both in decreasing length of stay and decreasing rates of hypoglycemia.  So we had a small group of information and then we combined that with looking at other medical centers. We partnered with Emory, Atlantic Diabetic Consultants, Sentara Healthcare system and then we were able to pull out almost 2,600 patients, about 1,500 with Glucommander, and compared that to usual care, which at those hospitals was generally an IV insulin through an order sent in Epic and those are the results we’re presenting here at AACE where we showed about a little bit more than a day, decreased length of stay by using Glucommander and across the board less hypoglycemia, either mild, moderate, or severe. If you look at the severe hypoglycemia, sugars of 40 or less, with Glucommander we saw less than a percent, and standard of care was about 5-6%, so a tenfold reduction and I think that’s what’s driving the decreased length of stay, so for the hospital administrators that translates into cost-savings, for the patient what we see is what’s happening in the usual care-group, IV insulin stopped prematurely and then it’s restarted or they get an episode of hypoglycemia, which slows down their plan to discharge. So the usual care, another way of looking at it is probably extending length of stay and Glucommander, by avoiding those two issues, helps reduce length of stay.

Steve: Is this mostly in-hospital? IV insulin?

Dr. Aloi: Yes, it’s all in-hospital. We do have some experience using Glucommander. We just have a paper coming out in Diabetes Technology, smaller group, looking at treating people with mild DKA and we can talk about what that means, but generally these are people that have missed a dose or two of insulin, are otherwise feeling well using Glucommander in the emergency room to discharge them before admitting them.  And we can do that successfully about three quarters of the time.

Steve: Treating DKA can be very complicated because it’s about more than just insulin.

Dr. Aloi: It’s a lot more than just about the insulin, yeah.

Steve: Where does it fall into that area, because there’s so many things involved in treating it.

Dr. Aloi: I think what Glucommander or other order sets or integrated things — to bring in all those other things — the other key components are what you’re using for IV fluids. So an important part of minimizing hypoglycemia is at some point having IV fluids that have dextrose in them.  That helps buffer against hypoglycemia. Watching the electrolytes, and actually there’s flags built into Glucommander for what the potassium is for when you’re starting insulin and so forth. And then the other piece is nutrition. So some people that have mild DKA are eating and eating on IV insulin makes it hard to adjust, but Glucommander actually allows you to do a carb specific insulin dosage estimate while they’re eating, so that also minimizes fluctuations. And what happens, I guess I would say traditionally is if a patient does eat, people are unsure where they increase the drip rate, give them a bolus of insulin, and generally what happens is the blood-sugar goes up, the drip rate goes up a little bit too rapidly and then they get hypoglycemic, they stop the drip, and again we’re sort of slowing down the treatment paradigm. So we can obviate a little bit of that with getting all the information funneled through one sort of common platform.

Steve: So it’s interesting, obviously there’s a lot more people at home, and there’s a lot more people that family practitioners work with that should be on insulin…

Dr. Aloi: Yes.

Steve: …but the doctor’s afraid of putting them on insulin because they run the risk of a lawsuit. They run the risk of hypoglycemia and basal-bolus becomes a little bit more complicated. So many doctors would prefer to see them on three oral drugs rather than putting them on insulin.  Does this kind of change that? That physicians working with an educator or a pharmacist, a diabetes educator?

Dr. Aloi: I think the short answer is anything that can help a provider feel a little bit more comfortable with the dosing is going to help lower the barrier and it may not be that actual initiation.  Because again, what do I do for a living? I get all the patients that say no. Right, so then I have to work with them. But you know a lot of times the barrier is either misunderstanding of the patient’s perception about insulin, or just kind of getting over the hurdle. I mean there’s two separate things, because then I also get the patients on insulin on a complicated insulin regimen, but nobody’s adjusting it. I think somehow tying that in with the message, you know something that the computer system’s sending you. Everything if you look at, you know, you’re an educator… there’s a lot of information about text messages, and I think it works well in adolescents. I can tell you if you’re text messaging me to remind me to check my blood sugar, I’d probably suppress it, but so for the right person, in the right situation, I think absolutely… it’s another tool in the box that can help people manage insulin more effectively and more – I think – importantly, safely. When you see the data for the hypoglycemia rates being so low… I think that will buy in some confidence both with the patient and the prescribers.

Steve: That is, certainly. Hypoglycemia is the number one issue when it comes to using insulin.

Dr. Aloi: I think it really is. I mean, you know we talk about blood sugar control… but certainly as a patient gets older, I’m really much, much more focused on hypoglycemia prevention.

Steve: I was recently invited to a special meeting where the FDA was there and all the insulin companies were there, they had doctors there, nurses, pharmacists. The whole purpose of this meeting… it was a 3-day event. The whole purpose of this meeting was to see how we could prevent insulin errors. Because insulin errors is the number 1 mistake for the last 50… since insulin was invented, it’s the most common error in issues when it comes to patients and making mistakes whether it be the wrong insulin or the wrong dosage of insulin and you had mentioned 3 when you first sat down here. You talked about not starting people on a basal insulin after they’ve taken off of IV. So I kind of see this is a way to prevent some of those mistakes.

Dr. Aloi: I agree and, you know, when I’m talking to my hospital administrator, they’re a little focused on this, and it’s frustrating as a person caring for diabetes as an educator… I mean we see… I have to review all of the hypoglycemic episodes in the hospital that result in, changing beds, there’s a couple qualifiers. But every time I look at them, I can see it coming. You know, patient just out of dialysis, not eating, they give them their home dose, which is not their correct dose. it’s the dose that’s prescribed. I mean there’s so many levels to this. Patients getting their meal time insulin when they’re NPO. So the electronic medical record has it’s good side/bad side, but there’s ways of partnering with that where you can at least try and prevent that. So we’ve been doing a lot of soul-searching about how to make the EMR work better to help prevent a lot of those errors. You can message somebody if they start to discontinue an insulin drip and there’s no basal insulin ordered. And we did that actually at Sentara and we used it for about 6 months and then we got rid of it. Because people quickly learned. I mean, providers will learn, and then when the nurses learn, they see an order coming for stopping the insulin, they haven’t started basal insulin, they’ll call the provider. So you can teach people through the MR how to adopt best practices.

Steve: A lot of mistakes are made in nursing homes.

Dr. Aloi: Long-term care facilities is if you think of that iceberg, I think of that as sort of a big piece that… it’s hard to… It’s hard to manage in long-term care facilities. Frequently there’s a provider coming in every few days. There’s one nurse for a multitude of patients. There is a little bit of information coming out, not that I’ve been involved with directly. But just doing some more simple insulin. And what I’m talking about is if you look at what’s being used in most of those facilities, it’s sliding scale regular. And so… very high risk for hypoglycemia, and if you can just get basal insulin and not worry about controlling those patients perfectly, you can minimize… a big driver for them is going to be just like hospitals are penalized for 30-day readmissions. Nursing homes if they’re sending the patient back are not going to get paid. So there’s going to be a big interest in learning how not to send a patient to the emergency, there blood sugars 300, to avoid blood sugar of 50 and sending the confused patient to the emergency room

Steve: So this is fairly new. Right now it’s only being used in hospitals and then I guess as far as research goes, it’s being used in a couple of endos offices?

Dr. Aloi: Yeah, it’s being used in some out-patient settings, yeah, it’s largely been used in hospitals and I think… I’ve been involved with Gly-Tech long enough… They had it embedded in Epic and when they launched it out and Sentara Health Care System is the largest health care system that did it… and I think that was now 3 years ago. So they’re in all the major EMRs now. We’ve got a lot of data we’re producing and publishing now. So I think it’s more the test of time, it’s a very good option and it works very well. It works pretty seamlessly with the MR.

Steve: You’ve seen this… Well, I have a hunch eventually it will be some kind of app for people to use on mobile. Because if the doctor has to send the patient home, because of the basal bolus, he probably gets a number of phone calls making adjustments, people are concerned that they’re told that insulin can be very dangerous, it can kill you.

Dr. Aloi: Well sure, they know that right, people kill people with insulin. So for patients new to insulin, it’s a big stress.

Steve: So, getting to another subject and why you’re here at AACE, have you had a chance to see any of the presentations?

Dr. Aloi: Oh yeah. Yeah, yeah.

Steve: Is there any that caught your interest?

Dr. Aloi: I think, I’m a history buff, so the history stuff is very interesting. But the late-breaking, I don’t want to say late-breaking, but the relatively new news about the SGLT2 inhibitors. Both on the upside for their, what seems like very powerful cardiovascular risk reduction, and on the difficulties with causing, in select patients, probably undiagnosed type 1 diabetics DKA. We’re actually looking at that in our hospital system, partnering with Duke and Chapel Hill, UNC.  Just looking at the MR between our three institutions, we’ve got about a 100 patients with the diagnosis of DKA on an SGLT2 inhibitor that has come through in the last year. I’ve personally seen six since I was looking for it in about September. Of the six I’ve seen, everyone has turned out to be type 1 that had gone on undiagnosed. Started on an SGLT2 inhibitor in the hospital with DKA with blood-sugars 150? 200?

Steve: Ok, so you’re seeing euglycemia with DKA?

Dr. Aloi: Yeah, this is what they’re phrasing euglycemia… now having said that, a couple people still come in with a 4-500.  But my… any time I do in-talk for primary care providers or anyone for that matter, if I talk about SGLT2 inhibitors, I throw that in as a “keep your eyes open.”

Steve: Do you see it in type 2s?

Dr. Aloi: Yes, is the short…but I’m not sure if there’s something different about their diabetes. Because the two patients I’ve seen that were not antibody positive for type 1 had measurable insulin levels… Uniformly, all the patients I’ve seen have been on insulin. So I haven’t seen it on people just on orals, and the type 2s that I’ve seen it in have been on insulin therapy for a long time, so maybe they’re relatively insulin deficient. But I’m not sure. There’s actually a talk in about 45 minutes that’s going to hopefully give some insights to that. I’m going to be going to that as well.