THE patient sitting across from you has type 2 diabetes and cardiovascular comorbidities. You want to prescribe them a sodium–glucose cotransporter type 2 (SGLT2) inhibitor. Their English is functional, but health literacy is not great. How do you explain diabetic ketoacidosis (DKA) to them without causing anxiety, while making it clear when it is vital that they stop the drug?

“The good news is SGLT2 inhibitors are really good drugs,” says Associate Professor Peter (Shane) Hamblin, a consultant endocrinologist at the Alfred Hospital in Melbourne, and Director of Diabetes and Endocrinology at Western Health.

“We want to encourage people with diabetes to use them and to be on them, in particular, if they already have cardiovascular disease.”

The problem, Associate Professor Hamblin says, is that although DKA is a rare adverse effect of SGLT2 inhibitors, it is not rare enough that it can be ignored when starting a patient on the drugs.

“Even though DKA is a very uncommon side effect – probably less than one in 1000 a year – if you have 200 000 people [on SGLT2 inhibitors], which we now do in Australia, then that’s a couple of hundred people a year potentially going to be presenting with DKA. Two hundred isn’t that many, but if you’re one of the unlucky ones, it can be quite serious.”

DKA comes about when, usually due to a lack of insulin, the body can’t metabolise glucose for energy, and starts to burn up fat instead. The by-product of burning fat is ketones. If the body builds up too many ketones, they form an acid, which can, effectively, poison cells.

“If you don’t have diabetes and you fast or go on a low [carbohydrate] diet, you will show ketones, but the body has enough insulin to naturally suppress the acid build-up,” says Associate Professor Hamblin in an exclusive InSight+ podcast.

“When you’ve got DKA, you don’t have that mechanism. So, you get an uncontrolled production of ketones. Acid builds higher and higher. And that’s why people get so sick.

“They get dehydrated because the glucose is high. They lack energy because the cells aren’t operating normally and then they start to breathe very fast, trying to compensate for the high acid by a respiratory mechanism that blows off acid effectively.

“You’ll see a patient who might be quite weak, dehydrated, might be feeling faint, they are breathing really fast and they can’t explain why they’re feeling that way.

“It then leads to other things like nausea and vomiting and so they’ll start throwing up and if they’ve got diabetes and they know about it, they may be too sick to do anything about it. Often what happens with traditional ketoacidosis is the person might just feel so unwell, they go to bed and they don’t check their blood sugars.

“Often, they’ll reduce their insulin because they think that they’re not eating, they’re vomiting. That’s the worst thing you can do because the insulin is the thing that turns it all off.”

SGLT2 inhibitors, which act to stop the reabsorption of glucose from urine, cause blood sugar levels to drop, and have been shown to have cardiovascular and renal benefits.

The problem comes if a patient is sick for another reason, or is preparing for surgery or, for example, a colonoscopy, and doesn’t come off the SGLT2 inhibitor beforehand.

“If you have a person [with diabetes] who is well and is taking the [SGLT2 inhibitor] along with all their other cholesterol drugs and blood pressure drugs, the risk [of DKA] is really, really low. Really low,” says Associate Professor Hamblin.

“In the majority of people, it’s not going to be an issue.

“However, if that same person then has to fast for, let’s say, a colonoscopy, or they have to come in to have an operation, or they might be unwell with pneumonia. They’re not eating, their carbohydrate intake goes down. That’s the first thing, because their body is going to start to form ketones.

“If the person continues to take their SGLT2 inhibitor they may then be heading towards ketoacidosis without knowing it, because most people with type 2 diabetes are going to have no knowledge of DKA in the usual way we’ve taught people [with type 1 diabetes].

“If a person comes into hospital with, for example, a very severe kidney infection, they’re off their food, they might be in pain. They’re not themselves. If the staff looking after them don’t stop the SGLT2 inhibitor, the risk is that the patient is going to be misdiagnosed because they will have nausea and perhaps vomiting because of their [kidney infection].

“Then as the antibiotics are given, usually the person would get better. But if you keep giving the medication for the diabetes thinking you’re doing the right thing because you want to control diabetes when you’re sick, that person might then start to get nausea and vomiting, not due to [the kidney infection], but due to the drug.”

DKA can be tricky to diagnose in the face of other symptoms, Associate Professor Hamblin says.

“One of the key learnings that we’ve had from studies in a number of clinical studies around the world is that the blood sugar, which is normally the key thing to look for with DKA, is not that high.

“It might only be 8 or 10 or 12 mmol/L. Typically, as medical students, we were taught the blood sugar would be 30 or 40 or 50 [mmol/L].

“So, what can happen is the patient is quietly getting worse under our very eyes and the person might end up in intensive care because on Day 4 finally someone does the appropriate tests. They might check ketones and blood gas, by which point the patient is really sick.”

The key, says Associate Professor Hamblin, is educating the patient about DKA and when to come off the SGLT2 inhibitor. The problem is how to tailor that message for different patient demographics, how to do it without putting people off the drug, and determining whose responsibility it is.

“How do you do that when you are talking to a person you’re about to put on these drugs?” he says.

“You don’t want to frighten them away from using the drugs because the benefits are clear, but at the same time, both to look after the patient and also medico-legally, you’ve got to talk about DKA.

“I’m trying to think, ‘how do I explain it to my mum? How would I explain it to my uncle? How do you talk to someone who isn’t medically trained?’

“Because trying to explain DKA to a medical student or to our advanced trainees is difficult enough.

“Our job as doctors is to work out how do we convey the information to a person so that they can safely take the drug and they know when not to take the drug. How often do we have to remind ourselves and remind the patient when they’re on that drug? Should we do it every 6 months? Every 12 months? Do we hand them a standard form of words, that everybody around the country is using? What about people who are not health literate? What about people who don’t speak English? Or don’t read English?

“This [type of] drug is going to be started by nephrologists, endocrinologists, cardiologists and GPs and probably others.

“We’ve all got a responsibility. We’ve all got to know about this small but critical risk and then we’ve got to work out, ‘what am I going to do about it?’ And what’s the health system going to do about it?

“Everybody who is in the field is aware of it, but we’re all going about it in a slightly different way. However, if we’re going to have 200 000, and soon 300 000 people on these drugs, we’ve just got to do something about it.”

Associate Professor Hamblin has co-authored a Perspective published in the MJA on SGLT2 inhibitors and the risk of DKA.

More from the MJA online:

Research letter: Exceedances of national air quality standards for particulate matter in Western Australia: sources and health-related impacts
Borchers Arriagada et al; doi: 10.5694/mja2.50547
Prescribed burning reduces wildfire risk but the health impacts of each fire type need to be better understood … OPEN ACCESS permanently

Research letter: Stereotactic radiosurgery for managing brain metastases in Victoria, 2012–2017
Ong et al; doi: 10.5694/mja2.50573
Patients in regional areas were as likely to receive SRS as those in metropolitan areas … OPEN ACCESS permanently

Research letter: Characteristics, treatment and complications of herpes zoster ophthalmicus at a tertiary eye hospital
Chakrabarti et al; doi: 10.5694/mja2.50554
Education of health care professionals caring for patients with HZO must be improved … OPEN ACCESS permanently

Podcast: Associate Professor Shane Hamblin, Director of Diabetes and Endocrinology at Western Health, talks about SGLT2 inhibitors and the risk of diabetic ketoacidosis … OPEN ACCESS permanently

4 thoughts on “Risk v reward: SGLT2 inhibitors and diabetic ketoacidosis

  1. Robertto Celada (GP) says:

    I agree that educating the patient about their diabetic management is of paramount importance. However, I continuously get so frustrated because the majority of diabetic patients do not want or are unable to learn about diabetes management. Every diabetic encounter I have I explain what HBA1C is and why we do it. I ask the patient next time what HBA1C is, and no one can tell me the answer. Explaining the serious possible complications of SGLT2 inhibitors and the most common: genital thrush is routine in my practice but I still have patients returning with severe genital thrush (males and females) still taking the drug!! despite being told to stop it immediately.

  2. Anonymous says:

    I agree with Ian. The number of proprietary brands of SGLT 2 inhibitors is increasing and difficult to keep track of. I am a surgeon, and routinely stop SGLT2 inhibitors prior to surgery. I have not treated a single patient using these drugs who recalls being made aware of the risks of DKA by their prescribing practitioner. These are only going to be safe drugs when steps are taken to ensure patients are universally warned of the risks, and there is an easy, reliable way of identifying them on a medication list where proprietary drug names are used. In the interim, this is a dangerous class of drugs in my opinion. Adverse events are inevitable until these safeguards are effectively implemented.

    At no stage was I or any of my colleagues warned about the peri-operative risks of DKA with this class of drugs. We discovered when the first case reports of adverse events started to emerge. Too little too late. The information should have been widely disseminated through the surgical colleges, the anaesthetic college and through surgical facilities, when the drugs first became available. Which begs the question; Why did this not occur?

  3. Anonymous says:

    Hello Prof Shane Hamblin! Can you recommend me an article where it is shown that the drug SGLT2 has cardiovascular benefits?
    Thank you very much.

  4. Ian Hargreaves says:

    Also tricky to explain to a simple surgeon. I had one recently where I didn’t recognise the trade name in a Medical Director printout (mainly looking for NOACs) and the anaesthetist spotted it on admission.

    Fortunately the operation could go ahead with a regional block, and rapid return to eating.

    I spotted a trade name last week and have advised the patient to stop preoperatively, but it would be a potential nightmare if the patient is cancelled during the COVID-19 crisis, or in the usual public hospital fashion of being admitted, then sent home with the promise of an operation in a couple of days’ time, or next week.

Leave a Reply

Your email address will not be published. Required fields are marked *