'Understand This Warning': SGLT2s and Ketoacidosis

Recognition Key in Emergency Departments and Urgent Care Centers

Anne L. Peters, MD


June 22, 2015

This feature requires the newest version of Flash. You can download it here.

Hi. I'm Dr Anne Peters. Today I'm going to talk about the US Food and Drug Administration's (FDA) warning about the development of diabetic ketoacidosis (DKA) in individuals taking SGLT2 inhibitors.[1] The most important headline about this is that it can occur and it seems to be a rare occurrence, therefore people miss it. It's important that physicians in the emergency department and urgent care settings, as well as the rest of us, realize that a patient with either type 1 or type 2 diabetes who presents with anion gap metabolic acidosis can have ketoacidosis. Once that is recognized, the treatment becomes simple, which is intravenous insulin and glucose. However, many of these patients are euglycemic, so providers aren't tipped into thinking that it's ketoacidosis. They think it's something else. It is very important that we be aware that this can happen so that we can recognize it and treat it appropriately.[2]

I also want to make it clear that I think SGLT2 inhibitors are great drugs for the treatment of type 2 diabetes and they should continue to be used, but people need to be aware that they can cause ketoacidosis. They are also used off label for the treatment of type 1 diabetes. Although I think that it's very effective in treating individuals with type 1 diabetes, it is currently not an FDA-approved use. Because of these concerns, I suggest not using this drug in patients with type 1 diabetes until we know more about this phenomenon.

It turns out that I have seen a number of cases of ketoacidosis in my experience with using these drugs. Most of the cases I've seen have been in patients with type 1 diabetes, although a few have been in patients with type 2 diabetes. I am going to discuss two cases of mine so that you understand what this phenomenon looks like.

SGLT2 Inhibitors and DKA in Type 1 Diabetes

I'll talk about type 1 diabetes because, in a way, it's simpler. The classic case in type 1 diabetes is seen in patients who have had type 1 diabetes for 5-10 years. They've never had DKA, although they are aware that it can happen. They expect DKA to occur when their blood sugars are high, they become dehydrated, they are polyuric and polydipsic, and they develop abdominal pain and headache—the classic signs and symptoms of DKA. They would then be seen and treated for the ketoacidosis.

With these drugs, however, what happens is that somebody has an illness such as bronchitis or an upper respiratory tract infection, or they are eating less because they're exercising more. Something is going on that is a little bit different. Then they develop a headache, perhaps nausea and vomiting, and show up in the emergency department. Oftentimes, it's missed that they have DKA because their blood glucose levels are relatively normal and they are not treated for their ketoacidosis.

I have had patients seen in urgent care who were given narcotics for their headache when it actually turned out that they had DKA. Other patients were given antiemetics for their nausea and vomiting, only to come back because they became "short of breath" due to Kussmaul respirations from the DKA. These individuals won't know that they are in ketoacidosis unless they've been instructed to check for urine ketones, and it may be missed by the emergency department.

Therefore, individuals with type 1 diabetes who are prescribed these drugs need to realize that they can develop ketoacidosis at a pretty normal blood sugar level. Some of my patients' blood sugars have been 110 or 125 mg/dL, and they were actually in ketoacidosis. These individuals should be instructed to check their urine or their blood for ketones and to seek appropriate treatment if they are ketone-positive with more than trace or small amounts of ketones.

A Case of SGLT2 Inhibitors and DKA in Type 2 Diabetes

In individuals with type 2 diabetes, this is even more puzzling. As we know, patients with type 2 diabetes do not routinely develop ketoacidosis. The three cases that I've seen in patients with type 2 diabetes have all been postoperative cases. In every case, the patient was anti-GAD-negative and had a measurable C-peptide level. They seemed to all the world like individuals with true type 2 diabetes.

I will give you the most extreme case that I've seen. The individual was morbidly obese and was 39 years old. He had a 9-year history of type 2 diabetes and went in for bariatric surgery. He had a gastric sleeve procedure done. He did well, went home, and about a day or two post-op, he resumed taking his SGLT2 inhibitor. Approximately 3 days later, he went to his local emergency department complaining of shortness of breath. The shortness of breath was from his acidosis and he was diagnosed with a profound anion gap metabolic acidosis.

The physicians in his local community believed that he was having some terrible postoperative complications and sent him back to our institution for evaluation. It was found that he did not have an anastomotic leak from his surgery, and he had no clear cause for this acidosis. He was intubated and was given intravenous sodium bicarbonate, but he only seemed to get worse. Fortunately, an endocrinologist had heard of the phenomenon of euglycemic DKA. They tested his serum ketone level, which was found to be high. They started him on intravenous insulin and glucose. With this treatment, he resolved his ketoacidosis. But, interestingly, it took a few days. It didn't happen as quickly as it did in my patients with type 1 diabetes and euglycemic DKA. In fact, it took him a couple of days to improve. But he did improve and now he is fine.

The key to all of these cases is recognition. People need to know that these drugs can be associated with ketoacidosis that occurs with quite a normal blood glucose level, which may not tip people into understanding that this is due to ketones. So please understand this warning. Understand what it means. Use these drugs safely. Again, I believe that these drugs are a great treatment for individuals with type 2 diabetes. However, they should be used with caution, if at all, in individuals with type 1 diabetes, knowing that it is an off-label use. I believe that we should wait until we have good data from clinical trials to understand this phenomenon in individuals with type 1 diabetes.

Thank you. This has been Dr Anne Peters for Medscape.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.