Memorandum
The FDA issued a drug safety communication today alerting providers and patients to the possibility of ketoacidosis for patients using SGLT-2 inhibitors (“euglycemic DKA”). Further down the page, there are drop-down panes with more details for physicians and patients. Notably, the warning applies to all diabetes patients using these agents – the initial reports we heard were from type 1 diabetes patients taking the drugs off-label, but most of the 20 cases reported to the FDA Adverse Event Reporting System (AERS) have been in type 2 diabetes patients. The FDA is continuing to investigate the issue and will determine whether label changes are necessary. In the meantime, the FDA recommends patients to pay close attention to the symptoms of ketoacidosis and immediately stop taking the drug and seek medical treatment if symptoms emerge. However, the communication recommends against stopping or changing diabetes medicines in the absence of symptoms without first consulting a provider. J&J’s Invokana (canagliflozin), AZ’s Farxiga (dapagliflozin), and Lilly/BI’s Jardiance (empagliflozin) – along with their fixed-dose combinations – were mentioned in the communication as they are the only SGLT-2 inhibitors on the market in the US. We have been anticipating this move ever since we learned about the issue from Dr. Anne Peters (USC, Los Angeles, CA) and Dr. Irl Hirsch (University of Washington, Seattle, WA) at this year’s ENDO in March. For background, this phenomenon involves the appearance of ion gap acidosis at normal blood sugar levels in patients taking an SGLT-2 inhibitor. The FDA communication did mention that most reported cases were linked with major illness, reduced food and fluid intake, and reduced insulin dose – this is important as it drastically narrows the at-risk population. Euglycemic DKA is serious because the presence of normal blood glucose often leads to misdiagnosis and a delay in treatment. On the plus side, once correctly diagnosed, treatment is fairly simple (IV glucose and insulin) and patients recover quickly. We will continue to monitor this closely – given how new and rare the issue is we imagine most providers are not aware of it, and there is room for this to put a dent in the new drug class’ ballooning popularity. We hope the FDA and scientific community continue to frame this issue in a non-alarmist manner – at this point this appears to be a serious but still fairly rare occurrence. For now, the most important goal is to educate patients and providers (including ER physicians) about the phenomenon and to make sure patients on SGLT-2 inhibitors have ketone testing strips on hand. As we learned from Dr. Hirsch, education will also be important for anesthesiologists and surgery teams, as surgery is potentially a high risk time when patients are not eating or are ketotic. Additionally, many general anesthesias inhibit endogenous insulin secretion. A slightly higher degree of caution may be appropriate for type 1 diabetes patients taking the drug off-label, for whom some KOLs like Dr. Peters have recommended discontinuing the drug until euglycemic DKA is better understood.
-- By Manu Venkat and Kelly Close