The essential truth is simple and serious: SGLT2 inhibitors can precipitate diabetic ketoacidosis that may be fatal, sometimes even when blood glucose is not markedly elevated—so recognition and swift action are non‑negotiable for anyone on dapagliflozin, empagliflozin, or canagliflozin.
The Short Version
- FDA added class-wide warnings in 2015 linking SGLT2 inhibitors to ketoacidosis; labels direct patients to stop the drug and seek urgent care if symptoms arise.
- Official labeling for empagliflozin (Jardiance) and dapagliflozin (Farxiga) states that fatal cases of ketoacidosis have been reported.
- Ketoacidosis can occur with glucose below 250 mg/dL (euglycemic DKA), which can mislead patients and clinicians; evaluate for acidosis based on symptoms, not glucose alone.
- Risk is substantially higher in insulin‑deficient states (notably in type 1 diabetes); if symptoms are consistent, stop the drug and get emergency care.
What the FDA actually warned—and why it matters
In late 2015, the FDA concluded a safety review and required new labeling for SGLT2 inhibitors—canagliflozin, dapagliflozin, empagliflozin, and others—warning about the risk of ketoacidosis and instructing patients to seek medical attention if they develop compatible symptoms. That action codified a pharmacovigilance signal into class‑wide communication, elevating awareness across primary care, endocrinology, emergency medicine, and pharmacy workflows. Importantly, this was not an abstract precaution: subsequent FDA‑approved labels for individual agents explicitly state that ketoacidosis can be life‑threatening and that fatal cases have occurred in users of these drugs.
Why that phrasing matters is twofold. First, the agency’s decision is a class statement—this is not a single‑molecule anomaly but a mechanism‑linked phenomenon observed across SGLT2 inhibitors. Second, the directive is practical: patients and clinicians are told what to do at the first sign of trouble—discontinue the drug and escalate care—rather than to watch and wait. That action orientation is the right design for a rare but dangerous syndrome.
Mechanism in plain clinical English: how SGLT2 inhibitors set the stage for DKA
SGLT2 inhibitors lower plasma glucose by blocking renal glucose reabsorption, causing glycosuria. The resulting drop in glucose can reduce endogenous insulin secretion; at the same time, counter‑regulatory hormones (notably glucagon) may rise, and the kidney’s handling of ketones shifts. In susceptible patients—especially those with limited insulin reserve—this metabolic milieu favors lipolysis and hepatic ketogenesis. The crucial twist is diagnostic: because glucose may stay below 250 mg/dL, the classic “high sugar” alarm for DKA fails, allowing euglycemic DKA to evolve unnoticed unless clinicians check ketones and acid–base status. FDA labeling for empagliflozin makes that point explicit: evaluate for ketoacidosis even if the blood glucose is less than 250 mg/dL.
Context from the literature is consistent with the regulator’s signal: academic reviews describe the DKA association as class‑level, with case clusters and elevated reporting rates compared with non‑SGLT2 comparators—while also emphasizing that the absolute event rate in type 2 diabetes is low in routine care. Both statements can be true at once: relative risk can be meaningfully higher while the baseline rate remains uncommon.
What the labels require patients and clinicians to do
Across products, the instructions are crisp. If symptoms consistent with ketoacidosis occur—nausea, vomiting, abdominal pain, malaise, unusual fatigue, shortness of breath—stop the SGLT2 inhibitor and seek medical attention immediately. For clinicians, the labels advise prompt assessment for acidosis whenever such symptoms appear, regardless of glucose level, and to institute standard DKA therapy as indicated. Empagliflozin labeling uses unambiguous language: ketoacidosis is a serious, life‑threatening condition requiring urgent hospitalization, and fatal cases have been reported in patients taking the drug. Dapagliflozin’s labeling contains the same fatality linkage and equivalent action guidance.
Patients should internalize two rules. First, don’t rely on fingerstick glucose to “rule out” DKA: normal or near‑normal readings do not exclude euglycemic DKA. Second, communicate perioperative plans, intercurrent illness, or major dietary changes to the prescribing clinician; insulin deficiency, fasting states, and surgical stress amplify risk and may warrant holding the drug temporarily under medical advice.
Who is most at risk: type 1 diabetes and other insulin‑deficient states
Although SGLT2 inhibitors are not approved for type 1 diabetes in the United States, some trials and off‑label use have occurred. In that setting the DKA signal is pronounced. Dapagliflozin’s FDA label is direct: in patients with type 1 diabetes, the drug significantly increases the risk of DKA beyond the background rate in placebo‑controlled trials. That pattern fits metabolic logic—limited insulin reserve leaves little buffer against the ketogenesis‑permissive environment created by SGLT2 inhibition.
By contrast, in typical type 2 diabetes use, the absolute incidence of DKA is low, though elevated compared with non‑SGLT2 agents in several observational and trial summaries. The absence of a simple denominator in postmarketing reports does not change the clinical imperative: rare and life‑threatening still demands vigilance. The practical upshot is targeted caution—screen aggressively when symptoms suggest acidosis, educate patients about warning signs, and hold the drug around major stressors that predictably lower insulin availability.
New—#Ketoacidosis with #SGLT2 inhibitors in routine clinical practice of type 2 #diabetes: Scandinavian cohort and nested case–control study https://t.co/Ob46G798FD #DKA #T2D
— The Lancet Diabetes & Endocrinology (@TheLancetEndo) September 2, 2026
Recognizing euglycemic DKA quickly: what to look for and how to test
Presentation is often nonspecific—nausea, vomiting, abdominal pain, malaise, dyspnea. Because polyuria and extreme hyperglycemia can be muted by glycosuria, clinicians should pivot early to acid–base testing and direct ketone measurement when an SGLT2 user is unwell. Bedside urine ketone strips can miss beta‑hydroxybutyrate, the predominant ketone in DKA; serum or capillary beta‑hydroxybutyrate assays and blood gas analysis are the reliable path to diagnosis. The labels’ directive to evaluate for ketoacidosis “regardless of presenting blood glucose levels” is the critical cue to bypass anchoring bias and test when symptoms warrant.
Balancing benefits and risks: why the warning coexists with broad clinical use
SGLT2 inhibitors earned their place in modern care because they do more than lower glucose. Across the class, trials have shown reductions in heart failure hospitalization and slowing of chronic kidney disease progression, reshaping standards of care for patients with cardiometabolic disease. The ketoacidosis warning does not negate those benefits; it calibrates the risk conversation. Drug classes with meaningful outcome benefits often carry serious but uncommon adverse events that demand literacy rather than fear. Clear label language, patient education, and clinician vigilance are how the field reconciles both truths.
What patients and caregivers should do today
If you take dapagliflozin, empagliflozin, or canagliflozin, learn the DKA symptom set, keep hydration and sick‑day plans current, and tell your care team before surgery, prolonged fasting, or major diet shifts. If compatible symptoms start, stop the drug and seek urgent care—do not wait for very high glucose to “confirm” trouble. That is the behavior the labels aim to drive, and it is how rare events stay rare and nonfatal.
Sources:
mirror.co.uk, pmc.ncbi.nlm.nih.gov, accessdata.fda.gov, fda.gov, scirp.org, jamanetwork.com
