Sodium-glucose co-transporter-2 inhibitors (SGLT2i) are medications used in type 2 diabetes mellitus (T2DM) that control high blood sugar levels and promote normoglycemia by preventing glucose reabsorption and facilitating glucosuria. Three SGLT2i (canagliflozin, dapagliflozin, and empagliflozin) are approved in the United States. Euglycemic ketoacidosis or ketoacidosis with a lower than expected hyperglycemia is a rare adverse event associated with SGLT2i. Low-carbohydrate ketogenic diet (LCKD) may lower the threshold for the development of SGLT2i-induced ketoacidosis in T2DM. We report a rare case of diabetic ketoacidosis in a type 2 diabetic patient with a blood glucose level of 159 mg/dl in the presence of empagliflozin and LCKD. He had recently started eating a LCKD. He presented with the nonspecific complaints of fatigue, headache, nausea, abdominal pain, and chest discomfort. He had run out of his anti-diabetic medications and started retaking them a week ago. Labs showed metabolic acidosis with ketonuria and glucosuria. His presentation was attributed to the reintroduction of empagliflozin in the presence of a LCKD. SGLT2i put patients at risk of euglycemic ketoacidosis in T2DM due to glucosuria, euglycemia, and suppressed insulin release. Lifestyle changes, such as LCKD, may lower the threshold further due to excessive lipolysis, beta-oxidation, and hepatic ketogenesis. Practitioners should educate patients about this adverse effect. Patients should consult with their physicians before incorporating a LCKD in their lifestyle if they are using SGLT2i.
Sodium-glucose co-transporter-2 inhibitors (SGLT2i) or Flozins are antidiabetic medicines that target high blood glucose levels in type 2 diabetes mellitus (T2DM) and reduce glucose reabsorption from proximal renal tubules, thus facilitating glucose excretion (glucosuria) and promoting euglycemia . In the United States, the Food and Drug Administration (FDA) has approved canagliflozin, dapagliflozin, and empagliflozin as SGLT2i . In Japan, other Flozins are also available, including ipragliflozin, luseogliflozin, and tofogliflozin . In May 2015, the FDA issued a warning about the rare occurrence of SGLT2i-induced ketoacidosis . By July 2015, 28 cases of ketoacidosis due to SGLT2i were reported .
Euglycemic ketoacidosis precipitated by SGLT2i is a rare condition, with an estimated incidence of 0.16–0.76 cases per 1000 patient-years [5, 6]. According to a Canadian study, the estimated incidence of SGLT2i-related ketoacidosis is 0.003% . Canagliflozin, the first available Flozin, is the most commonly reported SGLT2i associated with ketoacidosis. This is perhaps due to the longer availability of this drug in the market [7, 8]. Empagliflozin is a rare cause of SGLT2i-induced diabetic ketoacidosis (DKA) in T2DM compared with other Flozins .
Since patients with SGLT2i-related ketoacidosis usually have normal or slightly elevated blood glucose levels, the recognition of this condition could be missed or delayed, resulting in adverse outcomes. Euglycemic ketoacidosis may not present with typical DKA manifestations. Hence, a high index of clinical suspicion along with a careful review of patient’s home medications is critical to reaching a final diagnosis. We describe a rare case of euglycemic ketoacidosis due to empagliflozin and identify low-carbohydrate ketogenic diet (LCKD) as a potential trigger for excessive ketogenesis.
A 65-year-old Caucasian male with a past medical history of hypertension, T2DM, and benign prostatic hyperplasia presented to us with nausea, pounding headache, mild generalized abdominal pain, and some chest discomfort which he described as acid reflux. All these symptoms had been going on for four days. He also reported feeling diffusely achy and feeling overall very weak and fatigued. He denied urinary or respiratory complaints.
The patient reported no fever, headache, or neck stiffness. His vitals included pulse at 74/min, blood pressure at 131/56 mmHg, respiratory rate at 18/min, and temperature at 98°F. Physical examination was unremarkable. Hemoglobin and liver function tests were within normal limits.
Blood urea nitrogen was 24 mg/l (range, 9–21) and creatinine was 0.77 mg/dl (range, 0.80–1.40). Blood glucose was 159 mg/dl. Serum bicarbonate was 9 mmol/l (range, 21–29) and anion gap 28 mmol/l (range, 9–16). Blood gas performed showed a pH of 7.20 (range, 7.32–7.42). Urinalysis showed 4+ ketones and 2+ glucosuria (normal is negative).
An extensive workup focused on patient’s presenting complaints, including troponin, electrocardiogram, computed tomography (CT) scan of the head, CT angiogram of the chest, and CT of the abdomen and pelvis with intravenous contrast, was unremarkable. No infectious process, such as urinary tract infection, pneumonia, or septicemia, was detected. Hemoglobin A1c was 8.0. Serial lactic acid levels were within normal limits (1.1 and 0.8 mmol/l; range, 0.4–2.0). A preliminary diagnosis of metabolic acidosis due to an unclear etiology was made, and the patient was started on intravenous fluids and insulin therapy. We held his oral antihyperglycemics. He denied a history of alcohol or drug use. His home medications were reviewed for their potential contribution to the development of metabolic acidosis. He was on metformin 1000 mg twice daily. Though metformin can cause metabolic acidosis by raising the lactate levels, his lactic acid level was normal.
Other medications were aspirin 81 mg daily, pioglitazone 45 mg daily, tamsulosin 0.4 mg daily, and empagliflozin 25 mg daily. The patient denied intentional or unintentional overdose of any of his medications. He also denied taking over-the-counter medications.
Our patient had run out of all of his medications two weeks prior to the admission, including empagliflozin, and started retaking them a week ago. During the same period, he started eating a LCKD and ate 3–4 such meals; this lowered his threshold to develop ketoacidosis in the presence of reintroduction of empagliflozin. After ruling out various causes of metabolic acidosis, the patient’s presentation was attributed to empagliflozin and LCKD.
The patient received supportive care (intravenous fluids, insulin) and symptomatic therapy. His acidosis resolved within 24 h. Serum bicarbonate level increased from 9 to 22 mmol/l, and pH became normal at 7.41. We discontinued empagliflozin altogether as the patient intended to continue a LCKD in the future. We recommended the patient to have a repeat metabolic profile within a week post-discharge from the hospital, which showed a normal serum bicarbonate level of 24 mmol/l.
SGLT2i can cause ketoacidosis with a lower than expected hyperglycemia . Burke et al. performed a systematic literature review and found 34 published cases of SGLT2i-induced DKA, 25 of them occurred in T2DM . Empagliflozin caused DKA in 2 cases (after ipragliflozin with 1 case, not available in the United States), dapagliflozin in 5 cases, and canagliflozin in 26 cases . FDA Adverse Event Reporting System (FAERS) recorded 73 incidents of SGLT2i-related DKA between March 2013 and May 2015; four of DKA cases occurred with empagliflozin, 21 cases with dapagliflozin, and 48 cases with canagliflozin . In a randomized controlled trial of empagliflozin in T2DM, a total of four ketoacidosis cases (0.1%) occurred per 4687 patients, 3 cases with empagliflozin 10 mg (0.1%), and 1 case with empagliflozin 25 mg (< 0.1%) .
Since such patients are usually normoglycemic or minimally hyperglycemic, there are reports of delayed or missed diagnosis .
SGLT2i cause ketoacidosis by various mechanisms, including glucagon release and glucosuria, with subsequent euglycemia [3, 7, 9]. Euglycemia from SGLT2i that is mediated by glucosuria suppresses the insulin release and results in insulin deficiency. Insulin deficiency causes compensatory catabolism, increase lipolysis, and release free fatty acids (FFA) in the circulation . Suppressed insulin level also stimulates the glucagon release. Decreased insulin-to-glucagon ratio inhibits acetyl-CoA-carboxylase activity (insulin increases whereas glucagon decreases the activity of acetyl-CoA-carboxylase) and promotes hormone-sensitive lipase (HSL) activity (insulin decreases where glucagon increases the activity of HSL) [3, 11]. HSL liberates FFA of triglycerides from the adipose tissue, making them available for beta-oxidation and ketogenesis . Suppression of acetyl-CoA-carboxylase enzyme activates carnitine palmitoyl transferase-I (CPT-I), which transports FFA to the mitochondria, resulting in beta-oxidation and ketogenesis . Imbalance of ketogenesis and ketolysis results in ketoacidosis. Reduced insulin levels prevent peripheral utilization of ketones, therefore, tilting the balance toward ketoacidosis. Reduced glomerular filtration and diminished clearance of ketones may also play a role in the reduced elimination of ketones . Another proposed mechanism involves SGLT2i-induced osmotic diuresis, leading to hypovolemia and hormonal changes, causing ketogenesis .
The occurrence of ketoacidosis with SGLT2i is rare. Therefore, there are some other triggers responsible to induce ketoacidosis in the presence of SGLT2i. Risk factors cited in the literature are stress, fasting, dehydration, acute illness, surgery, low-carbohydrate diet, excessive alcohol intake, perioperative period, reduction in insulin dose, and discontinuation of diabetes treatment [6,7,8,9, 11]. It is essential to recognize these potential risk factors contributing to SGLT2i-related ketoacidosis.
Empagliflozin can shift fuel energetics from less efficient energy sources, such as fat or glucose, to more efficient sources, such as ketone bodies . Such a shift in fuel energetics due to empagliflozin has been speculated to confer cardiovascular and renal benefits in T2DM. This means that ketone production due to SGLT2i is not deleterious altogether. Excessive production of ketones due to pathophysiological triggers, such as prolonged fasting or LCKD, may precipitate ketoacidosis given the presence of calorie-restriction mimetics and pro-ketogenic effect of empagliflozin .
LCKD (defined as the diet that significantly restricts carbohydrate consumption, generally less than 50 g a day, contains mainly proteins and fats, and favors ketogenesis) is therapeutically beneficial to promote weight loss, reduce insulin resistance, and improve glycemic control in T2DM [14, 15]. LCKD mimics a prolonged fasting state in which the caloric sources are mainly “adipocentric” rather than “glucocentric” . Due to limited carbohydrate consumption, endogenous or exogenous fats are a preferential source of energy in LCKD, whereas a fewer number of calories gets derived from glucose . Ketone bodies generated during the metabolism of FFAs become a substitute for glucose and act as a substrate for gluconeogenesis. As there is a significant overlap between the mechanism of LCKD and SGLT2i, the presence of both factors in a single patient may predispose to overwhelming ketogenesis and cause ketoacidosis. Mechanistic similarities between LCKD and SGLT2i are given in Table 1.
We recognized two triggering factors of ketoacidosis in our patient. First, though he had used empagliflozin for years, interruption and then recontinuation following a gap of two weeks might have put him at the risk of ketoacidosis. This could be due to a maladaptive handling of ketone bodies. Second, he started using a LCKD during the same period with subsequent reduced insulin release followed by a shift of fuel energetics from carbohydrates to ketones. Both these factors led to excessive ketogenesis and then ketoacidosis. Perioperative euglycemic DKA due to SGLT2i has also been described . Pathophysiology of such euglycemic DKA in the perioperative period seems to be similar to that in LCKD. Perioperative factors, such as dehydration, prolonged fasting, reduced carbohydrate intake, stress, and pain, result in reduced insulin-to-glucagon ratio, cause lipolysis, and result in ketogenesis . As it is almost unavoidable to completely eliminate such perioperative factors, it is critical to discontinue SGLT2i at least 2–3 days prior to the surgery and allow enough time for the elimination of SGLT2i from the circulation .
Practitioners should be aware of SGLT2i-related DKA and its precipitating factors, such as LCKD. Patients on antihyperglycemics, such as SGLT2i and insulin, who start eating a LCKD are at risk of hypoglycemia due to low carbohydrates (reduced intake or urinary loss of glucose) in the presence of excess of the insulin.
For patients who wish to start LCKD, practitioners should recommend that they discontinue SGLT2i at least 3 days prior to the start of LCKD. Those who require insulin therapy may need to reduce the dose of insulin due to the risk of hypoglycemia. Inappropriate or rapid reduction and discontinuation of insulin on the other hand may cause insulin deficiency and promote further ketosis due to lipolysis, promoting ketoacidosis. Maintenance of adequate hydration should be encouraged as it promotes an adequate clearance of ketone bodies.
Blood sugar monitoring and careful adjustment of insulin therapy should be made with the help of the physician. Drastic and abrupt lifestyle changes should be avoided. Monitoring of blood or urinary ketones may help with early detection of ketoacidosis.
A high index of clinical suspicion is required to diagnose SGLT2i-induced-euglycemic DKA. As almost all patients who take SGLT2i have T2DM, the clinical manifestation of DKA may not be typical or anticipated at all. We report the rare presentation of DKA in a patient with T2DM who was taking empagliflozin and endorse the significance of a careful review of patient’s home medications. LCKD in the presence of SGLT2i can lower the threshold for SGLTi-induced DKA. Since the general practitioners commonly prescribe SGLT2i, they need to be well-informed of euglycemic DKA and its predisposing factors, such as LCKD. They should educate patients about this rare possibility and recommend stop using SGLT2i during a predisposing phase. Patients who experience any symptoms of DKA, such as persistent nausea or vomiting, should come to the emergency department for prompt treatment and laboratory workup.
Taylor SI, Blau JE, Rother KI. SGLT2 inhibitors may predispose to ketoacidosis. J Clin Endocrinol Metab. 2015;100(8):2849–52.
Sodium-glucose cotransporter-2 (SGLT2) inhibitors. US Food and Drug Administration. 2018. https://www.fda.gov/drugs/postmarket-drug-safety-information-patients-and-providers/sodium-glucose-cotransporter-2-sglt2-inhibitors. Accessed 02/25/2020.
Ogawa W, Sakaguchi K. Euglycemic diabetic ketoacidosis induced by SGLT2 inhibitors: possible mechanism and contributing factors. J Diabetes Investig. 2016;7(2):135–8.
FDA Drug Safety Communication: FDA warns that SGLT2 inhibitors for diabetes may result in a serious condition of too much acid in the blood. U.S. Food and Drug Administration. 2015. https://www.fda.gov/files/drugs/published/Drug-Safety-Communication%2D%2DFDA-warns-that-SGLT2-inhibitors-for-diabetes-may-result-in-a-serious-condition-of-too-much-acid-in-the-blood.pdf. Accessed 02/25/2020.
Erondu N, Desai M, Ways K, Meininger G. Diabetic ketoacidosis and related events in the canagliflozin type 2 diabetes clinical program. Diabetes Care. 2015;38(9):1680–6.
Goldenberg RM, Berard LD, Cheng AY, Gilbert JD, Verma S, Woo VC, et al. SGLT2 inhibitor-associated diabetic ketoacidosis: clinical review and recommendations for prevention and diagnosis. Clin Ther. 2016;38(12):2654–64.
Dizon S, Keely EJ, Malcolm J, Arnaout A. Insights into the recognition and management of sglt2-inhibitor-associated ketoacidosis: it's not just euglycemic diabetic ketoacidosis. Can J Diabetes. 2017;41(5):499–503.
Burke KR, Schumacher CA, Harpe SE. SGLT 2 inhibitors: a systematic review of diabetic ketoacidosis and related risk factors in the primary literature. Pharmacotherapy. 2017;37(2):187–94.
Masuta P, Johri G, Paul M. SGLT2 inhibitors and euglycemic ketoacidosis. Am J Ther. 2018;25(4):e498–500.
Zinman B, Wanner C, Lachin JM, Fitchett D, Bluhmki E, Hantel S, et al. Empagliflozin, cardiovascular outcomes, and mortality in type 2 diabetes. N Engl J Med. 2015;373(22):2117–28.
Qiu H, Novikov A, Vallon V. Ketosis and diabetic ketoacidosis in response to SGLT2 inhibitors: basic mechanisms and therapeutic perspectives. Diabetes Metab Res Rev. 2017;33(5):e2886.
Mudaliar S, Alloju S, Henry RR. Can a shift in fuel energetics explain the beneficial cardiorenal outcomes in the EMPA-REG OUTCOME study? A unifying hypothesis. Diabetes Care. 2016;39(7):1115–22.
Kalra S, Jain A, Ved J, Unnikrishnan AG. Sodium-glucose cotransporter 2 inhibition and health benefits: the Robin Hood effect. Indian J Endocrinol Metab. 2016;20(5):725–9.
Paoli A, Rubini A, Volek JS, Grimaldi KA. Beyond weight loss: a review of the therapeutic uses of very-low-carbohydrate (ketogenic) diets. Eur J Clin Nutr. 2013;67(8):789–96.
Yancy WS, Foy M, Chalecki AM, Vernon MC, Westman EC. A low-carbohydrate, ketogenic diet to treat type 2 diabetes. Nutr Metab (Lond). 2005;2(1):34.
Westman EC, Mavropoulos J, Yancy WS, Volek JS. A review of low-carbohydrate ketogenic diets. Curr Atheroscler Rep. 2003;5(6):476–83.
Tahrani AA, Barnett AH, Bailey CJ. SGLT inhibitors in management of diabetes. Lancet Diabetes Endocrinol. 2013;1(2):140–51.
Shintani H, Shintani T, Ashida H, Sato M. Calorie restriction mimetics: upstream-type compounds for modulating glucose metabolism. Nutrients. 2018;10(12):1821.
Lau A, Bruce S, Wang E, Ree R, Rondi K, Chau A. Perioperative implications of sodium-glucose cotransporter-2 inhibitors: a case series of euglycemic diabetic ketoacidosis in three patients after cardiac surgery. Can J Anaesth. 2018;65(2):188–93.
Conflict of Interest
Ethical Approval and Consent for Publication
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This article is part of the Topical Collection on Medicine
About this article
Cite this article
Khan, A., Mushtaq, K., Khakwani, M. et al. Metabolic Acidosis and its Predisposing Factor: Euglycemic Ketoacidosis Caused by Empagliflozin and Low-Carbohydrate Ketogenic Diet in Type 2 Diabetes Mellitus. Case Report. SN Compr. Clin. Med. (2020). https://doi.org/10.1007/s42399-020-00367-0
- Diabetes mellitus type 2
- Low-carbohydrate ketogenic diet