Current Diabetes Reports (2020) 20:25

https://doi.org/10.1007/s11892-020-01307-x

OTHER FORMS OF DIABETES AND ITS COMPLICATIONS (JJ NOLAN AND H THABIT, SECTION EDITORS)

Euglycemic Ketoacidosis
Benedetta Maria Bonora 1 & Angelo Avogaro 1 & Gian Paolo Fadini 1

# Springer Science+Business Media, LLC, part of Springer Nature 2020

Abstract
Purpose of Review Diabetic ketoacidosis is a life-threatening complication of diabetes characterized by hyperglycemia, acidosis,
and ketosis. Ketoacidosis may occur with blood glucose level < 200 mg/dl (improperly defined as euglycemic ketoacidosis,
euKA) and also in people without diabetes. The absence of marked hyperglycemia can delay diagnosis and treatment, resulting in
potential serious adverse outcomes.
Recent Findings Recently, with the wide clinical use of sodium glucose co-transporter 2 inhibitors (SGLT2i), euKA has come
back into the spotlight. Use of SGLT2i use can predispose to the development of ketoacidosis with relatively low or normal levels
of blood glucose. This condition, however, can occur, in the absence of diabetes, in settings such as pregnancy, restriction on
caloric intake, glycogen storage diseases or defective gluconeogenesis (alcohol abuse or chronic liver disease), and cocaine
abuse.
Summary euKA is a challenging diagnosis for most physicians who may be misled by the presence of normal glycemia or mild
hyperglycemia. In this article, we review pathophysiology, etiologies, clinical presentation and the management of euKA.

Keywords Euglycemic ketoacidosis . Ketoacidosis . Sodium-glucose cotransporter-2 inhibitors . Starvation . Pregnancy .

Alcohol

Introduction DKA suggest using a blood glucose (BG) threshold of

250 mg/dL and 200 mg/dL, respectively, to make the
It is common sense that most cases of ketoacidosis oc- diagnosis [1, 3].
cur in people with diabetes. Diabetic ketoacidosis Nevertheless, in 1973, Munro and colleagues reported that
(DKA) is a potentially life-threatening acute complica- among a series of 211 episodes of DKA in T1D, 37 had a BG
tion of diabetes that is still an important cause of mor- levels < 300 mg/dL and 16 episodes < 200 mg/dL [4]. They
bidity and mortality among people with diabetes. DKA referred this condition as “euglycemic ketoacidosis” (euKA).
may occurs both in people with type 1 (T1D) and rarely In a subsequent case series of 722 episodes of DKA in
type 2 diabetes (T2D) [1, 2]. DKA is a metabolic de- Birmingham, UK, 23 episodes of euKA were identified as
compensation caused from an absolute or severe relative defined by Munro’s criteria, but only eight presented with a
insulin deficiency and an increased concentration of BG < 200 mg/dL [5].
counterregulatory hormones (glucagon, catecholamines, This euglycemic presentation has been described in set-
cortisol, and growth hormone), resulting in a combina- tings such as pregnancy in women with or without diabetes
tion of hyperglycemia, ketosis, and metabolic acidosis. [6], restriction on carbohydrate intake or starvation [7], in
The US and the UK guidelines for the management of people with glycogen storage diseases [8, 9] or defective glu-
coneogenesis (alcohol abuse or chronic liver disease) [10],
This article is part of the Topical Collection on Other Forms of Diabetes sepsis [11], pancreatitis [12], and cocaine abuse [13].
and Its Complications Recently, interest in euKA has been renewed as euKA has
been associated with sodium glucose co-transporter 2 inhibi-
* Gian Paolo Fadini tors (SGLT2i) [14].
gianpaolo.fadini@unipd.it It presents a diagnostic and management challenge for the
1 physician, as the relatively low or normal BG may delay the
Department of Medicine, University of Padova, Via Giustiniani 2,
35128 Padova, Italy recognition, diagnosis, and treatment.
 25 Page 2 of 7 Curr Diab Rep (2020) 20:25

In this article, we briefly review pathogenesis, diagnosis,

causes, and treatment of euKA in people with and without
diabetes (Table 1).

Pathogenesis and Precipitating Factors

The pathogenesis of DKA is well established (Fig. 1). Briefly,

DKA is the result of an absolute or relative deficiency of
insulin associated with an increase of counter-regulatory hor-
mones (glucagon, cortisol, catecholamine, and growth hor-
mone). These metabolic changes promote an increase in he-
patic gluconeogenesis, an accelerated glycogenolysis, an im-
pairment of glucose utilization by peripheral tissues—which
causes hyperglycemia—and lipolysis with increased free fatty
acids (FFAs) mobilization from adipose tissue. In the liver,
FFAs are catabolized into ketone bodies (β-hydroxybutyrate
Fig. 1 Pathogenesis of diabetic ketoacidosis
and acetoacetate) [15]. Ketone bodies are the main contribu-
tors to metabolic acidosis. Hyperglycemia leads to glycosuria
and osmotic diuresis resulting into severe dehydration, hypo- Generally, DKA and euKA are induced by one or more of
volemia and a reduction in glomerular filtration, thus imped- the following precipitating factors: severe infection, discontin-
ing further glucose excretion [15]. uation of or inadequate insulin therapy, insulin pump failure,
In euKA, insulin deficiency and insulin resistance are often low carbohydrate intake or prolonged fasting, alcohol con-
milder, thereby limiting the surge in BG levels [16]. sumption, intercurrent acute events (pancreatitis, myocardial
Therefore, glucose overproduction and underutilization are infarction, cerebrovascular accident), drugs that affect carbo-
lesser than in DKA. In SGLT2i-induced euKA renal glucose hydrate metabolism (e.g., corticosteroids), surgery, and other
clearance is increased, contributing to lower BG levels [17]. stressful physical and medical conditions [18•]. These

Table 1 Causes, pathogenesis, and management of euKA

Causes Type of subjects Pathogenesis Treatment

SGLT2i Diabetic people Glycosuria ➔ reduction of blood glucose levels ➔ reduction in Stop SGLT2i, fluid replacement,
insulin production or self-reduction of insulin dose in insulin infusion and intravenous
insulin-treated patients + increase in glucagon concentrations ➔ glucose solution
Lipolysis and ketone production
Direct stimulation of alpha cells?
Strict diet/starvationPeople without Deficit carbohydrate ➔ shift to lipolysis Intravenous infusion of glucose
diabetes Depletion glycogen stores maintain BG level relatively low solution
Trigger in people Insulin only if diabetic patients
with diabetes
Pregnancy People with diabetes Nausea and vomiting Fluid replacement, insulin infusion
(T1D, T2D, Insulin resistance and intravenous glucose solution
GDM), usually Accelerated starvation Only intravenous glucose solution if
People without Respiratory alkalosis woman without diabetes
diabetes (rarely) Physiological hemodilution maintain BG level relatively low
Alcohol People without Oxidation ethanol ➔ ketones Intravenous glucose and large
diabetes Restriction on caloric intake amounts of intravenous saline
Trigger in people Depleted hepatic glycogen stores
with diabetes Impaired gluconeogenesis
Glycogen storage People with or Impaired glycogen storage Intravenous glucose solution, fluid
disease and chronic without diabetes Impaired gluconeogenesis replacement (insulin infusion)
liver disease
Cocaine abuse People with and Poor caloric intake Intravenous glucose solution, fluid
without diabetes Hypoglycemic effects of cocaine replacement, insulin infusion
Inherited metabolic People with or Defects of gluconeogenesis or glycogen metabolism Fluid replacement and intravenous
disease without diabetes glucose solution (insulin
infusion)
Curr Diab Rep (2020) 20:25 Page 3 of 7 25

precipitating factors are responsible for inducing significant numbers of DKA, with an incidence of less than 1 per 1000
increase in counter-regulatory hormones production. patient- or person-years [14]. However, the scenario might be
different in real life clinical practice, where conditions are far
from the “ideal” setting of randomized controlled trials
Clinical Presentation and Diagnosis (RCTs). In RCTs, patients are homogeneous, highly selected
and with close follow-up. A cohort study, using four large
Individuals with DKA and euKA typically present with the insurance claims US databases, reported an almost twofold
following clinical features: nausea, vomiting, diffuse abdom- increased DKA risk in new users of SGLT2i compared to
inal pain, polyuria, polydipsia, weight loss, dehydration, non-SGLT2i users [24]. Similarly, recent evidences derived from
weakness, fatigue, tachycardia, tachypnoea, and Kussmaul large claims database of commercially insured people in the USA,
breathing. Mental status can vary from full alertness to pro- suggests that the rate of DKA within 180 days after the initiation
found lethargy or coma [1]. of an SGLT2i was about twice the rate after the initiation of a
DKA is classically defined by the presence of the triad DPP4 inhibitor (4.9 events per 1000 person-years vs. 2.2 events
hyperglycemia (≥ 250 mg/dL), metabolic acidosis with elevat- per 1000 person-years; HR 2.2 95% CI 1.4–3.6) [25]. However,
ed anion gap (arterial pH < 7.3 and serum bicarbonate < in another cohort study, using a Korean Health Insurance data-
15 mEq/L, anion gap > 10–12 mEq/L) and elevated concen- base, SGLT2i treatment did not increase the risk of DKA com-
trations of ketone bodies. euKA is characterized by lower (< pared with DPP-4 inhibitor treatment (HR 0.96 95% CI 0.58–
200 mg/dL) BG levels; therefore, it is improperly defined as 1.57) [26]. To date, the exact rate at which this specific adverse
euglycemic because it also includes episode with BG above event occurs during SGLT2i therapy is difficult to establish.
the normal range. The severity of DKA is classified as mild, As expected, the risk of ketoacidosis is much greater in
moderate, or severe based on the severity of metabolic acido- people with T1D. SGLT2i phase 3 trials involving patients
sis (blood pH, bicarbonate, and anion gap) and the presence of with T1D have documented a significant increase in the risk
altered mental status [18]. for DKA in patients treated with SGLT2i, which appears to be
dose dependent. In DEPICT-1 (Dapagliflozin Evaluation in
Patients With Inadequately Controlled Type 1 Diabetes), at
Causes of euKA 52 weeks of follow-up, more patients in the dapagliflozin
groups had events adjudicated as DKA, at 4.0%, 3.4%, and
SGLT2i 1.9% in the dapagliflozin 5 mg, 10 mg, and placebo groups,
corresponding to incidence rate of 4.8, 3.7, and 2.2 per 100
SGLT2i constitute the latest developed class of glucose- patients-years, respectively [27]. In the EASE (Empagliflozin
lowering medications approved for the treatment of people as Adjunctive to inSulin thErapy) program, the rate of DKA in
with T2D. Recently, the clinical indication for SGLT2i patients on empagliflozin 2.5 mg was similar to placebo (0.8%
dapagliflozin was extended to people with T1D with BMI ≥ and 1.2%, respectively), while the rate was higher in the
27 kg/m2 [19] but only in Europe. To date, SGLT2i approval empagliflozin 10- and 25-mg groups compared with placebo
for T1D has been turned down by the US Food and Drug (4.3%, 3.3%, and 1.2%, respectively) corresponding to inci-
Administration (FDA), because available data on the dence rate of 5.9, 5.1, and 1.8 per 100 patients-years [28]. A
SGLT2i as an adjunct to insulin for the treatment of people similar trend was observed in the trials of sotagliflozin and
with T1D suggested that the overall benefits did not outweigh canagliflozin [29, 30].
the risks for potentially fatal DKA [20]. The proposed mechanism driving of SGLT2i-induced
SGLT2i block sodium-dependent glucose transporter-2 ketoacidosis is primarily related to a reduction in insulin concen-
(SGLT2) located in the early proximal renal tubule, which is trations and an increase in glucagon concentrations. The low plas-
responsible for reabsorption of most (80–90%) of the glucose ma glucose levels caused by glycosuria during SGLT2i therapy
filtered by the glomerulus. The resulting increase in urinary stimulates glucagon secretion and suppression of insulin produc-
glucose excretion lowers plasma glucose concentrations [17]. tion leading to an increase in the glucagon:insulin ratio and
In 2015, the US FDA and the European Medicines Agency, resulting in increased lipolysis, delivery of free fatty acids to the
based on the report of few cases, released a warning that liver, where they are oxidized. Lipid oxidation generates acetyl-
SGLT2i may cause DKA in both T1D and T2D people [21, CoA that is converted in ketone bodies when glycolysis interme-
22]. The number of reports is steadily increasing and several diates are unavailable due to reduced intracellular glucose oxida-
case reports of SGLT2i-associated DKA, as well as analysis of tion. These metabolic changes overall promote ketogenesis [31].
the FDA Adverse Event Reporting System have been pub- In addition, some studies have suggested that SGLT2 is also
lished in medical literature [14, 23]. The clinical trial programs expressed by human pancreatic alpha cells [32] and that
for the 3 most commonly used SGLT2i (canagliflozin, SGLT2i may exert a direct stimulatory effect on alpha cells [33,
dapagliflozin, and empagliflozin) reported numeratively small 34], although this finding is controversial and has not been
 25 Page 4 of 7 Curr Diab Rep (2020) 20:25

consistently replicated in all studies [35, 36, 37•]. Recently, Pregnancy

Capozzi et al. demonstrated, in diabetic mice, that SGLT2i-
induced ketosis is not affected by interruption of glucagon signal- DKA in pregnancy is a serious complication for the mother
ing by a glucagon receptor blocking antibody. Thus, the ketogenic and fetus and is associated with high mortality rates. DKA in
effects of SGLT2i probably occurs independently from glucagon pregnancy can occur in T1D, T2D, and in people with gesta-
[38]. tional diabetes (GDM) [44]. The exact rate at which this ad-
SGLT2i-associated DKA episodes sometimes presented verse event occurs during pregnancy is difficult to establish,
with relatively low or normal glucose levels because of gly- probably ranging between 0.5 and 10% of all diabetic gesta-
cosuria. Although the average glucose concentrations ob- tions, with 90% of these cases being women with GDM [45].
served during SGLT2i-induced DKA tended to be lower than Pregnancy is a ketosis-prone status due to intrinsic physi-
in classical DKA, most cases of SGLT2i-induced DKA had ological changes [6]. Especially in second and third trimester
elevated BG levels. In a review of 105 cases of SGLT2i- of pregnancy, a state of insulin resistance, accelerated starva-
induced DKA mean BG at admission was 294 ± 188 mg/dL tion, and respiratory alkalosis develops. In addition, nausea
and in 37 patients (35.2%) DKA was euglycemic, with an and vomiting are common in pregnancy. Maternal and placen-
admission BG of less than 200 mg/dL [14]. tal counter-regulatory hormones (progesterone, estrogen, cor-
Precipitating factors of SGLT2i-induced DKA include: ex- tisol, prolactin, human placental lactogen) and TNF-alpha
cessive reduction of insulin doses (> 50%) or insulin omis- play a key role in the decline in insulin sensitivity and com-
sion, especially if combined with a low-carbohydrate diet, pensatory increase in maternal insulin production to maintain
alcohol intake, acute event (e.g., myocardial infarction), infec- normal glucose metabolism. The decreased insulin sensitivity
tion or fever, trauma, surgery and all counter-regulatory states, is considered a physiological mechanism to deliver glucose to
insulin pump failure or malfunction. Some reports describe the fetus and the placenta, where it is used as a source of
the occurrence of DKA in people with presumed T2D who energy. This tends to reduce maternal BG concentrations
subsequently proved to be affected by autoimmune diabetes. [46]. On the other hand, maternal metabolism is diverted to
Nonetheless a minority of cases (22%) was not associated the use of FFA as primary fuel. Lipolysis makes pregnant
with any clear precipitating factor [14]. Duration of SGLT2i women more susceptible to ketosis. In addition, to compen-
treatment before DKA onset was extremely variable (0.3– sate for respiratory alkalosis, an increased renal excretion of
420 days) [14]. bicarbonate takes place, favoring metabolic acidosis [45].
The physiological hemodilution during pregnancy like-
ly contributes to the maintenance of relatively low or nor-
Low Caloric Intake or Starvation mal BG levels in women with diabetes [6]. In fact, DKA
in pregnant woman tends to occur at lower BG levels than
Restriction on carbohydrate/caloric intake or prolonged in non-pregnant women with diabetes. Several cases of
fasting with the maintenance of insulin therapy can be a pre- euKA with BG < 200 in pregnancy were reported in the
cipitating factor of DKA. Starvation or strict restriction of literature. In almost all cases, the episode of euKA oc-
carbohydrate intake is a cause of euKA also in individuals curred in the third trimester (32–37 week of gestation).
without diabetes [39, 40]. In fact, reduced dietary carbohy- The average age of this cohort was 30-years and average
drate leads to the use of alternative energy sources, FFA and BG level 110 mg/dl. Cases were equally distributed
ketones generated by lipolysis. It has been observed that dur- among T1D, T2D and GDM [47].
ing fasting, lipolysis and FFA production are accelerated [41], Fortunately, with improved prenatal counseling, under-
insulin is less effective at suppressing lipolysis, and ketogen- standing and management, the incidence of euKA/DKA in
esis occurs which leads to euKA [42]. In addition, prolonged pregnancy and its mortality have declined significantly over
fasting may result in a nearly total depletion of glycogen stores recent years [44].
and contribute to the euglycemic state [43]. Owen et al. de- Precipitating factors and clinical presentation of diabetic
scribed a case of a 24-year-old male without medical history ketoacidosis in pregnancy is similar to that of non-pregnant
who, due to nausea and vomiting, fasted for 3 days. He pre- diabetic people, except they tend to develop more rapidly.
sented at the emergency department with tachypnoea, sinus Treatment includes fluid replacement, insulin infusion, and
tachycardia, metabolic acidosis with high anion gap, low bi- intravenous glucose solution [48].
carbonate, substantial ketonuria and increased blood level of Adverse outcomes for the fetus include malformations, in-
β-hydroxybutyrate. His pH started to normalize immediately trauterine death, pre-term labor, hypoxia, distress, cardiac ar-
with intravenous infusion of dextrose [40]. However, starva- rhythmias, hyperbilirubinemia, hypoglycemia, and neurologic
tion or carbohydrate intake restriction can precipitate euKA complications. Maternal complications include preeclampsia,
more commonly in diabetic than in non-diabetic people and eclampsia, acute kidney injury, adult respiratory distress syn-
usually represents a precipitating factor of typical DKA. drome, coma, and death [44, 47].
Curr Diab Rep (2020) 20:25 Page 5 of 7 25

In addition, two cases of euKA have been reported to be a In individuals with cocaine abuse, poor caloric intake and
consequence of starvation in pregnant women without diabe- hypoglycemic effects of cocaine are potential causes of
tes or GDM. Both patients were treated with intravenous dex- euglycemia observed during ketoacidosis [13]. Ingestion of
trose [49, 50]. Thus, it is important to keep in mind that euKA drugs such as salicylate, methanol, ethylene glycol, and par-
can also occurs in pregnant women without diabetes and that aldehyde can also cause metabolic acidosis with euglycemia,
early diagnosis and treatment can reduce the risk of adverse accompanied by increased levels of ketone bodies [56, 57].
outcomes for both the mother and the fetus. Measurement of these drugs and intoxicants may be helpful
for differential diagnosis.
Alcoholic Ketoacidosis Ketoacidosis is also a feature of several inherited metabolic
disease seen mostly in childhood. When due to defects of
Alcoholic ketoacidosis (AKA) is a subtype of euKA that oc- gluconeogenesis or glycogen metabolism (e.g., glycogen stor-
curs in individuals without diabetes [10]. In 1940, Dillon and age disease), and maple syrup urine disease, ketogenesis is
colleagues described a series of nine individuals without dia- accompanied by euglycemia or even hypoglycemia [58].
betes, with a history of chronic alcohol consumption, that
experienced ketoacidosis [51]. Further case series were sub-
sequently reported. AKA is similar in presentation to euKA Mitigating euKA
although often with normal or low glucose values and no
glycosuria [52]. The acidosis is attributable to the accumula- Prevention of ketosis and progression to ketoacidosis is ideal-
tion of lactate and ketone bodies. The absence of hyperglyce- ly the first target to mitigate euKA. People with diabetes or
mia is the result of prolonged restriction on caloric intake, with other conditions that predispose to euKA, as well as
depleted hepatic glycogen stores, and impaired gluconeogen- physicians, should be instructed about precipitating factors
esis (caused by ethanol that raises the NADH/NAD ratio). and how to avoid them. They should be able to recognize
Ketones are derived by the oxidation of ethanol into acetalde- promptly signs and symptoms of ketoacidosis and be advised
hyde, which is, in turn, converted into acetate and then into that ketoacidosis can occur at any glycemic level. Individuals
acetoacetate [53]. with diabetes using interstitial glucose sensors may be misled
Patients typically present with classic DKA symptoms by glucose readings within or close to target range, thereby
which improve rapidly with intravenous glucose and large failing to recognize incipient euKA.
volumes of intravenous saline, usually without insulin. People with diabetes should be educated on how to adjust
Occasionally, intravenous bicarbonate may be needed to cor- their insulin dosage during illness. Self- measurement of cap-
rect the metabolic disturbance [10]. illary blood ketones in those using devices licensed for this
use should be recommended during illness or with signs and
Other Causes of euKA symptoms consistent with ketosis (malaise, nausea, and
vomiting).
Other causes of euKA that have been reported in literature can To minimize the risk of DKA/euKA associated with
occur in both people with and without diabetes. Prater and SGLT2i, international societies released a position statement
Chaiban described a case of euKA in a non-diabetic individ- that recommend: (i) stopping SGLT2i at least 24 h prior to
uals during an acute admission with pancreatitis [12]. elective surgery, planned invasive procedures, or anticipated
euKA in an individual without diabetes and without history severe stressful physical activity; (ii) stopping immediately for
of starvation/low-carbohydrate diet or alcohol abuse has been emergency surgery or any extreme stress event or any situa-
described in a 76-year-old woman with septic shock due to tion that might precipitate DKA including acute illness; (iii)
acute obstructive cholangitis [11]. The pathogenesis of the so avoid stopping insulin or decreasing the dose excessively;
called “septic ketoacidosis” is not completely understood. and(iv) avoid excess alcohol intake and very-low-carbohy-
People with sepsis have increased level of counter- drate/ketogenic diets [59•].
regulatory hormones and insulin resistance mediated by Careful management of DM in pregnant women, particu-
counter-regulatory hormones and pro-inflammatory cytokine larly in the last trimester, is very important to prevent the
through mechanisms involving hypothetically insulin receptor occurrence DKA/euKA.
and post-receptor signaling. These metabolic changes can in-
duce ketogenesis even in people without diabetes mellitus
[54]. Finally, septic shock induces kidney dysfunction that Management of euKA
may reduce ketone excretion [11].
Segebrecht et al. reported 2 cases of euKA in diabetic in- The treatment of euKA in diabetic individuals is similar to that
dividuals after abdominal surgery apparently not due to of hyperglycemic DKA. Guidelines for the management of
prolonged fasting [55]. DKA recommend rapid fluid replacement, followed by
 25 Page 6 of 7 Curr Diab Rep (2020) 20:25

continuous intravenous insulin infusion, correction of electro- 8. Stoimenis D, Spyridonidou C, Theofanidou S, Petridis N,
Papaioannou N, Iasonidou C, et al. Euglycemic ketoacidosis in
lyte imbalances and dextrose-containing solutions when BG
spinal muscular atrophy. Case Rep Pediatr. 2019;2019:2862916.
levels are below 250 mg/dL. The treatment should continue 9. Lee SH, Park JH, Hong M-K, Hyeon S, Kim MY, Lee SE, et al.
until resolution of ketoacidosis (normalization of acidosis and True euglycemic diabetic ketoacidosis in a person with type 2 dia-
ketone bodies). Bicarbonate therapy can be necessary for se- betes and Duchenne muscular dystrophy. Diabetes Res Clin Pract.
2011;92(1):e7–8.
vere acidosis [1]. In case of euKA in non-diabetic individuals,
10. McGuire LC, Cruickshank AM, Munro PT. Alcoholic ketoacidosis.
insulin infusion is not necessary, whereas fluid replacement, Emerg Med J. 2006 Jun;23(6):417–20.
correction of electrolyte imbalances and intravenous glucose 11. Nakamura K, Inokuchi R, Doi K, Fukuda T, Tokunaga K, Nakajima
solution are sufficient for the resolution of acidosis [60]. S, et al. Septic ketoacidosis. Intern Med. 2014;53(10):1071–3.
12. Prater J, Chaiban JT. Euglycemic diabetic ketoacidosis with acute
pancreatitis in a patient not known to have diabetes. AACE Clin
Case Rep. 2014;1(2):e88–91.
13. Abu-Abed Abdin A, Hamza M, Khan MS, Ahmed A. Euglycemic
Conclusions diabetic ketoacidosis in a patient with cocaine intoxication. Case
Rep Crit Care. 2016;2016:4275651.
Ketoacidosis is a rare but serious medical emergency that 14. Bonora BM, Avogaro A, Fadini GP. Sodium-glucose co-
transporter-2 inhibitors and diabetic ketoacidosis: an updated re-
requires immediate hospital care. EuKA is a diagnostic chal- view of the literature. Diabetes Obes Metab. 2018;20(1):25–33.
lenge due to the absence of marked hyperglycemia, which can 15. Kitabchi AE, Wall BM. Diabetic ketoacidosis. Med Clin North Am.
result in delayed recognition and treatment leading to poten- 1995;79(1):9–37.
tially serious complications and death. With the introduction 16. Rosenstock J, Ferrannini E. Euglycemic diabetic ketoacidosis: a
predictable, detectable, and preventable safety concern with
of SGLT2i for the management of diabetic subjects, the inci- SGLT2 inhibitors. Diabetes Care. 2015 Sep 1;38(9):1638–42.
dent of euKA has increased, but is still rare. People with dia- 17. Bakris GL, Fonseca VA, Sharma K, Wright EM. Renal sodium–
betes or with conditions that predispose to euKA, as well as glucose transport: role in diabetes mellitus and potential clinical
healthcare professionals looking after them, should be educat- implications. Kidney Int. 2009;75(12):1272–7.
18.• Nyenwe EA, Kitabchi AE. The evolution of diabetic ketoacidosis:
ed about precipitating factors, signs and symptoms of An update of its etiology, pathogenesis and management.
ketoacidosis and be advised that ketoacidosis can occur at Metabolism. 2016;65(4):507–21. Detailed and updated review
any glycemic levels. on etiology, pathogenesis and management of ketoacidosis.
19. Tahrani AA, Barnett AH, Bailey CJ. SGLT inhibitors in manage-
ment of diabetes. Lancet Diabetes Endocrinol. 2013;1(2):140–51.
20. Nathan DM. Adjunctive treatments for type 1 diabetes. N Engl J
Med. 2017 Sep 13;377(24):2390–1.
21. U.S. Food and Drug Administration. Drug Safety Communication.
References FDA warns that SGLT2 inhibitors for diabetes may result in a seri-
ous condition of too much acid in the blood. 2015. https://www.fda.
gov/media/92185/download
Papers of particular interest, published recently, have been 22. European Medicines Agency. Review of diabetes medicines called
highlighted as: SGLT2 inhibitors started Risk of diabetic ketoacidosis to be exam-
• Of importance ined. 2015;44(June):1–2. https://www.ema.europa.eu/en/documents/
referral/sglt2-inhibitors-article-20-procedure-review-started_en.pdf
1. Kitabchi AE, Umpierrez GE, Miles JM, Fisher JN. Hyperglycemic 23. Fadini GP, Bonora BM, Avogaro A. SGLT2 inhibitors and diabetic
crises in adult patients with diabetes. Diabetes Care. 2009 ketoacidosis: data from the FDA Adverse Event Reporting System.
Jul 1;32(7):1335–43. Diabetologia. 2017;60(8):1385–9.
24. Wang L, Voss EA, Weaver J, Hester L, Yuan Z, DeFalco F, et al.
2. Wang ZH, Kihl-Selstam E, Eriksson JW. Ketoacidosis occurs in
Diabetic ketoacidosis in patients with type 2 diabetes treated with
both type 1 and type 2 diabetes— a population-based study from
sodium glucose co-transporter 2 inhibitors versus other
northern Sweden. Diabet Med. 2008 Jul 1;25(7):867–70.
antihyperglycemic agents: An observational study of four US ad-
3. Joint British Diabetes Societies Inpatient Care Group The ministrative claims databases. Pharmacoepidemiol Drug Saf.
Management of Diabetic Ketoacidosis in Adults. 2013. https:// 2019;28(12):1620–8.
abcd.care/joint-british-diabetes-societies-jbds-inpatient-care-group 25. Fralick M, Schneeweiss S, Patorno E. Risk of diabetic ketoacidosis
4. Munro JF, Campbell IW, McCuish AC, Duncan LJP. Euglycaemic after initiation of an SGLT2 inhibitor. N Engl J Med. 2017
diabetic ketoacidosis. Br Med J. 1973 Jun 9;2(5866):578–80. Jun 7;376(23):2300–2.
5. Jenkins D, Close CF, Krentz AJ, Nattrass M, Wright AD. 26. Kim Y-G, Jeon JY, Han SJ, Kim DJ, Lee K-W, Kim HJ. Sodium-
Euglycaemic diabetic ketoacidosis : does it exist? Acta Diabetol. glucose co-transporter-2 inhibitors and the risk of ketoacidosis in
1993;30(4):251–3. patients with type 2 diabetes mellitus: a nationwide population-
6. Guo R-X, Yang L-Z, Li L-X, Zhao X-P. Diabetic ketoacidosis in based cohort study. Diabetes Obes Metab. 2018 Aug 1;20(8):1852–8.
pregnancy tends to occur at lower blood glucose levels: case–con- 27. Dandona P, Mathieu C, Phillip M, Hansen L, Tschöpe D, Thorén F,
trol study and a case report of euglycemic diabetic ketoacidosis in et al. Efficacy and safety of Dapagliflozin in patients with inade-
pregnancy. J Obstet Gynaecol Res. 2008 Jun 1;34(3):324–30. quately controlled type 1 diabetes: the DEPICT-1 52-week study.
7. Joseph F, Anderson L, Goenka N, Vora J. Starvation-induced true Diabetes Care. 2018 Dec 1;41(12):2552–9.
diabetic euglycemic ketoacidosis in severe depression. J Gen Intern 28. Rosenstock J, Marquard J, Laffel LM, Neubacher D, Kaspers S,
Med. 2008;24(1):129. Cherney DZ, et al. Empagliflozin as adjunctive to insulin therapy in
Curr Diab Rep (2020) 20:25 Page 7 of 7 25

type 1 diabetes: the EASE trials. Diabetes Care. 2018 Dec 1;41(12): 44. Sibai BM, Viteri OA. Diabetic ketoacidosis in pregnancy. Obstet
2560–9. Gynecol. 2014;123(1):167–78.
29. Sands AT, Zambrowicz BP, Rosenstock J, Lapuerta P, Bode BW, 45. Parker JA, Conway DL. Diabetic ketoacidosis in pregnancy. Obstet
Garg SK, et al. Sotagliflozin, a dual SGLT1 and SGLT2 inhibitor, as Gynecol Clin N Am. 2007;34(3):533–43.
adjunct therapy to insulin in type 1 diabetes. Diabetes Care. 2015 46. Ryan EA, Enns L. Role of gestational hormones in the induction of
Jul 1;38(7):1181–8. insulin resistance. J Clin Endocrinol Metab. 1988 Aug 1;67(2):
30. Henry RR, Thakkar P, Tong C, Polidori D, Alba M. Efficacy and 341–7.
safety of canagliflozin, a sodium–glucose cotransporter 2 inhibitor, 47. Jaber JF, Standley M, Reddy R. Euglycemic diabetic ketoacidosis
as add-on to insulin in patients with type 1 diabetes. Diabetes Care. in pregnancy: a case report and review of current literature. In:
2015 Dec 1;38(12):2258–65. Waxman KS, editor. Case Reports Crit Care, vol. 2019; 2019. p.
31. Ferrannini E, Baldi S, Frascerra S, Astiarraga B, Heise T, Bizzotto 8769714.
R, et al. Shift to fatty substrate utilization in response to sodium– 48. Kamalakannan D, Baskar V, Barton DM, Abdu TAM. Diabetic
glucose cotransporter 2 inhibition in subjects without diabetes and ketoacidosis in pregnancy. Postgrad Med J. 2003 Aug;79(934):
patients with type 2 diabetes. Diabetes. 2016 May 1;65(5):1190–5. 454–7.
32. Chen J, Williams S, Ho S, Loraine H, Hagan D, Whaley JM, et al.
49. Burbos N, Shiner AM, Morris E. Severe metabolic acidosis as a
Quantitative PCR tissue expression profiling of the human SGLT2
consequence of acute starvation in pregnancy. Arch Gynecol
gene and related family members. Diabetes Ther. 2010;1(2):57–92.
Obstet. 2008;279(3):399.
33. Ferrannini E, Muscelli E, Frascerra S, Baldi S, Mari A, Heise T,
et al. Metabolic response to sodium-glucose cotransporter 2 inhibi- 50. Patel A, Felstead D, Doraiswami M, Stocks GM, Waheed U. Acute
tion in type 2 diabetic patients. J Clin Invest. 2014;124(2):499–508. starvation in pregnancy: a cause of severe metabolic acidosis. Int J
34. Bonner C, Kerr-Conte J, Gmyr V, Queniat G, Moerman E, Obstet Anesth. 2011;20(3):253–6.
Thévenet J, et al. Inhibition of the glucose transporter SGLT2 with 51. Dillon E, Dyer W, Smelo L. Ketone acidosis of non-diabetic adults.
dapagliflozin in pancreatic alpha cells triggers glucagon secretion. Med Clin North Am. 1940;24:1813–22.
Nat Med. 2015 Apr 20;21:512–7. 52. Umpierrez GE, DiGirolamo M, Tuvlin JA, Isaacs SD, Bhoola SM,
35. Solini A, Sebastiani G, Nigi L, Santini E, Rossi C, Dotta F. Kokko JP. Differences in metabolic and hormonal milieu in
Dapagliflozin modulates glucagon secretion in an SGLT2- diabetic- and alcohol-induced ketoacidosis. J Crit Care.
independent manner in murine alpha cells. Diabetes Metab. 2000;15(2):52–9.
2017;43(6):512–20. 53. Lefèvre A, Adler H, Lieber CS. Effect of ethanol on ketone metab-
36. Kuhre RE, Ghiasi SM, Adriaenssens AE, Wewer Albrechtsen NJ, olism. J Clin Invest. 1970 Oct;49(10):1775–82.
Andersen DB, Aivazidis A, et al. No direct effect of SGLT2 activity 54. Chambrier C, Laville M, Rhzioual Berrada K, Odeon M,
on glucagon secretion. Diabetologia. 2019;62(6):1011–23. Boulètreau P, Beylot M. Insulin sensitivity of glucose and fat me-
37.• Wang M, Yu X, Lee Y, McCorkle SK, Chen S, Li J, et al. tabolism in severe sepsis. Clin Sci. 2000;99:321–8.
Dapagliflozin suppresses glucagon signaling in rodent models of 55. Segebrecht R, Moncure M, Bennett A, Geehan D, Van Way IIICW,
diabetes. Proc Natl Acad Sci. 2017;114(25):6611–6. Recently Weide L. Ketoacidosis in euglycemic patients with type 2 diabetes
published paper on the ketogenic effects of SGLT2i that chal- after abdominal surgery. J Parenter Enter Nutr. 2019;43(7):937–40.
lenges the dogma that glucagon is a primary factor in the reg- 56. Arena FP, Dugowson C, Saudek CD. Salicylate-induced hypogly-
ulation of ketone production. cemia and ketoacidosis in a nondiabetic adult. Arch Intern Med.
38. Capozzi ME, Coch RW, Koech J, Astapova II, Wait JB, Encisco 1978 Jul 1;138(7):1153–4.
SE, et al. The Limited role of glucagon for ketogenesis during 57. Kraut JA, Kurtz I. Toxic alcohol ingestions: clinical features, diag-
fasting or in response to SGLT2 inhibition. Diabetes. 2020;69(5): nosis, and management. Clin J Am Soc Nephrol. 2008 Jan 1;3(1):
882–92. 208–25.
39. Larroumet A, Camoin M, Foussard N, Alexandre L, Mesli S, 58. Alfadhel M, Babiker A. Inborn errors of metabolism associated
Redonnet I, et al. Euglycemic ketoacidosis induced by therapeutic with hyperglycaemic ketoacidosis and diabetes mellitus: narrative
fasting in a non-diabetic patient. Nutrition. 2020;72:110668. review. Sudan J Paediatr. 2018;18(1):10–23.
40. Owen D, Little S, Leach R, Wyncoll D. A patient with an unusual
59.• Handelsman Y, Henry RR, Bloomgarden ZT, Dagogo-Jack S,
aetiology of a severe ketoacidosis. Intensive Care Med. 2008;34(5):
DeFronzo RA, Einhorn D, et al. American Association of
971–2.
Clinical Endocrinologists and American College of Endocrinolgy
41. Wolfe RR, Peters EJ, Klein S, Holland OB, Rosenblatt J, Gary H.
position statement on the association of SGLT-2 inhibitors and
Effect of short-term fasting on lipolytic responsiveness in normal
diabetic ketoacidosis. Endocr Pract. 2016;22(6):753–62. Detailed
and obese human subjects. Am J Physiol Metab. 1987;252(2):
and comprehensive position statement specifically dedicated to
E189–96.
the association of SGLT2i and Diabetic Ketoacidosis and for
42. Jensen MD, Haymond MW, Gerich JE, Cryer PE, Miles JM.
management of DKA in patients taking SGLT2i.
Lipolysis during fasting. Decreased suppression by insulin and in-
creased stimulation by epinephrine. J Clin Invest. 1987 Jan;79(1): 60. Le Neveu F, Hywel B, Harvey JN. Euglycaemic ketoacidosis in
207–13. patients with and without diabetes. Pract Diabetes. 2013;30(4):
43. Burge MR, Hardy KJ, Schade DS. Short-term fasting is a mecha- 167–71.
nism for the development of euglycemic ketoacidosis during pe-
riods of insulin deficiency. J Clin Endocrinol Metab. 1993 Publisher’s Note Springer Nature remains neutral with regard to jurisdic-
May 1;76(5):1192–8. tional claims in published maps and institutional affiliations.