Letter to the Editor
DIABETIC KETOACIDOSIS IN PATIENTS WITH TYPE 2 DIABETES
RECENTLY COMMENCED ON SGLT-2 INHIBITORS:
AN ONGOING CONCERN
Mohamed Ahmed, PGD (Diab), MBBS; Malachi J. McKenna, MD; Rachel K. Crowley, MD
ABSTRACT
Objective: Sodium-glucose cotransporter 2 (SGLT-2)
inhibitors are increasingly used as an adjunctive treatment
for type 2 diabetes. We report the occurrence of diabetic
ketoacidosis (DKA) in 3 patients with type 2 diabetes
recently commenced on SGLT-2 inhibitors.
Methods: Clinical presentation, laboratory data, and
treatment outcomes of all 3 cases are described.
Results: All 3 patients had documented history of
longstanding type 2 diabetes. The presentation in all
patients was that of hyperglycaemia, acidosis, and ketosis
occurring within 4 weeks of commencing SGLT-2 inhibi-
tors. The risk factors for developing DKA were infection,
myocardial infarction, and alcohol excess. DKA resolved
within 24 hours of initiating intravenous fluids and insulin
in all cases.
Conclusion: This case series illustrates the impor-
tance of careful patient selection, education, and monitor-
ing when starting this group of antidiabetic medications.
(Endocr Pract. 2017;23:506-508)
Submitted for publication June 13, 2016
Accepted for publication January 5, 2017
From the Department of Endocrinology, St. Vincent’s University Hospital
and University College Dublin, Dublin, Ireland.
Address correspondence to Dr. Mohamed Ahmed, St. Vincent’s University
Hospital, Elm Park, Dublin 4, Ireland.
E-mail: mohamedomer07@gmail.com
Published as a Rapid Electronic Article in Press at http://www.endocrine
practice.org. DOI: 10.4158/EP161447.LT
To purchase reprints of this article, please visit: www.aace.com/reprints.
Copyright © 2017 AACE.
506 ENDOCRINE PRACTICE Vol 23 No. 4 April 2017
Abbreviations:
DKA = diabetic ketoacidosis; SGLT-2 = sodium-
glucose cotransporter 2; T2D = type 2 diabetes
INTRODUCTION
Sodium-glucose cotransporter 2 (SGLT-2) inhibi-
tors are a new class of oral hypoglycemics that are now
widely used for the treatment of type 2 diabetes (T2D).
They improve glycemic control by inhibiting renal glucose
re-absorption. They are oral agents, and other potential
benefits of the class include weight loss, lower risk of
hypoglycemia, and blood pressure reduction (1,2). In May
2015, based on the identification of 20 reported cases of
diabetic ketoacidosis (DKA) in patients treated with SGLT-
2 inhibitors, the U.S. Food and Drug Administration issued
a drug safety communication warning of this potential
complication of SGLT-2 inhibitors (3).
CASE SERIES
The first patient was a 76-year-old man. Three weeks
after canagliflozin was added, he presented with nausea and
vomiting; however, there was mild evidence of dehydra-
tion (elevated creatinine, see Table 1). His blood glucose
was 11.7 mmol/L with capillary ketones 3.4 mmol/L (refer-
ence range, <0.6 mmol/L). His troponin was elevated at
5.9 ng/mL (normal, <0.005 ng/mL), and he was treated for
acute coronary syndrome and DKA.
The second patient was a 75-year-old man with T2D
for 33 years and poor glycemic control who was not a
candidate for insulin due to cognitive impairment and poor
social circumstances. Four weeks after dapagliflozin was
commenced, he presented with anorexia, nausea, and confu-
sion. He was dehydrated, with dry mucous membranes and
Copyright © 2017 AACE
Copyright © 2017 AACE Letter to the Editor, Endocr Pract. 2017;23(No. 4) 507

tachycardia, and his blood glucose was 19.8 mmo/L with
serum ketones of 5.8 mmol/L.
The third patient was commenced on canagliflozin 1
week before admission. He had approximately 12 units
of alcohol the night before presentation and was admitted
with abdominal pain and vomiting. He was clinically dehy-
drated with hypotension, and his blood glucose was 18.4
mmol/L with capillary ketones of 4.1 mmol/L. He required
intensive care monitoring and inotropic support.
Ketoacidosis resolved within 24 hours after initiating
fluids and insulin in all cases. None of these patients had
a previous history of DKA, which developed relatively
rapidly after introduction of an SGLT-2 inhibitor; nor did
DKA recur after withdrawal of SGLT-2 inhibitor. Table
1 summarizes the clinical characteristics of all 3 patients
with DKA.
DISCUSSION
The exact mechanism of SGLT-2 inhibitor-related
DKA is not yet fully known. In subjects with type 2 diabe-
tes mellitus, SGLT-2 inhibitor use results in a significant
glucose excretion of 50 to 100 g/day (4), with a reduction
in glucose-mediated beta-cell stimulation and a consequent
fall in plasma insulin level. Plasma glucagon level increas-
es due to diminished insulin inhibition and the suppression
of SGLT-2 receptors located on the pancreatic alpha-cells
(5,6). Furthermore, animal studies reveal that glucagon

Table 1
Clinical Characteristics of Cases with DKA
Case patient 1 2 3
Age (years) 76 75 60
Sex Male Male Male
Duration of DM (years) 15 33 12
Sitagliptin/Metformin Metformin 1,000 mg
Metformin 1,000 mg
Oral hypoglycaemic Agents (50/1,000 mg) Gliclazide MR 120 mg
Linagliptin 5 mg
Gliclazide MR 120 mg Linagliptin 5 mg
Insulin No No No
BMI (kg/m2) 24.7 22.2 32.1
Prior A1C (mmol/L [%]) 79 (9.4) 110 (12.2%) 63 (7.9%)
Previous DKA No No No
SGLT-2 inhibitor/dose Canagliflozin/100 mg Dapagliflozin/10 mg Canagliflozin/100 mg
Duration of SGLT-2 (weeks) 3 4 1
Plasma glucose (mg/dL [mmol/L]) 210.8 (11.7) 356.7 (19.8) 331.5 (18.4)
Serum ketones
3.4 5.8 4.1
(β-hydroxybutyrate) (mmol/L)
pH 7.26 7.29 6.88
Bicarbonate (mEq/L) 17.0 16 8
pCO2 (kPa) 2.28 4.34 3.32
Chloride (mmol/L) 105 111 106
Sodium (mmol/L) 130 139 136
Potassium (mmol/L) 3.2 4.4 4.5
BUN 6.9 22.1 11.3
Creatinine 116 185 169
Anion gap (mEq/L) 11.2 16.4 26.5
Potential contributors Myocardial infarction Urinary tract infection Excess alcohol
Clinical area managed Coronary care unit Medical ward ICU
Abbreviations: A1C = hemoglobin A1C; BMI = body mass index; BUN = blood urea nitrogen; DKA = diabetic ketoacidosis; DM =
diabetes mellitus; ICU = intensive care unit; SGLT-2 = sodium-glucose cotransporter 2.
508 Letter to the Editor, Endocr Pract. 2017;23(No. 4)
inhibits glucose-stimulated insulin secretion via hepatic
kisspeptin 1 production (7). The net result of these changes
is increased glucagon concentration and a reduction in insu-
lin to glucagon ratio, which leads to enhanced fat oxidation,
lipolysis, and ketogenesis (8). Another proposed mechanism
is the increased levels of renal tubular fluid sodium concen-
tration due to the inhibition of SGLT-2–mediated sodium
re-absorption, which leads to an increased re-absorption of
the negatively charged ketone bodies through an electrome-
chanical gradient carrier-mediated mechanism, resulting in
increased levels of circulating ketone bodies and increasing
the risk of developing DKA (9).
External factors reported to increase the risk of DKA
associated with SGLT-2 inhibitors include reduction in
insulin dose following initiation of these agents, low-carbo-
hydrate diet, myocardial infarction, surgery, pancreatitis,
sepsis, alcohol use, and severe injury (10). Dehydration,
which was evident in all of our three cases, can result from
the osmotic diuresis associated with SGLT-2 inhibition,
leading to worsening of the hypovolemic state of DKA.
This in turn results in the stimulation of counterregulatory
hormones, namely epinephrine, glucagon, cortisol, and
growth hormone, leading to increased lipolysis, insulin
resistance, and ketogenesis (11).
The canagliflozin clinical trial program indicated that
the incidence of DKA in people with T2D was comparable
to that of the general population (12). The authors of the
trial report indicated that most subjects had risk factors for
DKA, such as those we have listed, or had been misdiag-
nosed with type 1 diabetes or latent autoimmune diabetes
in adults. All our three cases had documented follow-up
for T2D; however, the longstanding history and the poor
control in the first two cases suggest beta-cell failure as
a risk for DKA. Appropriate patient selection for SGLT-
2 inhibitors and follow-up are vital, and SGLT-2 inhibi-
tors should not be used as a substitution for insulin when
this is clinically indicated. In the third case, the triggering
factor was excess alcohol, and patients should be educated
to avoid this when commencing these agents.
All other oral hypoglycemic agents currently market-
ed for T2D have been commenced in patients with T2D
of long duration that could be regarded as having a risk
of DKA, but there has never been such a strong postmar-
keting signal of risk of DKA with any other agent. The
risk of DKA in T2D of long duration is readily apparent
to diabetologists but not to primary care physicians, who
were the prescribing physicians in two of these three cases.
Physicians are under considerable pressure to prescribe
these agents, given their advantages in weight loss and
potential in cardiovascular disease risk reduction.
Copyright © 2017 AACE
CONCLUSION
In conclusion, for the right patient, the SGLT-2 inhibi-
tor class may be a very useful tool, and we have found
this to be the case in our own practice. However, this case
series illustrates the importance of careful patient selection
when choosing SGLT-2 inhibitors. Patient selection and
safety issues should be highlighted both in guidelines and
in product information provided with this drug class.
DISCLOSURE
The authors have no multiplicity of interest to disclose.
REFERENCES
1. Plosker GL. Canagliflozin: a review of its use in patients
with type 2 diabetes mellitus. Drugs. 2014;74:807-824.
2. Oliva RV, Bakris GL. Blood pressure effects of sodi-
um-glucose co-transport 2 (SGLT2) inhibitors. J Am Soc
Hypertens. 2014;8:330-339.
3. FDA Drug Safety Communication. FDA warns that
SGLT2 inhibitors for diabetes may result in a serious condi-
tion of too much acid in the blood. Available at: http://www.
fda.gov/Drugs/DrugSafety/ucm446845.htm. Accessed
February 21, 2017.
4. Sha S, Devineni D, Ghosh A, et al. Canagliflozin, a novel
inhibitor of sodium glucose co-transporter 2, dose depend-
ently reduces calculated renal threshold for glucose excre-
tion and increases urinary glucose excretion in healthy
subjects. Diabetes Obes Metab. 2011;13:669-672.
5. Maruyama H, Hisatomi A, Orci L, Grodsky GM, Unger
RH. Insulin within islets is a physiologic glucagon release
inhibitor. J Clin Invest. 1984;74:2296-2299.
6. Bonner C, Kerr-Conte J, Gmyr V, et al. Inhibition of the
glucose transporter SGLT2 with dapagliflozin in pancre-
atic alpha cells triggers glucagon secretion. Nat Med.
2015;21:512-517.
7. Song WJ, Mondal P, Wolfe A, et al. Glucagon regulates
hepatic kisspeptin to impair insulin secretion. Cell Metab.
2014;19:667-681.
8. Rosenstock J, Ferrannini E. Euglycemic diabetic keto-
acidosis: a predictable, detectable, and preventable
safety concern with SGLT2 inhibitors. Diabetes Care.
2015;38:1638-1642.
9. Taylor SI, Blau JE, Rother KI. SGLT2 inhibitors may
predispose to ketoacidosis. J Clin Endocrinol Metab.
2015;100:2849-2852.
10. Handelsman Y, Henry RR, Bloomgarden ZT, et al.
American Association of Clinical Endocrinologists and
American College of Endocrinology position statement on
the association of SGLT-2 inhibitors and diabetic ketoaci-
dosis. Endocr Pract. 2016;22:753-762.
11. Peters AL, Buschur EO, Buse JB, Cohan P, Diner JC,
Hirsch IB. Euglycemic diabetic ketoacidosis: a potential
complication of treatment with sodium-glucose cotrans-
porter 2 inhibition. Diabetes Care. 2015;38:1687-1693.
12. Erondu N, Desai M, Ways K, Meininger G. Diabetic
ketoacidosis and related events in the canagliflozin type 2
diabetes clinical program. Diabetes Care. 2015;38:1680-
1686.