SGLT inhibitors and euglycaemic diabetic ketoacidosis: earlier observations rediscovered

Salem A Beshyah,1,2 Brian M Frier3

1 Department of Medicine, Dubai Medical College, Dubai, United Arab Emirates
2 Honorary Professor of Diabetes, The Queen’s Medical Research Institute, University of Edinburgh, Edinburgh, UK

Address for correspondence: Dr Salem A Beshyah
Adjunct Professor of Clinical Research and Metabolic Medicine, PO Box 59472, Abu Dhabi, United Arab Emirates
Tel: 00971 50 5662723

Br J Diabetes 2019;19:37-41


Diabetic ketoacidosis (DKA) is a life-threatening metabolic derangement, defined by the presence of severe hyperglycaemia, ketosis and metabolic acidosis. Recently, DKA was redefined to be present when the latter two abnormalities occur without severe hyperglycaemia. Munro and colleagues in Edinburgh described ‘euglycaemic diabetic ketoacidosis’ in 1973 and suggested this new terminology. The same year the critical feature of this subclass of DKA – that is, an increased urinary loss of glucose – was identified by Ireland and Thomson in Glasgow. In the subsequent 40 years (from 1974 to 2014), clinical interest in this condition was limited to a few case reports. The emergence of a new class of antidiabetes medications, the sodium-glucose cotransporter (SGLT) inhibitors, which promote profuse glycosuria, has reawakened interest in euglycaemic DKA, as this is a side effect of these drugs. The earlier perceptive observations of these physicians in Scotland deserve to be recognised for their contribution in identifying and describing euglycaemic DKA and correctly identifying its underlying pathogenesis. Recent international consensus has provided guidance to physicians to aid timely recognition of the condition by testing for ketosis in the appropriate clinical context and to manage it effectively by discontinuing the SGLT inhibitor and provision of insulin, carbohydrate and hydration (the STICH protocol). This may be particularly relevant in view of the recent licensing developments for use of certain members of the SGLT inhibitors in type 1 diabetes.

Key words: DKA, euglycaemic diabetic ketoacidosis, SGLT inhibitor, type 1 diabetes, urinary glucose


Diabetic ketoacidosis (DKA) is a life-threatening metabolic derangement of diabetes. It has long been characterised by the triad of hyperglycaemia, metabolic acidosis and ketosis. More recently, DKA has been redefined in the guidelines of the Joint British Diabetes Societies (JBDS) as follows: (i) ketonaemia 3 mmol/L and above or significant ketonuria (more than 2+ on standard urine sticks); (ii) blood glucose over 11 mmol/L or known diabetes mellitus; (iii) venous bicarbonate (HCO3) below 15 mmol/L and/or venous pH less than 7.3.1 Thus, this metabolic derangement can occur without marked elevation of blood glucose. However, this observation is not novel. The development of ‘euglycaemic diabetic ketoacidosis’ (euDKA) was described by a group of diabetologists in Scotland in the early 1970s, along with a putative interpretation of its likely pathogenesis. These observations have often been overlooked by authors of subsequent papers on DKA. Citations of the original report (including self-citation) barely exceeded 100 in total over the 45 years since it was published. In this brief commentary, credit is paid to the perception of the physicians who first described and named this metabolic derangement and others who helped to elucidate the underlying mechanism, demonstrating how it has been ‘rediscovered’ as a side effect of a new class of antidiabetes drugs.

‘Euglycaemic diabetic ketoacidosis’
In 1973, Dr John Munro (Figure 1, left) and his colleagues, which included Professor Ian Campbell, a former co-editor of this journal, while working in the Diabetic and Dietetic Department of the Royal Infirmary of Edinburgh published a case series of 211 episodes of diabetic metabolic decompensation.2 Thirty-seven of these patients had severe ketoacidosis with a plasma bicarbonate of 10 mmol/L or less but with a blood glucose level of less than 16.6 mmol/L. All of the patients were young and had type 1 diabetes mellitus, only one of whom had not been diagnosed previously. Vomiting was a common feature, and in all of these cases carbohydrate intake had either been reduced or stopped while administration of insulin had been maintained or increased. Treatment comprised typical parenteral fluid and electrolyte replacement and large doses of intravenous insulin and carbohydrate in the form of 10% dextrose. Alkali was either withheld or administered sparingly, and serial estimations were made of plasma bicarbonate to monitor replacement therapy. The authors coined the term ‘euglycaemic diabetic ketoacidosis’ (euDKA) to describe this metabolic derangement; this nomenclature had not been used previously in publications on diabetic ketoacidosis (Figure 2, upper).

403 Beshyah Figure 1

403 Beshyah Figure 2

The concept accepted but the interpretation debated
Dr John Ireland (Figure 1, right), a consultant diabetologist, and Dr William Thomson, a clinical biochemist in the Southern General Hospital in Glasgow immediately responded to the paper by Munro and colleagues in a letter to the British Medical Journal,3 and commended the authors for identifying a subgroup of young patients with insulin-dependent diabetes who had developed ketoacidosis without significant hyperglycaemia (Figure 2, lower). While acknowledging that this observation has extended the range of identifiable metabolic derangements associated with insulin-treated diabetes, they commented that, with the exception of vomiting, the clinical features were difficult to explain. They added that in their clinical experience such patients characteristically were young, had good renal function and, in some individuals, a massive urinary excretion of glucose had been observed. These patients appeared to have a lower renal threshold for the excretion of glucose. Since ketoacidosis is generally regarded to be the metabolic consequence of excessive and uncontrolled gluconeogenesis, coupled with increased plasma concentrations of free fatty acids, they observed that it was difficult to postulate that the relative euglycaemia was a consequence of a diminished production of glucose. As an alternative, they proposed that “a greater urinary loss of glucose was a more likely explanation”. It was not clear whether this could be a consequence of increased growth hormone secretion affecting renal function, related to the enhanced urinary clearance of glucose that is recognised to be present in early type 1 diabetes, or simply represented one end of the spectrum of the renal threshold for excretion of glucose.3

Reawakening of interest in euglycaemic DKA with the introduction of SGLT2 inhibitors
Subsequent discussion of euDKA and its pathogenesis was sparse. For instance, 88 publications were retrieved from the Scopus database in the last four years (2015–2018) compared with only 30 papers in the previous 41 years (1973–2014). This paucity of publications on euDKA for a protracted period after its original description is depicted in Figure 3. From 1974 to 2014, only a few isolated clinical cases were reported. These were associated with the effects of starvation, excessive alcohol consumption, chronic liver disease and glycogen storage disorders, pregnancy and substance abuse with cocaine. However, the introduction of a new class of antidiabetes medications, the sodium glucose cotransporter (SGLT) inhibitors,4 has reawakened interest in euDKA and shed further light on the underlying pathogenetic mechanisms.5,6 The upsurge in interest can be measured by the increasing number of publications from 2015 onwards.

403 Beshyah Figure 3

The association between the use of SGLT inhibitors and serious adverse events was highlighted in a Scandinavian register-based cohort study performed between July 2013 and December 2016.6 The primary outcomes identified from hospital records were lower limb amputation, bone fracture, diabetic ketoacidosis, acute kidney injury, serious urinary tract infection, venous thromboembolism and acute pancreatitis. In that analysis, the use of SGLT-2 inhibitors, as compared with glucagon-like peptide-1 receptor agonists, was associated with an increased risk of lower limb amputation and DKA, but not with other serious adverse events of current concern. A worrying side effect that has been reported with SGLT-2 inhibitors in patients with type 2 diabetes, and particularly during off-label use in patients with type 1 diabetes, is euDKA. This poses a challenge to physicians as ketoacidosis may be overlooked in sick patients presenting with normoglycaemia, leading to a delay in appropriate management.7 Although euglycaemia is relatively rare in association with ketoacidosis in people with diabetes, it remains a serious life-threatening medical emergency and must be treated as quickly and effectively as other forms of hyperglycaemic metabolic emergencies.8 Consequently, practising physicians should consider the presence of DKA in patients with ketonuria and biochemical evidence of acidosis even when blood glucose levels are not elevated. A high clinical suspicion is required to diagnose euDKA. Blood pH and blood or urine ketones should be measured in all ill patients with diabetes, irrespective of the presenting blood glucose level.8

SGLT inhibitor-associated DKA is likely in patients with insulin-deficient diabetes, including those with type 2 diabetes, and is typically precipitated either by the omission of insulin or overzealous reduction in the dose, or with acute illness, dehydration, extensive exercise, surgery, low-carbohydrate diets or excessive alcohol intake.6

SGLT inhibitor-induced euDKA may occur because of the effects on glucose clearance and volume depletion. Recent data from studies in rodents suggest that a combination of insulinopenia and dehydration are fundamental to the development of euDKA and may be potential therapeutic targets to prevent euDKA occurring in association with SGLT inhibitors.9 Patients with types 1 and 2 diabetes who experience nausea, vomiting or malaise while on SGLT inhibitor therapy should have urinary and blood ketones measured as a matter of urgency.7,8,10 If urinary ketones are ++ or plasma ketones >0.6 mmol/L, investigation should proceed to measuring venous or arterial blood gases. A recent international consensus statement provided a good working plan for the risk management of DKA in patients with type 1 diabetes who are being treated with SGLT inhibitors.11,12 Prompt verification of the diagnosis, invoking of the ‘STICH’ (STop the SGLT inhibitor, Inject bolus insulin, Consume 30 g carbohydrates and Hydrate) protocol, close monitoring and escalation of the care for non-responders should minimise the risk in this context (Table 1).

403 Beshyah Table 1

The licensing status of use of the various SGLT inhibitors in people with type 1 diabetes is controversial (Table 2). The European Commission has approved dapagliflozin for use in type 1 diabetes as an adjunct to insulin in people affected with a BMI of at least 27 kg/m2, making it the first approval of an oral therapy for type 1 diabetes in Europe.13 The favourable decision was presumably based on data derived from the DEPICT clinical programme in type 1 diabetes.14 SGLT-2 inhibitors are currently under regulatory review in the USA and Japan for use as an adjunctive treatment in adults with type 1 diabetes. However, in March 2019 the FDA in the USA rejected sotagliflozin, an oral dual inhibitor of SGLT-1 and SGLT-2. By contrast, sotagliflozin secured a positive opinion from the European Medicines Agency for is ability to lower HbA1c and reduce body weight, despite having an associated DKA rate of 3–4%.13 These licensing changes, while differing on each side of the Atlantic, may potentially place more patients at risk of developing euDKA. The international consensus recommendations are therefore timely and are particularly relevant in view of the recent licensing developments.12

403 Beshyah Table 2


In this short commentary we have highlighted the early Scottish observations on ‘euglycaemic DKA’ that have helped to identify this condition and to elucidate the potential pathogenetic mechanisms and the risk factors. Postulation of a pivotal role for the increased diuresis as a mechanism of euDKA was advanced several decades before the SGLT inhibitors came into clinical use. The recent development of this drug class for clinical application has provided an opportunity for the pathophysiology of DKA to be revisited and the pioneering work of these physicians to be more widely acknowledged. Knowledge of the risk factors, diagnostic criteria and management of euDKA is essential for clinicians who are required to manage such patients when they present as unselected medical emergencies. The ‘STICH’ protocol provides a very helpful working plan for clinicians. This is particularly important in view of the recent licensing developments that may potentially put a greater number of patients at risk of developing euDKA.

403 Beshyah Key Messages

Conflict of interest: None

Funding: Contributions

Contributions: SAB conceived and wrote the manuscript and BMF revised it extensively. Both authors approved the final version of the manuscript.


1.  Savage MW, Dhatariya KK, Kilvert A, et al for the Joint British Diabetes Societies. Joint British Diabetes Societies guideline for the management of diabetic ketoacidosis. Diabet Med 2011;28(5):508–15.

2.  Munro JF, Campbell IW, MacCuish AC, et al. Euglycaemic diabetic ketoacidosis. Br Med J 1973;2(5866):578–80.

3.  Ireland JT, Thomson WST. Euglycaemic diabetic ketoacidosis. Br Med J 1973;3(5871):107 (letter).

4.  Davies MJ, D'Alessio DA, Fradkin J, et al. Management of hyperglycaemia in type 2 diabetes, 2018. A consensus report by the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD). Diabetologia 2018;61(12):2461–98.

5.  Zelniker TA, Wiviott SD, Raz I, et al. SGLT2 inhibitors for primary and secondary prevention of cardiovascular and renal outcomes in type 2 diabetes: a systematic review and meta-analysis of cardiovascular outcome trials. Lancet 2019;393(10166):31–9.

6.  Ueda P, Svanström H, Melbye, et al. Sodium glucose cotransporter 2 inhibitors and risk of serious adverse events: nationwide register based cohort study. BMJ 2018;363:k4365.

7.  Modi A, Agrawal A, Morgan F. Euglycemic diabetic ketoacidosis: a review. Curr Diabetes Rev 2017;13(3):315–21.

8.  Rawla P, Vellipuram AR, Bandaru SS, et al. Euglycemic diabetic ketoacidosis: a diagnostic and therapeutic dilemma. Endocrinol Diabetes Metab Case Rep 2017;2017. pii: 17-0081.

9.  Perry RJ, Rabin-Court A, Song JD, et al. Dehydration and insulinopenia are necessary and sufficient for euglycemic ketoacidosis in SGLT2 inhibitor-treated rats. Nat Commun 2019;10(1):548.

10. Peters AL, Buschur EO, Buse JB, et al. Euglycemic diabetic ketoacidosis: a potential complication of treatment with sodium-glucose cotransporter 2 inhibition. Diabetes Care 2015;38(9):1687–93.

11. Garg SK, Peters AL, Buse JB, Danne T. Strategy for mitigating DKA risk in patients with type 1 diabetes on adjunctive treatment with SGLT inhibitors: a STICH protocol. Diabetes Technol Ther 2018;20(9):571–5.

12. Danne T, Garg S, Peters AL, Buse JB, et al. International consensus on risk management of diabetic ketoacidosis in patients with type 1 diabetes treated with sodium-glucose cotransporter (SGLT) inhibitors. Diabetes Care 2019 Feb 6. pii: dc182316.

13. European Medicines Agency. Committee for Medicinal Products for Human Use (CHMP). Summary of opinion (initial authorisation): Zynquista (sotagliflozin). Available at:  (last accessed 1 May 2019).

14. Dandona P, Mathieu C, Phillip M, et al; DEPICT-1 Investigators. Efficacy and safety of dapagliflozin in patients with inadequately controlled type 1 diabetes: the DEPICT-1 52-week study. Diabetes Care 2018; 41(12):2552–9.


  • There are currently no refbacks.

The Journal of the Association of British Clinical Diabetologists