Diabetic keto-acidosis secondary to sodium glucose co-transporter inhibitors: single-centre experience

Mikhail Nozdrin; Maria Selyanina; suhier elshowaya; Ghada Ahmed; Blazej Rzeszowski; Rafael Abreu; Simona Mihalikova; Gideon Mlawa

doi:10.1530/endoabs.90.P367

Background: SGLT-2 inhibitors (SGLT-2i) are used in treatment of type 2 diabetes mellitus (T2DM) through increasing urinary glucose secretion. Cases of euglycemic and hyperglycaemic diabetic ketoacidosis (DKA) have been reported in literature in patients using SGLT-2i. Increased glucose urinary excretion promoted by SGLT2i, decreases the production of insulin. In turn, lipolysis and ketone production increase predisposing the patient to developing DKA. It is hypothesised that SGLT2i, stimulate a cells of the pancreas, increasing the release of glucagon. The increased glucagon levels, further stimulate B-oxidation and ketone body production. In our case series we aim to describe the risk factors, demographics associated with development of SGLT-2i associated DKA and the course of disease in such patients.

Methods: All the T2DM patients with diabetic ketoacidosis on SGLT-2i who presented to the emergency department of a large hospital between Dec 26^th 2018 and Jan 1^st 2023 were included in the study.

Table 1 Comparison of CGM metrics before and after AID initiation, nZ185
	Before AID Initiation	After AID Initiation	P value
CGM wear time,%	81.25G16.9	88.44G12.71	<0.001
CGM data, days	76.39G18.8	81.11G16.47	0.002
Mean sensor glucose, mg/dl	167.57G29.23	150.32G18.36	<0.001
Glucose management indicator(GMI),%	7.3G0.7	6.90G0.44	<0.001
SD	59.40G13.17	49.16G10.12	<0.001
CV,%	35G5	33G4	<0.001
Time below range, <70 mg/dl(TBR),%	3.23G3.33	1.97G1.98	<0.001
Time between 54-70 mg/dl(TBR2),%	0.65G1.03	0.36G0.5	<0.001
Time below 54 mg/dl(TBR1),%	2.58G2.39	1.62G1.52	<0.001
Time in range, 70-180 mg/dl(TIR),%	58.95G17.29	73.56G11.66	<0.001
Time above range, >180 mg/dl(TAR),%	37.81G18.29	24.47G12.18	<0.001
Time between 181-250 mg/dl,(TAR1),%	24.99G8.92	18.87G7.75	<0.001
Time above 250 mg/dl(TAR2),%	12.83G11.26	5.60G5.15	<0.001
GRI	48.66G21.81	29.01G13.01	<0.001
GRI-Hypoglycemia component	2.72G2.85	1.65G1.68	<0.001
GRI-Hyperglycemia component	25.32G14.52	15.03G8.52	<0.001

Results: Ten patients were included in the study. Patient demographics are summarised in Table 1. All patients had confirmed T2DM, with negative GAD and anti-islet antibodies. Five patients were taking Canagliflozin, four-Empagliflozin, one-Dapagliflozin. Mean blood ketone level on admission was 4.18 mmol/l (2 St.Dev. ±0.274) and venous blood gas pH was 7.088 (2 St. Dev. ±0.155). Blood glucose on admission ranged from (4.7 to 28 mmol/l). Most common symptom at presentation was vomiting (n=5). Other common symptoms at presentation were abdominal pain (n=2), shortness of breath (n=2) and lethargy (n=2). Four patients had concurrent illness at presentation: urosepsis, recent knee surgery, community acquired pneumonia and acute coronary syndrome. All patients were treated with sliding scale, 1 patient was transferred to ITU, 1 required HDU support and 1 passed away from a cardiac arrest in ED. Mean duration of hospital stay was 5.3 days (2 St. Dev: 2.46 days).

Discussion: Diabetic ketoacidosis secondary to SGLT2i is a known side effect. In our cohort, we found 3 patients presenting with hyperglycaemia. DKA with SGLT2i is not uncommon, especially with intercurrent illness. Therefore patients on this therapy should be encouraged to monitor their blood ketone levels even if their blood glucose levels are within range.

Endocrine Abstracts

P367