Sodium-glucose cotransporter type 2 inhibitors (SGLT2i) are an attractive therapeutic option in diabetic patients due to a combination of cardiovascular and renal clinical benefits and an excellent safety profile. However, an increased risk of developing diabetic ketoacidosis has been reported in patients treated with these drugs, especially in those receiving concomitant insulin treatment.1 On the other hand, stress cardiomyopathy (SC) is a reversible acute heart disease that is triggered by physical or emotional stress. We describe a case of SC occurring in the context of euglycemic diabetic ketoacidosis (EDKA).
A 64-year-old woman presented to the emergency department with a 2-day history of nausea, vomiting and abdominal pain. Furthermore, in the last 24 h she reported a sharp increase in the rate of diuresis. She denied chest pain.
She had a history of type 2 diabetes with poor glycaemic control and was being treated with three oral antidiabetic drugs (metformin, dapagliflozin, sitagliptin) and insulin glargine.
On examination, she had tachypnoea and pain on abdominal palpation. Cardiopulmonary auscultation was unremarkable. An electrocardiogram revealed decreased voltages with pathological Q waves and a discrete ST-segment elevation in V1–V3, as well as generalised negative T waves. A blood test revealed an elevated ultrasensitive troponin-I (4836 ng/dl; reference value < 20 ng/dl), while the creatine kinase value was normal (162 mg/dl). Metabolic acidosis was also identified (pH 7.18; bicarbonate 14 mmol/l; PCO2: 30.1 mmHg; lactate 26 mmol/l). Blood glucose was slightly elevated (166 mg/dl). Urinalysis revealed the presence of severe ketonuria and glycosuria (++++). An abdominal CT scan showed unremarkable findings.
An echocardiogram was then requested, where a severely reduced left ventricular ejection fraction was identified at the expense of a mid-apical akinesia, with no associated valvular heart disease or other complications. An urgent coronary angiography was performed, in which the presence of obstructive coronary lesions was ruled out, while the ventriculography again showed apical ballooning.
Acidosis resolution was achieved with the infusion of saline and bicarbonate, and ketonuria became negative within 24 h.
The cardiac study was completed with an MRI (1.5 Tesla) performed on the eighth day after admission, where the absence of late gadolinium uptake at myocardial level was confirmed.
After 9 days of hospitalisation the patient was discharged. She was re-evaluated in an outpatient clinic one month later, with a normal electrocardiogram and echocardiogram, confirming the diagnosis of SC.
Different mechanisms have been suggested to be involved in the pathophysiology of SC: dysregulation of the central nervous system, increased plasma catecholamines, hyperproduction of stress hormones, etc.2 Although emotional triggers were first described as being associated with SC, current registries place physical triggers as the most common. Diabetic ketoacidosis has previously been described as a trigger for SC. The case described, however, developed in the context of EDKA, a rare form of diabetic ketoacidosis. Although formal definitions are lacking, EDKA is usually characterised by the concomitant presence of metabolic acidosis, ketonemia/ketonuria and normal - or slightly elevated - blood glucose.
SGLT-2i are considered the main factor associated with the development of EDKA in our setting. The glycosuria generated by these drugs leads to a decrease in blood glucose levels that stimulates the production of glucagon, which acts to promote lipolysis and the production of ketones in the liver. These mechanisms are in turn enhanced by an increase in preproglucagon production due to the direct action of SLGLT-2i on gene expression in pancreatic cellsα.3
The occurrence of EDKA is rare in patients treated with SGLT-2i monotherapy, with most cases reported with concomitant insulin therapy. Ketogenic states such as prolonged fasting, intense exercise or a critical clinical situation are often described as associated risk factors. It is therefore sensible to avoid treatment with SGLT-2i in these scenarios.
FundingThis research has not received any funding.
Conflict of interestsThe authors declare that they have no conflicts of interest related to this research work.