Sodium-glucose cotransporter 2 inhibitor associated diabetic ketoacidosis in chronic heart failure: a case series

Abstract

Abstract Funding Acknowledgements Type of funding sources: None. Background The sodium-glucose cotransporter 2 inhibitor (SGLT2i) medications have proven mortality and morbidity benefit for patients with type 2 diabetes mellitus (T2DM) and chronic heart failure (HF) over the previous decade leading to a rapid expansion of societal guideline recommendations and US FDA indications. With this rapid expansion of SGLT2i indications and prescribing, it is essential that we understand the safety and adverse effects of this drug class. Diabetic ketoacidosis (DKA) is a life-threatening adverse effect of the SGLT2i that has been described in post-marketing reports, prompting an FDA warning. The clinical characteristics of patients with T2DM and HF who develop SGLT2i associated DKA are lacking. Purpose To describe the clinical characteristics of patients with T2DM and chronic HF prescribed SGLT2i who develop DKA. Methods Data was collected retrospectively with Penn State Health TriNetX Local, a web-based tool for healthcare system research population queries, between January 1, 2013-December 31, 2020. Patients with T2DM and chronic HF who were prescribed empagliflozin, canagliflozin or dapagliflozin and subsequently developed DKA were eligible for inclusion. Patients with type 1 diabetes mellitus were excluded. Patient health records were reviewed to confirm SGLT2i use at the time of DKA hospitalization. Baseline characteristics, clinical features and outcome data of hospitalized patients were collected. Results Four unique patient records were identified. Mean age was 59.8 years, BMI 30.5 kg/m2, HbA1c 10.6%. Two patients had baseline heart failure mid-range ejection fraction and two patients had heart failure preserved ejection fraction. All patients were baseline NYHA class II, ACC/AHA stage C. Three patients were on a baseline diuretic, two were on ACE/ARB, two were on beta blocker, three were on metformin, two were on pioglitazone and none were on insulin. Two patients were on baseline empagliflozin 10 mg daily and two were on canagliflozin 100 mg daily. Two patients were on SGLT2i for less than 1 month prior to presentation. Three patients presented with SGLT2i associated DKA and one with euglycemic DKA. None of the patients presented with decompensated congestive HF. Three patients presented with concomitant acute medical conditions; Fournier’s gangrene, asthma exacerbation and alcoholic hepatitis. One patient developed acute stress induced cardiomyopathy while hospitalized. Another patient suffered cardiac arrest secondary to acidosis. Average duration of insulin infusion was 60 hours before resolution of DKA. All of the patients recovered. Conclusions In this single center study, we presented four unique cases of SGLT2i associated DKA in patients with T2DM and chronic HF. Diagnosis can be challenging as concomitant acute medical conditions and/or euglycemia can mask underlying DKA. Early diagnosis and management of SGLT2i associated DKA is important as this condition is life-threatening.