EUGLYCEMIC DIABETIC KETOACIDOSIS MASQUERADING AS TOXIC ALCOHOL INGESTION

Abstract

TOPIC: Critical Care TYPE: Medical Student/Resident Case Reports INTRODUCTION: First reported in 1973, euglycemic diabetic ketoacidosis (euDKA) historically accounted for 0.8-1.1% of diabetic ketoacidosis (DKA) cases [1]. Without hyperglycemia, the anion gap metabolic acidosis (AGMA) of euDKA may be attributed to other causes. We present a case of euDKA that was mistaken for toxic alcohol poisoning. CASE PRESENTATION: A 53-year-old woman with type 2 diabetes (T2D), for which she reported taking metformin and glyburide, presented with abdominal pain and nausea. She also reported drinking "vodka" given by a stranger the day prior. Labs included bicarbonate 12 mEq/L, anion gap (AG) 18, lactate 1.7 mmol/L, lipase 735 U/L, and urinalysis with 2+ glucose and 3+ ketones. CT confirmed pancreatitis. Correctional insulin was ordered but not given due to blood sugar of 121-225 mg/dL. Repeat labs overnight revealed bicarbonate <5, AG >18, and arterial pH 7.09. After toxicology consult, she was given IV fomepizole and transferred to the ICU for urgent hemodialysis (HD) for possible toxic alcohol poisoning. Just prior to placing HD access, records obtained revealed that she was prescribed empagliflozin. Ultimately, a toxic alcohol panel was negative and beta-hydroxybutyrate was elevated at 0.86 mmol/L. She was treated with infusions of crystalloid, insulin, and dextrose. Her ketosis resolved in 9 days, and she was discharged on insulin and metformin. DISCUSSION: Previously seen with low caloric intake, alcohol or cocaine use, liver disease, or pregnancy, euDKA has recently been linked to sodium glucose cotransporter 2 inhibitor (SGLT2I) use [2]. Since FDA approval in 2013, SGLT2I use has steadily risen as data on cardiovascular and renal benefits accumulate. SGLT2Is act independent of insulin by blocking proximal tubule reabsorption of glucose. This can lead to euglycemia despite insulin deficiency and the potential for an acute insult, like pancreatitis, to trigger euDKA. Along with lack of insulin, SGLT2I-related euDKA may be mediated by volume depletion, counterregulatory hormones, and enhanced lipolysis and fatty acid oxidation [3]. Notably, pancreatitis can trigger euDKA without SGLT2I use [2]. Though the initial ketonuria and lack of inebriation were, in retrospect, clues to the presence of DKA in this case, the report of a suspicious ingestion led to a presumptive diagnosis of toxic alcohol poisoning. Indeed, untreated toxic alcohol poisoning can lead to major disability or death, and prompt empiric treatment is vital if clinical suspicion exists. CONCLUSIONS: The rapid diagnosis of AGMA remains a challenge in intensive care. As SGLT2I use continues to grow, providers will increasingly encounter euDKA. Importantly, euglycemia does not exclude DKA, and euDKA must be routinely considered in the differential of unexplained severe AGMA to prevent adverse outcomes from delays in treatment. REFERENCE #1: Jenkins et al. Euglycaemic diabetic ketoacidosis: does it exist? Acta Diabetol 1993;30:251-3. REFERENCE #2: Bonora et al. Euglycemic Ketoacidosis. Review Curr Diab Rep 2020 May 19;20:25. REFERENCE #3: Qiu et al. Ketosis and diabetic ketoacidosis in response to SGLT2 inhibitors: Basic mechanisms and therapeutic perspectives. Diabetes Metab Res Rev 2017 Jul;33(5). DISCLOSURES: No relevant relationships by Christos Argyropoulos, source=Web Response No relevant relationships by Matt Bouchonville, source=Web Response No relevant relationships by Mitchell Byrd, source=Web Response No relevant relationships by Melissa Fang, source=Web Response No relevant relationships by Umair Khan, source=Web Response No relevant relationships by Joao Teixeira, source=Web Response