Abstract
INTRODUCTION: The risk of euglycemic DKA (EuDKA) is well-known with use of SGLT2i. We present two cases where EuDKA occurred despite discontinuing SGLT2i one week prior to bariatric surgery. CASE 1: A 43 year old male with type II diabetes mellitus and obesity class III presented with nausea and vomiting two weeks following gastric sleeve surgery. Preoperatively, the patient was taking glargine, lispro, empagliflozin and metformin. Empagliflozin was discontinued seven days before surgery. After surgery, patient was discharged on empagliflozin only. Labs showed blood glucose 266 mg/dl, anion gap (AG) 32 mmol/L, HCO3 11 mmol/L and betahydroxybutyrate > 6 mmol/L. He was admitted to the ICU and managed according to the guidelines for DKA. On discharge, his empagliflozin was discontinued. CASE 2: 31 year old female with type II diabetes, obesity class III developed nausea and vomiting a day after Roux-en-Y gastric bypass surgery. Preoperatively patient was taking glargine, metformin, dulaglutide and empagliflozin. Patient discontinued empagliflozin seven days before surgery. Labs showed glucose 232 mg/dl, AG 23 mmol/L, HCO3 6 mmol/L and betahydroxybutyrate 6.96 mmol/L. She was admitted to the ICU and managed according to the guidelines for DKA. Empagliflozin was discontinued on discharge. DISCUSSION: In case 1, EuDKA occurred two weeks postoperatively. In case 2, EuDKA occurred one day postoperatively. In both cases the SGLT2i was discontinued seven days before the surgery. The use of SGLT2i is a cause of EuDKA. The pathophysiology of EuDKA is not clear but thought to be related to a relative carbohydrate deficit, increased glucosuria and low insulin levels. Additionally, SGLT2i may directly stimulate glucagon release leading to increased lipolysis. There have been number of case reports suggestive of EuDKA after gastric bypass surgery in association with use of SGLT2i. Bariatric surgery causes changes in incretin pathways which in turn changes insulin: glucagon ratio and responses to blood glucose in immediate post-operative period. Additionally, preoperatively these patients are on water or ketogenic diet causing ketogenic state. However, in our cases the SGLT2i were discontinued seven days before surgery. It is not known if the use of SGLT2i therapy in the preoperative period might pose an increased risk of EuDKA due to changes in incretin based pathways associated with weight loss surgery and SGLT2i. CONCLUSION: It is important to recognize that use of SGLT2i in setting of bariatric surgery perioperatively increases the risk of EuDKA. Our cases suggest two things. One it may be necessary to discontinue SGLT2i more than seven days before bariatric surgery and second initiating SGLT2i for patients with history of bariatric surgery should be considered an additional risk factor for EuDKA due to permanent changes in incretin based pathways and other mechanisms. Presentation: No date and time listed