CHANGING THE DKA PROTOCOL: IATROGENIC REFRACTORY HYPOGLYCEMIA

SESSION TITLE: Critical Care Case Report Posters 60 SESSION TYPE: Case Report Posters PRESENTED ON: 10/11/2023 12:00 pm - 12:45 pm INTRODUCTION: Diabetic ketoacidosis (DKA) is a common complication and, occasionally, initial presentation of uncontrolled diabetes mellitus (DM). It is treated with intravenous (IV) insulin, IV fluids and electrolyte replacement, utilizing organization specific protocols. We present a case where a DKA patient developed refractory hypoglycemia due to shortcomings in the DKA protocol. CASE PRESENTATION: 33-year-old male presented with altered mentation, abdominal pain and dysarthria for twelve hours. He was tachypneic, hypotensive, and tachycardic. Labs showed a venous glucose level of >1500mg/dL, bicarbonate 10mmol/L, Anion Gap 35mmol/L, Creatinine 4.57mg/dL (baseline 0.70mg/dL), beta hydroxybutyrate 106.3mg/d, HgA1c >18.5%. His ABG was 7.09/41/44/13. Corrected sodium was 182mmol/L. Urine analysis showed 3+ glucose and ketones. He was diagnosed to be in diabetic ketoacidosis, admitted to the Medical Intensive Care Unit and started on appropriate treatment with our DKA protocol including nurse driven insulin titration based on hourly glucose checks. At hour 1, his POC glucose was >600 mg/dl, and per protocol his insulin rate was doubled. This treatment continued hourly until the insulin pump maxed out at 1000 units/hr. Eventually, this treatment resulted in severe refractory hypoglycemia requiring dextrose infusion, glucagon and steroids. It took 48 hours for his glucose level to stabilize. This incident led to a revision of our DKA protocol. DISCUSSION: The DKA protocol includes an insulin drip with scheduled basic metabolic profiles, electrolyte repletion, IV fluids and hourly Point of Care (POC) glucose checks. Recently, it was changed to a nurse driven protocol based on studies indicating that nurse-driven protocols are non-inferior to physician-driven protocols. It includes hourly insulin rate adjustments based on POC glucose without considering that the POC glucometer cannot detect glucose levels greater than 600mg/dL. Hence, even if the glucose levels are decreasing appropriately (i.e., 100mg/dL/h), the change will not be observed till the next BMP, which are usually scheduled at four-hour intervals, leading to inappropriate insulin dosing. This incident was identified as a systems failure. Apart from the deficiency in the protocol itself mentioned previously, the nursing staff had not been educated to identify a dangerously high rate of insulin and when to ask for physician guidance. The system flaw also extended to the pharmacy, from where multiple insulin bags were being dispensed without any secondary checks. In addition, the patient had acute kidney injury (AKI) that impeded insulin clearance, this was also not factored in. This eventually also made it harder to treat hypoglycemia. Following this incident, the DKA protocol at our institute was revised and now includes checking a venous glucose level after doubling the insulin dose once and specific physician's orders after the 30 units/hr mark. CONCLUSIONS: We emphasize using venous glucose levels for insulin dosing in critically ill patients with glucose levels exceeding maximum value of glucometer on initial presentation. This case highlights a systems failure that could also occur in other protocol-based treatments and we urge frequent auditing with a multidisciplinary team of all such protocols. We also propose that our changes be integrated into the standard DKA management at other sites to prevent similar incidents. REFERENCE #1: Abbas E. Kitabchi, Guillermo E. Umpierrez, John M. Miles, Joseph N. Fisher; Hyperglycemic Crises in Adult Patients With Diabetes. Diabetes Care 1 July 2009; 32 (7): 1335–1343. https://doi.org/10.2337/dc09-9032 REFERENCE #2: Gosmanov AR, Gosmanova EO, Dillard-Cannon E. Management of adult diabetic ketoacidosis. Diabetes Metab Syndr Obes. 2014 Jun 30;7:255-64. doi: 10.2147/DMSO.S50516. PMID: 25061324; PMCID: PMC4085289. REFERENCE #3: Anis TR, Boudreau M, Thornton T. Comparing the Efficacy of a Nurse-Driven and a Physician-Driven Diabetic Ketoacidosis (DKA) Treatment Protocol. Clin Pharmacol. 2021 Oct 7;13:197-202. doi: 10.2147/CPAA.S334119. PMID: 34675693; PMCID: PMC8504870. DISCLOSURES: No relevant relationships Added 04/13/2023 by Luis Arzeno Tejada, source=Web Response, value=Consulting fee Removed 04/13/2023 by Luis Arzeno Tejada, source=Web Response No relevant relationships by Rehan Saeed No relevant relationships by Sarah Schein No relevant relationships by Mishal Shaukat No relevant relationships by Peter Ucciferro