Safety and Statistical Analysis in Emergency Medicine Research

Date: August 30, 2023 Reference: Griffey et al. The SQuID protocol (subcutaneous insulin in diabetic ketoacidosis): Impacts on ED operational metrics. AEM August 2023 Dr. Suchismita Datta Guest Skeptic: Dr. Suchismita Datta. She is an Assistant Professor and Director of Research in the Department of Emergency Medicine at the NYU Grossman Long Island Hospital Campus. This is the last show for Season#11. It has been a great year with the addition of PedEM SuperHero Dr. Dennis Ren. We have some exciting news to cap off the end of this amazing year. Suchi will be joining the SGEM faculty as part of the Hot Off the Press team. Case: A 28-year-old male with a history of type-1 diabetes mellitus presents to the emergency department (ED) with increase in thirst and light headedness. He is otherwise healthy. Blood glucose in triage is 489 mg/dl (27.2 mmol/L). Venous blood gas (VBG) shows an acidosis with a pH of 7.21. Electrolytes show a gap of 21. The patient’s symptoms begin to improve after initial intravenous (IV) fluid administration of one litre of 0.9% saline. The patient states he has had multiple “diabetic emergencies” in the past and usually ends up in the intensive care unit (ICU) on a drip. He is wondering, “Hey doc, do I have to go back to the ICU strapped to an IV pole?” The flow nurse has similar questions for you and wants to know if she should clear out a bed in the critical care bay so that the patient can have appropriate nursing requirements for an insulin infusion. Your resident is eager to go ahead and sign off on the diabetic ketoacidosis (DKA) insulin order set and the ICU attending’s “Spidey senses” are going off. They are on the phone asking you if you already have another admission for them on this busy day. However, the ICU is full and the patient will likely be boarding in your ED for a bit before coming upstairs. Just as all this is happening, you notice how the waiting room is filling up and you can hear the sirens of approaching ambulances becoming louder. You take a deep breath, and you think to yourself…let the squid games begin. Background: DKA is a common yet potentially fatal condition seen in patients with type 1 diabetes. It accounted for roughly 8.9 ED visits /1000 adults with diabetes [1]. DKA results in over 500,000 annual hospital days with estimated annual hospital costs of over $5 billion [2]. Dr. Nathan Kuppermann Despite how common and expensive the management of DKA can be, we have only looked at it once on the SGEM. That was an episode covering the practice changing randomized control trial published in NEJM by Dr. Nathan Kuppermann from the PECARN Team for pediatric DKA [3]. They reported that the type of intravenous fluids (0.45% NaCl or 0.9% NaCl) or speed of infusion did not appear to make a clinically important difference (SGEM#255). Because of the complexity of care around managing DKA, the typical approach is an insulin drip with ICU level of care for all degrees of severity. Increased resource utilization around this can prolong ED length of stay, especially in the context of a busy hospital or a global pandemic. However, over the past 20 years, there is burgeoning evidence that fast-acting subcutaneous insulin analogs could be a potential treatment option for mild to moderate severity DKA including a 2016 Cochrane SRMA [4]. If proven to be a safe and effective management strategy, this would eliminate the need for an insulin drip and opens new options for management and disposition of DKA patients from the ED. Using fast-acting subcutaneous insulin could streamline care in the ED and decrease the length of stay (LOS) in the department. This reduction in LOS is desirable for many reasons including overcrowding, prolonged wait times, and the availability of ICU beds for other critical patients. Clinical Question: Can a patient with mild to moderate severity DKA be safely managed with subcutaneous fast acting insulin analogs on a non-ICU floor wit...