Carolinas Research Day 2021

Evaluation of the Diabetic Ketoacidosis Management in the Critically Ill Children and Adolescents Mackenzie Dreher, OMS-IV, MPH; Brittany Skaggs, OMS-IV; Christine Angeles, OMS-IV; Alyssa Davolio, OMS- IV; Jacob Andrews, OMS-IV; Ning Cheng, MS, PhD; Hanna S. Sahhar, MD, FAAP, FACOP Edward Via College of Osteopathic Medicine (VCOM)-Carolinas Campus & Spartanburg Regional Healthcare System (SRHS). Spartanburg, South Carolina .

CLIN-2

Table 1. Three bag protocol used at Spartanburg Regional Healthcare System to correct DKA

Table 2 . Demographics of study participants .

Conclusion

Abstract Study Aim: The purpose of this study is to re-evaluate diabetic ketoacidosis (DKA) management protocol for type 1 diabetes mellitus (T1DM) and type 2 diabetes mellitus (T2DM) patients in the pediatric intensive care unit (PICU) by evaluating the amount of fluids and insulin needed for resolution of metabolic acidosis and dehydration. Research Methods: A retrospective chart review of 0–17-year-old patients admitted to the PICU at Spartanburg Regional Health System (SRHS) for DKA. Results : A study population of 247 patients, predominantly female, white, ages 13-17 years with established T1DM had an average length of stay (LOS) in the PICU of 1.5 days. Mean values revealed about 6 hours spent in metabolic acidosis with an average of 114 mL/kg fluid and 2.6 units/kg of insulin to correct for dehydration. Complications of DKA incidents were cerebral edema at 4%, arrhythmias at 2% and hypoglycemia at 1.6%. There was a significance (p=0.034) of cerebral edema to the administration of sodium bicarbonate in the emergency department. Lastly, significance (p<0.001) was noted between the decrease in HR from baseline to correction of acidosis and baseline HR to resolution of dehydration. Conclusion: The DKA protocol used at SRHS showed improvement in PICU length of stay and amount of time spent in metabolic acidosis; with emphasis to continue fluid management until the resolution of dehydration. The proper resolution of DKA minimizes the complication of cerebral edema. The results expressed the relationship of cerebral edema to the administration of sodium bicarbonate in the emergency department. A retrospective chart review evaluating 3 bag protocol DKA management in the pediatric population admitted to the PICU at SRHS in Spartanburg, SC (Table 1). Electronic medical records were reviewed using the Epic EHR to gather data during two different time periods of patients admitted to the PICU: December 1, 2005 – July 15, 2014 and October 1, 2016 – April 1, 2020. All identifiers were omitted from data collection. Inclusion Criteria (1) Admitted with diagnosis of DKA (2) Between the ages of 0 – 17 years old Primary Outcomes (1) Time for resolution of metabolic acidosis from the initiation of insulin drip (2) Amount of fluids and insulin per kilogram of body weight used to correct the dehydration Secondary Outcomes (1) Length of stay (LOS) in the PICU (2) Changes in heartrate from baseline to resolution of metabolic acidosis and resolution of dehydration (3) Correlation of complications due to DKA and initial management in the emergency department Data was analyzed using descriptive statistical analysis, simple linear regression, and Fisher’s exact test to assess for significance (p ≤ 0.05) using SPSS Data Analysis software. Methods DKA is defined by the following diagnostic criteria: (1) Plasma glucose levels > 200 mg/dL (2) Metabolic acidosis with a pH ≤ 7.30 or a plasma bicarbonate level ≤ 15 mEq/L (3) Ketonemia/Ketonuria

Spartanburg Regional Healthcare System PICU utilizes a three-bag system for management of DKA in critically ill children and adolescents. The longer time a patient spends in metabolic acidosis and dehydration further increases their risk of complications and mortality. In addition, cerebral edema is the worst complication to become a patient in DKA, next to mortality. This study population was similar to other studies, analyzing children admitted for DKA with an increased incidence of cerebral edema. We further support the findings that a bolus of sodium bicarbonate correlates with the development of cerebral edema. However, no significance was found between administration of an insulin bolus, two or more fluid boluses at 20 mL/kg, or a fluid bolus >20 mL/kg with other DKA complications, such as arrythmias or hypoglycemia. The three-bag system brought about a quicker resolution of metabolic acidosis compared to other studies with similar insulin drip rates.

Results

The study population consisted of 247 patients; further demographics were analyzed in Table 2. Primary Outcomes The average time a patient spent in metabolic acidosis (HCO3 >14.99 mEq/L) was 5.98±9.58 hours. Resolution of dehydration (HCO3 >17.99 mEq/L) was analyzed by the mean amount of fluids and insulin per kilogram of body weight, 114.38±148.72 mL/kg and 2.60±5.51 units/kg, respectively; with an average drip rate at 0.098±0.174 units/kg/hour. (Figure 1)

Using simple linear regression analyses, with every one-unit of bodyweight increase, there would be a 44.78 increase in total IV fluids used to correct for dehydration (p=0.008). Also, for every one-unit increase in body weight, there would be a 0.001 decrease in insulin drip rate (p=0.026). Secondary Outcomes Total LOS in the PICU was an average of 1.54±1.06 days. Changes in heartrate from baseline to resolution of metabolic acidosis had an average decrease of 35.23±44.26 beats per minute (bpm) and baseline to resolution of dehydration averaged a 49.64±38.62 bpm decrease (Figure 2). In this study, 4% developed complications of cerebral edema, 2% arrythmias, and 1.6% hypoglycemia (Figure 3). Using Fisher’s exact test, a significant correlation was noted between the complication of brain edema and administration of sodium bicarbonate (p=0.034). Figure 1 . Average time spent in metabolic acidosis and the average fluids and insulin used to correct per kilogram of body weight. .

References

1. Gosmanov, Adair, MD, PhD, FACE, Gosmanova, Elvira O., MD, FASN, and Kitabchi, Abbas E., MD, PhD, MACE. Hyperglycemic Crises: Diabetic Ketoacidosis (DKA), And Hyperglycemic Hyperosmolar State (HHS). Endotext. May 2018. 2. Cashen, Katherine, DO and Petersen, Tara, MD, MSEd. Diabetic Ketoacidosis. Pediatrics in Review. August 2019. 40(8): 412- 20. 3. Wolfsdorf, Joseph, MB, BCH, Glaser, Nicole, MD, and Sperling, Mark A., MD. Diabetic Ketoacidosis in Infants, Children, and Adolescents. A consensus statement from the American Diabetes Association. Diabetes Care. May 2006. 29 (5). 1150-9. 4. George A Burghen, James N Etteldorf, Joseph N Fisher, and Abbas Q Kitabchi Comparison of High-Dose and Low-Dose Insulin by Continuous Intravenous Infusion in the Treatment of Diabetic Ketoacidosis in Children. Diabetes Care January/February 1980 3:1 15-20 5. Hughes, Helen K., Kahl, Lauren K. The Harriet Lane Handbook: Twenty-First Edition. Johns Hopkins Children’s Center. Chapter 10: Endocrinology. 257-259. 6. International Diabetes Federation “IDF Diabetes Atlas: 9th Edition 2019” 7. Malik, Faisal S., MD, MS, Hall, Matt, PhD, Mangione-Smith, Rita, MD, MPH, et al. Patient Characteristics Associated with Differences in Admission Frequency for Diabetic Ketoacidosis in United States Children’s Hospitals. The Journal of Pediatrics. 2016: 171:104-10. 8. Sherry, Nicole A. and Levitsky, Lynne L. Management of Diabetic Ketoacidosis in Children and Adolescents. Pediatric Drugs 2008: 10(4): 209-215. 9. Vellanki, P and Umpierrez, G. “Increasing Hospitalizations for DKA: A Need for Prevention Programs” Diabetes Care Volume 41, September 2018Nyenwe, Ebenezer A and Kitabchi, Abbas E. The evolution of diabetic ketoacidosis: An update of its etiology, pathogenesis and management. Metabolism. April 2016. 65(4): 507-21. We would like to express our gratitude and appreciation for Dr. Hanna Sahhar, MD, for his guidance and support throughout our research process. His care and enthusiasm for both his patients and his students is evident, and through his instruction, we were able to complete this project dedicated to the pediatric community. We would also like to acknowledge Dr. Ning Cheng, PhD, whose diligent work and expertise helped us to bring this project to life. Acknowledgements

Figure 3 . Complications reported in this study with significant (p<0.05) correlation of brain edema to administration of sodium bicarbonate in the emergency department.

Figure 2. Changes in heartrate from initial presentation to resolution of metabolic acidosis and resolution of dehydration.

21

2 0 2 1 R e s e a r c h R e c o g n i t i o n D a y