Hyperglycemic Emergencies in Minority Children and Adolescents With Diabetes Mellitus

Abstract

Abstract Introduction: Hyperglycemic emergencies in children with diabetes traditionally include diabetic ketoacidosis (DKA) predominantly associated with T1DM, whereas hyperosmolar hyperglycemic state (HHS) that is associated with relative insulin deficiency in T2DM, rarely occurs (2%). There have been increasing reports of mixed DKA-HHS affecting up to 27%. The purpose of this study is to identify clinical features, risk factors, complications and outcomes among minority children presenting with hyperglycemic emergencies at our center. Methods: This is a retrospective chart review of children and adolescents (ages 1-21 years) admitted for hyperglycemic emergencies including DKA [defined as glucose >200 mg/dL, metabolic acidosis (pH <7.3 or serum bicarbonate <15 mmol/L) and ketonemia (ß-hydroxybuyrate ≥3 mmol/L) or moderate to large ketonuria], or HHS [defined as glucose >600 mg/dL, effective serum osmolality >320 mOsm/kg, venous pH >7.25 or arterial pH >7.30 or serum bicarbonate >15 mmol/L, absent to mild ketonemia] or mixed DKA-HHS between 2004 and 2019. Descriptive statistics, chi-squared analysis and t-tests were used. Results: Of 322 patients, 92% were African American with mean age 13.6 ± 3.5 years, 39% males, consisting of 266 (83%) with DKA, 52 (16%) mixed DKA-HHS, and 4 (1%) HHS. Ninety-eight of the DKA and DKA-HHS groups had T1DM. All 4 patients with HHS had T2DM. Compared to the DKA group, the mixed DKA-HHS group required higher IV fluids rates (p<0.0001), 4.3-fold greater odds of acute kidney injury (AKI, mean serum creatinine of 1.6 mg/dl vs 1.1 mg/dl; ref 0.5-1.0 mg/dL) and 3.3-fold greater odds of developing altered mental status (AMS). Risk factors such as insulin adherence, and precipitating factors like infection or stress were not different between both groups (p=0.06). There was no significant difference in insulin rates or time to resolution (p=0.4) in both groups. Among the HHS group, 50% presented with AMS and AKI (mean Cr 1.1 mg/dl ± 0.31) due to severe dehydration and required higher IV fluids rates (≥2 x maintenance). Creatinine kinase in 2 patients were slightly elevated (mean 1643 units/L, ± 77; ref 39-309). Insulin drip at 0.05 u/kg/hour was started in all 4 patients. None had venous thrombosis, rhabdomyolysis, cerebral edema, or death. Average time of resolution was 10 ±1.63 hours. Conclusion: In this study, 16% of patients with hyperglycemic emergencies presented with mixed DKA-HHS which was associated with more complications compared to the DKA group. Serum osmolality should be calculated and checked at diagnosis. Identification of hyperosmolality whether with or without DKA in the emergency setting is important because treatment should be focused on the degree of dehydration (usually moderate to severe) and other complications such as AKI and mental status changes.