Hyponatremia and Severe Diabetic Ketoacidosis in Children with Type 1 Diabetes at an Indonesian Referral Hospital: A Cross-Sectional Study

Background: Diabetic ketoacidosis (DKA) is the leading acute cause of death in children with type 1 diabetes. Indonesian data on its severity and biochemical profile are scarce. Objective: To examine whether admission sodium, potassium and urine ketone status are associated with severe DKA. Methods: This retrospective cross-sectional study included children aged 1–18 years admitted in DKA (International Society for Pediatric and Adolescent Diabetes 2022 criteria) to a tertiary hospital in Bali in 2022–2024. Hyponatremia (<135 mmol/L), hypokalemia (<3.5 mmol/L) and urine ketone 3+ were compared between severe and mild–moderate DKA with exact tests and Firth-penalized logistic regression across all joint tables compatible with the observed pairwise counts. Results: Of 54 children (mean age 11.22 years; 74.1% girls), 37 (68.5%; 95% confidence interval [CI] 55.3–79.3) had severe DKA. Hyponatremia occurred in 24 of 37 children with severe DKA (64.9%) and 4 of 17 with mild–moderate DKA (23.5%) (odds ratio 6.00, 95% CI 1.62–22.20; Fisher p = 0.008; Holm-adjusted p = 0.023). In Model 1, the adjusted odds ratio for hyponatremia ranged from 4.83 to 4.90 and was significant in every admissible joint table. Hypokalemia (53.7%; odds ratio 1.05, p = 1.000) and urine ketone 3+ (57.4%; odds ratio 1.85, p = 0.379) were not associated with severity. Conclusion: Two-thirds of children admitted in DKA to this Indonesian referral hospital had severe acidosis. Measured hyponatremia at admission was associated with severe DKA. It should prompt calculation of the glucose-corrected sodium. Potassium and dipstick ketone grade were not associated with severity.