To develop a multivariable model to identify pediatric patients with diabetic ketoacidosis (DKA) and/or hyperglycemic hyperosmolar state (HHS) at increased risk of specific adverse outcomes, including death and prolonged ICU stays.
Key Findings:
1.5% of admissions had adverse outcomes from 2010-2019, increasing to 5.0% during 2020-2021.
Maximum serum glucose, initial pH, and diagnosis of type 2 diabetes were identified as strong predictors of adverse outcomes.
The model demonstrated excellent discrimination with AUC values of 0.948 for training, 0.960 for 2020-2021, and 0.873 for 2022-2023.
Interpretation:
Type 2 diabetes and the severity of initial hyperglycemia and acidosis are independent risk factors for adverse outcomes, which may explain the increased frequency observed during the COVID-19 pandemic.
Limitations:
Retrospective design may introduce bias, potentially affecting the reliability of the findings.
Data limited to a single institution may restrict the generalizability of the results.
Conclusion:
The predictive model can assist in identifying at-risk pediatric patients with DKA and HHS, particularly during periods of increased incidence such as the COVID-19 pandemic.
