(752) Hospital-Acquired Venous Thromboembolism Among Critically Ill Children With Diabetic Ketoacidosis

Introduction: Diabetic ketoacidosis (DKA) is considered a prothrombotic state as a result of hyperglycemic hyperosmolarity and severe dehydration at the time of presentation. We sought to characterize the prevalence of and risk factors for HA-VTE among children hospitalized for DKA.

Methods: We performed a retrospective, single center cohort study of critically ill children 0 - 17 years of age admitted for DKA from Oct 2020 - May 2023. The primary outcome was radiographically confirmed HA-VTE (i.e., pulmonary embolism, limb/neck deep venous thrombosis, and cerebral sinovenous thromboembolism). Secondary outcomes were HA-VTE data (i.e., timing, location, and central venous catheterization [CVC]-relatedness), DKA data (i.e., β-hydroxybutyrate levels and clinical cerebral edema), thromboprophylaxis (TP) data, and putative HA-VTE risk factors (e.g., age, inherited coagulopathy, acute and chronic immobility, CVC, mechanical ventilation [MV], and admission comorbidities [e.g., infectious diseases]). Descriptive (e.g., means, medians, and proportions), comparative (e.g., chi-square, student’s t, and Wilcoxon rank sum tests), and associative (e.g., logistic regression) statistics were applied as appropriate.

Results: During the study period, 275 children were hospitalized for DKA of which 2 (0.7%) developed a CVC-related HA-VTE. Both children, ages 15 and 14, were prescribed mechanical TP without anticoagulant TP, treated for cerebral edema, acidotic (β-hydroxybutyrate levels of 8.5 and 9.6 mmol/L), acutely immobilized (via Braden QD scores), receiving MV, hyperosmolar (339 and 342 mOsm/kg at presentation), and one had comorbid COVID-19 infection. For the study sample, only 59 (21.5%) were prescribed TP, 4 (1.5%) had a CVC, 29 (10.6%) had symptomatic cerebral edema, and the median β-hydroxybutyrate level on admission was 6.6 mmol/L (interquartile range: 4.7 - 8.8).

Conclusions: In this study sample, the estimated prevalence of HA-VTE among children with DKA was 0.7%. Large observational analyses are needed to verify putative risk factors for HA-VTE (e.g., CVC, acidosis, and hyperosmolarity) and assess TP efficacy, both mechanical and pharmacologic, in this unique pediatric subpopulation.