Hospital-Acquired Venous Thromboembolism (HA-VTE) Epidemiology and Risk Factors Among Critically Ill Children and Young Adults with Diabetic Ketoacidosis

Introduction: Children hospitalized for diabetic ketoacidosis (DKA) experience hyperglycemic hyperosmolarity and severe dehydration that are considered prothrombotic, heightening the risk of hospital-acquired venous thromboembolism (HA-VTE). Yet, the occurrence rate and risk factors for HA-VTE among cases of pediatric DKA remain ill-defined.

Methods: We performed a retrospective, single center cohort study including children and young adults 0 - 21 years of age hospitalized in the pediatric intensive care unit for DKA from Oct 2020 – December 2023. The primary outcome was radiographically confirmed HA-VTE (i.e., pulmonary embolism, limb/neck deep venous thrombosis, and cerebral sinovenous thromboembolism). Detailed data were collected on HA-VTE (timing, location, central venous catheterization [CVC]-relatedness), DKA (degrees of acidosis and hyperglycemia, β-hydroxybutyrate levels, clinical cerebral edema), thromboprophylaxis (TP) modalities, and putative HA-VTE risk factors (age, inherited coagulopathy, acute and chronic immobility, mechanical ventilation [IMV], infectious comorbidity). Descriptive and comparative (e.g., chi-square, student’s t, and Wilcoxon rank sum tests) statistics were performed.

Results: During the study period, 386 children and young adults were hospitalized for DKA of which 2 (0.5%) developed a HA-VTE (both were CVC-related). The children who developed HA-VTE were 13 and 14 years of age, prescribed mechanical TP without pharmacologic TP, treated for acute cerebral edema, profoundly acidotic (β-hydroxybutyrate: 8.5 and 9.6 mmol/L, respectively), acutely immobilized (per Braden QD values), and exhibited plasma hyperosmolarity (339 and 342 mOsm/kg, respectively). Rates of HA-VTE were more common among subpopulations with CVC (40% vs 0%), symptomatic cerebral edema (12.5% vs 0%), undergoing IMV (20% vs 0.2%), and COVID-19 infection (5.6% vs 0.3%) as compared to those without these clinical features.

Conclusion: The observed HA-VTE rate among critically ill children with DKA over a three-year period was 0.5%. Putative risk factors for HA-VTE in this population, including comorbid conditions and markers of disease severity at presentation such as symptomatic cerebral edema, degree of acidosis, CVC, and IMV, should be prospectively investigated.