632: HOSPITAL-ACQUIRED VENOUS THROMBOEMBOLISM AMONG CHILDREN UNDERGOING INVASIVE MECHANICAL VENTILATION

Abstract

Introduction: Hospital-acquired venous thromboembolism (HA-VTE) has an incidence of 2.2% among critically ill children. Although risk factors have been described (e.g., immobility, central venous catheterization [CVC], and systemic inflammation), insufficient data exists to recommend routine thromboprophylaxis (TP) in the pediatric intensive care unit (PICU). Children undergoing mechanical ventilation (MV) may represent an at-risk population due to illness severity, intentional immobility, and frequent CVC presence. We sought to estimate the rate and timing of HA-VTE for children undergoing MV and explore for variation in HA-VTE rates by MV parameters. Methods: We performed a single-center, retrospective cohort study of children < 18 years of age in the PICU undergoing MV from October 2020 - March 2022 excluding those with tracheostomy, HA-VTE prior to MV, and a total MV exposure of < 24 hours. The primary outcome was HA-VTE identified after intubation confirmed by imaging. Secondary outcomes were HA-VTE characteristics (i.e., timing, location, and CVC-related), MV parameters (i.e., barometric, volumetric, and compliance data within 72-hours of intubation), and other known HA-VTE risk factors. Descriptive and comparative statistics (Fisher’s exact, Wilcoxon rank sum, and Student’s t tests) were employed. Results: Of 170 subjects studied, 18 (10.6%) developed a HA-VTE (limb deep venous thromboses) at median of 4 (interquartile range [IQR]:1.4,6.4) days after intubation. Those with HA-VTE had a greater frequency of comorbid CVC (88.9% vs 61.8%, P=0.034) and prior history of HA-VTE (27.8% vs 8.6%, P=0.027). No differences in demographics, anthropometrics, severity of illness indices, immobility, applied TP, rates of comorbid hematologic malignancy, sepsis, COVID-19, trauma, or postoperative admission were noted. No differences were observed for rates of conventional MV, high-frequency oscillation, intubation timing, or MV duration. Ventilator parameters were not different between those with and without HA-VTE. Conclusions: In our study, the rate of HA-VTE among critically ill children undergoing MV was 10.6% and more common to children with comorbid CVC. Although HA-VTE rates were not observably different by MV exposure, duration, or intensity, prospective studies are still required.