(908) US Prescribing Trends and Efficacy of Mechanical Thromboprophylaxis Among Critically Ill Adolescents

Introduction: Mechanical thromboprophylaxis (mTP) is applied to critically ill children without trial-derived evidence of a risk reduction for hospital-acquired venous thromboembolism (HA-VTE). We sought to estimate the rate of mTP prescription among critically ill adolescents and determine if mTP is inversely associated with HA-VTE.

Methods: Using Pediatric Health Information System registry, we studied children aged 12-17 years hospitalized from 2016-2023 within 30 United States (US) pediatric intensive care units (PICUs). We excluded encounters with a VTE present on admission. Rates of mTP prescription were estimated and used to define study cohorts. The primary outcome was the development of a HA-VTE including limb/neck deep venous thromboses and pulmonary embolism. To evaluate the potential association of mTP and HA-VTE, adjusted logistic models were applied using three risk strata (low, moderate, high) for critically ill children by the presence of known prothrombotic risk factors (i.e., central venous catheterization [CVC], length of stay [LOS], and comorbid infection).

Results: 107,804 children were studied of which 21,124 (19.6%) were prescribed mTP. Hospital center mTP prescribing rates ranged from 1.4-65.4%. Compared to those without mTP prescription, those with more frequently had CVC (17.2% vs 9.4%, p< 0.001), underwent invasive mechanical ventilation, (37.4% vs 24.8%, p< 0.001), were admitted for postoperative care (30.9% vs12.7 %, p< 0.001), and had a longer median LOS (7 [interquartile range: 4-15] vs 4 [interquartile range: 2-9] days, p< 0.001). HA-VTE occurred in 2.7% of the sample and was more common among those with mTP prescription as compared to those without (4% vs 2.4%, p< 0.001). In multivariate logistic regression models for HA-VTE, the prescription of mTP was not associated with risk of HA-VTE among all three risk tiers.

Conclusions: mTP prescribing among critically ill adolescents is common (19.6%) and varies widely by institution (1.4-65.4%). We did not detect independent associations of mTP with HA-VTE risk after adjusting for prothrombotic factors and covariates. Until prospective trials are performed to validate our findings and address limitations of these methodology, clinicians must consider if mTP alone is sufficient as primary TP for HA-VTE.