Mechanical Thromboprophylaxis and Hospital-Acquired Venous Thromboembolism: A Report from the Children’s Healthcare Advances in Thrombosis Consortium

Background: Mechanical thromboprophylaxis (mTP) is applied to critically ill children without trial-derived or observational evidence to suggest a risk reduction for hospital-acquired venous thromboembolism (HA-VTE).

Aims: To assess potential associations between mTP and HA-VTE among critically ill adolescents.

Methods: We performed a multicenter case-control study as a secondary analysis of the Children’s Healthcare Advances in Thrombosis (CHAT) Consortium registry including children 12-19 years of age admitted to a CHAT participating pediatric intensive care unit (PICU). Cases with HA-VTE, defined as radiographically confirmed pulmonary embolism and/or limb/neck/abdominal deep venous thrombosis (DVT) excluding those present on admission, and controls without HA-VTE were identified for study. The primary exposure variable of interest was mTP administration. Data collected for analysis included demographics, body mass index, HA-VTE characteristics, pharmacologic TP (pTP) and mTP use, central venous catheterization (CVC), invasive mechanical ventilation (IMV) use and duration, hospital length of stay (LOS), Braden Q immobility scores, and concomitant disorders such as sepsis, hematologic malignancy, congenital heart disease, and autoimmune/rheumatologic disease. To assess a priori thrombotic risk, HA-VTE risk-tiers (low, moderate, and high) were calculated based on Arlikar et al. For each tier, multivariate logistic regression models for HA-VTE were employed, reporting adjusted odds ratios (aOR) with 95% confidence intervals (CIs).

Results: A total of 163 HA-VTE cases and 975 non-HA-VTE controls were identified for study, of which 23.5% received mTP, 7.6% pTP, and 9.2% both mTP and pTP. Most HA-VTE were lower/upper extremity DVT (89%) and CVC-related (59.5%). Compared to controls, cases more frequently had CVC (89% vs 21.1%), IMV (52.2% vs 11.8%), longer median LOS (29 [IQR:15-46] vs 6 [IQR:3-10] days), complete / very limited mobility by Braden Q score (72.6% vs 22.1%), and sepsis (48.5% vs 16%) (all p<0.001). In the high HA-VTE-risk tier, a multivariate logistic regression model for HA-VTE did not detect an association with mTP (aOR: 0.49, 95%CI: 0.16-1.46, p=0.2) but did detect associations with pTP (aOR: 0.18, 95%CI: 0.04-0.78, p=0.022), combined pTP and mTP (aOR: 0.20, 95%CI: 0.06-0.66, p=0.008), IMV >1 day (aOR: 4.25, 95%CI: 1.63-11.03, p=0.003), and complete / very limited mobility (aOR 3.72, 95%CI: 1.4-9.8, p=0.008).

Conclusions: Among critically ill adolescents at high risk for HA-VTE, pTP or the combination of pTP and mTP, but not mTP alone, was associated with a risk reduction for HA-VTE. By contrast IMV and immobility are also independently associated with increased HA-VTE risk. Taken in sum, these findings identify an important risk-stratified population for upcoming pediatric pTP clinical trials.