Overresuscitation with plasma is associated with sustained fibrinolysis shutdown and death in pediatric traumatic brain injury

Abstract

Background: Elevated International Normalized Ratio (INR) is a marker of poor outcome but not necessarily bleeding or clinical coagulopathy in injured children. Conversely, children with traumatic brain injury (TBI) tend to be hypercoagulable based on rapid thromboelastography (rTEG) parameters. Many clinicians continue to utilize INR as a treatment target.

Methods: Prospective observational study of severely injured children age < 18 with rTEG on arrival and daily thereafter for up to 7 days. Standard rTEG definitions of hyperfibrinolysis (LY30 ≥ 3), fibrinolysis shutdown (SD) (LY30 ≤ 0.8), and normal (LY30 = 0.9-2.9) were applied. The first 24-hour blood product transfusion volumes were documented. Abbreviated Injury Scale score ≥ 3 defined severe TBI. Sustained SD was defined as two consecutive rTEG with SD and no subsequent normalization. Primary outcomes were death and functional disability, based on functional independence measure score assessed at discharge.

Results: One hundred one patients were included: median age, 8 years (interquartile range, 4-12 years); Injury Severity Score, 25 (16-30); 72% blunt mechanism; 47% severe TBI; 16% mortality; 45% discharge disability. Neither total volume nor any single product volume transfused (mL/kg; all p > 0.1) differed between TBI and non-TBI groups. On univariate analysis, transfusion of packed red blood cells (p = 0.016), plasma (p < 0.001), and platelets (p = 0.006) were associated with sustained SD; however, in a regression model that included all products (mL/kg) and controlled for severe TBI (head Abbreviated Injury Scale score ≥ 3), admission INR, polytrauma, and clinical bleeding, only plasma remained an independent predictor of sustained SD (odds ratio, 1.17; p = 0.031). Patients with both severe TBI and plasma transfusion had 100% sustained SD, 75% mortality, and 100% disability in survivors. Admission INR was elevated in TBI patients, but did not correlate with rTEG activated clotting time (p = NS) and was associated with sustained SD (p = 0.006).

Conclusion: Plasma transfusion is independently associated with sustained fibrinolysis SD. Severe TBI is also associated with sustained SD; the combined effect of plasma transfusion and severe TBI is associated with extremely poor prognosis. Plasma transfusion should not be targeted to INR thresholds but rather to rTEG activated clotting time and clinical bleeding.

Level of evidence: Prognostic and epidemiological study, level III.