Whole blood versus balanced resuscitation in massive hemorrhage: six of one or half dozen of the other?

INTRODUCTION: Whole blood (WB) resuscitation is increasingly used at trauma centers. Prior studies investigating outcomes in WB versus component-only (CO) resuscitation have been limited by small cohorts, low volumes of WB resuscitation, and unbalanced CO resuscitation. This study aimed to address these limitations using data from a high-volume Level I trauma center, which adopted a WB-first resuscitation paradigm in 2018. We hypothesized that the resuscitation method, WB or balanced CO, would have no impact on patient mortality. METHODS: A single-center, retrospective cohort study of adults presenting as a trauma activation from July 2016 through July 2021 was performed. Receipt of 3 or more units of WB or packed red blood cells (RBC) within the first hour of resuscitation was required for inclusion. Patients were grouped into WB versus CO resuscitation and important clinical outcomes were compared. Mortality was evaluated with Kaplan-Meier analysis, log-rank testing, and multivariable Cox proportional hazards modeling. RESULTS: There were 180 patients in the WB group and 170 patients in the CO group. Of the 180 WB patients, 110 (61%) received only WB during the first 24 hours. The WB group received a median of 5.0 units (IQR 4.0-8.0) of WB and CO group received a median of 6.0 units (IQR 4.0-11.8) of RBCs during the first 24 hours of resuscitation. In the CO group, median RBC/plasma and RBC/platelet ratios approximated 1:1:1. Groups were similar in clinicopathologic characteristics including age, injury severity score, mechanism of injury, and requirement for hemorrhage control interventions (WB 55% vs CO 59%, p = 0.60). Unadjusted survival was equivalent at 24 hours (p = 0.52) and 30 days (p = 0.70) between both groups on Kaplan-Meier analysis with log-rank testing. On multivariable Cox regression, WB resuscitation was not independently associated with improved survival after accounting for age, ISS, mechanism of injury, and receipt of hemorrhage control procedure (HR 0.85, 95% CI 0.61-1.19, p = 0.34). CONCLUSIONS: Balanced CO resuscitation is associated with similar mortality outcomes to that of WB based resuscitation. LEVEL OF EVIDENCE: Level IV; Therapeutic/Care Management.

Comparison of bleeding and thrombotic outcomes in veno-venous extracorporeal membrane oxygenation: Heparin versus bivalirudin.

OBJECTIVES: We aimed to evaluate thrombotic and hemorrhagic complications with heparin versus bivalirudin use in veno-venous extracorporeal membrane oxygenation (V-V ECMO). METHODS: We performed a retrospective cohort study of adult patients placed on V-V ECMO with intravenous anticoagulation with either heparin or bivalirudin. Time to thrombotic event and major bleed were analyzed in addition to related outcomes. RESULTS: We identified 95 patients placed on V-V ECMO: 61 receiving heparin, 34 bivalirudin. The bivalirudin group had a higher rate of severe COVID-19, higher BMI, and longer ECMO duration. Despite this, bivalirudin was associated with reduced risk of thrombotic event (HR 0.14, 95% CI 0.06-0.32, p < .001) and increased average lifespan of the circuit membrane lung (16 vs. 10 days, p = 0.004). While there was no difference in major bleeding, the bivalirudin group required fewer transfusions of packed red blood cells and platelets per 100 ECMO days (means of 13 vs. 39, p = 0.004; 5 vs. 19, p = .014, respectively). Lastly, the bivalirudin group had improved survival to ECMO decannulation in univariate analysis (median OS 53 vs. 26 days, p = .015). CONCLUSIONS: In this real-world analysis of bivalirudin versus heparin, bivalirudin is a viable option for V-V ECMO and associated with lower risk of thrombotic complications and fewer transfusion requirements.

The impact of low-dose aspirin in the Brain Injury Guidelines on outcomes in traumatic brain injury: A retrospective cohort study.

BACKGROUND: Current Brain Injury Guidelines (BIG) characterize patients with intracranial hemorrhage taking antiplatelet or anticoagulant agents as BIG 3 (the most severe category) regardless of trauma severity. This study assessed the risk of in-hospital mortality or need for neurosurgery in patients taking low-dose aspirin who otherwise would be classified as BIG 1. METHODS: This was a retrospective study at an academic level 1 trauma center. Patients were included if they were admitted with traumatic intracerebral hemorrhage and were evaluated by the BIG criteria. Exclusion criteria included indeterminate BIG status or patients with missing primary outcomes documentation. Patients were categorized as BIG 1, BIG 2, BIG 3, or BIG 1 on aspirin (patients with BIG 1 features taking low-dose aspirin). The primary endpoint was a composite of neurosurgical intervention and all-cause in-hospital mortality. Key secondary endpoints include rate of intracranial hemorrhage progression, and intensive care unit- and hospital-free days. RESULTS: A total of 1,520 patients met the inclusion criteria. Median initial Glasgow Coma Scale was 14 (interquartile range [IQR], 12-15), Injury Severity Scale score was 17 (IQR, 10-25), and Abbreviated Injury Scale subscore head and neck (AIS Head ) was 3 (IQR, 3-4). The rate of the primary outcome for BIG 1, BIG 1 on aspirin, BIG 2, and BIG 3 was 1%, 2.2%, 1%, and 27%, respectively; the difference between BIG 1 on aspirin and BIG 3 was significant ( p < 0.001). CONCLUSION: Patients taking low-dose aspirin with otherwise BIG 1-grade injuries experienced mortality and required neurosurgery significantly less often than other patients categorized as BIG 3. Inclusion of low-dose aspirin in the BIG criteria should be reevaluated. LEVEL OF EVIDENCE: Therapeutic/Care Management; Level IV.

Incidence of traumatic intracranial hemorrhage expansion after stable repeat head imaging: A retrospective cohort study.

INTRODUCTION: Patients with traumatic intracranial hemorrhage (tICH) often require intensive care unit (ICU) admission until bleeding stability is demonstrated through interval head computed tomography (HCT). The brain injury guidelines (BIG) suggest a minimum 24-h ICU admission for severe patients (BIG 3) regardless of repeat CT stability. We sought to evaluate the rate of tICH expansion after an initial stable interval scan was obtained. METHODS: A single-center retrospective cohort study at a level 1 trauma center was performed. All adult patients with tICH evaluated using BIG criteria were included. The primary endpoint was incidence of tICH expansion after initial stability on interval HCT performed at approximately 6 h. Secondary endpoints included time to tICH stability, frequency of neurosurgical intervention, and time to surgical intervention. RESULTS: A total of 1517 patients met inclusion criteria. Of the 1121 patients with repeat imaging, 288 (25.7%) experienced progression with 94.4% detected on the initial 6-h interval scan. Of all patients with initially stable repeat imaging (n = 833), progression occurred in 16 (1.9%) patients. Of these patients, 5 required neurosurgical intervention, 4 received increased monitoring, 2 transitioned to comfort measures and 5 had no change in management. The median time from initial scan to expansion in these patients was 42.2 h. Median time to surgical intervention after post-stability expansion was 102 h. CONCLUSION: Patients who demonstrate bleeding stability on first interval HCT after tICH rarely experience expansion. Consideration should be given to discharging patients from the ICU when initial interval HCT shows no progression.

Thromboelastography with platelet mapping: Limited predictive ability in detecting preinjury antiplatelet agent use.

BACKGROUND: Preinjury antiplatelet agent (APA) use in trauma patients can increase traumatic hemorrhage and worsen outcomes. Thromboelastography with platelet mapping (TEGPM) has characterized platelet function via arachidonic acid (AA) and adenosine diphosphate (ADP) inhibition in nontrauma settings, but limited data exist in the acute trauma population. METHODS: A prospective observational study of adult trauma patients with suspected preinjury APA use who received TEGPM testing from 2017 to 2020 was performed. Patients on anticoagulants were excluded. Patients were grouped according to preinjury APA regimen: 81 mg or 325 mg of aspirin daily, 81 mg of aspirin and 75 mg of clopidrogrel daily, 75 mg of clopidrogrel daily, or no antiplatelet. Ability of TEGPM to detect APA use was assessed using predictive statistics and area under receiver operating characteristic curves (AUROCs). RESULTS: A total of 824 patients were included with most patients taking 81 mg of aspirin (n = 558). Patients on no antiplatelet were younger and had higher baseline platelet counts, while patients on 75 mg of clopidrogrel were more likely to be admitted after ground level fall. All other baseline characteristics were balanced. Admission TEG values were similar between groups. Median AA inhibition was higher in patients on aspirin containing regimens (p < 0.0001). Median ADP inhibition was higher in patients on clopidogrel containing regimens and those taking 325 mg of aspirin (p < 0.0001). Arachidonic acid inhibition accurately detected preinjury APA use and aspirin use (AUROC, 0.89 and 0.84, respectively); however, ADP inhibition performed poorly (AUROC, 0.58). Neither AA nor ADP inhibition was able to discern specific APA regimens or rule out APA use entirely. CONCLUSION: High AA inhibition accurately detects preinjury APA use in trauma patients. High ADP inhibition after trauma is common, limiting its utility to accurately identify preinjury APA use. Further study is needed to identify assays that can reliably detect and further characterize preinjury APA use in trauma populations. LEVEL OF EVIDENCE: Diagnostic test, level II.