AAST 2017: Utilité du TXA encore questionnée

Late TXA utilization is associated with Iincreased blood product transfusion: A secondary analysisof the pragmatic randomized optimal platelet and plasma ratios  (PROPPR) study

Aravind K. et AL . Session: XIIIA: Papers 45-54: Preclincial/Translational Science Paper 52

-----------------------------------

Ce poster présenté au dernier AAST 2017 est une ré-interprétation de données de l'étude POPR. Il semblerait qu'au delà des premières heures l'administration précoce (dans la 1ère heure) de TXA n'améliore pas la survie à 24h et un mois. L'administration tardive (après la 1ère heure mais avant la 3ème) majorerait les besoins transfusionnels. Les auteurs de ce poster attirent l'attention sur la disparité des groupes comparés, ce qui peut expliquer leurs résultats.

-----------------------------------

Introduction:

Exsanguination is the leading cause of preventable death after trauma. In addition to a balanced ratio blood component strategy, tranexamic acid (TXA) is used as an adjunct in hemorrhaging patients. This secondary analysis was performed to determine the incidence of TXA utilization and outcome in patients predicted to receive a massive transfusion (MT) in level 1 trauma centers.

Methods:

Trauma patients who were predicted to require a MT and admitted to 12 level I North American trauma centers were studied. Patients were divided into those who received TXA and those who did not. We examined 3 hour, 24 hour, and 30 day mortality. We also examined incidence of thromboembolic events, blood product administration within the first 24 hours, length of stay (hospital free days), ICU free days, as well as development of complications including acute respiratory distress syndrome (ARDS), acute kidney injury (AKI), sepsis, and multisystem organ failure (MOF). In our multivariate analysis, we controlled for Injury Severity Score (ISS), Glasgow Coma Scale (GCS), treatment group, mechanism of injury, hypotension and/or tachycardia on admission, geriatric patient (age > 65), and site as independent variables.

Results:

137 out of 680 (20.1%) patients in the PROPPR study received TXA with 130 patients receiving TXA within the first 3 hours after admission. Other adjunctive therapies administered included cryoprecipitate (25.4%), and others (6.6%). The incidence of TXA administration did not differ between the ratio groups (50.3% vs 47.4%, p=0.55), but patients receiving TXA were more severely injured with a median ISS of 34(21) vs 26(20), p<0.01  and a lower median(IQR) GCS of 9(12) vs 14(12),  p<0.01. Multivariate linear regression analysis revealed no association between TXA administration and blood transfusion requirements (Table 1). Further analysis revealed that patients who received late (from >1 hour to ≤3 hours after arrival) TXA (41 patients) experienced increased blood requirements in the first 24 hours (Table 1) compared to those remaining 543 patients that did not receive TXA. There was no difference in blood product requirement in those patients who received TXA early (≤1 hour) (89 patients) versus those that did not receive TXA. In patients that receiv ed TXA, there was an increased incidence of ARDS(OR (95% CI) 1.99 (1.06,3.73), p=0.03), AKI (1.90 (1.13,3.20), p=0.01), and MOF (4.18 (1.52,11.48), p<0.01) even when controlling for the factors mentioned above. There was also a difference in adjusted 3 hour mortality (OR (95% CI) 0.22 (0.07,0.73), p=0.01) but not 24 hour (0.61 (0.30,1.24), p=0.18) or 30 day mortality (1.42 (0.78,2.59), p=0.25) for any TXA administration. There was no difference in adjusted thromboembolic events or adjusted length of stay. Subgroup analysis with additional variables that were found to have a difference between groups with a p​<0.20 were added to the regression model including hematocrit, platelet count, international normalized ratio, creatinine, lactate, and R value on thrombelastography. This analysis showed an increase in PRBC transfusion with late TXA administration (16 out of 232 patients, 7.51 (0.46,14.56), p=0.04), but showed no difference in FFP or platelet administration.

Conclusion: Early TXA use was not associated with improved outcomes. Late TXA use was associated with increased blood product resuscitation. TXA administration in general was associated with improved 3 hour mortality. This did not translate to an improvement in mortality at 24 hours or 30 days. There was a significant increase in the incidence of ARDS, AKI, and MOF in patients who received TXA but this analysis is limited by the differences in the 2 populations despite attempts to control for them