Evidence in Intensive Care - Trauma
Haemostatic drugs for TBI
Cochrane review 2010
No reliable data.
Since this review CRASH-2 published (20000 patient multicentre RCT Lancet 2010) which showed tranexamic acid in 1st 3h of injury after trauma reduced death due to bleeding. Subgroup analysis on GCS suggests likely to be effective in all groups. Crash 2 intracranial bleeding study ongoing.
NB I seem to remember further analysis of crash 2 showed worse outcomes if TXA given late – need to check this – see ICM monitor notes.
Routine ICP monitoring in coma
Cochrane 2010
No adequate studies so no evidence whether beneficial.
Tranexamic acid
Effect of tranexamic acid on mortality in patients with traumatic bleeding: prespecified analysis of data from randomised controlled trial.
Roberts et al. BMJ 2012; 345
Original CRASH study showed a very small 0.8% reduction in mortality (so small as be be almost meaningless). After 3 hours increased mortality was observed.
The data suggests that TXA may improve survival following injury but the mechanism and patient subgroups that may benefit are unclear.
CRASH 2 (paper 2)
Lancet 2011;377:1096-1101
Original trial showed all cause mortality reduced if tranexamic acid given in 1st 8h.
This analysis shows:
Risk of death due to bleeding significantly reduced if given in 1st 3h.
Risk of death due to bleeding significantly increased if given later than 3h (all cause mortality not increased).
Note that absolute reduction in mortality was 0.8%.
Some statisticians argue that an absolute difference of <2% should be viewed with caution as impossible to account for all confounders.
75% of patients in the trial were recruited from the developing world where pre hospital times are prolonged.
The majority of patients did not suffer significant haemorrhage
Most trauma deaths from haemorrhage caused by solid organ or major vessel disruption in 1st few hours.
Factor 7 fails to improve survival.
Why should tranexamic acid which works on smaller vessels make a difference?
Conclusion
We probably shouldn’t use it at all and certainly not beyond 3h.
Military application of tranexamic acid in trauma emergency resuscitation (MATTERs) study.
Morrison et al. Arch Surg 2012;147:113-119.
Retrospective observational. Poor study. Big differences between groups.
No difference at 24 h.
Significant benefit after this.
Suggests TXA does not work by stopping bleeding. It has anti-inflammatory effects which may explain the benefit.
Incidence of DVT and PE increased in TXA group.
Effect of TXA in traumatic brain injury
CRASH 2 collaborators. BMJ 2011
No harm or benefit using TXA in TBI.
Blood products
Zink. Am J Surg 2009; 197: 565-570
Riskin. J Am Coll Surg 2009; 209: 198-205
PRC:FFP/PLT should be 1.5:1 with implementation of a massive transfusion protocol to reduce mortality (evidence level 4)
Discussion
Some of the coagulopathy of trauma is independent of fluid/PRC dilution of coag factors and acidosis and hypothermia. (Brohi. J Trauma 2008;64:1211-1217). 2 mechanisms:
- Activation of systemic anticoagulation (thrombomodulin-protein C pathway)
- Hyperfibrinolysis (increased tPA release from endothelium following hypoperfusion and ischaemia)
Out of hospital hypertonic saline...
Bulger et al Ann Surg 2011
Hypertonic saline (7.5%) vs 0.9% saline pre-hospital - no benefit (both equally safe)
OOH hypertonic fluid resus following TBI
Bulger et al JAMA 2010
Pre-hospital no benefit (or harm) with hypertonic saline bolus.
Splenectomy
ReCONECT trial
Arch Surg 2010;145:456-460
Substantial haemorrhage, grade splenic injury or contrast extravasation on CT mean splenectomy necessary. Otherwise can conservatively manage.
Damage control resuscitation….in damage control laparotomy patients.
Cotton et al Ann Surg 2011;254
Avoiding excessive cystalloid resuscitation fluids, early use of blood and blood products and avoiding chasing a normal blood pressure as an initial goal appears to confer a survival benefit in major trauma.
Hypotensive resus in theatre for trauma surgery.
J Trauma 2011;70:652-663
Crap study see ICMonitor.
No difference in MAP between the 2 groups.
‘High’ MAP cohort had significantly more blunt trauma.