Another trial of prehospital plasma for trauma (COMBAT)

A couple weeks ago I covered the PAMPer trial, which demonstrate a potentially “unbelievable” 10% decrease in mortality by using plasma as the primary resuscitation fluid prehospital in trauma patients being transferred by air. I was somewhat skeptical (surprising I know) of the result. This is a similar study, but with different conclusions.

The paper

Moore HB, Moore EE, Chapman MP, et al. Plasma-first resuscitation to treat haemorrhagic shock during emergency ground transportation in an urban area: a randomised trial. Lancet. 2018; 392(10144):283-291. PMID: 30032977

The Methods

This is a single-center pragmatic, randomised, placebo-controlled trial.

Patients: Adult trauma patients with a systolic blood pressure (SBP) below 70 or a SBP between 71 and 90 with a heart rate of 108 thought to be due to bleeding.

• Exclusions were: prisoners, pregnancy, isolated gunshot to the head, asystole or CPR prior to randomization, and known objection to blood products.

Intervention: 2 units of rapidly thawed AB plasma.

Comparison: 0.9% (normal) saline.

• Coolers for placebo contained frozen water, but the water was discarded and saline used for speed of administration and to avoid giving cold saline.
• Masking of the placebo was not done, because the FDA does not allow colouring of IV fluids.
• The volume of saline was determined by hemodynamic need (not a set volume).

Outcome: The primary outcome was 28 day all cause mortality.

The Results

144 patients were enrolled (out of 150 planned), 19 of whom were excluded after randomization because of exclusion criteria or lack of consent, leaving 125 patients. The groups look similar at baseline.

All 65 patients in the plasma group received the full 2 units of plasma. About ⅓ had both units transfused before hospital arrival. The total volume of saline used in the control group was pretty small (median 250 mL).

More patients died in the plasma group, but not statistically so (15% vs 10%; OR 1.54; 95% CI 0.60 – 3.98; p=0.37).

None of the secondary outcomes were statistically significant.

My thoughts

Though asking a similar question, this study seems to come to the opposite conclusion of PAMPer. 

Neither study was perfect. The biggest problem with this study is it’s sample size. The study was powered for a rather unrealistic 19% decrease in mortality. (Ie, based on their assumed baseline mortality of 25%, they assumed that plasma would decrease morality to 6%). I was surprised at the 10% decreased seen in PAMPer. A 19% decrease is essentially unheard of and a bit of a silly goal. The result is incredibly large confidence intervals on the primary outcome, meaning that plasma could be anything from incredibly beneficial to incredibly harmful. 

The small sample size also limits our assessment of harms. Although no differences were seen here, transfusion of FFP has numerous known harms that have to be considered.

One of the interesting discussions at the #resusTO conference 2 weeks ago was about opportunity cost in resuscitation. Thawing of plasma takes time, and paramedics are prevented from performing other interventions during the thawing procedures. Even if plasma is a better resuscitation fluid that crystalloid, it might not be as important as the many other things paramedics needs to get done to safely transport patients to the hospital.

Bottom line

Prehospital in trauma is not ready for prime time. Although there was no statistical difference, the 5% higher mortality in the plasma group in this trial is concerning. This study doesn’t alter any of the comments I made after PAMPer. I am still skeptical of the entire concept of balanced transfusion, but I also remain hopeful.