Erythropoietin in patients with traumatic brain injury and extracranial injury – A post hoc analysis of the EPO-TBI trial

BACKGROUND: Erythropoietin (EPO) may reduce mortality after traumatic brain injury (TBI). Secondary brain injury is exacerbated by multiple trauma, and possibly modifiable by EPO. We hypothesized that EPO decreases mortality more in TBI patients with multiple trauma, than in patients with TBI alone. METHODS: A post hoc analysis of the EPO-TBI randomised controlled trial conducted in 2009 to 2014. To evaluate the impact of injuries outside the brain, we calculated an extracranial injury severity score (ISS) that included the same components of the ISS, excluding head and face components. We defined multiple trauma as two injured body regions with an abbreviated injury score (AIS) of 3 or higher. Cox regression analyses, allowing for potential differential responses per the presence or absence of extracranial injury defined by these injury scores, was used to assess the effect of EPO on time to mortality. RESULTS: Of 603 included patients, the median extracranial ISS was 6 (inter quartile range 1-13) and 258 (43%) had AIS of 3 or higher in at least two body regions. On Cox regression, EPO was associated with decreased mortality in patients with greater extracranial ISS (interaction p 0.048) and weakly associated with differential mortality with multiple trauma (AIS>3 or in two regions, interaction p 0.17). At six months in patients with extracranial ISS higher than six, 10 (6.8%) out of 147 EPO treated patients compared to 26 (17%) of 154 placebo treated patients died (Risk reduction 10%, 95% CI 2.9% to 17%, p=0.007). CONCLUSIONS: In this post hoc analysis, EPO administration was associated with a potential differential improvement in 6-month mortality in TBI patients with more severe extracranial injury. These findings need confirmation in future clinical and experimental studies. LEVEL OF EVIDENCE: Therapeutic study, LEVEL II