Valproic acid decreases brain lesion size and improves neurologic recovery in swine subjected to traumatic brain injury, hemorrhagic shock, and polytrauma

BACKGROUND: We have previously shown that treatment with valproic acid (VPA) decreases brain lesion size in swine models of traumatic brain injury (TBI) and controlled hemorrhage. To translate this treatment into clinical practice, validation of drug efficacy and evaluation of pharmacologic properties in clinically realistic models of injury are necessary. In this study, we evaluate neurologic outcomes and perform pharmacokinetic analysis of a single dose of VPA in swine subjected to TBI, hemorrhagic shock, and visceral hemorrhage. METHODS: Yorkshire swine (n = 5/cohort) were subjected to TBI, hemorrhagic shock, and polytrauma (liver and spleen injury, rib fracture, and rectus abdominis crush). Animals remained in hypovolemic shock for 2 hours before resuscitation with isotonic sodium chloride solution (ISCS; volume = 3× hemorrhage) or ISCS + VPA (150 mg/kg). Neurologic severity scores were assessed daily for 30 days, and brain lesion size was measured via magnetic resonance imaging on postinjury days (PID) 3 and 10. Serum samples were collected for pharmacokinetic analysis. RESULTS: Shock severity and response to resuscitation were similar in both groups. Valproic acid-treated animals demonstrated significantly less neurologic impairment between PID 1 to 5 and smaller brain lesions on PID 3 (mean lesion size ± SEM, mm: ISCS = 4,956 ± 1,511 versus ISCS + VPA = 828 ± 279; p = 0.047). No significant difference in lesion size was identified between groups at PID 10 and all animals recovered to baseline neurologic function during the 30-day observation period. Animals treated with VPA had faster neurocognitive recovery (days to initiation of testing, mean ± SD: ISCS = 6.2 ± 1.6 vs ISCS + VPA = 3.6 ± 1.5; p = 0.002; days to task mastery: ISCS = 7.0 ± 1.0 vs ISCS + VPA = 4.8 ± 0.5; p = 0.03). The mean ± SD maximum VPA concentrations, area under the curve, and half-life were 145 ± 38.2 mg/L, 616 ± 150 hour·mg/L, and 1.70 ± 0.12 hours. CONCLUSIONS: In swine subjected to TBI, hemorrhagic shock, and polytrauma, VPA treatment is safe, decreases brain lesion size, and reduces neurologic injury compared to resuscitation with ISCS alone. These benefits are achieved at clinically translatable serum concentrations of VPA. LEVEL OF EVIDENCE: Therapeutic (preclinical study).