Alcohol and mortality after moderate to severe traumatic brain injury: a meta-analysis of observational studies.

OBJECT Experimental studies have shown numerous neuroprotective properties of alcohol ("ethanol") after TBI, but clinical studies have provided conflicting results. The authors aimed to assess the relationship between positive blood alcohol concentration (BAC) on hospital admission and mortality after moderate to severe traumatic brain injury (TBI). METHODS The authors searched 8 databases for observational studies reported between January 1, 1990, and October 7, 2013, and investigated the effect of BAC on mortality after moderate to severe TBI. Reviews of each study were conducted, and data were extracted according to the MOOSE and PRISMA guidelines. Study quality was assessed using the Newcastle-Ottawa scale. The Mantel-Haenszel fixed effect methodology was used to generate pooled estimates. Heterogeneity was dealt with by multiple sensitivity analyses. RESULTS Eleven studies with a total of 95,941 patients (42% BAC positive and 58% BAC negative) were identified for the primary analysis (overall mortality 12%). Primary analysis showed a significantly lower risk of death for BAC-positive patients compared with BAC-negative patients (crude mortality 11.0% vs 12.3%, pooled OR 0.84 [95% CI 0.81-0.88]), although flawed by heterogeneity (I(2) = 68%). Multiple sensitivity analyses, including 55,949 and 51,772 patients, yielded similar results to the primary analysis (crude mortality 12.2% vs 14.0%, pooled OR 0.87 [95% CI 0.83-0.92] and crude mortality 8.7% vs 10.7%, pooled OR 0.78 [95% CI 0.74-0.83]) but with good study homogeneity (I(2) = 36% and 14%). CONCLUSIONS Positive BAC was significantly associated with lower mortality rates in moderate to severe TBI. Whether this observation is due to selection bias or neuroprotective effects of alcohol remains unknown. Future prospective studies adjusting for TBI heterogeneity is advocated to establish the potential favorable effects of alcohol on outcome after TBI.