Traumatic brain injury (TBI) is one of the leading reasons for the increased burden on public health. Alcohol consumption is known to be a common risk factor for TBI. On the contrary, several studies have recently shown that higher alcohol levels in the blood at the time of injury could, in fact, promote short term survival in patients with TBI (Raj, Mikkonen et al. 2016). However, most of these studies have been underpowered with under-representation of certain age-groups as well as ethnicities. The present proposal aims to explore the influence of alcohol consumption on the prevalence of in-hospital mortality among patients with TBI using a large nationwide U.S. database, the National Trauma Data Bank (NTDB).

The objectives of this study are multifold.

1. As a primary objective, the study aims to explore the influence of alcohol consumption on in-hospital mortality in patients with TBI. I hypothesize decreased mortality in alcohol positive TBI patients.
2. As a secondary objective, the study aims to explore the effect of TBI severity on influencing mortality as a result of alcoholic consumption. I hypothesize an increased protective effect of alcohol on patients with severe TBI.

The present study could provide novel insights into the role of alcohol levels in blood in influencing the treatment outcome of patients with TBI admitted to the E.D., possibly leading to reduced morbidity and mortality among patients with TBI.

Background

Traumatic brain injury (TBI) occurs as a result of a severe hit or blow to the head, resulting in an alteration of brain function (Menon, Schwab et al. 2010). TBI leads to the excessive synthesis of the neurotransmitter glutamate, strongly linked with nerve cell death. It accounts for every 3rd injury-related death in the U.S. and causes an estimated loss of $76.5 billion every year (Coronado, McGuire et al. 2012). It is well established that 25-50% of patients with TBI admitted to hospitals are under the influence of above permissible alcohol levels in the blood. However, the effect of alcohol on in-hospital mortality of patients with TBI is highly controversial.

Several animal studies have previously shown that a low dose of alcohol offers protection against brain injury and a high dose aggravates brain injury (Asmaro, Fu et al. 2013). It has been hypothesized that alcohol induces neuroprotective effect by inhibiting N-methyl-D-aspartate (NMDA) receptor-mediated excitotoxicity, which attenuates the coupling between cerebral blood flow and metabolism of glucose, thereby delaying the hyperthermia induced by TBI (Talving, Plurad et al. 2010). Over the last two decades, several observational studies have attempted to explore alcohol’s influence on mortality after TBI. A recent comprehensive meta-analysis of 11 studies employing 95,941 patients with TBI showed a significantly lower risk of mortality in patients with severe or moderate TBI who tested positive for blood alcohol compared to those who tested negative for blood alcohol at the time of admission to the hospital (Mortality 11.0% vs 12.3%, pooled OR = 0.84 (95% CI 0.81–0.88)). The meta-analysis, however, had several limitations, including a high heterogeneity of 68%, which could be attributed to variability in study methodologies adopted by individual studies, including enrollment of indivuals with different age groups, failure to account for variability in ethnicity and stratification by disease severity.

In the past, several pharmaceuticals agents who have shown neuroprotective effects in preclinical studies have shown promising results in clinical trials in humans (Maas, Roozenbeek et al. 2010). As observed in animal models, the NMDA receptor blunting effect might be critical for nerve cell survival. Therefore, the present study could provide robust observational evidence by conducting one of the largest cross-sectional studies exploring influence of alcohol consumption on the prevalence of in-hospital mortality in patients with TBI to date with a uniform study design and accounting for ethnicity and disease severity, known to be potential confounding factors influencing mortality.

Methods

Study design

The present proposal aims to conduct a cross-sectional study to explore the relationship between TBI and mortality following alcohol consumption by collecting data on admissions to the emergency department (E.D.) across hospitals in the U.S. from the years 2010-2020, using the NTDB. Inclusion of patients with TBI will be based on criteria for identification of TBI by International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM). Patients with a history of chronic alcohol abuse or dependence will be excluded due to a potential differential effect on treatment outcome. Individuals under the age of 12 years will be further excluded as they are unlikely to be diagnosed with TBI as a result of alcohol consumption. And lastly, patients with missing information on main variables of interest were excluded.

Data collection

The primary outcome of the study will be the prevalence of in-hospital mortality among patients with TBI. Blood alcohol levels measured at the time of admission to the E.D. will be collected, and all the patients will be classified as ethanol positive (E.P.) or ethanol negative (EN), also considered the primary exposure in the present study. Additionally, the following data will be extracted from NTDB: age, sex, ethnicity, injury severity parameter (head Abbreviated injury scale or head ASI), Glasgow coma scale score, duration of stay in an intensive care unit (ICU) and will be judged for potential confounding in the association between TBI and mortality. Lastly, a patient was categorized as a severe TBI patient if he has a head AIS score of greater than 4. The E.P. and EN patients with TBI will be matched on age, gender and ethnicity to control for confounding.

Data analysis

The categorical data will be expressed in proportions, and continuous variables will be expressed in mean and standard deviation (S.D.). The distribution of categorical variables will be compared between patients who died, and patients who survived will be compared by Chi-square tests. The distribution of continuous variables between the same groups will be compared using the analysis of variance test (ANOVA). Confidence intervals will be computed at the 95% significance level. A p-value below 0.05 will be considered statistically significant. A binary logistic regression will be employed to check the association of mortality with the presence of alcohol in the blood outcome variable after adjusting the effect of potential confounding variables. Odds ratios (OR) will be computed for expressing the effect estimates. All statistical analyses will be performed using STATA 16.0 (Stata Corporation, College Station, Texas.

Strengths and Limitations

Despite being one of the most comprehensive representations of the U.S. population, the study’s cross-sectional nature poses certain inherent limitations. Firstly, only a small proportion of the patient population with TBI is likely to have information regarding alcohol status, thereby resulting in selection bias (Roudsari, Field et al. 2008). Nevertheless, we plan to match E.P. and EN patients with age, gender, and ethnicity, eliminating selection bias to a great extent. Secondly, blood samples are not available at the time of injury instead of at admission. The possibility of delay in hospital admission after the injury cannot be ruled out. Thirdly, the data on alcohol consumption history to exclude patients with a history of alcohol abuse may not be accurate as they are likely to be missing in patients who died immediately after admission. It is also likely that patients who provided information on alcohol consumption may have given wrong information due to recall bias associated with the injury. Fourthly, we do not plan to conduct the effect of alcohol dose on mortality in TBI due to time-delay in collecting blood samples after the injury. And lastly, NTDB sample collection is based on the individual participants and trauma centres’ informed consent, also called a convenient sample collection, the findings of our analysis planned for the present study might not be generalizable to centres not covered to NTDB. Nevertheless, we plan to stratify our patients according to disease severity, which may provide an accurate picture of centres, especially those treating less severe patients. Future multi-centric large scale prospective studies are required for overcoming several of these limitations to establish the protective effect of alcohol on TBI.

References

Asmaro, K., P. Fu and Y. Ding (2013). “Neuroprotection & mechanism of ethanol in stroke and traumatic brain injury therapy: new prospects for an ancient drug.” Curr Drug Targets 14(1): 74-80.

Coronado, V. G., L. C. McGuire, K. Sarmiento, J. Bell, M. R. Lionbarger, C. D. Jones, A. I. Geller, N. Khoury and L. Xu (2012). “Trends in Traumatic Brain Injury in the U.S. and the public health response: 1995-2009.” J Safety Res 43(4): 299-307.

Maas, A. I., B. Roozenbeek and G. T. Manley (2010). “Clinical trials in traumatic brain injury: past experience and current developments.” Neurotherapeutics 7(1): 115-126.

Menon, D., K. Schwab, D. Wright and A. Maas (2010). “Position statement: definition of traumatic brain injury.” Arch Phys Med Rehabil 91: 1637-1640.

Raj, R., E. D. Mikkonen, J. Siironen, J. Hernesniemi, J. Lappalainen and M. B. Skrifvars (2016). “Alcohol and mortality after moderate to severe traumatic brain injury: a meta-analysis of observational studies.” J Neurosurg 124(6): 1684-1692.

Roudsari, B., C. Field and R. Caetano (2008). “Clustered and missing data in the U.S. National Trauma Data Bank: implications for analysis.” Inj Prev 14(2): 96-100.

Talving, P., D. Plurad, G. Barmparas, J. Dubose, K. Inaba, L. Lam, L. Chan and D. Demetriades (2010). “Isolated severe traumatic brain injuries: association of blood alcohol levels with the severity of injuries and outcomes.” J Trauma 68(2): 357-362.