Keywords
alcohol dependence - head injury outcome - Glasgow Outcome Score
Introduction
According to the World Health Organization,[1] 370,000 deaths due to road traffic injuries result from harmful use of alcohol and
40% of injuries is attributable to alcohol as measured in disability-adjusted life
years (DALYs). Globally 2.3 billion people are current drinkers. Alcohol-associated
traumatic brain injury (TBI) is a major burden on health systems, caregivers, and
family members.
Alcohol intoxication as a risk factor for TBI and its related morbidity and mortality
has been studied by many with conflicting reports,[2] but there are not many studies on chronic alcohol-dependent patients with TBI.
Chronic alcohol consumption affects multiple systems. It may cause withdrawal effects,
thus increasing the risk of TBI. It leads to chronic malnutrition and decreases immunity
and therefore increases the risk of infection. Liver cirrhosis may lead to coagulopathy
with increased risk of intracranial bleeding.[3]
[4]
[5]
In this study, we aim to study the pattern of TBI and its outcome in alcohol-dependent
patients without the confounding effect of intoxication.
Alcohol Use Disorders
The term “alcoholism” or “alcohol addiction” has been replaced with the term “alcohol
use disorder” (AUD) in the current Diagnostic and Statistical Manual of Mental Disorders-5
(DSM-5). An individual with AUD (also known as alcohol dependence) is someone having
problems in multiple life areas due to prolonged continued use of alcohol. DSM-5 defines
AUD as repeated alcohol-related difficulties in at least 2 of 11 areas.[4] Anyone meeting 2 to 3 of 11 criteria would be receive a diagnosis of mild AUD. A
person would be diagnosed with moderate AUD if they met 4 to 5 of the 11 criteria,
while anyone meeting ≥6 of 11 criteria would receive a diagnosed of severe AUD.
Identification of patients with AUD is done using standardized questionnaires including
the following:
The CAGE questionnaire consists of four questions:
-
Have you ever had to Cut down on alcohol amount?
-
Have you ever been Annoyed by people's criticism of alcoholism?
-
Have you ever felt Guilty about drinking?
-
Have you ever needed an Eye opener drink (early morning drink)?
A score of 2 or more is indicative of an AUD.
WHO reports the per capita alcohol consumption in India in persons older than 15 years
increased from 4.3 L in 2010 to 5.7 L in 2016. The prevalence of episodic heavy drinking
(>60 gm alcohol in 30 days) in India was 28.4% in males and 5.4% in females. The prevalence
of AUD was 9.1% in males and 0.5% in females.[1]
Studies of Alcohol in Head Injury
Experimental animal studies of TBI have shown conflicting results. Some studies have
reported zero effect, some reported a negative effect, and some have shown neuroprotective
effects of alcohol. Likewise clinical studies have also shown varied results.[2]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
Clinical Studies of the Effect of Alcohol on Outcome in TBI
Older studies in patients with TBI showed negative effects of alcohol in higher doses
(>230 mg/dL),[15]
[16] while others showed no significant difference[17] or decreased mortality.[18]
The limitations of these clinical studies are that they are retrospective, as prospective
studies with alcohol use would be unethical. Also, the study population is heterogeneous
with different inclusion criteria causing selection bias.
Recent Studies
Recent studies on patients with TBI including a meta-analysis found a decreased mortality
associated with alcohol consumption although it was not significant,[19]
[20] while other meta-analyses showed poorer cognitive outcomes in the alcohol group.[21]
[22] The possible mechanisms of alcohol-related neuroprotection could be inhibition of
excitatory N-methyl-D-aspartate (NMDA) receptors and suppression of sympathetic response.[23]
[24]
[25]
[26]
[27]
[28]
Materials and Methods
In our prospective observational study, we studied the pattern of TBI in alcohol-dependent
patients admitted to a tertiary care center over 6 months. All the patients with head
injury admitted to our center were screened using the CAGE questionnaire to identify
alcohol dependence. A score of 2 or more on the CAGE criteria identified alcohol-dependent
patients. Approval of the institutional review board was obtained for the study.
The inclusion criteria were the following:
The exclusion criteria were the following:
History regarding the mode of injury was noted. The admission Glasgow Coma Scale (GCS),
vitals, and neurological deficits were documented. Associated injury to the spine,
chest, abdomen, and limbs were noted.
The admission computed tomography (CT) brain findings were noted. The admission Rotterdam
score was calculated as follows: basal cisterns—normal (0), compressed (1), and absent
(2); midline shift— < 5 mm (0) and >5 mm (1); subarachnoid hemorrhage (SAH)/intraventricular
hemorrhage (IVH)—absent (0) and present (1); epidural hematoma (EDH)—present (0) and
absent (1). +1 was added to get the total score out of 6. The patients with GCS score
<8 were tracheotomized.
Investigations
Hemoglobin and complete blood count along with complete liver function test were evaluated.
Coagulation workup including prothrombin time (PT)/international normalized ratio
(INR), activated partial thromboplastin time (aPTT), and clotting time (ct), if PT/INR
could not be done, was done. A clotting time of 2 to 8 minutes was considered normal.
The admission abdominal ultrasound was done to assess the liver morphology (fatty
liver, cirrhosis, etc.).
Treatment
All the admitted patients were treated according to the 2016 Brain Trauma Foundation
guidelines. Patients having deranged coagulation profile (PT/INR or ct0) were given
fresh frozen plasma (FFP). Platelets were transfused to patients with thrombocytopenia.
Patients requiring emergency intervention depending upon GCS and CT brain findings
were taken to appropriate emergency procedure. Patients not requiring emergency procedure
were managed conservatively. GCS monitoring was done in all the patients.
Postoperative scan was done in all the operated patients and the findings were documented
as good evacuation/insignificant residual bleed/significant residual or recurrent
bleed. Patients were taken for re-exploration, if required, based on the post-op brain
CT findings.
In patients on conservative treatment, a repeat brain CT was done at 6 and 48 hours
to document any change in CT findings such as increase in size of contusion, bleeding,
or edema. Patients were taken to appropriate emergency procedures, if required, depending
on the findings of the follow-up scan.
All the admitted patients were monitored for alcohol withdrawal symptoms and the following
complications: wound infection, meningitis, chest infection, hyponatremia, recurrent
convulsions, status epilepticus, progressive jaundice and liver failure, massive gastrointestinal
(GI) bleed, and hepatic encephalopathy.
At discharge, the patients GCS score and outcome using the “Extended Glasgow Outcome
Scale” (EGOS) was documented. The outcome was reassessed using EGOS at 1-month follow-up.
Observations and Results
Sixty alcohol-dependent patients with TBI were included in our prospective observational
study after screening for alcohol dependence.
All the patients in the study were males.
Age
Age distribution was as follows. The minimum age was 22 years, the maximum age was
80 years, and the average age was 47.8 years.
Duration of Alcohol Intake
Duration of alcohol intake was grouped in 5-year intervals and ranged from 5 to 30
years. The mean was 15.3 years, with most patients in the 15- to 20-year group. Increased
duration of intake was associated with decreased EGOS at 1 month (p = 0.61; [Table 1]).
Table 1
Duration of alcohol intake among patients stratified by 5-year intervals
|
Alcohol intake (y)
|
Number of patients
|
%
|
|
5–10
|
19
|
31.6
|
|
11–15
|
16
|
26.6
|
|
16–20
|
18
|
30
|
|
21–25
|
6
|
10
|
|
26–30
|
0
|
0
|
|
31–35
|
1
|
1.6
|
|
Total
|
60
|
|
Alcoholic Liver Disease
Previous evidence of alcoholic liver disease (ALD): 11 patients (20.8%) were diagnosed
as having ALD with a history of hospital admission for jaundice, ascites, and upper
or lower gastrointestinal bleed. In all, 88% of the patients were daily drinkers.
Mode of Injury
The most common form of injury was fall on level ground (32.1%), followed by seizure-related
falls (28.3%) and vehicular accidents (26.4%). Other modes were fall from height and
unknown.
Duration of Hospital Stay
The average duration of stay was 8.57 days, with a median of 8 and maximum of 35 days.
Head Injury Severity
In all, 39.6% of the patients had a severe head injury, 39.6% had a moderate head
injury, and only 20.8% of the people had a mild head injury.
Admission Rotterdam score
Increasing adjusted Rotterdam score related to reduced EGOS at 1 month (p ≤ 0.001; [Table 2]).
Table 2
Adjusted Rotterdam score among patients at admission
|
Adjusted Rotterdam score
|
Number of patients
|
%
|
|
1
|
7
|
13.2
|
|
2
|
12
|
22.6
|
|
3
|
11
|
20.8
|
|
4
|
15
|
28.3
|
|
5
|
5
|
9.4
|
|
6
|
3
|
5.7
|
|
Total
|
53
|
|
Treatment Received
In all, 30.2% of the patients were managed conservatively. The rest were operated
on either immediately (41.5%) or within 48 hours (13.2%). One patient refused consent.
Burr hole evacuation was done in 14.3% of the patients, 28.6% patients had decompressive
craniotomy, and 57.1% had craniotomy and evacuation.
Investigations
Blood investigations showed elevated liver function in 18 (38.3%) patients, thrombocytopenia
in 12 (23.5%) patients, prolonged prothrombin time in 20 (33.3%) patients, and deranged
clotting time in 10 (16.6%) patients.
Complications
Recurrent convulsions were seen in 25% of the patients, chest infection in 21.2% of
the patients, meningitis in 3.3% of the patients, and encephalopathy in 3.33% of the
patients.
Hospital Course
In the mild head injury group, all the patients improved. No one worsened or died.
In the moderate head injury group, 71.4% improved, 9.5% remained the same, 14.2% worsened,
and 4.7% died. In the severe head injury group, 38.0% improved, 61.9%worsened, and
47.6% expired.
Extended Glasgow Outcome Scale
The EGOS was measured at discharge and at 1 month from injury. The distribution is
shown in [Table 3].
Table 3
Extended Glasgow Outcome Score assessed at discharge and at 1 month after discharge
|
EGOS at discharge
|
EGOS at 1 mo
|
|
Death
|
13 (25%)
|
0
|
|
Vegetative state
|
6 (11.5%)
|
3 (7.5%)
|
|
Lower severe disability
|
8 (15.4%)
|
5 (12.5%)
|
|
Upper severe disability
|
6 (11.5%)
|
5 (12.5%)
|
|
Lower moderate disability
|
12 (23.1%)
|
3 (7.5%)
|
|
Upper moderate disability
|
6 (11.5%)
|
8 (20%)
|
|
Lower good recovery
|
0
|
4 (10%)
|
|
Upper good recovery
|
0
|
10 (25%)
|
|
Lost to follow-up
|
1 (1.9%)
|
2 (5%)
|
|
Total
|
52
|
40
|
Abbreviation: EGOS, Extended Glasgow Outcome Scale.
[Table 3]
Statistical analysis was performed using Jamovi open-source statistical software and
the following analysis was obtained.
Relation of EGOS at 1 Month to the Presence of ALD
[Fig. 1] shows the relation of ALD to outcome at 1 month via a regression plot.
Fig. 1 Regression plot showing relation of alcoholic liver disease to outcome at 1 month.
There is poor outcome in the ALD group.
The presence of ALD showed a tendency toward lower EGOS at 1 month.
Duration of Alcohol Use in Relation to EGOS at 1 Month
Increasing duration of alcohol use resulted in linear decline in EGOS at 1 month across
all severity classes.
[Fig. 2] shows the relationship between the duration of alcohol use and outcome at 1 month.
Fig. 2 Linear regression plots showing the relation of the duration of alcohol intake to
outcome at 1 month, combined and separately for each severity category.
Adjusted Rotterdam Score vs. EGOS
[Fig. 3] shows the relationship between adjusted Rotterdam score and outcome at 1 month.
Fig. 3 Graph showing the relation of the adjusted Rotterdam score with outcome at 1 month.
Showing a linear trend.
Increasing adjusted Rotterdam score significantly correlated with poor outcome at
1 month (p ≤ 0.001).
Effect of Re-Exploration
There was no significant effect of the duration of alcohol intake on re-exploration;
however, re-exploration showed a slight improvement in EGOS at 1 month.
Effect of Coagulation
There was a significant difference in outcome, with coagulopathy resulting in lower
EGOS scores ([Table 4]). Increased duration of alcoholism did not result in significant increase in coagulopathy.
Table 4
Significant relation of outcome at 1 month to abnormal coagulation
|
EGOS at 1 mo (EGOS1M) between normal and abnormal coagulation
|
|
|
Statistic
|
df
|
p
-Value
|
|
EGOS1M
|
Student's t-test
|
–5.62
|
56.0
|
<0.001
|
Abbreviation: EGOS, Extended Glasgow Outcome Scale.
Effect of Deranged Liver Function Test
The presence of deranged liver function test (LFT) significantly reduced the EGOS
score at 1 month ([Table 5]). Increased duration of alcohol use resulted in higher possibility of elevated LFT
insignificantly.
Table 5
Significant relation of outcome at 1 month to elevated LFT
|
EGOS at 1 mo (EGOS1M) between normal and elevated LFT
|
|
|
Statistic
|
df
|
p
-Value
|
|
EGOS1M
|
Student's t test
|
–5.76
|
50.0
|
< 0.001
|
Abbreviations: EGOS, Extended Glasgow Outcome Scale; LFT, liver function test.
Platelets and Outcome
Increasing alcohol use was associated with increased thrombocytopenia. The presence
of thrombocytopenia was also associated with poorer outcome at 1 month. These results
did not reach statistical significance.
Complications
Increased duration of alcohol intake was associated with increased risk of infection
(chest infection was the most common infection). There was increased risk of seizures
with increasing duration of alcohol use.
Relation of Computed Tomography Lesion
The distribution of various lesions like acute subdural hematoma, chronic subdural
hematoma, extradural hematoma, lobar hematoma, and single or multiple contusions did
not change with respect to the duration of alcoholism or coagulopathy or the presence
of liver dysfunction.
Discussion
Chronic alcoholism is common in several parts of the country including Goa.[29] Dhupdale et al found 10.2% of the participants were heavy drinkers and 9.5% were
exconsumers.[30] Alcoholism may predispose to head injury on account of high-risk behavior, withdrawal
symptoms, poor neuromuscular coordination, and impaired judgment.[31] It is also known to affect multiple organ systems, including the liver and immune
and hematopoietic system,[3]
[32] and may affect the outcome after head injury independently.
Effects of Alcohol
The effects of alcohol on the central nervous system are mediated by gamma-aminobutyric
acid (GABAA) receptor stimulation and inhibition of postsynaptic NMDA receptors. Chronic consumption
leads to upregulation of NMDA receptors, which may be the cause of withdrawal symptoms
through increased glutaminergic activity with decreased GABA stimulation.[4]
Long-term alcohol use affects the central and peripheral nervous system. There is
impaired higher mental function and poor sleep quality. Cerebellar atrophy causing
ataxia and nystagmus develops in 1% of cases. Wernicke–Korsakoff syndrome develops
in 1 of 500 individuals consisting of ophthalmoplegia, ataxia, encephalopathy, and
severe retrograde and anterograde amnesia. Peripheral neuropathy develops in 10% of
individuals.[4]
Chronic alcoholism causes impaired gluconeogenesis and fatty acid oxidation, leading
to fatty liver changes eventually causing perivenular sclerosis and cirrhosis.[4]
Folic acid deficiency associated with chronic alcoholism causes macrocytic anemia.
White blood cells (WBCs) have reduced counts, lower motility, and adherence, leading
to immune suppression.[4]
Jurkovich et al in 1993 reported that patients with chronic alcoholism develop chest
infections more commonly compared to those without.[5]
Animal Studies
Opreanu et al described the observations of various experimental and clinical studies
evaluating the effect of alcohol on TBI.[2]
Animal models have shown different effects of alcohol at low to moderate dose (<100 mg/dL)
compared to high dose (>100 mg/dL) on TBI.[2] Türeci et al,[6] Gottesfeld et al,[7] Kelly et al,[8] and Taylor et al[9] have evaluated the effects of low dose alcohol, and suggest a neuroprotective role
of alcohol in TBI. In contrast, experimental studies in which high alcohol dose were
used have shown deleterious effects on TBI.[2] Zink et al,[10]
[11]
[12] Katada et al,[13] and Yamakami et al[14] showed significant increase in mortality and neurological deficit after high alcohol
dose.
Recent Studies in Traumatic Brain Injury
Albrecht et al[19] performed a retrospective analysis of patients with TBI. They found higher blood
alcohol levels were associated with decreased risk of mortality compared to low level
of blood alcohol, but this association was not statistically significant. Raj et al[20] performed a meta-analysis of 11 studies. They found positive blood alcohol level
was associated with decreased risk of mortality, although this analysis was confounded
by heterogeneity. Unsworth and Mathias performed a meta-analysis of observational
studies comparing outcome after TBI. They reported a slightly poorer cognitive outcome
in those who had consumed alcohol before injury compared those who did not and poorer
performance in all cognitive domains in the high alcohol level group than in the low
alcohol level group.[21]
Possible mechanisms for neuroprotection from low doses of alcohol as seen in studies
are the following:
-
Inhibition of postsynaptic excitatory NMDA receptors. In experimental studies, pharmacological blockade of NMDA receptors improves brain
metabolic status and decreases neuronal damage and dysfunction. However, clinical
trials have failed to show the benefit of NMDA receptor antagonist possibly due to
a nonselective antagonist, which blocked both synaptic and extrasynaptic NMDA receptors
as reported by Hardingham et al.[22] Another explanation for failure, as given by Ikonomidou and Turski,[23] was “short therapeutic window” of effect. NMDA receptor activation and further progression
of injury was maximum during the initial 1 hour after TBI.
-
Suppression of sympathetic response. TBI causes initial surge of catecholamine followed by a prolonged state of elevation.
Alcohol causes suppression of sympathetic response. Hamill et al[24] found that patients with severe head injury had fivefold increase in catecholamine,
and persistent elevation correlated with poor outcome. Woolf et al[25]
[26] showed similar results. Ley et al[27]
[28] showed improved cerebral perfusion and decreased edema with the use of propranolol
in experimental animal studies. Retrospective clinical studies have also shown similar
benefits of propranolol in TBI.
Our study excluded intoxicated patients to avoid the confounding effect of alcohol
on TBI in the acute setting.[15]
[16]
[17]
[18]
[19]
[20]
[21] Patients with a history of alcohol dependence were included.
In our study, we found that outcome at 1 month as assessed by EGOS was correlated
with head injury severity at admission with significant result. There was also significant
correlation with the adjusted Rotterdam score at admission, which is consistent with
other studies.[33]
[34]
[35] Regression analysis showed a high Rotterdam score to be a predictor of poor outcome
at 1 month.
Elevated LFT and coagulopathy resulted in poorer EGOS scores at 1 month with significant
results (p < 0.05). Irrespective of the lesion type, there was a reduction of EGOS in those
with coagulopathy. Thus, aggressive correction of coagulopathy could improve outcome
in these patients.
Duration of alcohol use showed a linear trend toward a poorer EGOS at 1 month. This
trend was independent of injury severity, being consistent in all severity groups.
The numbers did not reach statistical significance; however, regression analysis showed
a clear trend. Increased duration of alcohol use also resulted in higher incidence
of coagulopathy, occurrence of thrombocytopenia, perioperative seizures, and requirement
of a second surgery. These results were not statistically significant.
Chest infection occurred in 21% of patients and was significantly correlated with
deranged liver function. However, it did not correlate with outcome at 1 month. Jurkovich
et al also found increased chest infection in alcoholics with head injury.[36]
Conclusion
It is difficult to assess the effects of chronic alcoholism, separate from acute intoxication,
as there is no measure like blood alcohol level that can be used. A detailed patient
history is important and hard to obtain.
What is clear, however, is that the outcome of head injury in chronic alcoholics follows
the usual trend with respect to head injury severity and CT scan at admission. In
addition, alcoholics with head injury have a tendency to poorer outcome with increasing
duration of alcoholism. This is seen across all levels of severity. This may be relevant
in prognosis and counseling. There is also increased incidence of complications, which
have a bearing on outcome. Some of the complications like coagulopathy and liver failure
are manageable with aggressive critical care and since they have a significant bearing
on outcome, their timely management may positively alter the outcome.
