Phenobarbital as Anticonvulsant Prophylaxis in Patients With Traumatic Brain Injury at Risk for Alcohol Withdrawal Syndrome

Abstract

Background: Anticonvulsant prophylaxis (ACP) for early post-traumatic seizures (PTS) is recommended in patients with traumatic brain injury (TBI). Phenobarbital (PB) may be used to prevent alcohol withdrawal syndrome (AWS) in at-risk patients. The dual-purpose use of PB in the TBI population would allow for consolidation of pharmacotherapy. Objective: The primary objective of this study was to determine the frequency of early PTS in TBI patients at risk of AWS treated with PB as ACP. Secondary objectives included determining rates of over sedation and endotracheal intubation. Methods: Patients received an intravenous (IV) loading dose of PB at 15-20 mg/kg followed by 1 mg/kg every 12 hours for 7 days with a goal level of 15-20 mcg/mL. Medication data, seizure frequency, and episodes of over sedation and endotracheal intubation were collected. Results: Eighty patients were treated with PB over a 1-year period. Thirty-nine patients were analyzed. Median loading dose was 19.9 (Interquartile Range 19.1-20.0) mg/kg with a median post load level of 21.7 mcg/mL (IQR 18.3-25.8) mcg/mL. One patient (2.6%) had electrographic evidence for early PTS. PB was discontinued in 4 (10.3%) patients out of concern for over sedation. One patient required endotracheal intubation after rapid PB loading. Conclusion: The frequency of early PTS was low when PB was used as primary ACP in patients with TBI at risk for AWS. Over sedation is a potential adverse effect that should be considered in the choice of ACP. No conclusions can be drawn as to the effectiveness of PB in preventing AWS.

Keywords

phenobarbital traumatic brain injury post-traumatic seizures alcohol withdrawal

Introduction

In the United States, there are approximately 288,000 hospitalizations for traumatic brain injury (TBI) each year.¹ TBI may lead to short- and long-term adverse health effects including early post-traumatic seizures (PTS), which are defined as seizures occurring within the first 7 days post-injury. Early PTS may occur in up to 10.8% of patients after TBI.² Risk factors for early PTS include TBI severity, prior medical comorbidities, presence of subdural or subarachnoid hemorrhage, and alcohol abuse.^2–4 Early PTS are associated with poor long-term functional outcomes, late PTS and increased mortality.³ The 2017 Brain Trauma Foundation’s Guidelines for the Management of Severe TBI recommend the use of anticonvulsant prophylaxis (ACP) with phenytoin for the first 7 days post-injury to decrease the incidence of early PTS when the overall benefit is thought to outweigh the risk of complications.⁵ Several alternative agents have been studied for prevention of early PTS including levetiracetam, carbamazepine, valproate and phenobarbital (PB).^6–8

Alcohol intoxication is present in one-third to one-half of patients hospitalized with TBI with pre-injury alcohol use disorder present in up to 66% of cases.^9,10 A subset of patients admitted after trauma will manifest alcohol withdrawal syndrome (AWS) which is associated with increased hospital and intensive care unit (ICU) length of stay (LOS), need for mechanical ventilation and pneumonia.¹¹ The American Society of Addiction Medicine recommends preventative therapy with benzodiazepines, or PB for patients with a contraindication to benzodiazepines, in patients at risk of developing severe or complicated AWS.¹² Recent clinical trials have shown PB to have similar or improved outcomes when compared to benzodiazepines for treatment of AWS.^13–16

PB is a long-acting barbiturate that binds to GABA-A receptors, resulting in sedative effects, and inhibits NMDA receptors, blocking excitatory glutamate signaling.¹⁷ In contrast to phenytoin, PB exhibits ideal pharmacokinetics including rapid absorption and a predictable linear dose and concentration relationship.^18–21 PB has been studied as ACP for early PTS and late PTS in prospective and retrospective studies and was shown to have no benefit on the frequency of early or late PTS when used alone.^22–26 TBI patients at our institution judged to be at risk for AWS have been treated with conventional ACP including (fos)phenytoin or levetiracetam combined with intramuscular (IM) PB loading followed by tapering doses over several days. We reasoned that PB could be used for both purposes to the extent that we could achieve and maintain PB levels within a therapeutic range over the duration of prophylaxis. The primary objective of this study was to evaluate the frequency of early PTS using PB dosed to achieve therapeutic levels in patients with TBI who were adjudged to be at risk for AWS.

Methods

This study was approved by the Quality Improvement (QI) Review Committee at a level 1 adult trauma center and was exempt from institutional review board review and approval.

Patient Selection

TBI patients eligible for ACP by our institutional guidelines included those with penetrating brain injury, depressed skull fracture, acute subdural hematoma, epidural hematoma, cortical or subcortical hemispheric contusion, subarachnoid hemorrhage, or intraventricular hemorrhage regardless of severity of injury. Patients eligible for inclusion into this study were admitted to the trauma or neurotrauma ICU with TBI and judged by the clinical team to be at risk for AWS from January 1, 2021 to December 31, 2021. Patients excluded from participation included those with hepatic dysfunction defined as transaminases greater than 3 times the upper limit of normal, a known diagnosis of cirrhosis of the liver, or who had marginal respiratory status without a secure airway. The ICU team caring for the patient ordered PB when the patient was judged to be at risk for AWS primarily based on information gathered from the patient’s chart, patient and/or family report of alcohol consumption as well as prior episodes of AWS. Patients who were administered IM PB, a PB taper or who were loaded with IV PB without the acquisition of subsequent PB levels were excluded from analysis.

Intervention

Patients meeting inclusion criteria for this protocol initially received a PB loading dose of 15 mg/kg IV infused over 2 hours followed by a maintenance regimen of PB 1 mg/kg IV or enteric twice daily to complete a total of 7 days of ACP. We targeted a PB level of between 15-20 mcg/mL. On January 15, 2021, the protocol was revised to increase the PB loading dose to 20 mg/kg in response to consistently subtherapeutic PB levels with a loading dose of 15 mg/kg. PB levels were obtained 2 to 4 hours after completion of the loading dose and on days 3 and 5 following the IV load. Dosing was based on actual body weight for non-obese patients while patients with a body mass index (BMI) greater than 30 kg/m² were dosed based on adjusted body weight. Conventional ACP, if initiated, was discontinued at the time of the PB loading dose.

Data Collection and Analysis

Patient demographics including TBI imaging characteristics, 24-hour best Glasgow Coma Scale (GCS), presence of ethanol detectable in blood, and endotracheal intubation status were obtained retrospectively through review of the electronic health record (EHR). The determination of seizure frequency was based on review of clinical records for reports of clinical seizures and from electroencephalograms (EEG) in those patients with continuous or spot EEG. Over sedation was an adverse effect that was defined as an event where PB was discontinued by the clinical team out of concern for over sedation as well as any event requiring endotracheal intubation clearly attributable to PB induced over sedation. Other data of interest were recorded including initial ACP, PB dosing, levels, duration of use and the use of adjunctive sedatives during the first 72 hours of admission. Time to extubation, rates of failed extubation, ICU and hospital LOS and overall hospital mortality were also determined. Data were analyzed using descriptive statistics. Continuous data were reported as median and interquartile range (IQR) or mean ± standard deviation (SD) as appropriate while categorical data were reported as frequencies.

Results

A total of 80 patients with TBI received PB from January 1, 2021, to December 31, 2021. Forty-one patients were excluded from analysis: 24 patients received IM PB or tapering PB, 7 patients had cirrhosis or elevated liver function tests, 3 patients did not have a PB level documented and 7 patients were excluded for other reasons. Demographic and injury characteristic data is shown in Table 1. The mean age was 53.6 ± 12.3 years, and the median weight was 92 (IQR 71-101) kilograms. Twenty-six (66.7%) patients were intubated at the time of ICU admission and the median 24-hour best GCS was 11 (IQR 10-15). Eight (20.5%) patients had a 24-hour best GCS ≤8. Almost 62% (24/39) of patients had ethanol detected via blood sample on admission. EEG monitoring was employed in 19 (48.7%) patients analyzed; continuous EEG was employed in 14/19 (73.7%) patients. The most common injuries were subarachnoid hemorrhage (74.4%), subdural hematoma (69.2%), and cortical contusion (53.8%). Eleven (28.2%) patients underwent craniotomy or craniectomy.

Table 1.

Demographics and Injury Characteristics (n = 39).

Demographics
Age, mean ± SD	53.6 ± 12.3
Male sex, N (%)	35 (89.7)
Weight (kg), median (IQR)	92 (71-101)
BMI (kg/m²), median (IQR)	26.5 (23.2-32.6)
Intubated, N (%)	26 (66.7)
Detectable blood ethanol, N (%)	24 (61.5%)
Best 24-hour GCS, median (IQR)	11 (10-15)
Best 24-hour GCS ≤8, N (%)	8 (20.5)
Best 24-hour GCS 9-12, N (%)	13 (33.3)
Best 24-hour GCS 13-15, N (%)	18 (46.2)
Injury pattern
Subarachnoid hemorrhage, N (%)	29 (74.4)
Subdural hematoma, N (%)	27 (69.2)
Cortical contusion, N (%)	21 (53.8)
Skull fracture, N (%)	20 (51.3)
Intraventricular hemorrhage, N (%)	9 (23.1)
Diffuse axonal injury, N (%)	9 (23.1)
Blunt cerebrovascular injury, N (%)	5 (12.8)
Epidural hematoma, N (%)	4 (1.0)
Surgical intervention, N (%)	11 (28.2)
EEG, N (%)	19 (48.7)

BMI, Body Mass Index; EEG, Electroencephalogram; GCS, Glasgow Coma Score.

While there were no clinical seizures observed in this cohort, 1 patient (2.6%) had electrographically confirmed seizures following PB loading. This patient had an initial 24-hour best GCS of 9T and a therapeutic PB level of 16.1 mcg/mL at the time the event. PB was discontinued in 4 (10.3%) patients due to concern for over sedation. Of the 4 patients with over sedation, 1 patient had an initial PB level of 27.2 mcg/mL, all patients received sedative IV infusions in the first 72 hours of ICU admission, and one patient received an additional 1040 mg in PB boluses. Additionally of those patients with over sedation, the 24-hour best GCS ranged from 10-13 and one patient underwent surgical intervention. Two patients were intubated after ICU admission: 1 for respiratory failure and the other due to over sedation from PB as a result of rapid IV PB loading over 30 minutes instead of the protocolized 2 hours (Table 2).

Table 2.

Rates of Early Post-Traumatic Seizure, Over Sedation and Endotracheal Intubation (n = 39).

Outcome
EEG confirmed early PTS, N (%)	1 (2.6)
Over sedation secondary to PB, N (%)	4 (10.3)
Endotracheal intubation after ICU admission, N (%)	2 (5.1)

EEG, Electroencephalogram; PB, Phenobarbital; PTS, post-traumatic seizure.

The median PB loading dose was 19.9 (IQR 19.1-20.0) mg/kg with median maintenance dose of 1 (IQR .98-1.02) mg/kg every 12 hours. The median post load level was 21.7 (IQR 18.3-25.8) mcg/mL with median day 3 and 5 levels of 20.6 (IQR 18.3-24.8) mcg/mL and 21.6 (IQR 20.0-24.7) mcg/mL respectively. The median duration of inpatient PB treatment was 5.4 (IQR 2.5-6.4) days. The median time from ICU admission to PB loading was 10.9 (IQR 6.8-16.0) hours. Sixty-four percent (25/39) of patients received sedative IV infusions and thirty percent (12/39) of patients received adjunctive sedatives during the first 72 hours of ICU admission. Eighty-four percent (33/39) of patients received ACP prior to PB loading. Most of these patients received (fos)phenytoin (24/39) or levetiracetam (11/39). The mean time to first extubation was 2.9 ± 2.7 days with 5 (17.9%) patients failing extubation. The median ICU and hospital LOS was 5 (IQR 2-8) and 12 (IQR 5-19) days respectively. In-hospital mortality was 7.7% (Table 3).

Table 3.

Medication, Extubation, Length of Stay and Mortality Data (n = 39).

Outcome
PB loading dose (mg/kg), median (IQR)	19.9 (19.1-20.0)
PB maintenance dose (mg/kg), median (IQR)	1 (.98-1.02)
Post load level (mcg/mL), median (IQR)	21.7 (18.3-25.8)
Day 3 level (mcg/mL), median (IQR)	20.6 (18.3-24.8)
Day 5 level (mcg/mL), median (IQR)	21.6 (20-24.7)
Time from ICU admission to PB load (hours), median (IQR)	10.9 (6.8-16.0)
Duration of PB treatment (days), median (IQR)	5.4 (2.5-6.4)
IV sedative infusions first 72 h (N, %)	25 (64.1)
Propofol in first 72 h, n (%)	25 (64.1)
Fentanyl in first 72 h, n (%)	19 (48.7)
Ketamine in first 72 h, n (%)	2 (5.1)
Dexmedetomidine in first 72 h, n (%)	1 (2.6)
Adjunctive sedatives^a first 72 h (N, %)	12 (30.8)
(Fos)phenytoin prior to PB, (N, %)	24 (61.5)
Levetiracetam prior to PB, (N, %)	11 (28.2)
Failed extubation, (N, %)	5 (17.9)
Time to first extubation (days), mean ± SD	2.9 ± 2.7
ICU LOS (days), median (IQR)	5 (2-8)
Hospital LOS (days), median (IQR)	12 (5-19)
Hospital mortality, (N, %)	3 (7.7)

ICU, Intensive care unit; LOS, Length of stay; PB, Phenobarbital.

^aIncluding bezodiazepines, olanzapine, phenobarbital, quetiapine, and trazodone.

Discussion

This retrospective study demonstrated low rates of early PTS, over sedation and endotracheal intubation with PB administration as ACP for early PTS in TBI patients at risk for AWS. Historically, all TBI patients with previously detailed injuries at our institution regardless of admission GCS received conventional ACP for early PTS with (fos)phenytoin or levetiracetam. The data supporting PB as ACP for early PTS is limited. Rish and Caveness reported use of PB as ACP in a small subset of Vietnam combat head injury victims which demonstrated no statistical difference when compared to placebo. Only 4 percent of patients received PB monotherapy in this study.²³ With our PB protocol, we were able to achieve and maintain a therapeutic PB level around 20 mcg/mL without a high incidence of seizure. In fact, only 1 patient (2.6%) had electrographic seizures following PB loading, which is significantly lower than what was reported by Ritter and Colleagues (10.8%) and similar to that reported by Glaser and Colleagues (4.7%).^2,27 Our cohort of patients met several of the documented risk factors for early PTS as 20.5% had a 24-hour best GCS ≤8 and roughly 70% of patients had subarachnoid or subdural hemorrhages.^2,3 The rate of early PTS at our institution over a 16-year period in patients treated with conventional ACP was recently reported as 4.6%, which is similar to that reported in our study (2.6%).²⁸ Although only roughly 50% of patients in our study underwent EEG monitoring, those who did not receive EEG monitoring were following commands demonstrating a low probability of subclinical seizure.

Moreover, our data have shown that a PB-based protocol is feasible and reasonably safe. PB was discontinued early in 4 patients (10.3%) due to concern for over sedation. Two patients were intubated; however 1 patient (2.6%) required endotracheal intubation due to rapid infusion of PB loading dose over 30 minutes which violated our protocol. Out of the 4 patients with documented over sedation, 1 had an initially supratherapeutic PB level and 1 received significant additional PB boluses. This suggests that the risk of endotracheal intubation and over sedation may be decreased by following a prescribed dosing protocol and suggested therapeutic range of 15-20 mcg/mL.

Ethanol intoxication is present in up to 1 half of hospitalized patients with TBI.⁹ Severe head injury is an independent predictor of progression to delirium tremens in patients with severe AWS.²⁹ TBI patients admitted to our institution at risk of AWS historically received ACP plus IM PB followed by a tapering regimen of PB for 7 days as prophylaxis of AWS. This regimen is associated with increased drug burden, risk of over sedation and the potential for drug interactions between (fos)phenytoin and PB.³⁰ Our study did not have predefined criteria establishing patients at risk for AWS and was based solely on the judgement of ICU providers. Due to the nature of these patient’s injuries, it was difficult to obtain history of alcohol use. Thus, the effectiveness of our PB protocol for AWS prophylaxis cannot be determined. However, our results suggest it might be effective given only twelve patients (30.8%) received additional sedative boluses including benzodiazepines, PB, quetiapine, olanzapine, or trazodone. Nejad and colleagues reported over fifty percent of patients managed with benzodiazepines for AWS received additional neuroleptics including quetiapine and haloperidol.¹⁶

This retrospective case series followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) criteria; however, it has several limitations.³¹ It is a single-center study at a large academic center, which limits its generalizability. Data were retrospectively collected, and incomplete or inadequate documentation of early PTS and AWS events might have obscured important variables that potentially contributed to shortened or prolonged PB therapy or usage of additional anticonvulsants or sedatives. Additionally, due to lack of protocolized AWS diagnosis or risk screening, we could only rely on collateral history from the family when available and clinical observation. Moreover, continuous IV sedative infusions such as propofol, ketamine and dexmedetomidine might have obscured AWS symptoms. Another limitation of our study is the relatively small sample size and the lack of cohort comparisons that could have allowed us to compare the effectiveness of PB to traditional therapies. Lastly, several patients received additional anticonvulsants prior to PB loading which may have influenced the rate of early PTS.

Conclusion

Although there are important limitations, our study demonstrated that an IV PB protocol can be used as ACP in TBI patients with high risk for AWS. We cannot draw conclusions from our data as to the efficacy of IV PB in the prevention of AWS. Larger prospective studies are necessary to determine the effectiveness of this strategy in preventing both early PTS and AWS.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Cory B. McGinnis

References

Centers for Disease Control and Prevention . Surveillance Report of Traumatic Brain Injury-Related Emergency Department Visits, Hospitalizations, and Deaths—United States, 2014. Centers for Disease Control and Prevention, U.S. Department of Health and Human Services; 2019.

Ritter

Wagner

Fabio

, et al. Incidence and risk factors of posttraumatic seizures following traumatic brain injury: A traumatic brain injury model systems study. Epilepsia. 2016;57(12):1968-7.

Laing

Gabbe

Chen

Perucca

Kwan

O’Brien

. Risk factors and prognosis of early posttraumatic seizures in moderate to severe traumatic brain injury. JAMA Neurol. 2022;79:334-1. doi:10.1001/jamaneurol.2021.5420

Sodal

Storvig

Tverdal

, et al. Early post-traumatic seizures in hospitalized patients with traumatic brain injury. Acta Neurol Scand. 2022;00:1.

Carney

Totten

O’Reilly

, et al. Guidelines for the management of severe traumatic brain injury, fourth edition. Neurosurgery. 2017;80(1):6-15. 14th edition.

Schierhout

Roberts

. Prophylactic antiepileptic agents after head injury: A systematic review. J Neurol Neurosurg Psychiatry. 1998;64:108-2.

Torbic

Forni

Anger

Degrado

Greenwood

. Use of antiepileptics for seizure prophylaxis after traumatic brain injury. Am J Health Syst Pharm. 2013;70(9):759-6.

Thompson

Pohlmann-Eden

Campbell

Abel

. Pharmacological treatments for preventing epilepsy following traumatic head injury. Cochrane Database Syst Rev. 2015;2015(8):CD009900. doi:10.1002/14651858.CD009900.pub2

Corrigan

. Substance abuse as a mediating factor in outcome from traumatic brain injury. Arch Phys Med Rehabil. 1995;76(4):302-9.

10.

Dikmen

Machamer

Donovan

Winn

Temkin

. Alcohol use before and after traumatic head injury. Ann Emerg Med. 1995;26(2):167-6.

11.

Jawa

Stothert

Shostrom

, et al. Alcohol withdrawal syndrome in admitted trauma patients. Am J Surg. 2014;208(5):781-7.

12.

Lindsay

Freedman

Jarvis

, et al. Executive summary of the American society of addiction medicine (ASAM) clinical practice guideline on alcohol withdrawal management. J Addiction Med. 2020;14(5):376-2.

13.

Oks

Cleven

Healy

, et al. The safety and utility of phenobarbital use for the treatment of severe alcohol withdrawal syndrome in the medical intensive care unit. J Intensive Care Med. 2020;35(9):844-50.

14.

Hammond

Rowe

Wong

Wiley

Lee

Kane-Gill

. Patient outcomes associated with phenobarbital use with or without benzodiazepines for alcohol withdrawal syndrome: A systematic review. Hosp Pharm. 2017;52(9):607-6.

15.

Tidwell

Thomas

Pouliot

Canonico

Webber

. Treatment of alcohol withdrawal syndrome: Phenobarbital vs CIWA-Ar protocol. Am J Crit Care. 2018;27(6):454-0.

16.

Nejad

Nisavic

Larentzakis

, et al. Phenobarbital for acute alcohol withdrawal management in surgical trauma patients – A retrospective comparison study. Psychosomatics. 2020;61(4):327-5.

17.

Brunton

Chabner

Knollman

Goodman and Gillman’s . The Pharmacological Basis of Therapeutics. 12 ed. New York, NY: McGraw-Hill; 2011.

18.

Nelson

Powell

Conrad

, et al. Phenobarbital pharmacokinetics and bioavailability in adults. J Clin Pharmacol. 1982;22:141-8.

19.

Richens

. Clinical pharmacokinetics of phenytoin. Clin Pharmacokinet. 1979;4(3):153-9.

20.

Patsalos

Spencer

Berry

. Therapeutic drug monitoring of antiepileptic drugs in epilepsy: A 2018 update. Ther Drug Monit. 2018;40(5):526-8.

21.

Tangmose

Nielsen

Allerup

Ulrichsen

. Linear correlation between phenobarbital dose and concentration in alcohol withdrawal patients. Dan Med Bull. 2010;57(8):A4141.

22.

Popek

Musil

. Clinical attempt to prevent post-traumatic epilepsy following severe brain injuries in adults. Cas Lek Cesk. 1969;108:133-7.

23.

Rish

Caveness

. Relation of prophylactic medication to the occurrence of early seizures following craniocerebral trauma. J Neurosurg. 1973;38:155-8.

24.

Penry

White

Brackett

. A controlled prospective study of the pharmacologic prophylaxis of post-traumatic epilepsy. Neurology. 1979;29:600-1.

25.

Manaka

. Cooperative prospective study on posttraumatic epilepsy: Risk factors and the effect of prophylactic anticonvulsant. Jpn J Psychiatry Neurol. 1992;46:311-5.

26.

Temkin

Haglund

Winn

. Post-traumatic seizures. In: Youmans

, ed. Neurological Surgery. Philadelphia, PA: W.B. Saunders Company; 1996.

27.

Glaser

Kanter

Martinez-Camblor

, et al. The effect of antiseizure medication administration on mortality and early post-traumatic seizures in critically ill older adults with traumatic brain injury. Neurocritical Care. 2022;37:538-6. doi:10.1007/s12028-022-01531-1

28.

Pease

Gonzalez-Martinez

Puccio

, et al. Risk factors and incidence of epilepsy after severe traumatic brain injury. Ann Neurol. 2022;92:663-9. doi:10.1002/ana.26443

29.

Salottolo

McGuire

Mains

van Doorn

Bar-Or

. Occurrence, predictors and prognosis of alcohol withdrawal syndrome and delirium tremens following traumatic injury. Crit Care Med. 2017;45(5):867-4.

30.

Perucca

. Clinically relevant drug interactions with antiepileptic drugs. Br J Clin Pharmacol. 2006;61(3):246-5.

31.

von Elm

Altman

Egger

, et al. Strengthening the reporting of observational studies in epidemiology (STROBE) statement: Guidelines for reporting observational studies. BMJ. 2007;335(7624):806-8.