I'm feeling a little shaky (EtOH Withdrawl)
Written by: Nathan Zarider MD, Edited by: Sunny Elagandhala
Alcohol withdrawal is a broad topic with variable management in the Emergency Department (ED). While diagnosis is mainly clinical and obviates treatment, sedative-hypnotics like lorazepam, chlordiazepoxide, and phenobarbital are well studied therapies for acute, in-patient alcohol withdrawal. Benzodiazepines can reduce seizures, delirium, and unpleasant symptoms of withdrawal. A standard practice at many EDs is safely discharging those with mild alcohol withdrawal home without any medications.
My theory for why this happens is threefold:
1) patients are deemed safe and stable for discharge, and are therefore unlikely to progress to severe withdrawal
2) stigmatization of alcohol use disorder and those who suffer from it
3) providers are uncomfortable prescribing a controlled substance with monetary street value and abuse potential to patients who are mistrusted
As someone with a family member who struggled with alcoholism, I have an inherent interest in treating each patient we discharge with the (some would call naïve) hope that I can help them quit. Therefore, I wanted to discuss gabapentin as an alternative to benzodiazepines.
Historically, antiepileptic drugs such as carbamazepine and valproic acid were once used to help with neurological symptoms of withdrawal. However, they have since fallen out of favor due to their adverse event profiles. Carbamazepine is known to cause bone marrow suppression (agranulocytosis, neutropenia, and thrombocytopenia), Stevens-Johnson’s syndrome, and has many drug interactions. Valproic acid is contraindicated in women of childbearing age (teratogenic), can cause hemorrhagic pancreatitis, and can be hepatotoxic. Both are hepatically metabolized and therefore not optimal in alcoholic patients in the ED (who may have concomitant hepatic dysfunction). Both are also dangerous in overdose.
By contrast, gabapentin is similar in structure to GABA, but its exact mechanism of action is not well elucidated. It is renally cleared and excreted in unchanged form. Its activity is not protein-mediated and there is no first-pass or hepatic metabolism. It is prescribed routinely for peripheral neuropathies and partial seizure disorders. Maximum therapeutic doses for some patients reach 1200mg three times daily (TID). In overdose, gastrointestinal absorption saturates limiting absorption of drug. Based on available poison center data, overdose syndromes are typically mild requiring observation for excess sedation [1]. These features make gabapentin preferable to alternative AEDs.
Myrick et al. 2009 [2] compared a regimen that consisted of a 4-day gabapentin taper and compared it to a 4-day Ativan fixed-dose taper. Briefly, their fixed dose tapers consisted of gabapentin 300mg or 400mg TID for the first three days, and 300mg twice daily (BID) or 400mg BID on day four, for each respective group. These were compared to a group that received lorazepam 2mg TID for the first three days, with lorazepam 2mg BID on day 4 (rescue dose packs were provided for breakthrough symptoms, see the primary source cited below for more details). Multimodal assessments were made with BAC at each follow up visit to assess for various symptoms of withdrawal. These authors noted a statistically significant improvement in CiWA scores for the high-dose gabapentin group compared to the lorazepam group. Secondary outcomes that favored gabapentin over lorazepam showed symptomatic improvement with respect to craving, anxiety, dysphoria, and somnolence.
Mason et al 2014 [3] published a double-blind, placebo-controlled trial of 150 chronic alcoholic patients treated as an outpatient with fixed-dose gabapentin for 12 weeks. Data were collected weekly, one-week post-treatment and one-year post-treatment. The gabapentin group had a statistically significantly higher rate of abstinence and less heavy drinking during the treatment period.
Before the above articles are accepted as “EBM” that supports gabapentin for alcohol withdrawal, it is important to understand numerous limitations apply to this literature. Many studied outcomes and psychiatric indices are fraught with inherently subjective reporting. This is one thing that makes alcohol withdrawal difficult to manage in practice, but also complicates literature because there are limited objective outcomes. There are newer scales being studied [4] that look at four objective measures to quantify/predict alcohol withdrawal (sweating, hallucinations, orientation, and tremors). This may help with the predicament that patients put us in when endorsing “feelings of withdrawal,” “anxiety” or vague shadows in their peripheral vision, for example.
Another limitation to keep in mind is all of those included in the above studies are not the patients we see in the ED. They’re a cohort of motivated alcoholics who desire abstinence enough to enroll in a study and follow strict follow up requirements, write in journals, and participate in lifestyle changes that go beyond stopping alcohol. Most patients seen recurrently in the ED would not be willing or capable of regular follow up and ongoing lifestyle modifications. Many of these studies reflect a similar issue with patients being lost to follow-up, which further compounds yet another limitation: small sample sizes.
Lastly, all studies used a fixed-dose taper schedule despite being shown to be inferior in the inpatient setting. Symptom-triggered therapy reduced intubations, total benzodiazepine dose, duration of treatment and hospitalization when compared to fixed-dose schedules. Unfortunately, there isn’t an easy way around fixed-dose schedules in the outpatient setting.
To summarize, trials comparing fixed-dose gabapentin and lorazepam for the outpatient management of alcohol withdrawal shows minor statistical benefits that favor gabapentin over lorazepam. The statistical significance can be misleading based on available evidence, but these studies act as proof of concept that gabapentin may be safe and effective at reducing symptoms of withdrawal. Finding the right patient is key, and there’s a large role motivational interviewing can play in identifying the right patient. In total, I have had 3 patients who I thought would be suitable candidates for outpatient alcohol withdrawal therapy. All were surprised by the effects of alcohol had on their bodies, demonstrated a desire to quit, and were accompanied by family members who were willing to provide emotional support. Their main complaint was something related to withdrawal and NOT intoxication. In these patients, chlordiazepoxide, phenobarbital, or diazepam could justifiably be sent (lets avoid naltrexone for now), but due to the cultural norms in our ED these patients were sent without prescriptions for their mild withdrawal. In these rare instances, I would consider gabapentin 400mg TID for 3 days and gabapentin 400mg BID on day 4. These patients need psychiatric and outpatient detox referral, but from my department, gabapentin is one of the few things I can ensure gets done to help someone who wants to recover.
Subjectivity: all alcohol withdrawal and psychiatric indices are fraught with inherently subjective measures. This is one of the things that makes treating alcohol withdrawal so frustrating in practice. *Stands on soapbox* One of my pet-peeves is having nurses start patients on alcohol withdrawal protocols or document a CIWA and recommend I start them on an alcohol withdrawal protocol prior to physician evaluation. The intent is noble, but alcohol withdrawal needs diagnosis before you attempt protocolized management. My threshold to start “CiWA protocol” is objective signs of withdrawal: tongue and extremity tremors, disorientation, diaphoresis, tachycardia, hypertension and hallucinosis. People who will likely be treated inpatient.
Outpatient vs inpatient difference: patients are seeking help in this study, likely to comply and are motivated to change their behavior. This is often different than our patient population in the emergency department. Queue: motivational interviewing (a topic for another post).
Fixed-dose taper: symptom triggered therapy in the inpatient setting reduces intubations, total benzodiazepine dose, duration of treatment and hospitalization when compared to fixed-dose schedules. There is no way around doing a fixed-dose taper in the outpatient setting.
Both groups had small sample sizes. Statistics drawn from this study are easily skewed toward outliers and are unlikely to catch rare events.
Take home and cautions:
Randomized trials comparing fixed-dose gabapentin and lorazepam for the outpatient management of alcohol withdrawal shows minor statistical benefits that favor gabapentin over lorazepam. These statistical differences are unlikely to be clinically significant. For me, this study functions as a proof of concept that gabapentin can improve symptoms of withdrawal.
References:
1) Klein-Schwartz W, Shepherd JG, Gorman S, Dahl B. Characterization of gabapentin overdose using a poison center case series. J Toxicol Clin Toxicol 2003; 41(1);11-5.
2) Myrick H, Malcolm R, Randall PK, et al. A double-blind trial of gabapentin versus lorazepam in the treatment of alcohol withdrawal. Alcohol Clin Exp Res 2009; 33:1582.
3) Mason BJ, Quello S, Goodell V, Shadan F, Kyle M, Begovic A. Gabapentin treatment for Alcohol Dependence: A Randomized Controlled Trial. JAMA Intern Med 2014. 174(1): 70-77.
4) Gray S, Borgundvaag B, Sirvastava A, Randall I, Kahan M. Feasibility and reliability of the SHOT: A short scale for measuring pretreatment severity of alcohol withdrawal in the emergency department. Acad Emerg Med. 2010 Oct;17(10):1048-54