IN THIS ISSUE: Posters |  Division Directors Forum | BZDs | Geriatric Psychiatry SIG 

An Alternative to BZDs in Alcohol Withdrawal Treatment

Positive results from a study on a long-acting barbiturate

Benzodiazepine (BZD)-based protocols are the most common strategies for managing alcohol withdrawal—but they may not be safe or appropriate for all patients. Phenobarbital, a long-acting barbiturate, is an alternative treatment. Research, however, offers only modest evidence regarding its safety and effectiveness in managing alcohol withdrawal syndrome (AWS) in general hospitals.

Therefore, Mladen Nisavic, MD, and his colleagues at the Massachusetts General Hospital, reviewed the records of 562 patients admitted to the hospital over a two-year period.

Mladen Nisavic, MD
Mladen Nisavic, MD

They examined AWS-related complications (seizures, alcoholic hallucinosis, and alcohol withdrawal delirium) post-treatment, and also measured hospital length of stay; intensive care admission rates and length of stay; medication-related adverse events; and discharges against medical advice.

Overall, they found that patients started on phenobarbital had similar primary and secondary treatment outcomes to those in a BZD treatment protocol.

Additionally, a subset of patients initially treated with BZDs and displaying signs of non-response to treatment, including significantly higher rates of AWS-related delirium and intensive care unit admissions, were well-managed following the transition to a phenobarbital protocol.

“This retrospective chart review lends further support to effectiveness and safety of phenobarbital for the treatment and management of AWS,” say the researchers in their paper in the September/October issue of Psychosomatics. “Further randomized controlled trials are warranted.”


The clinical diversity of alcohol withdrawal reflects the great variability in alcohol use across the population. Most patients with moderate to severe alcohol use disorder report only mild-to-moderate discomfort upon cessation of use, consistent with uncomplicated alcohol withdrawal (increased anxiety, restlessness, fine tremor, insomnia). A more complex AWS—including seizures, hallucinations, delirium, and/or autonomic instability—is seen in a small subset of patients. Alcohol withdrawal delirium, otherwise known as delirium tremens, is the most severe form of alcohol withdrawal and has historically carried a 5%-15% mortality rate if untreated, arising from multi-organ dysfunction and aspiration.

AWS is hard to detect as it shares many similarities with physiologic responses observed in severe medical or surgical illness. ”Prompt identification and management of AWS in the general hospital remains a challenge,” say the researchers.


“The treatment of AWS in patients who are intolerant or non-responsive to BZDs continues to plague our clinical management.”


Clinical effectiveness

Data shows that phenobarbital is a clinically safe, cost-effective, and easy-to-use medication. It has a notably long serum half-life of 53 to 140 hours, and thus remains effective for days following initial treatment. Much like some conventional BZDs, phenobarbital can be administered in a variety of formulations including intravenous, oral, and intramuscular, but unlike most BZDs, phenobarbital can be dosed to target a specific serum level.

“When used for AWS, phenobarbital has been commonly administered at doses below those utilized for chronic seizure management and has not been commonly associated with clinically-significant sedation despite theoretical concerns,” say the researchers.

Yet, “despite favorable qualities and common off-label use, there [has only been] modest clinical consensus as to the effectiveness of phenobarbital for AWS management, particularly in comparison to BZDs.”

The team examined the safety and effectiveness of phenobarbital compared to a conventional BZD treatment regimen. “We hypothesized that a phenobarbital-based alcohol withdrawal protocol would have comparable AWS treatment outcomes to a standard BZD-based treatment protocol. Secondarily, we hypothesized that phenobarbital would be an effective and well-tolerated alternative for management of complicated AWS in patients who could not tolerate and/or had limited response to BZDs.”

No significant differences were observed between the two treatment protocols pertaining to primary clinical outcomes, including seizures and hallucinosis. However, patients treated with BZDs had non-significantly higher rates of delirium, compared to patients initiated on phenobarbital. Similarly, no significant differences were found between groups in treatment-related side-effects, including over-sedation. Rates for other outcomes measured, including intensive care admission rates, and hospital and intensive care length of stay, were not significantly different.

“Our findings are overall consistent with the literature on phenobarbital and BZDs for alcohol withdrawal management,” say the researchers. “Despite theoretical concerns based on the long half-life of phenobarbital, overall rates of sedation appeared comparable between the two treatment groups—a finding that is consistent with prior studies. Moreover, utilization of a structured phenobarbital withdrawal protocol did not result in prolonged hospital length of stay compared to BZDs, as most patients successfully tapered off phenobarbital within five days of treatment initiation.”

Similar to other studies, “we identified a small subset of 16 patients initially treated with BZDs and transitioned to phenobarbital due to concerns for treatment-related side-effects or BZD non-response. Our data signals that despite being more medically complex and experiencing significantly worse outcomes before initiating phenobarbital (e.g., delirium rates, intensive care admission rates, and longer length of stay), these patients’ symptoms largely abated upon transitioning to phenobarbital, resulting in rapid improvement in clinical outcomes and medication-related adverse events.”

In addition, compared to most BZD-based standard withdrawal protocols, structured administration of phenobarbital (as three intramuscular injections) reduces the need for frequent assessments, thereby decreasing nursing burden. “Although initially scheduled as a seven-day taper, our results indicate that the oral component of the phenobarbital protocol can be safely abbreviated; in fact, most patients did not receive the full seven-day course of the medication. This finding is consistent with our clinical experience with the protocol since the original data collection and presents a potential point of improvement that may have further impact on hospital length of stay metrics.

“We believe these results reflect both the structure of our protocol (e.g., majority of the phenobarbital administered is provided within first 48 hours of treatment), as well as the pharmacokinetics of the medication (e.g., long half-life and prolonged clearance).”

The researchers add: “The treatment of AWS in patients who are intolerant or non-responsive to BZDs continues to plague our clinical management. Our data sheds light on this dilemma.”

Psychosomatics journal coverThe full Psychosomatics paper, Use of Phenobarbital in Alcohol Withdrawal Management  A Retrospective Comparison Study of Phenobarbital and Benzodiazepines for Acute Alcohol Withdrawal Management in General Medical Patientsis available here.


« « All 2019 ACLP News « October 2019 ACLP News