Does phenobarbital add benefit in treating alcohol withdrawal syndrome?

March 29, 2013, 7:43 pm


Phenobarbital for Acute Alcohol Withdrawal: A Prospective Randomized Double-Blind Placebo-Controlled Study. Rosenson J et al. J Emerg Med 2013;44:592-598.


This well done study looked primarily at the effect of adding phenobarbital to standardized symptom-triggered therapy (using lorazepam). The primary outcome was the rate of ICU admission.

One hundred and two patients were randomized to receive either the standard treatment (N=51) or standard treatment plus an initial dose of IV phenobarbital (10mg/kg). Although baseline characteristics were similar between the two groups, the placebo group had a 10% higher incidence of altered mental status.

The results showed that the group that received phenobarbital had significantly fewer ICU admissions (8% v 25%). There were no significant differences in the need for intubation (1 patient in each group) or incidence of other adverse outcomes.

Of course, this study took place at one institution (Highland Hospital in Oakland) that was following a very specific treatment protocol for alcohol withdrawal syndrome. It is not clear whether it would be at all generalizable to other settings. The authors note that phenobarbital might add benefit because it has a longer duration of action than does lorazepam. It is possible that using a longer-acting benzodiazepine such as diazepam might give the same advantage without complicating the treatment regimen.




  1. BAF Says:

    Do you really think that adding in different BZDs is beneficial? Adding in a different agent within the same class, just because the duration of action is longer, ie adding diazepam to lorazepam, seems rather pointless to me. I have heard this from a few different people, though, and would love to hear the argument in favor of it. I was very impressed with this study when I saw it and think it might be practice changing, especially if confirmed in a larger study.

  2. Leon Says:


    Thanks for the comment and giving me the opportunity to clarify my thoughts about the paper by Rosenson et al. I agree that when treating alcohol withdrawal syndrome with a benzodiazepine, it’s best to stick to a single agent, and not confuse the clinical pictures with multiple drugs having different onsets and durations. So I’m not at all in favor of using a farrago of different benzos. But in choosing a single agent to use, might it make more sense to use the longer-acting diazepam, and would that obviate the need for adding a dose of barbiturate.

    In any case, my main point was that the Rosenson paper looked at adding phenobarbital to a very specific treatment regimen using lorazepam. It is not at all clear to me that the clinical benefit without added risk their data suggest would hold up with different treatment algorithms using different sedative agents. To be fair, the authors discuss this limitation in the paper.