Phenobarbital, Propranolol, and Aggression

SIR: With great interest I read the recent article on treating aggression with propranolol.¹ As has been the experience of many of us who have worked for many years with individuals with mental retardation and other developmental disabilities, I have had several previously aggressive patients who did exceptionally well when treated with this medication. As I looked carefully, however, at the results of the treatment for the persons with epilepsy who were receiving phenobarbital at the time of the study (cases 4, 7, 11, and 12 at Creedmoor Psychiatric Center), I found that none had better than an equivocal response. (I assume that Case 12 at Creedmoor was the individual who was too violent to remain in the placement.)

To the best of my knowledge, phenobarbital is almost never used now in the “real world” except, for some reason, for people with mental retardation. My own data have shown that when people have been changed from phenobarbital to other anticonvulsant drugs, their behavior has clearly improved.² We first noticed this some years ago in an institution for people with mental retardation, when we began to simplify anticonvulsant regimens. Initially we thought that we were just stabilizing moods as we switched to valproic acid, but then we noted that anyone who was removed from phenobarbital was happier and easier to deal with, even if he or she was on phenytoin only. In an effort to remove our very strong observer bias, I went back and looked at data from years past for persons who, for one reason or another, had been taken off of barbiturates and who had a behavior program in place at the time. Without exception, we could look at the behavioral data and see when the phenobarbital taper had occurred. This factor was not noted at the time of the tapers—the behavior programs just “began working.” Behaviors most clearly affected were aggression and programmatically defined refusal to cooperate.

I note that case 4 at Creedmoor was receiving 300 mg phenobarbital a day, reportedly for complex partial seizures, as well as a quite large dose of fluphenazine. Might he not have had some degree of akathisia, as well as irritability from the barbiturate?

In my opinion the issue of adverse behavioral effects of barbiturate treatment is extremely important. A recent revision of the HCFA surveyors' manual³ on the use of psychotropic medications in facilities for people with mental retardation mandates close attention to the use of barbiturates, particularly for those persons with any sort of behavioral difficulties. I have personally yet to see anyone with behavior problems who did not improve markedly when barbiturates were removed from the treatment regimen. Perhaps the four persons in this study who were receiving barbiturates would have been less aggressive if they had been receiving a more modern antiepileptic regimen. Perhaps the propranolol would have been more effective if they had been barbiturate-free.