Butalbital is a barbiturate, most frequently prescribed in combination with acetaminophen or aspirin, and caffeine, for the treatment of migraine and tension-type headaches. Its use has waned over the years, in part because so many better remedies are available, and in part because of its reputation for habituation, rapid development of medication overuse headache, and a potentially fatal withdrawal syndrome.

This issue of Headache presents a case-controlled analysis of the associations between butalbital and a range of specific birth defects, mining data from the National Birth Defects Prevention study, which evaluates major birth defects across 10 states. Despite an analysis of 8373 unaffected controls and 21,090 case infants, it is an encouraging sign that only 73 case mothers and 15 control mothers reported periconceptional butalbital use. Of the 30 birth defect groups analyzed, statistical significance was found for 3 congenital heart defects: tetralogy of Fallot, pulmonary valve stenosis, and secundum-type atrial septal defect.[1]

The study is important despite its being underpowered by the lack of pregnant women using this barbiturate. Unfortunately, this medication is still a go-to drug for many prescribing doctors who mistakenly view it as safer than other alternatives. For those lacking prescription coverage, it is cheaper than many acute migraine treatments as well.

There are many reasons never to prescribe any butalbital combination during pregnancy or any other time. Analyzing data from the American Migraine Prevalence and Prevention Study, Bigal et al wrote a seminal paper examining the comparative risk for transformation to medication overuse headache developing from varying acute migraine medications. The authors found that across all acute medication types, barbiturate compounds led the pack in transformation risk, with an odds ratio (OR) of 1.73 (95% confidence interval [CI] 1.10–2.73), beating out even opiates, which had an OR for transformation risk of 1.44 (95% CI 1.10–2.08). The probability of developing transformed or chronic migraine occurred with only 5 days of barbiturate use per month, a remarkably low frequency of use associated with chronification, and clearly the worst provoker of rebound among all the acute migraine treatment options evaluated.[2]

Butalbital compounds carry particular risk for habituation. The barbiturate ingredient has a much longer half-life than the caffeine and acetaminophen components. There are 2 risks. First, as the shorter half-life components wear off, the headache returns, and the individual with headache is then prompted to repeat the dose, before the barbiturate has cleared the system. Second, the analgesic half-life for butalbital is in the 4-6 hour range, while the pharmacokinetic elimination half-life is from 35-88 hours.[3, 4] The barbiturate builds up, and the individual inadvertently becomes habituated to the drug with increasing blood levels, putting the patient at risk.

Monitoring butalbital usage has become increasingly difficult, as butalbital compounds have become easy to obtain over the internet. Prescription monitoring programs offered by many states may catch non-internet fills, but some of them do not routinely monitor butalbital compounds for reasons that are not clear.

A cautionary tale of problems resulting from internet purchase of a butalbital, caffeine, and acetaminophen compound was related in startling detail in a case report published by Romero et al. A patient was admitted to the hospital with intractable seizures, 48 hours after her last ingestion of the butalbital compound. She was treated with phenobarbital 100 mg 3 times per day, lorazepam, haloperidol, oxazepam, and olanzapine without apparent benefit. Finally, she required continuous intravenous midazolam for ongoing sedation until clearing on the fifth day. She had been getting prescribed butalbital for migraines, but supplemented this with unmonitored prescriptions from the internet.[5]

One of the issues to be considered in having a pregnant woman take a butalbital compound is the difficulty in handling any withdrawal issues without using medications that have potential harm to the fetus. The Food and Drug Administration (FDA) has listed phenobarbital, frequently used for prevention of these seizures, as pregnancy category D, meaning positive evidence of human fetal risk exists.

Pointing out the potential pitfalls of butalbital withdrawal, Loder and Biondi correctly pointed out that in the context of erroneous intake reported by the patient (either under- or overreporting), significant risk occurs. In the case of overreporting, patients can become intoxicated as they are tapered off the medication. In those who underreport, withdrawal seizures can occur. The authors devised a safe formula for phenobarbital loading and subsequent titration, the problems being both the lack of verifiable history of dosage intake and that phenobarbital remains a pregnancy category D medication, albeit an effective one to prevent seizures.[6] Also, if a pregnant woman goes into medication overuse headache, not only does the clinician have to create a safe wean, but also must have a plan for treatment to get the patient back to episodic migraine, an extremely difficult task during pregnancy.

Because of the many deviltries associated with butalbital compounds, including the problems with half-life, habituation, high risk of rebound, and risk of withdrawal seizures if quantities spiral out of control, this medication should not be prescribed to anyone, much less to a pregnant woman. The authors of this article recognize this, and pragmatically point out that nonetheless it is prescribed; therefore, the potential risk of birth defects needs to be studied. While they have selected a very large case-control cohort, the power of their study suffered from the lack of pregnant women using butalbital. They did find, however, that there appeared to be a risk of congenital heart defects with butalbital use, sufficient to recommend caution in its use by pregnant women, and they recommended the need for future study should this drug continue to be used. It would be ideal if there were no need for such analysis and future study. Unfortunately, I agree with their pragmatic approach, that evaluating the possible teratogenic properties of butalbital remains useful. An even better outcome would be for the FDA to recommend withdrawal of butalbital compounds from the market, given their danger up to and including lethality, and the absence of strong studies of efficacy or need.


  1. Top of page
  2. References