Most of us who prescribe benzodiazepines (BZs) have a love-hate relationship with them. On the one hand, they work quickly and effectively for anxiety and agitation, but on the other hand, we worry about sedative side effects and the fact that they can be difficult to taper because of withdrawal symptoms. We also fret about BZ dependency, tolerance, and abuse. In this article, we help guide you through these dilemmas.
First, some history. If you think BZs are problematic, consider barbiturates, the sedative of choice through the 1950s. Medications such as pentobarbital (Nembutal), secobarbital (Seconal), and phenobarbital were widely used as hypnoticsand anxiolytics for a variety of psychiatric disorders, including schizophrenia (López-Muñoz F et al, Neuropsychiatr Dis Treat 2005;1(4):329–343). While they often improved some symptoms, they were notoriously sedating, had a high abuse potential, and could easily be overdosed (Marilyn Monroe famously overdosed on barbiturates).
Benzodiazepines came on the scene as a replacement for barbiturates in the early 1960s. The first benzodiazepine, chlorodiazepoxide (Librium) was serendipitously discovered by a Roche chemist, Leo Sternbach in 1957. Diazepam (Valium) was introduced in 1963 and rocketed to stardom in the 1960s and 1970s for what was often called “anxiety neurosis,” a diagnostic category in the DSM-II. In 1981, after the publication of the DSM-III, alprazolam (Xanax) was aggressively marketed for the new diagnosis of “panic disorder,” followed eventually by clonazepam (Klonopin).
How Do They Work?
BZs work by affecting the receptor sites for GABA (gamma aminobutyric acid), which is our major inhibitory neurotransmitter. GABA normally attaches to postsynaptic GABA-A receptors, causing them to open chloride ion channels, slowing down neurotransmission. BZs attach to a specific benzodiazepine modulatory site next to GABA-A, and enhance the opening of the ion channel, essentially turbo-charging the efficacy of native GABA. This leads to decreased neuronal firing throughout the brain, which presumably then leads to its anti-anxiety effects, as well as its hypnotic, anticonvulsant, and muscle relaxant effects.
How do BZs differ from the “nonbenzodiazepines” such as zolpidem (Ambien)? It is thought that the alpha-1 subunit of the GABA-A receptor mediates sedation, while the alpha-2 subunit mediates anxiety. BZs work on both, while non-BZs work mostly on the alpha-1 (sedation) subunit. Alcohol mediates GABA receptor sites as well, but in ways that are more complex. (For a review, see Kumar S et al, Psychopharmacology (Berl) 2009;205(4):529–564.)
Which Psychiatric Disorders Do Benzodiazepines Work for?
BZs work for anxiety in all its manifestations, whether in the form of official DSM disorders, such as panic disorder, GAD, or social anxiety disorder, or in the more clinically common form of “mixed” disorders, such as “mixed depression/anxiety.” It’s nice to know which BZs are officially approved for what (if for medicolegal protection alone). Table 1 lists each drug’s official indication along with other practical tidbits, such as dosing, onset of action, mg equivalents, and clinical duration of action.
Pharmacokinetics of Benzodiazepines
The first step in drug metabolism is absorption from the gastrointestinal tract. Most BZs are swallowed and absorbed from the small intestine fairly quickly—within 20 to 30 minutes. Taking medications sublingually speeds up absorption and also sends the drugs directly to the brain, bypassing the first pass effect in the liver. While lorazepam (Ativan) is the only benzodiazepine with an official sublingual version, alprazolam is often used this way as well, and theoretically any of these medications could be dissolved under the tongue, though some will dissolve too slowly or taste too bad to make it worthwhile.
Sustained release alprazolam (marketed as Xanax XR) is encased in a fancy hyroxy-propyl-methylcellulose matrix. This allows sustained release alprazolam to be released slowly and more consistently over several hours, with benefits lasting more than 10 hours. There are randomized controlled trials (RCTs) to show this method of delivery works just as well as immediate release alprazolam for panic disorder (Pecknold J et al, J Clin Psychopharmacol 1994;14(5):314–321; Sheehan D and Raj B. Benzodiazepines. In: Schatzberg A and Nemeroff CB eds. The American Psychiatric Publishing Textbook of Psychopharmacology;2009:486). Tell your patients that eating food or taking an antacid before swallowing a benzodiazepine can slow the rate of absorption, therefore slowing the onset of action.
A common measure of the speed of metabolism is the “half-life,” defined as the time required for the body to metabolize half of the dose. But for many BZs, half life turns out to be a poor measure of how long the patient feels the effects of the medication. Consider immediate release alprazolam: the half-life of the medication is 10 to 15 hours, but in clinical practice it only feels like it works for about three or four hours. The reason is that a benzodiazepine’s actual duration of action is determined by its lipophilicity, or “lipid solubility.” Lipophilicity determines the rate a medication leaves the bloodstream and moves into fatty tissue, and it also determines how quickly a BZ crosses the blood brain barrier (Sheehan and Raj, ibid).
For example, diazepam (Valium) has a long half-life (26 to 50 hours), but because of its higher lipophilicity, it crosses the blood brain barrier more quickly than lorazepam (half life of 10 hours) and actually has a shorter duration of action clinically. Thus, diazepam’s onset of action is rapid, but its duration of action is short. The long half-life of diazepam can, however, become burdensome because it gradually accumulates in the fatty tissue and then can slowly cause more side effects when dosed long term for chronic anxiety (Sheehan and Raj, ibid).
BZs are rendered inactive by metabolism in the liver. Lorazepam, oxazepam (Serax) and temazepam (Restoril) (a useful acronym is “LOT”) are metabolized by the liver through glucuronidation. This has two important implications for clinicians: first, there are no active metabolites; and second, these drugs are rarely susceptible to drug-drug interactions. This means that the LOT drugs are especially appropriate for patients who are elderly, have cirrhosis, or have complex medical/pharmacological issues.
Drug-drug interactions. Several potential drug-drug interactions are relevant when choosing a benzodiazepine other than the “LOT” drugs. Potent inhibitors of the P450-3A4 enzyme like fluoxetine (Prozac), fluvoxamine (Luvox), and certain oral contraceptives can increase the plasma level of alprazolam and several other BZs, requiring dose reduction in some cases.
When switching from one BZ to another, refer to the dosage equivalence information in the table. A rule of thumb is to use lorazepam as the standard. Thus, 1 mg of lorazepam = 5 mg of diazepam = 0.25 mg of clonazepam = 0.5 mg alprazolam (these equivalencies are estimates and may vary from patient to patient).
Standing Doses Versus PRN
One question that often comes up in clinical practice is whether to dose these medications “standing,” ie, with a fixed schedule, or as a “PRN,” as needed. We are all tempted to use medications as a PRN for good reason: this allows patients to take the medication when they most need it and prevents too much medication from accumulating in the fatty tissues, hopefully preventing chronic side effects. On the other hand, a standing dose of a long acting medication, like clonazepam, is sometimes the best option, especially when you are starting treatment with a very anxious patient. This will predictably relieve symptoms and prevent clock-watching for the next dose.
Another important but often overlooked downside to using PRN dosing is that this may adversely affect cognitive behavioral therapy (CBT). The specific goal of CBT is to allow the patient to become more comfortable with sensations and emotions related to panic attacks and confront their automatic thoughts about how dangerous these feelings are. Reaching for a BZ, while giving a patient relief quickly, can interfere with the patient habituating to these “dangerous” feelings and sensations. It can also relieve anxiety to such an extent that the patient may lose motivation to continue the CBT (Cloos JM & Ferreira V, Curr Opinion Psychiatry;22(1):90–95).
In general, we recommend prescribing BZs as standing doses rather than PRN (or not prescribing them at all) with patients who are undergoing CBT psychotherapy for panic.
In most cases, BZs have a benign side effect profile. Patients often resist taking BZs during the day (when they most need them) because they fear sedation, but you can reassure them that this side effect is usually mild and goes awaywithin a few days. All BZs cause physiological dependence if a patient takes a high enough dose for several weeks.
“Dependence” in this context simply means that abrupt cessation may lead to withdrawal symptoms such as insomnia, anxiety, or tremor. Serious withdrawal symptoms, such as delirium tremens or seizures, are very rare among patients who have taken therapeutic doses of BZs without adding alcohol or illicit drugs.
The side effect situation is more worrisome among the elderly, who are at higher risk of falls (Woolcott JC et al, Arch Intern Med 2009;169(21):1952–1960) and delirium (Clegg A and Young JB, Age Ageing 2011;40(1):23–29) when using BZs.
In both the elderly and the young, BZs can cause cognitive impairment that can be overlooked (Barker MJ et al, CNS Drugs 2004;18(1):37–48). Most of us have had patients who stop their BZs after years of use and have an “awakening” experience of clear-mindedness. Consider tapering BZs in your patients from time to time to rule out the presence of occult side effects.
Tapering and Discontinuation of Benzodiazepines
How do you taper most successfully? BZ tapers are most successful in patients with lower baseline levels of anxiety, who have been on lower daily doses for fewer months. Regardless of the patient, the best way to taper is very, very slowly—it often takes five to six weeks but may take months. For example, one published slow-taper program for alprazolam recommends reducing the daily dose by 0.25 mg every two days for doses above 2 mg, and then reducing by 0.125 mg every two days once the patient is down to 2 mg or less. This taper schedule lasts about five weeks for patients taking a daily dose of 2 mg, and seven weeks for patients taking 4 mg per day (Otto MW & Pollack MH. Stopping Anxiety Medication. 2nd Edition. Oxford, UK: Oxford University Press; 2009). You can use this kind of schedule—about 5% reduction every two days—as guideline for other BZs. Most patients appreciate it if you write out the schedule in detail. Consider CBT during tapering for patients who have difficulty tolerating tapering and are motivated.
(Special thanks to Kate Salvatore, MD for her input on this article.)