Introduction

Before the modern era of psychopharmacology, patients seeking relief of anxiety symptoms were likely to receive one of an array of chemicals that act as generalized central nervous system (CNS) depressants. Since antiquity, alcoholic beverages and forms of laudanum, such as tincture of morphine, were commonly used to alleviate anxiety symptoms and to treat insomnia. Bromide, the first specific sedative-hypnotic agent, was introduced in the mid-19th century. Chloral hydrate and paraldehyde were discovered subsequently, followed by barbital and phenobarbital in the early 1900s. Because these sedative-hypnotic agents are nonspecific CNS depressants, they share a liability for developing tolerance and physical dependence. Furthermore, overdoses are potentially lethal while abrupt withdrawal following chronic administration can precipitate seizures.

Chlorpromazine and meprobamate, much more specifically-acting agents discovered in the 1950s, were found to have taming effects in animals without producing marked motor impairment. Thus, the pharmacologic potential for specific anxiolytic drugs lacking liability for depressing CNS function overall became a reality. Discovery of chlorpromazine and meprobamate also fostered development of more sophisticated laboratory methods for screening CNS-active compounds for desirable behavioral effects, including, for example, the conditioned avoidance test for potential anti-anxiety activity of new compounds.

The Era of Benzodiazepines

The era of benzodiazepines was ushered in when Randall and colleagues¹ discovered the uniquely specific anxiolytic but nonsedating profile of chlordiazepoxide, synthesized in 1957 by Sternbach.² Several thousand benzodiazepines have since been synthesized, with at least 35 compounds in clinical use around the world. While drugs of the benzodiazepine class possess varying admixtures of sedative-hypnotic, anticonvulsant, and muscle relaxant properties, they all share a core anti-anxiety profile. At least in principle, one benzodiazepine can substitute for another for treating anxiety symptoms or for preventing the drug withdrawal syndrome resulting from abrupt discontinuation.

Benzodiazepines have virtually no pharmacologic effects outside the CNS. Prominent effects include anxiolysis, sedation, muscle relaxation, anticonvulsant activity, and retrograde amnesia. All benzodiazepines act by promoting specific binding of g-aminobutyric acid (GABA), a major inhibitory neurotransmitter, to receptors widely distributed throughout the CNS. Because benzodiazepines bind specifically to the GABA_A subset of GABA receptors unless at high dosage, they do not depress all neuronal activity, unlike the sedative-hypnotic class of drugs and general anesthetics. By not inducing pronounced sedation or causing fatal CNS depression in overdose (except in combination with alcohol), the benzodiazepines rapidly gained acceptance and displaced older sedative-hypnotics with less favorable safety profiles from clinical practice.

This advantageous pharmacology, coupled with efficacy in relieving anxiety symptoms demonstrated in clinical trials, led to immense popularity and widespread prescription of benzodiazepines. Diazepam soon replaced chlordiazepoxide as the leading anti-anxiety drug because of greater potency and improved pharmacokinetic properties. Predictably, due to their popularity, other benzodiazepines of varying potency and half-lives followed the discovery of diazepam, with indicated uses dictated by their pharmacokinetic profile. While in principal the majority of benzodiazepines are interchangeable, a short elimination half-life favors hypnotic or anesthetic use rather than treatment of anxiety; a long half-life is important for an anticonvulsant indication, with rapid entry into the brain a key feature for treatment of status epilepticus. Nonbenzodiazepine drugs that are similarly acting but bind to a subunit of the GABA_A receptor, such as zolpidem and zaleplon, have recently gained widespread therapeutic acceptance as hypnotics, probably more due to the fact that they undergo rapid oral absorption and have a short half-life rather than because of unique pharmacology.

Alprazolam, a highly potent benzodiazepine with a shorter half-life than chlordiazepoxide and diazepam, replaced diazepam as the most widely prescribed psychotropic drug in the 1980s. In addition to proven efficacy in anxiety disorders, it also underwent study as an antidepressant by the pharmaceutical sponsor. Findings in a series of clinical trials suggested that alprazolam might have intrinsic antidepressant utility because it showed superiority compared with placebo on Hamilton Rating Scale for Depression (HAM-D) scores. In placebo-controlled trials also employing the standard antidepressant imipramine as an active control, alprazolam produced earlier symptomatic relief than the antidepressant; however, by the end of the treatment period (6 or 8 weeks), improvement in HAM-D ratings still tended to be inferior to those of the antidepressant control. It was hypothesized that effects primarily on anxiety and somatic items of the HAM-D rather than on core symptoms of depression accounted for the apparent superiority of alprazolam over placebo. Thus, the intrinsic antidepressant activity of alprazolam was not conclusively established. Because of the often indiscriminate use of benzodiazepines for presenting anxiety symptoms without patients undergoing proper psychiatric diagnosis, concerns emerged that benzodiazepine treatment might actually mask or worsen underlying depression in patients with a primary mood disorder.³

Liabilities of Benzodiazepines

While benzodiazepines supplanted the sedative-hypnotics in medical practice because of their effectiveness, benign side effects, and extreme safety in overdosage, the supposition that they produced little dependence or abuse potential was called into question with widening clinical experience. It became evident that long-term benzodiazepine treatment, in fact, had significant and largely unforeseen liabilities. Tolerance (diminished effects with chronic use requiring increased doses for clinical benefit) as well as other untoward effects emerged as concerns of governmental regulatory agencies in several countries.^4,5 Studies have shown that chronically anxious patients treated with benzodiazepines experienced withdrawal symptoms (often unrecognized) with dose reductions that were distinguishable from recrudescence of their original anxiety syndrome.

Drug dependence, either psychic or physical, is characterized by a compulsion to take drugs on a continuing basis to ameliorate anxiety symptoms, or perhaps to avoid the physical discomfort from the absence of drugs. While there is a propensity for rapidly developing dependence with the sedative-hypnotic class of drugs, especially at high doses, what surprised the medical profession were occurrences of dose dependency at therapeutic doses of benzodiazepines. A small but significant minority of patients with chronic anxiety became very dependent, as evidenced by a need to maintain daily dosing of benzodiazepine and extreme difficulty in weaning off therapy after a few months. Withdrawing benzodiazepines in such patients proved to be clinically challenging. Experience further showed that secondary drug dependence (multiple drug abuse) occurred frequently within the context of alcoholism so that benzodiazepine use in this patient population should be undertaken cautiously.

Despite the established effectiveness and popularity of alprazolam for treatment of anxiety, there were similar concerns about its widespread use in clinical practice. It became apparent that some patients experienced interdose interval anxiety precipitated by the relatively short half-life of alprazolam and resultant fluctuating plasma levels, which complicated clinical management. The often higher doses of alprazolam used in treating panic disorder carried a liability for potential psychomotor impairment and physical dependence as well as seizure risk with abrupt discontinuation.

With the ultra-short–acting benzodiazepine hypnotics, evidence emerged that they could cause a rebound terminal insomnia as well as transient amnesic episodes and daytime exacerbation of anxiety symptoms. Professional and regulatory bodies in assessing appropriate therapeutic uses issued guidelines for clinical use, including recommendations for using lower doses of benzodiazepines and minimizing long-term therapy.

The advent of the selective serotonin reuptake inhibitors (SSRIs) with their broad acceptance in medical practice has significantly changed patterns of benzodiazepine prescribing. Both short-term and long-term benzodiazepine use has declined from that of prior decades, in large part supplanted by a corresponding increased use of SSRIs (and other antidepressants), which have established efficacy not only for depression but also for such diverse disorders as bulimia, premenstrual dysphoric disorder, panic disorder, generalized anxiety disorder, social anxiety disorder, obsessive-compulsive disorder, and posttraumatic stress disorder.

Benzodiazepine Prescribing and Substance Abuse Disorder in Patients with Severe Mental Illness

A recent study⁶ examining prescribing patterns in patients with severe mental illnesses indicated that benzodiazepines are widely prescribed in this patient population, for both short-term and extensive long-term use. Epidemiologic and household surveys sponsored by the National Institute of Mental Health (NIMH) more than a decade ago had suggested that most medical uses of benzodiazepines were appropriate, with misuse infrequent and largely confined to patients with a history of substance abuse.⁷ Subsequent studies questioned these NIMH study findings and raised concerns about potential undesirable consequences of long-term treatment with benzodiazepines.^8,9

Investigators at Dartmouth Medical School⁶ studied benzodiazepine use within New Hampshire over a 5-year period among Medicaid beneficiaries with a diagnosis of schizophrenia, bipolar disorder, or major depressive disorder (MDD), and compared prescription rates to that of other psychiatric diagnoses. Patients with Alzheimer’s disease, dementia, or mental retardation were excluded from the survey. The authors examined outpatient benzodiazepine prescribing patterns by diagnostic group and by presence of comorbid substance use disorder (SUD). To assess whether persons with SUD were prescribed lower-risk benzodiazepines, as recommended under current guidelines, they also evaluated the prescription of “fast-acting/high potency” benzodiazepines (defined as alprazolam, estazolam, and triazolam) as well as duration of treatment.

The authors found that patients with serious mental disorders were more likely to be prescribed benzodiazepines (>60%) than patients with other psychiatric diagnoses (<50%). Bipolar patients had the highest rate of benzodiazepine prescription, followed by patients with MDD, who were also more likely to receive one of the fast-acting/high potency benzodiazepines. Long-term benzodiazepine treatment (>4 months) was more frequent for patients with bipolar disorder or MDD than for schizophrenic patients. Across all diagnoses, patients with comorbid SUD were more likely (P<.0001) to receive a benzodiazepine than those without diagnosed substance abuse. Irrespective of diagnosis, there was a trend for patients with comorbid SUD to receive higher doses of benzodiazepines than those without comorbid SUD, although the difference was not statistically significant. Use of benzodiazepines for detoxification of patients with comorbid SUD did not account for the higher utilization in these patients.

The investigators concluded that comorbidity of SUD and severe mental disorder were associated with very high rates of benzodiazepine prescribing (>50% of patients). Long-term benzodiazepine use (daily dosing for >4 months) exceeding American Psychiatric Association panel recommendations was also frequent. Among patients with comorbid SUD, 68% of subjects with schizophrenia, 82% with bipolar disorder, 65% with MDD, and 65% with other psychiatric disorders received benzodiazepine treatment for ≥4 months.

Conclusion

The Dartmouth study shows rates of benzodiazepine prescribing among patients with severe mental illness to be high, especially for patients with comorbid SUD. In this population it is unclear whether benzodiazepine use increased the risk of developing SUD, or conversely whether patients with comorbid SUD had symptoms appropriate for benzodiazepine treatment with resulting high prescribing rates. The investigators point out that because Medicaid beneficiaries incur minimal or no cost for drug prescriptions, it is possible that their drug use patterns differ from patients with other types of health insurance coverage.

Although benzodiazepines have been effectively used for more than 4 decades and remain important therapeutic agents, legitimate concerns about their use have emerged at various points over this interval, especially regarding excessive or long-term use because of potential undesirable sequelae. Even with several decades experience, questions about the optimal use of benzodiazepines persist and warrant investigation. PP

References

1. Randall LO, Schallek W, Heise GA, Keith EF, Bagdon RE. The psychosedative properties of methaminodiazepoxide. J Pharmacol Exp Ther. 1960;129:163-171.

2. Sternbach LH. The benzodiazepine story. Prog Drug Res. 1978;22:229-266.

3. Ives JO, Robinson DS, Nies A, et al. The ineffectiveness of chlordiazepoxide in depressive disorders. Psychiatr J Univ Ottawa. 1978;3:115-119.

4. Lader M. Long-term anxiolytic therapy: the issue of drug withdrawal. J Clin Psychiatry. 1987;48(suppl):12-16.

5. Noyes R Jr, Garvey MJ, Cook BL, Perry PJ. Benzodiazepine withdrawal: a review of the evidence. J Clin Psychiatry. 1988;49(10):382-389.

6. Clark RE, Xie H, Brunette MF. Benzodiazepine prescription practices and substance abuse in persons with severe mental illness. J Clin Psychiatry. 2004;65(2):151-155.

7. Woods JH, Katz JL, Winger G. Use and abuse of benzodiazepines. Issues relevant to prescribing. JAMA. 1988;260(23):3476-3480.

8. Martijena ID, Lacerra C, Bustos SG, Molina VA. Chronic benzodiazepine administration facilitates the subsequent development of ethanol dependence. Brain Res. 2001;891(1-2):236-246.

9. Brunette MF, Noordsy DL, Xie H, Drake RE. Benzodiazepine use and abuse among patients with severe mental illness and co-occurring substance use disorders. Psychiatr Serv. 2003;54(10):1395-1401.

Dr. Robinson is a consultant with Worldwide Drug Development in Melbourne, Florida.

Disclosure: Dr. Robinson is a consultant to Bristol-Myers Squibb, Genaissance, Organon, Somerset, and Takeda.