Abstract
Tibrewal et al. (2021a) recently argued that ‘benzodiazepines should be actively considered when SSRIs and CBT have initially failed for severe and disabling anxiety disorders’. To be clear, the authors were referring to long-term/chronic use of benzodiazepines, not short-term/acute use. Given the wide-ranging and potentially severe and disabling problems associated with benzodiazepines, this is a significant proposition. Crucially, their entire argument rests on the premise that benzodiazepines maintain their anxiolytic therapeuticity when used long-term. The authors confidently declared ‘studies have demonstrated that long-term treatment [with benzodiazepines] is associated with maintenance of therapeutic benefit and no dose escalation for anxiety disorders’, and they cited a recent paper by Edward Silberman et al. (2020) to support this assertion. Silberman’s paper, in turn, referenced a 1990 report published by the American Psychiatric Association (APA) Task Force on Benzodiazepine Dependence (the report). However, the report never actually concluded that long-term benzodiazepines maintain their therapeutic benefit at all. Rather, it stated that ‘there are conflicting reports about the development of tolerance to the anxiolytic effects of benzodiazepines’ (APA, 1990) and, naturally, it listed several conflicting studies—all of which are now at least 32 years old. To be fair to Silberman (and, in turn, Tibrewal et al), the report did list 4 ‘studies’ (term used loosely) that ‘do not report development of tolerance to therapeutic benefits’. However, when carefully examined, these “studies” do not credibly support the claim that chronic benzodiazepine use is associated with maintenance of therapeutic benefits either; or that there is “no dose escalation” over time (see Table 1). Indeed, arguably the first two sources suggest otherwise.
Sources referenced by the APA Task Force on Benzodiazepine Dependence to support the claim that long-term treatment with benzodiazepines is not associated with ‘tolerance to therapeutic benefits’.
A basic PubMed search for ‘benzodiazepines’ AND ‘tolerance’ produces 451 results, a multitude of which readily discuss the development of tolerance to the anxiolytic effects of benzodiazepines. Perhaps the most conclusive evaluation of this particular topic, however, is found within a ‘systematic review of the benzodiazepines’ produced by the United Kingdom (UK) Committee on the Review of Medicines (CRM, 1980) and published in the British Medical Journal (BMJ) at the peak of benzodiazepine prescribing (diazepam was the top-selling drug in the United States between 1968 and 1980 with a peak in sales in 1978—2.3 billion tablets (Calcaterra and Barrow, 2014)]. The CRM (1980) was explicit in its commentary on the lack of evidence for long-term benzodiazepine use and, considering their well-recognised adverse effects, recommended discontinuation of these drugs ‘as soon as possible’: ‘The committee took particular note of the lack of firm evidence of efficacy which might support the long-term use of benzodiazepines in insomnia and anxiety… there was little convincing evidence that benzodiazepines were efficacious in the treatment of anxiety after four months’ continuous treatment. It considered that an appropriate warning regarding long-term efficacy be included in the recommendations, particularly in view of the high proportion of patients receiving repeated prescriptions for extended periods of time’.
The CRM (1980) went on to suggest that the symptoms experienced by patients when coming off these drugs (i.e. withdrawal) were often confused with relapse of anxiety, prompting their re-prescription, and falsely ‘demonstrating’ their ‘long-term efficacy’. More recently, Vorspan et al. reported that people developed tolerance to benzodiazepines and Z-drugs within just 1 - 4 weeks, particularly to the hypnotic and anxiolytic effects. They similarly argued that so-called ‘long-term efficacy studies’ conflated withdrawal/rebound syndromes with psychiatric relapse (Vorspan et al., 2018) and recommended that where benzodiazepines are prescribed (e.g. for alcohol withdrawal), they should be time-limited and medically supervised—consistent with the CRM’s recommendations, albeit nearly 40 years later.
Tibrewal et al. were right to call for more research into the safety, efficacy and tolerability of psychotropic agents used to treat severe and disabling anxiety disorders (including second-generation antipsychotics such as quetiapine which, as they correctly identify, are not without risk). However, their push for the ready use of benzodiazepines, in this context, is based on a flawed rationale (i.e. persistent anxiolytic therapeuticity, despite tolerance to all other effects of benzodiazepines) and is not sufficiently supported by evidence. Given the authors’ strong advocacy for ‘second-line’ use of these potentially dangerous drugs (including in a person who has merely ‘failed’ SSRIs and CBT), it is perturbing to note that they did not scrutinise the primary sources underlying the reference that apparently validates their core argument. A contemporary risk-benefit analysis, informed by the available literature and clinical experience to date, frankly does not support the long-term use of benzodiazepines in this setting. The risks of inappropriate benzodiazepine prescribing are well known and have not been overestimated—that is precisely why the United States Food and Drug Administration (US FDA) has just updated the boxed warning.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
