Botulinum toxin: does the black box warning justify change in practice?

In January 2008, the Public Citizen's Health Research Group (a Washington, DC-based consumer advocacy group) petitioned the US Food and Drug Administration (FDA) to order black-box warnings on labels of botulinum toxin products, to issue 'Dear Doctor Letters' and require that every patient receive a pamphlet before an injection, warning of the risk of the spread of botulinum toxin beyond the injection site with the associated risks of dysphagia, aspiration and/or pneumonia and death.1 This request was based on similar warnings concerning botulinum toxin issued by the European Union between 2005 and 2007, and the Group's own analysis of the FDA Adverse Event database (AERS) for Botox and Myobloc. Their investigation found 180 such adverse event cases submitted by drug manufacturers, including 16 deaths, four of which occurred in children. In their preliminary response to this petition, the FDA, in February 2008, released an early communication about an ongoing safety review on the subject and notified the license holders of botulinum toxin products in the US to ensure that their product labeling included the risk of the spread of botulinum toxin effects from the area of injection.2 One year later, the FDA completed its safety review and in April 2009 announced the new safety label changes including a boxed warning, and a Medication Guide/Risk Evaluation and Mitigation Strategy (REMS) highlighting the risk of the spread of the toxin effect from the area of injection, causing symptoms similar to those of botulism.3–5 These symptoms had mostly been reported in children with cerebral palsy (CP) being treated for spasticity, an off-label (not FDA approved) use of these drugs. Similar investigations by Health Canada identified 10 reports of serious reactions following injections, including five deaths, two of which occurred in children with CP. Almost overnight, these warnings had a chilling effect on clinical practice in centres that for years had been using botulinum toxin injections for spasticity management. Particularly affected were children with more severe CP, who were felt to be at greatest risk of these complications. The change in practice seemed to be driven almost entirely by these 'new' safety concerns rather than a balanced consideration of the benefits and risks. The reports from which these concerns were derived demonstrated some worrying associations but could not (nor were they designed to) show causality. On one hand, patients at the highest risk of these adverse events are those with the poorest health, who are already at risk for such complications as part of their natural history. Thus, the chronology of the adverse event following a recent injection could have been coincidental. On the other hand, it is plausible that the baseline risk for such complications is indeed exacerbated by the effect of unintended local/distant spread of the toxin. Could this be related to the higher doses used to treat muscle spasticity or even accidental overdoses? Many of us were surprised by these reports, not least because we had not 'observed' these events in our own practice. Had these complications not caught our attention because isolated events might have been (wrongly) attributed to baseline risk? For those in the 'botulinum toxin is not deadly' camp, there is some good news. In their article, O'Flaherty et al.6 conducted a prospective pre–post cohort study of all children undergoing botulinum toxin A injections in a single children's rehabilitation department over a 16-month period, to determine the rate of adverse events in the month following injections. The subject of this paper is both important and timely. Patients served as their own controls with their health status in the month before their injection used as a baseline for comparison. Participants were distributed across all five Gross Motor Function Classification System levels. In the month following an injection episode, an adverse event occurred in 23% of injection episodes, 2% of these were potentially serious (lower respiratory tract infections/aspiration, worsening dysphagia, or generalized weakness), but this rate was slightly lower than, or no different from, the rate of these events in the month before the injections. Next to a randomized trial design, which is generally impractical for questions of this nature, the study design used was quite robust and superior to previous work pertinent to this subject. An observational study with a valid matched control group that did not receive botulinum toxin A would be difficult to find without significant selection biases. The specific adverse events considered were relevant and the ascertainment of these events either in person or by telephone was reasonable given the short time horizon of 1 month. One hundred per cent of participants were included. Readers will have to reflect on how closely the described practice in this centre (including the specific preparation of botulinum toxin and dosages used) resembles their own, when making judgments about the generalizability of these findings. The potential conflict of interest (declared) arising from unrestricted educational funding the authors have received from industry should of course be taken into consideration. In view of this study, I will be able to continue to offer (or not) my patients botulinum toxin injections based on the quality of evidence of effectiveness, rather than on safety concerns alone. Also, patients/parents will be able to make a more informed choice when they are presented with the warnings of potential adverse events. Of course, it behoves us to remain vigilant and report back to our respective health agencies about any of these serious adverse events. This study may not be the final word on the subject but is an important addition to the literature.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.