Response to: Stimulating dangerous argument?

Recent issues of the journal have contained an interesting, and we suspect somewhat entertaining, series of debate articles focused on the use of repetitive transcranial magnetic stimulation (rTMS) treatment for depression. Unfortunately, we are not sure if these have really advanced understanding of the field adequately. In the most recent follow-up article, ‘Stimulating dangerous argument?’ (we are not sure what is dangerous about this debate), Professor Malhi and colleagues (2019) have continued the debate but really failed to address the vast majority of factual or content issues that were raised by both of us in our responses that preceded this. Instead, they appear to have really focused on two points that we would like to address in turn:

• 1. That the evidence supporting the use of TMS is ‘meagre’ and potentially able to ‘support any position one chooses to adopt’

There is incontrovertibly a large and substantial literature that supports the use of rTMS and informs its use in clinical practice and is currently guiding its use by thousands of clinicians around the world who have integrated rTMS therapy into their clinical practice. This literature has been sufficient to satisfy the Food and Drug Administration (FDA) in the United States, most health insurers in that country and professional and regulatory bodies including the American Medical Association (AMA; US), The Royal Australian and New Zealand College of Psychiatrists (RANZCP) and National Institute for Health and Care Excellence (NICE; UK) that its routine clinical use should be supported and/or funded. Most notable in an Australian context, this evidence has also been sufficient to convince the Medicare Services Advisory Committee to recommend public funding of rTMS treatment in Australia in August 2019, a decision that by its nature reflects a deep investigation of the evidence supporting this therapy.

However, as is not necessarily surprising, given the relative newness of the treatment, the evidence that supports clinical decision-making is variable once we get beyond simple decisions about efficacy and, as we acknowledged previously, should be the focus (and in many areas is the focus) of ongoing research. However, this differs in absolutely no way from other areas of therapeutic practice in psychiatry and it is notable that the authors have chosen to single out rTMS as if it is somehow different from other forms of treatment. For example, lithium and benzodiazepines (BZDs) were co-prescribed in the treatment of bipolar disorder for many decades before the first study was published exploring the significance of this practice, it itself only a secondary analysis (Bobo et al., 2014). The lack of evidence addressing these types of clinical questions is seen across treatments in psychiatry, but is not raised by the authors as ‘shortcomings’. rTMS is singled out in a surprisingly narrowly focused manner. Given that two studies have already addressed BZD prescribing during rTMS (Hunter et al., 2019; Kaster et al., 2019) (and only one after 50 years of lithium treatment; Bobo et al., 2014), we think this might provide ‘more than a little reassurance’ that the rTMS community is dedicated to furthering the field with ‘rigorous research’ as is happening widely for rTMS but perhaps not so much in other areas.

Of note, although Professor Malhi and colleagues (2019) readdressed an earlier concern about concurrent BZD use, it is important to make completely clear that a statistical association with poorer response does in no way mean that the concurrent use of BZDs fully removes the possible beneficial effect of rTMS. Patients with more severe conditions are often on poly-pharmacy, reflecting illness severity or treatment resistance. These patients may indeed be the ones that need alternatives such as TMS (or electroconvulsive therapy [ECT] – although of note this may be especially problematic in BZD-dependent subjects). In established clinical practice, larger randomized and observational studies (e.g. Fitzgerald et al., 2016) have demonstrated substantive rTMS response rates without taking patients off BZDs for many years prior to these recent publications.

• 2. That offering rTMS to a patient at ‘the end of the line’ in regards to treatment options may be inappropriate

We found the logic of Professor Malhi and colleagues in this regards extremely confusing and quite contradictory. They initially point out that doing something in a highly treatment-resistant patient may be ‘very dangerous and inadvertently lead to harm’, making the reader believe that in fact to provide no treatment may be a better alternative. However, after talking about the use of rTMS and that this may not be effective, they state that providing it, or extending the treatment duration of application, is problematic, because it would potentially delay ‘trialing an alternative’ treatment. Professor Malhi and colleagues do not indicate what else they would propose to provide to patients as an alternative treatment given the circumstances but we would be extraordinarily surprised if a so-called end of the line patient would have a treatment alternative supported by anything like the evidence that supports the clinical use of rTMS treatment in patients with treatment-resistant depression. We do agree that patients with treatment-resistant depression are ‘desperate for any glimmer of hope’ and should be managed professionally. This applies to all forms of treatment, pharmacologic, TMS and ECT.

We would like to end with a brief scenario: if a patient now presents having failed two or more antidepressant (AD) medications, is it ethically appropriate not to mention that there is a safe treatment (rTMS), the efficacy of which is supported by over 40 clinical trials and at least 10 meta-analyses or just provide medication number 3 or 4 which is supported by no randomized placebo-controlled trials and probably has a likelihood of success of well less than 20%? If Professor Malhi and colleagues are ‘concerned about the reputation of our profession’, perhaps this concern would be better focused on the all-to-frequent repetition of antidepressant medication trials – something that is supported by much less evidence, but is also much more common, than the use of rTMS in treatment-resistant depression.

In conclusion, we would wholeheartedly agree with Malhi et al. that we should always make certain that we prescribe treatments that are ultimately in the best interest of patients. However, this comes with an obligation to ensure that as clinicians (and especially academics informing clinical practice) we are fully informed of the evidence base that underpins new treatments and do not judge new treatments with standards we do not impose on therapies to which we have an existing commitment such as pharmacotherapy or ECT.