We need to know more information about the type of electroconvulsive therapy given before concluding whether or not a trial of electroconvulsive therapy was adequate

Commentary on:

Malhi GS, Outhred T and Irwin L (2019) Do we need to know more about repetitive transcranial magnetic stimulation in the treatment of depression? Australian & New Zealand Journal of Psychiatry 53: 505–508. DOI: 10.1177/0004867419839784.

The article by Malhi et al. (2019), published on-line in April, opens up an important debate about the place of repetitive transcranial magnetic stimulation (rTMS) in the toolbox for treatment-resistant depression. They describe a case in which electroconvulsive therapy (ECT) was effective but was ceased due to intolerable cognitive side effects. However, we do not know what form of ECT was used, so we cannot really say whether or not these side effects could have been avoided or reduced by a change in the ECT prescription. ECT can no longer be considered a ‘unitary’ treatment, as we now know that electrode placement, pulse width and stimulus dose relative to seizure threshold (SDRST) all make a significant difference to both efficacy and cognitive side effects. For example, if this patient was being treated with bitemporal ECT, which has the highest risk of cognitive side effects, then a switch to a more cognitive sparing option such as right unilateral (RUL) brief pulse or even ultrabrief pulse ECT would have been a much more appropriate intervention before a trial of rTMS. We need to know these prescription parameters in order to make any judgement about the outcomes of a course of ECT. To do otherwise would be like declaring a medication to be ineffective or poorly tolerated, without knowing the dose used. It is also noted in the time-line chart that the most recent course of maintenance ECT was delivered concurrently with several medications, in particular lithium, which may have exacerbated cognitive side effects. Reduction and/or cessation of some of these medications may have been a viable alternative to ceasing the ECT.

The article also makes two interesting assertions about TMS, although neither was referenced in the article. The first is that ‘marked anxiety … is known to be an indicator of poor treatment response in general, and also for rTMS specifically’ and the second is that ‘benzodiazepines may also attenuate response to rTMS’. In respect of the first assertion, there have been some recent articles that challenge this notion. Clarke et al. (2019) recently published a study showing that, in patients with major depression and a comorbid anxiety disorder, there was no difference in response or remission rates to rTMS for depressive symptoms, compared to patients without comorbid anxiety. Chen et al.’s (2019) study showed that rTMS is effective in alleviating anxiety symptoms as well as depressive symptoms in patients with depression and comorbid anxiety. Therefore, it seems that the role of rTMS in depression with anxiety may be more robust than traditional wisdom has held.

In respect of the second assertion, until recently there was little evidence one way or another about how benzodiazepines might affect rTMS response. However, two recent studies have shone some light on this. The first (Kaster et al., 2018) showed that a more rapid response to rTMS was correlated with no benzodiazepine use, as well as a trend that benzodiazepine use was associated with no response, which just fell below statistical significance. The second (Hunter et al., 2019) showed that concurrently prescribed benzodiazepines led to less improvement at 2 weeks into the TMS course, and a lower response rate at the end of the 6-week course. These two studies show that, despite rTMS being a non-convulsive treatment, concurrent benzodiazepine use may, via other mechanisms, attenuate rTMS response. Further studies, such as a prospective randomised controlled trial, are required to confirm this.

Finally, the debate by Malhi et al. highlights an important issue. It has become a relatively common practice in Australia for rTMS to be given as an inpatient treatment. Patients are admitted, often to a private hospital, for a 4- to 6-week course of rTMS. The main driver for this seems to be funding. There is no Medicare funding for rTMS, and it is prohibitively expensive if the patient pays for the whole cost of outpatient treatment, usually several thousand dollars. An inpatient admission means that the hospital can charge the health fund for the cost of an inpatient stay and use this to fund the rTMS course. Understandably, health funds are beginning to question this, especially if there are no clinical grounds to justify admission, in terms of the patient’s acuity or risk. Transcranial magnetic stimulation can be given very easily as an outpatient treatment. Side effects and risks are few, and there is no anaesthetic involved, so patients can drive in and drive home, and even return to work the same day. There will be some patients receiving rTMS for whom admission is justified, based on the usual clinical indicators for inpatient admission, but for the majority, it should be an outpatient treatment. This funding driver for admissions would be resolved by the Federal Government agreeing to allocate a Medicare Item Number for rTMS, which is long overdue.