Sunday 3 March 2013

Research Briefing: How safe is transcranial magnetic stimulation?

Source Article: Maizey, L., Allen, C.P.G., Dervinis, M., Verbruggen, F., Varnava, A., Kozlov, M., Adams, R.C., Stokes, M., Klemen, J., Bungert, A., Hounsell, C.A., Chambers, C.D. (2013). Comparative incidence rates of mild adverse effects to transcranial magnetic stimulation. Clinical Neurophysiology, 124, 536-544.  [pdf] [monitoring forms]


When I moved to Cardiff University back in 2008, the first thing I did was set up two labs for doing human transcranial magnetic stimulation (TMS). I’d been using TMS since 2002 and it was (and continues to be) a major part of my research programme. Unlike brain imaging techniques such as fMRI or MEG, TMS interferes with brain activity. This means that the effect of TMS on behaviour can tell us which parts of the brain are necessary for different cognitive functions. In my lab we use TMS to study processes such as perception, attention, consciousness, decision-making, and response inhibition.

In the process of setting up TMS – a new technique for Cardiff at the time – I had to submit a lengthy application for ethics approval. After several weeks of discussion, the committee and I decided that building a TMS lab offered the opportunity to do some novel research on the side effects of brain stimulation.

Since it was developed in 1985, TMS has been generally considered safe for human use. Serious adverse effects, such as seizures, are rare, and few incidents have been reported since international guidelines for TMS safety were established in 1998 (updated in 2009). However, TMS has been suspected to cause a range of more mild adverse effects, such as headache and nausea. Much less is known about these lesser side effects, even though they can be very unpleasant for participants.

So back in 2008 we decided to put in place a system for monitoring side effects. After every experimental session involving TMS, participants were given a form to complete that listed a series of possible symptoms occurring within 24 hours of the session (the forms can be downloaded here). Then, when the participant returned for their next session, we collected and archived these forms. Over several years of TMS experiments – and many different variants of the technique – we amassed more than 1000 such forms from over 100 unique participants. Last year, after four years, we decided we had enough data to commence the analysis.

I’m now happy to report that the paper documenting this analysis has appeared in the journal Clinical Neurophysiology, written primarily by my PhD student, Leah Maizey. To our knowledge this paper reports the largest TMS safety study yet conducted by a single research team.

Overall, participants in our study reported mild adverse effects (or MAEs) following ~5% of sessions, although 39% of participants reported at least one MAE at some point during their experimental regime. When MAEs did occur, the most common was headache (41%). Rates of adverse effects were higher for active TMS compared to sessions involving ‘sham’ (placebo) TMS, although a small number of adverse effects could nevertheless be attributed to coincidence or placebo effects. 

Two other findings are notable and may be of special interest to TMS researchers. First, MAEs were more likely to occur following a participant’s first session, even controlling for various extraneous factors. We believe this tendency could be explained by anxiety when receiving TMS for the first time, so steps taken by researchers to ensure that participants are relaxed and comfortable are likely to help.

Second – and most striking – nearly 80% of MAEs were reported after participants had left the laboratory at the end of their session. We don’t have a good explanation for why this is, but 80% is too big to ignore. Maybe the physiological aftereffects of TMS are longer lasting than is generally assumed, or maybe the immediate aftereffects can have knock-on effects to other physiological systems. This was a serendipitous finding, so it will be important to see whether other researchers can independently replicate such long-lasting effects.

The good news for TMS researchers is that our study adds to a body of evidence that TMS is safe for human use under carefully controlled conditions. The adverse effects we did observe were mostly very minor (no seizures!) and only a few participants withdrew from the experiments. Our main recommendation is that it would be useful for the TMS community to monitor adverse effects more closely and to adopt standard methods for doing so. We provide relevant monitoring forms as part of our paper.


* Special thanks to Matthew Rushworth for helpful discussion at the outset of this project.