Use of suggestion methods to induce psychogenic non-epileptic seizures

We welcome the comprehensive and timely systematic review of suggestive seizure induction for psychogenic non-epileptic seizures (PNES) by Popkirov and colleagues [1]. As the authors of two of the studies cited in this review we would like comment on two issues. The first concerns the degree of honesty or deceptiveness of the communication strategies employed. We agree that avoiding deception is crucial for ethical reasons, and that an honest approach tends to reinforce a positive therapeutic relationship [2]. In our published studies, we told patients explicitly that a psychological cause for their attacks was being considered and that we wished to record attacks to confirm the diagnosis and thus facilitate appropriate treatment [3,4]. This is clearly stated in our first article [3], in the fifth paragraph of the discussion section, where we discuss the ethical issue: ‘‘. . . we told patients that psychological attacks were being considered, and that recording of these was necessary for diagnosis.’’ The fact that two thirds of our patients went on to have a habitual attack during video-EEG might be taken to indicate openness to a psychological diagnosis, as well as providing strong evidence against malingering [3]. The same approach continued to be used in our clinical practice [5] with a similar yield of recorded attacks and good diagnostic accuracy [6,7]; no episodes of ‘‘pseudostatus’’ have occurred in over 1000 recordings (unpublished observation). We would consider that our method involves honest communication without omission and would therefore suggest that, in the present paper by Popkirov et al., our studies should be included in the ‘‘explicitly open’’ rather than the ‘‘truthful but omissive’’ category. The second aspect relates to possible psychobiological mechanisms underlying the efficacy of suggestion in eliciting PNES, clearly a difficult and interesting question. As well as interactions with stress [8,9], arousal [10], hypnotizability [11] and dissociative tendencies [12], another possible framework for thinking about suggestion in PNES relates to the placebo effect [13], which has been increasingly well-characterised over the last decade [14,15]. This is described as a ‘‘psychobiological phenomenon attributable to the overall psychosocial therapeutic context’’ [15], and hinges upon expectation of a certain outcome. In studies of PNES, the environmental setting is the video-EEG recording and the expectation (of both patient and doctor) is that there is a greater chance than usual that a seizure will occur, especially in the context of specific provocation methods that are judged likely to trigger an attack. It is interesting that, in patients in our previous study in whom suggestion during video-EEG provoked an event, there was a prior history of events occurring in medical settings [3], implying an effect of situational context for this subgroup. The likelihood of placebo response is multi-factorial, depending on neurobiological, environmental and psychosocial factors [16]. It