Statistics from Altmetric.com
Research on the subjective experience of dissociative (psychogenic non-epileptic) seizures (DS) is dominated by that on objective semiology.1 ,2 Patients with DS tend not to spontaneously volunteer any warning symptoms, nor feelings of resistance to seizure onset, both of which are more common in epilepsy.3 But when an ‘aura’ is sought, studies in DS have found rates of 25–60%,4–7 typically with somatic symptoms of autonomic arousal such as dizziness, depersonalisation and chest tightness along with cognitive symptoms such as fear of losing control. This cluster of symptoms may sometimes meet criteria for a panic attack or an episode of hyperventilation, but in many cases, does not. This suggests that patients with DS do have prodromal symptoms but generally don't want to talk about them. Goldstein and Mellers6 proposed that DS are a “paroxysmal, dissociative response to heightened arousal in the absence of raised general anxiety levels”.
As we explored this ‘dissociative response’ in more detail with patients, we met some who were willing to admit that the attack was, at times, a welcome relief from their unbearable prodromal symptoms. This led us to discover that a number of our patients were aware of exerting conscious control over the onset of their blackouts.
Between August 2010 to August 2012 a retrospective audit identified 11 patients (10 female, age 17–54) who experienced highly unpleasant but brief warning symptoms prior to the majority of their DS and in addition subsequently admitted that, as their attacks recurred, they developed a feeling that they wanted the ‘seizure’ part of the attack to ‘hurry up’ in order to extinguish these unbearable prodromal symptoms. All patients were diagnosed with dissociative (non-epileptic) seizures by a consultant neurologist based on EEG or clinical features (table 1). Warning somatic symptoms consisted of variable mixture of lightheadedness, dissociation, warmth, sweating, chest tightness, trouble breathing (ie, hyperventilation), nausea or fear and lasted seconds to minutes. Prodromal symptoms typically shortened over time. Some patients described a more low grade uncomfortable feeling that could last several hours, which they knew from experience would be relieved by a seizure, again giving them a feeling that they wanted the seizure to happen in order to feel temporarily better.
Seizure prodromes and especially ‘wilful submission’ were generally described by the patient only after prompting and after several clinic visits. The common theme was that the rising somatic and cognitive symptoms were so intolerable that it was preferable to make a conscious choice to pass out rather than to put up with it.
Many patients described the sense of conflict that arose from the paradox of seeking help for the attacks to stop, whilst being aware that in an acute situation they wanted the attack to ‘hurry up’ to obliterate their unpleasant sensations. In contrast most patients described marked relief at being able to discuss this openly with their treating doctor, both to discover that other patients described similar experiences, and that criticism they feared about ‘giving in to the attacks’ did not materialise. Treatment was reframed at using distraction and cognitive techniques to extinguish the warning symptoms rather than having to rely on ‘having a blackout’ to get rid of them. It appeared in some cases that discussion of this problem was instrumental in accepting a diagnosis that the patient had previously been unhappy about. In many cases it appeared important in resolving or improving the occupational and social impact of attacks.
The idea that a ‘paroxysm’ can temporarily relieve ‘hysterical’ symptoms is centuries old. Freud hypothesised that the ‘unbearable idea’ was the cause of ‘conversion’ of intrapsychic stress into a physical symptom,8 although he was describing a more abstract psychic conflict rather than a specific attack of symptoms. Here we are suggesting something more straightforward. Somatic symptoms of panic and dissociation are common prodromal symptoms of DS,6 ,7 and in some cases, patients deliberately choose to have an attack in order to escape from the sensation. They feel compelled to do so, in the same way that someone with agoraphobia chooses to leave a cinema when they start to panic. It seems likely that the attack would happen anyway in most cases, even without encouragement, but we found that some patients learn that they can speed the process up by wilfully dissociating.
It is not difficult to see how dissociative attacks could, in this situation, become a learnt conditioned response in which the ‘reward’ of the attack is the relief of symptoms. This model also helps to understand why warnings of attacks shorten and disappear over time in some patients. Attacks that subsequently occur without warning may simply represent the conditioned response without the stimulus.
We believe that some patients’ knowledge of complicity in seizure production, but a simultaneous feeling of distress about having the attack, leads to particular anxiety and sensitivity around this issue, including a marked reluctance to reveal it to their treating clinicians. This may explain why patients with dissociative attacks are so characteristically silent when asked about their symptoms.3
This hypothesis is additive to one of the simplest and attractive cognitive models of DS.6 It also accounts for inexplicable features of some patient's attacks—for example, patients who are able to delay their attacks until they are in a place of safety. If some patients with DS are ‘hiding’ these cognitions it may be one factor in explaining why neurologists are often concerned about how genuine their patient's complaint is.9
These are preliminary observations that may only apply to a minority of patients with DS, may be a consequence rather than a cause of attacks, are at risk of being iatrogenically suggested to the patient by an over eager clinician and require further study.
We believe that a sensitive discussion of these issues can be incorporated within the cognitive behavioural model tested in a recent randomised trial.10 We caution, however, that this is an exceptionally sensitive topic, which clinicians should not rush to discuss with their patients until a therapeutic relationship has been established.
Contributors JS and AC both conceived the study, contributed data, drafted and revised the manuscript.
Competing interests None.
Ethics approval Audit of our routine practice to describe new observation.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.