EMDR to Treat Functional Neurological Disorder: A Review

Functional neurological disorder (FND), sometimes called conversion disorder, is a common presentation in neurology clinics (Snijders, de Leeuw, Klumpers, Kappelle, & van Gijn, 2004). The term refers to neurological symptoms, such as limb weakness or numbness, fits, dystonia, gait disturbance, episodes of altered awareness, and cognitive symptoms like memory disturbances or mental fogginess, unexplained by the presence of disease or injury to the body. Common presentations include functional non-epileptic attacks (FNEA), where a person presents with fits that resemble epileptic seizures but do not have associated epileptic activity; or functional motor symptoms, also referred to as functional movement disorder, such as gait disturbance, tremors, or weakness affecting one side of the body. FND is referred to as functional neurological symptom disorder in the Diagnostic and Statistical Manual of Mental Disorders (5th ed.; DSM-5; American Psychiatric Association [APA], 2013), dissociative disorder in the International Statistical Classification of Diseases and Related Health Problems, 10th Revision ICD-10 (ICD-10; World Health Organization [WHO], 1992), and other names including conversion disorder, medically unexplained symptoms (MUS), and psychogenic disorder; historically, it was referred to as hysteria. In the case of FNEA, other terms include: functional seizures, psychogenic non-epileptic seizures, pseudoseizures, and dissociative seizures.

Identification and communication regarding the diagnosis has improved (LaFrance, Reuber, & Goldstein, 2013), but evidence-based psychological treatments are still not clearly established (Martlew, Pulman, Marson, & Cochrane Epilepsy Group, 2014; Ruddy & House, 2005). There is some evidence for the use of cognitive-behavioral therapy (CBT) and psychodynamic interpersonal therapy (PIT) for FNEA (Goldstein et al., 2010; Howlett & Reuber, 2009; LaFrance et al., 2014), and physiotherapy is the most-established treatment for functional motor symptoms (Nielsen, Stone, & Edwards, 2013; Nielsen et al., 2015). Not providing effective treatments for FND is associated with significant costs, as without treatment many patients will be referred unnecessarily to multiple medical specialties and undergo repeated unnecessary medical investigations (Ahmedani et al., 2013; Magee, Burke, Delanty, Pender, & Fortune, 2014). Additionally, it has been estimated that patients with FND have similar or greater levels of distress compared to those with neurological disease (Carson et al., 2011).

Patients with FND are heterogeneous in their presentations, with comorbid mental health difficulties prominent in some, but absent in others (Brown & Reuber, 2016; Kranick et al., 2011; Rusch, Morris, Allen, & Lathrop, 2001). Posttraumatic stress disorder (PTSD) or posttraumatic symptoms are present in some patients with FND (Fiszman, Alves-Leon, Nunes, D’Andrea, & Figueira, 2004; Pick, Mellers, & Goldstein, 2017). Psychological therapists may commonly see patients presenting with distress and “medically unexplained” physical symptoms in their clinical practice, which can lead to interpretations that the physical symptoms are a physical manifestation of their distress and that their symptoms are due to trauma, in line with conversion theory and neurobiological accounts of the impact of trauma on the body (van der Kolk, 2003). This interpretation may be relevant for some patients, but there are many patients who present in neurology clinics without identifiable trauma in their background (Brown & Reuber, 2016; Edwards & Bhatia, 2012), and alternative theories such as attention to and expectations regarding symptoms may be a more fitting explanation in those cases (Edwards, Adams, Brown, Pareés, & Friston, 2012). Additionally, a patient may have identifiable trauma in his or her background, but it may not be related to their FND symptoms, although it could be considered a vulnerability factor.

There is debate in the literature regarding whether patients with FNEA and patients with functional motor symptoms should be grouped together, with evidence pointing toward functional motor symptoms more commonly occurring after a physical event, such as illness or surgery (Pareés et al., 2014). It is suggested that the experience of novel sensory data, alongside psychological factors like experiencing the symptoms as threatening, can lead to the motor symptoms developing. Some argue that FNEA and functional motor symptoms should be considered as having different psychological profiles, whereas others argue they have common psychological profiles (Demartini et al., 2016; Hopp, Anderson, Krumholz, Gruber-Baldini, & Shulman, 2012). In a systematic review of psychological and psychiatric aspects of FNEA alone, the authors conclude that patients presenting with FNEA are more likely to report higher physical symptoms generally; experiences of trauma are common but not always present; trait dissociation can be more common in FNEA patients, but not universally; and that inconsistent findings across studies are probably a reflection of the heterogeneous nature of patients experiencing FNEA (Brown & Reuber, 2016).

Description of EMDR

EMDR therapy was developed by Francine Shapiro in 1987 with the first study published in 1989 (Shapiro, 1989). It is a psychological therapy that has integrated cognitive-behavioral, psychodynamic, experiential, and gestalt ideas. The fundamental premise is that psychological distress (and somatic expressions of distress) originates from upsetting memories in a person’s past, and that targeting those key memories will result in resolution of the psychological distress. It was originally designed to focus on any distressing memories, but in order to test the therapy on a group of patients where upsetting memories are a problem, Francine Shapiro initially focused on the treatment of PTSD. But while EMDR is well known as an intervention to treat PTSD, it has applications beyond PTSD (Shapiro, 1999; Tesarz et al., 2014).

EMDR is designed to be utilized only by suitably qualified practitioner psychologists or psychotherapists, who have undergone the 7-day EMDR training required and who receive EMDR supervision. It is an eight-phase treatment, where the EMDR phases are the most distinctive part of the treatment (but this constitutes only three of the phases: 4–6). The theory behind EMDR therapy is the Adaptive Information Processing (AIP) model, which suggests that our information processing system assimilates new experiences into already existing memory networks. The AIP model proposes that pathology can occur when adverse life experiences are stored incorrectly into a state-specific form that is unable to connect to other memory networks that hold adaptive information. This subsequently disturbs the neurological system and can cause some of the symptomology experienced. The proposed mechanism of the EMDR treatment is an assimilation of adaptive information from other memory networks linking into the network holding the isolated adverse event, which allows learning to take place with the now adaptively stored event (Solomon & Shapiro, 2008). There are several theories regarding why the eye movements (or dual attention to auditory or tactile stimuli) may facilitate emotional processing (Andrade, Kavanagh, & Baddeley, 1997; Stickgold, 2002; van den Hout, Muris, Salemink, & Kindt, 2001), but like other psychological therapies, the specific mechanisms of action are not clear. It has been suggested that the eye movements allow the person to stay in an optimal zone for processing (i.e., not too distressed) (Jeffries & Davis, 2013). EMDR incorporates aspects of many psychological therapeutic approaches, and it is likely many parts of EMDR contribute to its effectiveness, not a single part. There has been controversy regarding whether the eye movements within EMDR are a necessary part of treatment, with disagreement among researchers (Davidson & Parker, 2001; Jeffries & Davis, 2013).

EMDR therapy is now an established and recommended treatment for PTSD (Bisson et al., 2013; National Institute for Clinical Excellence, 2005). Its use beyond PTSD has also been utilized. There is now accumulating evidence for its effectiveness in treating other disorders, including chronic pain (Tesarz et al., 2014) and anxiety disorders (Shapiro, 1999). Accordingly, modified EMDR protocols have been developed to support therapists (e.g., Grant’s EMDR pain protocol) (Grant & Threlfo, 2002).

Rationale for Using EMDR to Treat Functional Neurological Disorder

The history of trauma in patients with functional neurological symptoms is often significant. High rates of trauma and abuse, ranging from 44%–100% to 23%–77%, respectively, have been reported by patients with FNEA, 15%–40% higher than those found in control groups (Fiszman et al., 2004). The extent to which trauma is involved (if at all) may not be completely clear, but evidence indicates an increased risk compared to the general population. This suggests that the AIP model can potentially be applied to functional neurological symptoms.

For patients with comorbid PTSD and FND, EMDR would arguably already be a treatment option as it is already a treatment for the former. The gap in treatment, therefore, remains in those who have no comorbid PTSD, and whether EMDR therapy would still serve as a possible treatment option. These patients may have experienced “big T” trauma (required for the ICD-10 and the DSM-5 diagnosis of PTSD [APA, 2013; WHO, 1992]), but not have sufficient PTSD symptoms to meet the threshold for diagnosis. Patients can also report “small t” traumas associated with the start of their symptoms, such as conflict within their family or at work. Additionally, “small t” traumas that occurred in childhood (e.g., bullying) can have lasting negative effects upon a person and are known vulnerability factors in the development of mental health difficulties (Lereya, Copeland, Costello, & Wolke, 2015; Shapiro, 2014). There is evidence that EMDR can be effective at treating non-PTSD upsetting memories (Cvetek, 2008). Furthermore, memories of physical ill health, investigations, or operations could also be seen as “small t” trauma and may relate to symptomology; many patients with FND report a physical trigger to the start of their symptoms (Demartini et al., 2016). The AIP model posits that most forms of pathology are based on unprocessed memories, suggesting that EMDR could be useful for people with FND without comorbid PTSD linked to “small t” trauma.

The AIP model suggests that due to the state-specific nature the adverse event is held in, external and internal stimuli can continue to trigger the experience, which can result in inappropriate symptomology. In regards to MUS generally, van Rood and de Roos (2009) propose a triggering stimuli may result in physical symptomology in two ways: a physical re-experience (e.g., pain) of the adverse event triggered by associations to the event (e.g., loud noise), or the meaning of the somatic complaint (e.g., “I am helpless” in relation to fatigue) may remind (consciously or unconsciously) the patient of a previous traumatic event associated with a similar meaning (e.g., being sexually assaulted)—a cognitive and emotional re-experiencing. Therefore, EMDR therapy could target the traumatic memory, an upsetting memory related to the somatic symptom, or the somatic symptom itself during the desensitization phase. The traumatic memory could be a serious accident or physical abuse that is re-experienced with the physical complaint. The somatic symptom memory could be the upsetting experiences that surround the somatic symptom or the traumatic consequences of the somatic symptom (e.g., memory of having a FNEA). Lastly, the somatic symptom could be experienced in session and used (e.g., sensations of numbness), rather than the memory of the experience (van Rood & de Roos, 2009).

A systematic review by van Rood and de Roos investigated the treatment of EMDR and MUS. It focused on MUS generally and included a wide variety of somatic symptoms, from phantom limb pain to body dysmorphic disorder. The study concluded tentatively that EMDR might be useful in the treatment of MUS patients where the complaint is etiologically linked to or maintained by trauma.

Objectives

This review aims to access the literature regarding EMDR and functional neurological symptoms in order to examine the evidence base regarding effectiveness. MUS included in the van Rood and de Roos systematic review was a broad and heterogeneous collection of disorders (van Rood & de Roos, 2009); this review aims to narrow the scope to just functional neurological symptoms, in order to create a clear rationale for EMDR in the specific area of FND.

Method

Study Selection

The combination searching of both functional neurological symptoms (and its variant keywords) and EMDR totaled 108 references: 7 references from Medline with revisions, 18 references from Embase (14 were unique), 16 references from PsychINFO (13 were unique), 16 references from the Cochrane Library of Systematic Reviews, 3 references from the Cochrane Library of Controlled Trials, 14 references from CINAHL Plus (12 were unique), 24 references from Web of Science—one of which was later flagged by e-mail alert—(19 were unique), and 10 references from PubMed (0 were unique). Further screening of the references of any promising articles was carried out to search for more articles. This generated an additional two articles (Chemali & Meadows, 2004; Silver, Rogers, & Russell, 2008). Four articles were removed after full-text examination due to insufficient data (e.g., no clear pre-treatment and post-treatment data) or because they lacked original scientific data. All authors of the chosen three (Chemali & Meadows, 2004; Kelley & Benbadis, 2007; Silver et al., 2008) (Figure 1) were sent an e-mail enquiring of any further articles or ongoing projects they may have, but none responded.

Figure 1.

Flow diagram of search strategy and study selection.

Results

An overview of the papers included in this systematic review is displayed in Table 1 according to authors, study type, participant, functional neurological symptoms, duration of symptoms, number of EMDR sessions, follow-up, and success of treatment. The studies included are one case series (Kelley & Benbadis, 2007) and two case studies (Chemali & Meadows, 2004; Silver et al., 2008). This totaled five participants. One of the case studies was part of a larger case series but the other participants did not fit the inclusion criteria (Silver et al., 2008). The literature search included studies from 1989 to 2016, although no studies that met the inclusion criteria were found from before 2004, and the most recent study was from 2008. The studies followed the same design of pre-treatment and post-treatment, and follow-up data. The follow-up period ranged from 3 months to 18 months after treatment.

Outcome of Studies

The non-formal assessment of the clinically observed effect reported by the participants and therapists was the measurement used to decide the success of treatment. One study used standardized outcome measures to evaluate comorbid mental health difficulties (Silver et al., 2008).

Platinum Standard Assessment

The overall quality of the studies was assessed using the Hertlein and Ricci’s (2004) PS criteria for EMDR studies to create a PS score, as shown in Figure 2. One point, 0.5, or 0 is awarded for each of the 13 PS criteria (e.g., clearly defined target symptoms or treatment adherence) and these are summed to give the PS score (with 13 being the best score possible). All of the studies in this review received low scores. This is mainly due to the fact they are case studies and therefore don’t meet many of the criteria, such as use of control group or effect size reporting. Some information was missing in all the studies (e.g., all studies provided no information on assessor training), which resulted in 0 scoring for the relevant criterion because it must be assumed that it was not considered in the research design. It remains possible, of course, that the studies followed the PS criteria more closely than their score suggests but the information was not included in the articles. The average PS score from the Hertlein and Ricci review of empirical studies of EMDR treating PTSD symptoms was 7.75. This suggests that the quality of the studies, and their design, included in this review is below average.

Figure 2.

Platinum standard score for included studies.

Limitations

The overall quality of the studies, according to the Hertlein and Ricci (2004) PS criteria for EMDR studies, was low. Important details regarding treatment given were missing, including therapist qualifications, EMDR targets, and that the EMDR method varied across the case studies, with some patients receiving “counseling” sessions prior to EMDR. Additionally, the sample of patients included in the case reports was varied in terms of age, prescribed medication, and comorbid difficulties, while blinding was variable.

Case studies are often subject to publication and reporting bias, which can skew results. As only case studies and one case series—with a small number of cases—were available, no meta-analyses could be carried out; only descriptive analysis performed. Consequently, clear estimates of the effect size of EMDR treatment cannot be determined. The case study design also means that results may be due to placebo or other factors.

The measurements of outcome in the case studies included in this review were limited. Subjective reporting of FND symptoms was the primary outcome for all studies and this can only be primarily measured by patient report. Only one case study used standardized outcome measures to assess comorbid mental health difficulties, when clearly mood and comorbid mental health difficulties are important to assess before and after treatment. Additionally, arguably when assessing outcome of patients with functional neurological symptoms, pre- and post-assessment regarding the impact of symptoms on day-to-day functioning would be an important treatment outcome to measure, and no study administered standardized measures assessing daily function. Nevertheless, although multimodal measurement is a criterion in the platinum standard guidelines for EMDR studies, no association has been found between multimodal measurement and treatment effect (Maxfield & Hyer, 2002).

Discussion

This article aimed to systematically review the evidence regarding the use of EMDR as a treatment for FND. Of the five patients treated across three studies, four patients were successfully treated for their conditions: three who experienced FNEA and one who experienced a functional movement disorder. Despite the limitations of the studies included, the successes outweigh the failures in this review, and therefore point to the potential promise of EMDR treatment, or at least an avenue of treatment that warrants further investigation. All patients treated had a history of trauma. This may be indicative of the proposed link between FND and trauma, or the association of PTSD with EMDR may have directed treatment for the patients toward EMDR. All studies found were published between 2004 and 2008. The lack of newer EMDR research with the FND population is surprising given the advancements in the literature and research of other physical conditions and EMDR treatment such as chronic pain (including phantom limb pain and migraines), in which observational and controlled trials have been administered (Shapiro, 2014; Tesarz et al., 2014). The stigma of FND or EMDR may be a reason for this lack of research, both subject to controversy (Davidson & Parker, 2001; Jeffries & Davis, 2013; Nicholson, Stone, & Kanaan, 2011; Wessely & White, 2004). Another potential reason is that FND research is focusing on other nonpharmacological treatments, such as physiotherapy and cognitive-behavioral therapy (Goldstein et al., 2015; Nielsen et al., 2015).

Clinical Implications

This review demonstrates some promising evidence for the use of EMDR in treatment of FND with comorbid PTSD. This is not surprising given the proven effectiveness of EMDR to treat PTSD (Bisson et al., 2013; National Institute for Clinical Excellence, 2005). It suggests it may also be beneficial at treating associated somatic expressions, where there is a link between development of FND symptoms and traumatic experiences. Additionally, many patients with FND do not meet the criteria for a diagnosis of PTSD, but may have identifiable traumatic experiences in the background, such as experiences of childhood abuse (Fiszman et al., 2004; Myers, Perrine, Lancman, Fleming, & Lancman, 2013). The small amount of evidence from this review suggests EMDR could potentially be a helpful treatment for these presentations. Adding weight to this idea, EMDR is designed to treat traumatic memories (not just PTSD), so any presentation where the patient has identifiable trauma memories (whether “small t” or “big T” trauma) (Cvetek, 2008; Shapiro, 2014) that still distress him or her in the present day may be a suitable candidate for EMDR therapy (with or without comorbid PTSD).

An intriguing development of the usefulness of EMDR therapy is its use in treating somatic symptoms, such as chronic pain conditions like phantom limb pain, fibromyalgia, neuropathic pain, musculoskeletal pain, and headache (Tesarz et al., 2014). The target memories in EMDR treatment can be the pain itself or a distressing memory associated with the pain. de Jongh and colleagues demonstrated that in 64 patients—50% with PTSD, 50% without PTSD—there was no difference between groups in terms of reduction of vividness and emotionality associated with the target memory, suggesting that EMDR impacts on negative memories generally, not just those associated with PTSD (de Jongh, Ernst, Marques, & Hornsveld, 2013).

Therefore, we propose that EMDR therapy is possibly useful as a treatment for patients with FND in four possible ways: (a) comorbid PTSD associated with the FND; (b) distressing childhood trauma memories that may be relevant in terms of a person’s tendency to dissociate; (c) distressing memories associated with the FND symptoms, for example, memory of when the symptoms started (such as following a medical procedure); and (d) the FND symptoms (possibly using an adapted version of the pain protocol). This is summarized in Table 2. These potential uses of EMDR as a treatment for FND need to be properly researched, but we suggest that all four potential uses are examined.

Table 2.

Possible Uses of EMDR for Patients With Functional Neurological Disorder

1. Posttraumatic stress disorder

2. Childhood trauma memories

3. Distressing memories associated with FND

4. FND symptoms

It has been proposed that a one-treatment approach is not suitable for those with FND, and treatment should be individualized according to need (Agrawal, Gaynor, Lomax, & Mula, 2014; LaFrance, 2007). In Figure 3, we propose a FND pathway where EMDR treatment could potentially fit in. We suggest that after patients have had both neurological and neuropsychiatric assessment, they can attend group or individual psychoeducation regarding FND. Following this, depending on clinical need (and availability), they may then receive neuropsychiatric treatment, physiotherapy, psychological treatment (which could possibly include EMDR), and/or intensive inpatient or day-patient programs.

Figure 3.

Proposed functional neurological disorder care pathway, with possible role of EMDR.

The potentially short duration of EMDR therapy is useful, as is the nature of the therapy, which does not require the patient to speak in detail about associated distressing material. This could serve to reduce drop-out rates as difficulty identifying and discussing feelings can be a feature of patients with FNEA (Brown & Reuber, 2016). High drop-out rates were a problem in the Kelley and Benbadis case series (Kelley & Benbadis, 2007), but the study design included counseling sessions prior to the EMDR desensitization sessions, and this may have had the unintended consequence of causing people to disengage as they were not ready to consider possible links between their FND and past trauma. This was evidenced by one participant who dropped out prior to EMDR sessions, after the therapist suggested his FNEA were related to the death of his son.

Many patients with FND, in particular those with FNEA, have a general tendency to dissociate (Demartini et al., 2016; Pick et al., 2017). Although EMDR can be effective even when patients have dissociative tendencies, it should be noted that caution is required when working with this patient group. The desensitization phase requires accessing trauma memories, which may well trigger dissociation in the session. Clinicians using EMDR with patients who are highly dissociative, in particular patients with a diagnosis of dissociative identity disorder, need to have experience of working with this population, and be confident in their use of EMDR, in particular their ability to manage abreactions in session and the use of cognitive interweaves. Inadequate screening, preparation, or delivery of EMDR with this population can be destabilizing to patients’ mental health (Shapiro, 2001).

Future Research

The evidence presented from the case studies cannot be generalized, and clearly controlled trials are needed in order to properly establish EMDR therapy’s effectiveness with FND presentations. Given the small number of participants evaluated in the case studies presented, even an observational study with a larger sample size would be of benefit. Future trials would need to take into account the heterogeneous nature of the patients, perhaps examining subgroups of FND separately (e.g., FNEA vs. functional movement disorder), but then also the different presentations within each subgroup (e.g., associated with identifiable trauma or not). Non-trauma-based studies with FND patients are needed to determine whether the target can in fact be somatic symptom-based or needs to be a traumatic memory target. Van Rood and de Roos (van Rood & de Roos, 2009) noted in regards to a few studies on chronic pain that the pain as the target was not as effective as targeting the trauma (Mazzola et al., 2009; Wilensky, 2006).

In terms of measurement of outcomes, it will be important to examine not just reduction/resolution in FND symptoms, but also impact of symptoms on day-to-day functioning and mood.

Conclusions

Despite the limitations the case studies present, they offer a window into the possibilities that EMDR therapy could hold for patients suffering with FND. EMDR could potentially be a useful therapy for patients with FND, who have identifiable trauma in their background (with and without comorbid PTSD). Additionally, the burgeoning evidence for EMDR therapy’s effectiveness at treating chronic pain conditions suggests a utility for treating somatic symptoms alone, as well as the usefulness of targeting distressing memories associated with somatic symptoms. It is possible that EMDR could be used both as a treatment that targets distressing memories associated with the first experience of FND symptoms, as well as a treatment targeting FND symptoms alone, without targeting any traumatic or distressing memories. Further research, in particular controlled trials, is needed.