Epilepsy & Behavior 41 (2014) 140–143
Contents lists available at ScienceDirect
Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh
Brief Communication
Long-term outcome of psychogenic nonepileptic seizures: The role of induction by suggestion☆ Orsola Gambini a,1, Benedetta Demartini a,⁎,1, Valentina Chiesa b,1, Katherine Turner b,1, Valentina Barbieri a,1, Maria Paola Canevini b,1 a b
Cattedra di Psichiatria, Dipartimento di Scienze della Salute, Università degli Studi di Milano, Italy UO Neurologia 2, Centro Epilessia, A.O. San Paolo, Dipartimento di Scienze della Salute, Università degli Studi di Milano, Via A. Di Rudinì, 8, 20142 Milan, Italy
a r t i c l e
i n f o
Article history: Received 26 June 2014 Revised 19 August 2014 Accepted 27 September 2014 Available online xxxx Keywords: Psychogenic nonepileptic seizure Induction by suggestion Outcome Prognosis
a b s t r a c t Purpose: The aims of our retrospective observational study were to evaluate the long-term outcome of PNESs after communication of the diagnosis and to define predictors of good outcome. Method: Twenty-seven consecutive patients with a certain diagnosis of psychogenic nonepileptic seizures (PNESs) were included in the study. Follow-up information was obtained from each participant through a questionnaire designed for the study. Regarding seizure frequency, the patients were asked to report how many seizures they had experienced on average every month before the communication of the diagnosis and after it. Results: After the communication of the diagnosis, the median seizure frequency had dropped to 4 every month (p b 0.001). Seventeen participants (63%) were seizure-free at follow-up, and a further five (18.5%) showed a greater than 50% improvement in seizure frequency. Regarding the predictive value of clinical and sociodemographic variables for PNES global outcome, the factors gender, education, economic status, interval of time from onset, comorbidity with epilepsy, psychiatric history, mental retardation, psychological therapy, psychiatric therapy, and the presence of stressful and traumatic events were not related to prognosis; the only factor associated with a better outcome was the diagnosis made after the induction of PNESs by suggestion (p = 0.000, χ2 = 4.654). Conclusion: A substantial majority of our patients became seizure-free with communication of the diagnosis as the only intervention. The use of the induction by suggestion test was an important predictor of good outcome. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Psychogenic nonepileptic seizures (PNESs) can be defined as paroxysmal events that resemble or can be mistaken for epilepsy, without being associated with abnormal electrical activity in the brain [1]. Psychogenic nonepileptic seizures represent a puzzling clinical condition for which etiology, evidence-based treatments, and outcomes are not yet defined. Subjects with PNESs constitute a heterogeneous group; nevertheless, the majority of them fulfill the Diagnostic and Statistical Manual of Mental Disorders — fourth edition, text revision (DSM-IV-TR) criteria for conversion disorders. In addition, a high
☆ This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors. ⁎ Corresponding author at: Cattedra di Psichiatria, Dipartimento di Scienze della Salute, Università degli Studi di Milano, AO San Paolo, Via A. Di Rudinì, 8, 20142 Milan, Italy. Tel.: +39 3496375759. E-mail address: [email protected] (B. Demartini). 1 Each author has contributed in all the following: 1) conception and design, acquisition of data, or analysis and interpretation of data; 2) drafting the article or revising it critically for important intellectual content; and 3) final approval of the version to be published.
http://dx.doi.org/10.1016/j.yebeh.2014.09.076 1525-5050/© 2014 Elsevier Inc. All rights reserved.
percentage of patients affected by PNESs present a comorbid mood or anxiety disorder; according to recent studies, mood and anxiety disorders have been found in more than 50% of cases [2], posttraumatic stress disorders in up to 100% [3], and cluster A and B personality disorders in up to 40% [4]. The first step in PNES management is diagnosis, for which videoelectroencephalography (video-EEG) remains the gold standard; in their recent study, Syed et al. suggested that video-EEG, combined with the patient's history, has a very high sensitivity and specificity [5]. In addition to video-EEG, induction by suggestion has previously been reported to be effective in the diagnosis of PNESs. In the study by Lancman et al., the sensitivity of the induction test, which consisted in placing a colored patch imbibed in alcohol on the neck of the patients, was 77.4%, the specificity 100%, the positive predictive value 100%, and the negative predictive value 48.7% [6]. A second important step in PNES management resides in the communication of the diagnosis. Recently, Mayor et al. showed how communicating the diagnosis in a clear and supportive manner is the only intervention required to stop PNESs in at least 15–30% of cases [7]. Nevertheless, they did not identify specific predictors of good response to the sole communication of the diagnosis.
O. Gambini et al. / Epilepsy & Behavior 41 (2014) 140–143
After the diagnosis of PNESs and its communication to the patient, treatment options are presented. To date, no official guidelines for the treatment of PNESs are available. Psychological treatment and pharmacotherapy (mainly antidepressants) are widely used, but data regarding their efficacy are limited and not consistent. Efficacy of cognitive– behavioral therapy, which is the most used approach among psychotherapies, has been recently evaluated in two pilot randomized controlled studies [8,9]: Goldstein et al. found a significant reduction in seizure frequency at the end of the treatment period, but no changes were found in most of the psychological outcomes [8]; LaFrance et al. reported a significant improvement in a range of clinical and psychological factors at the end of the treatment [9]. However, neither study evaluated the long-term outcome of the psychological approach. Concerning pharmacotherapy, different drugs have been anecdotally used, but only a small controlled trial on sertraline, whose results are of partial efficacy, has been conducted [10]. Another controversial issue on PNESs concerns outcome and outcome measures. Among the outcome measures, PNES frequency after diagnosis and/or therapeutic program is often evaluated. However, recent studies showed how the reduction or cessation of seizures is useful as an objective clinical outcome measure but often do not correlate with psychosocial recovery or quality of life [11–13]. Data concerning outcome (assessed by the reduction of seizures) are often controversial and are mainly limited to the short term. Overall, prognosis of PNESs appears to be poor. The majority of studies found that 40% or less of patients with PNESs achieve seizure remission in the follow-up period [14–16]. The largest study (n = 260) [14] found that just 38% of patients were seizure-free 6–12 months after diagnosis. In one of the studies with the longest follow-up, Reuber et al. [15] found that 71% of patients were still having seizures 1–10 years after diagnosis. Nevertheless, long-term outcomes of PNESs after diagnosis have been rarely evaluated. The aims of our retrospective observational study were to evaluate the long-term outcome of PNESs after diagnosis communication and to define predictors of good outcome. 2. Methods Between October 2013 and February 2014, 27 consecutive patients with a certain diagnosis of PNESs seen at our Regional Epilepsy Center for follow-up were included in a retrospective outcome study. The original number of patients was 36, but nine of them have not been included because they were not reachable at the time of the present study. In order to focus on the long-term outcome, we recruited patients who had received the diagnosis at least 1 year before (in all cases, the diagnosis had been communicated by one of the authors). The following criteria were selected: age of 18–60 years, normal IQ, and no or mild intellectual disabilities. Twelve out of the 27 patients have also been previously enrolled in one of our previous studies [17]. All patients had given their written informed consent for the study. The Ethics Committee of San Paolo Hospital reviewed and approved the study protocol. The diagnosis of PNESs was made on the basis of the consensus of at least two epilepsy specialists based on the patients' clinical history and video-EEG monitoring. When no episode was registered during the video-EEG monitoring, an induction by suggestion test was also performed. The induction was carried out using a standardized protocol, according to Lancman et al. [6]. Patients with PNESs with a history of possible additional epileptic seizures were eligible to take part but only if the two seizure types could be distinguished clearly by the patients and caregivers. 2.1. Assessment All subjects were assessed by a psychiatrist at the Regional Epilepsy Center at the time of the PNES evaluation or diagnosis. Each patient
141
received a psychiatric assessment as follows. Diagnoses were made according to DSM-IV-TR criteria. All patients were evaluated using the Italian version of the SCID-I and SCID-II interview. Subjects were also interviewed about stressful and traumatic life events from childhood to present in the context of a nonstandardized interview. Follow-up information was obtained from each participant through a questionnaire designed for the study (Table 1). Seizure frequency data were collected retrospectively in one-month intervals for the duration of the follow-up period: patients were asked to report how many seizures they had experienced on average every month before the communication of the diagnosis and after it, relying on their seizure diaries (for patients with intellectual disabilities, seizure diaries were kept by caregivers). Patients who had experienced no seizures in this period of time were classed as “seizure-free”. 2.2. Statistical analysis Statistical analysis was performed using SPSS version 21 (Statistical Package for the Social Sciences). Categorical or ordinal variables were compared using the χ2 test, the Mann–Whitney U test, or the Kruskal–Wallis test where appropriate; continuous variables were compared using Student's t-test or oneway analysis of variance after applying Levene's test for equality of variance. If the variance was unequal, nonparametric tests were performed. Simultaneous logistic regression models were used to evaluate the ability of independent variables to predict outcomes. Seizure outcome data were expressed as a percentage of improvement in seizure frequency from before the communication of the diagnosis to follow-up (a 50% improvement represented 50% fewer seizures at follow-up). 3. Results Twenty-seven patients (21 females [77.8%]; mean age: 43.9 years [SD, 12.1 years]) took part in the study. Patients' demographic and clinical characteristics are shown in Table 2. Thirteen patients (48.1%) did not have PNESs during the video-EEG monitoring; subsequently, they underwent the induction test. Two patients (7.4%) underwent further investigations after the communication of the diagnosis, which consisted in brain MRI for both of them. Before the communication of the diagnosis, the median seizure frequency of the whole group was 11 per month. After the communication of the diagnosis [a mean of 21 months (SD, 12.1 months, range: 14–38 months)], the median seizure frequency had dropped to 4 per month (significant difference; p b 0.001). Seventeen participants (63%) were seizure-free at follow-up, and a further five (18.5%) showed a greater than 50% improvement in seizure frequency. The predictive value of a range of clinical and sociodemographic variables for PNES global outcome is summarized in Table 3: the factors gender, education, economic status, interval of time from onset, comorbidity with epilepsy, psychiatric history, mental retardation, psychological therapy, psychiatric therapy, and the presence of stressful and traumatic events were not related to prognosis; the only factor associated with a better outcome was the diagnosis made after the induction of PNESs by suggestion (p = 0.000, χ2 = 4.654): all 13 patients who received the induction of PNESs by suggestion were seizure-free at follow-up; this finding was independent from all the Table 1 Questions of the standardized interview designed for the study. 1. How many months ago did you receive the diagnosis of PNESs? 2. How many PNESs every three months did you have before the communication of the diagnosis? 3. How many PNESs did you have in the last three months? 4. Did you receive any specific treatment for PNESs? If yes, specify. 5. After the diagnosis of PNESs, did you have any other investigation?
142
O. Gambini et al. / Epilepsy & Behavior 41 (2014) 140–143
Table 2 Patients' demographic and clinical characteristics. Mean age, years (SD) Females (%) Mean age at PNES onset, years (SD) Education, years Economic status (% active)a Mean interval of time from onset, months (SD) Comorbidity with epilepsy (%) Psychiatric comorbidity (%) Intellectual disabilities (%) Psychological therapy (%) Psychiatric therapy (%) Stressful and traumatic life events Induction by suggestion (%)
43.9 (12.1) 77.8 32.56 (10.6) 9.7 (3.1) 44 21 (12.1) 51.9 44.4 29.6 18.5 40.7 29.6 48.1
a Patients were considered ‘economically active’ if they were either in full-time or parttime employment.
following variables: gender (p = 0.765, χ2 = 3.654), age (p = 0.643, χ2 = 7.744), education (p = 0.453, χ 2 = 3.455), economic status (p = 0.643, χ 2 = 2.854), interval of time from onset (p = 0.975, χ 2 = 6.865), comorbidity with epilepsy (p = 0.135, χ2 = 1.864), psychiatric history (p = 0.7453, χ 2 = 2.765), mental retardation (p = 0.432, χ 2 = 1.654), psychological therapy (p = 0.122, χ2 = 3.987), psychiatric therapy (p = 0.349, χ2 = 4.755), and the presence of stressful and traumatic events (p = 0.555, χ2 = 2.865). Despite the small sample size, we conducted the same analysis with the whole group divided into two subgroups: patients who underwent any psychological treatment (18.5% received a specific psychotherapy for PNESs, and 25.9% received a supportive psychological treatment which was very brief – number of sessions between one and six – and not specific for PNESs) and patients who did not. The only factor associated with a better outcome remained the diagnosis made after the induction of PNESs by suggestion in both subgroups (p = 0.000, χ2 = 5.877 for patients who received psychotherapy and p = 0.000, χ2 = 4.766 for those who did not receive psychotherapy). 4. Discussion Our retrospective observational study showed that patients can experience a significant reduction or even a cessation of seizures in the long-term follow-up after the communication of the diagnosis: 63% of our patients were seizure-free at follow-up, and a further 18.5% showed a greater than 50% improvement in seizure frequency. Our outcomes are not related to psychological intervention. These data are in keeping with previous data suggesting that a proportion of patients do well with communication of the diagnosis as the only intervention [14,18]. Factors determining short-term outcome may be different from those determining long-term outcome. This, as well as methodological differences, makes direct comparison of outcome studies difficult. Nevertheless, the global proportion of patients seizure-free 21 months (mean) after the communication of the diagnosis in our Table 3 Predictive value of a range of clinical and sociodemographic variables for PNES global outcome. Factor
p (χ2)
Gender Education Economic status Interval of time from onset Comorbidity with epilepsy Psychiatric comorbidity Intellectual disabilities Psychological therapy Psychiatric therapy Stressful and traumatic life events Induction by suggestion
0.417 (4.992) 0.765 (6.543) 0.543 (4.543) 0.893 (5.864) 0.747 (2.692) 0.401 (5.122) 0.313 (5.935) 0.172 (7.734) 0.087 (9.620) 0.733 (4.754) 0.000 (4.654)
Bold values indicate significance at p b0.005.
study is significantly higher than that found in previous studies. In fact, Mayor et al. [7] found that 16% of their patients were seizure-free six months after the communication of the diagnosis, and an additional 23% of patients reported a greater than 50% improvement in seizure frequency; Hall-Patch et al. [19] found that 14% of patients were seizure-free 3 months after the communication of the diagnosis, and a further 63% reported a greater than 50% improvement in seizure frequency. Our results are more similar to the ones described by Duncan et al. [20]: at 3 months postdiagnosis, 50% of their patients were seizure-free; for 88.9%, seizures ceased immediately upon communication of the diagnosis; at 6 months, 44.4% were seizure-free. An important difference between our study and those of Mayor et al. [7] and Hall-Patch et al. [19] consisted in the use of a specific diagnostic instrument for the diagnosis: in fact, video-EEG confirmation of PNESs was available for just a minority of their patients. In our study, all patients were diagnosed either after a video-EEG registration or after the induction test. In our population, induction of PNESs by suggestion is the only predictor of better outcome. All 13 patients who received the induction of PNESs by suggestion were seizure-free at follow-up. Despite the small sample size, which might result in a type II error, this is a completely new finding; in fact, none of the previous studies had used induction by suggestion as a variable to identify predictors of good outcome. Induction by suggestion has not been used in recent studies, although several data showed its efficacy (both sensitivity and specificity) in the diagnosis of PNESs. One possible issue is ethical. In fact, although the test is completely safe, it might produce potential harm because its use requires deception [21]. It is the deception implicit in the use of diagnostic placebos that raises ethical problems and produces their resulting harms (e.g., loss of confidence in the trustworthiness of physicians). Nevertheless, the placebo saline intravenous infusion test to deceptively provoke and prove the existence of PNESs has been used since 1982 [22] and has been considered safe, reliable, and effective, and its use is justified because it benefits patients by preventing them from being wrongly diagnosed with and treated for epilepsy [23]; in addition, provocative procedures for other conversion/ functional disorders and maneuvers such as the Hoover's sign (weakness of voluntary hip extension with normal involuntary hip extension during contralateral hip flexion against resistance) are currently used in the diagnostic process and explained to subjects as induction tools. Here, we speculate that the induction by suggestion test might not only have a role in the diagnostic process of PNESs but may also have relevance as a therapeutic intervention; in fact, patients' positive reaction to induction by suggestion might help them to accept the diagnosis and to understand the control they could have on their PNESs; further studies are needed to better clarify this finding on bigger samples. Regarding other predictors of good outcome, none of those we evaluated was significant. These data are in contrast with those of previous studies. Durrant et al. [24] have reviewed the literature regarding this aspect and found that, although results from several studies are very different, older patients with coexisting epilepsy and suffering from more dramatic seizures tend to have a poorer global outcome. The larger study available [14] found that previous psychiatric diagnoses, social security payments, and female genders were predictors of poor outcome. The main reason why our results are not in line with those of previous studies probably resides in the small sample of patients we have recruited. Our study has several limitations. First, we could only evaluate a relatively small number of patients with PNESs, which limits solid conclusions. In addition, we have carried out a relatively large number of statistical tests on a modestly sized data set. This raises the possibility of type II error. However, it is important to note that the statistical significance of our findings was very high. Second, it is possible that, despite our efforts to reflect clinical reality, our data are affected by selection bias and that some patient groups (including those who became seizure-free after the communication of the diagnosis) may be underrepresented in our sample. In fact, although 36 subjects were
O. Gambini et al. / Epilepsy & Behavior 41 (2014) 140–143
approached about this study, only 27/36 took part (nine were not in charge or not reachable by phone). Third, there is always the risk of bias in a questionnaire study when the study participants know they are being observed. This bias tends to lead to false positive results, providing the possibility of type I error. Fourth, differently from patients assessed in Mayor et al.'s [7] and Hall-Patch et al.'s [19] studies, all our patients were recruited from our Regional Epilepsy Center, which is a tertiary diagnostic unit, and are not necessarily representative of the population of PNESs in general neurological centers. Finally, there are other potential predictors of good or poor outcome that we did not measure; these might be social in nature such as the receipt of healthrelated financial benefit (differential diagnosis with malingering) or psychological. Future research should focus on using a wider variety of outcome measurements, overall level of functioning, and other quality-of-life indicators. Conflict of interests The authors have no conflict of interests to declare. References [1] Brown RJ, Syed TU, Benbadis S, LaFrance Jr WC, Reuber M. Psychogenic non epilepic seizure. Epilepsy Behav 2011;22(1):85–93. [2] D'Alessio L, Giagante B, Oddo S, Silva WW, Solis P, Consalvo D, et al. Psychiatric disorders in patients with psychogenic nonepileptic seizures, with and without comorbid epilepsy. Seizure 2006;15:333–9. [3] Fiszman A, Alves-Leon SV, Nunes RG, D'Andrea I, Figueira I. Traumatic events and posttraumatic stress disorder in patients with psychogenic nonepileptic seizures: a critical review. Epilepsy Behav 2004;5(6):818–25. [4] Reuber M, Pukrop R, Bauer J, Derfuss R, Elger CE. Multidimensional assessment of personality in patients with psychogenic non-epileptic seizures. J Neurol Neurosurg Psychiatry 2004;75:743–8. [5] Syed TU, LaFrance Jr WC, Kahriman ES, Hasan SN, Rajasekaran V, Gulati D. Can semiology predict psychogenic nonepileptic seizures? A prospective study. Ann Neurol 2011;69:997–1004. [6] Lancman ME, Asconape JJ, Craven WJ, Howard G, Penry J. Predictive value of induction of psychogenic seizures by suggestion. Ann Neurol 1994;35(3):359–61.
143
[7] Mayor R, Brown RJ, Cock H, House A, Howlett S, Singhal S, et al. Short-term outcome of psychogenic non-epileptic seizures after communication of the diagnosis. Epilepsy Behav 2012;25:676–81. [8] Goldstein LH, Chalder T, Chidwedere C. Cognitive–behavioural therapy for psychogenic nonepileptic seizures: a pilot RCT. Neurology 2010;74:1986–94. [9] LaFrance Jr WC, Miller IW, Ryan CE. Cognitive behavioural therapy for psychogenic nonepileptic seizures. Epilepsy Behav 2009;14:591–6. [10] LaFrance Jr WC, Keitner GI, Papandonatos GD, Blum AS, Machan JT, Ryan CE, et al. Pilot pharmacologic randomized controlled trial for psychogenic nonepileptic seizures. Neurology 2010;75:1166–73. [11] LaFrance Jr, Rusch D, Machan JT. What is ‘treatment as usual’ for nonepileptic seizures? Epilepsy Behav 2008;12:388–94. [12] Reuber M. Psychogenic nonepileptic seizures: answers and questions. Epilepsy Behav 2008;12:622–35. [13] Reuber M, Mitchell AJ, Howlett S, Elger CE. Measuring outcome in psychogenic nonepileptic seizures: how relevant is seizure remission? Epilepsia 2005;46(11): 1788–95. [14] McKenzie P, Oto M, Russell A, Pelosi A, Duncan R. Early outcomes and predictors in 260 patients with psychogenic nonepileptic attacks. Neurology 2010;74(1):64–9. [15] Reuber M, Pukrop R, Bauer J, Helmstaedter C, Tessendorf N, Elger CE. Outcome in psychogenic nonepileptic seizures: 1 to 10-year follow-up in 164 patients. Ann Neurol 2003;53(3):305–11. [16] O'Sullivan SS, Spillane JE, McMahon EM, Sweeney BJ, Galvin RJ, McNamara B, et al. Clinical characteristics and outcome of patients diagnosed with psychogenic nonepileptic seizures: a 5-year review. Epilepsy Behav 2007;11(1):77–84. [17] Turner K, Piazzini A, Chiesa V, Barbieri V, Vignoli A, Gardella E, et al. Patients with epilepsy and patients with psychogenic non-epileptic seizures: video-EEG, clinical and neuropsychological evaluation. Seizure 2011;20:706–10. [18] Kanner AM. More controversies on the treatment of psychogenic pseudoseizures: an addendum. Epilepsy Behav 2003;4:360–4. [19] Hall-Patch L, Brown R, House A, Howlett S, Kemp S, Lawton G, et al. Acceptability and effectiveness of a strategy for the communication of the diagnosis of psychogenic nonepileptic seizures. Epilepsia 2010;51:70–8. [20] Duncan R, Razvi S, Mulhern S. Newly presenting psychogenic nonepileptic seizures: incidence, population characteristics, and early outcome from a prospective audit of a first seizure clinic. Epilepsy Behav 2011;20:308–11. [21] Leeman BA. Provocative techniques should not be used for the diagnosis of psychogenic nonepileptic seizures. Epilepsy Behav 2009;15:110–4. [22] Cohen RJ, Suter C. Hysterical seizures: suggestion as a provocative EEG test. Ann Neurol 1982;11(4):391–5. [23] Walczak TS, Williams DT, Berten W. Utility and reliability of placebo infusion in the evaluation of patients with seizures. Neurology 1994;44(3 Pt 1):394–9. [24] Durrant J, Rickards H, Cavanna AE. Prognosis and outcome predictors in psychogenic nonepileptic seizures. Epilepsy Res Treat 2011:1–7 [Article ID 274736].
Contents lists available at ScienceDirect
Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh
Brief Communication
Long-term outcome of psychogenic nonepileptic seizures: The role of induction by suggestion☆ Orsola Gambini a,1, Benedetta Demartini a,⁎,1, Valentina Chiesa b,1, Katherine Turner b,1, Valentina Barbieri a,1, Maria Paola Canevini b,1 a b
Cattedra di Psichiatria, Dipartimento di Scienze della Salute, Università degli Studi di Milano, Italy UO Neurologia 2, Centro Epilessia, A.O. San Paolo, Dipartimento di Scienze della Salute, Università degli Studi di Milano, Via A. Di Rudinì, 8, 20142 Milan, Italy
a r t i c l e
i n f o
Article history: Received 26 June 2014 Revised 19 August 2014 Accepted 27 September 2014 Available online xxxx Keywords: Psychogenic nonepileptic seizure Induction by suggestion Outcome Prognosis
a b s t r a c t Purpose: The aims of our retrospective observational study were to evaluate the long-term outcome of PNESs after communication of the diagnosis and to define predictors of good outcome. Method: Twenty-seven consecutive patients with a certain diagnosis of psychogenic nonepileptic seizures (PNESs) were included in the study. Follow-up information was obtained from each participant through a questionnaire designed for the study. Regarding seizure frequency, the patients were asked to report how many seizures they had experienced on average every month before the communication of the diagnosis and after it. Results: After the communication of the diagnosis, the median seizure frequency had dropped to 4 every month (p b 0.001). Seventeen participants (63%) were seizure-free at follow-up, and a further five (18.5%) showed a greater than 50% improvement in seizure frequency. Regarding the predictive value of clinical and sociodemographic variables for PNES global outcome, the factors gender, education, economic status, interval of time from onset, comorbidity with epilepsy, psychiatric history, mental retardation, psychological therapy, psychiatric therapy, and the presence of stressful and traumatic events were not related to prognosis; the only factor associated with a better outcome was the diagnosis made after the induction of PNESs by suggestion (p = 0.000, χ2 = 4.654). Conclusion: A substantial majority of our patients became seizure-free with communication of the diagnosis as the only intervention. The use of the induction by suggestion test was an important predictor of good outcome. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Psychogenic nonepileptic seizures (PNESs) can be defined as paroxysmal events that resemble or can be mistaken for epilepsy, without being associated with abnormal electrical activity in the brain [1]. Psychogenic nonepileptic seizures represent a puzzling clinical condition for which etiology, evidence-based treatments, and outcomes are not yet defined. Subjects with PNESs constitute a heterogeneous group; nevertheless, the majority of them fulfill the Diagnostic and Statistical Manual of Mental Disorders — fourth edition, text revision (DSM-IV-TR) criteria for conversion disorders. In addition, a high
☆ This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors. ⁎ Corresponding author at: Cattedra di Psichiatria, Dipartimento di Scienze della Salute, Università degli Studi di Milano, AO San Paolo, Via A. Di Rudinì, 8, 20142 Milan, Italy. Tel.: +39 3496375759. E-mail address: [email protected] (B. Demartini). 1 Each author has contributed in all the following: 1) conception and design, acquisition of data, or analysis and interpretation of data; 2) drafting the article or revising it critically for important intellectual content; and 3) final approval of the version to be published.
http://dx.doi.org/10.1016/j.yebeh.2014.09.076 1525-5050/© 2014 Elsevier Inc. All rights reserved.
percentage of patients affected by PNESs present a comorbid mood or anxiety disorder; according to recent studies, mood and anxiety disorders have been found in more than 50% of cases [2], posttraumatic stress disorders in up to 100% [3], and cluster A and B personality disorders in up to 40% [4]. The first step in PNES management is diagnosis, for which videoelectroencephalography (video-EEG) remains the gold standard; in their recent study, Syed et al. suggested that video-EEG, combined with the patient's history, has a very high sensitivity and specificity [5]. In addition to video-EEG, induction by suggestion has previously been reported to be effective in the diagnosis of PNESs. In the study by Lancman et al., the sensitivity of the induction test, which consisted in placing a colored patch imbibed in alcohol on the neck of the patients, was 77.4%, the specificity 100%, the positive predictive value 100%, and the negative predictive value 48.7% [6]. A second important step in PNES management resides in the communication of the diagnosis. Recently, Mayor et al. showed how communicating the diagnosis in a clear and supportive manner is the only intervention required to stop PNESs in at least 15–30% of cases [7]. Nevertheless, they did not identify specific predictors of good response to the sole communication of the diagnosis.
O. Gambini et al. / Epilepsy & Behavior 41 (2014) 140–143
After the diagnosis of PNESs and its communication to the patient, treatment options are presented. To date, no official guidelines for the treatment of PNESs are available. Psychological treatment and pharmacotherapy (mainly antidepressants) are widely used, but data regarding their efficacy are limited and not consistent. Efficacy of cognitive– behavioral therapy, which is the most used approach among psychotherapies, has been recently evaluated in two pilot randomized controlled studies [8,9]: Goldstein et al. found a significant reduction in seizure frequency at the end of the treatment period, but no changes were found in most of the psychological outcomes [8]; LaFrance et al. reported a significant improvement in a range of clinical and psychological factors at the end of the treatment [9]. However, neither study evaluated the long-term outcome of the psychological approach. Concerning pharmacotherapy, different drugs have been anecdotally used, but only a small controlled trial on sertraline, whose results are of partial efficacy, has been conducted [10]. Another controversial issue on PNESs concerns outcome and outcome measures. Among the outcome measures, PNES frequency after diagnosis and/or therapeutic program is often evaluated. However, recent studies showed how the reduction or cessation of seizures is useful as an objective clinical outcome measure but often do not correlate with psychosocial recovery or quality of life [11–13]. Data concerning outcome (assessed by the reduction of seizures) are often controversial and are mainly limited to the short term. Overall, prognosis of PNESs appears to be poor. The majority of studies found that 40% or less of patients with PNESs achieve seizure remission in the follow-up period [14–16]. The largest study (n = 260) [14] found that just 38% of patients were seizure-free 6–12 months after diagnosis. In one of the studies with the longest follow-up, Reuber et al. [15] found that 71% of patients were still having seizures 1–10 years after diagnosis. Nevertheless, long-term outcomes of PNESs after diagnosis have been rarely evaluated. The aims of our retrospective observational study were to evaluate the long-term outcome of PNESs after diagnosis communication and to define predictors of good outcome. 2. Methods Between October 2013 and February 2014, 27 consecutive patients with a certain diagnosis of PNESs seen at our Regional Epilepsy Center for follow-up were included in a retrospective outcome study. The original number of patients was 36, but nine of them have not been included because they were not reachable at the time of the present study. In order to focus on the long-term outcome, we recruited patients who had received the diagnosis at least 1 year before (in all cases, the diagnosis had been communicated by one of the authors). The following criteria were selected: age of 18–60 years, normal IQ, and no or mild intellectual disabilities. Twelve out of the 27 patients have also been previously enrolled in one of our previous studies [17]. All patients had given their written informed consent for the study. The Ethics Committee of San Paolo Hospital reviewed and approved the study protocol. The diagnosis of PNESs was made on the basis of the consensus of at least two epilepsy specialists based on the patients' clinical history and video-EEG monitoring. When no episode was registered during the video-EEG monitoring, an induction by suggestion test was also performed. The induction was carried out using a standardized protocol, according to Lancman et al. [6]. Patients with PNESs with a history of possible additional epileptic seizures were eligible to take part but only if the two seizure types could be distinguished clearly by the patients and caregivers. 2.1. Assessment All subjects were assessed by a psychiatrist at the Regional Epilepsy Center at the time of the PNES evaluation or diagnosis. Each patient
141
received a psychiatric assessment as follows. Diagnoses were made according to DSM-IV-TR criteria. All patients were evaluated using the Italian version of the SCID-I and SCID-II interview. Subjects were also interviewed about stressful and traumatic life events from childhood to present in the context of a nonstandardized interview. Follow-up information was obtained from each participant through a questionnaire designed for the study (Table 1). Seizure frequency data were collected retrospectively in one-month intervals for the duration of the follow-up period: patients were asked to report how many seizures they had experienced on average every month before the communication of the diagnosis and after it, relying on their seizure diaries (for patients with intellectual disabilities, seizure diaries were kept by caregivers). Patients who had experienced no seizures in this period of time were classed as “seizure-free”. 2.2. Statistical analysis Statistical analysis was performed using SPSS version 21 (Statistical Package for the Social Sciences). Categorical or ordinal variables were compared using the χ2 test, the Mann–Whitney U test, or the Kruskal–Wallis test where appropriate; continuous variables were compared using Student's t-test or oneway analysis of variance after applying Levene's test for equality of variance. If the variance was unequal, nonparametric tests were performed. Simultaneous logistic regression models were used to evaluate the ability of independent variables to predict outcomes. Seizure outcome data were expressed as a percentage of improvement in seizure frequency from before the communication of the diagnosis to follow-up (a 50% improvement represented 50% fewer seizures at follow-up). 3. Results Twenty-seven patients (21 females [77.8%]; mean age: 43.9 years [SD, 12.1 years]) took part in the study. Patients' demographic and clinical characteristics are shown in Table 2. Thirteen patients (48.1%) did not have PNESs during the video-EEG monitoring; subsequently, they underwent the induction test. Two patients (7.4%) underwent further investigations after the communication of the diagnosis, which consisted in brain MRI for both of them. Before the communication of the diagnosis, the median seizure frequency of the whole group was 11 per month. After the communication of the diagnosis [a mean of 21 months (SD, 12.1 months, range: 14–38 months)], the median seizure frequency had dropped to 4 per month (significant difference; p b 0.001). Seventeen participants (63%) were seizure-free at follow-up, and a further five (18.5%) showed a greater than 50% improvement in seizure frequency. The predictive value of a range of clinical and sociodemographic variables for PNES global outcome is summarized in Table 3: the factors gender, education, economic status, interval of time from onset, comorbidity with epilepsy, psychiatric history, mental retardation, psychological therapy, psychiatric therapy, and the presence of stressful and traumatic events were not related to prognosis; the only factor associated with a better outcome was the diagnosis made after the induction of PNESs by suggestion (p = 0.000, χ2 = 4.654): all 13 patients who received the induction of PNESs by suggestion were seizure-free at follow-up; this finding was independent from all the Table 1 Questions of the standardized interview designed for the study. 1. How many months ago did you receive the diagnosis of PNESs? 2. How many PNESs every three months did you have before the communication of the diagnosis? 3. How many PNESs did you have in the last three months? 4. Did you receive any specific treatment for PNESs? If yes, specify. 5. After the diagnosis of PNESs, did you have any other investigation?
142
O. Gambini et al. / Epilepsy & Behavior 41 (2014) 140–143
Table 2 Patients' demographic and clinical characteristics. Mean age, years (SD) Females (%) Mean age at PNES onset, years (SD) Education, years Economic status (% active)a Mean interval of time from onset, months (SD) Comorbidity with epilepsy (%) Psychiatric comorbidity (%) Intellectual disabilities (%) Psychological therapy (%) Psychiatric therapy (%) Stressful and traumatic life events Induction by suggestion (%)
43.9 (12.1) 77.8 32.56 (10.6) 9.7 (3.1) 44 21 (12.1) 51.9 44.4 29.6 18.5 40.7 29.6 48.1
a Patients were considered ‘economically active’ if they were either in full-time or parttime employment.
following variables: gender (p = 0.765, χ2 = 3.654), age (p = 0.643, χ2 = 7.744), education (p = 0.453, χ 2 = 3.455), economic status (p = 0.643, χ 2 = 2.854), interval of time from onset (p = 0.975, χ 2 = 6.865), comorbidity with epilepsy (p = 0.135, χ2 = 1.864), psychiatric history (p = 0.7453, χ 2 = 2.765), mental retardation (p = 0.432, χ 2 = 1.654), psychological therapy (p = 0.122, χ2 = 3.987), psychiatric therapy (p = 0.349, χ2 = 4.755), and the presence of stressful and traumatic events (p = 0.555, χ2 = 2.865). Despite the small sample size, we conducted the same analysis with the whole group divided into two subgroups: patients who underwent any psychological treatment (18.5% received a specific psychotherapy for PNESs, and 25.9% received a supportive psychological treatment which was very brief – number of sessions between one and six – and not specific for PNESs) and patients who did not. The only factor associated with a better outcome remained the diagnosis made after the induction of PNESs by suggestion in both subgroups (p = 0.000, χ2 = 5.877 for patients who received psychotherapy and p = 0.000, χ2 = 4.766 for those who did not receive psychotherapy). 4. Discussion Our retrospective observational study showed that patients can experience a significant reduction or even a cessation of seizures in the long-term follow-up after the communication of the diagnosis: 63% of our patients were seizure-free at follow-up, and a further 18.5% showed a greater than 50% improvement in seizure frequency. Our outcomes are not related to psychological intervention. These data are in keeping with previous data suggesting that a proportion of patients do well with communication of the diagnosis as the only intervention [14,18]. Factors determining short-term outcome may be different from those determining long-term outcome. This, as well as methodological differences, makes direct comparison of outcome studies difficult. Nevertheless, the global proportion of patients seizure-free 21 months (mean) after the communication of the diagnosis in our Table 3 Predictive value of a range of clinical and sociodemographic variables for PNES global outcome. Factor
p (χ2)
Gender Education Economic status Interval of time from onset Comorbidity with epilepsy Psychiatric comorbidity Intellectual disabilities Psychological therapy Psychiatric therapy Stressful and traumatic life events Induction by suggestion
0.417 (4.992) 0.765 (6.543) 0.543 (4.543) 0.893 (5.864) 0.747 (2.692) 0.401 (5.122) 0.313 (5.935) 0.172 (7.734) 0.087 (9.620) 0.733 (4.754) 0.000 (4.654)
Bold values indicate significance at p b0.005.
study is significantly higher than that found in previous studies. In fact, Mayor et al. [7] found that 16% of their patients were seizure-free six months after the communication of the diagnosis, and an additional 23% of patients reported a greater than 50% improvement in seizure frequency; Hall-Patch et al. [19] found that 14% of patients were seizure-free 3 months after the communication of the diagnosis, and a further 63% reported a greater than 50% improvement in seizure frequency. Our results are more similar to the ones described by Duncan et al. [20]: at 3 months postdiagnosis, 50% of their patients were seizure-free; for 88.9%, seizures ceased immediately upon communication of the diagnosis; at 6 months, 44.4% were seizure-free. An important difference between our study and those of Mayor et al. [7] and Hall-Patch et al. [19] consisted in the use of a specific diagnostic instrument for the diagnosis: in fact, video-EEG confirmation of PNESs was available for just a minority of their patients. In our study, all patients were diagnosed either after a video-EEG registration or after the induction test. In our population, induction of PNESs by suggestion is the only predictor of better outcome. All 13 patients who received the induction of PNESs by suggestion were seizure-free at follow-up. Despite the small sample size, which might result in a type II error, this is a completely new finding; in fact, none of the previous studies had used induction by suggestion as a variable to identify predictors of good outcome. Induction by suggestion has not been used in recent studies, although several data showed its efficacy (both sensitivity and specificity) in the diagnosis of PNESs. One possible issue is ethical. In fact, although the test is completely safe, it might produce potential harm because its use requires deception [21]. It is the deception implicit in the use of diagnostic placebos that raises ethical problems and produces their resulting harms (e.g., loss of confidence in the trustworthiness of physicians). Nevertheless, the placebo saline intravenous infusion test to deceptively provoke and prove the existence of PNESs has been used since 1982 [22] and has been considered safe, reliable, and effective, and its use is justified because it benefits patients by preventing them from being wrongly diagnosed with and treated for epilepsy [23]; in addition, provocative procedures for other conversion/ functional disorders and maneuvers such as the Hoover's sign (weakness of voluntary hip extension with normal involuntary hip extension during contralateral hip flexion against resistance) are currently used in the diagnostic process and explained to subjects as induction tools. Here, we speculate that the induction by suggestion test might not only have a role in the diagnostic process of PNESs but may also have relevance as a therapeutic intervention; in fact, patients' positive reaction to induction by suggestion might help them to accept the diagnosis and to understand the control they could have on their PNESs; further studies are needed to better clarify this finding on bigger samples. Regarding other predictors of good outcome, none of those we evaluated was significant. These data are in contrast with those of previous studies. Durrant et al. [24] have reviewed the literature regarding this aspect and found that, although results from several studies are very different, older patients with coexisting epilepsy and suffering from more dramatic seizures tend to have a poorer global outcome. The larger study available [14] found that previous psychiatric diagnoses, social security payments, and female genders were predictors of poor outcome. The main reason why our results are not in line with those of previous studies probably resides in the small sample of patients we have recruited. Our study has several limitations. First, we could only evaluate a relatively small number of patients with PNESs, which limits solid conclusions. In addition, we have carried out a relatively large number of statistical tests on a modestly sized data set. This raises the possibility of type II error. However, it is important to note that the statistical significance of our findings was very high. Second, it is possible that, despite our efforts to reflect clinical reality, our data are affected by selection bias and that some patient groups (including those who became seizure-free after the communication of the diagnosis) may be underrepresented in our sample. In fact, although 36 subjects were
O. Gambini et al. / Epilepsy & Behavior 41 (2014) 140–143
approached about this study, only 27/36 took part (nine were not in charge or not reachable by phone). Third, there is always the risk of bias in a questionnaire study when the study participants know they are being observed. This bias tends to lead to false positive results, providing the possibility of type I error. Fourth, differently from patients assessed in Mayor et al.'s [7] and Hall-Patch et al.'s [19] studies, all our patients were recruited from our Regional Epilepsy Center, which is a tertiary diagnostic unit, and are not necessarily representative of the population of PNESs in general neurological centers. Finally, there are other potential predictors of good or poor outcome that we did not measure; these might be social in nature such as the receipt of healthrelated financial benefit (differential diagnosis with malingering) or psychological. Future research should focus on using a wider variety of outcome measurements, overall level of functioning, and other quality-of-life indicators. Conflict of interests The authors have no conflict of interests to declare. References [1] Brown RJ, Syed TU, Benbadis S, LaFrance Jr WC, Reuber M. Psychogenic non epilepic seizure. Epilepsy Behav 2011;22(1):85–93. [2] D'Alessio L, Giagante B, Oddo S, Silva WW, Solis P, Consalvo D, et al. Psychiatric disorders in patients with psychogenic nonepileptic seizures, with and without comorbid epilepsy. Seizure 2006;15:333–9. [3] Fiszman A, Alves-Leon SV, Nunes RG, D'Andrea I, Figueira I. Traumatic events and posttraumatic stress disorder in patients with psychogenic nonepileptic seizures: a critical review. Epilepsy Behav 2004;5(6):818–25. [4] Reuber M, Pukrop R, Bauer J, Derfuss R, Elger CE. Multidimensional assessment of personality in patients with psychogenic non-epileptic seizures. J Neurol Neurosurg Psychiatry 2004;75:743–8. [5] Syed TU, LaFrance Jr WC, Kahriman ES, Hasan SN, Rajasekaran V, Gulati D. Can semiology predict psychogenic nonepileptic seizures? A prospective study. Ann Neurol 2011;69:997–1004. [6] Lancman ME, Asconape JJ, Craven WJ, Howard G, Penry J. Predictive value of induction of psychogenic seizures by suggestion. Ann Neurol 1994;35(3):359–61.
143
[7] Mayor R, Brown RJ, Cock H, House A, Howlett S, Singhal S, et al. Short-term outcome of psychogenic non-epileptic seizures after communication of the diagnosis. Epilepsy Behav 2012;25:676–81. [8] Goldstein LH, Chalder T, Chidwedere C. Cognitive–behavioural therapy for psychogenic nonepileptic seizures: a pilot RCT. Neurology 2010;74:1986–94. [9] LaFrance Jr WC, Miller IW, Ryan CE. Cognitive behavioural therapy for psychogenic nonepileptic seizures. Epilepsy Behav 2009;14:591–6. [10] LaFrance Jr WC, Keitner GI, Papandonatos GD, Blum AS, Machan JT, Ryan CE, et al. Pilot pharmacologic randomized controlled trial for psychogenic nonepileptic seizures. Neurology 2010;75:1166–73. [11] LaFrance Jr, Rusch D, Machan JT. What is ‘treatment as usual’ for nonepileptic seizures? Epilepsy Behav 2008;12:388–94. [12] Reuber M. Psychogenic nonepileptic seizures: answers and questions. Epilepsy Behav 2008;12:622–35. [13] Reuber M, Mitchell AJ, Howlett S, Elger CE. Measuring outcome in psychogenic nonepileptic seizures: how relevant is seizure remission? Epilepsia 2005;46(11): 1788–95. [14] McKenzie P, Oto M, Russell A, Pelosi A, Duncan R. Early outcomes and predictors in 260 patients with psychogenic nonepileptic attacks. Neurology 2010;74(1):64–9. [15] Reuber M, Pukrop R, Bauer J, Helmstaedter C, Tessendorf N, Elger CE. Outcome in psychogenic nonepileptic seizures: 1 to 10-year follow-up in 164 patients. Ann Neurol 2003;53(3):305–11. [16] O'Sullivan SS, Spillane JE, McMahon EM, Sweeney BJ, Galvin RJ, McNamara B, et al. Clinical characteristics and outcome of patients diagnosed with psychogenic nonepileptic seizures: a 5-year review. Epilepsy Behav 2007;11(1):77–84. [17] Turner K, Piazzini A, Chiesa V, Barbieri V, Vignoli A, Gardella E, et al. Patients with epilepsy and patients with psychogenic non-epileptic seizures: video-EEG, clinical and neuropsychological evaluation. Seizure 2011;20:706–10. [18] Kanner AM. More controversies on the treatment of psychogenic pseudoseizures: an addendum. Epilepsy Behav 2003;4:360–4. [19] Hall-Patch L, Brown R, House A, Howlett S, Kemp S, Lawton G, et al. Acceptability and effectiveness of a strategy for the communication of the diagnosis of psychogenic nonepileptic seizures. Epilepsia 2010;51:70–8. [20] Duncan R, Razvi S, Mulhern S. Newly presenting psychogenic nonepileptic seizures: incidence, population characteristics, and early outcome from a prospective audit of a first seizure clinic. Epilepsy Behav 2011;20:308–11. [21] Leeman BA. Provocative techniques should not be used for the diagnosis of psychogenic nonepileptic seizures. Epilepsy Behav 2009;15:110–4. [22] Cohen RJ, Suter C. Hysterical seizures: suggestion as a provocative EEG test. Ann Neurol 1982;11(4):391–5. [23] Walczak TS, Williams DT, Berten W. Utility and reliability of placebo infusion in the evaluation of patients with seizures. Neurology 1994;44(3 Pt 1):394–9. [24] Durrant J, Rickards H, Cavanna AE. Prognosis and outcome predictors in psychogenic nonepileptic seizures. Epilepsy Res Treat 2011:1–7 [Article ID 274736].
