- Email: [email protected]
Utility of different seizure induction protocols in psychogenic nonepileptic seizures
+Model
ARTICLE IN PRESS
EPIRES-5110; No. of Pages 8 Epilepsy Research (2014) xxx, xxx—xxx
journal homepage: www.elsevier.com/locate/epilepsyres
Utility of different seizure induction protocols in psychogenic nonepileptic seizures Gourav Goyal, Jayantee Kalita, Usha K. Misra ∗ Department of Neurology, Sanjay Gandhi Post Graduate Institute of Medical Sciences, Lucknow, India Received 5 December 2012; received in revised form 3 February 2014; accepted 28 February 2014
KEYWORDS Pseudoseizure; Provocative test; Saline injection; Patch test; Tuning fork test; Epilepsy
Summary Psychogenic non epileptic seizure (PNES) can be induced by several induction tests but their relative usefulness has not been evaluated. In this study, we report the sensitivity and specificity of various induction tests in the diagnosis of PNES and assess their discomfort level. The induction tests were: (a) compression of temple region (CTR), (b) verbal suggestion (VS), (c) tuning fork application (TFA), (d) moist swab application (MSA), (e) torch light stimulation (TLS) and (f) saline injection (SI). Up to 3 trials were done for each test except for normal saline injection which was given once. For comparison of these tests, patients with epileptic seizures were included as controls. The time to precipitate PNES was recorded and patients’ discomfort levels were noted on a 0—10 scale. Video EEG was recorded in the PNES patients. 140 patients with PNES and 50 controls with epileptic seizures were included. The diagnostic yield of CTR was 65.7%, TFA 61.4%, MSA 60.7%, SI 55.6%, VS 54.3% and TLS 40.7%. These tests did not induce seizures in the controls. All these tests had 100% specificity and 100% positive predictive value in the diagnosis of PNES. The maximum discomfort was reported with SI and minimum with MSA. The similarity of efficacy and discomfort with CTR and TFA appear to be the most optimal induction techniques when compared with VS, AMS, TLS, and SI. © 2014 Elsevier B.V. All rights reserved.
∗
Corresponding author at: Department of Neurology, Sanjay Gandhi Post Graduate Institute of Medical Sciences, Raebareily Road, Lucknow 226014. Tel.: +91 522 2494167; fax: +91 0522 2668017. E-mail addresses: [email protected] (G. Goyal), [email protected] (J. Kalita), [email protected], [email protected] (U.K. Misra). http://dx.doi.org/10.1016/j.eplepsyres.2014.02.015 0920-1211/© 2014 Elsevier B.V. All rights reserved.
Please cite this article in press as: Goyal, G., et al., Utility of different seizure induction protocols in psychogenic nonepileptic seizures. Epilepsy Res. (2014), http://dx.doi.org/10.1016/j.eplepsyres.2014.02.015
+Model EPIRES-5110; No. of Pages 8
ARTICLE IN PRESS
2
G. Goyal et al.
Introduction Psychogenic non epileptic seizure (PNES) is defined as psychologically determined clinical events that resemble epileptic attacks but are not associated with paroxysmal physiologic cerebral dysfunction (Bodde et al., 2009). PNES is often misdiagnosed as epilepsy and may have prevalence as high as 33/100,000 population (Benbadis and Hauser, 2000). PNES accounts for 18—23% of patients at epilepsy referral centers (Smith et al., 1999; Benbadis and Hauser, 2000; Leis et al., 1992). Missing the diagnosis of PNES has significant impact on the patients as well as on the health care providers. It may expose the patient to antiepileptic drugs and their side effects and increase the health care cost. Failure to recognize psychological nature of PNES delays the implementation of correct treatment. Misdiagnosis of these patients as epilepsy results in social stigma and frustration (Abubakr et al., 2003). The diagnosis of PNES is challenging and requires detailed evaluation to exclude epileptic seizures. Sometimes, PNES may coexist with epileptic seizures. The seizure semiology is although helpful in distinguishing PNES from epileptic seizure but may be difficult at times. The features favoring PNES are bizarre side to side head movement, closed or forcibly closed eyes, presence of light reflex, vocalization, tongue bite at the tip rather than on the side, synchronized incoordinated limb movements, absence of incontinence, longer duration of attack up to several hours, post ictal shallow irregular breathing, ability to recall the events, and absence of confusion or lethargy (Ali et al., 2011). The interpretation of EEG in the patients with PNES may be difficult due to associated movement artifacts. The patients with epilepsy and PNES are best distinguished by capturing the clinical and EEG events by video-EEG. Video EEG although has high interrater reliability for the diagnosis of epileptic seizure but has only moderate reliability in the diagnosis of PNES (Benbadis et al., 2009a). In one study, epileptologists had higher sensitivity (96%) to identify the epileptic seizure but had low specificity (50%) highlighting that epileptologists were more likely to diagnose PNES as an epileptic seizure (Deacon et al., 2003). Sometimes, clinical attacks are not captured during video EEG. Postictal or interictal EEG is normal in 15—30% of patients with epileptic seizures (Binnie and Prior, 1994). In view of these limitations of video EEG in the diagnosis of PNES, a simple and cost effective clinical test may be helpful. Normal video-EEG recording during the typical spontaneous or provoked seizure is almost diagnostic of PNES (Krumholz, 1999). Sometimes, PNES does not occur spontaneously despite long video-EEG recording and leads to inconclusive result. In resource poor countries, video-EEG is not widely available. Combining induction tests with videoEEG improves the diagnostic yield and shortens the duration of recording, its cost as well as waiting period. In the literature, various induction tests such as compression of the temple, verbal suggestion, tuning fork application, moist swab application, torch light stimulation and saline injection have been used to diagnose PNES but there is paucity of studies evaluating the comparative usefulness of these tests (Walczak et al., 1994; Zaidi et al., 1999; Devinsky and Fisher, 1996; Benbadis et al., 2000). Only few studies have evaluated the sensitivity and specificity of normal saline injection
to induce the clinical attack (Lancman et al., 1994; Walczak et al., 1994; Bazil et al., 1994; Slater et al., 1995). The various induction tests used in the diagnosis of PNES may have different sensitivity and specificity as well as discomfort level. In this communication, we report the sensitivity, specificity and discomfort level of a number of induction tests in the patients with PNES.
Methods This is a prospective, single center study comparing the sensitivity, specificity and discomfort level of different induction tests in the patients with PNES. The research protocol was approved by the Institutional Ethics Committee. Informed consent was obtained from all the patients. For children (8—18 years of age), informed consent was obtained from their parents or guardians.
Inclusion criteria Consecutive patients with a possible diagnosis of PNES were included from the neurology out-patient and emergency service during 1st July 2009 to 31st December 2011. PNES was defined an observable abrupt paroxysmal change in the behavior or consciousness, that resembles an epileptic seizure, but not accompanied by the EEG changes of epileptic seizure or clinical evidence of epilepsy and there is positive evidence or a strong suspicion for psychogenic factors that may have caused the seizure (Bodde et al., 2009).
Exclusion criteria Children below 8 years and patients with mental retardation or dementia were excluded because of communication and assessment problem. The patients with underlying serious medical conditions such as cardiac, renal and liver failure and malignancy were also excluded. The patients with both PNES and epilepsy were excluded to avoid any conflicting results. The patients with underlying serious psychiatric disorders such as schizophrenia or mania and autistic spectrum disorders were also excluded.
Controls Age and gender matched patients with established diagnosis of epilepsy based on clinical and EEG findings with normal cognitive function were included as controls.
Evaluation Demographic information including age, gender, education, occupation and residence of the PNES patients and the controls (epileptic patients) were noted. Detailed medical history including duration of illness, number of attacks, precipitating factors and ictal semiology were recorded. Video EEG was performed in the patients with PNES.
Please cite this article in press as: Goyal, G., et al., Utility of different seizure induction protocols in psychogenic nonepileptic seizures. Epilepsy Res. (2014), http://dx.doi.org/10.1016/j.eplepsyres.2014.02.015
+Model EPIRES-5110; No. of Pages 8
ARTICLE IN PRESS
Psychogenic nonepileptic seizures
3
Induction protocol
Moist swab application (MSA)
The patients were explained about the induction tests and were told that they might experience an episode due to cardiac, respiratory, neurological or psychological factors. A number of commonly used tests will be administered which are likely to precipitate their usual attack that they are suffering from. Observation of these attacks will help in correct diagnosis and categorization of their seizure. At most these procedures may produce some discomfort but will produce no harm or lasting effect. Induction tests which were denied by the patients were not performed. All these tests were performed in a single session in a preset order by one investigator (GG) to avoid inter rater variability. The number of attempts of each test (maximum of 3) was noted except for saline injection. An alarm clock was used to record the time interval between the initiation of the test and precipitation of the event. Each attempt was terminated after 90 s if the event was not induced except normal saline injection which was given only one trial with waiting period of 10 min to avoid discomfort of repeated saline injections. Arbitrary limit of 90 s was chosen for each stimulus except normal saline injection to decrease the time of the procedure as 3 consecutive stimuli were used in the same sitting. Number of attempts for each stimulus was noted. After successful induction, subsequent attempt of the stimulus was not given. The discomfort level was assessed by a questionnaire which was developed on the basis of the other questionnaires used for the assessment of discomfort during diagnostic testing (Scarpero et al., 2005; Schonenberger et al., 2007; Vella, 1991; Al et al., 2009). For each test, a numerical scale of 0—10 was used to assess the level of physical and psychological discomfort, 0 suggested no discomfort and 10 maximum discomfort. The induction tests were performed in the following order:
Saline soaked moist swab was applied to the temple preceded by explanation that this maneuver will induce the event. Arbitrarily time limit of 90 seconds was kept to carry out an attempt and maximum 3 attempts were made before declaring induction failure.
Compression of temple region (CTR) Temple region was gently massaged by finger after explaining that this maneuver will induce the event. Arbitrarily time limit of 90 s was used to perform an attempt and maximum of 3 attempts were made before declaring induction failure.
Verbal suggestion (VS) Patient was asked to close eyes, to think about the event and verbally suggested that this will induce the event. Arbitrarily time limit of 90 s was kept to carry out the test and maximum of 3 attempts were made before declaring induction failure.
Torch light stimulation (TLS) Torch light was used for photic stimulation to avoid induction of epileptic seizures as stroboscopic photic stimulation may induce seizures in the patients with epilepsy. Torch light was put in front of both eyes approximately 10 cm away from the eye-line preceded by explanation that this maneuver will induce the event. Each attempt was completed within 90 s and maximum of 3 attempts were taken before declaring induction failure.
Saline injection (SI) 2 ml normal saline was injected in the antecubital vein over 1 min preceded by explanation that this maneuver will induce the event. Ten minutes were used arbitrarily as a waiting time before considering induction failure. Only 1 attempt was made to avoid patient’s discomfort by repeated injection.
Semiology The semiology of induced events was verified by a witness or on the basis of a recorded video of previous event. Any atypical feature in the semiology was noted. Induced events with different semiology were not considered as successful induction and have not been included in the analysis. The events were categorized into following six groups according to the semiology (Hubsch et al., 2011). (1) Hyperkinetic attacks (with or without hyperventilation), (2) Dystonic attacks (limb or axial) with or without primitive gestural activities. (3) Prolonged atonic seizures (with or without awareness). (4) Pseudosyncope. (5) Myoclonic pseudoseizure (6) Aphonic attacks.
Tolerability The physical and psychological discomfort level of different induction tests and patients’ perception for induction tests were assessed using a questionnaire (Tables 4A and 4B).
Tuning fork application (TFA)
Statistical analysis
Vibrating tuning fork (126 Hz) was applied to the vertex preceded by explanation that this maneuver will induce the event. Arbitrarily time limit of 15 s was kept to carry out an attempt as tuning fork usually vibrates up to 15 s and maximum of 3 attempts were made before declaring induction failure.
The results of various induction tests between the patients with PNES and epileptic seizures were compared by 2 test if the variables were categorical and independent t test if continuous. The variable was considered significant if the two tailed P value was <0.05. Sensitivity, specificity, positive predictive value and negative predictive value were
Please cite this article in press as: Goyal, G., et al., Utility of different seizure induction protocols in psychogenic nonepileptic seizures. Epilepsy Res. (2014), http://dx.doi.org/10.1016/j.eplepsyres.2014.02.015
+Model EPIRES-5110; No. of Pages 8
ARTICLE IN PRESS
4
G. Goyal et al. Table 1 Comparison of demographic and clinical findings between psychogenic nonepileptic seizure (PNES) and patients with epileptic seizure. Demographic variables
PNES n = 140 (%)
Epileptic seizures n = 50 (%)
P value
Age (years) Female Rural Education 1. Illiterate 2. Upto 12th class 3. >12th class Duration of illness (mean ± SD) months
21.38 ± 10.83 100 (71.4) 86 (61.4)
21.18 ± 9.98 17 (34) 27 (54)
0.91 <0.0001 0.62
11 (7.8) 106 (75.7) 23 (16.4) 20.2 ± 51.3 months
3 (6) 34 (68) 13 (926) 56.9 ± 46.6 months
62 (44.3) 41 (29.3) 37 (26.4) 76.22 ± 147.30
6 (12) 8 (16) 36 (72) 7.1 ± 5.18
Frequency of attacks 1. >1 episode/day 2. 1—7 episode/week 3. <1/week Duration of individual attack (minutes)
calculated for each induction test. The statistical analysis was done using SPSS 15 version software.
Results 140 patients with PNES were included and their findings were compared with 50 controls who had epileptic seizure. The patients with epilepsy were on carbamazepine, sodium valproate or clobazam in isolation or in combination and were seizure free for at least 6 months. The mean age of the patients with PNES was similar to the controls (21.5 vs 21.2; P = 0.91) but PNES patients were more commonly females (71% vs 34% P < 0.0001), had shorter duration of illness (mean 20.2 ± 51.3 months vs 56.9 ± 46.6 months, P < 0.0001) and more frequent (P < 0.001) and long lasting attacks (mean 76.2 vs 7.1 min, P = 0.001) compared to the patients with epileptic seizure. The demographic and clinical variables in the PNES and control groups are presented in Table 1.
Semiology The commonest semiology of PNES was hyperkinetic (with or without hyperventilation) in 41 (29.3%) followed by prolonged atonic seizures (with or without awareness) in 37 (26.4%), dystonic psychogenic seizure (with or without primitive gestures) in 32 (22.9%), pseudosyncope in 26 (18.6%), aphasia in 3 (2.1%) and myoclonic jerks in 1 (0.7%) patient. In the control group, 14 (28%) patients had idiopathic generalized tonic clonic seizure, 20 (40%) partial seizure with secondary generalized, 6 (12%) complex partial seizure and 10 (20%) had simple partial seizure.
Induction tests The seizures were precipitated by induction tests in 95 (67.9%) patients with PNES whereas none in the control group. The induced events were similar to the earlier events in 84 out of 95 (88.4%) patients and 11.6% patients had
0.48 <0.0001
<0.001 0.001
different semiology. In the control group, none had induction of seizures but 19 (38.8%) had nonspecific symptoms such as light headedness, uneasiness and palpitation. Amongst the induction test, CTR had the highest sensitivity (65.7%) and TLS the lowest (40.7%) (P < 0.001). The sensitivity of VS was 54.3%, TFA 61.4%, MSA 60.7% and SI 55.6%. Though sensitivity of CTR was more compared to TFA application, the difference was not statistically significant (P = 0.58). Difference between sensitivity of CTR and MSA was statistically not significant (P = 0.35). SI had significantly lower yield compared to CTR (P < 0.05) and TFA (P < 0.05). Twenty-three patients with PNES and 12 in control group did not consent for saline injection. All the induction tests had 100% specificity and 100% positive predictive value. The highest negative predictive value was 51% for CTR and the lowest negative predictive value for TLS (37.6%). The yield of different induction tests is presented in Table 2. The mean (SD) number of attempts to induce seizures was 1.08 ± 0.32 for VS, 1.14 ± 0.46 for CTR, 1.19 ± 0.47 for TFA, 1.06 ± 0.28 for MSA and 1.04 ± 0.19 for TLS. The number of repetitions needed to induce seizure was not significantly different in various induction tests (P = 0.48). Torch light stimulation needed maximum time (23.9 ± 9.7 s) and TFA test needed the least time (8.2 ± 1.9 s, P = 0.001). The number of attempts and time needed to induce PNES are presented in Table 3.
Tolerability of the induction tests The physical and psychological discomfort produced by different tests was assessed in 50 patients. Only 9 patients were anxious or apprehensive before the induction. Eleven patients did not report anxiety or psychological discomfort after the induction. Saline injection and TLS produced maximum anxiety or psychological discomfort in 13 and 10 patients respectively. The mean discomfort level during induction was 2.6 ± 2 and none of the patients had discomfort level above 5. The maximum discomfort was produced by SI (3.2 ± 3.0) and minimum by MSA (0.89 ± 1.46). This
Please cite this article in press as: Goyal, G., et al., Utility of different seizure induction protocols in psychogenic nonepileptic seizures. Epilepsy Res. (2014), http://dx.doi.org/10.1016/j.eplepsyres.2014.02.015
+Model
ARTICLE IN PRESS
EPIRES-5110; No. of Pages 8
5 difference is significant (P < 0.0001). The majority of subjects felt that the SI was most painful (P < 0.001) while 6 felt no pain by any of the induction tests. Most of the patients were satisfied with these induction tests, willing to undergo again if needed in future and none considered it psychologically traumatic. The results of the physical and psychological discomfort of the induction tests are summarized in Tables 4A and 4B.
51.02 43.86 48.07 47.62 49.02 37.59
NPV (%)
Discussion
100 100 100 100 100 100
PPV (%)
100 100 100 100 100 100
Specificity (%)
(100) (100) (100) (100) (100) (100) 50 50 50 50 38 50 PPV, positive predictive value; NPV, negative predictive value.
0 0 0 0 0 0 48 (34.3) 64 (45.7) 54 (38.6) 55 (39.3) 52 (44.4) 83 (59.3) 92 (65.7) 76 (54.3) 86 (61.4) 85 (60.7) 65 (55.6) 57 (40.7) Compression of temple region Verbal suggestions Tuning fork application Application of moist swab Saline injection Torch light stimulation
Induced N (%)
Not induced N (%)
Induced N (%)
Not induced N (%)
65.71 54.29 61.43 60.71 55.56 40.71
Sensitivity (%) Epileptic seizures PNES Induction methods
Table 2 Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) of different induction test in the patients with psychogenic nonepileptic seizure (PNES).
Psychogenic nonepileptic seizures
In the present study, CTR was the most effective in inducing PNES with a sensitivity of 65.7%, specificity of 100%, positive predictive value of 100% and negative predictive value of 51%. Tuning fork application test needed the least time for induction though number of attempts of different induction methods was not significantly different. The least discomfort was produced by MSA and maximum by SI. The present study for the first time evaluated the sensitivity, specificity and discomfort level of a number of induction tests in the patients with PNES. The higher specificity of induction tests in our study is due to lack of seizure induction in the control. The patients with epileptic seizure were on remission for at least 6 months. In an earlier study, the usefulness of patch and massage on the patch in precipitating PNES were evaluated in 91 PNES patients and 20 patients with epileptic seizure. The induction test was positive in 77.4% of PNES and none with epileptic seizure resulting in 100% specificity and 48.7% negative predictive value. The patients’ discomfort however was not evaluated in this study (Lancman et al., 1994). In another study, hypnosis was used to induce PNES in 69 patients of 3 groups (PNES, epilepsy, epilepsy + PNES). Hypnotic procedure had 77% sensitivity and 95% specificity for PNES (Barry et al., 2000). Placebo induced episodes were typical of prior PNES episodes in 90% whereas 10% had atypical episodes. Fifteen percent in epilepsy group had atypical and confusing events in the same study (Walczak et al., 1994). In another study, higher frequency (30%) of atypical attacks was reported in the patients with PNES (Drake, 1985). In the present study, 88.4% PNES patients had typical event and 11.6% atypical. Lesser proportion of atypical event in PNES group in our study may be due to uniform induction protocol to all the patients and verification of the induced event. The induced attack should be verified by a witness, family member or compared with the prior video records which are often available through mobile phone. Induction tests have been reported as a stressful experience and may lead to increased pulse rate, blood pressure and physiological tremor (Walczak et al., 1994). Emotional stress may trigger seizures in epilepsy patients as reported in various retrospective and prospective studies (Friis and Lund, 1974; Currie et al., 1971; Temkin and Davis, 1984). This may explain seizure induction even in some epilepsy patients. None of our epilepsy patient had induction of seizure though 38.8% had nonspecific symptoms such as light headedness, uneasiness and palpitation. These symptoms could be the manifestation of emotional stress of induction procedure and might be confused with aura of seizure attack.
Please cite this article in press as: Goyal, G., et al., Utility of different seizure induction protocols in psychogenic nonepileptic seizures. Epilepsy Res. (2014), http://dx.doi.org/10.1016/j.eplepsyres.2014.02.015
+Model
ARTICLE IN PRESS
EPIRES-5110; No. of Pages 8
6
G. Goyal et al. Table 3 Number of attempts and induction time of different induction tests in the patients with psychogenic nonepileptic seizure (PNES). Induction method
Total number of patients included (n)
Successful attempts N (%)
Consent was not given N (%)
Mean attempts
Verbal suggestion
140
76 (54.3)
—
1.08 ± 0.32
Compression of temple region
140
92 (65.7)
—
1.14 ± 0.46
Tuning fork application
140
86 (61.4)
—
1.19 ± 0.47
8.15 ± 1.93
Application of moist swab
140
85 (60.7)
—
1.06 ± 0.28
17.86 ± 9.32
Saline injection
117
65 (55.6)
23 (19.7)
1.00
17.46 ± 7.04
Torch light stimulation
140
57 (40.7)
—
1.04 ± 0.19
23.91 ± 9.74
Table 4A
P value
Mean time required to induce
P value
18.33 ± 9.63 0.48
17.64 ± 10.05
0.001
Assessment of discomfort of induction tests in the patients with psychogenic nonepileptic seizure.
Questions
CTR
VS
AMS
TFA
Which protocol produced maximum anxiety or psychological discomfort (no. of patients) Discomfort level (0—10 scale) Which protocol was most painful: (no. of patients; consent for SI was not given by 13)
8
0
3
5
1.33 ± 1.73
1.17 ± 1.99
0.89 ± 1.46
1.65 ± 2.23
1.76 ± 2.36
2
0
4
3
3
TLS 10
SI 13
3.24 ± 3.02 33
P value 0.04
0.01 <0.001
CTR, compression of temple region; VS, verbal suggestion; TFA, tuning fork application; AMS, application of moist swab; SI, saline injection; TLS, torch light stimulation.
Table 4B
Overall acceptability of the induction tests which was assessed in patients with psychogenic nonepileptic seizure.
Questions
Strongly agreed
Agreed
Acceptable
Disagree
Overall seizure induction protocol is patient friendly and satisfactory If needed in future, will you like to undergo such test? Are these procedures left you with mental trauma? Were the procedures fully explained to you?
15
27
5
3
0.001
21 0 8
20 4 39
3 2 3
6 31 0
0.001 <0.001 <0.001
P value
Please cite this article in press as: Goyal, G., et al., Utility of different seizure induction protocols in psychogenic nonepileptic seizures. Epilepsy Res. (2014), http://dx.doi.org/10.1016/j.eplepsyres.2014.02.015
+Model EPIRES-5110; No. of Pages 8
ARTICLE IN PRESS
Psychogenic nonepileptic seizures The majority of the studies have evaluated one or two tests for seizure induction in PNES. The sensitivity of SI ranges between 33% and 91% (Slater et al., 1995; Bhatia et al., 1997). In our study, the sensitivity of saline injection was 55.6% with 100% specificity. Thirty-five of our patients however refused to undergo this test. Invasiveness of the SI and associated discomfort may be the reason of refusal. This emphasizes need for other induction tests to substitute the widely used placebo SI for the diagnosis of PNES. Torch light stimulation although is a good alternative as it is painless and acceptable to most of the patients but is limited by longer procedural time and lower sensitivity of 40.7%. Moist swab application (60.7%) and TFA (61.4%) had somewhat similar sensitivity and specificity; however, TFA resulted in more anxiety and discomfort compared to MSA. In the reported literature, none of the studies have compared relative discomfort of various induction tests. The sensitivity of induction tests was higher in village dwellers, housewives and less educated (<12 standard) patients. These backgrounds may make them more suggestible to the provocative tests. We have not included the patients who had both epileptic seizures and PNES to avoid conflicting results. Little work has been done on underlying pathophysiology of inducibility of PNES. In our study, 67.9% patients with PNES had inducibility. It raises some questions like why some patients with PNES had inducibility and not the others, what are the predictors of inducibility and what is the impact of inducibility on long term outcome? In a recent study, hypermotor ictal semiology, self-reporting of uncommon cognitive and affective symptoms, higher tendency to rely on coping strategies of ‘‘instrumental support’’ and ‘‘active coping’’ have been reported as the predictors of successful induction (Chen et al., 2011). Use of induction test has raised ethical concerns as it may be associated with deception, violates patients’ autonomy, undermines the doctor-patient relationship and psychological harm to the patient (Leeman, 2009). The ethical practice is to do good to the patient (beneficence) and do no harm to the patient (malfeasance). Despite criticism and ethical concerns of nondisclosure that is inherent in provocative tests, violation of doctor patient trust and feeling of deception by some, it may be more harmful to miss a diagnosis when it can be made (Devinsky and Fisher, 1996). Missing the diagnosis of PNES may cause serious socioeconomic, psychological and medical consequences. In a study on 313 patients with PNES, there was an average delay of 7.2 years in the diagnosis (Reuber et al., 2002). Delay in the diagnosis of PNES may lead to inappropriate treatment, expose the patients to drug toxicity and rarely these patients wrongly intubated for psychogenic status epilepticus. Moreover there may be demoralization, low self-esteem and economic loss. These possibilities outweigh the ethical concern of induction of PNES. Ethical concern regarding psychological harm and damage to doctor patient relationship is rather theoretical. If the correct diagnosis is achieved by noninvasive or minimally invasive induction test without producing harm or much discomfort, it may be advantageous to the patients. The principle of autonomy is that if anyone who is influenced by the consequence of the decision should be involved in the decision making process. We have informed the patients about the induction test and they gave their consent.
7 Moreover the utility of these tests has been already reported. In the present study, we have not used any new induction test but compared the relative usefulness and discomfort of the various induction tests in the diagnosis of PNES. In a study, patients found these techniques useful and not harmful (Cohen et al., 1992). In the present study, most of the patients were comfortable during induction procedure and denied significant psychological trauma. Recently, provocative tests were advocated to use in the diagnosis of PNES (Benbadis, 2009b). Provocative tests during video EEG recording may shorten the evaluation time. In clinical settings, usefulness of these tests outweighs the ethical concerns by avoiding misdiagnosis of epilepsy in PNES patients, reducing uncertainty about the diagnosis and preventing long term cognitive and physical health hazards of unnecessary antiepileptic drug therapies. Devinsky and Fisher reviewed the ethical use of placebos and provocative tests in the diagnosis of PNES (Devinsky and Fisher, 1996). The present study has some limitations. All the induction tests were performed in prefixed order by a single investigator without blinding which may lead to a bias. Performing all the induction tests in a prefixed manner may lea d to higher sensitivity of the initial induction tests compared to the later. Similarly, performing a number of induction tests in a single setting may result in fatigue at the end and may influence the sensitivity of the tests by attenuating the response for later stimuli. There is also possibility of suggestibility through nonverbal cues in some of the induction tests such as CTR and application of TFA. Further studies are required to overcome these limitations and to confirm these findings.
Conclusion In a busy clinical practice, CTR and TFA are comparatively better induction tests in the diagnosis of PNES because these are sensitive, easy to perform and have minimal discomfort. Surprisingly widely used provocation test ‘‘SI’’ is significantly less efficacious, comparatively more cumbersome and least tolerable.
Ethics approval The research has been approved by the Institutional Ethics Committee, SGPGIMS, Lucknow.
Conflict of interest There is no conflict of interest to declare.
Funding support None.
Acknowledgement We acknowledge the help of Mr. Rakesh Nigam for secretarial help.
Please cite this article in press as: Goyal, G., et al., Utility of different seizure induction protocols in psychogenic nonepileptic seizures. Epilepsy Res. (2014), http://dx.doi.org/10.1016/j.eplepsyres.2014.02.015
+Model EPIRES-5110; No. of Pages 8
ARTICLE IN PRESS
8
G. Goyal et al.
References Abubakr, A., Kablinger, A., Caldito, G., 2003. Psychogenic seizures: clinical features and psychological analysis. Epilepsy Behav. 4, 241—245. Al, R.A., Yalvac, S., Altar, O.Y., Dolen, I., 2009. Perceived pain and anxiety before and after amniocentesis among pregnant Turkish women. Clin. Exp. Obstet. Gynecol. 36 (3), 184—186. Ali, S., Jabeen, S., Arian, A., Wassef, T., Ibrahim, A., 2011. How to use your clinical judgment to screen for and diagnose psychogenic nonepileptic seizures without video electroencephalogram. Innov. Clin. Neurosci. 8, 36—42. Barry, J.J., Atzman, O., Morrell, M.J., 2000. Discriminating between epileptic and non-epileptic events: the utility of hypnotic seizure induction. Epilepsia 41, 81—84. Bazil, C.W., Kothari, M., Luciano, D., Moroney, J., Song, S., Vasquez, B., Weinreb, H.J., Devinsky, O., 1994. Provocation of nonepileptic seizures by suggestion in a general seizure population. Epilepsia 35, 768—770. Benbadis, S.R., LaFrance Jr., W.C., Papandonatos, G.D., Korabathina, K., Lin, K., Kraemer, H.C., 2009a. NES Treatment Workshop. Interrater reliability of EEG-video monitoring. Neurology 73, 843—846. Benbadis, S.R., 2009b. Provocative techniques should be used for the diagnosis of psychogenic nonepileptic seizures. Epilepsy Behav. 15, 106—109. Benbadis, S.R., Hauser, W.A., 2000. An estimate of the prevalence of psychogenic non-epileptic seizures. Seizure 9, 280—281. Benbadis, S.R., Johnson, K., Anthony, K., Caines, G., Hess, G., Jackson, C., Vale, F.L., Tatum 4th, W.O., 2000. Induction of psychogenic nonepileptic seizures without placebo. Neurology 55, 1904—1905. Bhatia, M., Sinha, P.K., Jain, S., Padma, M.V., Maheshwari, M.C., 1997. Usefulness of short-term video EEG recording with saline induction in pseudoseizures. Acta Neurol. Scand. 95, 363—366. Binnie, C.D., Prior, P.F., 1994. Electroencephalography. J. Neurol. Neurosurg. Psychiatry 57, 1308—1319. Bodde, N.M.G., Brooks, J.L., Baker, G.A., Boon, P.A.J.M., Hendriksen, J.G.M., Mulder, O.G., Aldenkamp, A.P., 2009. Psychogenic non-epileptic seizures—–Definition, etiology, treatment and prognostic issues: a critical review. Seizure 18, 543—553. Chen, D.K., Izadyar, S., Collins, R.L., Benge, J.F., LeMaire, A.W., Hrachovy, R.A., 2011. Induction of psychogenic nonepileptic events: success rate influenced by prior induction exposure, ictal semiology, and psychological profiles. Epilepsia 52, 1063—1070. Cohen, L.M., Howard 3rd, G.F., Bongar, B., 1992. Provocation of pseudoseizures by psychiatric interview during EEG and video monitoring. Int. J. Psychiatry Med. 22 (2), 131—140. Currie, S., Heathfield, K.W.G., Hanson, R.A., Scott, D.F., 1971. Clinical course and prognosis of temporal lobe epilepsy. Brain 94, 173—190.
Deacon, C., Wiebe, S., Blume, W.T., McLachlan, R.S., Young, G.B., Matijevic, S., 2003. Seizure identification by clinical description in temporal lobe epilepsy: how accurate are we? Neurology 61, 1686—1689. Devinsky, O., Fisher, R., 1996. Ethical use of placebos and provocative testing in diagnosing nonepileptic seizures. Neurology 47, 866—870. Drake, M.E., 1985. Saline activation of pseudoepileptic seizures: clinical and neuropsychiatric observations. Clin. Electroencephalogr. 16, 171—176. Friis, M.L., Lund, M., 1974. Stress convulsions. Arch. Neurol. 31, 155—159. Hubsch, C., Baumann, C., Hingray, C., Gospodaru, N., Vignal, J.P., Vespignani, H., Maillard, L., 2011. Clinical classification of psychogenic non-epileptic seizures based on video-EEG analysis and automatic clustering. J. Neurol. Neurosurg. Psychiatry 82, 955—960. Krumholz, A., 1999. Nonepileptic seizures: diagnosis and management. Neurology 53, S76—S83. Lancman, M.E., Asconape, J.J., Craven, W.J., Howard, G., Penry, J.K., 1994. Predictive value of psychogenic seizures by suggestion. Ann. Neurol. 35, 359—361. Leeman, B.A., 2009. Provocative techniques should not be used for the diagnosis of psychogenic nonepileptic seizures. Epilepsy Behav. 15, 110—114. Leis, A.A., Ross, M.A., Summers, A.K., 1992. Psychogenic seizures: ictal characteristics and diagnostic pitfalls. Neurology 42, 95—99. Reuber, M., Fernández, G., Bauer, J., Helmstaedter, C., Elger, C.E., 2002. Diagnostic delay in psychogenic nonepileptic seizures. Neurology 58, 493—495. Scarpero, H.M., Padmanabhan, P., Xue, X., Nitti, V.W., 2005. Patient perception of videourodynamic testing: a questionnaire based study. J. Urol. 173 (2), 555—559. Schonenberger, E., Schnapauff, D., Teige, F., Laule, M., Hamm, B., Dewey, M., 2007. Patient acceptance of noninvasive and invasive coronary angiography. PLoS ONE 2 (2), e246. Slater, J.D., Brown, M.C., Jacobs, W., Ramsay, R.E., 1995. Induction of pseudoseizures with intravenous saline placebo. Epilepsia 36, 580—585. Smith, D., Defalla, B.A., Chadwick, D.W., 1999. The misdiagnosis of epilepsy and the management of refractory epilepsy in a specialist clinic. QJM 92, 15—23. Temkin, N.R., Davis, G.R., 1984. Stress as a risk factor for seizures among adults with epilepsy. Epilepsia 25, 450—456. Vella, P.V., 1991. A survey of women undergoing a pelvic examination. Aust. N. Z. J. Obstet. Gynaecol. 31, 355—357. Walczak, T.S., Williams, D.T., Berten, W., 1994. Utility and reliability of placebo infusion in the evaluation of patients with seizures. Neurology 44, 394—399. Zaidi, A., Crampton, S., Clough, P., Fitzpatrick, A., Scheepers, B., 1999. Head-up tilting is a useful provocative test for psychogenic non-epileptic seizures. Seizure 8, 353—355.
Please cite this article in press as: Goyal, G., et al., Utility of different seizure induction protocols in psychogenic nonepileptic seizures. Epilepsy Res. (2014), http://dx.doi.org/10.1016/j.eplepsyres.2014.02.015
ARTICLE IN PRESS
EPIRES-5110; No. of Pages 8 Epilepsy Research (2014) xxx, xxx—xxx
journal homepage: www.elsevier.com/locate/epilepsyres
Utility of different seizure induction protocols in psychogenic nonepileptic seizures Gourav Goyal, Jayantee Kalita, Usha K. Misra ∗ Department of Neurology, Sanjay Gandhi Post Graduate Institute of Medical Sciences, Lucknow, India Received 5 December 2012; received in revised form 3 February 2014; accepted 28 February 2014
KEYWORDS Pseudoseizure; Provocative test; Saline injection; Patch test; Tuning fork test; Epilepsy
Summary Psychogenic non epileptic seizure (PNES) can be induced by several induction tests but their relative usefulness has not been evaluated. In this study, we report the sensitivity and specificity of various induction tests in the diagnosis of PNES and assess their discomfort level. The induction tests were: (a) compression of temple region (CTR), (b) verbal suggestion (VS), (c) tuning fork application (TFA), (d) moist swab application (MSA), (e) torch light stimulation (TLS) and (f) saline injection (SI). Up to 3 trials were done for each test except for normal saline injection which was given once. For comparison of these tests, patients with epileptic seizures were included as controls. The time to precipitate PNES was recorded and patients’ discomfort levels were noted on a 0—10 scale. Video EEG was recorded in the PNES patients. 140 patients with PNES and 50 controls with epileptic seizures were included. The diagnostic yield of CTR was 65.7%, TFA 61.4%, MSA 60.7%, SI 55.6%, VS 54.3% and TLS 40.7%. These tests did not induce seizures in the controls. All these tests had 100% specificity and 100% positive predictive value in the diagnosis of PNES. The maximum discomfort was reported with SI and minimum with MSA. The similarity of efficacy and discomfort with CTR and TFA appear to be the most optimal induction techniques when compared with VS, AMS, TLS, and SI. © 2014 Elsevier B.V. All rights reserved.
∗
Corresponding author at: Department of Neurology, Sanjay Gandhi Post Graduate Institute of Medical Sciences, Raebareily Road, Lucknow 226014. Tel.: +91 522 2494167; fax: +91 0522 2668017. E-mail addresses: [email protected] (G. Goyal), [email protected] (J. Kalita), [email protected], [email protected] (U.K. Misra). http://dx.doi.org/10.1016/j.eplepsyres.2014.02.015 0920-1211/© 2014 Elsevier B.V. All rights reserved.
Please cite this article in press as: Goyal, G., et al., Utility of different seizure induction protocols in psychogenic nonepileptic seizures. Epilepsy Res. (2014), http://dx.doi.org/10.1016/j.eplepsyres.2014.02.015
+Model EPIRES-5110; No. of Pages 8
ARTICLE IN PRESS
2
G. Goyal et al.
Introduction Psychogenic non epileptic seizure (PNES) is defined as psychologically determined clinical events that resemble epileptic attacks but are not associated with paroxysmal physiologic cerebral dysfunction (Bodde et al., 2009). PNES is often misdiagnosed as epilepsy and may have prevalence as high as 33/100,000 population (Benbadis and Hauser, 2000). PNES accounts for 18—23% of patients at epilepsy referral centers (Smith et al., 1999; Benbadis and Hauser, 2000; Leis et al., 1992). Missing the diagnosis of PNES has significant impact on the patients as well as on the health care providers. It may expose the patient to antiepileptic drugs and their side effects and increase the health care cost. Failure to recognize psychological nature of PNES delays the implementation of correct treatment. Misdiagnosis of these patients as epilepsy results in social stigma and frustration (Abubakr et al., 2003). The diagnosis of PNES is challenging and requires detailed evaluation to exclude epileptic seizures. Sometimes, PNES may coexist with epileptic seizures. The seizure semiology is although helpful in distinguishing PNES from epileptic seizure but may be difficult at times. The features favoring PNES are bizarre side to side head movement, closed or forcibly closed eyes, presence of light reflex, vocalization, tongue bite at the tip rather than on the side, synchronized incoordinated limb movements, absence of incontinence, longer duration of attack up to several hours, post ictal shallow irregular breathing, ability to recall the events, and absence of confusion or lethargy (Ali et al., 2011). The interpretation of EEG in the patients with PNES may be difficult due to associated movement artifacts. The patients with epilepsy and PNES are best distinguished by capturing the clinical and EEG events by video-EEG. Video EEG although has high interrater reliability for the diagnosis of epileptic seizure but has only moderate reliability in the diagnosis of PNES (Benbadis et al., 2009a). In one study, epileptologists had higher sensitivity (96%) to identify the epileptic seizure but had low specificity (50%) highlighting that epileptologists were more likely to diagnose PNES as an epileptic seizure (Deacon et al., 2003). Sometimes, clinical attacks are not captured during video EEG. Postictal or interictal EEG is normal in 15—30% of patients with epileptic seizures (Binnie and Prior, 1994). In view of these limitations of video EEG in the diagnosis of PNES, a simple and cost effective clinical test may be helpful. Normal video-EEG recording during the typical spontaneous or provoked seizure is almost diagnostic of PNES (Krumholz, 1999). Sometimes, PNES does not occur spontaneously despite long video-EEG recording and leads to inconclusive result. In resource poor countries, video-EEG is not widely available. Combining induction tests with videoEEG improves the diagnostic yield and shortens the duration of recording, its cost as well as waiting period. In the literature, various induction tests such as compression of the temple, verbal suggestion, tuning fork application, moist swab application, torch light stimulation and saline injection have been used to diagnose PNES but there is paucity of studies evaluating the comparative usefulness of these tests (Walczak et al., 1994; Zaidi et al., 1999; Devinsky and Fisher, 1996; Benbadis et al., 2000). Only few studies have evaluated the sensitivity and specificity of normal saline injection
to induce the clinical attack (Lancman et al., 1994; Walczak et al., 1994; Bazil et al., 1994; Slater et al., 1995). The various induction tests used in the diagnosis of PNES may have different sensitivity and specificity as well as discomfort level. In this communication, we report the sensitivity, specificity and discomfort level of a number of induction tests in the patients with PNES.
Methods This is a prospective, single center study comparing the sensitivity, specificity and discomfort level of different induction tests in the patients with PNES. The research protocol was approved by the Institutional Ethics Committee. Informed consent was obtained from all the patients. For children (8—18 years of age), informed consent was obtained from their parents or guardians.
Inclusion criteria Consecutive patients with a possible diagnosis of PNES were included from the neurology out-patient and emergency service during 1st July 2009 to 31st December 2011. PNES was defined an observable abrupt paroxysmal change in the behavior or consciousness, that resembles an epileptic seizure, but not accompanied by the EEG changes of epileptic seizure or clinical evidence of epilepsy and there is positive evidence or a strong suspicion for psychogenic factors that may have caused the seizure (Bodde et al., 2009).
Exclusion criteria Children below 8 years and patients with mental retardation or dementia were excluded because of communication and assessment problem. The patients with underlying serious medical conditions such as cardiac, renal and liver failure and malignancy were also excluded. The patients with both PNES and epilepsy were excluded to avoid any conflicting results. The patients with underlying serious psychiatric disorders such as schizophrenia or mania and autistic spectrum disorders were also excluded.
Controls Age and gender matched patients with established diagnosis of epilepsy based on clinical and EEG findings with normal cognitive function were included as controls.
Evaluation Demographic information including age, gender, education, occupation and residence of the PNES patients and the controls (epileptic patients) were noted. Detailed medical history including duration of illness, number of attacks, precipitating factors and ictal semiology were recorded. Video EEG was performed in the patients with PNES.
Please cite this article in press as: Goyal, G., et al., Utility of different seizure induction protocols in psychogenic nonepileptic seizures. Epilepsy Res. (2014), http://dx.doi.org/10.1016/j.eplepsyres.2014.02.015
+Model EPIRES-5110; No. of Pages 8
ARTICLE IN PRESS
Psychogenic nonepileptic seizures
3
Induction protocol
Moist swab application (MSA)
The patients were explained about the induction tests and were told that they might experience an episode due to cardiac, respiratory, neurological or psychological factors. A number of commonly used tests will be administered which are likely to precipitate their usual attack that they are suffering from. Observation of these attacks will help in correct diagnosis and categorization of their seizure. At most these procedures may produce some discomfort but will produce no harm or lasting effect. Induction tests which were denied by the patients were not performed. All these tests were performed in a single session in a preset order by one investigator (GG) to avoid inter rater variability. The number of attempts of each test (maximum of 3) was noted except for saline injection. An alarm clock was used to record the time interval between the initiation of the test and precipitation of the event. Each attempt was terminated after 90 s if the event was not induced except normal saline injection which was given only one trial with waiting period of 10 min to avoid discomfort of repeated saline injections. Arbitrary limit of 90 s was chosen for each stimulus except normal saline injection to decrease the time of the procedure as 3 consecutive stimuli were used in the same sitting. Number of attempts for each stimulus was noted. After successful induction, subsequent attempt of the stimulus was not given. The discomfort level was assessed by a questionnaire which was developed on the basis of the other questionnaires used for the assessment of discomfort during diagnostic testing (Scarpero et al., 2005; Schonenberger et al., 2007; Vella, 1991; Al et al., 2009). For each test, a numerical scale of 0—10 was used to assess the level of physical and psychological discomfort, 0 suggested no discomfort and 10 maximum discomfort. The induction tests were performed in the following order:
Saline soaked moist swab was applied to the temple preceded by explanation that this maneuver will induce the event. Arbitrarily time limit of 90 seconds was kept to carry out an attempt and maximum 3 attempts were made before declaring induction failure.
Compression of temple region (CTR) Temple region was gently massaged by finger after explaining that this maneuver will induce the event. Arbitrarily time limit of 90 s was used to perform an attempt and maximum of 3 attempts were made before declaring induction failure.
Verbal suggestion (VS) Patient was asked to close eyes, to think about the event and verbally suggested that this will induce the event. Arbitrarily time limit of 90 s was kept to carry out the test and maximum of 3 attempts were made before declaring induction failure.
Torch light stimulation (TLS) Torch light was used for photic stimulation to avoid induction of epileptic seizures as stroboscopic photic stimulation may induce seizures in the patients with epilepsy. Torch light was put in front of both eyes approximately 10 cm away from the eye-line preceded by explanation that this maneuver will induce the event. Each attempt was completed within 90 s and maximum of 3 attempts were taken before declaring induction failure.
Saline injection (SI) 2 ml normal saline was injected in the antecubital vein over 1 min preceded by explanation that this maneuver will induce the event. Ten minutes were used arbitrarily as a waiting time before considering induction failure. Only 1 attempt was made to avoid patient’s discomfort by repeated injection.
Semiology The semiology of induced events was verified by a witness or on the basis of a recorded video of previous event. Any atypical feature in the semiology was noted. Induced events with different semiology were not considered as successful induction and have not been included in the analysis. The events were categorized into following six groups according to the semiology (Hubsch et al., 2011). (1) Hyperkinetic attacks (with or without hyperventilation), (2) Dystonic attacks (limb or axial) with or without primitive gestural activities. (3) Prolonged atonic seizures (with or without awareness). (4) Pseudosyncope. (5) Myoclonic pseudoseizure (6) Aphonic attacks.
Tolerability The physical and psychological discomfort level of different induction tests and patients’ perception for induction tests were assessed using a questionnaire (Tables 4A and 4B).
Tuning fork application (TFA)
Statistical analysis
Vibrating tuning fork (126 Hz) was applied to the vertex preceded by explanation that this maneuver will induce the event. Arbitrarily time limit of 15 s was kept to carry out an attempt as tuning fork usually vibrates up to 15 s and maximum of 3 attempts were made before declaring induction failure.
The results of various induction tests between the patients with PNES and epileptic seizures were compared by 2 test if the variables were categorical and independent t test if continuous. The variable was considered significant if the two tailed P value was <0.05. Sensitivity, specificity, positive predictive value and negative predictive value were
Please cite this article in press as: Goyal, G., et al., Utility of different seizure induction protocols in psychogenic nonepileptic seizures. Epilepsy Res. (2014), http://dx.doi.org/10.1016/j.eplepsyres.2014.02.015
+Model EPIRES-5110; No. of Pages 8
ARTICLE IN PRESS
4
G. Goyal et al. Table 1 Comparison of demographic and clinical findings between psychogenic nonepileptic seizure (PNES) and patients with epileptic seizure. Demographic variables
PNES n = 140 (%)
Epileptic seizures n = 50 (%)
P value
Age (years) Female Rural Education 1. Illiterate 2. Upto 12th class 3. >12th class Duration of illness (mean ± SD) months
21.38 ± 10.83 100 (71.4) 86 (61.4)
21.18 ± 9.98 17 (34) 27 (54)
0.91 <0.0001 0.62
11 (7.8) 106 (75.7) 23 (16.4) 20.2 ± 51.3 months
3 (6) 34 (68) 13 (926) 56.9 ± 46.6 months
62 (44.3) 41 (29.3) 37 (26.4) 76.22 ± 147.30
6 (12) 8 (16) 36 (72) 7.1 ± 5.18
Frequency of attacks 1. >1 episode/day 2. 1—7 episode/week 3. <1/week Duration of individual attack (minutes)
calculated for each induction test. The statistical analysis was done using SPSS 15 version software.
Results 140 patients with PNES were included and their findings were compared with 50 controls who had epileptic seizure. The patients with epilepsy were on carbamazepine, sodium valproate or clobazam in isolation or in combination and were seizure free for at least 6 months. The mean age of the patients with PNES was similar to the controls (21.5 vs 21.2; P = 0.91) but PNES patients were more commonly females (71% vs 34% P < 0.0001), had shorter duration of illness (mean 20.2 ± 51.3 months vs 56.9 ± 46.6 months, P < 0.0001) and more frequent (P < 0.001) and long lasting attacks (mean 76.2 vs 7.1 min, P = 0.001) compared to the patients with epileptic seizure. The demographic and clinical variables in the PNES and control groups are presented in Table 1.
Semiology The commonest semiology of PNES was hyperkinetic (with or without hyperventilation) in 41 (29.3%) followed by prolonged atonic seizures (with or without awareness) in 37 (26.4%), dystonic psychogenic seizure (with or without primitive gestures) in 32 (22.9%), pseudosyncope in 26 (18.6%), aphasia in 3 (2.1%) and myoclonic jerks in 1 (0.7%) patient. In the control group, 14 (28%) patients had idiopathic generalized tonic clonic seizure, 20 (40%) partial seizure with secondary generalized, 6 (12%) complex partial seizure and 10 (20%) had simple partial seizure.
Induction tests The seizures were precipitated by induction tests in 95 (67.9%) patients with PNES whereas none in the control group. The induced events were similar to the earlier events in 84 out of 95 (88.4%) patients and 11.6% patients had
0.48 <0.0001
<0.001 0.001
different semiology. In the control group, none had induction of seizures but 19 (38.8%) had nonspecific symptoms such as light headedness, uneasiness and palpitation. Amongst the induction test, CTR had the highest sensitivity (65.7%) and TLS the lowest (40.7%) (P < 0.001). The sensitivity of VS was 54.3%, TFA 61.4%, MSA 60.7% and SI 55.6%. Though sensitivity of CTR was more compared to TFA application, the difference was not statistically significant (P = 0.58). Difference between sensitivity of CTR and MSA was statistically not significant (P = 0.35). SI had significantly lower yield compared to CTR (P < 0.05) and TFA (P < 0.05). Twenty-three patients with PNES and 12 in control group did not consent for saline injection. All the induction tests had 100% specificity and 100% positive predictive value. The highest negative predictive value was 51% for CTR and the lowest negative predictive value for TLS (37.6%). The yield of different induction tests is presented in Table 2. The mean (SD) number of attempts to induce seizures was 1.08 ± 0.32 for VS, 1.14 ± 0.46 for CTR, 1.19 ± 0.47 for TFA, 1.06 ± 0.28 for MSA and 1.04 ± 0.19 for TLS. The number of repetitions needed to induce seizure was not significantly different in various induction tests (P = 0.48). Torch light stimulation needed maximum time (23.9 ± 9.7 s) and TFA test needed the least time (8.2 ± 1.9 s, P = 0.001). The number of attempts and time needed to induce PNES are presented in Table 3.
Tolerability of the induction tests The physical and psychological discomfort produced by different tests was assessed in 50 patients. Only 9 patients were anxious or apprehensive before the induction. Eleven patients did not report anxiety or psychological discomfort after the induction. Saline injection and TLS produced maximum anxiety or psychological discomfort in 13 and 10 patients respectively. The mean discomfort level during induction was 2.6 ± 2 and none of the patients had discomfort level above 5. The maximum discomfort was produced by SI (3.2 ± 3.0) and minimum by MSA (0.89 ± 1.46). This
Please cite this article in press as: Goyal, G., et al., Utility of different seizure induction protocols in psychogenic nonepileptic seizures. Epilepsy Res. (2014), http://dx.doi.org/10.1016/j.eplepsyres.2014.02.015
+Model
ARTICLE IN PRESS
EPIRES-5110; No. of Pages 8
5 difference is significant (P < 0.0001). The majority of subjects felt that the SI was most painful (P < 0.001) while 6 felt no pain by any of the induction tests. Most of the patients were satisfied with these induction tests, willing to undergo again if needed in future and none considered it psychologically traumatic. The results of the physical and psychological discomfort of the induction tests are summarized in Tables 4A and 4B.
51.02 43.86 48.07 47.62 49.02 37.59
NPV (%)
Discussion
100 100 100 100 100 100
PPV (%)
100 100 100 100 100 100
Specificity (%)
(100) (100) (100) (100) (100) (100) 50 50 50 50 38 50 PPV, positive predictive value; NPV, negative predictive value.
0 0 0 0 0 0 48 (34.3) 64 (45.7) 54 (38.6) 55 (39.3) 52 (44.4) 83 (59.3) 92 (65.7) 76 (54.3) 86 (61.4) 85 (60.7) 65 (55.6) 57 (40.7) Compression of temple region Verbal suggestions Tuning fork application Application of moist swab Saline injection Torch light stimulation
Induced N (%)
Not induced N (%)
Induced N (%)
Not induced N (%)
65.71 54.29 61.43 60.71 55.56 40.71
Sensitivity (%) Epileptic seizures PNES Induction methods
Table 2 Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) of different induction test in the patients with psychogenic nonepileptic seizure (PNES).
Psychogenic nonepileptic seizures
In the present study, CTR was the most effective in inducing PNES with a sensitivity of 65.7%, specificity of 100%, positive predictive value of 100% and negative predictive value of 51%. Tuning fork application test needed the least time for induction though number of attempts of different induction methods was not significantly different. The least discomfort was produced by MSA and maximum by SI. The present study for the first time evaluated the sensitivity, specificity and discomfort level of a number of induction tests in the patients with PNES. The higher specificity of induction tests in our study is due to lack of seizure induction in the control. The patients with epileptic seizure were on remission for at least 6 months. In an earlier study, the usefulness of patch and massage on the patch in precipitating PNES were evaluated in 91 PNES patients and 20 patients with epileptic seizure. The induction test was positive in 77.4% of PNES and none with epileptic seizure resulting in 100% specificity and 48.7% negative predictive value. The patients’ discomfort however was not evaluated in this study (Lancman et al., 1994). In another study, hypnosis was used to induce PNES in 69 patients of 3 groups (PNES, epilepsy, epilepsy + PNES). Hypnotic procedure had 77% sensitivity and 95% specificity for PNES (Barry et al., 2000). Placebo induced episodes were typical of prior PNES episodes in 90% whereas 10% had atypical episodes. Fifteen percent in epilepsy group had atypical and confusing events in the same study (Walczak et al., 1994). In another study, higher frequency (30%) of atypical attacks was reported in the patients with PNES (Drake, 1985). In the present study, 88.4% PNES patients had typical event and 11.6% atypical. Lesser proportion of atypical event in PNES group in our study may be due to uniform induction protocol to all the patients and verification of the induced event. The induced attack should be verified by a witness, family member or compared with the prior video records which are often available through mobile phone. Induction tests have been reported as a stressful experience and may lead to increased pulse rate, blood pressure and physiological tremor (Walczak et al., 1994). Emotional stress may trigger seizures in epilepsy patients as reported in various retrospective and prospective studies (Friis and Lund, 1974; Currie et al., 1971; Temkin and Davis, 1984). This may explain seizure induction even in some epilepsy patients. None of our epilepsy patient had induction of seizure though 38.8% had nonspecific symptoms such as light headedness, uneasiness and palpitation. These symptoms could be the manifestation of emotional stress of induction procedure and might be confused with aura of seizure attack.
Please cite this article in press as: Goyal, G., et al., Utility of different seizure induction protocols in psychogenic nonepileptic seizures. Epilepsy Res. (2014), http://dx.doi.org/10.1016/j.eplepsyres.2014.02.015
+Model
ARTICLE IN PRESS
EPIRES-5110; No. of Pages 8
6
G. Goyal et al. Table 3 Number of attempts and induction time of different induction tests in the patients with psychogenic nonepileptic seizure (PNES). Induction method
Total number of patients included (n)
Successful attempts N (%)
Consent was not given N (%)
Mean attempts
Verbal suggestion
140
76 (54.3)
—
1.08 ± 0.32
Compression of temple region
140
92 (65.7)
—
1.14 ± 0.46
Tuning fork application
140
86 (61.4)
—
1.19 ± 0.47
8.15 ± 1.93
Application of moist swab
140
85 (60.7)
—
1.06 ± 0.28
17.86 ± 9.32
Saline injection
117
65 (55.6)
23 (19.7)
1.00
17.46 ± 7.04
Torch light stimulation
140
57 (40.7)
—
1.04 ± 0.19
23.91 ± 9.74
Table 4A
P value
Mean time required to induce
P value
18.33 ± 9.63 0.48
17.64 ± 10.05
0.001
Assessment of discomfort of induction tests in the patients with psychogenic nonepileptic seizure.
Questions
CTR
VS
AMS
TFA
Which protocol produced maximum anxiety or psychological discomfort (no. of patients) Discomfort level (0—10 scale) Which protocol was most painful: (no. of patients; consent for SI was not given by 13)
8
0
3
5
1.33 ± 1.73
1.17 ± 1.99
0.89 ± 1.46
1.65 ± 2.23
1.76 ± 2.36
2
0
4
3
3
TLS 10
SI 13
3.24 ± 3.02 33
P value 0.04
0.01 <0.001
CTR, compression of temple region; VS, verbal suggestion; TFA, tuning fork application; AMS, application of moist swab; SI, saline injection; TLS, torch light stimulation.
Table 4B
Overall acceptability of the induction tests which was assessed in patients with psychogenic nonepileptic seizure.
Questions
Strongly agreed
Agreed
Acceptable
Disagree
Overall seizure induction protocol is patient friendly and satisfactory If needed in future, will you like to undergo such test? Are these procedures left you with mental trauma? Were the procedures fully explained to you?
15
27
5
3
0.001
21 0 8
20 4 39
3 2 3
6 31 0
0.001 <0.001 <0.001
P value
Please cite this article in press as: Goyal, G., et al., Utility of different seizure induction protocols in psychogenic nonepileptic seizures. Epilepsy Res. (2014), http://dx.doi.org/10.1016/j.eplepsyres.2014.02.015
+Model EPIRES-5110; No. of Pages 8
ARTICLE IN PRESS
Psychogenic nonepileptic seizures The majority of the studies have evaluated one or two tests for seizure induction in PNES. The sensitivity of SI ranges between 33% and 91% (Slater et al., 1995; Bhatia et al., 1997). In our study, the sensitivity of saline injection was 55.6% with 100% specificity. Thirty-five of our patients however refused to undergo this test. Invasiveness of the SI and associated discomfort may be the reason of refusal. This emphasizes need for other induction tests to substitute the widely used placebo SI for the diagnosis of PNES. Torch light stimulation although is a good alternative as it is painless and acceptable to most of the patients but is limited by longer procedural time and lower sensitivity of 40.7%. Moist swab application (60.7%) and TFA (61.4%) had somewhat similar sensitivity and specificity; however, TFA resulted in more anxiety and discomfort compared to MSA. In the reported literature, none of the studies have compared relative discomfort of various induction tests. The sensitivity of induction tests was higher in village dwellers, housewives and less educated (<12 standard) patients. These backgrounds may make them more suggestible to the provocative tests. We have not included the patients who had both epileptic seizures and PNES to avoid conflicting results. Little work has been done on underlying pathophysiology of inducibility of PNES. In our study, 67.9% patients with PNES had inducibility. It raises some questions like why some patients with PNES had inducibility and not the others, what are the predictors of inducibility and what is the impact of inducibility on long term outcome? In a recent study, hypermotor ictal semiology, self-reporting of uncommon cognitive and affective symptoms, higher tendency to rely on coping strategies of ‘‘instrumental support’’ and ‘‘active coping’’ have been reported as the predictors of successful induction (Chen et al., 2011). Use of induction test has raised ethical concerns as it may be associated with deception, violates patients’ autonomy, undermines the doctor-patient relationship and psychological harm to the patient (Leeman, 2009). The ethical practice is to do good to the patient (beneficence) and do no harm to the patient (malfeasance). Despite criticism and ethical concerns of nondisclosure that is inherent in provocative tests, violation of doctor patient trust and feeling of deception by some, it may be more harmful to miss a diagnosis when it can be made (Devinsky and Fisher, 1996). Missing the diagnosis of PNES may cause serious socioeconomic, psychological and medical consequences. In a study on 313 patients with PNES, there was an average delay of 7.2 years in the diagnosis (Reuber et al., 2002). Delay in the diagnosis of PNES may lead to inappropriate treatment, expose the patients to drug toxicity and rarely these patients wrongly intubated for psychogenic status epilepticus. Moreover there may be demoralization, low self-esteem and economic loss. These possibilities outweigh the ethical concern of induction of PNES. Ethical concern regarding psychological harm and damage to doctor patient relationship is rather theoretical. If the correct diagnosis is achieved by noninvasive or minimally invasive induction test without producing harm or much discomfort, it may be advantageous to the patients. The principle of autonomy is that if anyone who is influenced by the consequence of the decision should be involved in the decision making process. We have informed the patients about the induction test and they gave their consent.
7 Moreover the utility of these tests has been already reported. In the present study, we have not used any new induction test but compared the relative usefulness and discomfort of the various induction tests in the diagnosis of PNES. In a study, patients found these techniques useful and not harmful (Cohen et al., 1992). In the present study, most of the patients were comfortable during induction procedure and denied significant psychological trauma. Recently, provocative tests were advocated to use in the diagnosis of PNES (Benbadis, 2009b). Provocative tests during video EEG recording may shorten the evaluation time. In clinical settings, usefulness of these tests outweighs the ethical concerns by avoiding misdiagnosis of epilepsy in PNES patients, reducing uncertainty about the diagnosis and preventing long term cognitive and physical health hazards of unnecessary antiepileptic drug therapies. Devinsky and Fisher reviewed the ethical use of placebos and provocative tests in the diagnosis of PNES (Devinsky and Fisher, 1996). The present study has some limitations. All the induction tests were performed in prefixed order by a single investigator without blinding which may lead to a bias. Performing all the induction tests in a prefixed manner may lea d to higher sensitivity of the initial induction tests compared to the later. Similarly, performing a number of induction tests in a single setting may result in fatigue at the end and may influence the sensitivity of the tests by attenuating the response for later stimuli. There is also possibility of suggestibility through nonverbal cues in some of the induction tests such as CTR and application of TFA. Further studies are required to overcome these limitations and to confirm these findings.
Conclusion In a busy clinical practice, CTR and TFA are comparatively better induction tests in the diagnosis of PNES because these are sensitive, easy to perform and have minimal discomfort. Surprisingly widely used provocation test ‘‘SI’’ is significantly less efficacious, comparatively more cumbersome and least tolerable.
Ethics approval The research has been approved by the Institutional Ethics Committee, SGPGIMS, Lucknow.
Conflict of interest There is no conflict of interest to declare.
Funding support None.
Acknowledgement We acknowledge the help of Mr. Rakesh Nigam for secretarial help.
Please cite this article in press as: Goyal, G., et al., Utility of different seizure induction protocols in psychogenic nonepileptic seizures. Epilepsy Res. (2014), http://dx.doi.org/10.1016/j.eplepsyres.2014.02.015
+Model EPIRES-5110; No. of Pages 8
ARTICLE IN PRESS
8
G. Goyal et al.
References Abubakr, A., Kablinger, A., Caldito, G., 2003. Psychogenic seizures: clinical features and psychological analysis. Epilepsy Behav. 4, 241—245. Al, R.A., Yalvac, S., Altar, O.Y., Dolen, I., 2009. Perceived pain and anxiety before and after amniocentesis among pregnant Turkish women. Clin. Exp. Obstet. Gynecol. 36 (3), 184—186. Ali, S., Jabeen, S., Arian, A., Wassef, T., Ibrahim, A., 2011. How to use your clinical judgment to screen for and diagnose psychogenic nonepileptic seizures without video electroencephalogram. Innov. Clin. Neurosci. 8, 36—42. Barry, J.J., Atzman, O., Morrell, M.J., 2000. Discriminating between epileptic and non-epileptic events: the utility of hypnotic seizure induction. Epilepsia 41, 81—84. Bazil, C.W., Kothari, M., Luciano, D., Moroney, J., Song, S., Vasquez, B., Weinreb, H.J., Devinsky, O., 1994. Provocation of nonepileptic seizures by suggestion in a general seizure population. Epilepsia 35, 768—770. Benbadis, S.R., LaFrance Jr., W.C., Papandonatos, G.D., Korabathina, K., Lin, K., Kraemer, H.C., 2009a. NES Treatment Workshop. Interrater reliability of EEG-video monitoring. Neurology 73, 843—846. Benbadis, S.R., 2009b. Provocative techniques should be used for the diagnosis of psychogenic nonepileptic seizures. Epilepsy Behav. 15, 106—109. Benbadis, S.R., Hauser, W.A., 2000. An estimate of the prevalence of psychogenic non-epileptic seizures. Seizure 9, 280—281. Benbadis, S.R., Johnson, K., Anthony, K., Caines, G., Hess, G., Jackson, C., Vale, F.L., Tatum 4th, W.O., 2000. Induction of psychogenic nonepileptic seizures without placebo. Neurology 55, 1904—1905. Bhatia, M., Sinha, P.K., Jain, S., Padma, M.V., Maheshwari, M.C., 1997. Usefulness of short-term video EEG recording with saline induction in pseudoseizures. Acta Neurol. Scand. 95, 363—366. Binnie, C.D., Prior, P.F., 1994. Electroencephalography. J. Neurol. Neurosurg. Psychiatry 57, 1308—1319. Bodde, N.M.G., Brooks, J.L., Baker, G.A., Boon, P.A.J.M., Hendriksen, J.G.M., Mulder, O.G., Aldenkamp, A.P., 2009. Psychogenic non-epileptic seizures—–Definition, etiology, treatment and prognostic issues: a critical review. Seizure 18, 543—553. Chen, D.K., Izadyar, S., Collins, R.L., Benge, J.F., LeMaire, A.W., Hrachovy, R.A., 2011. Induction of psychogenic nonepileptic events: success rate influenced by prior induction exposure, ictal semiology, and psychological profiles. Epilepsia 52, 1063—1070. Cohen, L.M., Howard 3rd, G.F., Bongar, B., 1992. Provocation of pseudoseizures by psychiatric interview during EEG and video monitoring. Int. J. Psychiatry Med. 22 (2), 131—140. Currie, S., Heathfield, K.W.G., Hanson, R.A., Scott, D.F., 1971. Clinical course and prognosis of temporal lobe epilepsy. Brain 94, 173—190.
Deacon, C., Wiebe, S., Blume, W.T., McLachlan, R.S., Young, G.B., Matijevic, S., 2003. Seizure identification by clinical description in temporal lobe epilepsy: how accurate are we? Neurology 61, 1686—1689. Devinsky, O., Fisher, R., 1996. Ethical use of placebos and provocative testing in diagnosing nonepileptic seizures. Neurology 47, 866—870. Drake, M.E., 1985. Saline activation of pseudoepileptic seizures: clinical and neuropsychiatric observations. Clin. Electroencephalogr. 16, 171—176. Friis, M.L., Lund, M., 1974. Stress convulsions. Arch. Neurol. 31, 155—159. Hubsch, C., Baumann, C., Hingray, C., Gospodaru, N., Vignal, J.P., Vespignani, H., Maillard, L., 2011. Clinical classification of psychogenic non-epileptic seizures based on video-EEG analysis and automatic clustering. J. Neurol. Neurosurg. Psychiatry 82, 955—960. Krumholz, A., 1999. Nonepileptic seizures: diagnosis and management. Neurology 53, S76—S83. Lancman, M.E., Asconape, J.J., Craven, W.J., Howard, G., Penry, J.K., 1994. Predictive value of psychogenic seizures by suggestion. Ann. Neurol. 35, 359—361. Leeman, B.A., 2009. Provocative techniques should not be used for the diagnosis of psychogenic nonepileptic seizures. Epilepsy Behav. 15, 110—114. Leis, A.A., Ross, M.A., Summers, A.K., 1992. Psychogenic seizures: ictal characteristics and diagnostic pitfalls. Neurology 42, 95—99. Reuber, M., Fernández, G., Bauer, J., Helmstaedter, C., Elger, C.E., 2002. Diagnostic delay in psychogenic nonepileptic seizures. Neurology 58, 493—495. Scarpero, H.M., Padmanabhan, P., Xue, X., Nitti, V.W., 2005. Patient perception of videourodynamic testing: a questionnaire based study. J. Urol. 173 (2), 555—559. Schonenberger, E., Schnapauff, D., Teige, F., Laule, M., Hamm, B., Dewey, M., 2007. Patient acceptance of noninvasive and invasive coronary angiography. PLoS ONE 2 (2), e246. Slater, J.D., Brown, M.C., Jacobs, W., Ramsay, R.E., 1995. Induction of pseudoseizures with intravenous saline placebo. Epilepsia 36, 580—585. Smith, D., Defalla, B.A., Chadwick, D.W., 1999. The misdiagnosis of epilepsy and the management of refractory epilepsy in a specialist clinic. QJM 92, 15—23. Temkin, N.R., Davis, G.R., 1984. Stress as a risk factor for seizures among adults with epilepsy. Epilepsia 25, 450—456. Vella, P.V., 1991. A survey of women undergoing a pelvic examination. Aust. N. Z. J. Obstet. Gynaecol. 31, 355—357. Walczak, T.S., Williams, D.T., Berten, W., 1994. Utility and reliability of placebo infusion in the evaluation of patients with seizures. Neurology 44, 394—399. Zaidi, A., Crampton, S., Clough, P., Fitzpatrick, A., Scheepers, B., 1999. Head-up tilting is a useful provocative test for psychogenic non-epileptic seizures. Seizure 8, 353—355.
Please cite this article in press as: Goyal, G., et al., Utility of different seizure induction protocols in psychogenic nonepileptic seizures. Epilepsy Res. (2014), http://dx.doi.org/10.1016/j.eplepsyres.2014.02.015