- Top of page
Summary: Purpose: To describe male patients (pts) with psychogenic nonepileptic seizures (PNESs) followed up in a Veteran's Administration (VA) seizure clinic and to compare them with those with epileptic seizures (ESs) by using clinical, and psychosocial variables.
Methods: Adult male veterans seen between 1997 and 2000 with ESs were compared with those with PNESs with respect to clinical history (head trauma, antiepileptic drug exposure, depression, anxiety, substance abuse, seizure description), documented chronic pain, posttraumatic stress disorder (PTSD), compensation for diagnosis, neurologic examination, and test results including imaging and EEG data.
Results: Men with PNESs were younger and reported more frequent events, and diagnoses of chronic pain, anxiety, and PTSD were significantly greater. Neuroimaging [computed tomography (CT) or magnetic resonance imaging (MRI) of the brain] and neurologic examination were significantly more likely to be normal or nonspecific in pts with PNESs, although history of ictal urinary incontinence or service-connected compensation for diagnosis did not distinguish the groups.
Conclusions: Male veterans with PNESs have characteristics similar to those reported in the literature, even though younger women have dominated previously studied populations. Compared with men with ESs, those with PNESs are more likely to have chronic pain, anxiety, and PTSD, as well as normal examinations and brain imaging.
Psychogenic nonepileptic seizures (PNESs) are common in patients with seizure-like episodes that do not respond to antiepileptic drugs (AEDs) (1,2). Consequently, patients with PNESs are at risk of receiving inappropriate medications for prolonged treatment periods or invasive intervention such as vagal nerve stimulation (3). These treatments are costly and potentially dangerous. Diagnosis is typically delayed until referral for video-EEG monitoring (4,5). Presence of PNESs has been shown to be associated with a lower quality of life and higher level of stress compared with the presence of epileptic seizures (ESs) (6–8). This further highlights the importance of accurate diagnosis.
Most of what is understood about PNESs is derived from studies conducted at tertiary epilepsy centers where 10 to 58% of adult patients with intractable spells are ultimately diagnosed with PNESs (9,10). In these studies, 65–80% of patients are young and female, which has led to the commonly held opinion that PNESs are less important in men. Furthermore, a small number of studies have demonstrated that significant gender differences may exist in background and clinical manifestations (11,12). For example, compared with women, men with PNESs are more likely to be substance abusers, report minor head injury, have a documented financial gain from having seizures, and worse emotional adjustment by MMPI-II testing (12–14). Having a history of physical or sexual abuse appears to be more common in women with PNES, but is still more likely in men with PNESs than in the general population of men. Abuse is correlated with conversion disorder, the primary psychopathologic correlate of PNESs; nonconversion etiologies such as anxiety, psychosis, and impulse-control disorders, interestingly, have shown similar frequencies in men and women (15,16) Other studies suggest that PNESs may appear clinically different in men than in women (12,16), with more dramatic-appearing motor movements and less affective change in men (17). Gender also may be an important prognostic factor, because the probability of a favorable outcome of PNESs is higher in women than in men (18). Gender differences in the use of health care services and the widespread belief that PNESs is primarily a woman's diagnosis may impede referral to epilepsy centers and lead to delayed diagnosis of PNESs in men (19,20).
The Veteran's Administration (VA) population consists largely of men older than 50 years. Three years of video-EEG monitoring experience has convinced us that PNESs are not rare in the VA population. We describe the characteristics of male veterans with probable or definite PNESs evaluated in a VA Hospital Seizure Clinic, and compare them with male veterans with ESs with respect to clinical, psychiatric, and psychosocial variables.
- Top of page
Subjects were consecutive adult male veterans evaluated for spells in a VA seizure clinic or admitted to the VA neurology ward for long-term video-EEG monitoring (LTM) between 1997 and 2000. This study was approved by the Institutional Review Board of the Boston VA Healthcare System. Of 2,443 total patient encounters, 288 unique patients were seen. Female patients were excluded (n = 37); 140 male patients were excluded on the basis of incomplete data, limited mental status examination, or lack of evidence for an episodic disorder. Of 111 eligible patients, we excluded those with probable PNESs but definite interictal epileptiform discharges (n = 3), no clinical events captured during the monitored period (n = 45), only subjective events during monitoring (n = 4), or purely physiological spells [vasovagal syncope, paroxysmal coughing, hypnic (sleep onset) jerks; n = 3]. Of the remaining patients, 22 had PNESs based on a history of paroxysmal behavioral events to suggest epileptic seizures, and at least one typical event captured on monitoring either spontaneously or with suggestion strongly suggesting PNESs without ictal, interictal, or postictal EEG change, no response to AEDs, and no events suggestive of ESs, as defined by Vanderzant et al. (21). The ESs group totaled 34 and was defined as patients with an unequivocal history of epilepsy, no events to suggest PNESs, and definite epileptiform discharges on EEG, ictally (n = 12) or interictally (n = 22) or both, interpreted by a certified electroencephalographer (B.D.). Duration of disease for this group on average was 18.3 years, with a range from 1 to 56 years. Classification of seizure types and epilepsy syndromes are shown in Table 1. Clinical history and neuroimaging data extracted from the electronic medical record were used to compare PNES and ES patients. A registered nurse (J.P.) interviewed all patients and recorded on a standardized intake form information regarding history of chronic pain, sleep problems, substance abuse, prior head trauma, family history of seizures, history of childhood epilepsy, febrile convulsions, and physical or sexual abuse. Neuropsychological assessment was reviewed where available but was too incomplete for formal analysis. AED therapy and service-connected compensation was ascertained from the medical record. Psychiatric diagnoses, including posttraumatic stress disorder (PTSD), anxiety, and depression, were included if they were documented in the medical record by a psychiatrist, psychologist, or social worker. All data were analyzed by using two-tailed t tests for continuous variables and Fisher's exact tests for categorical variables.
Table 1. Classification and etiologies of seizures and epilepsy in male veterans
|Simple partial seizures|| 3|
|Complex partial seizures||18|
|Secondarily generalized seizures|| 8|
|Idiopathic generalized epilepsy|| 4|
|Idiopathic partial epilepsy|| 3|
|Symptomatic partial epilepsy||27|
| Traumatic brain injury||14|
| Neoplasm|| 4|
| Vascular|| 7|
- Top of page
Compared to male veterans with ESs, those with PNESs were ∼10 years younger (Table 2). Handedness, education level, employment status, and race did not significantly differ between groups (see Table 2). PNES patients were more likely to have a normal neurologic examination and a normal magnetic resonance imaging (MRI) scan of the brain (see Table 2). PNES patients had more-frequent events than did ES patients, with the majority complaining of daily episodes (Table 3). The groups reported similar clinical symptoms, with the exception of more staring spells for the ES group (see Table 3). Furthermore, report of incontinence by history was similar in both groups. Both groups had similar reports of substance abuse, sleep-related problems, and depressive symptoms, whereas PNES patients were significantly more likely to report chronic pain, anxiety, and PTSD (Table 4). More than 95% of PNES patients were treated with AEDs before receiving their diagnosis. Of these, 77% received monotherapy. ES patients were more likely to be receiving polytherapy (see Table 2). PNES and ES groups did not significantly differ with respect to the proportion of patients who received compensation for their diagnosis or in the number of years that a patient experienced the seizures before entry into this study (see Table 2).
Table 2. Patient characteristics of male veterans, obtained through medical record and self-report
| ||ESs (n = 34)||PNESs (n = 22)||Fisher's Exact|
| Right||31||20||p = NS|
| yr (SD)||61.2 (14)||50.1 (13)||p < 0.01a|
| White||26||21||p = NS|
| Nonwhite|| 8|| 1|| |
| yr (SD)||11.4 (2.5)||12.4 (3)||p = NSa|
|Time with symptoms before diagnosis in study (mean no. yr)|
| ≤1 yr||10|| 6||p = NS|
| >1 yr||24||16|| |
| No therapy|| 1|| 1||p < 0.002|
| Monotherapy||13||17|| |
| >1 drug||20|| 4|| |
|History of febrile seizure|
| Yes|| 0|| 1||p = NS|
| Yes||13||11||p = NS|
|Employment history||n = 32||n = 21|| |
| Full time/Part time|| 5|| 4|| |
| Retired|| 5|| 2||p = NS|
| Disabled||22||15|| |
|Neurologic examination||n = 32|| |
| Abnormal||19|| 1||p < 0.001|
| Normal||13||15|| |
|Brain imaging (MRI/CT)||n = 31|| |
| Abnormal||26||11||p < 0.01|
| Normal|| 5||11|| |
Table 3. Seizure characteristics, from clinical history of male veterans
| ||ESs (n = 34)||PNESs (n = 22)||Fisher's Exact|
|Seizure frequency|| n = 29|| n = 19|| |
| None/yr|| 8|| 0|| |
| ≤2 /mo|| 12|| 5||p < 0.0001|
| Weekly|| 7|| 3|| |
| Daily|| 2||11|| |
|Limb movements||n = 30|| |
| No|| 8||11||p = NS|
| Yes||22||11|| |
|Staring||n = 29|| |
| Yes||18|| 8||p < 0.05|
|Lip smacking||n = 30|| |
| Yes|| 5|| 1||p = NS|
|Sensory symptoms||n = 27|| |
| Yes|| 4|| 3||p = NS|
|Aura||n = 33|| |
| Yes||15||13||p = NS|
|Incontinence||n = 31|| |
| Yes|| 8|| 5||p = NS|
Table 4. Behavioral and neuropsychiatric profile of male veterans, obtained from medical records and self-report
| ||ESs (n = 34)||PNESs (n = 22)||Fisher's Exact|
|Substance abuse||n = 33|| |
| Yes (any, incl. EtOH)||18||14||p = NS|
|Sleep problems||n = 27|| |
| Yes||14||10||p = NS|
|Chronic pain||n = 26|| |
| Yes||13||19||p < 0.008|
|Anxiety||n = 32|| |
| Yes||12||16||p < 0.009|
|PTSD||n = 28|| |
| Yes|| 3||14||p < 0.0001|
|Depression||n = 33|| |
| Yes||18||16||p = NS|
- Top of page
This study offers a unique opportunity to focus on differences between individuals with ESs and PNESs in a relatively older male population, a group that has been underrepresented in the epilepsy literature as well as in practice in epilepsy monitoring units. McBride and colleagues (22), publishing retrospective data on PNES in older populations, reported that 27% of patients older than 60 years who undergo diagnostic video-EEG monitoring are found to have NES; 13%, PNES; and 14%, of physiologic cause (i.e., cataplexy, hypotension). This and other recent studies suggest that PNESs, either psychogenic or physiologic, should be strongly considered in older patients with episodic seizure-like events (23,24).
Despite reported gender differences, the clinical characteristics of male veterans with NESs are similar to what has been shown in the literature for women. When compared with patients with ESs, male veterans with PNESs in this investigation were, on average, 10 years younger and reported a higher frequency of episodes. However, the groups did not differ with respect to education or employment history. Incidence of PTSD, a clear risk factor for PNESs in women, has been found in a significantly higher proportion of the male PNES than ES patients as well (25–27). The association of chronic pain, painful “auras,” especially headaches, and posttraumatic stress with PNESs has been demonstrated for younger female populations; this same association was found with our male veterans. This suggests that chronic pain and PTSD may be general predictors of PNESs, regardless of patients' gender or age (28). Urinary incontinence, traditionally used as a predictor to support a diagnosis of ESs, was found not to be associated with ESs in our study, because no difference was found between groups. This is similar to what Peguero et al. (29) found by telephone survey. As in previous studies (16–18,30), we found a similarly elevated rate of occurrence of sleep problems and depression in both groups.
We believe that the majority of PNES patients in our study were overtreated, with 77% taking one AED and 18% taking two or more. We also found 96% were receiving compensation, with nearly 42% of the total receiving it specifically for their diagnosis. This compensation rate was similar for both groups, which on first glance may seem surprising, but might imply a similar disability. Unnecessary treatments are concerning because of potential unwanted side effects from the drugs themselves or from drug–drug interactions, which can cause serious consequences, especially in the elderly (31). These can lead to unnecessary medical costs for visits, blood draws, and treatment of side effects. In a small study of involving 20 patients with diagnostic LTM, an 84% average reduction was found in total seizure-related medical charges within 6 months of making the diagnosis (5). Significantly fewer young female patients with PNESs are treated with AEDs (32,33), perhaps because young women's seizures are perceived as more likely to have a psychological etiology or because of possible risk of the drug should they become pregnant. Additionally, men with PNESs have been shown in the literature to exhibit more tonic–clonic-like events than do women (12), and perhaps this display leads to a higher likelihood of therapy, in that events appear more severe.
Our study supports the observations of others that PNES patients are less likely than ES patients to have abnormal neuroimaging or neurologic examinationss, despite the confirmation of a previously reported high incidence of head trauma in PNES (range, 32–70%) (14,34). Additionally, in individuals with PNESs, head injury has been found to be associated with poor long-term outcome, including long-lasting disability (14), despite being relatively mild.
Substance abuse was common in both groups. Previous studies have shown this to be more common in men with PNESs (18); however, substance abuse, in particular with alcohol, is elevated in the population of male veterans in general (35). Studies also have suggested that men may use PNESs to avoid responsibilities (16). It is possible that service-connected compensation for illness may increase the likelihood for treatment in patients attending VA clinics. However, in our study, the proportion of patients with possible secondary gain is not different between groups. Compensation may have other effects, such as perpetuating the misdiagnosis of ongoing seizures. However, it is important to note that reimbursement does not imply conscious malingering, and it is likely that PNESs represent a form of conversion disorder in our population, as in most women (36). It is notable that a relatively high percentage of VA PNES patients are categorized as disabled, based on the diagnosis of seizure disorder; as has been reported, an individual can be similarly or even more disabled by PNESs than by ESs (6–8,37). Higher ratings of disability, lower self-reported quality-of-life scores, greater impact on families, and overall misdiagnosis and mistreatment have been reported in patients with PNESs (6,8,38,39).
Our study demonstrates that male veterans with PNESs have risk factors similar to those reported in the literature, even though younger females have been the predominant group studied to date. Although this study uses limited neuropsychiatric assessments, it nevertheless introduces data that suggest important distinguishing characteristics for men with PNESs. Thus we believe that screening male patients with frequent uncontrolled seizures should include questions designed to detect various stressors, including a history of PTSD and chronic pain. Further studies are needed to confirm that our results are generalizable to the population of civilian men and to investigate the influence of gender and age on the expression, treatment, and outcome of PNESs.