The answer to this question is not known. If the cause of a particular patient’s seizures is unclear, is it possible that two different types of seizure are present, one epileptic, the other not? The literature suggests that, in the assessment of a patient with an unknown attack disorder that has two different presentations, 5% to 30% of the time both with epilepsy and psychogenic PNES may be present.⁶⁹ The prevalence rates of concurrent ES and PNES may vary from one study to another according to whether a diagnosis of concurrent epilepsy was based on the capturing of actual seizures or on the recording of interictal epileptiform discharges only. For example, Devinsky et al. identified 20 patients with ES among a group of 99 PNES patients.²¹ In contrast, in a study of 1,590 patients who underwent a video-electroencephalographic (V-EEG) study, Martin et al. found 514 (32.3%) diagnosed with PNES, of whom 29 (5.3%) were found to have both PNES and epilepsy.⁵⁵ Lesser et al. concluded that PNES and concurrent epilepsy occurred in about 10% of patients with PNES.⁵⁴ Using stringent criteria (as opposed to higher estimates in the past that based abnormalities—including slowing on EEG—as ictal evidence), Benbadis et al. identified three patients with interictal epileptiform activity among 32 patients (9.4%) with PNES.⁴ These investigators did not record ES so it is not clear whether the seizure disorder was concurrent or the seizures had remitted with pharmacotherapy.

Studying the surgical population, Henry and Drury conducted a V-EEG in 145 patients who had temporal interictal EEG spikes, and they reported ictal semiology characteristic of temporal lobe seizures for presurgical evaluation of medically refractory seizures. PNES were unexpectedly identified in 12 (8%) of these patients.³⁷ A low prevalence was reported even in patient populations at greater risk. For example, Lelliott et al. found that, over a 5-year period, 7% of admissions to an epilepsy unit in a psychiatric hospital had PNES and concurrent epilepsy.⁵³

An exception must be made among patients with cognitive developmental delay. Indeed, these are patients with epilepsy who may have found that their seizure activity can be reinforced or rewarded (i.e., having a seizure allows them to get out of unpleasant situations at school or workshop). As a result, they “learn to precipitate their epileptic seizures” at times when they try to avoid uncomfortable situations. These patients, therefore, are found to have higher prevalence rates of PNES and concurrent ES. For example, Neil and Alvarez
evaluated 124 mentally retarded persons with behaviors suggestive of epilepsy.⁵⁷ Patients were monitored using an eight-channel radio-telemetered V-EEG recording system. Twenty patients (16%) were found to have only ES, 50 (40.5%) to have PNES, 11 to have both epileptic and PNES (9%), and 43 were classified as inconclusive. Among the patients with only PNES 15 had abnormal EEGs and four (37%) had epileptiform EEGs. Thus, in this study, among the 61 patients with PNES, 24.5% had a concurrent or past seizure disorder.

As for the previous question, no clear answer is obtained from the available literature. For example, in some studies it is also sometimes unclear, when the prevalence of the two conditions is compared, whether the investigator is referring to patients with PNES and concurrent ES or merely a history of past epilepsy. It is likely, as Ramsey et al. have concluded,⁶⁹ that if the history is one of past as opposed to present epilepsy, the proportion of patients with both epilepsy and nonepilepsy is likely to be higher, particularly in patients with cognitive developmental delay.⁵⁷

A review of the literature^8,69 and the clinical studies^42,47,71 suggest that 10% to 40% of patients who appear to have established PNES also have a past history of ES. Sometimes the question can only be answered by withdrawing antiepileptic drugs (AEDs) in those patients whose interictal recordings reveal epileptiform activity or in whom a history of paroxysmal events with clinical manifestations is highly suggestive of ES (also see the next section). If epilepsy re-emerges upon discontinuation of AEDs, the possibility exists of a patient who has a seizure disorder that remitted with the use of AEDs, and who, at some point, went on to develop PNES. On the other hand, failure to record an ES upon discontinuation of AEDs does not rule out a seizure disorder that remitted with AEDs, above all in the presence of epileptiform activity in interictal recordings.

One example of patients with both PNES and a past history of epilepsy is found in those patients who have had successful temporal lobectomies and then went on to develop PNES, presumably because of an inability to adjust to a seizure-free life. Fortunately, this is not a common occurrence. In a study of 166 consecutive patients who underwent epilepsy surgery, Parra et al. found three patients (2%) who experienced postsurgical de novo PNES documented by V-EEG.⁶² The interval between the date of surgery and the development of the symptoms was variable (8–47 months range). Two of the three patients had become seizure free after surgery, and one had significant improvement of her seizures. The clinical phenomena of PNES differed from those seen for the ES preceding surgery. The diagnosis of PNES had not been suspected before the diagnostic V-EEG. Following the diagnosis of PNES, spells stopped in two patients and recurred rarely in one. Davies et al. found de novo postsurgical PNES in eight (3.5%) of 228 patients who underwent epilepsy surgery.¹⁴ PNES occurred between 6 weeks to 6 years (mean, 23 months) after surgery. Six had undergone a resection, and two had complete callosotomy. There was a significant excess of postoperative interictal dysphoric disorder (IDD) and operative complications (bone flap infections) in the PNES group. Ney et al. identified five patients (5%) with PNES among a group of 96 consecutive patients who underwent epilepsy surgery.⁵⁸ Two patients experienced operative complications. Compared with the surgical cohort, patients had a higher frequency of preoperative psychopathologic conditions, lower mean FSIQ, and a greater occurrence of operative complications. In a recent study completed at the Rush Epilepsy Center, Kanner et al. found that seven of 94 patients (7%) who underwent an anterotemporal lobectomy developed de novo PNES. As in other studies, most patients (n = 6) were women; six of the seven patients had a pre- and postsurgical psychiatric depression/anxiety disorder, whereas the seventh patient had a history of attention deficit disorder and personality disorder that was present before surgery and worsened after surgery. Three patients had gainful employment before and after surgery, whereas the remaining four did not seek or obtain work either before or after surgery. Given the small number of patients with de novo PNES, no statistical analyses were carried out (Kanner AM, unpublished data).

Glosser et al. tried to identify risk factors of postsurgical PNES by comparing the demographic, neurologic, and psychiatric variables of 22 medically refractory epilepsy patients in whom PNES was documented by electroencephalogram (EEG) after resective surgery to those of a larger series of epilepsy surgery patients.³³ Patients with PNES were significantly more likely to have a seizure onset later in life, to have undergone epilepsy surgery in the right hemisphere, and to be women, but these subjects failed to differ with respect to age, IQ, or preoperative psychiatric diagnoses. PNES tended to become apparent in the first few months after surgery.

This question is difficult to answer, because it depends on the accuracy of the diagnosis of PNES, which is often impossible to determine in published series. In the literature, completely different methods have been used for diagnosing PNES, based sometimes on EEG criteria (but not necessarily EEGs recorded during the ictus itself), clinical criteria, or a mixture of both. For example, people with epilepsy often show emotional reactions to their auras, so that they may become afraid of them and develop hyperventilation or panic attacks or have an emotional reaction during the ES itself. When auras fail to evolve to complex partial seizures, and their scalp recordings fail to reveal an electrographic ictal pattern,²² clinicians may be prone to misdiagnose these patients with simple partial seizures as having PNES. In fact, in 1885, Gowers suggested that minor seizures can elaborate into hysterical seizures.³⁴ A century later, Devinsky and Gordon reported the case of four patients in whom V-EEG–documented ES were temporally associated with PNES.²⁰ In one woman, the nonepileptic event followed an absence seizure, whereas in the other three patients, the seizures were partial and arose from right frontotemporal regions. Kapur et al. described three patients who convincingly elaborated simple partial seizures during EEG monitoring in order to ensure a result that would lead to surgery.⁴³ Clearly, one of the problems of assessing the cause of a seizure is that, during a simple partial seizure, some of what is observed may well be a behavioral manifestation of the patient’s emotional reactions to an unpleasant epileptic experience. This understandable emotional reaction may be mistaken for a nonepileptic event.

Paradoxically, as clinicians have become increasingly aware of the existence of PNES, patients with certain types of seizure disorders are increasingly being erroneously diagnosed as having PNES. For example, Parra et al. studied 100 consecutive patients who were undergoing a diagnostic V-EEG.⁶¹ Referring physicians correctly suspected a diagnosis of ES in only nine (43%) of 21 patients, whereas 12 (57%) patients were incorrectly thought to have PNES. This misdiagnosis was especially likely in patients with clinical seizures of mesial frontal or parietal lobe origin and stemmed from their “bizarre” or “atypical” manifestations or from the absence of any electrographic ictal activity concurrent with the event during the V-EEG.^41,56,84 Furthermore, it should be remembered that the electrographic
ictal pattern on scalp recordings of simple partial seizures is identified in only 25% of these seizures.²² The error rate in diagnosing PNES probably lies between 5% and 10% (i.e., in up to 10% of patients firmly diagnosed with PNES, the patient really has epilepsy).^69,85 This topic is reviewed in greater detail in Chapter 284.

Reaching a correct diagnosis of “only” PNES versus mixed PNES with concurrent or past ES is of the essence if a patient’s AEDs are to be discontinued without placing her at risk of (epileptic) seizure recurrence. Clearly, a carefully obtained history and detailed description of all events witnessed by family members and the patient’s own recollections are the first step in the diagnostic process and are of great importance when the existence of the two disorders in the same patient is suspected. The next step consists of a V-EEG to confirm the diagnosis. Unfortunately, few studies have investigated in a systematic manner the degree to which PNES and ES differ from one another in the same patient. The available data suggest that the two types of event differ clinically to a significant degree. In the Devinsky et al. study, analysis of PNES and ES on V-EEG revealed that the clinical features of PNES clearly differed from ES in 18 of 20 PNES cases.²¹

Although differentiating the two often is attempted by history, or even with observation of the ictal semiology, many epileptologists and neurologists know of cases in which a high suspicion for ES and V-EEG revealed PNES, or vice versa. Therefore, PNES is best diagnosed using the “gold standard” V-EEG, performed with the greatest humility because even the best have been fooled.

This topic is reviewed in great detail in Chapter 282. We will therefore only refer to some of the salient clinical differences here. In many clinicians’ minds, PNES presents as opisthotonic posturing with pelvic thrusting in a young female. Gates et al. showed that any presentation that epilepsy can have, PNES may have also,³⁰ but bedside examination reveals differences in the characteristics of the signs and movements.

During the ictus, some signs of PNES include geotropic eye movements, in which the eyes deviate downward to the side toward which the head is turned.³⁶ Eyelids are typically closed at the onset of PNES, and for a longer duration when as compared to temporal lobe epilepsy (TLE) or frontal lobe seizures (FLS) (20 seconds vs. ∼2 seconds, respectively).²⁴ Along with ictal eye closure, weeping is associated with PNES.^7,27 Postictal nose rubbing and postictal cough are found in TLE, but not in PNES.⁸² Pelvic thrusting has been reported to occur as commonly in FLS as it does in PNES. PNES and FLS differ, however, in that FLS are rarely associated with the overtly sexualized, often prolonged, display seen in some cases of PNES.³²

It was once thought that absence of physical injury sustained during a seizure was a diagnostic indicator differentiating PNES from ES; however, more than half of all patients with PNES actually have a physical injury associated with their PNES.⁴⁰ PNES patients may have urinary incontinence and may injure themselves during the ictus. Tongue biting, self-injury, and incontinence are commonly associated with generalized seizures; however, two thirds of patients with PNES report one of these three signs, typically associated with ES.¹⁵ Rug or floor burns on a patient’s cheeks or body is viewed as being pathognomonic for PNES.⁸⁰