The Challenge of Diagnosing and Managing Psychogenic Nonepileptic Seizures

Psychogenic nonepileptic seizures (PNES) are characterized by paroxysmal changes in behavior, motor activity, or sensation that may clinically resemble epileptic seizures but do not result from abnormal electrical activity in the brain. It is understood that PNES arises from an array of psychological factors and is often preceded by physical or sexual abuse (1, 2). The prevalence of PNES has been calculated to be between 2 and 33 persons per 100,000. However, there remains a scarcity of data about the epidemiology of PNES, and existing studies have significant limitations (3). The coexistence of PNES with epileptic seizures poses significant diagnostic and therapeutic challenges given the high prevalence rate of coexisting epilepsy and PNES (5.2% among epilepsy monitoring unit admissions in a review of 1,567 patient medical records [4]). Therefore, the diagnosis of PNES must be made first by excluding epilepsy. Beyond this, no standard or comprehensive theoretical framework exists to aid in the diagnosis of PNES (2). However, it is considered to be a conversion disorder that falls under the diagnostic category of somatic symptom disorders in DSM-5 (5). The present case report highlights the common features of PNES and outlines the typical processes involved in making a diagnosis, followed by a review of the literature on the most effective treatments.

Case

"Mark" is a 19-year-old Caucasian male with a past medical history of refractory Hodgkin's lymphoma and hypothyroidism and past psychiatric history significant for depression and anxiety. The patient was in his usual state of health when he was started on a new monoclonal antibody chemotherapy treatment called nivolumab. Immediately following infusion of this medication, the patient reportedly began experiencing episodes of full-body rigors with diaphoresis, tachycardia, chills, and general malaise. He presented to another hospital where he was found to have a temperature of 101°F. His thyroid-stimulating hormone level was 20.7 mcg/dL, and his ECG showed sinus tachycardia. He was diagnosed with an infusion reaction, which is not an uncommon complication associated with administration of monoclonal antibodies, and he was transferred to the emergency department at our hospital for further monitoring.

Upon arrival at our hospital, the patient was afebrile, and a repeat thyroid-stimulating hormone level test showed normal levels. However, the patient began developing worsening rigors, characterized by episodes of upper- and lower-extremity shaking lasting 2–3 minutes at a time. During these episodes, he remained alert and able to converse appropriately. Loss of bowel or bladder function, tongue biting, dysarthria, loss of consciousness, or postictal symptoms were not noted. The patient's physical examination was remarkable for slight left lower-extremity weakness and left-nasolabial fold flattening. However, a computerized tomography (CT) of the patient's head showed no evidence of stroke. Empiric antibiotics were started, an infectious workup was initiated, and the patient was admitted to the pediatric intensive care unit for further monitoring.

While in the pediatric intensive care unit, a standard EEG with video revealed no EEG changes before, during, or after the shaking episodes. Given the generalization of the episodes and normal mental status examination, seizure activity was thought to be unlikely. MRI of the brain and ECG with bubble study were unremarkable. CT of the chest, abdomen, and pelvis indicated good response to chemotherapy, with no evidence of progression of the Hodgkin's lymphoma. An exhaustive infectious workup, including lumbar puncture and CSF paraneoplastic studies, were all negative.

The patient's abnormal jerking movements progressed to severe whole-body jarring. He developed worsening musculoskeletal pain, along with increasing creatine kinase levels, and he eventually lost all desire to get out of bed. He began urinating and defecating on himself. With concerns that an undetected neurologic process had occurred secondary to administration of nivolumab, a short course of prednisone was started but without improvement in symptoms.

Psychiatry was consulted to investigate the possibility of conversion disorder. On initial assessment, the patient was soft-spoken, highly guarded, and minimizing of any psychiatric component. Over the course of several days, he endorsed a great deal of anxiety related to his mortality and fear of implications of recurrent Hodgkin's lymphoma. He seemingly lacked a forum to express his inner conflict and felt compelled to appear outwardly strong for the sake of his mother and sister. He eventually spoke of his father, who had abandoned the family around the time of the initial lymphoma diagnosis, and he expressed a great deal of anger and resentment about this situation. During each subsequent encounter with the psychiatry consultation service, the patient's entire body began to shake and jar when topics of emotional salience were brought up.

After a comprehensive medical workup revealing no abnormal findings and over 2 weeks of psychiatric consultation, the patient was diagnosed with PNES. Upon comprehensive and compassionate communication of the diagnosis, he showed significant improvement in symptoms of whole-body jarring and twitching. Over the next few days, he eventually got out of bed, began using the commode, and was able to carry out basic functions independently. Upon discharge from the hospital, the family reluctantly agreed to participate in outpatient psychiatric treatment.

Discussion

Although the initial presentation of our patient is consistent with cytokine release syndrome following the infusion of a monoclonal antibody, the progressively worsening seizure-like activity and decompensation several days postinfusion indicate a distinctly different etiology (6). One of the biggest challenges in diagnosing PNES is the lack of uniformly agreed-upon diagnostic criteria and varied presentation. Although video EEG is generally considered to be the gold standard in the diagnosis of PNES, it is important to understand that negative findings only prove that there are no epileptic seizures occurring (2, 7, 8). Several underlying conditions can potentially mimic the symptoms of PNES. As with the present case, the medical causes of PNES-like symptoms are numerous and must be ruled out before considering a diagnosis of PNES. In addition, N-methyl-d-aspartate receptor autoimmune encephalitis, frontal lobe seizures, and acute disseminated encephalomyelitis must be kept on the differential (8–10). Other methods used to help diagnose PNES include clinical features of the seizure, levels of serum prolactin immediately following a seizure, seizure provocation, and various psychological techniques, such as hypnotic regression and avoidance reactions (2, 11).

Many clinicians rely on the clinical features of seizure activity to distinguish PNES from epileptic seizures (see box) (12). Although there is literature to support this, it is important to remember that the behavioral characteristics of PNES are variable and can, at times, mimic epileptic seizures. Moreover, both PNES and epileptic seizures can simultaneously occur in the same patient. Therefore, no single feature of a seizure should be thought of as sensitive or specific for differentiating PNES versus epileptic seizures (2, 12). An acute stressor frequently precedes the onset of PNES. Past psychiatric history will often reveal physical or sexual abuse as well as a variety of comorbid disorders, such as somatoform disorders, dissociative disorders, affective disorders, personality disorders, posttraumatic stress disorder, and other anxiety disorders (1, 2). Again, the difficulty of diagnosing PNES arises from the fact that the comorbidities are numerous, and stressors can take on a variety of forms. In summary, the diagnosis of PNES should encompass all available data and should not rely on any single observation.

Management of PNES is multidisciplinary and long-term. Similar to diagnosis of PNES, there is no single or agreed-upon method of treatment. The most serious morbidity involved in the treatment of PNES is from inappropriate treatment with antiepileptic drugs. Accumulating toxic levels, the danger posed to pregnant patients, and other iatrogenic harm has been implicated in unnecessary treatment with antiepileptic drugs (10). Currently, there is no high-quality evidence to support the use of antidepressants in the treatment of PNES, and studies with large sample sizes are still needed to clarify the role of psychotropics (7). A review of multiple studies that focused on comprehensive and compassionate communication of PNES diagnosis showed promising results in decreasing seizure-like activity. In one study, up to 14% of patients were seizure-free at a 3-month follow up, and 63% reported more than 50% reduction in seizure frequency (7). Interestingly, the efforts made in the careful deliberation of our patient's diagnosis resulted in immediate resolution of his seizure-like spells.

The use of psychological treatments, including cognitive-behavioral therapy, psychodynamic therapy, and group therapy has shown promising results. A recent meta-analysis that evaluated evidence for the efficacy of psychological interventions showed that 82% of patients with PNES who completed psychotherapy reported a reduction in seizures ≥50% (13). This analysis highlights the potential role of psychotherapy in the treatment of PNES. Other augmenting therapy options, such as hypnosis, mindfulness-based therapy, and eye-movement desensitization and reprocessing, have been introduced but lack high-quality evidence for efficacy (7, 11, 14). Although there is still much to learn in the realm of the pathophysiology and treatment of PNES, maintaining flexibility in the delivery of treatment options and incorporating a multidisciplinary approach can result in successful outcomes in clinical practice.

Key Points/Clinical Pearls