Biomarkers in the diagnosis and study of psychogenic nonepileptic seizures: A systematic review

Video electroencephalography (vEEG) is the gold-standard method for diagnosing psychogenic nonepileptic seizures (PNES), but such assessment is expensive, unavailable in many centers, requires prolonged hospitalization, and many times is unable to capture an actual seizure episode. This paper systematically reviews other non-vEEG candidate biomarkers that may facilitate both diagnosis and study of PNES as differentiated from epileptic seizures (ES).

PubMed database was searched to identify articles between 1980 and 2015 (inclusion: adult PNES population with or without controls, English language; exclusion: review articles, meta-analyses, single case reports).

A total of 49 studies were examined, including neuroimaging, autonomic nervous system, prolactin, other (non-prolactin) hormonal, enzyme, and miscellaneous marker studies. Functional MRI studies have shown PNES is hyperlinked with dissociation and emotional dysregulation centers in the brain, although conflicting findings are seen across studies and none used psychiatric comparators. Heart rate variability suggests increased vagal tone in PNES when compared to ES. Prolactin is elevated in ES but not PNES, although shows low diagnostic sensitivity. Postictal cortisol and creatine kinase are nonspecific. Other miscellaneous biomarkers (neuron specific enolase, brain derived neurotropic factor, ghrelin, leptin, leukocytosis) showed no conclusive evidence of utility. Many studies are limited by lack of psychiatric comparators, size, and other methodological issues.

No single biomarker successfully differentiates PNES from ES; in fact, PNES is only diagnosed via the negation of ES. Clinical assessment and rigorous investigation of psychosocial variables specific to PNES remain critical, and subtyping of PNES is warranted. Future investigational and clinical imperatives are discussed.