Vulnerability to psychogenic non-epileptic seizures is linked to low neuropeptide Y levels
Abstract
Psychogenic non-epileptic seizures (PNES) is a conversion disorder that reflects underlying psychological distress. Female patients with PNES often present with a history of prolonged stressors, especially sexual abuse. In the current study, we studied the relationship between neuropeptide Y (NPY) and PNES symptoms in women with a history of sexual abuse. NPY has been associated with resilience to stress and we hypothesized that low levels would increase the extent and severity of PNES symptoms in this patient population. Serum levels of NPY, and related hormones were measured in fifteen female PNES patients and sixty female controls. PNES patients reported more severe abuse histories, feeling of abandonment, and decreased perception of quality of life than controls. Importantly, they also had lower NPY levels. Our analysis indicates that low levels of NPY in PNES may confer greater vulnerability to exhibit seizure-like symptoms and lower quality of life.
Introduction
Psychogenic non-epileptic seizures (PNES) are paroxysmal events characterized by changes in responsiveness, movements, or behavior that resemble epileptic seizures but are not accompanied by the synchronous electrophysiological signals that characterize epilepsy. Serious sexual, emotional, or physical abuse is reported in the childhood histories of many patients with PNES (Alsaadi & Marquez, ۲۰۰۵).
Perhaps the best-studied component of the stress response in humans is activation of the HPA axis where the anterior pituitary gland produces and secretes adrenocorticotropic hormone (ACTH) into the general circulation, which in turn induces cortisol (CORT) synthesis and release from the adrenal glands. An important feature of a healthy stress response is that it terminates when the stressor has ended. Termination of the stress response is critical because the prolonged activation of the sympathetic nervous system (SNS) and the HPA axis can lead to long-lasting adverse health consequences such as autoimmune diseases and impaired growth (Guilliams & Edwards, ۲۰۱۰). However, stress-related health problems in PNES patients could also be marked by the alteration of other signaling pathways. Higher levels of testosterone (T) and lower levels of oxytocin (OXT) have been found in women with chronic stress and history of abuse (Heim et al., ۲۰۰۹). Estradiol, which modulates the HPA axis, is lower in women during prolonged stressful situations (Hertting & Theorell, ۲۰۰۲). In addition, high levels of prolactin (PRL) have been found to be elevated in post-ictal events in epilepsy (Bauer et al., ۱۹۹۲).
Another component of the SNS, NPY, acts as a neurotransmitter in the brain and in the autonomic nervous system, modulating the HPA axis through release of ACTH from the anterior pituitary. Due to its action on the SNS, NPY may reverse hyperarousal (Ehlers, Somes, Seifritz, & Rivier, ۱۹۹۷). In addition, the interaction of NPY with CORT helps individuals to cope with stress and sustain health, while hyperarousal and decreased NPY are associated with post-traumatic stress disorder (PTSD) symptoms (Rasmusson et al., ۲۰۰۰). The role of NPY in PNES remains to be elucidated.
Here, we focus on the concomitant activity of the SNS and HPA as parts of the stress management system in PNES through the assessment of basal serum levels of NPY, ACTH, CORT, and related hormones. We hypothesize that low NPY levels, due to their important role in resilience to stress, will be diminished in patients with PNES.
Methods
Participants
Fifteen women diagnosed with PNES (mean age 38 years) were recruited from the non-epilepsy outpatient registry at Loma Linda University Medical Center (LLUMC) and participated in the study. A member of the research team telephoned all the women listed on the relevant LLUMC registry and explained to each prospective subject that she was being invited to participate in this research project because she has seizure symptoms. All patients had been diagnosed with PNES using video EEG monitoring by an epileptologist (L.U.-Z.) to rule out organic causes of seizures and were evaluated for psychiatric symptoms by a clinical psychologist (A.C.) following best practices.
The control group of 60 healthy women (mean age 22 years) were randomly recruited from Scripps College and Claremont Graduate University (CGU) through flyers posted around the Claremont Colleges. Based on the self-reported abuse scores (J-NAAP; Jacobs, ۲۰۰۲), the control group was further divided in healthy controls without abuse (HC), and healthy controls with abuse (HA) scoring higher than 0. Only women were tested in this study because the incidence of PNES is three-fold higher than in men (Alsaadi & Marquez, ۲۰۰۵). Women are also more likely to report significant physical and sexual abuse histories (Alsaadi & Marquez, ۲۰۰۵).
Exclusion criteria were: exhibiting clinically significant suicidal ideation; exhibiting psychotic features; substance abuse disorder (per DSM-4 TR) within the past 6 months; any current or past psychiatric disorder that could interfere with diagnostic assessment or study adherence; treatment with psychoactive medications (other than mood stabilizers or Ambien); clinically unstable medical conditions; clinically significant abnormal laboratory results; clinically verified mixed epileptic and non-epileptic seizures; treatment with an experimental drug or device within 60 days of study enrollment. Only participants (PNES and controls) who did not show any depressive tendency, according to an abridged version of the Beck Depression Inventory (BDI), were included. A clinical psychologist (A.C.) interviewed all participants to rule out the above-mentioned co-morbidities prior to inclusion. The Institutional Review Boards of LLUMC, CGU, and Scripps College approved the study. All participants gave written informed consent prior to commencing participation.
Measures
Psychological measures
All participants enrolled in the study completed a shortened version of the Jacobs Neglect, Abandonment, and Abuse Protocol (J-NAAP; Jacobs, ۲۰۰۲) administered by a clinician to determine the presence and severity of abuse/neglect. The J-NAAP assesses traumatic or stressful life events such as loss or abandonment, serious neglect, physical abuse, emotional abuse, and sexual abuse. Participants also completed the Quality of Life in Epilepsy (QOLIE-89) questionnaire (Devinsky et al., ۱۹۹۵). The QOLIE-89 measures overall quality of life, emotional well-being, role limitations due to emotional problems, social support, social isolation, energy/fatigue, worry about seizures medication effects, health discouragement, work/driving/social functions, attention/concentration, language, memory, physical function, pain, role limitations due to physical problems, and health perceptions. Participants also completed the Experiences in Close Relationships-Revised (ECR-R) questionnaire, a 36-item measure of adult attachment style (Fraley, Waller, & Brennan, ۲۰۰۰). The ECR-R measures individuals on two subscales of attachment: avoidance and anxiety. In general, avoidant individuals find discomfort with intimacy and seek independence, whereas anxious individuals tend to fear rejection and abandonment. To control for a possible impact of depression on the HPA response, the BDI (Beck, Ward, Mendelson, Mock, and Erbaugh, ۱۹۶۱) was used. Scores higher than 20 represent clinically relevant depressive symptoms and therefore excluded from the current study. To further check the presence and the nature of traumatic events, the Life Stressor Checklist-Revised (LSC-R; Wolfe & Kimerling, ۱۹۹۷) was administered. The LSC-R focuses on 30 traumatic events relevant to females, such as abortion, sexual abuse, natural disasters, physical abuse, and others. Participants were asked to indicate if each event occurred in their lives, and for those, they were asked to provide additional information about the age at which the event occurred, when it ended, if they felt they were harmed, if they felt helpless about the event, and how upsetting the event was to them.
Serological measurements
All testing was done between 6 pm and 8 pm to control for the diurnal variations of hormones. A total of 28 ml of blood was drawn from an antecubital vein, using two 8-ml EDTA whole-blood tubes and one 12 ml serum-separator tube using Vacutainer® blood collection kits. After phlebotomy, each tube was immediately stored on ice. The tubes were then placed in a refrigerated centrifuge and spun at 1500 rpm at 4 °C for 12 min. Serum was withdrawn and placed into 2-ml polypropylene Fisherbrand screw cap with O-ring microtubes. The microtubes were immediately placed on dry ice and then transferred to a −۷۰ °C freezer until analysis. All tests were performed at the Endocrine Core Laboratory of the Yerkes National Primate Research Center at Emory University, Atlanta, Georgia, USA. Commercial RIA kits from American Laboratory Products Company, Windham, NH (NPY), DiaSorin, Inc., Stillwater, MN (ACTH), Diagnostic Systems Laboratories, Webster, TX (CORT; PRL; estradiol, E), Beckman Coulter, Webster, TX (testosterone, T), Siemens, Los Angeles, CA (progesterone), and Assay Designs, Ann Arbor, MI (OXT), were used. All inter-assay and intra-assay coefficients of variation were within acceptable bounds ( <15%).
Statistical analyses
All hormone data were log transformed before analysis and values that deviated more 2.5 standard deviations were considered outliers and removed. A total of 19 outliers were removed (NPY = ۲, ACTH = ۰, CORT = ۱, T = ۵, OXT = ۲, progesterone = ۵, estradiol = ۰, PRL = ۴). Hormone ratios were calculated by dividing hormone X by hormone Y followed by the log transformation in order to maintain a linear relationship and to avoid problems related to the choice of ratio. Analysis of variances (ANOVAs) were computed to analyze endocrine and psychological measures. All post hoc analyses were performed with Tukey’s honestly significant difference (HSD) test. Correlations were computed by Pearson product–moment correlations; no covariates were included in the analysis. For all analyses, the significance level was set at α = ۰.۰۵. Results are expressed as mean ± the standard error of the mean (SEM).
Results
The PNES group was significantly older than both HA and HC (p < .۰۰۰۱), but there was no age difference between HA and HC. Every measure of history of stress (LSC-R), including length and type, was substantially more severe in the patient group, χ۲(۱۴) = ۸۱.۷۱, p < .۰۰۰۱ (Table 1). ANOVAs revealed significant differences in abuse (F۲,۷۲=۵۱.۶۹, p < .۰۰۰۱), abandonment (F۲,۷۲=۳۱.۶۲, p < .۰۰۰۱), and QOLIE (F۲,۶۳=۱۱.۰۵, p < .۰۰۰۱) between groups, with PNES patients reporting higher abuse and abandonment (all p < .۰۰۱) and lower QOLIE-89 (p < .۰۰۱) when compared to both HC and HA. Except for abuse (p < .۰۰۰۱), HC and HA did not differ in other psychometric measures. PNES patients also scored higher than HC, but not HA, in attachment avoidance (F۲,۶۰=۳.۷۷, p < .۰۵), yet no differences were observed for attachment anxiety. Details are reported in Figure 1 and Table 2.
Vulnerability to psychogenic non-epileptic seizures is linked to low neuropeptide Y levels
Figure 2. Hormone levels of PNES, HC, and HA. Error bars represent SEM; significance codes: ***p < .۰۰۱, **p < .۰۱, *p < .۰۵ (p-adjusted with Tukey HSD).
Discussion
An altered SNS and HPA axis in PNES patients, marked by lowered basal levels of NPY and CORT as well as higher basal levels of ACTH, were the key findings of this study. We interpret our results in the light of resilience. Resilience is the capacity to cope with adversities, traumas, and tragedies. Greater resilience is associated with better health, so genetic and neurobiological factors might contribute to resilience to stress (Southwick, Southwick, & Charney, ۲۰۱۲). Transgenic rats with overexpression of NPY in the hippocampus show greater resilience (Thorsell et al., ۲۰۰۰). Special Forces soldiers trained to be resilient to stress had significantly higher NPY levels than typical soldiers after the Prisoners of War experience (Morgan et al., ۲۰۰۰). In a stress study with healthy subjects, resilience was positively correlated with urinary CORT and negatively correlated with childhood trauma (Simeon et al., ۲۰۰۷).
Our findings suggest that the endocrine dysregulation is perhaps derived from a long-term exposure to stress consistent with patients’ self-reported abuse. Despite a significant number of PNES patients having been subjected to abuse (Alsaadi & Marquez, ۲۰۰۵), most previous studies did not report the life history of abuse exposure. PNES patients often suffer from chronic stress as a consequence of severe physical and emotional abuse, and our findings indicate that PNES patients may experience permanent damages to the HPA axis. Approximately half of individuals with PNES have co-morbid psychological disorders associated with trauma, including PTSD, anxiety, and depression (Alsaadi & Marquez, ۲۰۰۵). Moreover, the physical manifestation of PNES may be a result of psychological distress (Alsaadi & Marquez, ۲۰۰۵) perhaps driven by physiologic dysfunctions we document here. However, despite the fact that PNES patients were not depressed at the time of the study, we did not check for other co-morbidities such as PTSD or anxiety.
Hormone levels
A key finding from this study is that serum levels of NPY were lower in PNES patients than in the control groups (Table 2). Low NPY levels have previously been reported in individuals with PTSD (Charney, ۲۰۰۴; Rasmusson et al., ۲۰۰۰), but, in line with our results, such a reduction was more accentuated following trauma exposure rather than with PTSD per se (Morgan et al., ۲۰۰۳). Blunted NPY may be linked to the development of stress-related conditions such as PTSD (Rasmusson et al., ۲۰۰۰) or PNES, as in our case, for vulnerable individuals. NPY is also linked to depression (Charney, ۲۰۰۴). Yet, despite a well-known link between depression and PNES (Alsaadi & Marquez, ۲۰۰۵), the patients in our study were not depressed at the time of data collection as confirmed by both the BDI and the lack of depressive symptoms during clinical evaluation. However, given the potential antidepressant action of NPY in the brain (Morales-Medina, Dumont, & Quirion, ۲۰۱۰), the low NPY in the PNES population might have contributed to decreasing their resilience to stress and accentuating the consequences of their traumatic experiences, manifesting as conversion disorder. Interestingly, low NPY has been also linked a higher incidence of epileptic seizures; by reducing neuronal excitability (Reibel et al., ۲۰۰۱), NPY has been found effective in suppressing epileptic seizures in animal models (Stroud, O’Brien, Jupp, Wallengren, & Morris, ۲۰۰۵). We do not exclude that other pathological pathways are operating in PNES, but to our knowledge no studies to date have investigated NPY in PNES, therefore it is possible that lowered NPY levels might have contributed to the emergence/occurrence of PNES symptoms.
A trend, although non-significant, was found for ACTH that increased with abuse scores across the three groups (Figure 2), with PNES patients having significantly higher ACTH than HC. When compared to epilepsy, patients with PNES have a higher risk of developing PTSD (Dikel, Fennell, & Gilmore, ۲۰۰۳). Previous studies reported no difference in basal ACTH levels in PTSD patients when compared to healthy controls, even though CORT levels were significantly lower (Kanter et al., ۲۰۰۱). Yet, interestingly, women who experienced a history of sexual abuse without having developed major depression—as in our patient population—produce more ACTH and lower CORT in response to CRH compared to controls and women without sexual abuse but with major depression (Heim, Newport, Bonsall, Miller, & Nemeroff, ۲۰۰۱). Given that our patients also reported significantly higher rates of sexual abuse, but they did not develop major depression, we suggest that a history of sexual abuse per se might be a causative factor for an increased ACTH in PNES patients. Lower NPY levels may exacerbate this vulnerability.
We also found significantly decreased CORT levels in PNES patients. The literature is inconsistent when testing CORT in PNES subjects; it has been reported as increased (Bakvis et al., ۲۰۱۰) or equal (Bakvis, Spinhoven, & Roelofs, ۲۰۰۹) to that of healthy controls. Bakvis et al. (۲۰۱۰) found increased basal diurnal CORT in PNES patients with history of sexual abuse compared to both controls and PNES without abuse. Inconsistencies might be caused by sampling (saliva versus serum), time of the day (early versus late afternoon), number of measures (single versus multiple), group size, and sex ratio (only women versus men and women). An alternative explanation for such conflicting results could be that the high-CORT phase that follows an intensely stressful situation may last years before resulting in hypocortisolism (Guilliams & Edwards, ۲۰۱۰). Thus, CORT levels may depend on the time between the stressful event and CORT assessment, or with the intensity of abuse. Accordingly, our HA group had higher CORT levels than HC, though not significantly. We thus speculate that our low CORT levels might be due to the early and extended (>2 months) occurrence of abuse in our patients (Table 1). In line with our results, women with a history of intimate partner violence and rape have lower plasma CORT levels than controls (Seedat, Stein, Kennedy, & Hauger, ۲۰۰۳). Hypocortisolism may therefore be a result of traumatic early life experiences (Gunnar & Vazquez, ۲۰۰۱).
We found lower PRL in PNES patients compared to HC and HA subjects. Peripheral PRL can cross the blood–brain barrier and can act as a neuropeptide in the brain protecting the hippocampus from the anti-proliferative effects of glucocorticoids (Torner et al., ۲۰۰۹). PRL has been found to attenuate hormonal and neuronal responses to stressors reducing gastric ulcers and protecting neurogenesis in chronically stressed mice (Torner et al., ۲۰۰۹). Low peripheral PRL levels have been found in women with post-partum depression compared to non-depressed post-partum women (Abou-Saleh, Ghubash, Karim, Krymski, & Bhai, ۱۹۹۸). As such, lowered levels of PRL in PNES patients might result from chronic stress and decreased resilience to stress.
Hormone interactions
In healthy individuals, the balance between hormones within the SNS and HPA systems is maintained by negative feedback. Because of this interaction, a change in a single hormone produces a cascade of effects within the same system and/or between the SNS and HPA axis. We found an increased ACTH to CORT (ACTH/CORT) ratio in PNES patients (Table 3 and Figure 2). Dissociations in the HPA axis are quite common, especially when patients face high stress (Bornstein, Engeland, Ehrhart-Bornstein, & Herman, ۲۰۰۸). Consistent with other findings (Yehuda, Golier, Halligan, Meaney, & Bierer, ۲۰۰۴, we did not detect a correlation between ACTH and CORT in any of the groups. In fact, factors other than pituitary ACTH—such as other neuropeptides, neurotransmitters, and sexual hormones—might affect CORT secretion, leading to dissociations between ACTH and CORT (Bornstein et al., ۲۰۰۸.
The interaction between SNS and the HPA axis and their feedback is consistent with the reduced NPY and CORT, and higher ACTH in PNES patients. A possible explanation is that NPY and CORT cease to be (fully) regulated by ACTH. Indeed, besides the endocrine ACTH-mediated mechanism, CORT production can be stimulated by both neural and immune pathways (Bornstein et al., ۲۰۰۸ Hypothalamic NPY neurons are equipped with glucocorticoid receptors (Härfstrand et al., ۱۹۸۹), and when dexamethasone (a selective glucocorticoid receptor agonist) is administered, hypothalamic NPY levels increase in rats (Corder et al., ۱۹۸۸). In this fashion, a negative feedback may occur between NPY and CORT: the low CORT contributes to lower NPY levels, which in turn further reduces CORT production. Indeed, the adrenal glands are connected to the SNS through the splanchnic nerve, on which activation depends on the production and release of both NPY (Renshaw & Hinson, ۲۰۰۱) and CORT (Bornstein et al., ۲۰۰۸).
Although not significant, we found decreased estradiol levels in PNES participants. When the ACTH to estradiol (ACTH/E) ratio was computed, PNES patients showed significantly lower values than HC but not HA. Chronic stress has been shown to reduce estradiol levels (Hertting & Theorell, ۲۰۰۲). At the same time, protecting the hippocampus, estradiol modulates psychosocial stress (Albert, Pruessner, & Newhouse, ۲۰۱۵). In addition, estradiol is able to modulate CORT and ACTH in response to stress (Kirschbaum, Kudielka, Gaab, Schommer, & Hellhammer, ۱۹۹۹). Lastly, it has been suggested that estrogens and progesterone might promote resilience in women (Charney, ۲۰۰۴). Thus, it appears that the balance between HPA and hypothalamic–pituitary–gonadal (HPG) axis, measured by the ACTH/E ratio, may be a resilience marker.
PNES subjects had an altered CORT to T (CORT/T) ratio though not higher free T compared to controls. In women, T is produced in equal quantity by both adrenal zona fasciculata and ovarian stroma, cumulatively contributing to 50% of the total circulating T (Burger, ۲۰۰۲). Although the relationship between T and physical/psychological stress remains to be elucidated, some studies found diminished T in PTSD (Charney, ۲۰۰۴). Yet, in soldiers, when PTSD was not accompanied by other comorbid conditions, T levels were reported to be higher than controls as well as soldiers with both PTSD and major depression (Karlović et al., ۲۰۱۲). This finding is consistent with an elevated CORT/T, a phenomenon that follows a prolonged period of stress in which the CORT/T ratio is high (Doan, Newton, Kraemer, Kwon, & Scheet, ۲۰۰۷). Deriving from the same biochemical precursor (cholesterol), an increase of one corresponds to a decrease of the other because of a competitive reaction process (Lee et al., ۲۰۱۶). In addition, CORT may reduce T production by blocking T effects in target tissues. In turn, T may reduce CORT levels through the hypothalamus (Romero-Martínez, González-Bono, Lila, & Moya-Albiol, ۲۰۱۳).
Adrenal selective damages
Inhibited negative feedback from ACTH to CORT might be an inadequate explanation for our results as a previous study found that the dexamethasone suppression test did not reduce CORT in PNES patients (Bakvis et al., ۲۰۱۰). Indeed, the CORT negative feedback has not been associated with an altered ACTH/CORT ratio (Yehuda et al., ۲۰۰۴). An alternative explanation is reduced adrenal sensitivity, when the adrenal glands become unresponsive to ACTH and do not release enough CORT. Low circulating CORT is then an insufficient signal to the pituitary to stop producing ACTH. Considering this as a valid alternative explanation for our findings, further evidence comes from the primary adrenal insufficiency that is characterized by high ACTH and low CORT levels (Arlt & Allolio, ۲۰۰۳). Women with adrenal insufficiency are at increased risk for the development of immune-related disorders or pain syndromes (Heim et al., ۲۰۰۱). It has also been reported that adrenal insufficiency might lead to some forms of seizure (Arlt & Allolio, ۲۰۰۳). Although it has been suggested that NPY is unlikely a product of the adrenal glands (Renshaw & Hinson, ۲۰۰۱), our results are somewhat inconsistent with the literature showing that central NPY stimulates ACTH production and inhibits secretion of PRL (Crowley, ۲۰۰۴). If it was so, the lowered NPY would not have signaled ACTH production as well as PRL reduction. The fact that NPY production in the adrenal glands is modulated by ACTH in the zona glomerulosa and by glucorticoids in the inner zones and medulla (Renshaw & Hinson, ۲۰۰۱), and that T is likely to be stimulated by an action of ACTH —accounting for the 25% of circulating T— in the adrenal zona fasciculata (Burger, ۲۰۰۲), we suggest that selective damages to the adrenal glands might have occurred due to over production of tissue-specific hormones, as already seen in other stress-related pathologies such as anorexia nervosa (Lanfranco et al., ۲۰۰۴
Aging, NPY, and self-reported abuse
Several studies report higher NPY levels in older versus younger healthy individuals (Baranowska, Radzikowska, Wasilewska-Dziubinska, Roguski, & Polonowski, ۲۰۰۰. We found a positive correlation between NPY and age in the PNES group, but not in the control groups. This absence of correlation could be due to a significant difference in age distribution in our groups. We also found that in the PNES group ACTH decreased with age. In accordance with previous literature (Heuser et al., ۱۹۹۴), we did not find a decrease of ACTH in healthy controls. Consistent with the linear regression (Table 4) in which age, NPY, and ACTH levels predict abuse, we suggest that as the endocrine profile moves towards normal values with age, the self-reported level of abuse is perceived to be less severe. We interpret our results as the increased NPY and decreased ACTH in PNES patients indicates that with age these two indices of SNS and HPA axis tend to move toward normal values, perhaps suggesting a natural but long recovery from trauma. While we are unable to test this hypothesis, it is consistent with changes in recalled traumatic memories over time in PTSD patients (Vanderkolk et al., ۱۹۹۵).
Limitations and further studies
Some limitations of the present study must be addressed. First, there is a selection bias in our participant sample as only female patients were recruited from an outpatient clinic and volunteered for the study. Since we could not perform a random assignment to condition, PNES patients may have pre-abuse vulnerabilities that explain our findings. Another limitation is our sample size and age distribution. This could be a reason why we did not find hormonal changes in response to age in the control group as other studies have reported. Yet, this weakness disappeared when the whole population was considered, and the regression model revealed indeed an effect of age as well as NPY and ACTH on abuse history, suggesting a more robust and generalizable effect of NPY and ACTH on self-reported levels of abuse. We also acknowledge that single-point measures of CORT and ACTH are not sufficient to extensively assess the HPA axis tone. Contemporary assessment of HPA axis activity typically includes consideration of daily rhythms, with a special emphasis placed on the waking response that is strongly linked to pathology (see e.g. Bakvis et al., ۲۰۱۰).
We also did not take into account the phase of the menstrual cycle and/or hormonal contraceptive use. This might have hindered finding significant differences in estradiol, progesterone, OXT, and testosterone. However, NPY appears unaffected by menstrual cycle (Lewandowski et al., ۱۹۹۸) and no ACTH and CORT differences are evident between luteal and follicular phase of the menstrual cycle (Kirschbaum et al., ۱۹۹۹). Future studies obtain additional measures of hormones, especially for HPG hormones that depend on the ovarian cycle.
Unfortunately, we did not record the onset and duration of illness, the frequency of seizures or the time of the initial abuse. It would also be of interest to check whether hormone levels and ratios are modulated by the time since the trauma(s) and illness occurred as well as a correlation with seizure frequency. Following a traumatic experience, CORT rises and remains stable for a certain amount of time, after which it falls below typical physiological levels (Guilliams & Edwards, ۲۰۱۰) perhaps due to an adrenal tissue damage. This likely affects other SNS and HPA hormones due to feedback effects. Thus, fluctuating interactions within the hormonal environment are expected and most likely depend on when the stressor occurred.
Lastly, given that in our study NPY, PRL, and estradiol seemed to characterize PNES itself, while ACTH and T seemed to vary with the levels of abuse, testing those hormones and their ratios in PNES patients without history of abuse against PNES with abuse would be important in discriminating distinctive features of PNES (see e.g. Bakvis et al., ۲۰۱۰).
Conclusions
The SNS and HPA axis appear to be altered in our cohort of PNES patients. Specifically, PNES patients, compared to healthy controls with and without exposure to abuse, had lower NPY and CORT and increased ACTH plasma levels. The vulnerability to PNES appears to be linked to low NPY levels, indicating that NPY could be a potential biomarker of PNES, while ACTH and CORT might be related to the level of abuse. Thus, we hope that future testing of hormones in the clinic may aid in the detection of PNES among patients presenting with epileptic-like symptoms, and at a lower cost than vEEG. Measuring hormones may also help move PNES patients into appropriate treatment more rapidly to alleviate suffering. Future larger studies involving subjects of both sexes and patients with epileptic seizures are warranted.
