Clinical relevance of interictal dysphoric disorder and its impact on quality of life in drug-resistant epilepsy

Objective: This study aims to assess the prevalence of Interictal Dysphoric Disorder (IDD) in drug- resistant epilepsy (DRE) and to describe its clinical and psychopathological proﬁle, including personality, as well as its impact on quality of life (QOL). Method: A retrospective cross-sectional study from an Epilepsy Unit from January 2007 to December 2017. All patients were diagnosed with DRE. Patients underwent a battery of tests (HADS, SCL-90R, PDQ-4+, QOLIE-31) and a psychiatrist assessed the presence of Axis-I disorders and IDD. Statistical procedures were carried out using R-4.0.1 software. Results: A total of 282 patients were included. A statistically signiﬁcant association was found between IDD and mood and anxiety disorders ( p < 0.001 and p < 0.05 respectively), and between IDD and higher scores in all HADS and SCL-90-R items compared to subjects without IDD ( p < 0.001). A statistically signiﬁcant association was also found between IDD and obsessive–compulsive, borderline and depressive personality disorder ( p < 0.05). Scores in all QOLIE-31 items except for ‘medication effects’ were signiﬁcantly lower in subjects with IDD compared with subjects without IDD ( p < 0.001). Conclusions: In DRE, IDD subjects show differences in the psychopathological proﬁle and QOL scores compared to subjects without a diagnosis of IDD. An early diagnosis of IDD could facilitate prompt interven- tions which might positively impact QOL. (cid:1)


Introduction
Epilepsy is a major health problem as studies point to an estimated prevalence rate of 6.38 per 1,000 persons [1]. Drugresistant epilepsy (DRE) accounts for around 30% of patients with epilepsy [2,3] and implies most of the burden of epilepsy in the population as a result of psychological dysfunction, social stigmatization, cognitive deficits, reduced quality of life, increased risk of mortality and presence of comorbid illnesses [4,5].
Among comorbidities, special focus must be placed on Interictal Dysphoric Disorder (IDD), the concept of which primarily derives from Kraepelin and Bleuler's observations of patients with untreated epilepsy [6,7]. The term IDD was latterly coined by Blumer and described as a pleomorphic affective disorder that appears in epilepsy and is characterized by labile depressive symptoms (depressive mood, anergia, pain, insomnia), labile affective symptoms (fear, anxiety), and the specific symptoms euphoria and paroxysmal irritability [8]. These symptoms occur intermittently, without external triggers, and have a duration that may last from a few hours up to two days, with a rapid beginning and end. The pharmacological treatment for IDD includes antiepileptic drugs, antidepressants, and, sometimes, low doses of neuroleptics.
Despite the high variability in epidemiological data, the actual prevalence of IDD is expected to be high and it is thought to be present to a higher extent in DRE [9] -some authors even consider it the key psychiatric syndrome associated with epilepsy [10].
There is a small body of evidence that describes an existing relationship between IDD, clinical aspects, and quality of life (QOL) in patients with epilepsy [11][12][13], suggesting that IDD is associated with the presence of psychiatric disorders, depressive episodes, and lower QOL. Moreover, no studies have attempted to define or quantify psychiatric symptoms in DRE, such as depression, anxiety, or personality traits, amongst others, and there are no investigations on QOL in this specific group of patients with DRE and IDD. Also, psychiatric symptoms in epilepsy are often difficult to classify into any of the standardized diagnostic systems, as they might not meet full diagnostic criteria. This can lead to underdiagnosis and undertreatment, and thus lower quality of life, as psychiatric symptoms have been proven to be the strongest predictors of QOL in DRE [14].
The current study aimed to assess the occurrence of IDD in DRE and to describe its clinical and psychopathological profile, including personality, as well as its impact on QOL in DRE.

Material and methods
We carried out a retrospective cross-sectional study of patients with DRE from an Epilepsy Unit at University Hospital Clínic de Barcelona from January 2007 to December 2017.

Patients
Adult subjects with DRE that had been admitted to an Epilepsy Unit between January 2007 and December 2017 were enrolled in the study. All patients were considered to have DRE according to the definition of the International League Against Epilepsy (ILAE), which defines DRE as ''failure of adequate trials of two tolerated and appropriately chosen and used AED schedules (whether as monotherapies or in combination) to achieve sustained seizure freedom" [15]. Patients were excluded according to the following criteria: diagnosis of mental retardation, primary sensory or motor impairment altering test performance, non-Spanish speakers, and those who had undergone prior intracranial surgery.
The study was approved by the Human Research Ethics Committee of University Hospital Clínic de Barcelona. All patients were recruited upon written consent.

Assessment
During the subjects' admission to the Epilepsy Unit, they were evaluated by neurologists, psychiatrists, and neuropsychologists. They all underwent video-encephalography monitoring and comprehensive neuropsychological testing. Sociodemographic information (age, gender, education level, employment) and epilepsy characteristics (frequency of seizures, age of onset, frequency, duration, focus, AEDs, neuroimaging) were extracted from clinical records.

Psychiatric assessment
Axis-I disorders were assessed by a psychiatrist following the Structured Clinical Interview for DSM-IV (SCID-IV) [16]. Disorders were grouped into four clusters: mood disorders, anxiety disorders, psychotic disorders, and other disorders.
Anxiety and depressive symptoms were assessed using the Spanish version of the Hospital Anxiety and Depression Scale (HADS) [17]. This scale was designed to measure anxiety and depressive symptoms in non-psychiatric populations during the week prior to the evaluation, to facilitate the detection and management of mood and anxiety disorders [18] and has been established as an efficient screening instrument in people with epilepsy [19]. It consists of 14 multiple-choice items divided into anxiety and depression subscales. The items are scored from 0 to 3 and the final score ranges from 0 to 21.
Psychological distress was ascertained by the Spanish version of the Symptom Checklist-90-R questionnaire (SCL-90R) [20]. This scale assesses symptoms over the previous week, and it is comprised of 90 items, rated 0 to 4, divided into nine primary symptom scales including somatization, obsessive-compulsiveness, interpersonal sensitivity, depression, anxiety, hostility, phobic anxiety, paranoid ideation, and psychoticism. It provides three indexes: the Global Severity Index, which measures overall mental distress, the Positive Symptom Total Score, which considers the total number of symptoms independent of their severity, and the Positive Symptom Distress Index, which indicates the measure of the intensity of symptoms perceived by patients. Higher scores indicate higher psychological distress.
Personality was assessed using the Spanish version of the Personality Diagnostic Questionnaire-4+ (PDQ-4+) [21,22]. It is a self-report questionnaire developed for the evaluation of personality disorders based on DSM-IV criteria. It is made up of a total of 99 items distributed throughout 12 subscales where items are answered using a yes/no format. It assesses ten specific personality disorders (Paranoid, Schizoid, Schizotypal, Antisocial, Borderline, Histrionic, Narcissistic, Avoidant, Dependent, and Obsessive-Compulsive) along with two personality disorders proposed in the DSM-IV Appendix B (Negativistic (Passive-Aggressive) and Depressive). It has demonstrated satisfactory psychometric reliability [23][24][25].
The presence of an Interictal Dysphoric Disorder was evaluated according to the diagnostic criteria established by Blumer [10] following the questions incorporated into the Seizure Questionnaire. These require at least three of eight main symptoms -depressive mood, anergia, pain, insomnia, fear, anxiety, euphoria, or paroxysmal irritability-of moderate or serious severity, leading to moderate or severe limitation. The symptoms must have been present in the previous 12 months.

Quality of life assessment
Quality of life was evaluated using the Spanish version of the Quality of Life in Epilepsy Inventory-31 (QOLIE-31), which evaluates patients' subjective feelings towards their quality of life in various aspects related to epilepsy [26,27]. It consists of seven multi-item subscales that assess the Overall Quality of Life (QOL), energy/fatigue (EF), emotional well-being (EWB), social functioning (SF), cognition (COG), medication effects (ME) and seizure worry (SW). Each subscale has an item measuring the level of subjective distress of that particular subscale. The overall score ranges from 1 to 100. Higher scores indicate higher quality of life. An overall score (QOLIET) results from a weighted average of the multi-item scale scores.

Data analysis
We compared whether patients with IDD differed from patients without IDD in the following variables: the presence of a psychiatric disorder (PD) diagnosis according to SCID-IV, three HAD variables, sixteen PDQ variables, thirteen SCL-90R variables, and eight QOLIE-31 variables. To conduct these comparisons, we used chi-squared tests for categorical variables (e.g., PD diagnosis) and t-tests for numeric variables (e.g., HADA score). To correct for multiple comparisons, we applied Bonferroni correction within each scale. Corrected P values < 0.05 were regarded as statistically significant. All statistical procedures were carried out using R-4.0.1 software.

Sociodemographic and clinical characteristics
A total number of 282 patients met the inclusion criteria. From the total sample, 98 patients had a diagnosis of IDD (34.8%), while 184 (65.2%) did not. Sociodemographic and clinical characteristics, along with differences between groups are summarized in Table 1. The only statistically significant difference found among both groups is marital status.

SCID-IV
As previously stated, 98 patients had a diagnosis of IDD (34.8%), 70 (71.4%) of which were diagnosed with a psychiatric disorder (PD) according to SCID-IV, in contrast to 44 (23.9%) subjects in the group without IDD being diagnosed with a PD (p-value < 0.0 01), suggesting a higher prevalence of any psychiatric comorbidity in IDD patients with an odds ratio (OR) of 7.88. Patients with IDD were also more likely to have a diagnosis of mood or anxiety disorder. See results in Table 2.

HADS
In the analysis of the association between depressive and anxious symptoms and IDD, statistically significant differences were found in all HADS items. Scores in all HADS items were higher in subjects with IDD compared with subjects without IDD. See results in Table 3.

SCL-90-R
In the analysis of the association between different psychiatric symptoms and the presence of IDD diagnosis, scores in all the items of SCL-90-R were statistically significantly higher in subjects with IDD compared with subjects without IDD. See results in Table 3.

PDQ
In the analysis of the association between different personality traits and the presence of IDD diagnosis, scores in the overall cluster C group, and the subgroups obsessive-compulsive, borderline and depressive personality disorder were statistically significantly higher in IDD patients compared with non-IDD patients. See results in Table 3.

Quality of life
In the analysis of the association between QOL and IDD, scores in all items of QOLIE-31 except for ''medication effects" were statistically significantly lower in subjects with IDD compared with subjects without IDD. Results from the total sample of the evaluation of the quality of life using QOLIE-31 are shown in Table 4.

Discussion
The main finding of our study is that DRE subjects with IDD show differences in the clinical and psychopathological profiles compared to subjects without a diagnosis of IDD, and these differences have a relevant impact on QOL, showing lower QOL scores in IDD in comparison with subjects without IDD.
Firstly, from the scarce previous studies that have estimated the prevalence of IDD in epilepsy (without distinction between DRE and not DRE), most results have ranged from 17% to 21.3% [9,12,13,28]. The fact that we have observed a higher occurrence might be explained by the consideration that our sample was composed only of subjects with DRE. De Araujo et al. [9] estimated a prevalence of 18.4% in a sample of DRE. However, it must be noted that they did not perform a diagnosis of IDD if participants met diagnostic criteria for an Axis I disorder according to DSM, which we believe may have contributed to the discrepancy among prevalences. In our sample, one-third of patients with DRE had a diagnosis of IDD, which we consider to be high. Given our results, the presence of IDD should always be considered by physicians when attending patients with DRE, in an attempt to reduce its misdiagnose and possible undertreatment.
Furthermore, it has been suggested that psychiatric symptomatology in epilepsy does not find a place in the current standardized classificatory systems [29]. However, we aimed to evaluate the association between IDD and PD, in order to characterize the disorders that are most frequently associated with IDD in DRE. Previous studies in epilepsy have found high comorbidity of IDD with other psychiatric disorders, especially mood and anxiety disorders [12,28,30], while some have even encouraged future research to validate whether IDD is nosologically independent of other psychiatric conditions. Nonetheless, no investigations are assessing the presence of IDD and other PD in DRE. We report an association between IDD and resistant epilepsy and PD, as more than twothirds of subjects with IDD from our sample (71.4%) had at least one psychiatric diagnosis, indicating that patients with DRE and IDD are at a higher risk of suffering from comorbid PDs, in particular anxiety or mood disorders. Previous studies have stated that psychiatric comorbidities are underrecognized and go untreated in patients with epilepsy [31]. Given the high comorbidity found in our study, one of the key points when assessing these patients should be recognizing comorbid psychopathology.
Whether existing psychiatric classificatory systems are adequate or not in epilepsy, a greater understanding of the psychopathology in these subjects through psychometric scales could also contribute to improving clinical practice in this specific subgroup of patients. We assessed anxiety and depressive symptoms using HADS and found statistically significant differences between patients with IDD and without IDD, where the group with IDD showed higher scores in the anxiety and depression subscales, as well as in the total scale. Previous literature has shown that DRE is significantly associated with anxious and depressive symptoms, which are in turn associated with impaired QOL [32], but no studies have addressed these symptoms in IDD with resistant epilepsy. Continuing with this approach, psychological distress was evaluated with SCL-90-R, which has not been used in precedent literature in the assessment of patients with DRE and IDD. We found that subjects with IDD scored significantly higher in all subscales compared to those without a diagnosis of IDD. As mentioned in the introduction, IDD is characterized by eight main symptoms: depressive mood, anergia, pain, insomnia, fear, anxiety, euphoria, and paroxysmal irritability [8]. However, our results show significant differences in symptoms addressed in SCL-90-R that are not included in the main assessment established for IDD, such as obsessive-compulsiveness, interpersonal sensitivity, paranoid  ideation, and psychoticism. To our knowledge, no previous research has investigated these symptoms in IDD. These results possibly support the pleomorphic nature of IDD, which along with the intermittent duration of its symptoms might favor its frequent failure to meet criteria for standardized diagnostic systems. We believe that this broad range of symptoms has an impact on QOL and could be considered a predictor of lower QOL [14]. Therefore, the identification and management of not only anxious and depressive symptoms, but a wider range of psychiatric symptoms, should become a crucial issue when attending to these patients.
Moreover, research on how epilepsy can affect neural circuits mediating personality has shown that seizures may lead to the development of maladaptive personality traits [33,34]. The presence of abnormal personality profiles in patients with epilepsy is high, being more severe in DRE [35]. An interictal personality has been described [36], formerly known as Gastaut-Geschwind Syndrome and frequently found in temporal lobe epilepsy, but there is limited investigation on further personality traits on IDD. Suda et al [12] found no association between IDD and antisocial personality disorder assessed with the Mini-International Neuropsychi-  atric Interview [37] but did not assess other personality disorders and the sample was not DRE specific. We evaluated all personality disorders using PDQ-4+ and found an association between IDD and Borderline Personality Disorder, Obsessive-Compulsive Personality Disorder, and Depressive Personality Disorder. As personality disorders could influence the ability to treat underlying epilepsy [34], identifying different personality profiles should be of concern, paying specific attention to Borderline, Obsessive-Compulsive, and Depressive personality disorders when treating patients with IDD in DRE, according to our results. Also, personality has been found to play an important role in adjusting to epilepsy, so its assessment is also relevant in the recognition of patients at risk of poorer QOL [38].
Although an extensive literature has examined the impact of epilepsy in QOL, there are relatively few studies investigating the impact of IDD on QOL, none of them being performed on patients with DRE. On one hand, investigations have aimed to relate DRE to impaired QOL [39], highlighting psychiatric symptoms as predominant determinants of QOL [32,40,41]. On the other hand and as aforementioned, patients with IDD have significantly lower QOL [12,13]. Our findings suggest the presence of a strong negative impact of IDD in resistant epilepsy on QOL, in comparison to DRE without IDD. We found differences among all the subscales examined by QOLIE-31 except for ''medication effects". Among these subscales, we find energy/fatigue emotional well-being, cognition, seizure worry, and social functioning, all of which might be accompanied by distress at a psychopathological level. Thus, addressing not only psychiatric comorbidity but specifically IDD is likely to become a beneficial intervention in this group of patients, raising the possibility of novel therapeutic approaches aiming to improve impaired QOL in these subjects.
This study has limitations. Firstly, it is an observational study and lacks a control group. Results should not be generalized as it is a monocentric study. Moreover, the type and duration of AED treatment were not controlled. AEDs have psychotropic effects that might result in psychopathological manifestations, being difficult to differentiate from symptoms secondary to epilepsy itself. Despite using the Seizure Questionnaire to assess IDD, the Interictal Dysphoric Disorder Inventory [42] could have brought greater consistency to the objective assessment of IDD. Finally, subjects diagnosed with different types of epilepsy were included in the study which may lead to certain heterogeneity. However, we controlled for types of epilepsy (idiopathic/secondary), types of seizures, hemispheres, and locus.

Conclusions
We believe our findings have relevant implications for clinical practice as well as research. Our results indicate that the prevalence of IDD in DRE is particularly high and that psychiatric symptoms are determinant in this subgroup of patients when compared to patients without IDD. These results may guide clinicians to a better understanding of the psychiatric comorbidity in DRE, particularly in those patients presenting with IDD, as it must be noticed that despite depressive and anxious symptoms are the most frequently evaluated in IDD, a broader range of psychiatric symptoms (including personality) should be considered, which all contribute to an impaired QOL. Furthermore, many symptoms of IDD might be easily managed with antidepressants and sometimes neuroleptics.
Therefore, and especially in case of inability to achieve seizure freedom, we encourage clinicians to include interventions focused on assessing the presence of IDD as well as addressing accompanying psychiatric symptoms as they are likely to improve QOL.