صفات اختلال شخصیت در بیماران مبتلا به صرع

Abstract Objective and methods: The Questionnaire on Personality Traits (VKP: Vragenlijst voor Kenmerken van de Persoonlijkheid) was used to investigate personality disorder (PD) traits in 203 patients with epilepsy and a control group of 332 subjects from the general population. Furthermore, the association of PD traits with epilepsy-related variables was studied, as well as the association between PD traits and level of psychopathology. Results: The results showed that, compared with the control group, patients with epilepsy had higher dimensional VKP scores for several Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) and International Classification of Diseases (ICD-10) PDs. Associations were found between PD traits and age at onset of epilepsy, duration of epilepsy, seizure frequency and number of anti-epileptic drugs. Anxiety and depression were not associated with PD traits. Conclusion: It is likely that suffering from epileptic seizures negatively influences personality development and can result in the development of maladaptive PD traits. The results also support the idea that PD traits are not (completely) covered by axis I psychopathology and therefore should be separately investigated.

INTRODUCTION The relationship between epilepsy and personality disorders (PDs) is not often the subject of scientific investigation. Much research concentrates on psychiatric disorders, such as psychotic disorders, anxiety disorders and mood disorders, whereas PDs are less frequently studied in epilepsy. Most of the literature concerning psychopathology in epilepsy fails to make the distinction between psychiatric disorders and PDs. According to the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV)1 an axis II PD is ‘an enduring pattern of inner experience and behaviour that deviates markedly from the expectations of the individual’s culture, is pervasive and inflexible, has an onset in adolescence or early adulthood, is stable over time, and leads to distress or impairment’. An axis I psychiatric disorder is an illness appearing any time in life, with its own characteristic features, course and prognosis, which usually can be treated successfully. When psychopathology and personality in epilepsy is studied, the Minnesota Multiphasic Personality Inventory (MMPI) is frequently used2., 3., 4. and 5.. The MMPI has been criticised by several authors as an inappropriate instrument for the detection of interictal behaviour and personality disturbances in epilepsy6. and 7.. Bear and Fedio8 who noted the shortcomings of the MMPI, developed the Bear-Fedio Inventory (BFI), a rating scale of 18 behavioural traits that previously had been associated in literature with temporal lobe epilepsy (TLE). They found an increased frequency of all 18 traits in patients with TLE, compared with normal and neurological controls, including obsessionality, dependency, emotionality, irritability, religiosity and philosophic interest. Other studies using the BFI showed mixed results, with generally increased behavioural traits in epilepsy patients (both TLE and primarily generalised epilepsy) compared with normal controls9., 10. and 11.. Some investigators use alternative methods like classification by a psychiatrist, to assess psychosis, depressive disorders, anxiety disorders and PDs. In a study by Schwartz and Cummings12 the medical records of 21 epilepsy patients and 24 neurological control patients were reviewed. A psychiatrist classified PDs according to DSM-III in eight patients with epilepsy (38%) and one control patient (4%). Fiordelli et al. 13 studied psychiatric disturbances in 100 epilepsy patients and 100 matched control patients. After administration of a psychiatric interview (Clinical Interview Schedule: CIS), 19 epilepsy patients and 15 control patients were identified as having psychiatric disorders. Subsequently, a psychiatrist classified these patients following DSM-III-R criteria. PDs were found in four patients with epilepsy (21%) and in none of the control patients. Only few studies exist assessing PDs by means of standardised diagnostic instruments based on objective diagnostic criteria. Lopez-Rodriguez et al. 14 used the Structured Clinical Interview for DSM-III-R PDs (SCID-II) for investigating PDs in 52 epilepsy patients. They found PDs in 11 patients (21%), especially cluster C disorders (15%). Avoidant and dependent PDs were the most common diagnoses. Also, Victoroff 15 found axis II PDs in 11 subjects out of 60 epilepsy patients by using the patient version of the SCID. Personality disorder Not Otherwise Specified (NOS) prevailed. Manchanda et al. 16 investigated both DSM-III-R axis I and II disorders in 300 epilepsy patients who where candidates for epilepsy surgery. They found PDs in 18% of the patients, especially dependent and avoidant PDs. Also, Arnold and Privitera 17 found axis II PDs in 18% of the epilepsy patients using the epilepsy version of the SCID. The most common diagnosis was the avoidant PD, which was present in all patients. Comparing previous study results of interictal PDs in epilepsy patients is complicated, due to the use of different patient samples. Additionally, patient groups are usually small and frequently no control group is included. Even more important is the use of a variety of diagnostic instruments: many studies use instruments such as the MMPI and BFI which assess personality ‘traits’ underlying personality psychopathology, and not PDs as defined in widely accepted categorical systems such as the DSM and International Classification of Diseases (ICD). This study investigates PDs in Dutch patients with epilepsy by means of a self-report questionnaire assessing PDs according to diagnostic criteria of the DSM-IV and ICD-10. Patients are assessed for each of the PDs but each separate disorder is conceptualised as a continuum. In this way the degree to which a patient exhibits the traits of a particular PD is determined. The results of the epilepsy patients are compared with a control group consisting of people from the general population. Because patients with epilepsy often experience mood and anxiety disturbances18, we also investigated their level of general psychopathology. Furthermore, the association of PD traits with epilepsy-related variables was explored. Finally, the relationship between general psychopathology and PD traits was determined in order to investigate whether it is worthwhile to assess PD traits independent of and above assessing the level of psychopathology.

RESULTS As shown in Table 1, the epilepsy group significantly differed from the control group on the demographic variables age, sex and education: they were significantly older (P<0.05), had a lower level of education (P<0.001) and were predominantly male (P<0.01). Because of these differences in demographic characteristics, a covariance-analysis was done correcting for age, sex, and education in order to investigate differences on PD traits between both groups. The results of this covariance-analysis are shown in Table 2. Compared with the control group, patients with epilepsy showed significantly higher dimensional scores on seven DSM-IV PDs, namely schizoid, antisocial, histrionic, avoidant, dependent, passive aggressive and depressive. Concerning ICD-10 PD traits, they scored significantly higher on the paranoid, schizoid, dyssocial, histrionic, anxious and dependent PD. For both DSM-IV and ICD-10 a significantly higher total dimensional score was found in the epilepsy patients. Besides these group-effects, also a certain number of significant age-, sex- and education effects were found. Table 2. Covariance analysis (ANCOVA) with group, sex and education as fixed factors, age as covariant and PD traits as dependent variable. PD traits Corrected means (SD)a F-values Epilepsy Controls Group Age Sex Education DSM-IV Paranoid 1.48 (1.70) 1.37 (1.49) 0.58 1.11 7.58** 1.54 Schizoid 1.14 (1.29) 0.68 (1.14) 15.51*** 30.60*** 6.60* 2.08 Schizotypical 1.31 (1.60) 1.04 (1.40) 3.69 8.06** 0.03 1.02 Antisocial 1.20 (2.18) 0.77 (1.90) 4.74* 9.38** 35.73*** 0.37 Borderline 1.54 (1.77) 1.44 (1.54) 0.33 7.48** 1.56 3.28* Histrionic 1.29 (1.35) 0.76 (1.19) 19.13*** 0.80 0.44 1.32 Narcissistic 1.00 (1.53) 1.07 (1.35) 0.28 5.13* 9.54** 0.65 Avoidant 1.77 (2.08) 1.29 (1.81) 6.48* 0.12 17.14*** 4.59* Dependent 1.72 (1.74) 0.93 (1.53) 25.50*** 0.59 6.39* 5.12** Obsessive-compulsive 1.84 (1.76) 1.71 (1.54) 0.65 9.35** 1.22 0.82 Passive aggressive 0.99 (1.27) 0.71 (1.10) 5.99* 3.51 0.02 0.65 Depressive 1.67 (1.76) 0.97 (1.54) 19.46*** 0.99 13.31*** 0.28 NOS 15.76 (12.88) 12.73 (11.25) 6.88** 1.61 0.50 2.90 ICD-10 Paranoid 1.75 (1.59) 1.45 (1.38) 4.63* 11.35*** 10.20*** 2.65 Schizoid 1.45 (1.58) 0.96 (1.37) 12.11*** 19.98*** 3.97* 0.92 Dyssocial 0.81 (0.98) 0.58 (0.85) 6.85** 6.33* 28.45*** 0.44 Impulsive type 1.03 (1.31) 1.02 (1.15) 0.02 3.41 3.09 6.07** Borderline 1.90 (1.90) 1.72 (1.77) 0.99 8.44** 5.50* 5.40** Histrionic 1.03 (1.11) 0.59 (0.96) 19.42*** 7.34** 3.64 0.79 Anankastic 1.67 (1.66) 1.62 (1.46) 0.09 7.00** 0.70 0.35 Anxious 1.36 (1.58) 0.78 (1.37) 17.16*** 0.03 19.52*** 4.59* Dependent 1.60 (1.58) 1.05 (1.38) 15.08*** 0.09 18.79*** 4.91** NOS 12.73 (8.98) 8.75 (7.84) 24.36*** 0.29 4.37* 3.87* SD: standard deviation. a Mean dimensional scores for the PDs, corrected for age, sex and education. * P<0.05. ** P<0.01. *** P<0.001. Table options Compared with normative data of the general population (n=1009) 20, epilepsy patients scored significantly higher on the SCL-90 sub-scales for anxiety (P<0.001), agoraphobia (P<0.02), depression (P<0.001), somatisation (P<0.02), insufficiency (P<0.001) and psychoneuroticism (P<0.001). Within the epilepsy group, patients with TLE scored significantly higher on the SCL-90 sub-scales for insufficiency (P<0.05) and sensitivity (P<0.05), compared with extra-TLE patients. The results of the linear multiple regression analyses, in which the association of PD traits with epilepsy-related variables was explored, are shown in Table 3. Cluster C PD traits avoidant and obsessive-compulsive were positively associated with number of anti-epileptic drugs, duration of epilepsy, seizure frequency (avoidant only) and age at onset (obsessive-compulsive only). Age at onset and duration of epilepsy were also positively associated with the schizoid PD traits. Regarding ICD-10, schizoid and anankastic PD traits were positively associated with age at onset and duration of epilepsy. Impulsive type, borderline and anxious PD traits were positively associated with seizure frequency, whereas anxious was also positively associated with duration of epilepsy and number of anti-epileptic drugs. Table 3. Multiple regression (forced entry) with significant regression coefficients (Beta’s) and squared multiple correlations (R2) with the VKP PD traits as dependent variables and the epilepsy-related variables as independent variables. PD traits Epilepsy-related variables Age at onset epilepsy Duration epilepsy Seizure frequency Number of AEDs Localisation epileptogenic zone R2 DSM-IV Paranoid Schizoid 0.31 0.41 0.10 Schizotypical Antisocial Borderline Histrionic Narcissistic Avoidant 0.25 0.18 0.20 0.10 Dependent OCD 0.21 0.24 0.17 0.09 Passive aggressive Depressive ICD-10 Paranoid Schizoid 0.29 0.36 0.09 Dyssocial Impulsive type 0.18 0.06 Borderline 0.16 0.04 Histrionic Anankastic 0.24 0.24 0.10 Anxious 0.24 0.20 0.19 0.07 Dependent AEDs: anti-epileptic drugs, OCD: obsessive-compulsive disorder. Table options For the epilepsy group, we also computed Pearson correlations between the VKP and SCL-90 sub-scales. Significant positive correlations were found between most scales for PD traits and SCL-90 sub-scales. In particular, the SCL-90 sub-scale sensitivity correlated high with most of the VKP sub-scales. Highest correlations were found between avoidant and sensitivity (r=0.62, P<0.001), anxious and sensitivity (r=0.59, P<0.001) and schizotypical and sensitivity (r=0.57, P<0.001). Multiple stepwise regression was carried out with the VKP sub-scales for PD traits as dependent variables and the SCL-90 sub-scales as independent variables ( Table 4). Regarding DSM-IV PD traits, the avoidant PD is the best represented by the SCL-90 sub-scales (R2=0.45). For the ICD-10 PD traits, highest R2s were found for the impulsive type (R2=0.35) borderline (R2=0.36) and anxious (R2=0.38) PD. Furthermore, the results showed that the SCL-90 sub-scale sensitivity is a strong predictor for all VKP sub-scales. The sub-scales agoraphobia, anxiety, depression, sleeping problems and somatisation had minimal predictive value. Table 4. Multiple regression (stepwise) with significant regression coefficients (Beta’s) and squared multiple correlations (R2) with the VKP PD traits as dependent variables and the SCL-90 sub-scales as independent variables. Personality disorder traits SCL-90 sub-scales ANG AGO DEP SOM IN SEN HOS SLA R2 DSM-IV Paranoid 0.49 0.24 Schizoid 0.36 0.13 Schizotypical 0.57 0.32 Antisocial 0.22 0.05 Borderline 0.29 0.28 0.25 Histrionic 0.24 0.20 0.15 Narcissistic −0.20 −0.19 0.47 0.29 0.24 Avoidant 0.14 0.25 0.56 −0.24 −0.15 0.45 Dependent 0.20 0.38 0.28 OCD 0.25 0.21 0.17 Passive aggressive −0.18 0.33 0.41 0.31 Depressive 0.40 −0.25 0.32 0.32 ICD-10 Paranoid 0.38 0.14 Schizoid 0.42 0.18 Dyssocial −0.35 0.32 0.28 0.15 Impulsive type 0.30 0.37 0.35 Borderline 0.22 0.26 0.22 0.36 Histrionic 0.33 0.20 0.22 Anankastic 0.32 0.23 0.25 Anxious 0.20 0.49 0.38 Dependent 0.45 0.21 ANG: anxiety; AGO: agoraphobia; DEP: depression; SOM: somatisation; IN: insufficiency; SEN: sensitivity; HOS: hostility; SLA: sleeping problems; R2: squared multiple correlation. OCD: obsessive-compulsive disorder.