Comprehensive analysis of presurgical factors predicting psychiatric disorders in patients with refractory temporal lobe epilepsy and mesial temporal sclerosis underwent cortico‐amygdalohippocampectomy

Background This study aimed to evaluate the predictive value of presurgical factors for psychiatric disorders (PD) in refractory temporal lobe epilepsy and mesial temporal sclerosis (TLE‐MTS) patients underwent cortico‐amygdalohippocampectomy (CAH). Methods A total of 98 refractory TLE‐MTS patients underwent CAH were consecutively enrolled in this cohort study. Several presurgical factors were recorded, such as married status, employment status, highest education, disease duration, family history of epilepsy, and disorganized VEEG background activity. Results There were 17 (17.3%) refractory TLE‐MTS patients occurring PD after CAH, including 8 (8.2%) mood disorders, 7 (7.1%) anxiety disorders, 8 (8.2%) psychoses, and 1 (1.0%) interictal dysphoric disorder. Employed status correlated with low PD occurrence, while disease duration and asymmetric VEEG background activity positively correlated with PD occurrence. Multivariate logistic analysis revealed employed status (P = 0.009) could independently predict lower PD occurrence, while highest education (P = 0.027), disease duration (P = 0.028), seizure frequencies (P = 0.015), and asymmetric VEEG background activity (P = 0.034) could independently predict higher PD occurrence. Receiver operating characteristic curve showed combination of these five factors (area under curve (AUC) = 0.871, 95%CI: 0.783‐0.960) disclosed a great predictive value of PD occurrence. The sensitivity and specificity were 70.6% and 92.6% at the best cutoff point. In addition, the percentage of PD was increased with higher Engel classification (P = 0.003). Conclusion Employed status, highest education, disease duration, seizure frequencies, and asymmetric VEEG background activity correlate with PD occurrence independently in epileptic patients.


| INTRODUC TI ON
Epilepsy, a transient occurrence of signs and/or symptoms owning to abnormal excessive or synchronous neuronal activity in the brain, is a kind of chronic disorder that frequently causes long-term medical, psychological, and social sequelae. 1,2 According to the report of World Health Organization (WHO), there are estimated more than 50 million epilepsy patients and 2.4 million new cases occurring worldwide per year. 3 For adult epilepsy patients, approximately 60% patients are diagnosed as partial-onset epilepsy, whose most common subtype is temporal lobe epilepsy (TLE), and mesial temporal sclerosis (MTS) is considered as the most common pathological substrate of TLE. 4,5 For these TLS-MTS patients, most of them have been reported to be refractory to antiepileptic drugs (AEDs), but have favorable responses to surgery, including cortico-amygdalohippocampectomy (CAH), which is a type of anterior and mesial temporal lobectomy (ATL) and one of optimal choices in patients with refractory epilepsy. 5,6 Although epilepsy surgery contributes to estimated 60%-70% remission rate for long-term clinical symptoms, the quality of life (QoL) is still difficult to be improved, partially due to high prevalence rate of psychiatric comorbidity in TLE-MTS patients. There are about 20%-40% TLE-MTS patients combining with psychiatric disorders (PD), and more than 70% refractory patients also occur this seizure disorder. [7][8][9][10][11][12] High prevalence of PD might contribute to worse outcomes of surgery in refractory TLE-MTS patients. Thus, more understanding about the presurgical factors predicting PD in refractory TLE-MTS patients underwent CAH is necessary.
The risk factors related to PD in epilepsy patients after surgery have raised concern, and some researchers suggest that several sociodemographic and clinical factors are correlated with PD among epilepsy patients, including employed status, seizure-free period, and marital problems. 13,14 However, little is known about the presurgical predictors for PD in refractory TLE-MTS patients received CAH. Thus, the purpose of this study was to evaluate the predictive value of presurgical factors for PD in refractory TLE-MTS patients underwent CAH.

| Procedures
Before CAH, patients received continuous video-electroencephalographic (VEEG) monitoring with 32-channel EEG recording for two-six days, accompanied with electrodes placed according to the 10-10 system on the temporal lobe, including sphenoidal electrodes. The CAH procedure consisted of an en block resection of the superior, middle, inferior temporal, and fusiform gyri, with a posterior limit of 4.5 cm from the tip of the temporal lobe, and after opening the temporal horn, the mesial temporal structures (hippocampus, amygdala, and parahippocampal gyrus) were also resected as described in previous studies. 16,17

| Data collection
Comprehensive factors preoperation were recorded including age at surgery, gender, married status, employment status, highest education, history of smoke, history of drink, age at epilepsy onset, disease duration, family history of epilepsy, family history of PD, seizure frequencies, presence of febrile seizures, presence of left-sided MTS, disorganized VEEG background activity, asymmetric VEEG background activity, contralateral slow waves on VEEG, and contralateral epileptiform discharges on VEEG.

| Assessments and definitions
(a) Refractoriness to medical treatment was defined as seizures persisted after the utilization of at least two first-line medications for partial seizures at the highest tolerated doses for at least 6 months.  (Table 1) 22 In addition, other epilepsy commonly related to PD such as psychoses and interictal dysphoric disorder, not covered by the DSM-IV, was also assessed using the ILAE criteria. 23

| Statistics
Statistical analysis was performed using SPSS 22.0 software (IBM, Armonk, NY) and Graphpad Prism 6 software (GraphPad Software Inc, New York, NY, USA). Data were presented as mean ± SD or count (percentage). Comparison was determined by t test, chi-square test, or Wilcoxon rank sum test. Baseline characteristics affecting PD occurrence were determined by univariate logistic regression analysis, while all factors with P value no more than 0.1 were further detected by multivariate logistic regression analysis. Furthermore, receiver operating characteristic (ROC) curve was used to assess the predictive value of the factors affecting PD occurrence. P value <0.05 was considered significant.

| Engel classification of postoperative outcome
Engel classification was used for the assessment of the treatment efficiency. As presented in Table 3, the number of refractory TLE-MTS

| Baseline characteristics affecting PD occurrence after CAH by logistic regression model analysis
Univariate logistic regression analysis was performed to evaluate baseline characteristics affecting PD occurrence in refractory TLE-MTS patients underwent CAH (Table 6). Employed status (P = 0.048) was associated with lower PD occurrence, while high-

| Correlation of PD occurrence after CAH with Engel classification
In order to further evaluate the influence of treatment efficacy on occurrence of PD in refractory TLE-MTS patients underwent CAH, we performed Wilcoxon rank sum test, which revealed that the occurrence of PD was correlated with increased Engel classification (P = 0.003), and the occurrence of PD among refractory TLE-MTS patients with Engel classification Class I, Class II, Class II, and Class IV were 11.0%, 29.4%, 50.0%, and 50.0%, respectively (Figure 2). epilepsy patients. 25 Meanwhile, short-term seizure-free period and employed status also have been reported to predict the presence of psychiatric disorders, including depression, anxiety, and psychoses in epilepsy patients. 14 Another retrospective study discloses an obvious association of a reduced cognitive function with a pejorative psychiatric outcome after the operation in adult patients with MTLE associated with hippocampal sclerosis (HS) that is the most common focal, drug-resistant epilepsy syndrome. 26 Furthermore, other studies indicate that the stresses of family members correlate with the increased risk of behavior problems and psychological disorders in epilepsy patients. 27 Although some interesting results were found in this study, some limitations still existed. Firstly, sample size in this cohort study was relatively small, which only enrolled 98 refractory TLE-MTS patients underwent CAH; further study with a large sample size was necessary. Secondly, all patients recruited in our study were from monocentric, thus, more patients from multicentre are needed to be enrolled in additional study. Thirdly, these patients had only MTS findings on their MRI, and these patients might have dual pathology.

| D ISCUSS I ON
In conclusion, employed status, highest education, disease duration, seizure frequencies, and asymmetric VEEG background activity correlate with PD occurrence independently in epileptic patients.

ACK N OWLED G M ENTS
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.