Do auras predict seizure outcome after temporal lobe epilepsy surgery?

PURPOSE: The success of epilepsy surgery lies in identifying the ictal onset zone accurately. The significance of auras has little been explored on surgical outcome in drug-resistant epilepsy. This study focuses on the clinicopathological correlation of aura(s) and its role in predicting surgical outcome in drug-resistant temporal lobe epilepsy (TLE). We compared surgical outcome in TLE between patients with and without aura and identified the clinico-pathological, radiological and surgical differences between the two groups. METHODS: Consecutive patients who underwent presurgical evaluation from January 2009 to December 2014 for drug-resistant TLE who underwent anterior temporal lobectomy (ATL) were included. Patients were followed up at 3months, 12 months and then annually. RESULTS: Among 456 patients, 344(75%) had aura. Multivariate logistic regression showed that prototype EEG pattern at ictal onset (OR 2.12, 95% CI 1.18-3.06, p = 0.012) and right sided epileptogenic zone (OR 1.82 95% CI 1.18-3.78, p = 0.007) were significantly associated with presence of aura. There was no difference in surgical outcome between those with and without aura. But patients with auditory aura (OR 7.28, CI 2.80-18.95, p = 0.0002) and vertiginous aura (OR 3.01, CI 1.55-7.85, p = 0.028) had a poor surgical outcome. Bivariate analysis showed that normal MRI (p = 0.028) and normal/indeterminate pathology (p = 0.001) were significantly more common with auditory/vertiginous auras. CONCLUSIONS: Mere presence of aura does not affect outcome after TLE surgery. However, auditory and vertiginous auras are predictors of poor surgical outcome. These patients require more extensive screening for an ictal onset zone beyond standard limits of ATL before surgery.

"Time is Brain"-How early should surgery be done in drug-resistant TLE?

OBJECTIVES: To explore the effect of duration of epilepsy and delay in surgery on seizure outcome in patients operated for drug-resistant temporal lobe epilepsy (TLE). MATERIALS & METHODS: A total of 664 consecutive patients who underwent anterior temporal lobectomy (ATL) for TLE from 1995 to 2008 formed the study cohort. We divided them into two, one as seizure-free with or without antiepileptic drugs after ATL as "good outcome" (Engel class I a) and seizures of any type, any time after surgery as "poor outcome." The probability of seizure freedom/seizure recurrence based on the duration of epilepsy was compared using Kaplan-Meier curves, univariate Cox regression survival analysis, and multivariate Cox proportional hazards regression model. RESULTS: A total of 136 children and 528 adults underwent ATL during this period. Mean duration of epilepsy pre-ATL was 17.1 + 9.4 years. At mean follow-up of 8.5 years, 331 patients (49.8%) had good outcome and 333 (50.2%) had poor outcome. The hazard of seizure recurrence linearly increased with duration of epilepsy pre-ATL, from 1.5 (duration of epilepsy, 5-10 years) to 1.9 (duration of epilepsy, 10-15 years) to 2 (duration of epilepsy over 15 years). In addition, encephalitis as antecedent, bilateral mesial temporal sclerosis in MRI, normal histopathology, and spikes in postoperative EEG at 3 months and 1 year predicted poor seizure outcome. CONCLUSIONS: "Epilepsy duration" independently predicted both short- and long-term seizure outcome after surgery in TLE. "Lost years" translate into poor seizure outcome after ATL. Therefore, all cases of drug-resistant TLE should be referred to a surgical center at the earliest.

Predictors of outcome after surgery in 134 children with drug-resistant TLE.

PURPOSE: To determine the factors predicting seizure outcome in children who underwent anterior temporal lobectomy (ATL) for drug-resistant temporal lobe epilepsy (TLE). METHODS: Among the 664 patients who underwent ATL for TLE from 1995 to 2008, there were 134 children. "Excellent" outcome in them was defined as seizure freedom during the entire follow-up (Engel class I a); "good outcome", if in remission last 2 years, and the rest as "unfavorable outcome". To identify the potential predictors of seizure recurrence, the attributes of recurred and non-recurred groups was compared by univariate and multivariate analysis. RESULTS: Of the 134 children, at a mean post-operative follow-up of 8.1 years, there were 82 (61.1%) with excellent outcome and 26 (19.4%) with good outcome (category-1; seizure-free, 80.6%). Drugs could be successfully withdrawn in 69 (63.9%). 26 patients (22.4%) had seizure recurrence, (category-2, treatment/surgical failure).Whereas, of the 530 adults who underwent ATL during this time period, only 46.8% never had seizures (category-1) and 53.2% had seizure recurrence in the form of acute post-operative seizures or auras or habitual or non-habitual seizures (category-2).The hazard of seizure recurrence in children increased with positive family history of seizures and /or epilepsy, normal neuroimaging, spikes in post-operative EEG at 3-months, normal histopathology and duration of epilepsy ≥5 years. CONCLUSIONS: The information on the predictive factors causing seizure recurrence/freedom in children with drug-resistant TLE is important in surgical selection. Surgery for pediatric TLE results in favorable outcome, hence one should subject children for resective surgery at the earliest especially in lesions known to cause refractoriness.

Early resective surgery causes favorable seizure outcome in malformations of cortical development.

PURPOSE: We analyzed consecutive cases of a large cohort of the spectrum of malformations of cortical development (MCDs) including focal cortical dysplasias (FCDs) who underwent presurgical evaluation through our epilepsy program from January 2000-December 2010. We analyzed factors predicting surgical candidacy, predictors of seizure outcome and reasons for deferring surgery. METHODS: 148 patients with MCD underwent detailed presurgical evaluation and 69 were operated. MCD was diagnosed based on characteristic findings in MRI and re-confirmation by histopathology in operated patients. Post-operative seizure outcome of non-operated and operated patients were assessed every 3 and 12 months and yearly intervals. Multivariate analysis and backward step-wise logistic regression analyzed factors predicting seizure outcome. Kaplan-Meier analysis predicted seizure-free survival rates. RESULTS: 66.67% patients were seizure-free and aura-free at last follow-up. On multivariate logistic regression, the predictors of seizure freedom in operated MCDs were completeness of resection (odds ratio 8.2; 95% CI 1.43-64.96, p=0.01), shorter duration of epilepsy (odds ratio 1.19, 95% CI 1.02-1.39, p=0.02), and absence of spikes in post-operative EEG at one year (odds ratio 4.2; 95% CI 2.52-16.6; p<0.002). In FCD sub-group, shorter duration of epilepsy (11.1 versus 16.1 years, p=0.03), absence of secondary generalized seizures (p=0.05), absence of spikes in post-operative EEG on seventh day (p=0.009) and one year (p=0.002) were associated with favorable seizure outcome. CONCLUSION: Majority of patients with MCD and refractory epilepsy when operated early remains seizure-free. Shorter duration of epilepsy is the single most important pre-operative variable and absence of spikes in post-operative EEG, predicts a long-term favorable seizure outcome.

Surgery for "Long-term epilepsy associated tumors (LEATs)": Seizure outcome and its predictors.

OBJECTIVES: "Long-term epilepsy associated tumors (LEATs)" by definition are tumors primarily causing drug-resistant seizures for two years or more. They include low-grade glial and glioneuronal tumors with normal life expectancy. We studied a large cohort of patients with LEATs who underwent surgery through our epilepsy program. PATIENTS & METHODS: From 1998-2011, 105 patients with LEATs underwent surgery in our center. We utilized their data archived in a prospective registry to evaluate their electro-clinical-imaging characteristics affecting the long-term seizure outcome. RESULTS: Of 105 patients (age 3-50 years), mean age at surgery was 20 years and mean pre-surgical duration of epilepsy was 10.9 years. 66 (62.8%) had secondary generalized seizures. 82 had temporal tumors, 23 had extra temporal (13 frontal, 3 parietal, 2 occipital and 5 multilobar lesions) and four had associated hippocampal sclerosis. The interictal discharges and ictal onset were concordant to the lesion in 82 (78%) and 98 (93%) patients respectively. Lesionectomy and/or adjoining corticectomy or temporal lobectomy was done. Ganglioglioma was the most dominant pathological substrate in 61 (58%). During a mean follow-up of 7.5 years (range 3-16 years), 78/105 (74.2%) were seizure-free and 45 (57.4%) were totally off drugs. Secondary generalized seizures (p-0.02), temporal location of tumor (p-0.008) and spikes in third month post-operative EEG (p-0.03) caused unfavorable seizure outcome. A pre-surgical duration of epilepsy of more than 6.6 years caused less than optimal surgical outcome CONCLUSIONS: Early surgery should be considered a priority in LEATs. Presence of secondary generalized seizures is the single most important predictor of a poor seizure outcome.

Does F-18 FDG-PET substantially alter the surgical decision-making in drug-resistant partial epilepsy?

OBJECTIVE: There is a dearth of information on the critical utility of positron emission tomography (PET) in choosing candidates for epilepsy surgery especially in resource-poor countries where it is not freely available. This study aimed to critically analyze the utility of FDG-PET in the presurgical evaluation and surgical selection of patients with DRE based on the results obtained through its use in our comprehensive epilepsy program. METHODS: From 2008 to 2012, 117 patients with drug-resistant epilepsy underwent F-18 fluoro-deoxy-glucose (FDG) PET in our center. We utilized their data to audit the utility of PET in choosing/deferring patients for surgery. RESULTS: Of the 117 patients (age: 5-42years) who underwent F-18 FDG-PET, 64 had normal MRI, and 53 had lesions. Electroclinical data favored temporal ictal onset in 48 (41%), extratemporal in 60 (51.3%), and uncertain lobar localization in 9 (7.7%). The topography of PET hypometabolism was localizing in 53 (45.3%), lateralizing in 12 (10.3%), and 52 (44.4%) had either normal or discordant results. In the nonlesional group, focal hypometabolism was concordant to the area of ictal onset in 27 (41.5%) versus 38 (58.5%) in the lesional group (p=0.002). Greater concordance was noted in temporal lobe epilepsy (TLE) (78.0%) as compared to extratemporal epilepsy (ETPE) (28.6%) (p<0.001). Positron emission tomography was more concordant in patients with mesial temporal sclerosis than in those with other lesions (82.8% versus 50%) (p=0.033). Positron emission tomography helped in surgical decision-making in 68.8% of TLE and 23.3% of ETPE cases. Overall, 37 patients (31.6%) were directly selected for resective surgery based on PET results. CONCLUSIONS: Positron emission tomography, when utilized judiciously, remained an ancillary tool in the surgical selection of one-third of patients with drug-resistant partial epilepsy, although its utility as an independent tool is not very promising.

Diffusion tensor imaging tractography of Meyer's loop in planning resective surgery for drug-resistant temporal lobe epilepsy.

PURPOSE: Whether Meyer's loop (ML) tracking using diffusion tensor imaging tractography (DTIT) can be utilized to avoid post-operative visual field deficits (VFD) after anterior temporal lobectomy (ATL) for drug-resistant temporal lobe epilepsy (TLE) using a large cohort of controls and patients. Also, we wanted to create a normative atlas of ML in normal population. METHODS: DTIT was used to study ML in 75 healthy subjects and 25 patients with and without VFD following ATL. 1.5T MRI echo-planar DTI sequences with DTI data were processed in Nordic ICE using a probabilistic method; a multiple region of interest technique was used for reconstruction of optic radiation trajectory. Visual fields were assessed in patients pre- and post-operatively. RESULTS: Results of ANOVA showed that the left ML-TP distance was less than right across all groups (p = 0.01). The average distance of ML from left temporal pole was 37.44 ± 4.7 mm (range: 32.2-46.6 mm) and from right temporal pole 39.08 ± 4.9 mm (range: 34.3-49.7 mm). Average distance of left and right temporal pole to tip of temporal horn was 28.32 ± 2.03 mm (range: 26.4-32.8 mm) and was 28.92 ± 2.09 mm, respectively (range: 25.9-33.3 mm). If the anterior limit of the Meyer's loop was ≤38 mm on the right and ≤35 mm on the left from the temporal pole, they are at a greater risk of developing VFDs. CONCLUSIONS: DTIT is a novel technique to delineate ML and plays an important role in planning surgical resection in TLE to predict post-operative visual performance and disability.