Search

Background and Objectives: A new frontier in diagnostic radiology is the inclusion of machine-assisted support tools that facilitate the identification of subtle lesions often not visible to the human eye. Structural neuroimaging plays an essential role in the identification of lesions in patients with epilepsy, which often coincide with the seizure focus. In this study, we explored the potential for a convolutional neural network (CNN) to determine lateralization of seizure onset in patients with epilepsy using T1-weighted structural MRI scans as input., Methods: Using a dataset of 359 patients with temporal lobe epilepsy (TLE) from 7 surgical centers, we tested whether a CNN based on T1-weighted images could classify seizure laterality concordant with clinical team consensus. This CNN was compared with a randomized model (comparison with chance) and a hippocampal volume logistic regression (comparison with current clinically available measures). Furthermore, we leveraged a CNN feature visualization technique to identify regions used to classify patients., Results: Across 100 runs, the CNN model was concordant with clinician lateralization on average 78% (SD = 5.1%) of runs with the best-performing model achieving 89% concordance. The CNN outperformed the randomized model (average concordance of 51.7%) on 100% of runs with an average improvement of 26.2% and outperformed the hippocampal volume model (average concordance of 71.7%) on 85% of runs with an average improvement of 6.25%. Feature visualization maps revealed that in addition to the medial temporal lobe, regions in the lateral temporal lobe, cingulate, and precentral gyrus aided in classification., Discussion: These extratemporal lobe features underscore the importance of whole-brain models to highlight areas worthy of clinician scrutiny during temporal lobe epilepsy lateralization. This proof-of-concept study illustrates that a CNN applied to structural MRI data can visually aid clinician-led localization of epileptogenic zone and identify extrahippocampal regions that may require additional radiologic attention., Classification of Evidence: This study provides Class II evidence that in patients with drug-resistant unilateral temporal lobe epilepsy, a convolutional neural network algorithm derived from T1-weighted MRI can correctly classify seizure laterality., (© 2023 American Academy of Neurology.)

Background and Objectives: Temporal lobe epilepsy (TLE) is the most common adult form of epilepsy and is associated with a high risk of cognitive deficits and depressed mood. However, little is known about the role of environmental factors on cognition and mood in TLE. This cross-sectional study examined the relationship between neighborhood deprivation and neuropsychological function in adults with TLE., Methods: Neuropsychological data were obtained from a clinical registry of patients with TLE and included measures of intelligence, attention, processing speed, language, executive function, visuospatial skills, verbal/visual memory, depression, and anxiety. Home addresses were used to calculate the Area Deprivation Index (ADI) for each individual, which were separated into quintiles (i.e., quintile 1 = least disadvantaged and quintile 5 = most disadvantaged). Kruskal-Wallis tests compared quintile groups on cognitive domain scores and mood and anxiety scores. Multivariable regression models, with and without ADI, were estimated for overall cognitive phenotype and for mood and anxiety scores., Results: A total of 800 patients (median age 38 years; 58% female) met all inclusion criteria. Effects of disadvantage (increasing ADI) were observed across nearly all measured cognitive domains and with significant increases in symptoms of depression and anxiety. Furthermore, patients in more disadvantaged ADI quintiles had increased odds of a worse cognitive phenotype ( p = 0.013). Patients who self-identified as members of minoritized groups were overrepresented in the most disadvantaged ADI quintiles and were 2.91 (95% CI 1.87-4.54) times more likely to be in a severe cognitive phenotype than non-Hispanic White individuals ( p < 0.001). However, accounting for ADI attenuated this relationship, suggesting neighborhood deprivation may account for some of the relationship between race/ethnicity and cognitive phenotype (ADI-adjusted proportional odds ratio 1.82, 95% CI 1.37-2.42)., Discussion: These findings highlight the importance of environmental factors and regional characteristics in neuropsychological studies of epilepsy. There are many potential mechanisms by which neighborhood disadvantage can adversely affect cognition (e.g., fewer educational opportunities, limited access to health care, food insecurity/poor nutrition, and greater medical comorbidities). Future research will seek to investigate these potential mechanisms and determine whether structural and functional alterations in the brain moderate the relationship between ADI and cognition., (© 2023 American Academy of Neurology.)

Background and Objectives: The aim of this study was to characterize short-term outcomes in episodic memory, as assessed by the Children's Memory Scale (CMS), after temporal lobe resection in children with epilepsy using empirical methods for assessing cognitive change (i.e., reliable change indices [RCI] and standardized regression-based change scores [SRB]) and develop and internally validate clinically applicable models to predict postoperative memory decline., Methods: This retrospective cohort study included children aged 6-16 years who underwent resective epilepsy surgery that included the temporal lobe (temporal only: "temporal" and multilobar: "temporal plus") and who completed preoperative and postoperative neuropsychological assessments including the CMS. Change scores on the CMS delayed memory subtests (Faces, Stories, and Word Pairs) were classified as decline, no change, or improvement using epilepsy-specific RCI and SRB. Logistic regression models for predicting postoperative memory decline were developed and internally validated with bootstrapping., Results: Of the 126 children included, most of them demonstrated either no significant change (54%-69%) or improvement (8%-14%) in memory performance using RCI on individual measures at a median of 7 months after surgery. A subset of children (23%-33%) showed postoperative declines. Change distributions obtained using RCI and SRB were not statistically significantly different from each other. Preoperative memory test score, surgery side, surgery extent, and preoperative full-scale IQ were predictors of memory decline. Prediction models for memory decline included subsets of these variables with bias-corrected concordance statistics ranging from 0.70 to 0.75. The models were well calibrated although slightly overestimated the probability of verbal memory decline in high-risk patients., Discussion: This study used empiric methodology to characterize memory outcome in children after temporal lobe resection. Provided online calculator and nomograms may be used by clinicians to estimate the risk of postoperative memory decline for individual patients before surgery., (© 2023 American Academy of Neurology.)

Background and Objectives: Naming decline after left temporal lobe epilepsy (TLE) surgery is common and difficult to predict. Preoperative language fMRI may predict naming decline, but this application is still lacking evidence. We performed a large multicenter cohort study of the effectiveness of fMRI in predicting naming deficits after left TLE surgery., Methods: At 10 US epilepsy centers, 81 patients with left TLE were prospectively recruited and given the Boston Naming Test (BNT) before and ≈7 months after anterior temporal lobectomy. An fMRI language laterality index (LI) was measured with an auditory semantic decision-tone decision task contrast. Correlations and a multiple regression model were built with a priori chosen predictors., Results: Naming decline occurred in 56% of patients and correlated with fMRI LI ( r = -0.41, p < 0.001), age at epilepsy onset ( r = -0.30, p = 0.006), age at surgery ( r = -0.23, p = 0.039), and years of education ( r = 0.24, p = 0.032). Preoperative BNT score and duration of epilepsy were not correlated with naming decline. The regression model explained 31% of the variance, with fMRI contributing 14%, with a 96% sensitivity and 44% specificity for predicting meaningful naming decline. Cross-validation resulted in an average prediction error of 6 points., Discussion: An fMRI-based regression model predicted naming outcome after left TLE surgery in a large, prospective multicenter sample, with fMRI as the strongest predictor. These results provide evidence supporting the use of preoperative language fMRI to predict language outcome in patients undergoing left TLE surgery., Classification of Evidence: This study provides Class I evidence that fMRI language lateralization can help in predicting naming decline after left TLE surgery., (© 2022 American Academy of Neurology.)

Objective: To develop and externally validate models to predict the probability of postoperative verbal memory decline in adults after temporal lobe resection (TLR) for epilepsy using easily accessible preoperative clinical predictors., Methods: Multivariable models were developed to predict delayed verbal memory outcome on 3 commonly used measures: Rey Auditory Verbal Learning Test (RAVLT) and Logical Memory (LM) and Verbal Paired Associates (VPA) subtests from Wechsler Memory Scale-Third Edition. With the use of the Harrell step-down procedure for variable selection, models were developed in 359 adults who underwent TLR at the Cleveland Clinic and validated in 290 adults at 1 of 5 epilepsy surgery centers in the United States or Canada., Results: Twenty-nine percent of the development cohort and 26% of the validation cohort demonstrated significant decline on at least 1 verbal memory measure. Initial models had good to excellent predictive accuracy (calibration [c] statistic range 0.77-0.80) in identifying patients with memory decline; however, models slightly underestimated decline in the validation cohort. Model coefficients were updated with data from both cohorts to improve stability. The model for RAVLT included surgery side, baseline memory score, and hippocampal resection. The models for LM and VPA included surgery side, baseline score, and education. Updated model performance was good to excellent (RAVLT c = 0.81, LM c = 0.76, VPA c = 0.78). Model calibration was very good, indicating no systematic overestimation or underestimation of risk., Conclusions: Nomograms are provided in 2 easy-to-use formats to assist clinicians in estimating the probability of verbal memory decline in adults considering TLR for treatment of epilepsy., Classification of Evidence: This study provides Class II evidence that multivariable prediction models accurately predict verbal memory decline after TLR for epilepsy in adults., (© 2021 American Academy of Neurology.)

Objective: With this prospective, observational study, we aimed to determine whether noninvasive language tasks, developed specifically for children, could reliably identify the hemisphere of seizure onset in pediatric epilepsy., Methods: Seventy-eight children with unilateral epilepsy (44 left), aged 6-15 years (mean age = 11.8, SD = 2.6), completed the Children's Auditory Naming and Visual Naming Tests, the Boston Naming Test, and other verbal and nonverbal tasks. Multivariate analysis of variance was used to compare test performance between left and right hemisphere epilepsy groups, and χ 2 analyses and odds ratios were used to examine classification of left vs right hemisphere epilepsy for individual patients based on test performance., Results: Group comparisons revealed poorer auditory naming in children with left hemisphere epilepsy ( p = 0.02), yet no significant differences on measures of visual naming, general intelligence, or other cognitive functions. Moreover, χ 2 analyses using auditory naming cutoff scores to define intact vs impaired performance correctly classified seizure laterality in a significant proportion of children ( p = 0.004). The odds of left hemisphere epilepsy were 4.2 times higher (95% confidence interval 1.4-11.7) than the odds of right hemisphere epilepsy with poor auditory naming performance. In the subset of patients with temporal lobe epilepsy (TLE), the odds of left TLE were 11.3 times higher (95% confidence interval 2.00-63.17) than the odds of right TLE with poor auditory naming performance., Conclusion: Contrary to previous findings, naming performance can lateralize hemisphere of seizure onset in children with epilepsy, thereby assisting in the preoperative workup for pediatric epilepsy surgery., (© 2018 American Academy of Neurology.)

Objective: To develop and externally validate models to predict the probability of postoperative naming decline in adults following temporal lobe epilepsy surgery using easily accessible preoperative clinical predictors., Methods: In this retrospective, prediction model development study, multivariable models were developed in a cohort of 719 patients who underwent temporal lobe epilepsy surgery at Cleveland Clinic and externally validated in a cohort of 138 patients who underwent temporal lobe surgery at one of 3 epilepsy surgery centers in the United States (Columbia University Medical Center, Emory University School of Medicine, University of Washington School of Medicine)., Results: The development cohort was 54% female with an average age at surgery of 36 years (SD 12). Twenty-six percent of this cohort experienced clinically relevant postoperative naming decline. The model included 5 variables: side of surgery, age at epilepsy onset, age at surgery, sex, and education. When applied to the external validation cohort, the model performed very well, with excellent calibration and a c statistic (reflecting discriminatory ability) of 0.81. A second model predicting moderate to severe postoperative naming decline included 3 variables: side of surgery, age at epilepsy onset, and preoperative naming score. This model generated a c statistic of 0.84 in the external validation cohort and showed good calibration., Conclusion: Externally validated nomograms are provided in 2 easy-to-use formats (paper version and online calculator) clinicians can use to estimate the probability of naming decline in patients considering epilepsy surgery for treatment of pharmacoresistant temporal lobe epilepsy., (© 2018 American Academy of Neurology.)

Objective: This retrospective cohort study characterized cognitive and motor outcomes in a large sample of adults who underwent frontal lobe resections for treatment of pharmacoresistant epilepsy., Methods: Ninety patients who underwent unilateral frontal lobe resection for epilepsy (42 language-dominant hemisphere/48 nondominant hemisphere) between 1989 and 2014 completed comprehensive preoperative and postoperative neuropsychological evaluations that included measures of verbal and nonverbal intellectual functioning, attention/working memory, processing speed, language, executive functioning, verbal and visual memory, and motor functioning. Objective methods were used to assess meaningful change across a wide range of abilities and to identify factors associated with neuropsychological decline following frontal lobectomy. Detailed postoperative neuroimaging analysis was conducted to characterize region, extent, and volume of resection., Results: Forty-eight percent of patients did not demonstrate meaningful postoperative declines in cognition and an additional 42% demonstrated decline in 1 or 2 cognitive domains. When cognitive decline was observed, it usually occurred on measures of intelligence, visuomotor processing speed, or executive functioning. Side and site of resection were unrelated to cognitive outcome, but played a role in decline of contralateral manual dexterity following supplementary motor area resection. Higher preoperative ability, older age at surgery, absence of a malformation of cortical development on MRI, and poor seizure outcome were related to cognitive decline on some measures, but had poor sensitivity in identifying at-risk patients., Conclusions: The vast majority of patients who undergo frontal lobectomy for treatment of pharmacoresistant epilepsy demonstrate good cognitive and motor outcomes., (© 2017 American Academy of Neurology.)

Objective: This retrospective, observational study examined the frequency and magnitude of change in naming ability as a function of side/site of epilepsy surgery and identified predictive factors to assist clinicians in identifying patients at low, moderate, or high risk of postoperative naming decline., Methods: A total of 875 adults with pharmacoresistant epilepsy (454 left/421 right; 763 temporal/87 frontal/25 posterior quadrant) met inclusion criteria and completed the Boston Naming Test before and after surgery. Clinically meaningful change in naming ability was assessed using reliable change indices for epilepsy. Demographic, cognitive, and seizure variables were examined to determine factors most predictive of naming decline and to develop a decision tree to assist with clinical decision-making., Results: Naming decline was rare in right-sided resections and did not exceed the level expected by chance (5% overall; 90% confidence interval [CI] ± 2%). Naming decline occurred in 41% (CI ± 5%) of patients after left temporal resection (TLR) compared to 10%-12% (CI ± 10%-19%) in other left-sided surgical groups. A sizable proportion of left TLR patients (17%; CI ± 4%) showed substantial declines in naming (>11 points). Decline following left TLR was related to later age at seizure onset, older age at surgery, and higher preoperative naming ability. These factors correctly predicted naming decline in 68% of patients and were associated with degree of decline following left TLR. A decision tree is provided to assist clinicians in identifying patients at low, moderate, or high risk for postoperative naming declines., Conclusions: In addition to discussions regarding risk for memory decline following left TLR, patients should be counseled about potential decline in word-finding ability., (© 2016 American Academy of Neurology.)

Objectives: The objective of this cohort study was to compare neuropsychological outcomes following left temporal lobe resection (TLR) in patients with epilepsy who had or had not undergone prior invasive monitoring., Methods: Data were obtained from an institutional review board-approved, neuropsychology registry for patients who underwent epilepsy surgery at Cleveland Clinic between 1997 and 2013. A total of 176 patients (45 with and 131 without invasive EEG) met inclusion criteria. Primary outcome measures were verbal memory and language scores. Other cognitive outcomes were also examined. Outcomes were assessed using difference in scores from before to after surgery and by presence/absence of clinically meaningful decline using reliable change indices (RCIs). Effect of invasive EEG on cognitive outcomes was estimated using weighting and propensity score adjustment to account for differences in baseline characteristics. Linear and logistic regression models compared surgical groups on all cognitive outcomes., Results: Patients with invasive monitoring showed greater declines in confrontation naming; however, when RCIs were used to assess clinically meaningful change, there was no significant treatment effect on naming performance. No difference in verbal memory was observed, regardless of how the outcome was measured. In secondary outcomes, patients with invasive monitoring showed greater declines in working memory, which were no longer apparent using RCIs to define change. There were no outcome differences on other cognitive measures., Conclusions: Results suggest that invasive EEG monitoring conducted prior to left TLR is not associated with greater cognitive morbidity than left TLR alone. This information is important when counseling patients regarding cognitive risks associated with this elective surgery., (© 2015 American Academy of Neurology.)

Objective: Misleading manifestations of common epilepsy syndromes might account for some epilepsy surgery failures, thus we sought to characterize patients with difficult to diagnose (atypical) mesial temporal lobe epilepsy (mTLE)., Methods: We retrospectively reviewed our surgical database over 12 years to identify patients who underwent a standard anterior temporal lobectomy after undergoing intracranial EEG (ICEEG) evaluation with a combination of depth and subdural electrodes. We carefully studied electroclinical manifestations, neuroimaging data, neuropsychological findings, and indications for ICEEG., Results: Of 835 patients who underwent anterior temporal lobectomy, 55 were investigated with ICEEG. Ten of these had atypical mTLE features and were not considered to have mTLE preoperatively. All of them had Engel class I outcome for 3 to 7 years (median 3.85). Five reported uncommon auras, and 3 had no auras. Scalp-EEG and nuclear imaging studies failed to provide adequate localization. None had MRI evidence of hippocampal sclerosis. However, ICEEG demonstrated exclusive mesial temporal seizure onset in all patients. Clues suggesting the possibility of mTLE were typical auras when present, anterior temporal epileptiform discharges or ictal patterns, small hippocampi, asymmetrical or ipsilateral temporal hypometabolism on PET, anterior temporal hyperperfusion on ictal SPECT, and asymmetry of memory scores. Histopathology revealed hippocampal sclerosis in 6 patients and gliosis in 2., Conclusions: Atypical electroclinical presentation may be deceptive in some patients with mTLE. We emphasize the importance of searching for typical mTLE features to guide ICEEG study of mesial temporal structures in such patients, who may otherwise mistakenly undergo extramesial temporal resections or be denied surgery.