Search Results

You are looking at 1 - 10 of 16 items for

  • Author or Editor: Orrin Devinsky x
  • Refine by Access: all x
Clear All Modify Search
Full access

Neural network analysis of preoperative variables and outcome in epilepsy surgery

Jeffrey E. Arle, Kenneth Perrine, Orrin Devinsky, and Werner K. Doyle

Because appropriate patient selection is essential for achieving successful outcomes after epilepsy surgery, the need for more robust methods of predicting postoperative seizure control has been created. Standard multivariate techniques have been only 75 to 80% accurate in this regard. Recent use of artificial intelligence techniques, including neural networks, for analyzing multivariate clinical data, has been successful in predicting medical outcome. The authors applied neural network techniques to 80 consecutive patients undergoing epilepsy surgery in whom demographic, seizure, operative, and clinical variables to predict postoperative seizures data were obtained.

Neural networks were able to predict postoperative seizures in up to 98% of cases. Student's t tests or chi-square analysis performed on individual variables revealed that only the preoperative medication index was significantly different (p = 0.02) between the two outcome groups. Six different combinations of input variables were used to train the networks. Neural network accuracies differed in their ability to predict seizures using all data (96%); all data minus electroencephalography concordance and operative side (93%); all data except intra- or postoperative variables such as tissue pathology (98%); all data excluding pathology, intelligence quotient (IQ) data, and Wada results (84%); only using demographics and tissue pathology (65%); and only using IQ data (63%).

Analysis of the results reveals that several networks that are trained with the usual accepted variables characterizing the typical evaluation of epilepsy patients can predict postoperative seizures with greater than 95% accuracy.

Restricted access

Preoperative predictors of anterior temporal language areas

Theodore H. Schwartz, Orrin Devinsky, Werner Doyle, and Kenneth Perrine

Object. Although it is known that 5 to 10% of patients have language areas anterior to the rolandic cortex, many surgeons still perform standard anterior temporal lobectomies for epilepsy of mesial onset and report minimal long-term dysphasia. The authors examined the importance of language mapping before anterior temporal lobectomy.

Methods. The authors mapped naming, reading, and speech arrest in a series of 67 patients via stimulation of long-term implanted subdural grids before resective epilepsy surgery and correlated the presence of language areas in the anterior temporal lobe with preoperative demographic and neuropsychometric data.

Naming (p < 0.03) and reading (p < 0.05) errors were more common than speech arrest in patients undergoing surgery in the anterior temporal lobe. In the approximate region of a standard anterior temporal lobectomy, including 2.5 cm of the superior temporal gyrus and 4.5 cm of both the middle and inferior temporal gyrus, the authors identified language areas in 14.5% of patients tested. Between 1.5 and 3.5 cm from the temporal tip, patients who had seizure onset before 6 years of age had more naming (p < 0.02) and reading (p < 0.01) areas than those in whom seizure onset occurred after age 6 years. Patients with a verbal intelligence quotient (IQ) lower than 90 had more naming (p < 0.05) and reading (p < 0.02) areas than those with an IQ higher than 90. Finally, patients who were either left handed or right hemisphere memory dominant had more naming (p < 0.05) and reading (p < 0.02) areas than right-handed patients with bilateral or left hemisphere memory lateralization. Postoperative neuropsychometric testing showed a trend toward a greater decline in naming ability in patients who were least likely to have anterior language areas, that is, those with higher verbal IQ and later seizure onset.

Conclusions. Preoperative identification of markers of left hemisphere damage, such as early seizure onset, poor verbal IQ, left handedness, and right hemisphere memory dominance should alert neurosurgeons to the possibility of encountering essential language areas in the anterior temporal lobe (1.5–3.5 cm from the temporal tip). Naming and reading tasks are required to identify these areas. Whether removal of these areas necessarily induces long-term impairment in verbal abilities is unknown; however, in patients with a low verbal IQ and early seizure onset, these areas appear to be less critical for language processing.

Restricted access

Neural network analysis of preoperative variables and outcome in epilepsy surgery

Jeffrey E. Arle, Kenneth Perrine, Orrin Devinsky, and Werner K. Doyle

Object. Because appropriate patient selection is essential for achieving successful outcomes after epilepsy surgery, the need for more robust methods of predicting postoperative seizure control has been created. Standard multivariate techniques have been only 75 to 80% accurate in this regard. Recent use of artificial intelligence techniques, including neural networks, for analyzing multivariate clinical data has been successful in predicting medical outcome.

Methods. The authors applied neural network techniques to 80 consecutive patients undergoing epilepsy surgery in whom data on demographic, seizure, operative, and clinical variables to predict postoperative seizures were collected.

Neural networks could be used to predict postoperative seizures in up to 98% of cases. Student's t-tests or chi-square analysis performed on individual variables revealed that only the preoperative medication index was significantly different (p = 0.02) between the two outcome groups. Six different combinations of input variables were used to train the networks. Neural network accuracies differed in their ability to predict seizures: using all data (96%); all data minus electroencephalography concordance and operative side (93%); all data except intra- or postoperative variables such as tissue pathological category (98%); all data excluding pathological category, intelligence quotient (IQ) data, and Wada results (84%); only demographics and tissue pathological category (65%); and only IQ data (63%).

Conclusions. Analysis of the results reveals that several networks that are trained with the usual accepted variables characterizing the typical evaluation of epilepsy patients can predict postoperative seizures with greater than 95% accuracy.

Restricted access

Coagulation abnormalities in children undergoing epilepsy surgery

Clinical article

Donato Pacione, Francine Blei, Orrin Devinsky, Howard L. Weiner, and Jonathan Roth

Object

Surgery is increasingly used to treat children with refractory epilepsy. Before surgery, the authors routinely evaluated the coagulation profile to identify coagulation abnormalities not established by personal and family history, physical examination, and routine screening tests.

Methods

Thirty-nine consecutive children undergoing testing prior to epilepsy surgery were prospectively evaluated. The authors evaluated a detailed hematological history and an elaborative hematological panel including complete blood count, hepatic panel, anticoagulant levels, coagulation profile (prothrombin time, partial thromboplastin time, international normalized ratio, fibrinogen, thrombin time, von Willebrand antigen, ristocetin cofactor, factor VIII, and individual factor assays when indicated) and platelet aggregation studies (in the presence of adenosine diphosphate, epinephrine, collagen, and ristocetin). Patient variables included tuberous sclerosis complex (TSC), age at epilepsy onset, age at surgery, seizure frequency, number and type of antiepileptic drugs, recent or present ketogenic diet, and use of selective serotonin reuptake inhibitors.

Results

Ten children (25.6%) had either coagulation or platelet function abnormalities. Abnormal coagulation was identified in 5 children, and abnormal platelet function was discovered in 6. A diagnosis of TSC was associated with a platelet function abnormality (p = 0.012), whereas children without TSC had a higher rate of coagulopathy (p = 0.041). None of the other characteristics reached statistical significance. In 2 patients (5.1%) with TSC and platelet aggregation abnormalities, the authors noted normal standard screening laboratory studies and an uneventful detailed personal and family history. One of these 2 patients developed a significant intraoperative bleeding complication.

Conclusions

A preoperative screening with standard laboratory studies and detailed history may not be adequate to fully examine underlying coagulation abnormalities in children with refractory epilepsy. Platelet aggregation studies should be considered in patients with TSC.

Restricted access

Electrocorticographic evidence of perituberal cortex epileptogenicity in tuberous sclerosis complex

Clinical article

Tracy S. Ma, Robert E. Elliott, Véronique Ruppe, Orrin Devinsky, Ruben Kuzniecky, Howard L. Weiner, and Chad Carlson

Object

Tuberous sclerosis complex (TSC) is a multisystem autosomal dominant disorder resulting in hamartomas of several organs. Cortical tubers are the most prominent brain lesions in TSC. Treatment-resistant epilepsy often develops early in life in patients with TSC and is associated with severe intellectual and behavioral impairments. Seizures may remit following epilepsy surgery in selected cases, yet it remains unclear whether the tuber or the perituberal cortex is the source of seizure onset. In this study, the authors reviewed the onset of seizures in patients in whom depth electrodes had been placed within or adjacent to cortical tubers.

Methods

After obtaining institutional review board approval, the authors retrospectively reviewed data from 12 pediatric patients with multifocal TSC and treatment-resistant epilepsy who had undergone invasive intracranial electroencephalographic monitoring. Tubers were identified on postimplantation MRI, and all depth electrodes were located. Depth electrode contacts were classified visually as either tuber/perituberal cortex or nontuber/nonperituberal cortex. Board-certified clinical neurophysiologists reviewed the seizures to identify all electrodes involved in the ictal onset.

Results

Among 309 recorded seizures, 104 unique ictal onset patterns were identified. Of the 11 patients with electrodes recording in a tuber, 9 had seizure onsets involving the tuber. Similarly, of the 9 patients with perituberal recording electrodes, 7 had perituberal ictal onsets. Overall, there was no difference in the percentage of contacts involved in seizure onset between the tuber and perituberal cortex. In a subset of 7 patients in whom at least 1 depth electrode contact was within the tuber and 1 was in the perituberal cortex, there was no difference between the percentage of tuber and perituberal onsets.

Conclusions

Findings demonstrated heterogeneity in the ictal onset patterns as well as involvement of the tuber and perituberal cortex within and between patients. Although the data are limited by the restricted region(s) sampled with intracranial electrodes, they do suggest that cortical hyperexcitability in TSC may derive from the tuber or surrounding cortex.

Full access

Epilepsy surgery and tuberous sclerosis complex: special considerations

A review

Robert J. Bollo, Stephen P. Kalhorn, Chad Carlson, Veronique Haegeli, Orrin Devinsky, and Howard L. Weiner

Epilepsy surgery for medically refractory seizures among patients with tuberous sclerosis complex (TSC) is a well-accepted treatment option. Many epilepsy centers around the world have published their experience over the past several years, supporting the idea that the best seizure control is obtained when a single tuber and associated epileptogenic zone is documented and targeted surgically. Recent advances in imaging and physiological techniques that reveal the epileptogenic zone have been used successfully in children with TSC who are being evaluated for surgery. As a result, a number of different surgical strategies have emerged, each reflecting the experience, strengths, and referral biases of the individual treating teams. Experience suggests that some patients with TSC who present with seizures that are difficult to localize and do not meet the classic selection criteria for epilepsy surgery may, nevertheless, benefit from sugery. Tuberectomy alone is often not sufficient for obtaining seizure control. Intracranial electrode recordings performed in a large number of children with TSC undergoing epilepsy surgery have raised new questions about the relationship of the cortical tuber to the epileptogenic zone in TSC. A careful assessment of the risks and benefits of any surgical strategy, compared with those associated with continued refractory epilepsy, should be considered by the treating team in conjunction with the patient's family. Epilepsy surgery has not only benefited many children with TSC, but it also facilitates the understanding of epileptogenesis in TSC.

Free access

Letter to the Editor: Anterior temporal lobectomy

Marco Giulioni, Matteo Martinoni, and Gianluca Marucci

Restricted access

Intraventricular lesions in tuberous sclerosis complex: a possible association with the caudate nucleus

Clinical article

Joel S. Katz, Sarah S. Milla, Graham C. Wiggins, Orrin Devinsky, Howard L. Weiner, and Jonathan Roth

Object

Tuberous sclerosis complex (TSC) can manifest with 3 principal intracranial pathological entities: cortical tubers, subependymal nodules (SENs), and subependymal giant cell astrocytomas (SEGAs). The authors analyzed the location and growth of intraventricular lesions in a large cohort of patients with TSC.

Methods

After institutional review board protocol approval, the authors retrospectively reviewed brain MRI scans of TSC patients for whom at least 1 electronically stored cranial MRI study was available. Collected data included location, size, and growth over time of all intraventricular lesions.

Results

The authors reviewed 560 scans in 103 patients, who harbored 496 intraventricular lesions. Of the 496 lesions, 157 lesions were located along the caudate-thalamic groove (CTG) in 88 patients. Twenty SEGAs were operated on. The remaining 339 lesions were distributed along the lateral ventricle, always in contact with the course of the caudate nucleus, and were presumed to be SENs. Twenty-two patients with more than 4 years of follow-up had 34 lesions along the CTG, of which 23 were stable in size and 11 grew. All other intraventricular lesions were stable. Seven-Tesla MRI showed the intimate association of SENs and the caudate nucleus in 1 patient.

Conclusions

Intraventricular lesions in TSC patients are located throughout the lateral ventricular wall. Their location exclusively follows the course of the caudate nucleus. Only lesions along the CTG showed the potential to grow, and these were then identified as SEGAs. The remaining lesions were SENs. Understanding why these lesions develop in relation to the caudate nucleus may offer insights into therapy.

Restricted access

The ability of high field strength 7-T magnetic resonance imaging to reveal previously uncharacterized brain lesions in patients with tuberous sclerosis complex

Clinical article

Jason R. Chalifoux, Nissa Perry, Joel S. Katz, Graham C. Wiggins, Jonathan Roth, Daniel Miles, Orrin Devinsky, Howard L. Weiner, and Sarah S. Milla

Object

Tuberous sclerosis complex (TSC) brain pathology is characterized on MRI by cortical tubers, subependymal nodules, and subependymal giant cell astrocytomas. Seizures, the prominent feature of TSC, are frequently intractable to medical therapy and, in many patients, resection of tubers results in seizure control. However, in approximately 40% of patients, resection of tubers does not control seizures. This fact, as well as evidence from invasive electrophysiological recordings and experimental animal models, suggests that in patients with TSC, there may be extratuberal epileptogenic brain that does not display any apparent abnormality on conventional MRI. The authors hypothesized that high field strength MRI might uncover lesions not seen on conventional MRI in these patients.

Methods

Institutional review board approval was obtained to scan 4 patients with TSC (ages 18–26 years) in a 7-T MR unit. Optimized 7-T sequences, including T1- and T2-weighted, FLAIR, SPACE FLAIR, T2*, and MPRAGE studies, were performed. Imaging studies were compared with identical sequences performed using a conventional 1.5-T MR scanner.

Results

In all 4 patients, there was improved visualization of the findings demonstrated on conventional imaging. Importantly, new lesions were detected in all 4 patients, which were not well visualized with conventional MRI. Newly detected lesions included microtubers, radial glial signal abnormalities, subependymal nodules arising from the caudate nucleus, and caudate nucleus lesions.

Conclusions

High field strength MRI detects previously uncharacterized lesions in patients with TSC and allows better detection and delineation of subtle abnormalities. In addition, the data demonstrate a compelling relationship between intraventricular lesions and the caudate nucleus. These data support previous electrophysiological and animal-model findings that demonstrate neurological pathology beyond the conventionally detected lesions in TSC.

Restricted access

Longitudinal quantitative analysis of the tuber-to-brain proportion in patients with tuberous sclerosis

Clinical article

David S. Hersh, Jonathan Chun, Howard L. Weiner, Steven Pulitzer, Henry Rusinek, Jonathan Roth, Orrin Devinsky, and Sarah S. Milla

Object

In patients with tuberous sclerosis complex (TSC), the tuber-to-brain proportion (TBP) is a marker of seizure severity and cognitive function. However, few studies have quantified the TBP. Furthermore, authors of these studies have measured the TBP at only a single time point, despite the fact that tuber cells were found to express proliferation markers, suggesting that they may be dynamic lesions. Authors of the present study used a semi-automated tuber segmentation program to determine whether the TBP changes over time.

Methods

Axial FLAIR MR images were retrospectively identified for patients with TSC who had undergone imaging at the authors' institution between February 1998 and June 2009. Using FireVoxel software, the TBP was measured for each patient at a minimum interval of 2 years.

Results

Twelve patients meeting the study inclusion criteria were identified. The mean TBP was 1.88% (range 0.38%–3.70%). Eight patients demonstrated minimal changes and 3 patients demonstrated small increases in TBP. The remaining patient exhibited a decrease of 1.00%, which correlated with a visible decrease in the size of 2 cerebellar lesions.

Conclusions

Semi-automated brain segmentation is a valuable tool in the longitudinal study of tubers. A subset of patients with TSC, particularly those with cerebellar lesions, may exhibit changes in the TBP over time.