July 28 2011
Mentor: Vivek Prabhakaran, M.D. Ph.D.
Background: The intracarotid amobarbital (or Wada) test is currently the gold standard for clinically lateralizing cerebral functions to the cerebral hemispheres in pre-operative evaluation. Recently, clinic use of functional magnetic resonance imaging(fMRI) is increasingly used to lateralize language function. In this study, the concordance between fMRI and the Wada test is reported in patients with medically intractable seizures. Also, the correlation of the distance between the epileptic focus to the functional activation area with the patients’ post-operative deficits in language is analyzed. Finally, the relationship between post-operative seizure improvement and non-fMRI factors, including age, gender, handedness and diagnosis, are examined.
Methods: This study includes 17 patients with preoperative fMRI and Wada test (n =17 for language tests, n=9 for memory, age range: 12-49). The fMRI images were obtained and analyzed using established fMRI paradigms for language and memory. Activation of Broca’s and Wernicke’s areas was measured three-dimensionally. A threshold-dependent lateralization index (LI) was calculated for language areas: LI = [(L-R)/(L+R)] * ½. A calculated LI > 0.25 is considered left-hemisphere dominant, 0.25 < LI < -0.25 is considered bilateral, and LI < -0.25 is considered right dominant. An experienced technologist’s interpretation of memory lateralization was used for analysis. Standard neuropsychiatry Wada test procedures were used for comparison. The shortest distance between language area to the border of surgical focus (LAD) was measured using the PACS computer software, categorized (, 10mm, 10-20mm, or > 20mm), and compared with postoperative language deficits. The effects of the non-fMRI factors on postoperative seizure improvement was individually analyzed using improvement defined by the postoperative neurologic assessment at 6 months and/or reduction in seizure medication dose at 6 months. Statistical analysis included Chi square test, Fisher’s exact test, ANOVA, and Student’s t-test.
Results: Concordance between fMRI and the Wada test is 82.4% (p = 0.88) for language dominance and 66% (p = 0.64) for memory. No correlation was found between LAD and postoperative language or memory deficit (p = 0.46 for language; p = 0.37 for memory). Females demonstrated increased postoperative seizure improvement (Fisher’s p-value = 0.009; female = 12; male = 7). Other factors are still being analyzed for effects on post-operative seizure improvement.
Conclusion: The language concordance between fMRI and Wada found in this study is similar to the data reported in several prior studies. Targets for future studies or the extension of this study include further examination of the concordance of fMRI with Wada for memory paradigms and further explore the correlation of LAD with language and memory. This study, however, clearly demonstrates fMRI as an effective modality for preoperative evaluation of epileptic patients and a predictive tool for postoperative patient morbidity.