Authors :
Presenting Author: Hui Ming Khoo, MD, PhD – Osaka University, Graduate School of Medicine
Takuto Emura, MD – Department of Neurosurgery – Osaka University, Graduate School of Medicine; Yuto Satake, MD, PhD – Department of Psychiatry – Osaka University, Graduate School of Medicine; Takamitsu Iwata, MD – Department of Neurosurgery – Osaka University, Graduate School of Medicine; Yuya Fujita, MD, PhD – Department of Neurosurgery – Osaka University, Graduate School of Medicine; Yuki Kimoto, MD – Department of Neurosurgery – Osaka University, Graduate School of Medicine; Takahiro Matsuhashi, MD – Department of Neurosurgery – Osaka University, Graduate School of Medicine; Shimpei Miura, MD – Department of Neurosurgery – Osaka University, Graduate School of Medicine; Shunsuke Sato, MD, PhD – Department of Psychiatry – Osaka University, Graduate School of Medicine; Takashi Suehiro, MD, PhD – Department of Psychiatry – Osaka University, Graduate School of Medicine; Hideki Kanemoto, MD, PhD – Department of Psychiatry – Osaka University, Graduate School of Medicine; Kenji Yoshiyama, MD, PhD – Department of Psychiatry – Osaka University, Graduate School of Medicine; Takufumi Yanagisawa, MD, PhD – Department of Neurosurgery – Osaka University, Graduate School of Medicine; Koichi Hosomi, MD, PhD – Department of Neurosurgery – Osaka University, Graduate School of Medicine; Naoki Tani, MD, PhD – Department of Neurosurgery – Osaka University, Graduate School of Medicine; Satoru Oshino, MD, PhD – Department of Neurosurgery – Osaka University, Graduate School of Medicine; Manabu Ikeda, MD, PhD – Department of Psychiatry – Osaka University, Graduate School of Medicine; Haruhiko Kishima, MD, PhD – Department of Neurosurgery – Osaka University, Graduate School of Medicine
Rationale:
Temporal lobe resection offers the most effective treatment to achieve seizure freedom in drug-resistant temporal lobe epilepsy with a cost of potential verbal memory decline especially when the speech dominant hemisphere is involved. We aim to analyze the prognostic value of verbal memory functional mapping using electrical stimulation (ESM), also known as electrical Wada test, in addition to common neuropsychological and clinical variables for post-operative verbal memory changes.
Methods:
Among patients who were considered for left or speech-dominant temporal lobe resection from 2004 to 2023, twenty nine who underwent verbal memory ESM were identified. Twelve were excluded for no WMS-R or Wada test data. Patients were presented with one to three words while a 50-Hz bipolar stimulation that lasted for five to six seconds was administered via depth electrodes implanted in the hippocampus or parahippocampal gyrus. Those who failed to recall or recognize any presented word following a one minute distraction was considered at risk of verbal memory decline. Postoperative verbal memory change (z-Score) was derived from the difference between pre and postoperative verbal memory sub-score of WMS-R. First, a stepwise regression was used to select relevant predictors among age at surgery, epilepsy duration, preoperative verbal memory performance, hippocampus hypometabolism, hippocampus resection, ESM finding, and Wada test verbal memory lateralization. Then, a hierarchical linear regression analysis was set up to examine the prognostic relevance of ESM findings when added on to other predictors that are obtainable prior to ESM.
Results:
Relevant predictors were hippocampus resection, ESM finding, and Wada test verbal memory lateralization. In the hierarchical linear regression, ESM findings had an additional prognostic value for verbal memory change (F3,13 = 5.91, p = 0.03) when added to the combination of hippocampus resection and Wada test lateralization (F2,14 = 6.20, p = 0.03). The variance explained (adjusted R2) and the estimated AUC increased by adding ESM finding to the model. Although ESM finding remained its additional prognostic value without Wada test lateralization (F2,14 = 7.56, p = 0.02), the estimated AUC was smaller, which suggests that ESM finding is likely dependent on Wada test lateralization. For instance, no memory decline was seen in all seven patients considered to be at low risk with Wada test (symmetrical verbal memory on both hemispheres) regardless of ESM finding. Conversely, among ten patients considered to be at high risk with Wada test, ESM was able to correctly identify two with a false positive Wada test while being confirmative in six.
Conclusions:
Electrical Wada test contributes to predicting postoperative verbal memory decline following temporal lobe resection especially in patients considered to be at high risk with conventional Wada test.
Funding:
Grant-in-Aid for Scientific Research (No. 20K09368) from the Ministry of Education, Culture, Sports, Science and Technology of Japan