The Relationship Between Quantitative T2 Relaxometry and Memory in Nonlesional Temporal Lobe Epilepsy.
Wendel JD, Trenerry MR, Xu YC, Sencakova D, Cascino GD, Britton JW, Lagerlund TD, Shin C, So EL, Sharbrough FW, Jack CR
Epilepsia 2001; 42:863–868
PURPOSE: We investigated the relationship between preoperative quantitative magnetic resonance imaging (MRI) T2 relaxometry and volumetry of the hippocampi and pre- and postoperative verbal memory in temporal lobectomy patients who had nonlesional temporal lobe epilepsy.
METHODS: Pre- and postoperative memory data based on the Logical Memory (LM) subtest of the Wechsler Memory Scale-Revised (WMS-R) and the 30-min delayed recall trial of the Rey Auditory Verbal Learning Test (AVLT) were obtained from 26 left and 15 right temporal lobectomy patients. Coronal MRI T2 maps were generated for these 41 temporal lobectomy patients as well as 61 control patients. Hippocampal T2 relaxation times and hippocampal volumes, converted to z scores using control group data, were correlated with neuropsychological performance in the patients.
RESULTS: In left temporal lobe-onset patients, high T2 in the left hippocampal body predicted higher LM performance after surgery. Asymmetrically high T2 in the left hippocampal body (i.e., the right-minus-left difference), compared with the right hippocampal body, also predicted higher LM performance after surgery. In right temporal lobe-onset patients, high T2 in the left hippocampal body predicted relatively lower AVLT performance after surgery. Multiple regression analysis in left temporal-onset patients revealed that high T2 in the left hippocampal body together with higher preoperative LM performance predict higher postoperative LM performance.
CONCLUSIONS: Our findings suggest that elevated (i.e., abnormal) hippocampal T2 signal is associated with memory ability (or hippocampal functional capacity) independent of MRI-determined hippocampal atrophy. Therefore, our findings support the use of quantitative T2 relaxometry as an independent predictor of verbal memory outcome in both left and right TLE patients who are candidates for temporal lobectomy.
COMMENTARY
It is widely appreciated that memory decline represents the primary neuropsychological morbidity following anterior temporal lobectomy (ATL). Postoperative memory difficulties are observed in some, but certainly not all, patients and are more commonly seen after left than right ATL. Considerable research has been devoted to identifying the risk factors for postoperative memory decline. The risk of postoperative verbal memory decline following left ATL has been associated with the absence of histopathologically confirmed left hippocampal sclerosis, a lack of preoperative magnetic resonance imaging volume reduction in the to-be-resected left hippocampus, intact memory performance on the Wada Test, intact verbal memory performance on a preoperative neuropsychological assessment, and a lack of focal left temporal lobe/hippocampal positron emission tomography with flurodeoxyglucose (FDG-PET) hypometabolism. These and other findings all suggest that an absence of structural and functional compromise of the to-be-resected left mesial temporal region is associated with an increased risk of postoperative verbal memory decline following left ATL. The picture is not as clear regarding memory decline following right ATL.
In this context, the authors note that quantitative T2 abnormalities have been associated with hippocampal cell loss, and given that relationship, an examination of quantitative T2 abnormalities might also relate to the risk of postoperative memory decline and might therefore be worth examining. This is especially the case in that the relationship between quantitative T2 abnormalities and memory performance, particularly preoperative to postoperative memory decline, has not been fully characterized. Twenty-six left and 18 right temporal lobe epilepsy patients underwent careful preoperative workup, including magnetic resonance imaging morphometry and T2 imaging of the head and body of the hippocampus, as well as a preoperative and postoperative verbal memory assessment. Among left ATL patients, multiple regression analysis showed verbal memory decline to be associated with better preoperative performance on neuropsychological memory tests and less left hippocampal body T2 signal, both of which suggest a more intact left hippocampus, consistent with the literature reviewed here. For right ATL patients, multiple regression analysis showed verbal memory decline to be associated with left hippocampal body volume reduction and increased T2 signal, suggesting that greater pathology in the contralateral left hippocampus is associated with a risk of postoperative verbal memory decline.
In summary, for left ATL candidates, an absence of ipsilateral left hippocampal pathology (i.e., less hippocampal T2 signal, better preoperative verbal memory performance) was associated with an increased risk of postoperative verbal memory decline. For right ATL candidates, the presence of contralateral left hippocampal pathology (i.e., decreased left hippocampal volume and increased T2 signal) was associated with an increased risk of verbal memory decline. Quantitative T2 relaxometry appears to be a useful predictor of verbal memory decline for both left and right ATL candidates.