NeuroImage 45 (2009) 286–297
Contents lists available at ScienceDirect
NeuroImage j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / y n i m g
Prediction of visual field deficits by diffusion tensor imaging in temporal lobe epilepsy surgery Xiaolei Chen a,c, Daniel Weigel a, Oliver Ganslandt a, Michael Buchfelder a, Christopher Nimsky a,b,⁎ a b c
Department of Neurosurgery, University Erlangen-Nürnberg, Schwabachanlage 6, 91054 Erlangen, Germany Department of Neurosurgery, University Marburg, Baldingerstrasse, 35033 Marburg, Germany Department of Neurosurgery, First Affiliated Hospital, Sun Yat-sen University, Zhongshan Er Road 58, 510080 Guangzhou, China
a r t i c l e
i n f o
Article history: Received 18 October 2007 Revised 19 June 2008 Accepted 27 November 2008 Available online 16 December 2008 Keywords: Diffusion tensor imaging Fiber tracking Optic radiation Visual field deficits Temporal lobectomy Epilepsy surgery
a b s t r a c t Visual field deficits due to optic radiation injury are a common complication of temporal lobectomy in epilepsy surgery. In this prospective study, diffusion tensor imaging (DTI) based fiber tracking was performed on 48 patients who had temporal lobectomy for pharmaco-resistant epilepsy. Pre- and intra-operative DTI based fiber tracking was used to visualize the optic radiation and to predict the post-operative visual field defects. The course of the optic radiation could be successfully reconstructed by DTI based fiber tracking. There was a significant correlation between the fiber tracking estimation and the outcome of visual field deficits after surgery. The Receiver Operating Characteristic (ROC) curve analysis confirmed the accuracy and validity of prediction of the post-operative visual field deficits comparing pre- and intra-operative fiber tracking results. Intra-operative visualization of the optic radiation may help in avoiding post-operative visual field deficits. © 2008 Elsevier Inc. All rights reserved.
Introduction Visual field defects (VFDs) due to optic radiation (OR) injury are among the commonest complications after anterior temporal lobectomy for temporal lobe epilepsy (Anderson et al., 1989; Falconer and Wilson, 1958; Hughes et al., 1999; Krolak-Salmon et al., 2000; Wieshmann et al., 1999). Typically, VFDs occur in the superior homonymous field contralateral to the resection and are due to disruption of the OR, especially the anterior bundle, which is also known as Meyer's loop. Meyer's loop is the most anterior portion of the optic radiation. The fibers of Meyer's loop project from the lateral geniculate body (LGB), running anteriorly across the superior aspect of the anterior tip of the lateral ventricle's temporal horn before making a sharp turn to join the dorsal bundle fibers of the optic radiation in their course towards the calcarine cortex (Choi et al., 2006; Rubino et al., 2005). Meyer's loop transmits visual information from the contralateral superior field of both eyes, and damage to its fibers is one cause of a homonymous superior quadrantanopia (Jacobson, 1997). Estimates of quadrantanopia complicating temporal lobectomy range from 50–70% (Katz et al., 1989; Marino and Rasmussen, 1968; Tecoma et al., 1993) to 90–100% (Bjork and Kugelberg, 1957; Falconer and Wilson, 1958; Hughes et al., 1999). Although many studies have been done to explore the anatomy of Meyer's loop, there still has been considerable disagreement on the ⁎ Corresponding author. Fax: +49 6421 5866415. E-mail address:
[email protected] (C. Nimsky). 1053-8119/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.neuroimage.2008.11.038
location, course, and anatomy of Meyer's loop in human beings. For example, the anterior limit of the Meyer's loop has not been well localized. It has been estimated at anywhere from 20 to 60 mm posterior of the temporal pole, with a tendency to lower estimates in more recent studies (Krolak-Salmon et al., 2000; Nilsson et al., 2004). There is also controversy on the intersubject variability, even on the intrasubject hemispheric asymmetries in Meyer's loop (Ebeling and Reulen, 1988). As a result, so far to now, the occurrence and extent of a postoperative VFD cannot be accurately predicted by conventional imaging methods or by measuring the extent of the resection. Diffusion tensor imaging (DTI) is a MRI technique that evaluates brain structure by measuring tissue water diffusion in 3-dimensional (3-D) space (Pierpaoli et al., 1996). It is based on the general principle that diffusion is directed by the anatomical microstructure, i.e. by white matter (WM) fibers (Melhem et al., 2002; Mori et al., 2002; Mori and Van Zijl, 2002; Wiegell et al., 2000). Fiber tracking (FT), also known as “tractography”, allows to non-invasively visualize the course of major white matter tracts based on DTI technique (Basser et al., 2000; Conturo et al., 1999; Lazar et al., 2006; Lazar et al., 2003; Mori et al., 1999; Mori et al., 2002). It can provide information about the course, the displacement, or interruption of white matter tracts. Fiber tracts are estimated by FT with algorithms that detect longrange patterns of continuity in the diffusion tensor field. Multiple studies have demonstrated that FT can reconstruct the major WM fiber structures in the healthy brain (Mori et al., 2002). However, to date, the role of DTI based FT in detection of the optic radiation, including Meyer's loop, has not yet been well addressed. Only a few
X. Chen et al. / NeuroImage 45 (2009) 286–297
studies with very small cohorts have demonstrated that DTI based FT can reconstruct the major fiber structures of the optic radiation (Powell et al., 2005; Yamamoto et al., 2005). On the other hand, some studies correlating clinical outcome data have recently questioned the accuracy and validity of DTI based FT depicting the major WM tracts (Kinoshita et al., 2005). In a prospective study we sought to relate the status of Meyer's loop, assessed by pre- and intra-operative DTI based FT, to the extent of VFD following anterior temporal lobectomy. By doing this, we try to determine changes in the size and course of Meyer's loop due to surgical resection, and to address first, whether DTI based FT can precisely depict Meyer's loop; second, if DTI based tractography can accurately predict the resection effects on the visual field; and third, what can be the best strategy to minimize the injury of Meyer's loop during anterior temporal lobectomy? Materials and methods Patients Our study consecutively enrolled 48 patients with pharmacoresistant temporal lobe epilepsy and they all underwent standard tailored anterior temporal lobectomy from January 2003 to May 2007. Conventional anatomical MRI and DTI data were prospectively collected. All patients had both pre- and intra-operative MR imaging and pre- and post-operative visual field examination. We excluded those patients with other ophthalmic or neurological causes of visual loss. Of the 48 patients, 25 had left- and 23 had right-sided resections. Average age of the patients in our study cohort was 35.4 years (SD 12.5, range 8–59 years), with no significant difference between right and left lobectomy groups (P = 0.14). The local ethical committee of the University Erlangen-Nuremberg approved intra-operative MRI, and signed informed consent was provided by all patients or appropriate family members. Conventional MR imaging and diffusion tensor imaging (DTI) Both pre- and intra-operative MR imaging were performed on a 1.5T Siemens Sonata scanner (Siemens Healthcare, Erlangen, Germany) with the same protocol. Details of the intra-operative MRI setting are published (Nimsky et al., 2005; Nimsky et al., 2006a,b). A T1-weighted three-dimensional (3-D) magnetization prepared rapid acquisition gradient echo (MPRAGE) sequence was measured with an echo time (TE) of 4.38 ms, repetition time (TR) of 2020 ms, matrix size of 256 × 256, field of view (FOV) of 250 × 250 mm, slice thickness of 1 mm, and slab of 16 cm. T2-weighted images (TE 98 ms, TR 6490 ms, matrix size 512 × 307, FOV 230 × 183 mm, slice thickness 3 mm) were scanned as well. Intra-operative scans were performed immediately after anterior temporal lobectomy, prior to head closure. For DTI we applied a single-shot spin-echo diffusion-weighted echo planar imaging (EPI) sequence (TE 86 ms, TR 9200 ms, matrix size 128× 128, FOV 240 mm, slice thickness 1.9 mm, bandwidth 1502 Hz/Px, using b values of 0 and 1000 s/mm2, 60 slices, no intersection gap, measurement time 5 min 31 s at 5 averages) for DTI. This sequence is based on a balanced diffusion gradient design which strongly minimizes eddy-current artifacts compared to a single-refocused design. One image without diffusion weighting (b = 0 s/mm2) and six diffusionweighted images were obtained with the diffusion-encoding gradients directed along the following axes (±1, 1, 0), (±1, 0, 1), and (0, 1, ±1). Fiber tracking (FT) For reconstruction and visualization of the fiber tracts, we used the “Fiber Tracking” module of the navigation planning software iPlan 2.5 (BrainLab, Feldkirchen, Germany). For this purpose, we implemented a tracking algorithm based on a tensor deflection algorithm
287
(Nimsky et al., 2006a,b). Fiber tracking was performed by the first author blinded to the results of the patient's ophthalmological evaluation. Before tracking is initiated, the user can adjust the FA threshold and the minimum fiber length (stop criteria). Our default FA threshold is 0.15 and the minimum fiber length is 50 mm. Tract seeding is performed by defining a rectangular volume of interest (VOI) either in the FA maps or in the co-registered standard anatomical datasets. We used a multi-VOI algorithm for the fiber tracking of the optic radiation. For the anterior bundle of the optic radiation (Meyer's Loop), VOI 1 was placed on the lateral geniculate body (LGB) on the resection side and VOI 2 was placed at the level of the lower lip of the visual occipital cortex (calcarine cortex) on the same side. We identified the LGB by selecting the axial slice at the level of the transition from the posterior limb of the internal capsule to the cerebral peduncle. At this level, the LGB is visible posterolateral to the peduncles. For the rest part of optic radiation (central and posterior bundle), VOI 1 was placed on the LGB on the resection side and VOI 2 was placed at the level of the middle and upper lip of the visual occipital cortex on the same side. Fiber tracts which passed through both VOIs were the final tracts of interest. The final result of the tracking calculation is a parametric display of fibers, which are represented as streamlines, using the standard direction color encoding: left–right oriented fibers are displayed in red, anterior–posterior in green, and craniocaudal in blue, with all other orientations represented by a mixture of these colors (Pajevic and Pierpaoli, 2000; Pierpaoli et al., 1996). After selecting the appropriate fiber bundle, a 3-D object is generated automatically by wrapping neighboring fibers with a hull. The closing lines around all fibers from all slices together result in the 3-D object. The 3-D object is generated in the 3-D space of one of the regular MRI datasets, such as the MPRAGE dataset, which has a higher resolution than the DTI data, so that in every second slice of the highresolution dataset, the wrapping contours are interpolated, resulting in a smoother surface of the 3-D object. The time required to process the DTI data and to acquire the preliminary fiber-tracking images was approximately 15 min. The distance between the anterior tip of Meyer's loop and the ipsilateral temporal pole was measured at the level of midbrain on the axial plane. It was compared with published measurements obtained by previous studies of the optic radiation. The width of Meyer's loop at the resection site was also measured, both preoperatively and intra-operatively, so that the preserve ratio of Meyer's loop (post-operative width/pre-operative width) could be calculated and recorded. The injury fraction of the Meyer's loop (1-preserve ratio) was also calculated. For the evaluation of Meyer's loop injury, a five-point injury score was defined to reflect the DTI based FT findings: 1, definitely negative injury (preserve ratio ≥80%); 2, probably negative (80%N preserve ratio≥60%); 3, indeterminate (60%N preserve ratio ≥ 40%); 4, probably positive (40%N preserve ratio≥20%); 5, definitely positive (preserve ratiob 20%). Brain shift evaluation After rigid registration of the pre-operative and intra-operative optic radiation tractography data with the same pre-operative T1 dataset, we measured the shifting distance of the fiber tracts, both vertically and horizontally at the resection site. To evaluate the maximum extent of shifting, the inner borders of the reconstructed optic radiation tracts of the pre- and intraoperative data were segmented. Then, the maximum distance between the corresponding pre- and intraoperative contours was measured. According to the direction of shifting, which was referred to the craniotomy opening, positive or negative values were assigned: horizontally, positive for a movement towards the surface (i.e., swelling), negative for inward
288
X. Chen et al. / NeuroImage 45 (2009) 286–297
movements; vertically, positive for a movement towards cranial direction, negative for a movement towards caudal direction. Visual field perimetry All patients had visual field examination both pre-operatively and post-operatively, done by an experienced ophthalmologist with a Haag Streit projective kinetic perimeter (Model 940, Haag-Streit, Bern, Switzerland). To avoid the influence from “Wallerian degeneration” (Castillo and Mukherji, 1999; Mukherjee et al., 2000; Taoka et al., 2005), all the visual field examinations were done within two weeks after the surgery. The results were recorded by Goldmann perimetry in a standardized fashion for each eye. Two examiners blinded to the results of neuroimaging findings evaluated the degree of visual field loss in terms of area. The defect area on the Goldmann perimetry was assessed in the same way as it was reported by literature (Barton et al., 2005). If the VFDs are different between the two eyes of the same subject,
the VFD level of the better eye will be taken as the final VFD level of that subject. Because both the examination and the evaluation of VFD are not strictly objective, after the two examiners had reached consensus, a five-point degree score system was used to reflect the visual field findings: 1, definitely negative VFD (VFD b 20% of a quadrant); 2, mild VFD (40% N VFD ≥ 20%); 3, medium VFD (60% N VFD ≥ 40%); 4, severe VFD (80% N VFD ≥ 60%); 5, complete VFD (VFD ≥ 80%). Statistical analysis To determine the relationship between visual field loss and the status of Meyer's loop injury, we performed a linear regression analysis of VFD against injury fraction of Meyer's loop. To evaluate the validity of DTI based FT visualizing the optic radiation for prediction of the post-operative VFD, the five-point probability scores of DTI based FT findings and Goldmann perimetry scores were analyzed by a Receiver Operating Characteristic (ROC) curve. We used
Table 1 Patients undergoing temporal lobectomy for pharmaco-resistant epilepsy (n = 48) No.
Age [years]
Gender
Resection side
Resection size [mm]
ML distance [mm]
Vertical shift [mm]
Horizontal shift [mm]
preOP ML width [mm]
intraOP ML width [mm]
Injury fraction
Injury grade
VFD grade
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48
8 37 38 12 20 30 44 31 23 18 49 31 59 33 43 47 19 39 26 40 22 23 16 55 37 29 41 41 47 39 57 32 59 29 38 23 58 39 33 51 32 30 37 31 22 45 38 50
Male Male Male Female Male Female Female Female Male Male Male Female Female Male Female Male Male Male Male Male Female Female Female Male Male Male Male Male Female Male Female Male Male Female Male Male Female Male Female Female Male Female Female Female Male Female Male Female
Right Left Right Right Left Left Left Right Left Left Right Right Left Right Left Left Right Left Right Left Right Left Right Right Right Left Left Left Left Left Left Right Right Left Left Right Right Right Left Right Right Left Left Left Left Right Right Right
36.9 41.8 44.9 25.1 45.4 39.1 43.9 42.1 41.0 39.3 47.8 32.8 21.5 44.8 31.9 40.9 20.2 39.7 44.0 46.1 34.5 27.7 26.8 37.1 42.8 43.7 40.2 30.8 41.2 49.4 39.0 49.2 43.5 41.1 43.1 38.6 38.3 42.8 20.0 48.0 21.3 37.7 31.9 32.0 34.9 48.8 31.8 32.7
29.3 36.9 38.1 33.9 23.9 26.5 26.2 29.2 42.8 26.0 33.4 36.6 26.7 35.9 26.9 35.4 35.6 35.7 39.1 29.5 33.5 34.2 27.1 51.5 34.1 42.8 27.5 35.6 31.5 29.7 39.9 33.8 27.9 28.9 32.0 32.1 31.9 31.5 23.9 28.1 30.7 31.5 26.9 35.9 29.8 29.6 31.0 20.9
−3.8 −4.0 −2.0 3.5 2.5 −3.1 −2.7 0.0 1.5 −2.3 0.0 1.0 3.0 −2.0 4.0 −1.5 −1.0 0.0 1.1 −3.4 0.0 −5.8 −3.0 0.0 7.8 −1.0 1.9 3.9 2.6 0.0 −1.4 −2.4 4.3 0.0 3.0 3.0 3.1 2.4 0.0 4.3 −1.0 6.7 −2.6 −2.6 −2.0 1.9 2.1 6.7
−8.0 −4.0 2.0 −4.3 3.5 −2.7 −3.8 −7.4 0.0 2.5 3.0 1.2 −3.0 −1.2 −2.3 −4.0 0.0 −2.0 −2.5 −2.9 −6.5 −6.3 −4.5 −4.1 −8.6 0.0 −5.7 −1.3 −4.1 −11.0 −4.8 −11.1 −8.8 0.0 −3.0 −2.6 −4.6 −4.3 0.0 0.0 −2.1 −6.3 −3.1 −5.6 4.1 −2.2 0.0 −5.1
3.0 4.8 4.3 6.0 4.3 3.5 3.1 3.9 5.8 3.0 3.4 4.1 4.4 4.7 5.0 3.5 5.1 3.9 4.0 4.8 4.0 5.0 5.0 3.2 4.9 2.0 3.5 3.2 4.2 5.8 2.4 1.9 6.2 3.7 5.1 4.6 3.4 6.3 3.3 5.0 2.9 3.9 4.3 6.7 5.5 5.9 3.0 3.9
2.5 1.1 1.7 5.8 2.0 3.4 0.0 1.1 3.0 0.0 2.8 0.0 3.8 4.7 2.6 1.0 5.0 3.9 1.8 2.0 1.9 4.8 0.8 2.8 4.8 2.0 0.0 3.1 3.1 0.0 2.4 1.8 1.0 3.7 2.5 4.1 0.0 3.0 3.3 2.6 2.9 2.9 1.4 6.7 2.5 4.4 2.9 3.0
0.17 0.77 0.60 0.03 0.53 0.03 1.00 0.72 0.48 1.00 0.18 1.00 0.14 0.00 0.48 0.71 0.02 0.00 0.55 0.58 0.52 0.04 0.84 0.12 0.02 0.00 1.00 0.03 0.26 1.00 0.00 0.05 0.84 0.00 0.51 0.11 1.00 0.53 0.00 0.48 0.00 0.26 0.67 0.00 0.55 0.25 0.03 0.23
1 4 4 1 3 1 5 4 3 5 1 5 1 1 3 4 1 1 3 3 3 1 5 1 1 1 5 1 2 5 1 1 5 1 3 1 5 3 1 3 1 2 4 1 3 2 1 2
2 3 4 1 2 1 5 3 4 5 2 5 2 1 3 4 1 1 4 4 4 1 5 1 1 2 5 1 2 5 1 1 5 3 3 1 5 4 1 2 1 2 4 1 5 2 1 2
ML: Meyer's loop VFD: visual field defect.
X. Chen et al. / NeuroImage 45 (2009) 286–297
Fig. 1. The relationship of Meyer's loop injury fraction to visual field defect (VFD) grade. Linear regression line is shown, with equations in the lower right corner of the graph. For each patient, the VFD grade is plotted against the Meyer's loop injury fraction. A VFD score of 5 indicates complete quadrantanopia. Spearman correlation analysis showed significant correlations of visual field loss (VFD grade) with optic radiation/Meyer's loop injury fraction (P b 0.001), with an r value of 0.889.
Goldmann perimetry VFD score of 5 as true positive VFD and a score of 1 as true negative VFD. The area under the curve was calculated. All statistical analysis was done with software “SPSS 13.0” (SPSS inc., Chicago, USA). A threshold of P b 0.05 was used. Results With our methods, the full courses of the optic radiations were successfully identified in all 48 subjects, with the dorsal and Meyer's loops visible; an overview of the measured data is depicted in Table 1. The extent of the resections in all the patients was comparable (mean 37.8 mm; SD 7.9 mm; range 20.0–49.4 mm). Our measure of the anterior tip of the optic radiation was 32.1 mm (range 20.9–51.5 mm; SD 5.6 mm) from the temporal pole. This agreed quite well with the reported independent fiber-dissection studies using Klingler's fiber-dissection technique (Choi et al., 2006; Rubino et al., 2005).
289
Fig. 3. Scatterplot graph depicting the extents of “brain shift”. For each patient, the horizontal and vertical extents of the shifting of optic radiation are plotted. X axis indicates the extent of horizontal shifting and Y axis indicates the extent of vertical shifting. This graph demonstrated that although all the patients have similar positioning and approach for surgery, there was no fixed pattern for “brain shift”. Neither the extent nor the direction of shifting was predictable. Tractography with a similar configuration showed different behavior.
The Spearman correlation analysis showed significant correlations of field loss (VFD grade) with optic radiation/Meyer's loop injury fraction (P b 0.001), with an r value of 0.889 (Fig. 1). For ROC curve analysis, there were 9 (18.8%) subjects that had a score of 5 for VFD by Goldmann perimetry, and 8 (16.7%) subjects had a score of 5 by DTI based FT estimation. Using Goldmann perimetry VFD Grade 5 as true positive, the ROC curve is shown in Fig. 2a. The area under the curve of DTI based FT was 0.97 [95% confidence interval (CI): 0.92–1.03] (P b 0.001). On the other hand, there were 16 (33.3%) subjects that had a score of 1 for VFD by Goldmann perimetry, and 21 (43.8%) subjects had a score of 1 by DTI based FT estimation. Using VFD Grade 1 as true negative, the ROC curve is depicted in Fig. 2b. The area under the curve of DTI based FT was 0.92 [95% CI: 0.84–1.00] (P b 0.001). For “brain shift” analysis, co-registration of preoperative and intraoperative tractographies depicted a marked shifting of the optic
Fig. 2. (a) ROC curve by using VFD grade = 5 as true positive. The area under the curve of DTI based FT was 0.97 [95% confidence interval (CI): 0.92–1.03] (P b 0.001). (b) ROC curve by using VFD grade = 1 as true negative. The area under the curve of DTI based FT was 0.92 [95% CI: 0.84–1.00] (P b 0.001). The ROC curves indicates that DTI based FT can accurately predict the visual field defect after surgery.
290
X. Chen et al. / NeuroImage 45 (2009) 286–297
radiation tracts after temporal lobectomy (Fig. 3). Optic radiation tract shifting ranged horizontally from 0 to 11.1 mm (mean 3.75 mm; SD 2.77 mm). In 35 patients (72.9%), an inward shifting and in 6 (12.5%), an outward shifting was detected. In the remaining 7 patients, no intraoperative horizontal shifting of the optic radiation could be measured. Vertically, the optic radiation tract shifting ranged from 0 to 7.8 mm (mean 2.46 mm; SD 1.83 mm). In 17 patients (35.4%), a caudal and in 23 (47.9%), a cranial shifting was detected. In the remaining 8 patients, no intraoperative vertical shifting of the optic radiation could be measured. In contrary to the expectation that shifting would be mainly influenced by gravity and patient positioning, the extent and direction of shifting was not predictable. Tractography with a similar configuration showed different behavior (Fig. 3). Illustrative cases Case 1 A 41-year-old male suffered from pharmaco-resistant temporal lobe epilepsy (patient no. 27). The diagnosis was left hippocampal sclerosis. Preoperative DTI based FT showed that the anterior tip of the left Meyer's loop was approximately 27.5 mm posterior to the left
temporal pole (Figs. 4a, d and g). After the left anterior temporal lobectomy, intraoperative MR imaging revealed that the maximum resection size is about 40.2 mm from the left temporal pole. Intraoperative DTI based FT revealed that no Meyer's loop can be traced, while the dorsal bundle of the left optic radiation still appeared to have the same size compared with the pre-operative one (Figs. 4c, f and i). Overlay of the pre-operative optic radiation tractography on the intra-operative image demonstrates that the tip of the left Meyer's loop was completely exposed in the cavity caused by the resection (Figs. 4b, e and h), which implied that most probably the surgery had caused disruption of the left Meyer's loop while the rest of the left optic radiation was spared. According to the image findings, the Meyer's loop preserve fraction is 0 and the injury score is 5, which means definitely positive injury to Meyer's loop and a complete superior quadrantanopia on the right side would be expected. Postoperative perimetry was examined 8 days after the operation. The result revealed a contralateral superior quadrantanopia. So, the VFD score was 5 and it correlates well with the prediction based on FT (Fig. 5). In this case, the patient's head was placed in the horizontal position for surgery. The inward shifting of the optic radiation after
Fig. 4. Fiber tracking in a patient with left hippocampal sclerosis and pharmaco-resistant temporal lobe epilepsy (Case 1). Note that right on images is patients' left. The color bar represents the optic radiation (OR) (red for preoperative Meyer's loop; blue for pre-operative dorsal bundle of OR; brown for intra-operative Meyer's loop; green for intra-operative dorsal bundle of OR). (a) (d) (g) Pre-operative 3-D, axial and coronal tractography of the left optic radiation overlaid on preoperative T1 anatomical dataset. (b) (e) (h) Preoperative 3-D, axial and coronal tractography of the left OR overlaid on intra-operative T1 anatomical dataset. The whole tip of Meyer's loop (red) overlies the resected anterior temporal lobe, which implies that the tract may have been completely injured by the resection. (c) (f) (i) Intra-operative 3-D, axial and coronal tractography of the left OR overlaid on intra-operative T1 anatomical dataset, showing that only the dorsal bundle of the OR can be delineated by FT. The fiber tract injury grade is 5.
X. Chen et al. / NeuroImage 45 (2009) 286–297
291
Fig. 5. Goldmann perimetry of visual fields of the same patient (Case 1) as in Fig. 4. Pre-operative left and right perimetries are displayed without visual field defect. Post-operative left and right perimetries demonstrate a complete superior homonymous quadrantanopia. The post-operative VFD score is 5.
resection amounted 5.7 mm. Simultaneously, a cranial shifting of 1.9 mm of the tract was observed (compare Figs. 10a and b). Case 2 In a 38-year-old male (patient no. 35), a left hippocampal sclerosis had resulted in refractory seizures. Tractography of the left optic radiation revealed that the distance between the anterior tip of the Meyer's loop and the temporal pole was 32 mm (Figs. 6a, d and g). After the resection, intraoperative MR imaging revealed that the maximum resection size is about 43.1 mm from the left temporal pole. Overlay of the preoperative optic radiation tractography on the intra-operative (post-resection) image demonstrated that the lateral half of the tip of the left Meyer's loop exposed in the resection cavity (Figs. 6b, e and h), which implied that probably the surgery had caused partial disruption of the left Meyer's loop while part of the Meyer's loop was spared. After the resection, tractography revealed loss of the lateral half of the anterior edge of Meyer's loop (Figs. 6c, f and i). After we overlaid both the pre-operative and intra-operative tractography in the same T1 dataset, we measured and compared the diameter of the Meyer's loop before and after the resection at the resection site. Before the resection, the diameter of Meyer's loop was 5.1 mm. And after the resection, the diameter was only 2.5 mm. Therefore, the injury fraction of the Meyer's loop was about 0.51 and the injury grade was 3, which means the patient would possibly have
a partial (50%–60%) superior quadrantanopia on the right side after the operation. Postoperative perimetry was examined 7 days after the surgery. The result revealed a contralateral superior partial quadrantanopia. The VFD score was 3 and it correlates well with the prediction based on FT (Fig. 7). This patient had almost the same positioning as the first illustrated case for surgery. The inward shifting of the optic radiation after resection amounted 3 mm. Simultaneously, a cranial shifting of 3 mm was observed (compare Figs. 10c and d). Case 3 In a 41-year-old male (patient no. 28), a small left temporal WHO Grade I ganglioglioma (diameter 0.8 cm) had resulted in medically refractory seizures. Tractography of the optic radiation revealed that the distance between the anterior tip of the Meyer's loop and the temporal pole was 35.6 mm (Figs. 8a, d and g). After the resection, intraoperative MR imaging revealed that the maximum resection size was about 30.8 mm from the left temporal pole. Overlay of the pre-operative optic radiation tractography on the intra-operative images demonstrates that the anterior extent of the Meyer's loop did not overlie in the resected area of the anterior temporal lobe (Figs. 8b, e and h). This result implied that probably the surgery had caused no damage to the left Meyer's loop. The full course
292
X. Chen et al. / NeuroImage 45 (2009) 286–297
Fig. 6. Fiber tracking in Case 2, a patient with left hippocampal sclerosis and refractory seizures. Note that right on images is patients' left. The color bar represents the optic radiation (OR) (red for preoperative Meyer's loop; blue for pre-operative dorsal bundle of OR; brown for intra-operative Meyer's loop; green for intra-operative dorsal bundle of OR). (a) (d) (g) Pre-operative 3-D, axial and coronal tractography of the left optic radiation overlaid on preoperative T1 anatomical dataset. (b) (e) (h) Pre-operative 3-D, axial and coronal tractography of the left OR overlaid on intra-operative T1 anatomical dataset. About half of lateral part of the anterior edge of Meyer's loop (red) is exposed in the resected anterior temporal lobe, which implies probably that the resection has caused partial disruption of the left Meyer's loop while part of the Meyer's loop was spared. (c) (f) (i) Intra-operative 3-D, axial and coronal tractography of the left OR overlaid on intra-operative T1 anatomical dataset, showing that the left Meyer's loop is partially disrupted. The fiber tract injury grade is 3.
of the left optic radiation was visible on intraoperative FT images, and the Meyer's loop looks almost of the same size as the preoperative one (Figs. 8c, f and i). After we overlaid both the pre-operative and intraoperative tractography in the same pre-operative T1 dataset, we measured and compared the diameter of the Meyer's loop before and after the resection at the resection site. Before the resection, the diameter of Meyer's loop was 3.2 mm, while after the resection, the diameter is 3.1 mm. Therefore, the injury fraction of the Meyer's loop is 0.03 and the injury score is 1, which means no injury to Meyer's loop. The patient was supposed to have no VFD. Postoperative perimetry was examined 7 days after the resection. The result revealed no quadrantanopia. The VFD score was 1 and it correlates well with the prediction based on FT (Fig. 9). This patient also had the same positioning for operation as the cases described above. The inward shifting of the optic radiation after resection amounted 1.3 mm. Simultaneously, a cranial shifting of 3.9 mm of the tract was observed (compare Figs. 10e and f). Discussion Determining the exact course of the optic radiation, especially of Meyer's loop in individual patients may help to improve neurosurgical
approaches and to avoid damage to the optic radiation. It may contribute not only to better post-operative life-quality in epilepsy patients, but also to the outcome in stroke patients or in other patients with lesions involving the optic radiation. In the last 50 years, knowledge about the anatomy of the optic radiation, including Meyer's loop, has been gained by the development of anterior temporal lobectomy as a treatment for medical refractory epilepsy. The size of the resection involved is tailored to available data concerning the location of seizure foci in the temporal lobe, and this variation in turn generates variability in the frequency and size of the visual field defects due to associated damage to Meyer's loop. Besides, this variation has also allowed estimating the anatomy of Meyer's loop in human beings. However, a review of the literature shows that anatomically, the course of the optic radiation, especially Meyer's loop, varies considerably (Ebeling and Reulen, 1988). Such individual anatomical variation results in a great variety of visual field defects following anterior temporal lobe resection (Falconer and Wilson, 1958; Hughes et al., 1999; Krolak-Salmon et al., 2000; Marino and Rasmussen, 1968; Van Buren and Baldwin, 1958). There also has been considerable disagreement on many other points, like the location, course, and retinotopic anatomy of the optic radiation.
X. Chen et al. / NeuroImage 45 (2009) 286–297
293
Fig. 7. Goldmann perimetry of visual fields in Case 2 (same patient as in Fig. 6). Pre-operative left and right perimetries are displayed without any visual field defect. Post-operative left and right perimetries demonstrate a partial superior homonymous VFD. The post-operative VFD score is 3.
The lack of consensus in the literature may reflect the limitations in methodology. In our study, we evaluated the functional anatomy of Meyer's loop by correlating VFDs with the actual change of Meyer's loop which was depicted by DTI based FT after resection. With the advancement of DTI based FT it is possible to achieve a more accurate assessment of the changing course, size, and configuration of optic radiation or Meyer's loop. By using DTI based FT, we are the first to present a prospective study to evaluate the feasibility, accuracy, and validity of DTI based FT for depicting the optic radiation, especially the Meyer's loop, in relation to clinical outcome. We also demonstrated how surgical resections close to these pathways, like anterior temporal lobectomy, may affect superior temporal quadrantanopia. To summarize our findings, there was a significant correlation between visual field loss and the degree of optic radiation injury depicted by DTI based FT. With Goldmann perimetry scores of 5 and 1 as true positive and true negative for VFD, we found that FT had areas under the ROC curve of 0.97 [95% CI: 0.92–1.03] and 0.92 [95% CI: 0.84–1.00], respectively. This indicated that FT was accurate in predicting VFDs after anterior temporal lobectomy for temporal lobe epilepsy. Thus, it also proved that DTI based fiber tracking can precisely depict the optic radiation, including the Meyer's loop.
The boundary of the anterior fibers of Meyer's loop and their relationship to the temporal pole remains controversial. A series of studies have reported that the location of the anterior tip of the Meyer's loop is from 30 mm to 45 mm posterior to the temporal pole (Bjork and Kugelberg, 1957; Falconer and Wilson, 1958; Marino and Rasmussen, 1968). All of these older studies used intraoperative estimates of resection size and they believed that the occurrence of VFDs correlated with the extent of the resection. An anatomical dissection of 25 brains found a mean value of 27 mm (SD 3.5) but emphasized the variability between subjects (Ebeling and Reulen, 1988). Two studies with MRI and automated perimetry moved the estimation of the anterior limit of Meyer's loop to between 20 and 30 mm (Barton et al., 2005; Krolak-Salmon et al., 2000). Recently, two anatomical studies using Klingler's fiber-dissection technique revealed that the distance from the pole of the temporal lobe to the anterior edge of the optic radiations (Meyer's loop) ranges from 28.0 to 34.0 mm (average, 31.4 mm) (Choi et al., 2006; Rubino et al., 2005), with a considerable intersubject variability. Using DTI based FT data, we found the anterior limits of Meyer's loop are between 20.9 mm and 51.5 mm (average, 32.1 mm). There is considerable inter-subject variability, which must be kept in mind for surgical planning and prognosis of potential post-operative deficits.
294
X. Chen et al. / NeuroImage 45 (2009) 286–297
Fig. 8. Fiber tracking in Case 3, a patient with left temporal WHO Grade I ganglioglioma and refractory seizures. Note that right on images is patients' left. The color bar represents the optic radiation (OR) (red for preoperative Meyer's loop; blue for pre-operative dorsal bundle of OR; brown for intra-operative Meyer's loop; green for intra-operative dorsal bundle of OR). (a) (d) (g) Pre-operative 3-D, axial and coronal tractography of the left optic radiation overlaid on preoperative T1 anatomical dataset. (b) (e) (h) Pre-operative 3-D, axial and coronal tractography of the left OR overlaid on intra-operative T1 anatomical dataset. The full course of Meyer's loop (red) doesn't overlie in the resected cavity, which implies probably that the left Meyer's loop is still intact after the resection. (c) (f) (i) Intra-operative 3-D, axial and coronal tractography of the left OR overlaid on intra-operative T1 anatomical dataset, showing that the left Meyer's loop is intact. The fiber tract injury grade is 1.
The depicted bundles of the optic radiation agreed well with the classic “textbook” anatomic topography of the visual pathway. Our findings are also consistent with the modern dissection and MRI reports (Barton et al., 2005; Choi et al., 2006; Rubino et al., 2005) and contrary to older reports which were mainly based on studies applying relatively rough intraoperative measures of resection size and field loss assessments (Babb et al., 1982; Falconer and Wilson, 1958; Tecoma et al., 1993). DTI based FT can precisely depict the optic radiation and reliably predict the VFDs after temporal lobectomy. Our results have shown a significant correlation, and we believe that three critical factors allowed us to show a good correlation between visual field loss and optic radiation injury depicted by DTI based FT. First, our optic radiation injury assessments were based on the change of the actual size of the Meyer's loop depicted by DTI based FT. This allowed a more fine-grained measure than the cruder resection size assessments on conventional MRI images or the even more rough intraoperative estimates of resection extent reported by the older literature. Actual display of the optic radiation fiber tracts allowed us to determine the most anterior boundary of Meyer's loop, which may be damaged by
the resection and cause post-operative VFD. It also allowed us to compare the change of the size of the fibers pre- and post-resection. Second, comparing the change of the size of Meyer's loop pre- and post-resection for each subject allowed calculating the individual preserve ratio for the prediction of the post-operative VFDs, thus reduced one source of intersubject variability. Furthermore, our study has revealed considerable shifting of the optic radiation during resection, without a common pattern of the shifting. So, those attempts trying to estimate the fiber tracts' status according just to the resection extent (Barton et al., 2005; Powell et al., 2005), will be influenced by the effects of brain shifting and may lead to erroneous assumptions. On the contrary, our study evaluated the injury degree of the fiber tracts according to the change of the actual status of the optic radiation depicted by comparing pre- and post-resection DTI based FT. By doing this, we also compensated for the effects of “brain shift”. The clear correlation between the changes of pre- and intraoperative DTI with postoperative visual field defects in temporal lobe resections proves that the reconstructed fiber tract data are a valuable representation of the actual course of the optic radiation. As a consequence for the neurosurgical situation it can be concluded, that a
X. Chen et al. / NeuroImage 45 (2009) 286–297
295
Fig. 9. Goldmann perimetry of visual fields in Case 3 (same patient as in Fig. 8). Pre-operative left and right perimetries are displayed without any visual field defect. Post-operative left and right perimetries also demonstrate no VFD. The post-operative VFD score is 1.
preservation of the reconstructed major white matter tracts of interest (e.g. optic radiation) should result in a better neurological outcome (e.g. less VFDs). Despite quantitative information of the anterior boundary of Meyer's loop was available from the literature, it has been very difficult to measure the optic radiation in the actual surgical field and it was also very hard to intra-operatively determine the distance needed to avoid Meyer's loop and the optic tract until now. Then, what can be the best strategy to minimize the injury of the Meyer's loop during the anterior temporal lobectomy? As we have reported, DTI based FT can be implemented into a standard neuro-navigation system, which makes it possible to visualize the course of the major white matter tracts intraoperatively (Nimsky et al., 2006a,b). Thus, the next logical step will be to implement the optic radiation depicted with DTI based FT into a neuro-navigation system, so that the course of the optic radiation can be visualized in the surgical field intraoperatively. It should allow tailoring the resection extent to the individual course. However, the extent of resection, determined by the goal to make the patient seizure free, may be in conflict with preservation of the optic radiation. At least preoperative DTI should help to better estimate the risk of visual field defects, so that the most appropriate surgical approach can be chosen to minimize neurological deficits, as well as the risk can be discussed with the patient before surgery. This is even more crucial clinically, in surgery for lesions close
to the pyramidal tract or language related major white matter tracts, to minimize the risk for a paresis or aphasia. The ultimate goal of our study is to further improve preoperative imaging, intraoperative monitoring, neurological function protection and outcome prediction not only in patients with epilepsy, but also in patients with lesions involving optic radiation. For example, DTI based FT may yield useful anatomical information for the relationship between the lesion and the eloquent tracts, even for the assessment of the degree of tract involvement (Kikuta et al., 2006). These FT images may be useful not only for the surgical planning, but also for the diagnosis of various pathologic conditions involving the optic radiation, including infarcts and hemorrhages. A limitation of our study may be that we performed the DTI measurement with only 6 gradient directions, so that, it might have been possible that a part of the optic radiation with a severely curved course (Meyer's loop) could have not be estimated well. Some investigators may recommend a better performance of the tractography in a higher number of gradient directions instead of 6 directions in describing the tracts with severely curved or tortuous projection, such as Meyer's loop in the visual pathway. Fortunately, in this study, our results indicated that our description of the optic radiation, including Meyer's loop by tractography correlates well with the VFD, which possibly means that the tractography in 6 directions
296
X. Chen et al. / NeuroImage 45 (2009) 286–297
Fig. 10. Evaluation of shifting of OR tracts in the 3 illustrated cases. Note that the color bar represents the optic radiation (OR) (red for preoperative Meyer's loop; blue for preoperative dorsal bundle of OR; orange for intra-operative Meyer's loop; green for intra-operative dorsal bundle of OR). (a) and (b) Axial and left lateral 3-D view of the tractography of the dorsal bundle of OR in case 1. Because the Meyer's loop cannot be traced by FT in this case, we have to compare the pre-operative and intra-operative dorsal bundle of OR to demonstrate the “Brain shift” effect. (c) and (d) Axial and left lateral 3-D view of the tractography of Meyer's loop in case 2. (e) and (f) Axial and left lateral 3-D view of the tractography of Meyer's loop in case 3. Notice that although in these 3 cases, the operations were done in very similar patient positioning and approach, the extents of “brain shift” were different.
may be sufficient for this purpose. Besides, we are also supported by a recent study (Yamamoto et al., 2007). In that study, the investigators compared the tractography generated with 6-, 12-, 40-, and 81directional motion-probing gradients, and their findings revealed that motion-probing gradients number did not exert any significant effect on visualization of the optic radiation, and 6-directional motionprobing gradients did not significantly differ from by the tractography in 12 directions or even more. However, it still has to be confirmed by larger cohort studies. And this issue will be resolved in the future by the accumulation of data regarding correlation between the function of the tracts and the findings of tractography generated with different motion-probing gradients. In summary, our study demonstrates that DTI based FT can reliably depict the optic radiation, including Meyer's loop. It can also accurately predict the VFD, caused by anterior temporal lobectomy. Thus, we anticipate that visualizing the optic radiation by fiber tracking navigation in the surgical field may reduce post-operative VFD, like visualizing the pyramidal tract for the motor system has reduced post-operative paresis.
Acknowledgments This work was supported in part by the Deutsche Forschungsgemeinschaft (German Research Foundation) in the context of project C9 of SFB603 and by the National Natural Science Foundation (NSFC) of China (No:30800349). Xiaolei Chen was supported by the China Scholarship Council (CSC No. 2006101030). We are very grateful to the colleagues of the department of ophthalmology of the University Erlangen-Nuremberg (chairman: Prof. Dr. F. Kruse) for providing the visual field investigations. We express our special thanks to G. Ogrezeanu, U. Mezger, A. Dombay and T. Seiler (BrainLAB, Feldkirchen, Germany) for the implementation of the fiber tracking software in the navigation environment. References Anderson, D.R., Trobe, J.D., Hood, T.W., Gebarski, S.S., 1989. Optic tract injury after anterior temporal lobectomy. Ophthalmology 96 (7), 1065–1070. Babb, T.L., Wilson, C.L., Crandall, P.H., 1982. Asymmetry and ventral course of the human geniculostriate pathway as determined by hippocampal visual evoked potentials
X. Chen et al. / NeuroImage 45 (2009) 286–297 and subsequent visual field defects after temporal lobectomy. Exp. Brain Res. 47 (3), 317–328. Barton, J.J., Hefter, R., Chang, B., Schomer, D., Drislane, F., 2005. The field defects of anterior temporal lobectomy: a quantitative reassessment of Meyer's loop. Brain 128 (9), 2123–2133. Basser, P.J., Pajevic, S., Pierpaoli, C., Duda, J., Aldroubi, A., 2000. In vivo fiber tractography using DT-MRI data. Magn. Reson. Med. 44 (4), 625–632. Bjork, A., Kugelberg, E., 1957. Visual field defects after temporal lobectomy. Acta Ophthalmol. (Copenh) 35 (3), 210–216. Castillo, M., Mukherji, S.K., 1999. Early abnormalities related to postinfarction Wallerian degeneration: evaluation with MR diffusion-weighted imaging. J. Comput. Assist. Tomogr. 23 (6), 1004–1007. Choi, C., Rubino, P.A., Fernandez-Miranda, J.C., Abe, H., Rhoton Jr, A.L., 2006. Meyer's loop and the optic radiations in the transsylvian approach to the mediobasal temporal lobe. Neurosurgery 59 (4 Suppl 2), ONS228–ONS236. Conturo, T.E., Lori, N.F., Cull, T.S., Akbudak, E., Snyder, A.Z., Shimony, J.S., McKinstry, R.C., Burton, H., Raichle, M.E., 1999. Tracking neuronal fiber pathways in the living human brain. Proc. Natl. Acad. Sci. U.S.A. 96 (18), 10422–10427. Ebeling, U., Reulen, H.J., 1988. Neurosurgical topography of the optic radiation in the temporal lobe. Acta. Neurochir. (Wien) 92 (1–4), 29–36. Falconer, M.A., Wilson, J.L., 1958. Visual field changes following anterior temporal lobectomy: their significance in relation to Meyer's loop of the optic radiation. Brain 81 (1), 1–14. Hughes, T.S., Abou-Khalil, B., Lavin, P.J., Fakhoury, T., Blumenkopf, B., Donahue, S.P., 1999. Visual field defects after temporal lobe resection: a prospective quantitative analysis. Neurology 53 (1), 167–172. Jacobson, D.M., 1997. The localizing value of a quadrantanopia. Arch. Neurol. 54 (4), 401–404. Katz, A., Awad, I.A., Kong, A.K., Chelune, G.J., Naugle, R.I., Wyllie, E., Beauchamp, G., Luders, H., 1989. Extent of resection in temporal lobectomy for epilepsy. II. Memory changes and neurologic complications. Epilepsia 30 (6), 763–771. Kikuta, K., Takagi, Y., Nozaki, K., Hanakawa, T., Okada, T., Miki, Y., Fushimi, Y., Fukuyama, H., Hashimoto, N., 2006. Early experience with 3-T magnetic resonance tractography in the surgery of cerebral arteriovenous malformations in and around the visual pathway. Neurosurgery 58 (2), 331–337. Kinoshita, M., Yamada, K., Hashimoto, N., Kato, A., Izumoto, S., Baba, T., Maruno, M., Nishimura, T., Yoshimine, T., 2005. Fiber-tracking does not accurately estimate size of fiber bundle in pathological condition: initial neurosurgical experience using neuronavigation and subcortical white matter stimulation. Neuroimage 25 (2), 424–429. Krolak-Salmon, P., Guenot, M., Tiliket, C., Isnard, J., Sindou, M., Mauguiere, F., Vighetto, A., 2000. Anatomy of optic nerve radiations as assessed by static perimetry and MRI after tailored temporal lobectomy. Br. J. Ophthalmol. 84 (8), 884–889. Lazar, M., Weinstein, D.M., Tsuruda, J.S., Hasan, K.M., Arfanakis, K., Meyerand, M.E., Badie, B., Rowley, H.A., Haughton, V., Field, A., Alexander, A.L., 2003. White matter tractography using diffusion tensor deflection. Hum. Brain Mapp. 18 (4), 306–321. Lazar, M., Alexander, A.L., Thottakara, P.J., Badie, B., Field, A.S., 2006. White matter reorganization after surgical resection of brain tumors and vascular malformations. AJNR Am. J. Neuroradiol. 27 (6), 1258–1271. Marino Jr, R., Rasmussen, T., 1968. Visual field changes after temporal lobectomy in man. Neurology 18 (9), 825–835. Melhem, E.R., Mori, S., Mukundan, G., Kraut, M.A., Pomper, M.G., Van Zijl, P.C., 2002. Diffusion tensor MR imaging of the brain and white matter tractography. AJR Am. J. Roentgenol. 178 (1), 3–16.
297
Mori, S., Van Zijl, P.C., 2002. Fiber tracking: principles and strategies — a technical review. NMR Biomed. 15 (7–8), 468–480. Mori, S., Crain, B.J., Chacko, V.P., Van Zijl, P.C., 1999. Three-dimensional tracking of axonal projections in the brain by magnetic resonance imaging. Ann. Neurol. 45 (2), 265–269. Mori, S., Frederiksen, K., van Zijl, P.C., Stieltjes, B., Kraut, M.A., Solaiyappan, M., Pomper, M.G., 2002. Brain white matter anatomy of tumor patients evaluated with diffusion tensor imaging. Ann. Neurol. 51 (3), 377–380. Mukherjee, P., Bahn, M.M., McKinstry, R.C., Shimony, J.S., Cull, T.S., Akbudak, E., Snyder, A.Z., Conturo, T.E., 2000. Differences between gray matter and white matter water diffusion in stroke: diffusion-tensor MR imaging in 12 patients. Radiology 215 (1), 211–220. Nilsson, D., Malmgren, K., Rydenhag, B., Frisen, L., 2004. Visual field defects after temporal lobectomy — comparing methods and analysing resection size. ActaNeurol. Scand. 110 (5), 301–307. Nimsky, C., Ganslandt, O., Hastreiter, P., Wang, R., Benner, T., Sorensen, A.G., Fahlbusch, R., 2005. Intraoperative diffusion-tensor MR imaging: shifting of white matter tracts during neurosurgical procedures—initial experience. Radiology 234 (1), 218–225. Nimsky, C., Ganslandt, O., Fahlbusch, R., 2006a. Implementation of fiber tract navigation. Neurosurgery 58 (4 Suppl. 2), ONS292–ONS294. Nimsky, C., Ganslandt, O., Merhof, D., Sorensen, A.G., Fahlbusch, R., 2006b. Intraoperative visualization of the pyramidal tract by diffusion-tensor-imaging-based fiber tracking. Neuroimage 30 (4), 1219–1229. Pajevic, S., Pierpaoli, C., 2000. Color schemes to represent the orientation of anisotropic tissues from diffusion tensor data: application to white matter fiber tract mapping in the human brain. Magn. Reson. Med. 43 (6), 921. Pierpaoli, C., Jezzard, P., Basser, P.J., Barnett, A., Di Chiro, G., 1996. Diffusion tensor MR imaging of the human brain. Radiology 201 (3), 637–648. Powell, H.W., Parker, G.J., Alexander, D.C., Symms, M.R., Boulby, P.A., Wheeler-Kingshott, C.A., Barker, G.J., Koepp, M.J., Duncan, J.S., 2005. MR tractography predicts visual field defects following temporal lobe resection. Neurology 65 (4), 596–599. Rubino, P.A., Rhoton Jr, A.L., Tong, X., Oliveira, E., 2005. Three-dimensional relationships of the optic radiation. Neurosurgery 57 (4 Suppl.), 219–227. Taoka, T., Sakamoto, M., Iwasaki, S., Nakagawa, H., Fukusumi, A., Hirohashi, S., Taoka, K., Kichikawa, K., Hoshida, T., Sakaki, T., 2005. Diffusion tensor imaging in cases with visual field defect after anterior temporal lobectomy. AJNR Am. J. Neuroradiol. 26 (4), 797–803. Tecoma, E.S., Laxer, K.D., Barbaro, N.M., Plant, G.T., 1993. Frequency and characteristics of visual field deficits after surgery for mesial temporal sclerosis. Neurology 43 (6), 1235–1238. Van Buren, J.M., Baldwin, M., 1958. The architecture of the optic radiation in the temporal lobe of man. Brain 81 (1), 15–40. Wiegell, M.R., Larsson, H.B., Wedeen, V.J., 2000. Fiber crossing in human brain depicted with diffusion tensor MR imaging. Radiology 217 (3), 897–903. Wieshmann, U.C., Symms, M.R., Clark, C.A., Lemieux, L., Franconi, F., Parker, G.J., Barker, G.J., Shorvon, S.D., 1999. Wallerian degeneration in the optic radiation after temporal lobectomy demonstrated in vivo with diffusion tensor imaging. Epilepsia 40 (8), 1155–1158. Yamamoto, T., Yamada, K., Nishimura, T., Kinoshita, S., 2005. Tractography to depict three layers of visual field trajectories to the calcarine gyri. Am. J. Ophthalmol. 140 (5), 781–785. Yamamoto, A., Miki, Y., Urayama, S., Fushimi, Y., Okada, T., Hanakawa, T., Fukuyama, H., Togashi, K., 2007. Diffusion tensor fiber tractography of the optic radiation: analysis with 6-, 12-, 40-, and 81-directional motion-probing gradients, a preliminary study. AJNR Am. J. Neuroradiol. 28 (1), 92–96.
