Clinical Neurology and Neurosurgery 124 (2014) 59–65

Contents lists available at ScienceDirect

Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro

Application of diffusion tensor imaging and tractography of the optic radiation in anterior temporal lobe resection for epilepsy: A systematic review Rory J. Piper a,b, * , Michael M. Yoong a,b,c , Jothy Kandasamy c, Richard F. Chin a,b,c a

College of Medicine and Veterinary Medicine, University of Edinburgh, Chancellor’s Building, 49 Little France Crescent, Edinburgh EH16 4SB, UK Muir Maxwell Epilepsy Centre, Child Life and Health, School of Clinical Sciences and Community Health, The Royal Hospital for Sick Children, University of Edinburgh, 20 Sylvan Place, Edinburgh EH9 1UW, UK c Department of Paediatric Neurosciences, Royal Hospital for Sick Children, 9 Sciennes Road, Edinburgh EH9 1LF, UK b

A R T I C L E I N F O

A B S T R A C T

Article history: Received 20 March 2014 Received in revised form 8 May 2014 Accepted 9 June 2014 Available online 17 June 2014

Background: Approximately 50–100% of patients with temporal lobe epilepsy undergoing anterior temporal lobe resection (ATLR) will suffer a postoperative visual field defect (VFD) due to disruption of the optic radiation (OpR). Objective: We conducted a systematic review of the literature to examine the role of DTI and tractography in ATLR and its potential in reducing the incidence of postoperative VFD. Methods: We conducted an electronic literature search using PubMed, Embase, Web of Science and BMJ case report databases. Eligibility for study inclusion was determined on abstract screening using the following criteria: the study must have been (1) an original investigation or case report in humans; (2) investigating the OpR with DTI in cases of ATLR in temporal lobe epilepsy; (3) investigating postoperative VFD. All forms of ATLR and ways of assessing VFD were included to reflect clinical practice. Results: 13 studies (four case reports, eight prospective observational studies, one prospective comparative trial) were included in the review, 179 (mean  SD, 13.8  12.6; range, 1–48) subjects were investigated using DTI. The time of postoperative VFD measurement differed between the detected studies, ranging from two weeks to nine years following ATLR. A modest number of studies and insufficient statistical homogeneity precluded meta-analysis. However, DTI methods were consistently accurate at quantifying and predicting postoperative damage to the OpR. These methods revealed a correlation between the extent of OpR damage and the severity of postoperative VFD. The first and only trial with 15 subjects compared to 23 controls reported that using intraoperative tractography in ATLR significantly reduces the occurrence of postoperative VFD on comparison to conventional surgical planning. Conclusions: DTI shows potential to be an effective method used in planning ATLR. Findings from a single modest sized study suggest that tractography may be employed as part of intraoperative navigation techniques in order to avoid injury to the OpR. Further research needs to be conducted to ensure the applicability and effectiveness of this technology before implementation in routine clinical practice. ã 2014 Elsevier B.V. All rights reserved.

Keywords: Anterior temporal lobectomy Epilepsy surgery Visual field defects Diffusion tensor imaging Tractography

Contents 1. 2.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Materials and methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Search strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.

1 2 2

Abbreviations: ADC, apparent diffusion coefficient; A–P, antero–posterior; ATLR, anterior temporal lobe resection; DTI, diffusion tensor imaging; FA, fractional anisotropy; MD, mean diffusivity; MRI, magnetic resonance imaging; n/a, not applicable; OpR, optic radiation; ROI, region of interest; SAH, selective amygdalohippocampectomy; TLE, temporal lobe epilepsy; VFD, visual field defect; 3D, three-dimensional. * Corresponding author at: College of Medicine and Veterinary Medicine, University of Edinburgh, Chancellor’s Building, 49 Little France Crescent, Edinburgh, Midlothian EH16 4SB, UK. E-mail address: [email protected] (R.J. Piper). http://dx.doi.org/10.1016/j.clineuro.2014.06.013 0303-8467/ ã 2014 Elsevier B.V. All rights reserved.

60

3.

4. 5.

R.J. Piper et al. / Clinical Neurology and Neurosurgery 124 (2014) 59–65

2.2. Selection criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Search results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Study demographics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Study designs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Methods of DTI assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4. Methods of VFD assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5. Findings on DTI ROI analysis in subjects with and without postoperative VFD . . . . 3.6. 3.7. Findings on tractography analysis in subjects with and without postoperative VFD Intraoperative methods in ATLR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.8. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conflicts of interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Financial disclosures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Introduction Temporal lobe epilepsy (TLE) is the most common type of surgically treated epilepsy in adults and has a high rate of seizure freedom following surgery [1]. Anterior temporal lobe resection (ATLR) procedures (including but not restricted to en bloc resection and selective amygdalohippocampectomy (SAH)) can ameliorate drug-resistant TLE [2], however 50–100% [3,4] of patients will suffer a postoperative visual field defect (VFD) [5]. VFD are the most common complications following ATLR [6]. Even if rendered seizure-free from ATLR, patients who suffer VFD may not be able to drive. This has psychosocial implications, particularly in younger patients where the inability to drive may inhibit social independence or may disable certain occupations [7]. Pathak-Ray et al. revealed in their study that 7/14 (50%) patients who underwent ATLR for TLE failed to meet UK Driving and Vehicle Licensing Agency standards [3]. Furthermore, a larger study of patients undergoing temporal lobe surgery (n = 135) found that 64% of patients suffered a VFD and 50% of these patients had a VFD that prevented them from meeting the German legal requirements to drive [8]. Taylor et al. revealed that the ability to drive is considered by patients to be one of the five most important outcomes following epilepsy surgery [9]. Consequently, a priority for ATLR candidates is avoiding postoperative VFD [10]. VFD occurs when patients suffer disruption to the optic radiation (OpR), including the ‘Meyers loop’ (the most anterior portion of the OpR), during ATLR. Therefore, employing imaging techniques in surgical planning and guidance that delineate the OpR in relation to epileptogenic foci could facilitate the prevention of postoperative VFD. The anatomy and definitions of the OpR remain controversial. In part this is because the structures of the temporal stem lack anatomic landmarks [11], either in surgical or structural imaging investigation, and in part because the exact position and limits of the OpR have been found to be highly variable between individuals [12]. However, in recent studies, diffusion tensor imaging (DTI) has been used to more reliably reveal the anatomy of the OpR. DTI is a relatively novel application of magnetic resonance imaging (MRI) in the planning of epilepsy surgery and is not routinely utilised alongside conventional imaging modalities. DTI exploits the diffusion principles of water in the cerebral white matter tracts in order to create a three-dimensional (3D) representation of diffusion within each voxel [13]. Commonly, scalar metrics such as fractional anisotropy (FA) and mean diffusivity (MD) are derived from this diffusion ‘tensor’ in order to provide further insight into the microstructural architecture of the brain [14]. Further manipulation of the diffusion tensor can also allow the 3D visualisation of the white matter tract anatomy,

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. 2 . 2 . 2 . 3 . 3 . 3 . 3 . 3 . 3 . 5 . 5 . 7 . 7 . 7 . 00 . 7

coined ‘tractography’ or ‘fibre tracking’. DTI can be used to reveal the neuroradiological relationships and proximity between epileptogenic brain regions and important white matter tracts. Thus, the deviation, displacement and destruction of OpR white matter fibres caused either by epileptogenic regions or surgical intervention may be directly assessed using DTI. We conducted a systematic review of the literature in order to examine the potential use of DTI and tractography in surgical planning and intraoperative guidance to correlate, predict and/or prevent VFD in ATLR. 2. Materials and methods 2.1. Search strategy One investigator (RJP) conducted an electronic literature search using PubMed, Embase, Web of Science and the BMJ case report databases. Searching strategies used a combination of the following terms: epilepsy, diffusion tensor imaging, tractography, fibre tracking, surgery, visual field deficit/defect, optic radiation, and Meyer’s loop. The citations listed in the found studies and reviews were manually screened for relevant studies not detected using the electronic search. In cases where the full-text was not available, most commonly in conference abstracts, personal communication was made to the corresponding author and the full-text or relevant data was requested. Alternatively, the full text not available electronically was sought out in paper copy. The search aimed to detect all of the literature meeting our selection criteria and was initially carried out on the 19th of June, 2013, and then updated on the 5th of March, 2014. 2.2. Selection criteria Eligibility for inclusion in the review was determined on abstract screening using the following inclusion criteria: (1) the study reports an original investigation or case report in humans (i.e. not a review, comment or response); (2) the study investigated the OpR using pre-, intra- or postoperative DTI assessment in the management of ATLR in TLE; and (3) the study must have investigated a measure of VFD. We intentionally included all forms of ATLR and ways of assessing VFD to reflect clinical practice. 3. Results 3.1. Search results 46 studies were initially detected using our search strategy (nine PubMed, seven Embase, 29 Web of Science, zero BMJ case

R.J. Piper et al. / Clinical Neurology and Neurosurgery 124 (2014) 59–65

reports and one from reference searching). After the search was completed the following number of results were removed: 12, duplicates; 21, the study/report aims did not match our inclusion criteria. The full text was not available for the study by Siu et al. [15], but the information required was obtained from a recording of a presentation delivered at an international meeting. On re-searching in March 2014, one more conference proceeding abstract by Zhang et al. [4] was found, but since access to full text, presentation or correspondence was not available, the analysis of the results included in the abstract was insufficient and the study was excluded from analysis. Thirteen studies were finally included in the review. 3.2. Study demographics Collectively, the studies included 179 (mean  SD, 13.8  12.6; range, 1–48) subjects. The results from the systematic search have been summarised in Table 1. 3.3. Study designs We identified three case–control studies [5,16,17], one single case report within a larger study [18] and eight prospective (noncomparative) studies [19–26]. Two studies used DTI to assess the extent of the OpR in order to avoid VFD [15,24]. One prospective study used intraoperative tractography in SAH to avoid VFD, but this study did not include a control group [24]. We identified one prospective trial comparing the VFD outcomes after ATLR with intraoperative DTI guidance vs. ATLR with no DTI guidance [15]. All studies investigated the relationships between injury to the OpR and postoperative VFD. Two attempted to predict postoperative VFD [16,19] and two used intraoperative methods to prevent VFD by preserving the integrity of the OpR [15,24]. 3.4. Methods of DTI assessment The first report of DTI employed in this particular paradigm (and that met our search criteria) was by Wieshmann et al. [17] in which FA data was compared in three subjects undergoing either en bloc ATLR or SAH. This method of using FA was similarly employed by subsequent investigators [21,23]. In addition, region of interest (ROI) analysis of apparent diffusion coefficient (ADC) measures was conducted by Taoka et al. [23]. In addition to ROI analysis, Wieshmann et al. used voxel-based analysis of FA to correlate the visible evidence of OpR damage with VFD, using postoperative data coregistered with the data from healthy controls [17]. However, this qualitative method has subsequently been superseded for two main reasons. Firstly, subsequent studies of tractography have developed methods to quantitatively assess OpR injury, allowing objective comparisons of OpR between individuals. This has allowed investigators to control for wide intersubject variability of OpR anatomy, which is not taken into account when comparing individual OpR measurements to data from healthy controls. Seven studies adopted coregistration and comparison of preoperative tractography with intraoperative tractography or postoperative structural MRI [18–20,22,24–26] to avoid this pitfall. Secondly, recent advances in MRI, such as the utilisation of high-resolution 3 T scanners in research, increasing numbers of diffusion directions and the move from deterministic to probabilistic tractography, have substantially enhanced the ability of DTI to accurately delineate the OpR [27]. Methods of measuring and quantifying OpR injury differed across the studies. Several investigators used preoperative and intra- or postoperative tractography to measure injury to the OpR by difference in width [19] or antero–posterior (A–P) distance

61

(distance between the Meyer’s loop and temporal horn or occipital pole) [5,18,20,25,26]. 3.5. Methods of VFD assessment Goldmann perimetry, by acquisition and analysis of pre- and postoperative measures of VFD, was employed in 10 of the studies, but used variable methods of interpretation and statistical analysis. Three studies [15,21,24] employed Humphrey perimetry, an automated method with quantitative output. As well as qualitative reporting of VFD [5,16–18,24], investigators used semi-quantitative rating scales [19,22,23] and fully quantitative measurement [20,21,25,26]. Winston et al. [25] and Daga et al. [20] used a custom statistical method, since in some cases preoperative perimetry data was not available: VFD = 1 – [area of upper quadrants contralateral to resection/area of upper quadrants ipsilateral to resection]. In the detected studies the time of postoperative VFD measurement ranged from two weeks to nine years following ATLR. These data are detailed in Table 1. Table 2 details the different methods of DTI and VFD assessment used in the studies revealed by this systematic review. 3.6. Findings on DTI ROI analysis in subjects with and without postoperative VFD In a case control study in which three patients with temporal lobe resections were compared with 22 control subjects, one patient suffered postoperative homonymous hemianopia after ATLR surgery. In this case, DTI ROI analyses of the OpR showed a significant reduction in FA consistent with Wallerian degeneration extending from the temporal lobe to the occipital lobe [17]. In the same study, two subjects who underwent ATLR with no postoperative VFD showed no significant changes in FA, compared to FA data from healthy controls. In concurrence, Taoka et al., in their prospective study, measured the mean FA ratio from operated-to-intact temporal lobes and found a significant decrease according to a gradient of VFD severity [23]. ADC measures were not significantly different and thus a similar relationship was not reliably found between ADC and VFD severity. However, Taoka et al. admit their study may have been underpowered (n = 21). McDonald et al. conducted a prospective study using pre- and postoperative whole-brain voxel-wise DTI analysis in TLE subjects (n = 7) undergoing ATLR and found significant FA reduction in the temporo–occipital tract ipsilateral to surgical resection [21]. This change was significantly correlated (r = 0.83, p < 0.05) with the VFD measured by automated Humphrey analysis. This investigation extended the literature in the regard that a novel automated probabilistic atlas was employed to assess fibre location and orientation, one that can be applied to individual patient data. 3.7. Findings on tractography analysis in subjects with and without postoperative VFD Three studies demonstrated that visible disruption of the anterior edge of Meyer’s loop on postoperative tractography correlates with contralateral superior quadrantanopia [5,16,18]. Other subjects in whom postoperative tractography showed absence of injury to the OpR showed no VFD on perimetry [5,16,17]. However, these reports of tractography have only indicated the correlation between OpR damage and potential VFD in relatively small numbers of subjects. Chen et al. [19] conducted a prospective study in a larger cohort (n = 48), examining the injury fraction of the OpR on pre- and intraoperative tractography (1 (postoperative OpR width preoperative OpR width)) and VFD severity (grades 1–5). A significant correlation was found between these parameters, and ROC curve analysis found the technique to be valid

62

Table 1 Summary of search results and study outcomes. Refs. Study type

Sample size

Age (mean, range)

Surgical method

DTI method

VFD method

Time of VFD measurement after ATLR

Results

Correlation data (between OpR damage and VFD severity)

Chen et al., 2009

[19]

Prospective

48

35.4, 8–59

ATLR

Significant correlation detected between OpR injury and VFD severity. ROC curve analysis indicates accuracy and validity of the method.

Spearman correlation: r = 0.9, p < 0.001.

[20]

Prospective

20

Not ATLR reported

3–5 months

High correlation between A–P distance and Pearson correlation: r = 0.9 VFD severity.

McDonald [21] et al., 2010

Prospective

7

31.1

ATLR

Postoperative FA (whole-brain, ROI, voxel-wise) analysis at two months and one year.

Goldmann perimetry. Five-point degree score. Goldmann perimetry. % of upper quadrant. Automated, quantitative Humphrey perimetry.

3 months

Powell et al., [16] 2005

Case–control

2

42.0, 37–47

ATLR

Pre- and postoperative tractography with qualitative reporting.

Goldmann perimetry.

2 months

Siu et al., 2012

[15]

Prospective comparative trial

39, 15 17–57 (vs. 23 controls)

No significant FA changes from two months Significant Spearman correlation to one year. between ipsilateral FA change and VFD (r = 0.83, p < 0.05), but not between contralateral FA change and VFD (r = 0.7). n/a One subject postoperative contralateral superior quadrantanopia suffered disruption to anterior Meyer’s loop. Another subject had no apparent damage to the OpR and had no postoperative VFD. n/a One subject postoperative contralateral superior quadrantanopia suffered disruption to anterior Meyer’s loop. Another subject had no apparent damage to the OpR and had no postoperative VFD. n/a 66% of tractography patients had no postoperative deficit vs. 30% in the control group.

Taoka et al., 2005

[23]

Prospective

14

31.9

Taoka et al., 2008

[22]

Prospective

14

Thudium [24] et al., 2010

Prospective

Wieshmann [17] et al., 1999

Case–control

Preoperative tractography was used Standard vs. tractography- in intraoperative guidance for ATLR guided ATLR and was compared against results from ATLR with no DTI guidance. ATLR ADC and FA (ROI) postoperative analysis in the intact and operated OpR.

Not reported Automated, quantitative Humphrey perimetry. 3 weeks, 9 Goldmann years perimetry. VFD severity groups – A to C.

13–66

ATLR

12

33.4, 15–49

SAH

3

34.0, 20–50

One SAH, two Postoperative FA (ROI) comparison ATLR with healthy controls.

Goldmann perimetry. VFD severity groups – A to D. Automated, quantitative Humphrey perimetry. Goldmann perimetry.

Preoperative tractography coregistered with postoperative structural MRI. Injury to OpR determined by overlap between resection area and preoperative OpR. Preoperative tractography used in surgical planning operative neuronavigation.

Operated:intact ratio between FA operated:intact ratio was significantly decreased in subjects with severe VFD. The group A and C: p < 0.05. ADC ratio was not significantly different.

3 months Goldmann perimetry.

Significant correlation between injury to the Spearman correlation: r = 0.8, OpR and VFD severity. p = 0.001 Median 4 months (range, 3–12 months) Goldmann perimetry. % of upper quadrant.

ATLR 37, 18–62 Prospective Yogarajah [26] et al., 2009

21

ATLR 35.6, 17–56 Prospective Winston [25] et al., 2012

20

ATLR Case report Winston [18] et al., 2011

1

17

Preoperative tractography coregistered with postoperative structural MRI. Injury to OpR determined by overlap between resection area and preoperative OpR. Preoperative tractography coregistered with postoperative structural MRI. Injury to OpR determined by overlap between resection area and preoperative OpR. Preoperative tractography coregistered with postoperative structural MRI. Injury to OpR determined by overlap between resection area and preoperative OpR.

n/a Resection of lesion confluent with the Meyer’s loop resulted in superior quadrantanopia. Not recorded Goldmann perimetry.

R.J. Piper et al. / Clinical Neurology and Neurosurgery 124 (2014) 59–65

63

Table 2 Summary of the methods of diffusion tensor imaging (DTI) assessment and visual field defect (VFD) assessment used amongst studies included in the systematic review. Method of DTI assessment Region of interest (ROI) analysis using DTI Qualitative tractography

Quantitative tractography

Method VFD assessment Goldmann perimetry Humphrey perimetry

Comparison of DTI data (ADC, FA, or MD) measured at the OpR between pre- and postoperative time images or against control data. Observation of disruption to the anatomy of the OpR on postoperative tractography [5]. Comparison to either healthy controls or preoperative imaging. Calculation of OpR damage using preoperative imaging (width [19] or antero–posterior distance of the OpR [5,18,20,25,26]) and the overlapping resection margin on postoperative structural MRI.

Manual method of visual field assessment. Employed by 10/13 of the included studies. Automated method of visual field assessment with quantitative output. Employed by 3/14 of the included studies [15,21,24].

and reproducible. Several other studies detected a significant correlation between the severity of VFD and either the resection margin to OpR overlap [22] or the A–P distance [20,22,25,26]. 3.8. Intraoperative methods in ATLR Our literature search identified two prospective studies that employed intraoperative methods of DTI and tractography in the surgical management of TLE. Thudium et al. [24] employed tractography in image-guided neuronavigation in 12 patients with SAH, attempting to plan safe trajectories of resection to preserve Meyer’s loop. Postoperative perimetry found 75% of subjects to have no postoperative VFD. No internal control sample was used; however this result compares favourably to the results reported in a previous study in which 53% of subjects suffered major VFD after SAH [28]. A study conducted by Siu et al. [15] marks the first prospective comparative trial in this area. The open labelled study compared the outcomes of using intraoperative tractography in ATLR (15 subjects) to conventional ATLR image guidance (23 controls). 66% of those in the tractography group had no VFD vs. 30% with no VFD in the control group, No data analysis was reported but we calculate the percentage difference is 36% (95% CI 4–60%). Limitations of the study are the modest sample size, no evidence of treatment randomisation and no allocation concealment by those carrying out the VFD assessment or data analysis. 4. Discussion A recently written review by Winston has already commented on the study of epilepsy surgery, the OpR and VFD [27]. In a recent study, DTI has been shown to be helpful in the neurosurgical planning of patients with cerebral tumours in proximity to eloquent white matter tracts, such as the OpR [29]. However, our review is the first systematic review, to our knowledge, that has collated information from all of the published studies on epilepsy surgery that have investigated the clinical implications of using DTI and tractography to correlate, predict and/or prevent VFD in ATLR. It is hypothesised that a parameter responsible for the variable VFD outcomes in ATLR with comparable resection may be the intersubject variability of the anatomy of the Meyer’s loop [22]. It is well reported that the dimensions of OpR, as well as the Meyer’s

64

R.J. Piper et al. / Clinical Neurology and Neurosurgery 124 (2014) 59–65

loop individually, show significant intersubject heterogeneity [5,16,22,26]. The intrasubject variability of these structures has also been reported, i.e. there is no strict symmetry between the dimensions of the left and right OpR in individuals [5]. It is therefore considered that analyses of the OpR are required to be corrected by individual parameters. This has been largely achieved by the co-registration of pre- and postoperative imaging for measurement of OpR-specific injury, as opposed to previous techniques of measuring crude temporal lobe resection size [22]. These individualised methods allow increased sensitivity in detecting the correlations between VFD and are suggested to be more reliable in using DTI when predicting the postoperative risks of VFD [5]. In the studies reviewed the most recent and robust methods used quantification of the A–P distance, measuring the distance between the anterior border of Meyer’s loop to the edge of the resection margin. The optimum time to asses VFD after ATLR has been disputed. It has been suggested that VFD occurring early after ATLR may be due to transient oedema or inflammation of the OpR [21] and therefore studies measuring postoperative VFD at an early stage may overestimate long term outcome. However, measuring VFD within two weeks avoids the influence of Wallerian degeneration [19]. In this systematic review nine studies reported on the timing of VFD measurement. Times reported ranged considerably from within two weeks to nine years post procedure. The largest prospective study (n = 48) included in this review measured VFD within two weeks postoperatively in all subjects and therefore may overestimate the presence or severity of VFD. There are a number of variations in DTI methodology used by investigators, which would allow variable predictions of OpR anatomy in a single subject. Disagreement of OpR anatomy is exacerbated since there are very few studies comparing anatomy on imaging and pathology. Benjamin et al. demonstrate this problem in their experiment using a number of different ROI to seed tractography streamlines of the OpR [30]. They found that each attempt had a significantly different anatomical result. It may be considered that using multiple ROI on seeding the OpR will generate a more accurate prediction of OpR anatomy. There are a number of developments that may increase the accuracy of OpR tractography. For example, it may be possible to further optimise tractography of the OpR by employing multi-tensor diffusion imaging [26,31]. This method combines tensors in order to overcome the limitations of single-tensor diffusion imaging which may be susceptible to partial volume artifact or error in reconstructing crossing fibres. Furthermore, constrained spherical harmonics, q-ball imaging, or diffusion spectrum imaging may offer more accurate modelling of complex fibre orientation. The challenges in current DTI methods and the potential for future improvements are discussed fully in a review by Mandelstam [32]. Surgical strategy was not heterogenous across the studies. For example, Thudium et al. report a novel basal temporal approach [24], as opposed to the conventional transsylvian approach for ATLR. However, our review captures the lack of consensus in current practice for the most effective approach for ATLR. Considerations that influence the decision of surgical approach include likelihood of seizure freedom and neuropsychological outcome. In terms of VFD, there has been suggestion that both transsylvian SAH and transcortical SAH are less likely to cause VFD than standard ATLR, but this suggestion is based on a small sample in one centre [4]. Further study of the OpR in ATLR using DTI may allow further understanding of the anatomy of OpR. For example, Jeelani et al. suggest that a left/right hemispherical asymmetry of the geniculocalcarine tracts causes the risk of VFD after left ATLR to be 3.5 times higher than that after right ATLR [33]. In addition, studies that investigate the anatomical and VFD consequences of

selective ATLR procedures may allow us to be informed on the potential benefits and risks compared to standard ATLR [34,35]. Furthermore, the study by Thudium et al. highlights that brain shift is a significant confound of using preoperative imaging alone in neuronavigation. Brain shift is soft tissue deformation that typically occurs after craniotomy and penetration of the ventricular system. Chen et al. [19] detected significant brain shift (horizontal range 0–11.1 mm) of the OpR between pre- and intraoperative imaging and claim that methods that only use resection size to determine OpR damage may be inaccurate. However, further advances in intraoperative imaging may offer a solution to this problem. Accurate intraoperative localisation of the OpR using tractography has been shown to be possible and effective at various time points through ATLR, as presented by Daga et al. in a recently published interventional imaging workflow [20]. Intraoperative images can be taken after craniotomy, initial resection and post resection prior to full surgical closure. These techniques may prove useful in designing surgical strategies to avoid OpR injury in this paradigm and other pathologies such as cerebral tumours. Further improvements to intraoperative DTI assessment may involve coregistration with fMRI [36] or the employment of automated methods to reduce operator time [31]. Despite the increasing employment of DTI and tractography in this field, these imaging modalities are used mostly in research and are not currently considered conventional methods of surgical planning or guidance. As discussed, DTI findings have so far been shown to correlate with and predict postoperative VFD. However, it has been suggested by a number of authors that DTI and tractography may provide significant benefit in preventing postoperative VFD when used in intraoperative neuronavigation [19,25]. The potential benefits of intraoperative imaging are supported in the study conducted by Thudium et al., who appear to present the first study to use preoperative tractography in influencing surgical decision [24]. The study showed that it is possible to use these methods in an intraoperative setting, however a limitation of advocating the benefits (75% of subjects VFD free) of this technique were only weighed against a previously published study [28] and not a control sample. A recent study by Siu et al. marks the first study to date, to our knowledge, that has used a control sample [15]. The study found that additionally using intraoperative tractography data, the percentage of patients without postoperative VFD increased from 30% (control group) to 66%. A limitation of this review is that the application of DTI in epilepsy surgery is being intensely researched with most of this research occurring over the last decade. In a few years time there is likely to be an increase in the number of attempts to use intraoperative tractography in ATLR, requiring further review. Furthermore, because of the modest number of studies on the topic and the variation in clinical practice, there was not sufficient statistical homogeneity (due to different methodology of measurement and statistical analysis employed between studies) to allow metaanalytical comparison of the relationship between OpR injury and VFD. The surgical approach also differed between studies, preventing statistical comparison of VFD outcomes among different approaches. There have only been two studies to date that have directly assessed the additional benefits of using intraoperative DTI and tractography compared to conventional image guidance in ATLR, and both have been in adults. We advocate there is a need for further robust and large-scale clinical trial of this approach in adults and in children before introduction of this technique into conventional clinical practice. Alternatively, an observational case– control study of visual field outcomes after ATLR in centres that do and do not use these techniques would add useful information to the debate on the role of DTI in reducing VFD after ATLR.

R.J. Piper et al. / Clinical Neurology and Neurosurgery 124 (2014) 59–65

Very few paediatric subjects were found in the studies detected in our search. While some work has been carried out using tractography to delineate the OpR in a paediatric population [37,38], the utility of these approaches in epilepsy surgery requires further investigation. 5. Conclusion In summary, DTI methods have potential to be effectively integrated with other neuroimaging modalities in the planning of ATLR. Specifically, these methods may be employed intraoperatively to avoid injury to the OpR. However, only one small trial has tested the potential benefit of DTI guidance in addition to conventional imaging. Further research needs to be conducted to ensure the applicability and effectiveness of DTI techniques. Conflicts of interest None. Financial disclosures None. Acknowledgments Michael M. Yoong is supported by an NHS Research Scotland/ University of Edinburgh Clinical Lectureship. Jothy Kandasamy holds an NHS Research Scotland Career Researcher Fellowship. The Muir Maxwell Epilepsy Centre receives support from the Muir Maxwell Trust. References [1] Wiebe S, Blume WT, Girvin JP, Eliasziw M. A randomized, controlled trial of surgery for temporal-lobe epilepsy. N Engl J Med 2001;345:311–8. [2] Al-Otaibi F, Baeesa SS, Parrent AG, Girvin JP, Steven D. Surgical techniques for the treatment of temporal lobe epilepsy. Epilepsy Res Treat 2012;2012:374848. [3] Pathak-Ray V, Ray A, Hatfield R. Detection of visual field defects in patients after anterior temporal lobectomy for mesial temporal sclerosis –. Eye 2002;16:744–8. [4] Zhang H, Wu W, Fang Y, Zhou D, Lei D. Changes in optic radiation integrity and visual fields afer three different temporal lobe epilepsy surgeries. 30th International Epilepsy Conference. p. 179. [5] Nilsson D, Starck G, Ljungberg M, Ribbelin S, Jönsson L, Malmgren K, et al. Intersubject variability in the anterior extent of the optic radiation assessed by tractography. Epilepsy Res 2007;77:11–6. [6] Anderson DR, Trobe JD, Hood TW, Gebarski SS. Optic tract injury after anterior temporal lobectomy. Ophthalmology 1989;96:1065–70. [7] Manji H, Plant GT. Epilepsy surgery, visual fields, and driving: a study of the visual field criteria for driving in patients after temporal lobe epilepsy surgery with a comparison of Goldmann and Esterman perimetry. J Neurol Neurosurg Psychiatry 2000;68:80–2. [8] Beisse F, Lagreze W, Schmitz J, Schulze-Bonhage A. Visual field defects after epilepsy surgery: implications for driving license tenure. Ophthalmologe 2014 [Epub ahead of print]. [9] Taylor DC, McMacKin D, Staunton H, Delanty N, Phillips J. Patients’ aims for epilepsy surgery: desires beyond seizure freedom. Epilepsia 2001;42:629–33. [10] Gilliam F, Kuzniecky R, Faught E, Black L, Carpenter G, Schrodt R. Patientvalidated content of epilepsy-specific quality-of-life measurement. Epilepsia 1997;38:233–6. [11] Kier EL, Staib LH, Davis LM, Bronen RA. MR imaging of the temporal stem: anatomic dissection tractography of the uncinate fasciculus, inferior occipitofrontal fasciculus, and Meyer’s loop of the optic radiation. Am J Neuroradiol 2004;25:677–91. [12] Ebeling U, Reulen HJ. Neurosurgical topography of the optic radiation in the temporal lobe. Acta Neurochir (Wien) 1988;92:29–36. [13] Basser PJ, Mattiello J, LeBihan D. MR diffusion tensor spectroscopy and imaging. Biophys J 1994;66:259–67.

65

[14] Beaulieu C. The basis of anisotropic water diffusion in the nervous system – a technical review. NMR Biomed 2002;15:435–55. [15] Siu D, Zhu X, Abrigo J, Wang Y, Leung J, Mak C, et al. The use of diffusion tensor tractography to prevent visual field defect in epilepsy surgery – a prospective study. J Med Imaging Radiat Oncol 2012 63rd Annual Scientific Meeting of the Asian Oceanic Congress of Radiology. [16] Powell HWR, Parker GJM, Alexander DC, Symms MR, Boulby PA, WheelerKingshott CAM, et al. MR tractography predicts visual field defects following temporal lobe resection. Neurology 2005;65:596–9. [17] Wieshmann UC, Symms MR, Clark CA, Lemieux L, Franconi F, Parker GJ, et al. Wallerian degeneration in the optic radiation after temporal lobectomy demonstrated in vivo with diffusion tensor imaging. Epilepsia 1999;40:1155–8. [18] Winston GP, Yogarajah M, Symms MR, McEvoy AW, Micallef C, Duncan JS. Diffusion tensor imaging tractography to visualize the relationship of the optic radiation to epileptogenic lesions prior to neurosurgery. Epilepsia 2011;52:1430–8. [19] Chen X, Weigel D, Ganslandt O, Buchfelder M, Nimsky C. Prediction of visual field deficits by diffusion tensor imaging in temporal lobe epilepsy surgery. Neuroimage 2009;45:286–97. [20] Daga P, Winston G, Modat M, White M, Mancini L, Cardoso MJ, et al. Accurate localization of optic radiation during neurosurgery in an interventional MRI suite. IEEE Trans Med Imaging 2012;31:882–91. [21] McDonald CR, Hagler DJ, Girard HM, Pung C, Ahmadi ME, Holland D, et al. Changes in fiber tract integrity and visual fields after anterior temporal lobectomy. Neurology 2010;75:1631–8. [22] Taoka T, Sakamoto M, Nakagawa H, Nakase H, Iwasaki S, Takayama K, et al. Diffusion tensor tractography of the Meyer loop in cases of temporal lobe resection for temporal lobe epilepsy: correlation between postsurgical visual field defect and anterior limit of Meyer loop on tractography. Am J Neuroradiol 2008;29:1329–34. [23] Taoka T, Sakamoto M, Iwasaki S, Nakagawa H, Fukusumi A, Hirohashi S, et al. Diffusion tensor imaging in cases with visual field defect after anterior temporal lobectomy. Am J Neuroradiol 2005;26:797–803. [24] Thudium MO, Campos AR, Urbach H, Clusmann H. The basal temporal approach for mesial temporal surgery: sparing the Meyer loop with navigated diffusion tensor tractography. Neurosurgery 2010;67:385–90. [25] Winston GP, Daga P, Stretton J, Modat M, Symms MR, McEvoy AW, et al. Optic radiation tractography and vision in anterior temporal lobe resection. Ann Neurol 2012;71:334–41. [26] Yogarajah M, Focke NK, Bonelli S, Cercignani M, Acheson J, Parker GJM, et al. Defining Meyer’s loop-temporal lobe resections, visual field deficits and diffusion tensor tractography. Brain 2009;132:1656–68. [27] Winston GP. Epilepsy surgery, vision, and driving: what has surgery taught us and could modern imaging reduce the risk of visual deficits? Epilepsia 2013;54:1877–88. [28] Renowden SA, Matkovic Z, Adams CB, Carpenter K, Oxbury S, Molyneux AJ, et al. Selective amygdalohippocampectomy for hippocampal sclerosis: postoperative MR appearance. Am J Neuroradiol 1995;16:1855–61. [29] Yu C, Li K, Ji X, Qin W. Diffusion tensor tractography in patients with cerebral tumors: a helpful technique for neurosurgical planning and postoperative assessment. Eur J Radiol 2005;56:197–204. [30] Benjamin CFA, Singh JM, Prabhu SP, Warfield SK. Optimization of tractography of the optic radiations. Hum Brain Mapp 2014;35:683–97. [31] Winston GP, Mancini L, Stretton J, Ashmore J, Symms MR, Duncan JS, et al. Diffusion tensor imaging tractography of the optic radiation for epilepsy surgical planning: a comparison of two methods. Epilepsy Res 2011;97:124– 32. [32] Mandelstam SA. Challenges of the anatomy and diffusion tensor tractography of the Meyer loop. Am J Neuroradiol 2012;33:1204–10. [33] Jeelani NUO, Jindahra P, Tamber MS, Poon TL, Kabasele P, James-Galton M, et al. ‘Hemispherical asymmetry in the Meyer’s Loop’: a prospective study of visualfield deficits in 105 cases undergoing anterior temporal lobe resection for epilepsy. J Neurol Neurosurg Psychiatry 2010;81:985–91. [34] Mengesha T, Abu-Ata M, Haas K, Lavin P, Sun D, Konrad P, et al. Visual field defects after selective amygdalohippocampectomy and standard temporal lobectomy. J Neuroophthalmol 2009;29:208–13. [35] Yeni S, Tanriover N, Uyanik O, Ulu M, Ozkara C, Karaagac N, et al. Visual field defects in selective amygdalohippocampectomy for hippocampal sclerosis: the fate of Meyer’s loop during the transsylvian approach to the temporal horn. Neurosurgery 2008;63:507–13. [36] Tax C, Duits R, Haar Romenij B, Vilanova A, Ossenblok P. Tractography of the optic radiation for vision sparing epilepsy surgery. Proceedings of the 2012 international conference on information and automation (ICIA, Shenyang, China, June 6–8, 2012). Shenyang, China. p. 441–5. [37] Glass HC, Berman JI, Norcia AM, Rogers EE, Henry RG, Hou C, et al. Quantitative fiber tracking of the optic radiation is correlated with visual-evoked potential amplitude in preterm infants. Am J Neuroradiol 2010;31:1424–9. [38] Xie S, Gong GL, Xiao JXI, Ye JT, Liu HH, Gan XL, et al. Underdevelopment of optic radiation in children with amblyopia: a tractography study. Am J Ophthalmol 2007;143:642–6.

Application of diffusion tensor imaging and tractography of the optic radiation in anterior temporal lobe resection for epilepsy: a systematic review.

Approximately 50-100% of patients with temporal lobe epilepsy undergoing anterior temporal lobe resection (ATLR) will suffer a postoperative visual fi...
353KB Sizes 1 Downloads 5 Views