Original Article
Quantification of Corticospinal Tracts with Diffusion Tensor Imaging in Brainstem Surgery: Prognostic Value in 14 Consecutive Cases at 3T Magnetic Resonance Imaging Yuqiang Yao1,2, Nils H. Ulrich1,3, Roman Guggenberger1, Yahea A. Alzarhani1, Helmut Bertalanffy4, Spyros S. Kollias1
OBJECTIVE: In this study, we investigated the potential prognostic role of morphologic and quantitative diffusion tensor imaging (DTI) in patients with brainstem cavernoma (BSC) in terms of postoperative outcome.
-
METHODS: In this retrospective study of 14 brainstem cavernoma patients, we analyzed pre- and postoperative DTI data. White matter tractography of corticospinal tracts (CSTs) was performed with the Fiber Assignment by Continuous Tracking algorithm, and morphologic characteristics of CSTs were compared with clinically assessed motor strength. Fractional anisotropy (FA) and apparent diffusion coefficient (ADC) were measured in ipsi- and contralesional regions of interest at the lesion level, as well as levels caudal and rostral to the lesion. Correlation analysis was performed between lateral index (LI) of ipsi-/contralateral FA and ADC values and patients’ motor function.
-
RESULTS: Preoperatively, normal morphologic features of CSTs corresponded to normal motor function in 4 patients. The other 10 morphologically abnormal preoperative CSTs didn’t show corresponding motor impairment either in pre- or postoperative follow-up period. The sensitivity, specificity, positive predictive, and negative predictive values of white matter tractographic morphology on preoperative motor function were 100%, 57.1%, 70%, and 100%, respectively. The corresponding values on follow-up motor function were 100%, 33.3%, 20%, and 100%, respectively.
-
Key Words Apparent diffusion coefficient - Brainstem cavernoma - Corticospinal tract - Diffusion tensor imaging - Fractional anisotropy - White matter tractography -
Abbreviations and Acronyms BSC: Brainstem cavernoma CST: Corticospinal tract DTI: Diffusion tensor imaging FA: Fractional anisotropy LI: Lateral index MRC: Medical Research Council MRI: Magnetic resonance imaging NPV: Negative predictive value
1006
www.SCIENCEDIRECT.com
The mean FA at lesion level was significantly lower than the corresponding FA at the contralateral hemisphere (P [ 0.009). In areas caudal to the lesion the same trend yet without statistic significance was observed (P [ 0.23). There was no significant laterality difference of mean FAs rostral to the level of the lesion. No correlation in LI of FA or ADC in the 3 anatomic levels with simultaneous and long-term follow up motor function was observed. Restoration of the morphology of the affected CST postoperatively was associated with a trend for decreasing ADC compared with the preoperative measurements. CONCLUSION: Intact CST morphology in DTI predicts a favorable postoperative outcome in patients with BSC. Interrupted CSTs and decreased FA values correlate well within BSC lesion level; nevertheless, morphologic characteristics and diffusion parameter changes at lesion level cannot predict poor prognosis. Caudal and rostral diffusion parameters can provide more information of the integrity of CSTs compared with morphologic study alone.
-
INTRODUCTION
I
n recent years, the incidence of brain cavernomas has increased as the result of advances in neuroimaging, particularly the widespread use of magnetic resonance imaging (MRI)
PPV: Positive predictive value ROI: Region of interest WMT: White matter tractography From the 1Department of Neuroradiology, University Hospital, University of Zurich, Zurich, Switzerland; 2Department of Neurosurgery, Beijing Jishuitan Hospital, Beijing, China; 3 Department of Neurosurgery, University Hospital, University of Zurich, Zurich, Switzerland; and 4Department of Neurosurgery, International Neuroscience Institute, Hannover, Germany To whom correspondence should be addressed: Yuqiang Yao, M.D. [E-mail: [email protected]] Citation: World Neurosurg. (2015) 83, 6:1006-1014. http://dx.doi.org/10.1016/j.wneu.2015.01.045 Journal homepage: www.WORLDNEUROSURGERY.org Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2015 Elsevier Inc. All rights reserved.
WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2015.01.045
ORIGINAL ARTICLE YUQIANG YAO ET AL.
QUANTIFICATION OF CST WITH DTI IN BRAINSTEM SURGERY
Table 1. General Information and Morphologic Changes of CST Motor Scale*
Morphologic Changes Patient No.
Age (Years)/Sex
Lesion Level
Side
Approach
Preoperative
Postoperative
0
1
2
1
25/f
Pons
R
2
39/f
Pons
L
Suboccipital (midline)
Interrupted
Interrupted
3
4
5
Retromastoidal
Deformed
Deviated
5
5
5
3
32/f
Pons þ mesencephalon
R
Subtemporal
Interrupted
Deformed
4
4
5
4
44/f
Pons
5
22/f
Pons
L
Suboccipital (midline)
Deformed
Deviated
4
5
5
R
Suboccipital (midline)
Normal
Normal
5
5
5
6
52/f
Pons
M
Suboccipital (midline)
Normal
Normal
5
5
5
7
19/f
Medulla oblongata
L
Retromastoidal
Interrupted
Interrupted
4
4
4
8
6/m
Pons
R
Retromastoidal
Interrupted
N/A
4
4
5
Interrupted
N/A
4
4
4
Normal
N/A
5
5
5
9
50/m
Pons
R
Suboccipital (midline)
10
30/m
Pons
L
Retromastoidal
11
26/f
Pons
R
Suboccipital (midline)
Interrupted
N/A
5
5
5
12
51/f
Mesencephalon
L
Supracerebellar-infratentorial
Deformed
N/A
4
5
5
13
63/f
Medulla oblongata
L
Suboccipital (midline)
Deviated
N/A
5
5
5
14
45/f
Pons
L
Retromastoidal
Normal
N/A
5
5
5
CST, corticospinal track; f, female; R, right, L, left; m, male; N/A, not available; MRC, Medical Research Council. *MRC scale 0: preoperative; 1: postoperative; 2: follow-up.
in clinical practice. The prevalence of intracranial cavernomas ranges from 0.4% to 0.9% (8, 28), with the ones being localized in the brainstem representing 8%e22% (11). This subgroup of cavernomas has a substantially greater tendency for bleeding (up to 30%) (25), which is more likely to result in severe neurologic deficits (7, 11, 25) and moreover has a greater incidence of recurrent hemorrhage than those in other locations (26, 34). Surgical extirpation is the major treatment option for the symptomatic brainstem cavernomas (BSCs). Because of their location deep within a relatively small anatomic region, highly compact in functionally important white matter tracts and grey matter nuclei, BSCs to date remain a major challenge for neurosurgeons. They also represent a challenge for a functionally relevant neuroradiologic description because of the poor differentiation of individual anatomical structures on conventional MRI. Diffusion tensor imaging (DTI) is a recently introduced technique that allows evaluation of white matter integrity by virtue of its ability to visualize water diffusion characteristics. White matter tractography (WMT), also known as “fiber tracking,” allows 3-dimensional visualization of white matter tracts and accurate description of their architecture and integrity in the brainstem (2, 12, 21). The validity and reliability of WMT via the use of DTI for brainstem lesions in preoperative planning have been well demonstrated in previous studies (6, 22). Recent studies in patients with brain tumors have used quantitative DTI parameters, namely apparent diffusion coefficient (ADC) and fractional anisotropy (FA), for histologic prediction of the tumor (15, 19, 29)
WORLD NEUROSURGERY 83 [6]: 1006-1014, JUNE 2015
and to characterize the effects of the tumor on the surrounding white matter tracts (5, 10). On the basis of these studies, several institutions have adopted DTI as an increasingly important component of preoperative planning in patients with supratentorial tumors (35) and brainstem lesions (5, 31, 32). Currently, there are no clinical or imaging findings that can reliably predict postoperative motor function in patients with BSCs. Our hypothesis is that the structural morphology and quantitative measures of DTI in these patients might provide prospectively prediction of postoperative clinical findings and outcome and, thus, hold certain prognostic value for more informed therapy decision making. To test this hypothesis, we investigated the prognostic value of morphologic and quantitative imaging parameters generated from WMT of corticospinal tract (CST) using pre- and postoperative DTI in 14 patients with BSCs. MATERIALS AND METHODS DTI data were analyzed in a retrospective study of 14 patients with BSC, which were operated between January of 2008 and June of 2010 at the department of neurosurgery. Each diagnosis was confirmed by histology. Patients’ details are listed in Table 1. The mean age of the patients was 36 years (range, 6e63 years). Most of the lesions located in the pons (10/14, 71.4%); 1 lesion in the mesencephalon (7.1%); 2 lesions in the medulla oblongata (14.2%); and 1 lesion involved both pons and mesencephalon (7.1%). Six lesions were mainly on the right side, 7 mainly on the left and one in the midline of the
www.WORLDNEUROSURGERY.org
1007
ORIGINAL ARTICLE YUQIANG YAO ET AL.
QUANTIFICATION OF CST WITH DTI IN BRAINSTEM SURGERY
Figure 1. Illustration of a patient with a BSC in the right cerebral peduncle (CP) and anterior part of the pons (PN) (Patient 3). (AeB) Pre-operative structural magnetic resonance (MR) imaging. (CeE) Preoperative FA map (blue: contralateral reconstructed corticospinal tract [CST]; red and pink: ipsilateral reconstructed CST); (FeG): postoperative structural MR, (HeJ): postoperative FA map (blue: contralateral reconstructed CST; yellow and red: ipsilateral reconstructed CST). The respective colors of fiber tracts were chosen to distinguish ipsilateral from contralateral side and pre-from postoperative situations. Fiber tracts generated from DTI data and consecutively calculated FA maps nicely followed the normal anatomy of
brainstem. All patients underwent examination via MRI, including DTI, as part of the routine preoperative protocol at the department of neuroradiology. Seven patients also had DTI acquired during their postoperative follow-up examinations. All subjects underwent microsurgical resection of brainstem cavernoma. The surgical approach was determined by the size, location of the lesion, and the spatial relationship to critical structures such as the corticospinal and sensory tracts. For lesions located in lower pons and medulla oblongata, the midline suboccipital approach was used for medially located lesions; the retromastoidal approach was used for laterally located lesions. For lesions located in upper pons and mesencephalon, the subtemporal approach was used for the anterior and lateral lesions and supracerebellar-infratentorial approach for posterior ones. The fasciculi were protected carefully under intraoperative DTI navigation. Multimodality intraoperative monitoring, such as motor-evoked potentials, somatosensory-evoked potential, acoustic-evoked potentials, neuronavigation, and mapping of the rhomboid fossa also was used to support the surgical intervention.
1008
www.SCIENCEDIRECT.com
CST. Preoperative FA map showed severely interrupted CST of the lesion side; postoperative FA map showed remarkable improvement of lesion insulted CST. The MRC grades were 4, 4, and 5 at preoperative, postoperative, and follow-up time points. This figure also illustrates how ROIs were drawn at 4 different anatomic levels in rostrocaudal direction (The color of region of interests [ROIs] were randomly assigned just for distinction with each other): the posterior limbs of the internal capsule (PL), the mid-portion of the CP, the anterior part of PN, and upper medullaoblongata (UM).
Each patient’s motor grade was assessed using the Medical Research Council (MRC) scale of 0e5 (23) at 3 time points: preoperatively (MRC0), postoperatively (MRC1), and in 3e6 months’ follow-up (MRC2). Conventional MRI and DTI Acquisition MRI was performed at a 3-Tesla whole-body magnetic resonance system (PhilipsAchieva, Best, Netherlands), equipped with 80 mT/ m/ms gradient coils and an 8-element receive head coil array (MRI Devices Corp., Waukesha, WI). T2-weighted images (echo time 85ms, repetition time 4900 ms, matrix size 320 256, field of view 220 mm, slice thickness 4 mm), as well as T1-weighted images (echo time 10 ms, repetition time 400 ms, matrix size 256 192, field of view 220 mm, slice thickness 4 mm) before and after intravenous contrast administration were acquired. DTIs were acquired, as part of the routine preoperative protocol via the use of a single-shot, spin-echo, diffusion-weighted, echo planar imaging sequence with the following parameters: in-plane matrix ¼ 96 96, reconstructed to 128 128, 60 contiguous slices
WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2015.01.045
ORIGINAL ARTICLE YUQIANG YAO ET AL.
QUANTIFICATION OF CST WITH DTI IN BRAINSTEM SURGERY
Table 2. Accuracy of WMT TP
FN
FP
TN
Se %
Sp %
PPV %
NPV %
MRC0
7
0
3
4
100
57.1
70.0
100
MRC2
2
0
8
4
100
33.3
20.0
100
WMT, white matter tractography; TP, true positive; FN, false negative; FP, false positive; TN, true negative; Se, sensitivity; Sp, specificity; PPV, positive predictive value; NPV, negative predictive value; MRC, Medical Research Council.
of 2.1-mm thickness, TE ¼ 50 ms, number of signal averages ¼ 2, and 60% partial k-space acquisition. Diffusion weighting with a maximal b-factor of 1000 s/mm2 was carried out along 15 icosahedral directions complemented by one scan with b ¼ 0. WMT and Morphologic Study DTIs were processed off-line using commercially available software on an “Extended MR Working Station” (Philips Achieva). White matter fiber-tracking was performed by the use of the fiber assignment by continuous tracking algorithm with a fractional anisotropy of 0.15 and an angle of 27. The fiber tracking of CST was obtained by placing 3 different regions of interest (ROIs) along the pathway. All ROIs were identified by anatomical landmarks and color-coded FA maps in which red indicates fibers running along the right-left direction, green anterior-posterior, and blue inferior-superior. The first ROI was placed in the upper medulla, which is above the decussation of the pyramidal tracts; the second in the posterior limb of the internal capsule; and the third, in the upper part of the semioval center, adjacent to the primary motor cortex. Tracking of the CST was performed by
selecting only the fibers passing through all 3 ROIs. After the first reconstruction, some fibers, which by their anatomic distribution were obviously not belonging to the CSTs, were removed by additional operations (using exclusive ROIs). The reconstructed fibers were visually evaluated for similarity to reference atlas images and conformity to the above-mentioned anatomic landmarks. The CSTs were reconstructed in all subjects on both sides. We adapted the criteria developed by Lazar et al. (17) and revised by Kovanlikaya et al. (16) to classify fiber tract involvement into 4 groups: normal, deviated, deformed, or interrupted. The patterns of tract alteration were identified similar to Kovanlikaya’s description as follows: Group I: “normal” Group II: “deviated” if the tract was in abnormal location and/or direction due to mass effect of the lesion. Group III: “deformed” if there was a partial defect or interruption in parts of tract while the rest of it was identifiable on FA color maps or WMT. Group IV: “interrupted” if the tract was discontinuous or defective both on FA color maps and WMT. ROI Analysis and Quantitative Study We quantified diffusion tensor imaging using the FA index and the ADC index in specific ROIs drawn at four levels along the reconstructed CSTs: the posterior limbs of the internal capsule, the mid-portion of the cerebral peduncle, the anterior part of the pons, and the upper medulla oblongata (Figure 1). To include only the region of CST, the typical ROI size was set between 2 and 9 voxels. ROIs were specified on axial slices and symmetric on ipsilateral and contralateral sides at the same levels. For
Figure 2. Mean Fractional anisotropy (FA) value of lesion insulted regions of interest was significantly lower than contralateral sides (P < 0.05). Mean FA value of ipsilateral sides is lower than contralateral sides on caudal direction, however, is not statistically significant (P ¼ 0.23).There was no significant difference of mean FA between ipsilateral sides and contralateral sides on rostral direction. Although tendentially greater mean Apparent diffusion coefficient (ADC) values were seen in ipsilateral sides compared with the contralateral sides, they were not statistically significant on all levels.
WORLD NEUROSURGERY 83 [6]: 1006-1014, JUNE 2015
www.WORLDNEUROSURGERY.org
1009
ORIGINAL ARTICLE YUQIANG YAO ET AL.
QUANTIFICATION OF CST WITH DTI IN BRAINSTEM SURGERY
Table 3. Quantitative Results of FAs and ADCs at 3 Different Levels in Relation to the Lesion FA Ipsi- vs. Contralateral
Caudal level
Correlation Analysis
Pre- vs. Postoperative
Ipsi
Contralateral
P Value
LI
p (LI vs MRC0)
p (LI vs MRC2)
Pre
Post
P Value
0.424 0.133
0.492 0.095
0.230
0.083 0.141
0.959
0.272
0.381 0.099
0.334 0.064
0.206
Lesion level
0.364 0.215
0.596 0.126
0.009
0.302 0.231
0.547
0.631
0.396 0.105
0.458 0.146
0.182
Rostral level
0.600 0.140
0.620 0.107
0.615
0.022 0.116
0.516
0.633
0.614 0.134
0.636 0.142
0.493
FA, fractional anisotropy; ADC, apparent diffusion coefficient; LI, lateral index ¼ (ipsi-contra)/(ipsi þ contra); MRC, Medical Research Council.
intersubject averaging, the ROIs sizes were similar in all subjects and on both sides at the same levels. FA and ADC were calculated for each ROI. The symmetry of FA and ADC between ipsi- and contralateral ROIs was determined by calculating the Laterality Index (LI) ¼ (ipsi contra) / (ipsi þ contra). Negative values indicate a smaller parameter on the ipsilateral ROI, a 0 value indicates equal parameters on both sides, and positive values indicate a smaller parameter on the contralateral ROI. Data Processing and Statistics Statistical analysis included calculations of sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of WMT, with neurologic findings accepted as the gold standard. Because the number of patients in this study is limited, the outcomes from WMT and neurologic examinations were dichotomized as normal/abnormal to calculate sensitivity, specificity, PPV, and NPV value. For WMT, deviated, deformed, and interrupted CSTs (Groups IIeIV) were defined as abnormal versus normal. For grading the neurologic status, patients with motor deficit (Grade 0e4) were defined as abnormal versus normal (patients without motor deficit [Grade 5]). In our quantitative evaluation we divided all ROIs into lesion level, caudal level, and rostral level to lesion. We compared FA/ ADC differences between ipsilateral sides and contralateral sides at the various levels using paired t-test. We also compared FA/ADC differences between preoperative and postoperative DTI at the different levels using paired t-test. We used Spearman’s test to analyze correlation of FA/ADC symmetry (LI) with motor scales. The threshold for statistical significance was set at P < 0.05. RESULTS All patients showed either improvement of their symptoms or had stable neurological status without any additional complications. The 14 patients’ contralateral motor functions at 3 time points are shown in Table 1. Morphologic Study Preoperatively, 4 normal, 1 deviated, 3 deformed, and 6 interrupted CSTs were found on WMT and FA maps of the 14 patients. All morphologically normal CSTs (4 patients) corresponded to normal motor function on neurologic examination. Three patients having presurgically abnormal CSTs on WMT (1 deviated, 1
1010
www.SCIENCEDIRECT.com
deformed, and 1 interrupted) also showed normal motor function both on the presurgical (MRC0) and follow-up (MRC2) neurologic examinations. Seven patients with deformed (n ¼ 2) and interrupted (n ¼ 5) CSTs had MRC Grade 4 or MRC Grade 3 on presurgical neurologic examinations (MRC0) on the contralateral sides. Five of these (2 deformed, 3 interrupted) recovered to normal muscle strength on follow-up evaluation (MRC2), whereas the other 2 with interrupted CSTs showed persistent MRC Grade 4 on follow-up evaluation (MRC2). The WMT findings for these 14 patients are summarized in Table 1. The sensitivity, specificity, PPV, and the NPV of WMT on preoperative MRC were 100%, 57.1%, 70%, and 100%, respectively, and the corresponding values of WMT on follow-up MRC were 100%, 33.3%, 20%, and 100%, respectively (Table 2). Of 7 patients with both pre- and postoperative DTI data, 1 patient (case 6) had normal CST and neurological function in both pre- and postsurgical WMT; 4 patients showed improved morphology in postoperative WMT compared with preoperative WMT and also had normal muscle strength on the follow-up evaluation (Figure 1 shows an illustrated case). The other 2 patients showed severely interrupted CSTs both in pre- and postoperative WMT, with one patient recovering to normal muscle strength on follow-up evaluation (MRC2) and the other one showing persistent grade 4 muscle strength (MRC2).
Quantitative Study The mean FA of ROIs at the level and side of the lesions were significantly lower than the contralateral ones at the same level (paired t-test, P ¼ 0.009). Caudal to the lesion level the mean FA ipsilesional was lower than the uninvolved side but didn’t reach statistical significance (paired t-test, P ¼ 0.23). There was no significant difference of mean FA between contralesional and ipsilesional sides rostral to the lesion level (Figure 2). Mean ADC showed no significant differences between ipsilateral and contralateral sides at any level (Figure 2). All quantitative results are shown in Table 3. Correlation analyses performed between LI of FA/ADC at the 3 levels and simultaneous motor function (MRC0) and long-term follow up motor function (MRC2) using Spearman’s test, showed no correlation between the LI of FA or ADC in these 3 levels and the corresponding preoperative motor function (MRC0) and long-term follow-up motor function (MRC2).
WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2015.01.045
ORIGINAL ARTICLE YUQIANG YAO ET AL.
QUANTIFICATION OF CST WITH DTI IN BRAINSTEM SURGERY
Table 3. Continued ADC Ipsi- vs. Contralateral
Correlation Analysis
Pre- vs. Postoperative
Ipsi
Contra
P Value
LI
p (LI vs MRC0)
p (LI vs MRC2)
Pre
Post
P Value
1.113 0.262
1.061 0.303
0.705
0.028 0.133
0.089
0.725
1.201 0.231
1.077 0.177
0.495
0.996 0.354
0.823 0.106
0.155
0.087 0.192
0.307
0.324
1.007 0.298
0.815 0.239
0.028
0.816 0.153
0.791 0.142
0.593
0.015 0.072
0.458
0.119
0.835 0.166
0.795 0.146
0.552
There was a slightly increasing trend of FA (paired t-test, P ¼ 0.182) and a decreasing trend of ADC (paired t-test, P ¼ 0.028) of ROIs at the level of the lesions after operation as compared with the preoperative ones (Figure 3). There were no significant differences of FA/ADC values between pre- and postoperative DTI at the caudal and rostral levels.
DISCUSSION DTI-WMT is currently the only noninvasive method that can visualize the human white matter tracts in vivo. This technique has a proven value for planning a safer surgical approach but also a
Figure 3. The fractional anisotropy/apparent diffusion coefficient (FA/ADC) changes of lesion levels after operation comparing with preoperative evaluation. There was an increasing trend of FA (paired t-test, P ¼ 0.182) and a decreasing trend of ADC (paired t-test, P ¼ 0.028) of lesion insulted regions of interest after operation compared with preoperative imaging.
WORLD NEUROSURGERY 83 [6]: 1006-1014, JUNE 2015
prospective potential for predicting postoperative motor function. To date, this has been reported mostly in supratentorial tumor surgery (18, 36). Previous studies have reported that for supratentorial lesions, preoperative CST involvement may be a predictor of motor deficit and that postoperative improvement, or normalization, of the CST was a predictor of clinical improvement (3, 4, 9). To our knowledge, only a few published series have reported CST alteration using DTI-WMT in patients with brainstem lesions especially cavernomas. Chen et al. (5) studied the ability of DTI-WMT to visualize white matter tract involvement before and after surgery in 10 patients with brainstem lesions, of which 7 were BSC. Only one patient with deviating and partially interrupting CST showed a motor deficit preoperatively. Four months postoperatively, the motor deficit completely resolved and the CST normalized on DTI-WMT. The other 6 patients with normal CST morphology showed normal motor function. Kovanlikaya et al. (16) reported a series of 14 patients with brainstem lesions, of which 6 were BSCs. Examining the CSTs of the BSC patients, they found 7 normal CSTs with normal motor function, 5 abnormal CSTs, 2 associated with normal motor function and 3 with preoperative motor deficit (2 deformed CSTs and 1 interrupted CST). All 3 patients with preoperative motor deficits showed improved motor performance after surgery (followup 2e8 months). The patient with interrupted CST had normalization of the DTI findings whereas 2 deformed tracts did not change their appearance. McLaughlin and Kelly (20) presented a case supporting the use of DTI as predictor of motor recovery after BSC evacuation. The patient had interrupted CST preoperatively with mild right pronator drift and, 2-months postoperatively, DTI showed normal and symmetric CST that correlated to grossly normal motor function. Our study included 14 BSC patients with preoperative DTI. We were able to show similar results as in previous reports. The negative predictive value of preoperative DTI-WMT was very high (100%) both for MRC0 and MRC2, indicating that normal appearance of the CSTs is a highly predictive indicator of tract preservation. Compared with the high negative predictive value (100%), PPV was somewhat limited, especially on the follow-up evaluation (only 20%). On presurgical evaluation 7 of 10 abnormal CSTs showed abnormal motor function (positive predictive value 70%). On the follow-up evaluation, 5 of these 7 patients fully recovered to MCR Grade 5, which points out that presurgical DTI-WMT disruption should not necessarily be
www.WORLDNEUROSURGERY.org
1011
ORIGINAL ARTICLE YUQIANG YAO ET AL.
QUANTIFICATION OF CST WITH DTI IN BRAINSTEM SURGERY
considered as an indicator of definite tract loss and worst prognosis. Hence, caution is warranted in cases with abnormal CSTs with only mild motor deficit in preoperative planning. In fact, despite complete interruption of the CSTs at the lesion injured level, 4 of 6 patients recovered to normal motor function, confirming the integrity of their ipsilateral CSTs. We agree with Kovanlikaya et al.’s opinion that the mismatch between DTI findings and motor functions are considered to relate to lesion characteristics and especially to the presence of artifacts, including interfering hematoma, edema and vascular pulsation (16). This is particularly true for cavernomas which, despite their well-defined margins, are invariably hemorrhagic lesions, associated with hemosiderin deposition in the adjacent brain parenchyma causing pronounced magnetic susceptibility artifacts and thus signal loss in MRI. Comparison of pre- and postoperative DTI results also indicated that selection of the appropriate surgical approach should result to the benefit of the patients’ function. In our group, 4 patients showed improved morphology in postoperative DTI compared with preoperative DTI and their motor function all recovered to normal status postoperatively. Quantitative analysis of diffusion parameters at the level of the lesions showed similar results to our morphologic evaluation. Although there was significant difference between ipsilateral and contralateral FAs at lesion level, correlation analysis showed that lower FA values do not necessarily correlate with unfavorable prognosis. The postoperative trend of increasing FA (P > 0.05) and decreasing ADC (P < 0.05) values in ROIs at the level of the cavernomas paralleled the morphological improvement of CSTs compared with the preoperative DTI. It is well known that abnormal brain parenchyma attributable to edema or gliosis usually shows decreased FA and increased ADC (13, 30). Therefore, a trend to normalization of the FA (i.e., increasing values) and ADC (i.e., decreasing values) postoperatively may reflect recovery of infiltrated fasciculi. However, it is important to mention that the inherent sensitivity of MRI to susceptible artifacts arising from the presence of hemorrhagic products, as discussed in the previous paragraph, is an important limiting factor in the quantification of diffusion parameters. In our study, we additionally investigated potential anterograde and retrograde FA changes of CST and their relationship with prognosis. Degeneration of white matter fibers at a distance from a primary lesion is a common finding in many diseases of the central nervous system. These secondary white matter degeneration is classically divided into retrograde and anterograde (or Wallerian) degeneration. From the primary lesion, retrograde degeneration proceeds proximally toward the cell body, while anterograde degeneration proceeds distally toward the axon’s terminals. The diffusion parameters of anterograde and retrograde degeneration following trauma, stroke, demyelinating disease, have been demonstrated in several previous studies (1, 27, 33). However, very few reports discussed the anterograde or retrograde FA/ADC changes, reflecting secondary axonal degeneration, associated with brainstem lesions, especially BSCs. By analyzing anterograde and retrograde diffusion parameters of 10 morphologically abnormal CSTs, we found lower FA of
1012
www.SCIENCEDIRECT.com
ipsilateral CSTs compared with contralateral CSTs in the caudal direction that, however, did not reach statistical significance. There were no significant differences regarding FAs and ADCs in the rostral direction on both sides. The relatively comparable diffusion parameters of CSTs in both anterograde and retrograde directions in relation to the lesion level may be linked to the relatively good prognosis. In this group, only 2 of 10 patients showed MRC grade 4 on follow-up period, the motor grade of the other 8 patients either improved to or remained at grade 5. It has been previously demonstrated that secondary degeneration of CSTs from cerebral or spinal primary insults at chronic stage can cause anterograde or retrograde alterations of diffusion parameters, e.g., FA decreased and/or ADC increased, which may correlate to changes in corresponding motor functions (14, 24). Thus, the good postoperatively clinical motor function both ipsi- and contralaterally observed in our patient cohort may be reflected in the lack of significant differences of diffusion parameters. We herein present the first report discussing morphologic characteristics and anterograde/retrograde changes of diffusion parameters along the CSTs of patients with BSCs. Morphologically interrupted CSTs and decreased FA correlate well at the BSC lesion level. Morphologic characteristics and diffusion parameter changes of CSTs, however, do not allow a strong prediction of patients’ motor outcome. The relatively unchanged anterograde and retrograde diffusion parameters of CSTs confirmed the integrity of the corresponding CSTs as reflected in good motor function, although we didn’t get positive correlation between diffusion parameters and clinical manifestations. A major limitation of the present study is the limited number of subjects and their relatively homogeneous clinical picture. Most patients returned to or remained at normal motor function postoperatively causing difficult the detection of potential subtle quantitative diffusion differences. In this series, no patients had a worsened postoperative outcome, which renders difficult to conclude with any confidence that normal CST morphology predicts favorable postoperative outcome when all patients remained the same/improved clinically. However, these data represented our clinical experience from a series of consecutive cases and patients haven’t been selected on the basis of their clinical outcome. Studying patients with persisting, or worsened, neurologic/motor deficits would probably allow more accurate information on the prognostic significance of morphologic and quantitative DTI changes. Nevertheless, the trends found in our study, i.e., high sensitivity and NPV of pre-operative DTI-WMT, are strong indicators for good outcome of pre-peratively detected intact WMT. Despite the standardized procedures and imaging protocols applied in this study, its retrospective nature is another limitation related to limited availability of postoperative DTI information in only half of the patients. However, the type of pathology we addressed is relatively rare and handled in centers with increased expertise in brainstem surgery. Our study reports on a large number of BSCs with DTI analysis collected during a period of 2 years. A larger number of subjects and more diversity in clinical manifestation would certainly lead to more reliable conclusions and need to be
WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2015.01.045
ORIGINAL ARTICLE YUQIANG YAO ET AL.
QUANTIFICATION OF CST WITH DTI IN BRAINSTEM SURGERY
addressed in future prospective, potentially multicenter studies or meta-analysis reports.
changes at lesion level cannot predict poor prognosis. Caudal and rostral diffusion parameters can provide more information of the integrity of CSTs compared with morphologic study alone.
CONCLUSIONS Normal CST morphology generated from WMT using DTI predicts a favorable postoperative outcome in patients with BSC. Interrupted CSTs and decreased FA values correlate well within BSC lesion level. However, morphologic characteristics and diffusion parameter
REFERENCES 1. Ahn YH, Kim SH, Han BS, Kim OL, Ahn SH, Cho YW, Kwon YH, Jang SH: Focal lesions of the corticospinal tract demonstrated by diffusion tensor imaging in patients with diffuse axonal injury. NeuroRehabilitation 21:239-243, 2006. 2. Basser PJ, Pajevic S, Pierpaoli C, Duda J, Aldroubi A: In vivo fiber tractography using DTMRI data. Magn Reson Med 44:625-632, 2000. 3. Bello L, Gambini A, Castellano A, Carrabba G, Acerbi F, Fava E, Giussani C, Cadioli M, Blasi V, Casarotti A, Papagno C, Gupta AK, Gaini S, Scotti G, Falini A: Motor and language DTI Fiber Tracking combined with intraoperative subcortical mapping for surgical removal of gliomas. Neuroimage 39:369-382, 2008. 4. Castellano A, Bello L, Michelozzi C, Gallucci M, Fava E, Iadanza A, Riva M, Casaceli G, Falini A: Role of diffusion tensor magnetic resonance tractography in predicting the extent of resection in glioma surgery. Neuro Oncol 14:192-202, 2012. 5. Chen X, Weigel D, Ganslandt O, Buchfelder M, Nimsky C: Diffusion tensor imaging and white matter tractography in patients with brainstem lesions. Acta Neurochir (Wien) 149:1117-1131; discussion 1131, 2007. 6. Ciccarelli O, Parker GJ, Toosy AT, WheelerKingshott CA, Barker GJ, Boulby PA, Miller DH, Thompson AJ: From diffusion tractography to quantitative white matter tract measures: a reproducibility study. Neuroimage 18:348-359, 2003. 7. Ciurea AV, Nastase C, Tascu A, Brehar FM: Lethal recurrent hemorrhages of a brainstem cavernoma. Neurosurg Rev 30:259-262; discussion 262, 2007. 8. Del Curling O Jr, Kelly DL Jr, Elster AD, Craven TE: An analysis of the natural history of cavernous angiomas. J Neurosurg 75:702-708, 1991. 9. Ellis MJ, Rutka JT, Kulkarni AV, Dirks PB, Widjaja E: Corticospinal tract mapping in children with ruptured arteriovenous malformations using functionally guided diffusion-tensor imaging. J Neurosurg Pediatr 9:505-510, 2012. 10. Field AS, Alexander AL, Wu YC, Hasan KM, Witwer B, Badie B: Diffusion tensor eigenvector directional color imaging patterns in the evaluation of cerebral white matter tracts altered by tumor. J Magn Reson Imaging 20:555-562, 2004.
ACKNOWLEDGMENTS The authors are very grateful to all patients for participating in this study and thank Dr. Uzeyir Ahmadli for neuroradiological consultation in fiber tracking.
11. Fritschi JA, Reulen HJ, Spetzler RF, Zabramski JM: Cavernous malformations of the brain stem. A review of 139 cases. Acta Neurochir (Wien) 130: 35-46, 1994. 12. Glenn OA, Henry RG, Berman JI, Chang PC, Miller SP, Vigneron DB, Barkovich AJ: DTI-based three-dimensional tractography detects differences in the pyramidal tracts of infants and children with congenital hemiparesis. J Magn Reson Imaging 18: 641-648, 2003. 13. Gulani V, Webb AG, Duncan ID, Lauterbur PC: Apparent diffusion tensor measurements in myelin-deficient rat spinal cords. Magn Reson Med 45:191-195, 2001. 14. Guleria S, Gupta RK, Saksena S, Chandra A, Srivastava RN, Husain M, Rathore R, Narayana PA: Retrograde Wallerian degeneration of cranial corticospinal tracts in cervical spinal cord injury patients using diffusion tensor imaging. J Neurosci Res 86:2271-2280, 2008. 15. Guo AC, Cummings TJ, Dash RC, Provenzale JM: Lymphomas and high-grade astrocytomas: comparison of water diffusibility and histologic characteristics. Radiology 224:177-183, 2002. 16. Kovanlikaya I, Firat Z, Kovanlikaya A, Ulug AM, Cihangiroglu MM, John M, Bingol CA, Ture U: Assessment of the corticospinal tract alterations before and after resection of brainstem lesions using Diffusion Tensor Imaging (DTI) and tractography at 3T. Eur J Radiol 77:383-391, 2011. 17. Lazar M, Alexander AL, Thottakara PJ, Badie B, Field AS: White matter reorganization after surgical resection of brain tumors and vascular malformations. AJNR Am J Neuroradiol 27:1258-1271, 2006. 18. Lo CY, Chao YP, Chou KH, Guo WY, Su JL, Lin CP: DTI-based virtual reality system for neurosurgery. Conf Proc IEEE Eng Med Biol Soc 2007:1326-1329, 2007. 19. Ma L, Song ZJ: Differentiation between low-grade and high-grade glioma using combined diffusion tensor imaging metrics. Clin Neurol Neurosurg 115:2489-2495, 2013. 20. McLaughlin N, Kelly DF: Corticospinal tractography as a prognosticator for motor improvement after brainstem cavernoma resection. Br J Neurosurg 27:108-110, 2013. 21. Mori S, Crain BJ, Chacko VP, van Zijl PC: Threedimensional tracking of axonal projections in the brain by magnetic resonance imaging. Ann Neurol 45:265-269, 1999.
WORLD NEUROSURGERY 83 [6]: 1006-1014, JUNE 2015
22. Nguyen TH, Yoshida M, Stievenart JL, IbaZizen MT, Bellinger L, Abanou A, Kitahara K, Cabanis EA: MR tractography with diffusion tensor imaging in clinical routine. Neuroradiology 47:334-343, 2005. 23. Paternostro-Sluga T, Grim-Stieger M, Posch M, Schuhfried O, Vacariu G, Mittermaier C, Bittner C, Fialka-Moser V: Reliability and validity of the Medical Research Council (MRC) scale and a modified scale for testing muscle strength in patients with radial palsy. J Rehabil Med 40:665-671, 2008. 24. Pierpaoli C, Barnett A, Pajevic S, Chen R, Penix LR, Virta A, Basser P: Water diffusion changes in Wallerian degeneration and their dependence on white matter architecture. Neuroimage 13:1174-1185, 2001. 25. Porter PJ, Willinsky RA, Harper W, Wallace MC: Cerebral cavernous malformations: natural history and prognosis after clinical deterioration with or without hemorrhage. J Neurosurg 87:190-197, 1997. 26. Porter RW, Detwiler PW, Spetzler RF, Lawton MT, Baskin JJ, Derksen PT, Zabramski JM: Cavernous malformations of the brainstem: experience with 100 patients. J Neurosurg 90:50-58, 1999. 27. Reich DS, Zackowski KM, Gordon-Lipkin EM, Smith SA, Chodkowski BA, Cutter GR, Calabresi PA: Corticospinal tract abnormalities are associated with weakness in multiple sclerosis. AJNR Am J Neuroradiol 29:333-339, 2008. 28. Robinson JR, Awad IA, Little JR: Natural history of the cavernous angioma. J Neurosurg 75: 709-714, 1991. 29. Sinha S, Bastin ME, Whittle IR, Wardlaw JM: Diffusion tensor MR imaging of high-grade cerebral gliomas. AJNR Am J Neuroradiol 23:520-527, 2002. 30. Thomas B, Eyssen M, Peeters R, Molenaers G, Van Hecke P, De Cock P, Sunaert S: Quantitative diffusion tensor imaging in cerebral palsy due to periventricular white matter injury. Brain 128: 2562-2577, 2005. 31. Ulrich NH, Ahmadli U, Woernle CM, Alzarhani YA, Bertalanffy H, Kollias SS: Diffusion tensor imaging for anatomical localization of cranial nerves and cranial nerve nuclei in pontine lesions: initial experiences with 3T-MRI. J Clin Neurosci 21:1924-1927, 2014. 32. Ulrich NH, Kockro RA, Bellut D, Amaxopoulou C, Bozinov O, Burkhardt JK, Sarnthein J, Kollias SS, Bertalanffy H: Brainstem cavernoma surgery with
www.WORLDNEUROSURGERY.org
1013
ORIGINAL ARTICLE YUQIANG YAO ET AL.
QUANTIFICATION OF CST WITH DTI IN BRAINSTEM SURGERY
the support of pre- and postoperative diffusion tensor imaging: initial experiences and clinical course of 23 patients. Neurosurg Rev 37:481-491; discussion 92, 2014. 33. Vargas P, Gaudron M, Valabregue R, Bertasi E, Humbert F, Lehericy S, Samson Y, Rosso C: Assessment of corticospinal tract (CST) damage in acute stroke patients: comparison of tractspecific analysis versus segmentation of a CST template. J Magn Reson Imaging 37:836-845, 2013. 34. Wang CC, Liu A, Zhang JT, Sun B, Zhao YL: Surgical management of brain-stem cavernous malformations: report of 137 cases. Surg Neurol 59:444-454; discussion 454, 2003.
35. Witwer BP, Moftakhar R, Hasan KM, Deshmukh P, Haughton V, Field A, Arfanakis K, Noyes J, Moritz CH, Meyerand ME, Rowley HA, Alexander AL, Badie B: Diffusion-tensor imaging of white matter tracts in patients with cerebral neoplasm. J Neurosurg 97:568-575, 2002.
commercial or financial relationships that could be construed as a potential conflict of interest. Yuqiang Yao and Nils H. Ulrich are coefirst authors. Received 6 July 2014; accepted 20 January 2015 Citation: World Neurosurg. (2015) 83, 6:1006-1014. http://dx.doi.org/10.1016/j.wneu.2015.01.045 Journal homepage: www.WORLDNEUROSURGERY.org
36. Yu CS, Li KC, Xuan Y, Ji XM, Qin W: Diffusion tensor tractography in patients with cerebral tumors: a helpful technique for neurosurgical planning and postoperative assessment. Eur J Radiol 56:197-204, 2005.
Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2015 Elsevier Inc. All rights reserved.
Conflict of interest statement: The authors declare that the article content was composed in the absence of any
WORLD FEDERATION OF NEUROSURGICAL SOCIETIES
Mission Statement
To facilitate the personal association of neurological surgeons throughout the world To aid in the exchange and dissemination of knowledge and ideas in the field of neurological surgery To encourage research in neurological surgery and allied sciences To address issues of neurosurgical demography To address issues of Public Health To implement, improve and promote the standards of neurosurgical care and training worldwide
1014
www.SCIENCEDIRECT.com
WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2015.01.045
Quantification of Corticospinal Tracts with Diffusion Tensor Imaging in Brainstem Surgery: Prognostic Value in 14 Consecutive Cases at 3T Magnetic Resonance Imaging Yuqiang Yao1,2, Nils H. Ulrich1,3, Roman Guggenberger1, Yahea A. Alzarhani1, Helmut Bertalanffy4, Spyros S. Kollias1
OBJECTIVE: In this study, we investigated the potential prognostic role of morphologic and quantitative diffusion tensor imaging (DTI) in patients with brainstem cavernoma (BSC) in terms of postoperative outcome.
-
METHODS: In this retrospective study of 14 brainstem cavernoma patients, we analyzed pre- and postoperative DTI data. White matter tractography of corticospinal tracts (CSTs) was performed with the Fiber Assignment by Continuous Tracking algorithm, and morphologic characteristics of CSTs were compared with clinically assessed motor strength. Fractional anisotropy (FA) and apparent diffusion coefficient (ADC) were measured in ipsi- and contralesional regions of interest at the lesion level, as well as levels caudal and rostral to the lesion. Correlation analysis was performed between lateral index (LI) of ipsi-/contralateral FA and ADC values and patients’ motor function.
-
RESULTS: Preoperatively, normal morphologic features of CSTs corresponded to normal motor function in 4 patients. The other 10 morphologically abnormal preoperative CSTs didn’t show corresponding motor impairment either in pre- or postoperative follow-up period. The sensitivity, specificity, positive predictive, and negative predictive values of white matter tractographic morphology on preoperative motor function were 100%, 57.1%, 70%, and 100%, respectively. The corresponding values on follow-up motor function were 100%, 33.3%, 20%, and 100%, respectively.
-
Key Words Apparent diffusion coefficient - Brainstem cavernoma - Corticospinal tract - Diffusion tensor imaging - Fractional anisotropy - White matter tractography -
Abbreviations and Acronyms BSC: Brainstem cavernoma CST: Corticospinal tract DTI: Diffusion tensor imaging FA: Fractional anisotropy LI: Lateral index MRC: Medical Research Council MRI: Magnetic resonance imaging NPV: Negative predictive value
1006
www.SCIENCEDIRECT.com
The mean FA at lesion level was significantly lower than the corresponding FA at the contralateral hemisphere (P [ 0.009). In areas caudal to the lesion the same trend yet without statistic significance was observed (P [ 0.23). There was no significant laterality difference of mean FAs rostral to the level of the lesion. No correlation in LI of FA or ADC in the 3 anatomic levels with simultaneous and long-term follow up motor function was observed. Restoration of the morphology of the affected CST postoperatively was associated with a trend for decreasing ADC compared with the preoperative measurements. CONCLUSION: Intact CST morphology in DTI predicts a favorable postoperative outcome in patients with BSC. Interrupted CSTs and decreased FA values correlate well within BSC lesion level; nevertheless, morphologic characteristics and diffusion parameter changes at lesion level cannot predict poor prognosis. Caudal and rostral diffusion parameters can provide more information of the integrity of CSTs compared with morphologic study alone.
-
INTRODUCTION
I
n recent years, the incidence of brain cavernomas has increased as the result of advances in neuroimaging, particularly the widespread use of magnetic resonance imaging (MRI)
PPV: Positive predictive value ROI: Region of interest WMT: White matter tractography From the 1Department of Neuroradiology, University Hospital, University of Zurich, Zurich, Switzerland; 2Department of Neurosurgery, Beijing Jishuitan Hospital, Beijing, China; 3 Department of Neurosurgery, University Hospital, University of Zurich, Zurich, Switzerland; and 4Department of Neurosurgery, International Neuroscience Institute, Hannover, Germany To whom correspondence should be addressed: Yuqiang Yao, M.D. [E-mail: [email protected]] Citation: World Neurosurg. (2015) 83, 6:1006-1014. http://dx.doi.org/10.1016/j.wneu.2015.01.045 Journal homepage: www.WORLDNEUROSURGERY.org Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2015 Elsevier Inc. All rights reserved.
WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2015.01.045
ORIGINAL ARTICLE YUQIANG YAO ET AL.
QUANTIFICATION OF CST WITH DTI IN BRAINSTEM SURGERY
Table 1. General Information and Morphologic Changes of CST Motor Scale*
Morphologic Changes Patient No.
Age (Years)/Sex
Lesion Level
Side
Approach
Preoperative
Postoperative
0
1
2
1
25/f
Pons
R
2
39/f
Pons
L
Suboccipital (midline)
Interrupted
Interrupted
3
4
5
Retromastoidal
Deformed
Deviated
5
5
5
3
32/f
Pons þ mesencephalon
R
Subtemporal
Interrupted
Deformed
4
4
5
4
44/f
Pons
5
22/f
Pons
L
Suboccipital (midline)
Deformed
Deviated
4
5
5
R
Suboccipital (midline)
Normal
Normal
5
5
5
6
52/f
Pons
M
Suboccipital (midline)
Normal
Normal
5
5
5
7
19/f
Medulla oblongata
L
Retromastoidal
Interrupted
Interrupted
4
4
4
8
6/m
Pons
R
Retromastoidal
Interrupted
N/A
4
4
5
Interrupted
N/A
4
4
4
Normal
N/A
5
5
5
9
50/m
Pons
R
Suboccipital (midline)
10
30/m
Pons
L
Retromastoidal
11
26/f
Pons
R
Suboccipital (midline)
Interrupted
N/A
5
5
5
12
51/f
Mesencephalon
L
Supracerebellar-infratentorial
Deformed
N/A
4
5
5
13
63/f
Medulla oblongata
L
Suboccipital (midline)
Deviated
N/A
5
5
5
14
45/f
Pons
L
Retromastoidal
Normal
N/A
5
5
5
CST, corticospinal track; f, female; R, right, L, left; m, male; N/A, not available; MRC, Medical Research Council. *MRC scale 0: preoperative; 1: postoperative; 2: follow-up.
in clinical practice. The prevalence of intracranial cavernomas ranges from 0.4% to 0.9% (8, 28), with the ones being localized in the brainstem representing 8%e22% (11). This subgroup of cavernomas has a substantially greater tendency for bleeding (up to 30%) (25), which is more likely to result in severe neurologic deficits (7, 11, 25) and moreover has a greater incidence of recurrent hemorrhage than those in other locations (26, 34). Surgical extirpation is the major treatment option for the symptomatic brainstem cavernomas (BSCs). Because of their location deep within a relatively small anatomic region, highly compact in functionally important white matter tracts and grey matter nuclei, BSCs to date remain a major challenge for neurosurgeons. They also represent a challenge for a functionally relevant neuroradiologic description because of the poor differentiation of individual anatomical structures on conventional MRI. Diffusion tensor imaging (DTI) is a recently introduced technique that allows evaluation of white matter integrity by virtue of its ability to visualize water diffusion characteristics. White matter tractography (WMT), also known as “fiber tracking,” allows 3-dimensional visualization of white matter tracts and accurate description of their architecture and integrity in the brainstem (2, 12, 21). The validity and reliability of WMT via the use of DTI for brainstem lesions in preoperative planning have been well demonstrated in previous studies (6, 22). Recent studies in patients with brain tumors have used quantitative DTI parameters, namely apparent diffusion coefficient (ADC) and fractional anisotropy (FA), for histologic prediction of the tumor (15, 19, 29)
WORLD NEUROSURGERY 83 [6]: 1006-1014, JUNE 2015
and to characterize the effects of the tumor on the surrounding white matter tracts (5, 10). On the basis of these studies, several institutions have adopted DTI as an increasingly important component of preoperative planning in patients with supratentorial tumors (35) and brainstem lesions (5, 31, 32). Currently, there are no clinical or imaging findings that can reliably predict postoperative motor function in patients with BSCs. Our hypothesis is that the structural morphology and quantitative measures of DTI in these patients might provide prospectively prediction of postoperative clinical findings and outcome and, thus, hold certain prognostic value for more informed therapy decision making. To test this hypothesis, we investigated the prognostic value of morphologic and quantitative imaging parameters generated from WMT of corticospinal tract (CST) using pre- and postoperative DTI in 14 patients with BSCs. MATERIALS AND METHODS DTI data were analyzed in a retrospective study of 14 patients with BSC, which were operated between January of 2008 and June of 2010 at the department of neurosurgery. Each diagnosis was confirmed by histology. Patients’ details are listed in Table 1. The mean age of the patients was 36 years (range, 6e63 years). Most of the lesions located in the pons (10/14, 71.4%); 1 lesion in the mesencephalon (7.1%); 2 lesions in the medulla oblongata (14.2%); and 1 lesion involved both pons and mesencephalon (7.1%). Six lesions were mainly on the right side, 7 mainly on the left and one in the midline of the
www.WORLDNEUROSURGERY.org
1007
ORIGINAL ARTICLE YUQIANG YAO ET AL.
QUANTIFICATION OF CST WITH DTI IN BRAINSTEM SURGERY
Figure 1. Illustration of a patient with a BSC in the right cerebral peduncle (CP) and anterior part of the pons (PN) (Patient 3). (AeB) Pre-operative structural magnetic resonance (MR) imaging. (CeE) Preoperative FA map (blue: contralateral reconstructed corticospinal tract [CST]; red and pink: ipsilateral reconstructed CST); (FeG): postoperative structural MR, (HeJ): postoperative FA map (blue: contralateral reconstructed CST; yellow and red: ipsilateral reconstructed CST). The respective colors of fiber tracts were chosen to distinguish ipsilateral from contralateral side and pre-from postoperative situations. Fiber tracts generated from DTI data and consecutively calculated FA maps nicely followed the normal anatomy of
brainstem. All patients underwent examination via MRI, including DTI, as part of the routine preoperative protocol at the department of neuroradiology. Seven patients also had DTI acquired during their postoperative follow-up examinations. All subjects underwent microsurgical resection of brainstem cavernoma. The surgical approach was determined by the size, location of the lesion, and the spatial relationship to critical structures such as the corticospinal and sensory tracts. For lesions located in lower pons and medulla oblongata, the midline suboccipital approach was used for medially located lesions; the retromastoidal approach was used for laterally located lesions. For lesions located in upper pons and mesencephalon, the subtemporal approach was used for the anterior and lateral lesions and supracerebellar-infratentorial approach for posterior ones. The fasciculi were protected carefully under intraoperative DTI navigation. Multimodality intraoperative monitoring, such as motor-evoked potentials, somatosensory-evoked potential, acoustic-evoked potentials, neuronavigation, and mapping of the rhomboid fossa also was used to support the surgical intervention.
1008
www.SCIENCEDIRECT.com
CST. Preoperative FA map showed severely interrupted CST of the lesion side; postoperative FA map showed remarkable improvement of lesion insulted CST. The MRC grades were 4, 4, and 5 at preoperative, postoperative, and follow-up time points. This figure also illustrates how ROIs were drawn at 4 different anatomic levels in rostrocaudal direction (The color of region of interests [ROIs] were randomly assigned just for distinction with each other): the posterior limbs of the internal capsule (PL), the mid-portion of the CP, the anterior part of PN, and upper medullaoblongata (UM).
Each patient’s motor grade was assessed using the Medical Research Council (MRC) scale of 0e5 (23) at 3 time points: preoperatively (MRC0), postoperatively (MRC1), and in 3e6 months’ follow-up (MRC2). Conventional MRI and DTI Acquisition MRI was performed at a 3-Tesla whole-body magnetic resonance system (PhilipsAchieva, Best, Netherlands), equipped with 80 mT/ m/ms gradient coils and an 8-element receive head coil array (MRI Devices Corp., Waukesha, WI). T2-weighted images (echo time 85ms, repetition time 4900 ms, matrix size 320 256, field of view 220 mm, slice thickness 4 mm), as well as T1-weighted images (echo time 10 ms, repetition time 400 ms, matrix size 256 192, field of view 220 mm, slice thickness 4 mm) before and after intravenous contrast administration were acquired. DTIs were acquired, as part of the routine preoperative protocol via the use of a single-shot, spin-echo, diffusion-weighted, echo planar imaging sequence with the following parameters: in-plane matrix ¼ 96 96, reconstructed to 128 128, 60 contiguous slices
WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2015.01.045
ORIGINAL ARTICLE YUQIANG YAO ET AL.
QUANTIFICATION OF CST WITH DTI IN BRAINSTEM SURGERY
Table 2. Accuracy of WMT TP
FN
FP
TN
Se %
Sp %
PPV %
NPV %
MRC0
7
0
3
4
100
57.1
70.0
100
MRC2
2
0
8
4
100
33.3
20.0
100
WMT, white matter tractography; TP, true positive; FN, false negative; FP, false positive; TN, true negative; Se, sensitivity; Sp, specificity; PPV, positive predictive value; NPV, negative predictive value; MRC, Medical Research Council.
of 2.1-mm thickness, TE ¼ 50 ms, number of signal averages ¼ 2, and 60% partial k-space acquisition. Diffusion weighting with a maximal b-factor of 1000 s/mm2 was carried out along 15 icosahedral directions complemented by one scan with b ¼ 0. WMT and Morphologic Study DTIs were processed off-line using commercially available software on an “Extended MR Working Station” (Philips Achieva). White matter fiber-tracking was performed by the use of the fiber assignment by continuous tracking algorithm with a fractional anisotropy of 0.15 and an angle of 27. The fiber tracking of CST was obtained by placing 3 different regions of interest (ROIs) along the pathway. All ROIs were identified by anatomical landmarks and color-coded FA maps in which red indicates fibers running along the right-left direction, green anterior-posterior, and blue inferior-superior. The first ROI was placed in the upper medulla, which is above the decussation of the pyramidal tracts; the second in the posterior limb of the internal capsule; and the third, in the upper part of the semioval center, adjacent to the primary motor cortex. Tracking of the CST was performed by
selecting only the fibers passing through all 3 ROIs. After the first reconstruction, some fibers, which by their anatomic distribution were obviously not belonging to the CSTs, were removed by additional operations (using exclusive ROIs). The reconstructed fibers were visually evaluated for similarity to reference atlas images and conformity to the above-mentioned anatomic landmarks. The CSTs were reconstructed in all subjects on both sides. We adapted the criteria developed by Lazar et al. (17) and revised by Kovanlikaya et al. (16) to classify fiber tract involvement into 4 groups: normal, deviated, deformed, or interrupted. The patterns of tract alteration were identified similar to Kovanlikaya’s description as follows: Group I: “normal” Group II: “deviated” if the tract was in abnormal location and/or direction due to mass effect of the lesion. Group III: “deformed” if there was a partial defect or interruption in parts of tract while the rest of it was identifiable on FA color maps or WMT. Group IV: “interrupted” if the tract was discontinuous or defective both on FA color maps and WMT. ROI Analysis and Quantitative Study We quantified diffusion tensor imaging using the FA index and the ADC index in specific ROIs drawn at four levels along the reconstructed CSTs: the posterior limbs of the internal capsule, the mid-portion of the cerebral peduncle, the anterior part of the pons, and the upper medulla oblongata (Figure 1). To include only the region of CST, the typical ROI size was set between 2 and 9 voxels. ROIs were specified on axial slices and symmetric on ipsilateral and contralateral sides at the same levels. For
Figure 2. Mean Fractional anisotropy (FA) value of lesion insulted regions of interest was significantly lower than contralateral sides (P < 0.05). Mean FA value of ipsilateral sides is lower than contralateral sides on caudal direction, however, is not statistically significant (P ¼ 0.23).There was no significant difference of mean FA between ipsilateral sides and contralateral sides on rostral direction. Although tendentially greater mean Apparent diffusion coefficient (ADC) values were seen in ipsilateral sides compared with the contralateral sides, they were not statistically significant on all levels.
WORLD NEUROSURGERY 83 [6]: 1006-1014, JUNE 2015
www.WORLDNEUROSURGERY.org
1009
ORIGINAL ARTICLE YUQIANG YAO ET AL.
QUANTIFICATION OF CST WITH DTI IN BRAINSTEM SURGERY
Table 3. Quantitative Results of FAs and ADCs at 3 Different Levels in Relation to the Lesion FA Ipsi- vs. Contralateral
Caudal level
Correlation Analysis
Pre- vs. Postoperative
Ipsi
Contralateral
P Value
LI
p (LI vs MRC0)
p (LI vs MRC2)
Pre
Post
P Value
0.424 0.133
0.492 0.095
0.230
0.083 0.141
0.959
0.272
0.381 0.099
0.334 0.064
0.206
Lesion level
0.364 0.215
0.596 0.126
0.009
0.302 0.231
0.547
0.631
0.396 0.105
0.458 0.146
0.182
Rostral level
0.600 0.140
0.620 0.107
0.615
0.022 0.116
0.516
0.633
0.614 0.134
0.636 0.142
0.493
FA, fractional anisotropy; ADC, apparent diffusion coefficient; LI, lateral index ¼ (ipsi-contra)/(ipsi þ contra); MRC, Medical Research Council.
intersubject averaging, the ROIs sizes were similar in all subjects and on both sides at the same levels. FA and ADC were calculated for each ROI. The symmetry of FA and ADC between ipsi- and contralateral ROIs was determined by calculating the Laterality Index (LI) ¼ (ipsi contra) / (ipsi þ contra). Negative values indicate a smaller parameter on the ipsilateral ROI, a 0 value indicates equal parameters on both sides, and positive values indicate a smaller parameter on the contralateral ROI. Data Processing and Statistics Statistical analysis included calculations of sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of WMT, with neurologic findings accepted as the gold standard. Because the number of patients in this study is limited, the outcomes from WMT and neurologic examinations were dichotomized as normal/abnormal to calculate sensitivity, specificity, PPV, and NPV value. For WMT, deviated, deformed, and interrupted CSTs (Groups IIeIV) were defined as abnormal versus normal. For grading the neurologic status, patients with motor deficit (Grade 0e4) were defined as abnormal versus normal (patients without motor deficit [Grade 5]). In our quantitative evaluation we divided all ROIs into lesion level, caudal level, and rostral level to lesion. We compared FA/ ADC differences between ipsilateral sides and contralateral sides at the various levels using paired t-test. We also compared FA/ADC differences between preoperative and postoperative DTI at the different levels using paired t-test. We used Spearman’s test to analyze correlation of FA/ADC symmetry (LI) with motor scales. The threshold for statistical significance was set at P < 0.05. RESULTS All patients showed either improvement of their symptoms or had stable neurological status without any additional complications. The 14 patients’ contralateral motor functions at 3 time points are shown in Table 1. Morphologic Study Preoperatively, 4 normal, 1 deviated, 3 deformed, and 6 interrupted CSTs were found on WMT and FA maps of the 14 patients. All morphologically normal CSTs (4 patients) corresponded to normal motor function on neurologic examination. Three patients having presurgically abnormal CSTs on WMT (1 deviated, 1
1010
www.SCIENCEDIRECT.com
deformed, and 1 interrupted) also showed normal motor function both on the presurgical (MRC0) and follow-up (MRC2) neurologic examinations. Seven patients with deformed (n ¼ 2) and interrupted (n ¼ 5) CSTs had MRC Grade 4 or MRC Grade 3 on presurgical neurologic examinations (MRC0) on the contralateral sides. Five of these (2 deformed, 3 interrupted) recovered to normal muscle strength on follow-up evaluation (MRC2), whereas the other 2 with interrupted CSTs showed persistent MRC Grade 4 on follow-up evaluation (MRC2). The WMT findings for these 14 patients are summarized in Table 1. The sensitivity, specificity, PPV, and the NPV of WMT on preoperative MRC were 100%, 57.1%, 70%, and 100%, respectively, and the corresponding values of WMT on follow-up MRC were 100%, 33.3%, 20%, and 100%, respectively (Table 2). Of 7 patients with both pre- and postoperative DTI data, 1 patient (case 6) had normal CST and neurological function in both pre- and postsurgical WMT; 4 patients showed improved morphology in postoperative WMT compared with preoperative WMT and also had normal muscle strength on the follow-up evaluation (Figure 1 shows an illustrated case). The other 2 patients showed severely interrupted CSTs both in pre- and postoperative WMT, with one patient recovering to normal muscle strength on follow-up evaluation (MRC2) and the other one showing persistent grade 4 muscle strength (MRC2).
Quantitative Study The mean FA of ROIs at the level and side of the lesions were significantly lower than the contralateral ones at the same level (paired t-test, P ¼ 0.009). Caudal to the lesion level the mean FA ipsilesional was lower than the uninvolved side but didn’t reach statistical significance (paired t-test, P ¼ 0.23). There was no significant difference of mean FA between contralesional and ipsilesional sides rostral to the lesion level (Figure 2). Mean ADC showed no significant differences between ipsilateral and contralateral sides at any level (Figure 2). All quantitative results are shown in Table 3. Correlation analyses performed between LI of FA/ADC at the 3 levels and simultaneous motor function (MRC0) and long-term follow up motor function (MRC2) using Spearman’s test, showed no correlation between the LI of FA or ADC in these 3 levels and the corresponding preoperative motor function (MRC0) and long-term follow-up motor function (MRC2).
WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2015.01.045
ORIGINAL ARTICLE YUQIANG YAO ET AL.
QUANTIFICATION OF CST WITH DTI IN BRAINSTEM SURGERY
Table 3. Continued ADC Ipsi- vs. Contralateral
Correlation Analysis
Pre- vs. Postoperative
Ipsi
Contra
P Value
LI
p (LI vs MRC0)
p (LI vs MRC2)
Pre
Post
P Value
1.113 0.262
1.061 0.303
0.705
0.028 0.133
0.089
0.725
1.201 0.231
1.077 0.177
0.495
0.996 0.354
0.823 0.106
0.155
0.087 0.192
0.307
0.324
1.007 0.298
0.815 0.239
0.028
0.816 0.153
0.791 0.142
0.593
0.015 0.072
0.458
0.119
0.835 0.166
0.795 0.146
0.552
There was a slightly increasing trend of FA (paired t-test, P ¼ 0.182) and a decreasing trend of ADC (paired t-test, P ¼ 0.028) of ROIs at the level of the lesions after operation as compared with the preoperative ones (Figure 3). There were no significant differences of FA/ADC values between pre- and postoperative DTI at the caudal and rostral levels.
DISCUSSION DTI-WMT is currently the only noninvasive method that can visualize the human white matter tracts in vivo. This technique has a proven value for planning a safer surgical approach but also a
Figure 3. The fractional anisotropy/apparent diffusion coefficient (FA/ADC) changes of lesion levels after operation comparing with preoperative evaluation. There was an increasing trend of FA (paired t-test, P ¼ 0.182) and a decreasing trend of ADC (paired t-test, P ¼ 0.028) of lesion insulted regions of interest after operation compared with preoperative imaging.
WORLD NEUROSURGERY 83 [6]: 1006-1014, JUNE 2015
prospective potential for predicting postoperative motor function. To date, this has been reported mostly in supratentorial tumor surgery (18, 36). Previous studies have reported that for supratentorial lesions, preoperative CST involvement may be a predictor of motor deficit and that postoperative improvement, or normalization, of the CST was a predictor of clinical improvement (3, 4, 9). To our knowledge, only a few published series have reported CST alteration using DTI-WMT in patients with brainstem lesions especially cavernomas. Chen et al. (5) studied the ability of DTI-WMT to visualize white matter tract involvement before and after surgery in 10 patients with brainstem lesions, of which 7 were BSC. Only one patient with deviating and partially interrupting CST showed a motor deficit preoperatively. Four months postoperatively, the motor deficit completely resolved and the CST normalized on DTI-WMT. The other 6 patients with normal CST morphology showed normal motor function. Kovanlikaya et al. (16) reported a series of 14 patients with brainstem lesions, of which 6 were BSCs. Examining the CSTs of the BSC patients, they found 7 normal CSTs with normal motor function, 5 abnormal CSTs, 2 associated with normal motor function and 3 with preoperative motor deficit (2 deformed CSTs and 1 interrupted CST). All 3 patients with preoperative motor deficits showed improved motor performance after surgery (followup 2e8 months). The patient with interrupted CST had normalization of the DTI findings whereas 2 deformed tracts did not change their appearance. McLaughlin and Kelly (20) presented a case supporting the use of DTI as predictor of motor recovery after BSC evacuation. The patient had interrupted CST preoperatively with mild right pronator drift and, 2-months postoperatively, DTI showed normal and symmetric CST that correlated to grossly normal motor function. Our study included 14 BSC patients with preoperative DTI. We were able to show similar results as in previous reports. The negative predictive value of preoperative DTI-WMT was very high (100%) both for MRC0 and MRC2, indicating that normal appearance of the CSTs is a highly predictive indicator of tract preservation. Compared with the high negative predictive value (100%), PPV was somewhat limited, especially on the follow-up evaluation (only 20%). On presurgical evaluation 7 of 10 abnormal CSTs showed abnormal motor function (positive predictive value 70%). On the follow-up evaluation, 5 of these 7 patients fully recovered to MCR Grade 5, which points out that presurgical DTI-WMT disruption should not necessarily be
www.WORLDNEUROSURGERY.org
1011
ORIGINAL ARTICLE YUQIANG YAO ET AL.
QUANTIFICATION OF CST WITH DTI IN BRAINSTEM SURGERY
considered as an indicator of definite tract loss and worst prognosis. Hence, caution is warranted in cases with abnormal CSTs with only mild motor deficit in preoperative planning. In fact, despite complete interruption of the CSTs at the lesion injured level, 4 of 6 patients recovered to normal motor function, confirming the integrity of their ipsilateral CSTs. We agree with Kovanlikaya et al.’s opinion that the mismatch between DTI findings and motor functions are considered to relate to lesion characteristics and especially to the presence of artifacts, including interfering hematoma, edema and vascular pulsation (16). This is particularly true for cavernomas which, despite their well-defined margins, are invariably hemorrhagic lesions, associated with hemosiderin deposition in the adjacent brain parenchyma causing pronounced magnetic susceptibility artifacts and thus signal loss in MRI. Comparison of pre- and postoperative DTI results also indicated that selection of the appropriate surgical approach should result to the benefit of the patients’ function. In our group, 4 patients showed improved morphology in postoperative DTI compared with preoperative DTI and their motor function all recovered to normal status postoperatively. Quantitative analysis of diffusion parameters at the level of the lesions showed similar results to our morphologic evaluation. Although there was significant difference between ipsilateral and contralateral FAs at lesion level, correlation analysis showed that lower FA values do not necessarily correlate with unfavorable prognosis. The postoperative trend of increasing FA (P > 0.05) and decreasing ADC (P < 0.05) values in ROIs at the level of the cavernomas paralleled the morphological improvement of CSTs compared with the preoperative DTI. It is well known that abnormal brain parenchyma attributable to edema or gliosis usually shows decreased FA and increased ADC (13, 30). Therefore, a trend to normalization of the FA (i.e., increasing values) and ADC (i.e., decreasing values) postoperatively may reflect recovery of infiltrated fasciculi. However, it is important to mention that the inherent sensitivity of MRI to susceptible artifacts arising from the presence of hemorrhagic products, as discussed in the previous paragraph, is an important limiting factor in the quantification of diffusion parameters. In our study, we additionally investigated potential anterograde and retrograde FA changes of CST and their relationship with prognosis. Degeneration of white matter fibers at a distance from a primary lesion is a common finding in many diseases of the central nervous system. These secondary white matter degeneration is classically divided into retrograde and anterograde (or Wallerian) degeneration. From the primary lesion, retrograde degeneration proceeds proximally toward the cell body, while anterograde degeneration proceeds distally toward the axon’s terminals. The diffusion parameters of anterograde and retrograde degeneration following trauma, stroke, demyelinating disease, have been demonstrated in several previous studies (1, 27, 33). However, very few reports discussed the anterograde or retrograde FA/ADC changes, reflecting secondary axonal degeneration, associated with brainstem lesions, especially BSCs. By analyzing anterograde and retrograde diffusion parameters of 10 morphologically abnormal CSTs, we found lower FA of
1012
www.SCIENCEDIRECT.com
ipsilateral CSTs compared with contralateral CSTs in the caudal direction that, however, did not reach statistical significance. There were no significant differences regarding FAs and ADCs in the rostral direction on both sides. The relatively comparable diffusion parameters of CSTs in both anterograde and retrograde directions in relation to the lesion level may be linked to the relatively good prognosis. In this group, only 2 of 10 patients showed MRC grade 4 on follow-up period, the motor grade of the other 8 patients either improved to or remained at grade 5. It has been previously demonstrated that secondary degeneration of CSTs from cerebral or spinal primary insults at chronic stage can cause anterograde or retrograde alterations of diffusion parameters, e.g., FA decreased and/or ADC increased, which may correlate to changes in corresponding motor functions (14, 24). Thus, the good postoperatively clinical motor function both ipsi- and contralaterally observed in our patient cohort may be reflected in the lack of significant differences of diffusion parameters. We herein present the first report discussing morphologic characteristics and anterograde/retrograde changes of diffusion parameters along the CSTs of patients with BSCs. Morphologically interrupted CSTs and decreased FA correlate well at the BSC lesion level. Morphologic characteristics and diffusion parameter changes of CSTs, however, do not allow a strong prediction of patients’ motor outcome. The relatively unchanged anterograde and retrograde diffusion parameters of CSTs confirmed the integrity of the corresponding CSTs as reflected in good motor function, although we didn’t get positive correlation between diffusion parameters and clinical manifestations. A major limitation of the present study is the limited number of subjects and their relatively homogeneous clinical picture. Most patients returned to or remained at normal motor function postoperatively causing difficult the detection of potential subtle quantitative diffusion differences. In this series, no patients had a worsened postoperative outcome, which renders difficult to conclude with any confidence that normal CST morphology predicts favorable postoperative outcome when all patients remained the same/improved clinically. However, these data represented our clinical experience from a series of consecutive cases and patients haven’t been selected on the basis of their clinical outcome. Studying patients with persisting, or worsened, neurologic/motor deficits would probably allow more accurate information on the prognostic significance of morphologic and quantitative DTI changes. Nevertheless, the trends found in our study, i.e., high sensitivity and NPV of pre-operative DTI-WMT, are strong indicators for good outcome of pre-peratively detected intact WMT. Despite the standardized procedures and imaging protocols applied in this study, its retrospective nature is another limitation related to limited availability of postoperative DTI information in only half of the patients. However, the type of pathology we addressed is relatively rare and handled in centers with increased expertise in brainstem surgery. Our study reports on a large number of BSCs with DTI analysis collected during a period of 2 years. A larger number of subjects and more diversity in clinical manifestation would certainly lead to more reliable conclusions and need to be
WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2015.01.045
ORIGINAL ARTICLE YUQIANG YAO ET AL.
QUANTIFICATION OF CST WITH DTI IN BRAINSTEM SURGERY
addressed in future prospective, potentially multicenter studies or meta-analysis reports.
changes at lesion level cannot predict poor prognosis. Caudal and rostral diffusion parameters can provide more information of the integrity of CSTs compared with morphologic study alone.
CONCLUSIONS Normal CST morphology generated from WMT using DTI predicts a favorable postoperative outcome in patients with BSC. Interrupted CSTs and decreased FA values correlate well within BSC lesion level. However, morphologic characteristics and diffusion parameter
REFERENCES 1. Ahn YH, Kim SH, Han BS, Kim OL, Ahn SH, Cho YW, Kwon YH, Jang SH: Focal lesions of the corticospinal tract demonstrated by diffusion tensor imaging in patients with diffuse axonal injury. NeuroRehabilitation 21:239-243, 2006. 2. Basser PJ, Pajevic S, Pierpaoli C, Duda J, Aldroubi A: In vivo fiber tractography using DTMRI data. Magn Reson Med 44:625-632, 2000. 3. Bello L, Gambini A, Castellano A, Carrabba G, Acerbi F, Fava E, Giussani C, Cadioli M, Blasi V, Casarotti A, Papagno C, Gupta AK, Gaini S, Scotti G, Falini A: Motor and language DTI Fiber Tracking combined with intraoperative subcortical mapping for surgical removal of gliomas. Neuroimage 39:369-382, 2008. 4. Castellano A, Bello L, Michelozzi C, Gallucci M, Fava E, Iadanza A, Riva M, Casaceli G, Falini A: Role of diffusion tensor magnetic resonance tractography in predicting the extent of resection in glioma surgery. Neuro Oncol 14:192-202, 2012. 5. Chen X, Weigel D, Ganslandt O, Buchfelder M, Nimsky C: Diffusion tensor imaging and white matter tractography in patients with brainstem lesions. Acta Neurochir (Wien) 149:1117-1131; discussion 1131, 2007. 6. Ciccarelli O, Parker GJ, Toosy AT, WheelerKingshott CA, Barker GJ, Boulby PA, Miller DH, Thompson AJ: From diffusion tractography to quantitative white matter tract measures: a reproducibility study. Neuroimage 18:348-359, 2003. 7. Ciurea AV, Nastase C, Tascu A, Brehar FM: Lethal recurrent hemorrhages of a brainstem cavernoma. Neurosurg Rev 30:259-262; discussion 262, 2007. 8. Del Curling O Jr, Kelly DL Jr, Elster AD, Craven TE: An analysis of the natural history of cavernous angiomas. J Neurosurg 75:702-708, 1991. 9. Ellis MJ, Rutka JT, Kulkarni AV, Dirks PB, Widjaja E: Corticospinal tract mapping in children with ruptured arteriovenous malformations using functionally guided diffusion-tensor imaging. J Neurosurg Pediatr 9:505-510, 2012. 10. Field AS, Alexander AL, Wu YC, Hasan KM, Witwer B, Badie B: Diffusion tensor eigenvector directional color imaging patterns in the evaluation of cerebral white matter tracts altered by tumor. J Magn Reson Imaging 20:555-562, 2004.
ACKNOWLEDGMENTS The authors are very grateful to all patients for participating in this study and thank Dr. Uzeyir Ahmadli for neuroradiological consultation in fiber tracking.
11. Fritschi JA, Reulen HJ, Spetzler RF, Zabramski JM: Cavernous malformations of the brain stem. A review of 139 cases. Acta Neurochir (Wien) 130: 35-46, 1994. 12. Glenn OA, Henry RG, Berman JI, Chang PC, Miller SP, Vigneron DB, Barkovich AJ: DTI-based three-dimensional tractography detects differences in the pyramidal tracts of infants and children with congenital hemiparesis. J Magn Reson Imaging 18: 641-648, 2003. 13. Gulani V, Webb AG, Duncan ID, Lauterbur PC: Apparent diffusion tensor measurements in myelin-deficient rat spinal cords. Magn Reson Med 45:191-195, 2001. 14. Guleria S, Gupta RK, Saksena S, Chandra A, Srivastava RN, Husain M, Rathore R, Narayana PA: Retrograde Wallerian degeneration of cranial corticospinal tracts in cervical spinal cord injury patients using diffusion tensor imaging. J Neurosci Res 86:2271-2280, 2008. 15. Guo AC, Cummings TJ, Dash RC, Provenzale JM: Lymphomas and high-grade astrocytomas: comparison of water diffusibility and histologic characteristics. Radiology 224:177-183, 2002. 16. Kovanlikaya I, Firat Z, Kovanlikaya A, Ulug AM, Cihangiroglu MM, John M, Bingol CA, Ture U: Assessment of the corticospinal tract alterations before and after resection of brainstem lesions using Diffusion Tensor Imaging (DTI) and tractography at 3T. Eur J Radiol 77:383-391, 2011. 17. Lazar M, Alexander AL, Thottakara PJ, Badie B, Field AS: White matter reorganization after surgical resection of brain tumors and vascular malformations. AJNR Am J Neuroradiol 27:1258-1271, 2006. 18. Lo CY, Chao YP, Chou KH, Guo WY, Su JL, Lin CP: DTI-based virtual reality system for neurosurgery. Conf Proc IEEE Eng Med Biol Soc 2007:1326-1329, 2007. 19. Ma L, Song ZJ: Differentiation between low-grade and high-grade glioma using combined diffusion tensor imaging metrics. Clin Neurol Neurosurg 115:2489-2495, 2013. 20. McLaughlin N, Kelly DF: Corticospinal tractography as a prognosticator for motor improvement after brainstem cavernoma resection. Br J Neurosurg 27:108-110, 2013. 21. Mori S, Crain BJ, Chacko VP, van Zijl PC: Threedimensional tracking of axonal projections in the brain by magnetic resonance imaging. Ann Neurol 45:265-269, 1999.
WORLD NEUROSURGERY 83 [6]: 1006-1014, JUNE 2015
22. Nguyen TH, Yoshida M, Stievenart JL, IbaZizen MT, Bellinger L, Abanou A, Kitahara K, Cabanis EA: MR tractography with diffusion tensor imaging in clinical routine. Neuroradiology 47:334-343, 2005. 23. Paternostro-Sluga T, Grim-Stieger M, Posch M, Schuhfried O, Vacariu G, Mittermaier C, Bittner C, Fialka-Moser V: Reliability and validity of the Medical Research Council (MRC) scale and a modified scale for testing muscle strength in patients with radial palsy. J Rehabil Med 40:665-671, 2008. 24. Pierpaoli C, Barnett A, Pajevic S, Chen R, Penix LR, Virta A, Basser P: Water diffusion changes in Wallerian degeneration and their dependence on white matter architecture. Neuroimage 13:1174-1185, 2001. 25. Porter PJ, Willinsky RA, Harper W, Wallace MC: Cerebral cavernous malformations: natural history and prognosis after clinical deterioration with or without hemorrhage. J Neurosurg 87:190-197, 1997. 26. Porter RW, Detwiler PW, Spetzler RF, Lawton MT, Baskin JJ, Derksen PT, Zabramski JM: Cavernous malformations of the brainstem: experience with 100 patients. J Neurosurg 90:50-58, 1999. 27. Reich DS, Zackowski KM, Gordon-Lipkin EM, Smith SA, Chodkowski BA, Cutter GR, Calabresi PA: Corticospinal tract abnormalities are associated with weakness in multiple sclerosis. AJNR Am J Neuroradiol 29:333-339, 2008. 28. Robinson JR, Awad IA, Little JR: Natural history of the cavernous angioma. J Neurosurg 75: 709-714, 1991. 29. Sinha S, Bastin ME, Whittle IR, Wardlaw JM: Diffusion tensor MR imaging of high-grade cerebral gliomas. AJNR Am J Neuroradiol 23:520-527, 2002. 30. Thomas B, Eyssen M, Peeters R, Molenaers G, Van Hecke P, De Cock P, Sunaert S: Quantitative diffusion tensor imaging in cerebral palsy due to periventricular white matter injury. Brain 128: 2562-2577, 2005. 31. Ulrich NH, Ahmadli U, Woernle CM, Alzarhani YA, Bertalanffy H, Kollias SS: Diffusion tensor imaging for anatomical localization of cranial nerves and cranial nerve nuclei in pontine lesions: initial experiences with 3T-MRI. J Clin Neurosci 21:1924-1927, 2014. 32. Ulrich NH, Kockro RA, Bellut D, Amaxopoulou C, Bozinov O, Burkhardt JK, Sarnthein J, Kollias SS, Bertalanffy H: Brainstem cavernoma surgery with
www.WORLDNEUROSURGERY.org
1013
ORIGINAL ARTICLE YUQIANG YAO ET AL.
QUANTIFICATION OF CST WITH DTI IN BRAINSTEM SURGERY
the support of pre- and postoperative diffusion tensor imaging: initial experiences and clinical course of 23 patients. Neurosurg Rev 37:481-491; discussion 92, 2014. 33. Vargas P, Gaudron M, Valabregue R, Bertasi E, Humbert F, Lehericy S, Samson Y, Rosso C: Assessment of corticospinal tract (CST) damage in acute stroke patients: comparison of tractspecific analysis versus segmentation of a CST template. J Magn Reson Imaging 37:836-845, 2013. 34. Wang CC, Liu A, Zhang JT, Sun B, Zhao YL: Surgical management of brain-stem cavernous malformations: report of 137 cases. Surg Neurol 59:444-454; discussion 454, 2003.
35. Witwer BP, Moftakhar R, Hasan KM, Deshmukh P, Haughton V, Field A, Arfanakis K, Noyes J, Moritz CH, Meyerand ME, Rowley HA, Alexander AL, Badie B: Diffusion-tensor imaging of white matter tracts in patients with cerebral neoplasm. J Neurosurg 97:568-575, 2002.
commercial or financial relationships that could be construed as a potential conflict of interest. Yuqiang Yao and Nils H. Ulrich are coefirst authors. Received 6 July 2014; accepted 20 January 2015 Citation: World Neurosurg. (2015) 83, 6:1006-1014. http://dx.doi.org/10.1016/j.wneu.2015.01.045 Journal homepage: www.WORLDNEUROSURGERY.org
36. Yu CS, Li KC, Xuan Y, Ji XM, Qin W: Diffusion tensor tractography in patients with cerebral tumors: a helpful technique for neurosurgical planning and postoperative assessment. Eur J Radiol 56:197-204, 2005.
Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2015 Elsevier Inc. All rights reserved.
Conflict of interest statement: The authors declare that the article content was composed in the absence of any
WORLD FEDERATION OF NEUROSURGICAL SOCIETIES
Mission Statement
To facilitate the personal association of neurological surgeons throughout the world To aid in the exchange and dissemination of knowledge and ideas in the field of neurological surgery To encourage research in neurological surgery and allied sciences To address issues of neurosurgical demography To address issues of Public Health To implement, improve and promote the standards of neurosurgical care and training worldwide
1014
www.SCIENCEDIRECT.com
WORLD NEUROSURGERY, http://dx.doi.org/10.1016/j.wneu.2015.01.045
