Diffusion tensor imaging for anatomical localization of cranial nerves and cranial nerve nuclei in pontine lesions: Initial experiences with 3T-MRI

Journal of Clinical Neuroscience 21 (2014) 1924–1927

Contents lists available at ScienceDirect

Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn

Clinical Study

Diffusion tensor imaging for anatomical localization of cranial nerves and cranial nerve nuclei in pontine lesions: Initial experiences with 3T-MRI Nils H. Ulrich a,b,⇑, Uzeyir Ahmadli b, Christoph M. Woernle a, Yahea A. Alzarhani b, Helmut Bertalanffy c, Spyros S. Kollias b a

Department of Neurosurgery, University Hospital, University of Zurich, Zurich, Switzerland Department of Neuroradiology, University Hospital, University of Zurich, Frauenklinikstrasse 10, 8091 Zurich, Switzerland c International Neuroscience Institute, INI, Hannover, Germany b

a r t i c l e

i n f o

Article history: Received 15 December 2013 Accepted 23 March 2014

Keywords: Brainstem Cranial nerve Cranial nerve nuclei Cavernoma DTI MRI Pons

a b s t r a c t With continuous refinement of neurosurgical techniques and higher resolution in neuroimaging, the management of pontine lesions is constantly improving. Among pontine structures with vital functions that are at risk of being damaged by surgical manipulation, cranial nerves (CN) and cranial nerve nuclei (CNN) such as CN V, VI, and VII are critical. Pre-operative localization of the intrapontine course of CN and CNN should be beneficial for surgical outcomes. Our objective was to accurately localize CN and CNN in patients with intra-axial lesions in the pons using diffusion tensor imaging (DTI) and estimate its input in surgical planning for avoiding unintended loss of their function during surgery. DTI of the pons obtained pre-operatively on a 3 Tesla MR scanner was analyzed prospectively for the accurate localization of CN and CNN V, VI and VII in seven patients with intra-axial lesions in the pons. Anatomical sections in the pons were used to estimate abnormalities on color-coded fractional anisotropy maps. Imaging abnormalities were correlated with CN symptoms before and after surgery. The course of CN and the area of CNN were identified using DTI pre- and post-operatively. Clinical associations between post-operative improvements and the corresponding CN area of the pons were demonstrated. Our results suggest that pre- and post-operative DTI allows identification of key anatomical structures in the pons and enables estimation of their involvement by pathology. It may predict clinical outcome and help us to better understand the involvement of the intrinsic anatomy by pathological processes. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction Due to the complex and dense anatomy of the brainstem, surgery on and visualization of pontine lesions in relation to cranial nerves (CN) and cranial nerve nuclei (CNN) remain challenging for neurosurgeons and neuroradiologists. With continuous development of surgical techniques and neuroimaging, the management of those lesions is improving. MRI techniques, in particular diffusion tensor imaging (DTI), have been proven to provide reliable anatomical–structural information in various areas of the brain [1–5]. The region of the pons, which represents an exceptional visualization challenge due to its densely packed anatomical structures, has been the focus of several investigations using DTI [3,6–8]. The brainstem encompasses areas with highly eloquent functions that are under constant risk of damage by surgical manipula⇑ Corresponding author. Tel.: +41 4 4255 1111; fax: +41 4 4255 4505. E-mail address: [email protected] (N.H. Ulrich). http://dx.doi.org/10.1016/j.jocn.2014.03.027 0967-5868/Ó 2014 Elsevier Ltd. All rights reserved.

tion. Among them, certain CN and CNN play a critical role in normal functioning. Their pre-operative localization in the presence of pathology is challenging, but nevertheless can contribute to avoiding unintended loss of function and undoubtedly creates an advantage for surgical outcomes. Although DTI provides valuable information about the distortion and interruption of corticospinal and sensory tracts in the infratentorial area, the exact anatomical localization of CN and CNN in relation to brainstem lesions has not yet been systematically studied to our knowledge. According to current literature [9] and anatomical textbooks [10] CN and CNN are symmetrically located around and within the pons. In recent years we have been using DTI routinely for pre- and post-operative visualization and surgical planning in patients harboring lesions in the infratentorial area of the brainstem. We present our experience using DTI to visualize certain CN and CNN in seven patients with brainstem cavernomas (BSC) in regard to pre-operative planning and post-operative clinical outcome.

1925

N.H. Ulrich et al. / Journal of Clinical Neuroscience 21 (2014) 1924–1927

Our goal was to determine whether it is possible to identify CN and CNN for surgical guidance and improved clinical outcomes with current DTI technology. 2. Patients and methods 2.1. Patient characteristics This retrospective single-center study included seven patients (six women, one man) with a mean age of 36 years (standard deviation 15 years; range 19 to 50 years, Table 1). Each underwent microsurgical resection of a symptomatic BSC. All procedures and imaging were performed in accordance with the routine institutional guidelines after receiving the patient’s written consent. All BSC were located in the pons (Table 1). 2.2. Clinical assessment and cranial nerve deficits Patient data, including CN deficits, were retrospectively obtained upon admission and once again on discharge with a special focus on CN V–VII. The neurological status of all patients during the pre-operative period and upon discharge was classified with a modified Patzold Rating [11,12]. This rating tests CN, motility, motor disturbances, pathological reflexes, sensory deficit, gait disturbances, and psychological symptoms to achieve a grading of functional disturbances. Patzold Rating scores represent a weighted sum of clinical symptoms. The rating was assessed 1 day before surgery and 1 day after surgery. All clinical information was retrospectively derived from official admission reports, clinical notes and discharge letters. Histopathological analysis of the resected lesions revealed cavernoma malformations in all seven patients.

128  128 mm, 60 contiguous slices, slice thickness = 2.1 mm, echo time = 50 ms, number of signal averages = 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 = 0. The DTI time was approximately 5 minutes. Axial, coronal and sagittal images of the color-coded fractional anisotropy (FA) maps were analyzed for identification of consistent anatomy and structures of CN and CNN V, VI and VII. A standard color scheme was used in the Philips software to encode the FA maps, with blue indicating superior to inferior direction, red indicating transverse direction and green indicating anterior-posterior direction. Anatomical key locations of CN and CNN V, VI and VII and white matter tracts were identified by comparing MRI and color-coded FA maps. Figure 1 illustrates the estimated location and extent of CNN, CN and major tracts that are critical in the lower pons. Images were subsequently compared with anatomical sections of the pons from an anatomical brainstem atlas [10]. 2.4. Surgical procedure All patients underwent microsurgical resection of the symptomatic BSC in the pons. Surgical approaches depended on the size and location of the lesion and included four suboccipital, two retromastoid and one temporobasal approaches. Multimodal intra-operative monitoring was used to support the surgeon and included motor evoked potentials [13], somatosensory evoked potentials [14], acoustic evoked potentials, neuronavigation and mapping of the rhomboid fossa [15] 3. Results 3.1. Illustrative patients

2.3. Conventional MRI data acquisition and DTI data post-processing All radiological images were provided and analyzed by the Department of Neuroradiology, University of Zurich. T1- and T2-weighted MRI, as well as DTI studies, were obtained as part of the pre- and post-operative routine at this institution. A 3 Tesla (3T) whole body MRI system (Philips Achieva, Best, Netherlands), equipped with 80 mT/m/ms gradient coils and an eight element receive head coil array (MRI Devices, Waukesha, WI, USA), was used for imaging studies in patients with BSC. Each imaging session included DTI and an anatomical imaging study, including a gadolinium enhanced scan for intra-operative navigation. The field of view for all scans was defined as 200  200 mm2. For the DTI series, a whole brain diffusion-weighted single-shot spin-echo echo planar imaging sequence was applied with the following parameters: in-plane matrix = 96  96 mm, reconstructed to

3.1.1. Illustrative Patient 1 Supplementary Figure 1A illustrates the typical case of Patient 6. This 22-year-old woman had a BSC malformation (3.8 mm3) in the right central part of the pons. On admission she presented with hemiparesis and CN deficit VI and VII. CN V was intact. On axial FA color-coded maps CN V showed up as sagittal fibers in green. The major FA map findings were distortion in the area of CN VI and VII. This correlated well with the space occupying effect of the lesion and the CN deficits of the patient. Furthermore, the intact cisternal course of CN V was displayed. This was reflected by the absence of relevant clinical symptoms. The transverse pontocerebellar fibers formed red bands as they cross the pons, but turned green (indicating sagittal direction) as they curved dorsally and merged into the sagittal fibers of the middle cerebellar peduncles. The corticospinal tracts formed blue bundles as they descended

Table 1 Clinical and surgical characteristics of patients Patient

Age

1 2 3 4 5 6 7

47 39 32 44 50 22 19

Minimum Median Maximum

19 39 50

Sex

F F F F M F F

Location

Pons Pons Pons Pons Pons Pons Pons

Volume (mm3)

21.1 17.4 15.7 14.9 9.2 3.8 0.6

Approach

Suboccipital Retromastoidal Temperobasal Suboccipital Suboccipital Suboccipital Retromastoidal

Cranial nerve deficits

Patzold

0 CN V

1 CN V

0 CN VI

1 CN VI

0 CN VII

1 CN VII

0

1

Y Y Y N N N N

Y Y Y N N N N

Y N N Y Y N Y

Y N N Y Y Y Y

Y N Y Y Y Y N

Y N Y N N N Y

17 7 18 21 23 13 10

17 6 6 7 19 9 8

7 17 23

6 8 19

0.6 6.1 28.9

CN = cranial nerve, Patzold = Patzold Rating score, N = no, Y = yes, 0 = preoperative, 1 = postoperative.

1926

N.H. Ulrich et al. / Journal of Clinical Neuroscience 21 (2014) 1924–1927

Fig. 1. Sectional level of the lower pons. The approximate locations and extent of cranial nerve nuclei and major tracts that are important in the lower pons are simplified in this axial illustration of the brainstem section.

through the mid-pons. The ascending medial lemnisci also formed blue bands (indicating superior to inferior direction). Post-operatively, our imaging data (Supp. Fig. 1B) showed that the lesion had been removed and no damage to the surrounding structures was seen. The space occupying effect of the lesion in the area of cranial nerve VI and VII was significantly reduced. The CN VI deficit resolved and CN VII deficit remained but no additional CN deficits were seen. 3.1.2. Illustrative Patient 2 Our second illustrative patient (Supp. Fig. 2A, Patient number 7) was a 19-year-old woman with a 0.6 mm3 lesion at the pontomedullary junction. Pre-operatively, she presented with CN VI and VII deficits. The major DTI finding was the course of CN VII as it merged dorsally to enter/exit the brainstem just below the middle cerebellar peduncle. The clearly visible inferior cerebellar peduncles show an anterosuperior course (blue). Successful removal of the lesion without damage to the brainstem was identified on post-operative MRI. The DTI color-coded map showed preservation of CN VII. The patient still had a deficit of CN VII function in the early post-operative phase, which later resolved. The space occupying effect was still visible and parts of the corticospinal tracts were deviated into a more anterior posterior direction, as indicated in green (Supp. Fig. 2B). On the post-operative FA map the lesion appeared to be larger. 3.2. Effect of surgery on neurological outcome To further evaluate the effect of surgery on neurological outcome we performed a broad classification of all seven patients using the Patzold Rating. This assessment was done on admission and on discharge. The Patzold Rating improved in six out of seven patients (86%, p = 0.03). One patient remained unchanged (Fig. 2). 3.3. FA color differentiation during clinical course In Illustrative Patient 1, the pre-operative FA map showed a rim of greenish color at the periphery around the lesion in the key regions of CN VI and VII. The band of the bluish lemniscus medialis and parts of CNN VI and VII were pushed laterally and were more green in appearance (indicating sagittal movement), depending on the direction of the diffusion of water molecules. In comparison to the pre-operative image, the post-operative FA map showed reorganization in the area of interest with less space occupying effect and clear bluish delineation. The greenish appearance disappeared.

Fig. 2. Follow-up of Patzold Rating. Patzold Rating improved in six out of seven patients (86%, grey lines), and remained equal in one (14%, black line).

In all patients, color-coded FA maps helped to identify key regions of interest in the pons. Pre-operatively, the space occupying effect of the lesion showed variable degree of distortion and displacement in the region of interest, related to specific CN deficits. Post-operative FA maps were taken and consistent patterns of key anatomical landmarks were visualized. After lesion removal the key anatomy showed structural redistribution comparable to the normal atlas anatomy. We found consistent patterns in color-coded FA maps across all patients during the clinical course. The anatomical resolution provided by the clinical MRI protocol was sufficient for the detection of small lesions in the pons and their relationship to adjacent structures. We found consistent patterns of white matter tracts, namely corticospinal tracts and medial lemnisci. Further CN and CNN location of V, VI and VII could be displayed in our FA maps and this was also reflected in the neurological status of the patient. The post-operative improvement of CN symptoms could be related to the repositioning of the CNN in all of our patients. 4. Discussion The aim of our study was to investigate the feasibility of incorporating clinical DTI into routine pre-operative imaging to identify key anatomical structures in patients with lesions in the pons and to possibly offer guidance for neurosurgical procedures. Generating precise imaging is difficult in the area of the pons where vital key structures such as fiber tracts, CN and CNN are in close proximity. Preserving the integrity of these structures within the pons is challenging for successful surgery. Other studies have shown reproducible visualization of brainstem structures, namely the corticospinal tracts and medial lemnisci, with DTI and in particular in the pons [9,16]. In the present study, we additionally analysed the usefulness of DTI color-coded FA maps in patients with pons lesions during the pre- and post-operative course. DTI color-coded FA maps are the method of choice in visualizing white matter tracts in vivo. This visualization technique has multiple applications in neurosurgery and neuroradiology. It depicts major white matter tracts, and can also be applied in gray matter partitioning [14,17,18]. To our knowledge, color-coded FA maps have not been studied to identify key structures such as CN and CNN while a lesion distorted the intrinsic anatomy. In this study we demonstrated that 3T DTI color-coded FA maps may increase the information available to the surgeon before and after surgery. We have shown that color-coded FA maps identify key structures in the pons that are not obvious on conventional T2-weighted sequences. Furthermore, they depict changes in the intrinsic anatomy relevant to the clinical symptoms and post-operative course

N.H. Ulrich et al. / Journal of Clinical Neuroscience 21 (2014) 1924–1927

of each patient. Although we have demonstrated a proof-of-concept in this retrospective series with 3T MRI, we acknowledge several shortcomings in our analysis and interpretation. One of our major weak points is image resolution. DTI sequences are acquired with a voxel resolution of 2  2  2 mm3. Therefore, spatial resolution and accuracy is limited to some degree. Patient motion, large BSC and hemosiderin-related artifacts are other factors that might change FA maps and cause limitations as well. Cavernomas tend to show increased susceptibility artifacts on DTI, probably due to concentrated hemosiderin in and around the lesion. This was seen in our second illustrative patient, where the lesion appeared to be larger after removal on our FA maps. Furthermore, we did not estimate our results using quantitative measurements such as FA or apparent diffusion coefficient. Nevertheless, we found associations between the clinical status and the structural distortion demonstrated by the color-coded FA maps. Future prospective studies with high resolution DTI are required to evaluate the impact of better spatial resolution. High resolution DTI and higher field strengths, including 7T MRI, 9T MRI and diffusion kurtosis imaging (DKI) [19,20], will further increase precision in structure definition. Further, one can increase the signal by collecting more directions (e.g. 16) or increasing the number of times each direction is sampled. However, additional directions or samples do increase the acquisition time. Naidich et al. demonstrated stippled zones of mixed coloration in CN VI and VII at the level of the mid pons with 9T MRI in a post mortem specimen. In comparison, our resolution was not high enough to confirm this coloration but we found the coloration in the area of CN VI and VII was associated with the clinical course and outcome. Visualization of CN and CNN, especially before surgery, offers a surgical planning advantage by determining the exact point of entry into the brainstem. Therefore, we strongly encourage further studies with higher field strength and DKI, involving larger numbers of subjects and clinical correlation, in the near future. Other groups studying supratentorial and infratentorial surgery [3,5,21,22,23] have included DTI derived tractography of fiber tracts into their planning process. Incorporation of this technology in neurosurgery should not only rely on visual inspection of colorcoded images but also on correlations between quantifiable DTI parameters and clinical deficits [24]. In particular, surgery in areas with low contrast resolution on conventional MRI, such as the brainstem, is expected to greatly benefit from the introduction of this technology. We believe that our study represents a helpful contribution to improve safety through the more visible and intuitive modality of DTI compared with electrophysiology and neuronavigation alone. 4.1. Limitations and outlook Despite current limitations in image resolution, our experiences with incorporating DTI color-coded maps in routine pre-operative imaging planning in pons surgery are encouraging. Further refinements of this technique in both resolution as well as automation of fiber reconstruction and data quantification promise a more widespread application in pre-surgical planning and estimation of prognosis. Probabilistic DTI and DKI with higher field strengths will be crucial for the understanding of the intrinsic anatomy of the pons. 5. Conclusion DTI color-coded FA maps in patients with intrinsic lesions in the pons accurately depict key anatomical structures and the distortion caused by the lesion not visible on conventional MRI. These changes correlate with patient symptoms and their post-operative clinical course. Therefore FA maps increase the information avail-

1927

able to the surgeon before and after surgery. In the future higher spatial resolution DTI may be of importance when planning these patients for surgery. Conflicts of interest/disclosures The authors declare that they have no financial or other conflicts of interest in relation to this research and its publication. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jocn.2014.03.027. References [1] Arfanakis K, Gui M, Lazar M. Optimization of white matter tractography for pre-surgical planning and image-guided surgery. Oncol Rep 2006;15(Spec no.):1061–4. [2] Barboriak DP. Imaging of brain tumors with diffusion-weighted and diffusion tensor MR imaging. Magn Reson Imaging Clin N Am 2003;11:379–401. [3] Chen X, Weigel D, Ganslandt O, et al. Diffusion tensor imaging and white matter tractography in patients with brainstem lesions. Acta Neurochir (Wien) 2007;149:1117–31 [discussion 1131]. [4] Coenen VA, Krings T, Mayfrank L, et al. Three-dimensional visualization of the pyramidal tract in a neuronavigation system during brain tumor surgery: first experiences and technical note. Neurosurgery 2001;49:86–92 [discussion 92–3]. [5] Witwer BP, Moftakhar R, Hasan KM, et al. Diffusion-tensor imaging of white matter tracts in patients with cerebral neoplasm. J Neurosurg 2002;97: 568–75. [6] Cao Z, Lv J, Wei X, et al. Appliance of preoperative diffusion tensor imaging and fiber tractography in patients with brainstem lesions. Neurol India 2010;58:886–90. [7] Holodny AI, Schwartz TH, Ollenschleger M, et al. Tumor involvement of the corticospinal tract: diffusion magnetic resonance tractography with intraoperative correlation. J Neurosurg 2001;95:1082. [8] Kovanlikaya I, Firat Z, Kovanlikaya A, et al. 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 2011;77:383–91. [9] Nagae-Poetscher LM, Jiang H, Wakana S, et al. High-resolution diffusion tensor imaging of the brain stem at 3 T. AJNR Am J Neuroradiol 2004;25:1325–30. [10] Naidich TP, Duvernoy HM, Delman BN, et al. Duvernoy’s Atlas of the human brain stem and cerebellum: high-field MRI, surface anatomy, internal structure, vascularization and 3D sectional anatomy. Wien New York: Springer; 2009. [11] Price SJ, Peña A, Burnet NG, et al. Tissue signature characterisation of diffusion tensor abnormalities in cerebral gliomas. Eur Radiol 2004;14:1909–17. [12] Masur H. Skalen und Scores in der Neurologie. Stuttgart, BadenWürttemberg: Thieme; 2000. [13] Sarnthein J, Bozinov O, Melone AG, et al. Motor-evoked potentials (MEP) during brainstem surgery to preserve corticospinal function. Acta Neurochir (Wien) 2011;153:1753–9. [14] Little DM, Kraus MF, Joseph J, et al. Thalamic integrity underlies executive dysfunction in traumatic brain injury. Neurology 2010;74:558–64. [15] Bertalanffy H, Tissira N, Krayenbühl N, et al. Inter- and intrapatient variability of facial nerve response areas in the floor of the fourth ventricle. Neurosurgery 2011;68:23–31 [discussion 31]. [16] Stieltjes B, Kaufmann WE, van Zijl PC, et al. Diffusion tensor imaging and axonal tracking in the human brainstem. Neuroimage 2001;14:723–35. [17] Yeo SS, Kim SH, Ahn YH, et al. Anatomical location of the pedunculopontine nucleus in the human brain: diffusion tensor imaging study. Stereotact Funct Neurosurg 2011;89:152–6. [18] Sedrak M, Gorgulho A, Frew A, et al. Diffusion tensor imaging and colored fractional anisotropy mapping of the ventralis intermedius nucleus of the thalamus. Neurosurgery 2011;69:1124–9 [discussion 1129–30]. [19] Jensen JH, Helpern JA. MRI quantification of non-Gaussian water diffusion by kurtosis analysis. NMR Biomed 2010;23:698–710. [20] Wu EX, Cheung MM. MR diffusion kurtosis imaging for neural tissue characterization. NMR Biomed 2010;23:836–48. [21] Nimsky C, Ganslandt O, Fahlbusch R. Implementation of fiber tract navigation. Neurosurgery 2006;58:ONS-292–303 [discussion ONS-303–4]. [22] Nimsky C, Grummich P, Sorensen AG, et al. Visualization of the pyramidal tract in glioma surgery by integrating diffusion tensor imaging in functional neuronavigation. Zentralbl Neurochir 2005;66:133–41. [23] Ulrich NH, Kockro RA, Bellut D, et al. Brainstem cavernoma surgery with the support of pre- and postoperative diffusion tensor imaging: initial experiences and clinical course of 23 patients. Neurosurg Rev 2014;37:481–92. [24] Kleiser R, Staempfli P, Valavanis A, et al. Impact of fMRI-guided advanced DTI fiber tracking techniques on their clinical applications in patients with brain tumors. Neuroradiology 2010;52:37–46.