- Email: [email protected]
Patterns of neurovascular compression in patients with classic trigeminal neuralgia: A high-resolution MRI-based study
European Journal of Radiology 81 (2012) 1851–1857
Contents lists available at ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Patterns of neurovascular compression in patients with classic trigeminal neuralgia: A high-resolution MRI-based study José Lorenzoni a,∗ , Philippe David b , Marc Levivier c a
Department of Neurosurgery, School of Medicine, Pontificia Universidad Católica de Chile, Chile Department of Radiology, Hôpital Erasme, Université Libre de Bruxelles, Brussels, Belgium c Department of Neurosurgery, Centre Hopitalier Universitaire Vaudois, Université de Lausanne, Switzerland b
a r t i c l e
i n f o
Article history: Received 22 June 2009 Received in revised form 16 September 2009 Accepted 17 September 2009 Keywords: Trigeminal neuralgia Nerve compression syndromes Anatomy Magnetic resonance imaging
a b s t r a c t Purpose: To describe the anatomical characteristics and patterns of neurovascular compression in patients suffering classic trigeminal neuralgia (CTN), using high-resolution magnetic resonance imaging (MRI). Materials and methods: The analysis of the anatomy of the trigeminal nerve, brain stem and the vascular structures related to this nerve was made in 100 consecutive patients treated with a Gamma Knife radiosurgery for CTN between December 1999 and September 2004. MRI studies (T1, T1 enhanced and T2-SPIR) with axial, coronal and sagital simultaneous visualization were dynamically assessed using the software GammaPlanTM . Three-dimensional reconstructions were also developed in some representative cases. Results: In 93 patients (93%), there were one or several vascular structures in contact, either, with the trigeminal nerve, or close to its origin in the pons. The superior cerebellar artery was involved in 71 cases (76%). Other vessels identified were the antero-inferior cerebellar artery, the basilar artery, the vertebral artery, and some venous structures. Vascular compression was found anywhere along the trigeminal nerve. The mean distance between the nerve compression and the origin of the nerve in the brainstem was 3.76 ± 2.9 mm (range 0–9.8 mm). In 39 patients (42%), the vascular compression was located proximally and in 42 (45%) the compression was located distally. Nerve dislocation or distortion by the vessel was observed in 30 cases (32%). Conclusions: The findings of this study are similar to those reported in surgical and autopsy series. This non-invasive MRI-based approach could be useful for diagnostic and therapeutic decisions in CTN, and it could help to understand its pathogenesis. © 2009 Elsevier Ireland Ltd. All rights reserved.
1. Introduction The existence of neurovascular compression (NVC, known as neurovascular contact or neurovascular conflict) between a blood vessel and the trigeminal nerve (TN) has been proposed as a possible etiology of classic trigeminal neuralgia; however, the exact pathogenic mechanism remains not completely understood [1–3]. Magnetic Resonance Imaging (MRI) has been used successfully in patients with trigeminal neuralgia to study the anatomy of the trigeminal nerve, the brain stem, and its vascular relationships, being this technique nowadays well validated for these purposes [4–15].
∗ Corresponding author at: Departamento de Neurocirugía, Hospital Clínico, Pontificia Universidad de Chile, Marcoleta 352, 2◦ piso, Santiago, Chile. Tel.: +0056 2 3543465. E-mail addresses: [email protected], [email protected] (J. Lorenzoni), [email protected] (P. David), [email protected] (M. Levivier). 0720-048X/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ejrad.2009.09.017
The aim of this study is to analyze retrospectively on highresolution MRI the anatomy of the trigeminal nerve, the brainstem and their neurovascular contacts in 100 consecutive patients with classic trigeminal neuralgia studied with high-resolution MRI and treated with Gamma Knife radiosurgery. Emphasis was done on the different patterns of neurovascular compression and their relative frequencies.
2. Materials and methods Between December 1999 and September 2004 one hundred consecutive patients with classic trigeminal neuralgia were treated with Gamma Knife radiosurgery (Elekta Instruments AB, Stockholm, Sweden). In all cases, the treatment was based on highresolution MRI. MRI studies were acquired with a Philips (Best, the Netherlands), Model: Intera 1.5 T, obtaining axial slices parallel to the orbitomeatal plane. Sequences obtained were T1-weighted, T1-weighted
1852
J. Lorenzoni et al. / European Journal of Radiology 81 (2012) 1851–1857
Fig. 1. (A) Multiplanar (axial, coronal and sagital) dynamic display of the MRI (T2-SPIR and T1 enhanced). (A) Delimiting the structures of interest and (B) acquisition of a 3D model.
enhanced and T2-weighted selected partial inversion recovery (T2SPIR). T1-weighted acquisition protocol was 3D Turbo field echo T1, time of acquisition: 7 min 13 s, FOV 270, Rectangular FOV 80%, Matrix 256, spatial resolution 1.05 mm × 1.42 mm × 1.3 mm, 100 slices, 1.3 mm in thickness, TR (ms)/TE/Angle 20/4.6/25◦ . T2-SPIR acquisition protocol was 3D Turbo spin echo T2 with Flow compensation SPIR FAT Suppression, time of acquisition: 9 min 3 s, FOV 270, Rectangular FOV 50%, Matrix 512 spatial resolution 0.53 mm × 0.84 mm × 1.0 mm, 50 slices, 1 MM thickness, TR (ms)/TE 3000/180. Stereotactic co-registered images were analyzed with GammaPlanTM software, version 5.34 software (Elekta Instruments AB, Stockholm, Sweden). Imaging visualization was done in the workstation in a dynamic manner, with multiplanar and multisequence display at once on the screen (Figs. 1–5). In complex and illustrative cases, 3D models were built. T2-SPIR sequence was initially used for a general anatomic approach, identifying the brainstem, the trigeminal nerve, and all the “vascular-like structures” close to the nerve and, then, T1weighted and T1-weighted enhanced sequences were used for vessel confirmation. Arteries and veins were differentiated analyzing their anatomical characteristics following each structure dynamically on the screen. To generate 3D reconstructions, all significant structures (brain stem, trigeminal nerve, arteries and veins) were outlined on the axial T2-SPIR slices (Fig. 1A). Each structure was outlined separately as a partial volume with a specific color, and then, all these volumes were displayed together, generating the 3D model (Fig. 1B). No MRI angiography was employed. The 3D model can be displayed on the screen also dynamically, this way; it can be rotated in any direction and studied from any perspective. NVC was defined as the existence of a vessel in contact with TN, according with two of the criteria described by Masur et al. [9]: 1- simple contact; 2- contact with nerve dislocation. If a layer of cerebrospinal fluid was identified between the nerve and a vessel, the neurovascular contact was not considered. Another criterion used was a contact between the vessel and the brain stem close to the nerve origin in the pons without a direct contact with the nerve itself. Nerve dislocation or distortion was defined as a nerve angulation or displacement by the vessel at level of contact.
Multiple vascular contacts were considered if two or more vessels contacted the nerve. The location of NVC was classified in 2 categories: proximal: when distance between NVC and brain stem surface was less than 3 mm [16] or when there was direct contact with the brain stem surface (but not with the nerve itself), and distal: when this distance was 3 mm or more. The frequency of each type of neurovascular contact was calculated using as a denominator the number of patients having a vascular contact with the nerve. To compare 2 × 2 contingency tables the two-sided Fisher’s exact test was used. A p value ≤0.05 was considered significant. 3. Results Among the 100 patients studied, in 93 (93%), a neurovascular contact was found on MRI in the symptomatic side. Conversely, on the contralateral asymptomatic side, only 55 cases (55%) had a neurovascular contact (p < 0.0001). 3.1. Neurovascular compression characteristics (symptomatic side) The mean distance between the nerve compression and the origin of the nerve in the brainstem was 3.76 ± 2.9 mm (range 0–9.8 mm). This vascular compression was located anywhere along the trigeminal nerve (juxtapontine, midcisternal or juxtapetrous). In 39 patients (42%) the vessel contacted the nerve in the first 3 mm of the nerve from its origin in the pons, and in 42 (45%), it was distal than 3 mm. In seven cases (7.5%) multiple contacts were found. In five patients (5%), there was an obvious contact between the vessel and the brain stem close to the trigeminal nerve origin, but not with the nerve itself, these cases were considered as proximal neurovascular contacts, i.e., in Fig. 4B, the brainstem is compressed and grooved by the vertebral artery without any vessel contacting the trigeminal nerve. The superior cerebellar artery (in two patients), the antero-inferior cerebellar artery (in one case) and a vein (in one case) were also identified. Nerve dislocation or distortion by the vessel was observed in 30 cases (32%) on the symptomatic side, on the contrary, on the asymptomatic side there was always a simple contact (p < 0.0001).
J. Lorenzoni et al. / European Journal of Radiology 81 (2012) 1851–1857
1853
Fig. 2. Examples of neurovascular compression by the superior cerebellar artery (SCA) (A and B). (In orange, the brain stem, in pink, both trigeminal nerves, in red, arterial vessels, and in blue, veins. The vascular contact is showed with an arrow). VA: vertebral artery, BA: basilar artery, V: trigeminal nerve, PCA: posterior cerebral artery, SCA: superior cerebellar artery, AICA: antero-inferior cerebellar artery, PICA: postero-inferior cerebellar artery, CPCV: cerebello-pontine scissure vein, SPV: superior petrous vein.
3.2. Vessel patterns involved in the neurovascular compression Table 1 summarizes the vessels involved in the nerve compression and their patterns. 3.2.1. Superior cerebellar artery (SCA) Usually the SCA is located medial to the nerve with a descending proximal portion, and then, a loop and an ascending distal portion. The size of this loop is variable, being more pronounced in patients with compression involving this artery, in comparison with other vessels (Figs. 2–5), the main patterns of vascular contact between the SCA with the nerve observed were: (a) Compression by the main trunk of the SCA (Fig. 2A). This pattern was the most frequent, 43 of 71 patients. (b) Contact at level of the bifurcation of the SCA. This pattern was observed in 15 cases. In two patients the bifurcation trapped the nerve with the rostral branch located medial and the caudal branch located lateral to the trigeminal nerve (Fig. 2A). (c) Contact of one or both branches of the artery after the bifurcation in 13 cases. 3.2.2. Antero-inferior cerebellar artery (AICA) The AICA is involved in the vascular compression at the main trunk. It shows a proximal ascending portion, then a loop and a descending distal portion. Usually the point of contact was at the
level of this loop (Fig. 3 A and B). This finding was isolated in seven patients, but in four cases an association with the SCA was present. 3.2.3. Basilar artery (BA) One patient presented a dolicho-ectatic basilar artery with nerve dislocation and brainstem deformation (Fig. 4A), other patient presented a tortuous artery. 3.2.4. Vertebral artery (VA) Two patients presented a tortuous vertebral artery grooving the pons just below the nerve origin but not contacting the nerve (Fig. 4B). No patient in this study presented a neurovascular compression by the postero-inferior cerebellar artery (PICA), the labyrinthine or a pontine artery. 3.2.5. Veins Vein patterns that caused a neurovascular contact were more inconsistent. The normal anatomy of the veins in this region is quite variable and it includes the superior petrous vein that arises after the junction of a transverse pontine vein, the vein of the medial cerebellar peduncle and the cerebellar-pontine scissure vein among others. The superior petrous vein goes up and laterally to find the superior petrous sinus. The venous patterns identified in contact with the nerve were:
1854
J. Lorenzoni et al. / European Journal of Radiology 81 (2012) 1851–1857
Fig. 3. Examples of neurovascular compression by the antero-inferior cerebellar artery (AICA), A and B (In orange, the brain stem, in pink, both trigeminal nerves, in red, arterial vessels, and in blue, veins. The vascular contact is showed by the arrow). VA: vertebral artery, BA: basilar artery, V: trigeminal nerve, PCA: posterior cerebral artery, SCA: superior cerebellar artery, AICA: antero-inferior cerebellar artery, PICA: postero-inferior cerebellar artery, CPCV: cerebello-pontine scissure vein, TVP: transverse pontine vein, SPV: superior petrous vein.
(a) The superior petrous vein in four patients. (b) A transverse pontine vein contacted perpendicular to the nerve in four patients (Fig. 5A). (c) The ponto-cerebellar scissure vein was present in three patients. (d) A trigeminal vein embedded the nerve running longitudinally in one case. (e) Anastomotic venous plexuses surrounding the nerve were present in one case (Fig. 5B). (f) A pontine venous angioma close to the nerve origin was present in one patient. 3.2.6. Multiple contacts In seven patients (7.5%), two or more vessels were in contact with the nerve. In four cases, it was the superior cerebellar artery and the antero-inferior cerebellar artery. In three cases, a mixed pattern existed (arterial and venous) 4. Discussion The present study confirms that high-resolution magnetic resonance allows the identification of a neurovascular compression in the majority of patients (93%) and that it is possible to study the anatomical features of these contacts as it has been done by other authors who have validated this approach [4–15]. Nevertheless, because of the nature of this study, there is not a systematic surgical or autopsy confirmation of the findings. Masur et al. [9] described more than one decade ago the use of MRI to identify neurovascular compression and proposed several
criteria. They found that nerve compression and nerve dislocation were always associated to clinical neuralgia, simple contact on the other hand was often associated to trigeminal neuralgia, but it was also present in asymptomatic patients. We used in this study two of the criteria defined by this author: simple contact and contact with nerve dislocation or distortion. The differentiation between a simple contact and a nerve compression was many times difficult. Patel et al. [12] studied 92 patients preoperatively with trigeminal neuralgia with MRI and MR angiography. In 76 (83%), the study showed a neurovascular compression in accordance with the intraoperative findings. Eight patients had non-vascular compression neither on MRI nor at surgery. On the contrary in 9 cases MRI was negative but at surgery a vascular compression was found. It is because of the fact that in this study the sensitivity of the method for the identification of a neurovascular compression was 90.5% and its specificity was 100%. Anderson et al. [5] compared the intraoperative findings during microvascular decompression surgery in 48 patients with classic trigeminal neuralgia; with the preoperative study based in high-resolution MRI (3D time-of-flight angiography and Gadolinium-enhanced 3D spoiled gradient recalled imaging). The MRI was able to identify vascular compression in 91% of the cases. The offending vessel was correctly identified in 31 of 41 cases. The author reports a sensitivity of 76% and a specificity of 75% for this method. Akimoto et al. [4] compared surgical findings with preoperative 3D reconstructions based on 3D-FISP and 3D-CISS MRI sequences
J. Lorenzoni et al. / European Journal of Radiology 81 (2012) 1851–1857
1855
Fig. 4. Examples of neurovascular compression by the basilar artery (A) and by the vertebral artery (B). (In orange, the brain stem, in pink, both trigeminal nerves, in red, arterial vessels. The vascular contact is showed by the arrow). VA: vertebral artery, BA: basilar artery, V: trigeminal nerve, PCA: posterior cerebral artery, SCA: superior cerebellar artery, AICA: antero-inferior cerebellar artery, PICA: postero-inferior cerebellar artery.
in 24 patients, in all but one case there was an excellent correlation (96%). In one case the correlation was partial, because MRI identified the SCA as a vessel contacting the nerve, but at surgery the vessel contacting the nerve was a branch of the SCA. In regard of the vessels in contact with the symptomatic nerve, the present study shows comparable findings to surgical series. Levi and Jannetta [1] in a series of 1204 operated patients, found the superior cerebellar artery (SCA) as the main vessel in contact with the nerve (75.5%), the AICA was involved in 9.6%, the vertebral artery in 1.6%, the basilar and the posterior-inferior cerebellar artery (PICA) in 0.7% and the labyrinthine artery in 0.2%. Veins were involved in 68.2% of the cases. Sindou et al. [2] report that the SCA is also the most frequent vessel contacting this nerve, ranging between 66.5% and 88%; the AICA was involved in 5.7–25%, the posterior-inferior cerebellar artery (PICA) in 0–1%, the vertebral artery in 1–3.5%, veins alone were present in 5.5–16.6%, and veins associated to any artery in 6.3–56%. The main difference in our study with the surgical series is the apparent lower proportion of veins contacting the nerve, only 14 patients (15%) in the present study; 10 patients with venous contact alone, three patients with veins and SCA contacting the nerve and one case of a venous angioma. One explanation of this difference could be the lower sensibility of MRI for detecting small venous structures. In fact, if the vessel diameter is less than the pixel size of the acquisition (1 mm) it can be missed because of the partial volume effect. Another explanation could be the lower signal of veins on MRI specially T2-SPIR because of their low flow velocity.
The present study shows that neurovascular contact can be located anywhere along the trigeminal nerve, and that these findings are in accordance with the results reported by Sindou et al. [2] who, in an anatomical study during microvascular decompression in 579 patients, found proximal or juxtapontine contacts in 52% of the patients, midcisternal contacts in 54% and juxtapetrous contacts in 10%. In our study, proximal contact was found in 42% of the patients and distal (midcisternal and juxtapetrous) in 45%. The first 3 mm was considered as proximal because in this part of the nerve the myelin sheet is formed by central oligodendroglia that has been considered more sensitive to vascular compressions [16]. In the present study, the existence of a vascular contact was more frequent in the symptomatic side 93%, but it was also present in the asymptomatic side (55%), and this difference was significant. None of the vascular contacts in the asymptomatic side produced nerve dislocation or distortion. Masur et al. [9] in a study including 18 patients with trigeminal neuralgia found a vascular compression of the nerve in 10 of the 18 (55%) asymptomatic sides, but none of them with nerve deformation or dislocation. In the same series, the symptomatic side was present in 12 of 18 patients (67%) with neurovascular contact and seven (58%) of them had a nerve compression or dislocation. The lower sensibility of the MRI in Masur series for detecting vascular contacts with the nerve could be explained because of the quality and resolution the MRI had one decade ago. Yousry et al. [14] studied 33 asymptomatic patients (66 sides) using high-resolution MRI with 3D-CISS and 3D MR angiography.
1856
J. Lorenzoni et al. / European Journal of Radiology 81 (2012) 1851–1857
Fig. 5. Examples of neurovascular compression by veins. Transverse pontine vein (A) and anastomotic plexus around the trigeminal nerve (B) (In orange, the brain stem, in pink, both trigeminal nerves, in red, arterial vessels, and in blue, veins. The vascular contact is showed by the arrow). VA: vertebral artery, BA: basilar artery, V: trigeminal nerve, PCA: posterior cerebral artery, SCA: superior cerebellar artery, AICA: antero-inferior cerebellar artery, PICA: postero-inferior cerebellar artery, CPCV: cerebello-pontine scissure vein, TVP: transverse pontine vein, SPV: superior petrous vein.
The aim of that study was to identify the sensory as well as the motor roots of the trigeminal nerve and their vascular relationships. The sensory root of the trigeminal nerve was contacted by the SCA in 45.5% of the sides, by the AICA in 4.5% of the sides and by veins in 54.5% of the sides. Motor roots were in contact only with the SCA in 48.5% of the sides analyzed. These findings were similar to Table 1 Vessels involved in the nerve compression and their patterns. Pattern of vessels involved
Number
Percentage
Superior cerebellar artery (SCA) Main trunk Bifurcation Post bifurcation Antero-inferior cerebellar artery (AICA) Main trunk Basilar artery Dolicho-ectatic vessel Tortuous vessel Vertebral artery Tortuous vessel Veins Superior petrous vein Transverse pontine vein Cerebello-pontine-scissure vein Trigeminal vein Anastomotic venous plexus Venous angioma
71 43 15 13 11 11 2 1 1 2 2 14 4 4 3 1 1 1
71 43 15 13 11 11 2 1 1 2 2 14 4 4 3 1 1 1
his findings at autopsy in seven specimens (14 sides) that served as controls. In all patients studied, the root was not distorted or grooved in any instance. Kakizawa et al. [8] studied 220 sides in 110 asymptomatic patients who underwent high-resolution 3-T MRI for unrelated to trigeminal neuralgia reasons. The sequences used were the FIESTA axial sequences of the entire brain and Time-of-flight spoiled gradient recalled acquisitions. In 99 nerves (45%) there was one vascular contact and in nine (4.1%) there were two. In 99 cases, the contact was without nerve deviation, and in 17, a mild deviation was described. There were no moderate or severe deviations in any individual in the study. According to the results reported by Kakizawa et al. [8], Masur et al. [9], Yousry et al. [14] and the present study as well, it seems that consistent nerve dislocation or distortion by the offending vessel is very specific for trigeminal neuralgia. The present study identified some cases in which a good familiarity with the patient anatomy could be important to plan the surgical treatment. The most significant cases were four patients with a basilar or a vertebral artery grooving the brainstem (Fig. 4A and B). One of these patients had surgery in other institution with a microvascular decompression without any effect on pain control (Fig. 4B). Another patient had an intraaxial pontine venous angioma without any vascular contact with the nerve, and four others presented
J. Lorenzoni et al. / European Journal of Radiology 81 (2012) 1851–1857
vascular contacts at the brain stem close to the nerve origin but not with the nerve itself. An additional application of anatomical knowledge in patients suffering trigeminal neuralgia is the possibility to establish prognosis. For example, we have reported [15] that the existence of a large vessel (basilar or vertebral artery) contacting the nerve or a proximal nerve compression is associated with a worse pain outcome after Gamma Knife treatment. 5. Conclusions The anatomical findings in this MRI-based study are comparable with other surgical series. The vascular compression is evidenced in the majority of the patients in the symptomatic side and it can be located anywhere along the trigeminal nerve. In one half of the patients it is possible to identify a vascular contact affecting the trigeminal nerve in the asymptomatic side, but the existence of nerve dislocation or distortion of the nerve by the vessel was not present in the asymptomatic side. This approach could be useful to evaluate the unique anatomy of each patient suffering classic trigeminal neuralgia, for diagnostic purposes, and to assist the selection of the best surgical treatment. This could be particularly useful in presurgical planning of a microsurgical vascular decompression of the nerve. This method could be also useful for research, helping to understand the complex and not yet well-known mechanisms involved in the pathogenesis of classic trigeminal neuralgia. Conflicts of interest None of the authors have any conflict of interest neither with Gamma Knife or Elekta A.B. Sweden nor with any company that manufactures magnetic resonance equipments. Acknowledgments Jaques Brotchi MD PhD: for providing all the facilities for the realisation of the present study. Christian Cantillano MD: for his help in assessing the redaction in English. Mr. Frederic Schoovaerts: for his help in image files recollection. The materials of this study is based on patients treated at the Gamma Knife and Department of Neurosurgery, Hôpital Erasme, Université Libre de Bruxelles, Brussels, Belgium.
1857
References [1] Levy EI, Jannetta PJ. Microvascular decompression. In: Kim J, Burchiel, editors. Surgical management of pain. New York: Thieme; 2002. p. 878–86. [2] Sindou M, Howeidy T, Acevedo G. Anatomical observations during microvascular decompression for Idiopathic Trigeminal Neuralgia (with correlations between topography of pain and site of the neurovascular conflict) Prospective Study in a series of 579 patients. Acta Neurochir (Wien) 2002;144:1–13. [3] Tyler-Kabara EC, Kassam AB, Horowitz MH, et al. Predictors of outcome in surgically managed patients with typical and atypical trigeminal neuralgia: comparison of results following microvascular decompression. J Neurosurg 2002;96:527–31. [4] Akimoto H, Nagaoka T, Nariai T, Takada Y, Ohno K, Yoshino N. Preoperative evaluation of neurovascular compression in patients with trigeminal neuralgia by use of three-dimensional reconstruction from two types of high-resolution magnetic resonance imaging. Neurosurgery 2002;51(October (4)):956–61. [5] Anderson VC, Berryhill PC, Sandquist MA, Ciavarella DP, Nesbit GM, Burchiel KJ. High-resolution three-dimensional magnetic resonance angiography and three-dimensional spoiled gradient-recalled imaging in the evaluation of neurovascular compression in patients with trigeminal neuralgia: a double-blind pilot study. Neurosurgery 2006;58:666–73. [6] Brisman R, Khandji AG, Mooij RR. Trigeminal nerve-blood vessel relationship as revealed by high-resolution magnetic resonance imaging and its effect on pain relief after gamma knife radiosurgery for trigeminal neuralgia. Neurosurgery 2002;50(June (6)):1261–7. [7] Erbay SH, Bhadelia RA, O’callaghan M, et al. Nerve atrophy in severe trigeminal neuralgia: noninvasive confirmation at MR imaging; initial experience. Radiology 2006;238:689–92. [8] Kakizawa Y, Seguchi T, Kodama K, Ogiwara T, Sasaky T, Goto THongo K. Anatomical study of the trigeminal and facial cranial nerves with the aid of 3.0-tesla magnetic resonance imaging. J Neurosurg 2008;108:483–90. [9] Masur H, Papke K, Bongartz G, Vollbretcht K. The significance of threedimensional MR-defined neurovascular compression for the pathogenesis of trigeminal neuralgia. J Neurol 1995;242:93–8. [10] Miller J, Acar F, Hamilton B, Burchiel K. Preoperative visualization of neurovascular anatomy in trigeminal neuralgia. J Neurosurg 2008;108:477–82. [11] Naraghi R, Hastreiter P, Tomandi B, Bonk A, Huk W, Fahlbusch R. Threedimensional visualization of neurovascular relationships in the posterior fossa: technique and clinical application. J Neurosurg 2004;100(June):1025–35. [12] Patel NK, Aquilina K, Clarke Y, Renowden SA, Coakham HB. How accurate is magnetic resonance angiography in predicting neurovascular compression in patients with trigeminal neuralgia? A prospective, single-blinded comparative study. Br J Neurosurg 2003;17(February (1)):60–4. [13] Satoh T, Onoda K, Date I. Preoperative simulation for microvascular decompression in patients with idiopathic trigeminal neuralgia: visualization with three-dimensional magnetic resonance cisternogram and angiogram fusion imaging. Neurosurgery 2007;60:104–13. [14] Yousry I, Moriggi B, Holtmannspoetter M, Schmid UD, Naidich TP, Yousry TA. Detailed anatomy of the motor and sensory roots of the trigeminal nerve and their neurovascular relationships: a magnetic resonance imaging study. J Neurosurg 2004;101(September):427–34. [15] Lorenzoni J, Massager N, David F, et al. Neurovascular compression anatomy and pain outcome in patients with classic trigeminal neuralgia treated by radiosurgery. Neurosurgery 2008;62:368–76. [16] De Ridder D, Moller A, Verlooy J, Cornelissen M, De Ridder L. Is the root/exit zone important in microvascular compression syndromes? Neurosurgery 2002;51:427–33.
Contents lists available at ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Patterns of neurovascular compression in patients with classic trigeminal neuralgia: A high-resolution MRI-based study José Lorenzoni a,∗ , Philippe David b , Marc Levivier c a
Department of Neurosurgery, School of Medicine, Pontificia Universidad Católica de Chile, Chile Department of Radiology, Hôpital Erasme, Université Libre de Bruxelles, Brussels, Belgium c Department of Neurosurgery, Centre Hopitalier Universitaire Vaudois, Université de Lausanne, Switzerland b
a r t i c l e
i n f o
Article history: Received 22 June 2009 Received in revised form 16 September 2009 Accepted 17 September 2009 Keywords: Trigeminal neuralgia Nerve compression syndromes Anatomy Magnetic resonance imaging
a b s t r a c t Purpose: To describe the anatomical characteristics and patterns of neurovascular compression in patients suffering classic trigeminal neuralgia (CTN), using high-resolution magnetic resonance imaging (MRI). Materials and methods: The analysis of the anatomy of the trigeminal nerve, brain stem and the vascular structures related to this nerve was made in 100 consecutive patients treated with a Gamma Knife radiosurgery for CTN between December 1999 and September 2004. MRI studies (T1, T1 enhanced and T2-SPIR) with axial, coronal and sagital simultaneous visualization were dynamically assessed using the software GammaPlanTM . Three-dimensional reconstructions were also developed in some representative cases. Results: In 93 patients (93%), there were one or several vascular structures in contact, either, with the trigeminal nerve, or close to its origin in the pons. The superior cerebellar artery was involved in 71 cases (76%). Other vessels identified were the antero-inferior cerebellar artery, the basilar artery, the vertebral artery, and some venous structures. Vascular compression was found anywhere along the trigeminal nerve. The mean distance between the nerve compression and the origin of the nerve in the brainstem was 3.76 ± 2.9 mm (range 0–9.8 mm). In 39 patients (42%), the vascular compression was located proximally and in 42 (45%) the compression was located distally. Nerve dislocation or distortion by the vessel was observed in 30 cases (32%). Conclusions: The findings of this study are similar to those reported in surgical and autopsy series. This non-invasive MRI-based approach could be useful for diagnostic and therapeutic decisions in CTN, and it could help to understand its pathogenesis. © 2009 Elsevier Ireland Ltd. All rights reserved.
1. Introduction The existence of neurovascular compression (NVC, known as neurovascular contact or neurovascular conflict) between a blood vessel and the trigeminal nerve (TN) has been proposed as a possible etiology of classic trigeminal neuralgia; however, the exact pathogenic mechanism remains not completely understood [1–3]. Magnetic Resonance Imaging (MRI) has been used successfully in patients with trigeminal neuralgia to study the anatomy of the trigeminal nerve, the brain stem, and its vascular relationships, being this technique nowadays well validated for these purposes [4–15].
∗ Corresponding author at: Departamento de Neurocirugía, Hospital Clínico, Pontificia Universidad de Chile, Marcoleta 352, 2◦ piso, Santiago, Chile. Tel.: +0056 2 3543465. E-mail addresses: [email protected], [email protected] (J. Lorenzoni), [email protected] (P. David), [email protected] (M. Levivier). 0720-048X/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ejrad.2009.09.017
The aim of this study is to analyze retrospectively on highresolution MRI the anatomy of the trigeminal nerve, the brainstem and their neurovascular contacts in 100 consecutive patients with classic trigeminal neuralgia studied with high-resolution MRI and treated with Gamma Knife radiosurgery. Emphasis was done on the different patterns of neurovascular compression and their relative frequencies.
2. Materials and methods Between December 1999 and September 2004 one hundred consecutive patients with classic trigeminal neuralgia were treated with Gamma Knife radiosurgery (Elekta Instruments AB, Stockholm, Sweden). In all cases, the treatment was based on highresolution MRI. MRI studies were acquired with a Philips (Best, the Netherlands), Model: Intera 1.5 T, obtaining axial slices parallel to the orbitomeatal plane. Sequences obtained were T1-weighted, T1-weighted
1852
J. Lorenzoni et al. / European Journal of Radiology 81 (2012) 1851–1857
Fig. 1. (A) Multiplanar (axial, coronal and sagital) dynamic display of the MRI (T2-SPIR and T1 enhanced). (A) Delimiting the structures of interest and (B) acquisition of a 3D model.
enhanced and T2-weighted selected partial inversion recovery (T2SPIR). T1-weighted acquisition protocol was 3D Turbo field echo T1, time of acquisition: 7 min 13 s, FOV 270, Rectangular FOV 80%, Matrix 256, spatial resolution 1.05 mm × 1.42 mm × 1.3 mm, 100 slices, 1.3 mm in thickness, TR (ms)/TE/Angle 20/4.6/25◦ . T2-SPIR acquisition protocol was 3D Turbo spin echo T2 with Flow compensation SPIR FAT Suppression, time of acquisition: 9 min 3 s, FOV 270, Rectangular FOV 50%, Matrix 512 spatial resolution 0.53 mm × 0.84 mm × 1.0 mm, 50 slices, 1 MM thickness, TR (ms)/TE 3000/180. Stereotactic co-registered images were analyzed with GammaPlanTM software, version 5.34 software (Elekta Instruments AB, Stockholm, Sweden). Imaging visualization was done in the workstation in a dynamic manner, with multiplanar and multisequence display at once on the screen (Figs. 1–5). In complex and illustrative cases, 3D models were built. T2-SPIR sequence was initially used for a general anatomic approach, identifying the brainstem, the trigeminal nerve, and all the “vascular-like structures” close to the nerve and, then, T1weighted and T1-weighted enhanced sequences were used for vessel confirmation. Arteries and veins were differentiated analyzing their anatomical characteristics following each structure dynamically on the screen. To generate 3D reconstructions, all significant structures (brain stem, trigeminal nerve, arteries and veins) were outlined on the axial T2-SPIR slices (Fig. 1A). Each structure was outlined separately as a partial volume with a specific color, and then, all these volumes were displayed together, generating the 3D model (Fig. 1B). No MRI angiography was employed. The 3D model can be displayed on the screen also dynamically, this way; it can be rotated in any direction and studied from any perspective. NVC was defined as the existence of a vessel in contact with TN, according with two of the criteria described by Masur et al. [9]: 1- simple contact; 2- contact with nerve dislocation. If a layer of cerebrospinal fluid was identified between the nerve and a vessel, the neurovascular contact was not considered. Another criterion used was a contact between the vessel and the brain stem close to the nerve origin in the pons without a direct contact with the nerve itself. Nerve dislocation or distortion was defined as a nerve angulation or displacement by the vessel at level of contact.
Multiple vascular contacts were considered if two or more vessels contacted the nerve. The location of NVC was classified in 2 categories: proximal: when distance between NVC and brain stem surface was less than 3 mm [16] or when there was direct contact with the brain stem surface (but not with the nerve itself), and distal: when this distance was 3 mm or more. The frequency of each type of neurovascular contact was calculated using as a denominator the number of patients having a vascular contact with the nerve. To compare 2 × 2 contingency tables the two-sided Fisher’s exact test was used. A p value ≤0.05 was considered significant. 3. Results Among the 100 patients studied, in 93 (93%), a neurovascular contact was found on MRI in the symptomatic side. Conversely, on the contralateral asymptomatic side, only 55 cases (55%) had a neurovascular contact (p < 0.0001). 3.1. Neurovascular compression characteristics (symptomatic side) The mean distance between the nerve compression and the origin of the nerve in the brainstem was 3.76 ± 2.9 mm (range 0–9.8 mm). This vascular compression was located anywhere along the trigeminal nerve (juxtapontine, midcisternal or juxtapetrous). In 39 patients (42%) the vessel contacted the nerve in the first 3 mm of the nerve from its origin in the pons, and in 42 (45%), it was distal than 3 mm. In seven cases (7.5%) multiple contacts were found. In five patients (5%), there was an obvious contact between the vessel and the brain stem close to the trigeminal nerve origin, but not with the nerve itself, these cases were considered as proximal neurovascular contacts, i.e., in Fig. 4B, the brainstem is compressed and grooved by the vertebral artery without any vessel contacting the trigeminal nerve. The superior cerebellar artery (in two patients), the antero-inferior cerebellar artery (in one case) and a vein (in one case) were also identified. Nerve dislocation or distortion by the vessel was observed in 30 cases (32%) on the symptomatic side, on the contrary, on the asymptomatic side there was always a simple contact (p < 0.0001).
J. Lorenzoni et al. / European Journal of Radiology 81 (2012) 1851–1857
1853
Fig. 2. Examples of neurovascular compression by the superior cerebellar artery (SCA) (A and B). (In orange, the brain stem, in pink, both trigeminal nerves, in red, arterial vessels, and in blue, veins. The vascular contact is showed with an arrow). VA: vertebral artery, BA: basilar artery, V: trigeminal nerve, PCA: posterior cerebral artery, SCA: superior cerebellar artery, AICA: antero-inferior cerebellar artery, PICA: postero-inferior cerebellar artery, CPCV: cerebello-pontine scissure vein, SPV: superior petrous vein.
3.2. Vessel patterns involved in the neurovascular compression Table 1 summarizes the vessels involved in the nerve compression and their patterns. 3.2.1. Superior cerebellar artery (SCA) Usually the SCA is located medial to the nerve with a descending proximal portion, and then, a loop and an ascending distal portion. The size of this loop is variable, being more pronounced in patients with compression involving this artery, in comparison with other vessels (Figs. 2–5), the main patterns of vascular contact between the SCA with the nerve observed were: (a) Compression by the main trunk of the SCA (Fig. 2A). This pattern was the most frequent, 43 of 71 patients. (b) Contact at level of the bifurcation of the SCA. This pattern was observed in 15 cases. In two patients the bifurcation trapped the nerve with the rostral branch located medial and the caudal branch located lateral to the trigeminal nerve (Fig. 2A). (c) Contact of one or both branches of the artery after the bifurcation in 13 cases. 3.2.2. Antero-inferior cerebellar artery (AICA) The AICA is involved in the vascular compression at the main trunk. It shows a proximal ascending portion, then a loop and a descending distal portion. Usually the point of contact was at the
level of this loop (Fig. 3 A and B). This finding was isolated in seven patients, but in four cases an association with the SCA was present. 3.2.3. Basilar artery (BA) One patient presented a dolicho-ectatic basilar artery with nerve dislocation and brainstem deformation (Fig. 4A), other patient presented a tortuous artery. 3.2.4. Vertebral artery (VA) Two patients presented a tortuous vertebral artery grooving the pons just below the nerve origin but not contacting the nerve (Fig. 4B). No patient in this study presented a neurovascular compression by the postero-inferior cerebellar artery (PICA), the labyrinthine or a pontine artery. 3.2.5. Veins Vein patterns that caused a neurovascular contact were more inconsistent. The normal anatomy of the veins in this region is quite variable and it includes the superior petrous vein that arises after the junction of a transverse pontine vein, the vein of the medial cerebellar peduncle and the cerebellar-pontine scissure vein among others. The superior petrous vein goes up and laterally to find the superior petrous sinus. The venous patterns identified in contact with the nerve were:
1854
J. Lorenzoni et al. / European Journal of Radiology 81 (2012) 1851–1857
Fig. 3. Examples of neurovascular compression by the antero-inferior cerebellar artery (AICA), A and B (In orange, the brain stem, in pink, both trigeminal nerves, in red, arterial vessels, and in blue, veins. The vascular contact is showed by the arrow). VA: vertebral artery, BA: basilar artery, V: trigeminal nerve, PCA: posterior cerebral artery, SCA: superior cerebellar artery, AICA: antero-inferior cerebellar artery, PICA: postero-inferior cerebellar artery, CPCV: cerebello-pontine scissure vein, TVP: transverse pontine vein, SPV: superior petrous vein.
(a) The superior petrous vein in four patients. (b) A transverse pontine vein contacted perpendicular to the nerve in four patients (Fig. 5A). (c) The ponto-cerebellar scissure vein was present in three patients. (d) A trigeminal vein embedded the nerve running longitudinally in one case. (e) Anastomotic venous plexuses surrounding the nerve were present in one case (Fig. 5B). (f) A pontine venous angioma close to the nerve origin was present in one patient. 3.2.6. Multiple contacts In seven patients (7.5%), two or more vessels were in contact with the nerve. In four cases, it was the superior cerebellar artery and the antero-inferior cerebellar artery. In three cases, a mixed pattern existed (arterial and venous) 4. Discussion The present study confirms that high-resolution magnetic resonance allows the identification of a neurovascular compression in the majority of patients (93%) and that it is possible to study the anatomical features of these contacts as it has been done by other authors who have validated this approach [4–15]. Nevertheless, because of the nature of this study, there is not a systematic surgical or autopsy confirmation of the findings. Masur et al. [9] described more than one decade ago the use of MRI to identify neurovascular compression and proposed several
criteria. They found that nerve compression and nerve dislocation were always associated to clinical neuralgia, simple contact on the other hand was often associated to trigeminal neuralgia, but it was also present in asymptomatic patients. We used in this study two of the criteria defined by this author: simple contact and contact with nerve dislocation or distortion. The differentiation between a simple contact and a nerve compression was many times difficult. Patel et al. [12] studied 92 patients preoperatively with trigeminal neuralgia with MRI and MR angiography. In 76 (83%), the study showed a neurovascular compression in accordance with the intraoperative findings. Eight patients had non-vascular compression neither on MRI nor at surgery. On the contrary in 9 cases MRI was negative but at surgery a vascular compression was found. It is because of the fact that in this study the sensitivity of the method for the identification of a neurovascular compression was 90.5% and its specificity was 100%. Anderson et al. [5] compared the intraoperative findings during microvascular decompression surgery in 48 patients with classic trigeminal neuralgia; with the preoperative study based in high-resolution MRI (3D time-of-flight angiography and Gadolinium-enhanced 3D spoiled gradient recalled imaging). The MRI was able to identify vascular compression in 91% of the cases. The offending vessel was correctly identified in 31 of 41 cases. The author reports a sensitivity of 76% and a specificity of 75% for this method. Akimoto et al. [4] compared surgical findings with preoperative 3D reconstructions based on 3D-FISP and 3D-CISS MRI sequences
J. Lorenzoni et al. / European Journal of Radiology 81 (2012) 1851–1857
1855
Fig. 4. Examples of neurovascular compression by the basilar artery (A) and by the vertebral artery (B). (In orange, the brain stem, in pink, both trigeminal nerves, in red, arterial vessels. The vascular contact is showed by the arrow). VA: vertebral artery, BA: basilar artery, V: trigeminal nerve, PCA: posterior cerebral artery, SCA: superior cerebellar artery, AICA: antero-inferior cerebellar artery, PICA: postero-inferior cerebellar artery.
in 24 patients, in all but one case there was an excellent correlation (96%). In one case the correlation was partial, because MRI identified the SCA as a vessel contacting the nerve, but at surgery the vessel contacting the nerve was a branch of the SCA. In regard of the vessels in contact with the symptomatic nerve, the present study shows comparable findings to surgical series. Levi and Jannetta [1] in a series of 1204 operated patients, found the superior cerebellar artery (SCA) as the main vessel in contact with the nerve (75.5%), the AICA was involved in 9.6%, the vertebral artery in 1.6%, the basilar and the posterior-inferior cerebellar artery (PICA) in 0.7% and the labyrinthine artery in 0.2%. Veins were involved in 68.2% of the cases. Sindou et al. [2] report that the SCA is also the most frequent vessel contacting this nerve, ranging between 66.5% and 88%; the AICA was involved in 5.7–25%, the posterior-inferior cerebellar artery (PICA) in 0–1%, the vertebral artery in 1–3.5%, veins alone were present in 5.5–16.6%, and veins associated to any artery in 6.3–56%. The main difference in our study with the surgical series is the apparent lower proportion of veins contacting the nerve, only 14 patients (15%) in the present study; 10 patients with venous contact alone, three patients with veins and SCA contacting the nerve and one case of a venous angioma. One explanation of this difference could be the lower sensibility of MRI for detecting small venous structures. In fact, if the vessel diameter is less than the pixel size of the acquisition (1 mm) it can be missed because of the partial volume effect. Another explanation could be the lower signal of veins on MRI specially T2-SPIR because of their low flow velocity.
The present study shows that neurovascular contact can be located anywhere along the trigeminal nerve, and that these findings are in accordance with the results reported by Sindou et al. [2] who, in an anatomical study during microvascular decompression in 579 patients, found proximal or juxtapontine contacts in 52% of the patients, midcisternal contacts in 54% and juxtapetrous contacts in 10%. In our study, proximal contact was found in 42% of the patients and distal (midcisternal and juxtapetrous) in 45%. The first 3 mm was considered as proximal because in this part of the nerve the myelin sheet is formed by central oligodendroglia that has been considered more sensitive to vascular compressions [16]. In the present study, the existence of a vascular contact was more frequent in the symptomatic side 93%, but it was also present in the asymptomatic side (55%), and this difference was significant. None of the vascular contacts in the asymptomatic side produced nerve dislocation or distortion. Masur et al. [9] in a study including 18 patients with trigeminal neuralgia found a vascular compression of the nerve in 10 of the 18 (55%) asymptomatic sides, but none of them with nerve deformation or dislocation. In the same series, the symptomatic side was present in 12 of 18 patients (67%) with neurovascular contact and seven (58%) of them had a nerve compression or dislocation. The lower sensibility of the MRI in Masur series for detecting vascular contacts with the nerve could be explained because of the quality and resolution the MRI had one decade ago. Yousry et al. [14] studied 33 asymptomatic patients (66 sides) using high-resolution MRI with 3D-CISS and 3D MR angiography.
1856
J. Lorenzoni et al. / European Journal of Radiology 81 (2012) 1851–1857
Fig. 5. Examples of neurovascular compression by veins. Transverse pontine vein (A) and anastomotic plexus around the trigeminal nerve (B) (In orange, the brain stem, in pink, both trigeminal nerves, in red, arterial vessels, and in blue, veins. The vascular contact is showed by the arrow). VA: vertebral artery, BA: basilar artery, V: trigeminal nerve, PCA: posterior cerebral artery, SCA: superior cerebellar artery, AICA: antero-inferior cerebellar artery, PICA: postero-inferior cerebellar artery, CPCV: cerebello-pontine scissure vein, TVP: transverse pontine vein, SPV: superior petrous vein.
The aim of that study was to identify the sensory as well as the motor roots of the trigeminal nerve and their vascular relationships. The sensory root of the trigeminal nerve was contacted by the SCA in 45.5% of the sides, by the AICA in 4.5% of the sides and by veins in 54.5% of the sides. Motor roots were in contact only with the SCA in 48.5% of the sides analyzed. These findings were similar to Table 1 Vessels involved in the nerve compression and their patterns. Pattern of vessels involved
Number
Percentage
Superior cerebellar artery (SCA) Main trunk Bifurcation Post bifurcation Antero-inferior cerebellar artery (AICA) Main trunk Basilar artery Dolicho-ectatic vessel Tortuous vessel Vertebral artery Tortuous vessel Veins Superior petrous vein Transverse pontine vein Cerebello-pontine-scissure vein Trigeminal vein Anastomotic venous plexus Venous angioma
71 43 15 13 11 11 2 1 1 2 2 14 4 4 3 1 1 1
71 43 15 13 11 11 2 1 1 2 2 14 4 4 3 1 1 1
his findings at autopsy in seven specimens (14 sides) that served as controls. In all patients studied, the root was not distorted or grooved in any instance. Kakizawa et al. [8] studied 220 sides in 110 asymptomatic patients who underwent high-resolution 3-T MRI for unrelated to trigeminal neuralgia reasons. The sequences used were the FIESTA axial sequences of the entire brain and Time-of-flight spoiled gradient recalled acquisitions. In 99 nerves (45%) there was one vascular contact and in nine (4.1%) there were two. In 99 cases, the contact was without nerve deviation, and in 17, a mild deviation was described. There were no moderate or severe deviations in any individual in the study. According to the results reported by Kakizawa et al. [8], Masur et al. [9], Yousry et al. [14] and the present study as well, it seems that consistent nerve dislocation or distortion by the offending vessel is very specific for trigeminal neuralgia. The present study identified some cases in which a good familiarity with the patient anatomy could be important to plan the surgical treatment. The most significant cases were four patients with a basilar or a vertebral artery grooving the brainstem (Fig. 4A and B). One of these patients had surgery in other institution with a microvascular decompression without any effect on pain control (Fig. 4B). Another patient had an intraaxial pontine venous angioma without any vascular contact with the nerve, and four others presented
J. Lorenzoni et al. / European Journal of Radiology 81 (2012) 1851–1857
vascular contacts at the brain stem close to the nerve origin but not with the nerve itself. An additional application of anatomical knowledge in patients suffering trigeminal neuralgia is the possibility to establish prognosis. For example, we have reported [15] that the existence of a large vessel (basilar or vertebral artery) contacting the nerve or a proximal nerve compression is associated with a worse pain outcome after Gamma Knife treatment. 5. Conclusions The anatomical findings in this MRI-based study are comparable with other surgical series. The vascular compression is evidenced in the majority of the patients in the symptomatic side and it can be located anywhere along the trigeminal nerve. In one half of the patients it is possible to identify a vascular contact affecting the trigeminal nerve in the asymptomatic side, but the existence of nerve dislocation or distortion of the nerve by the vessel was not present in the asymptomatic side. This approach could be useful to evaluate the unique anatomy of each patient suffering classic trigeminal neuralgia, for diagnostic purposes, and to assist the selection of the best surgical treatment. This could be particularly useful in presurgical planning of a microsurgical vascular decompression of the nerve. This method could be also useful for research, helping to understand the complex and not yet well-known mechanisms involved in the pathogenesis of classic trigeminal neuralgia. Conflicts of interest None of the authors have any conflict of interest neither with Gamma Knife or Elekta A.B. Sweden nor with any company that manufactures magnetic resonance equipments. Acknowledgments Jaques Brotchi MD PhD: for providing all the facilities for the realisation of the present study. Christian Cantillano MD: for his help in assessing the redaction in English. Mr. Frederic Schoovaerts: for his help in image files recollection. The materials of this study is based on patients treated at the Gamma Knife and Department of Neurosurgery, Hôpital Erasme, Université Libre de Bruxelles, Brussels, Belgium.
1857
References [1] Levy EI, Jannetta PJ. Microvascular decompression. In: Kim J, Burchiel, editors. Surgical management of pain. New York: Thieme; 2002. p. 878–86. [2] Sindou M, Howeidy T, Acevedo G. Anatomical observations during microvascular decompression for Idiopathic Trigeminal Neuralgia (with correlations between topography of pain and site of the neurovascular conflict) Prospective Study in a series of 579 patients. Acta Neurochir (Wien) 2002;144:1–13. [3] Tyler-Kabara EC, Kassam AB, Horowitz MH, et al. Predictors of outcome in surgically managed patients with typical and atypical trigeminal neuralgia: comparison of results following microvascular decompression. J Neurosurg 2002;96:527–31. [4] Akimoto H, Nagaoka T, Nariai T, Takada Y, Ohno K, Yoshino N. Preoperative evaluation of neurovascular compression in patients with trigeminal neuralgia by use of three-dimensional reconstruction from two types of high-resolution magnetic resonance imaging. Neurosurgery 2002;51(October (4)):956–61. [5] Anderson VC, Berryhill PC, Sandquist MA, Ciavarella DP, Nesbit GM, Burchiel KJ. High-resolution three-dimensional magnetic resonance angiography and three-dimensional spoiled gradient-recalled imaging in the evaluation of neurovascular compression in patients with trigeminal neuralgia: a double-blind pilot study. Neurosurgery 2006;58:666–73. [6] Brisman R, Khandji AG, Mooij RR. Trigeminal nerve-blood vessel relationship as revealed by high-resolution magnetic resonance imaging and its effect on pain relief after gamma knife radiosurgery for trigeminal neuralgia. Neurosurgery 2002;50(June (6)):1261–7. [7] Erbay SH, Bhadelia RA, O’callaghan M, et al. Nerve atrophy in severe trigeminal neuralgia: noninvasive confirmation at MR imaging; initial experience. Radiology 2006;238:689–92. [8] Kakizawa Y, Seguchi T, Kodama K, Ogiwara T, Sasaky T, Goto THongo K. Anatomical study of the trigeminal and facial cranial nerves with the aid of 3.0-tesla magnetic resonance imaging. J Neurosurg 2008;108:483–90. [9] Masur H, Papke K, Bongartz G, Vollbretcht K. The significance of threedimensional MR-defined neurovascular compression for the pathogenesis of trigeminal neuralgia. J Neurol 1995;242:93–8. [10] Miller J, Acar F, Hamilton B, Burchiel K. Preoperative visualization of neurovascular anatomy in trigeminal neuralgia. J Neurosurg 2008;108:477–82. [11] Naraghi R, Hastreiter P, Tomandi B, Bonk A, Huk W, Fahlbusch R. Threedimensional visualization of neurovascular relationships in the posterior fossa: technique and clinical application. J Neurosurg 2004;100(June):1025–35. [12] Patel NK, Aquilina K, Clarke Y, Renowden SA, Coakham HB. How accurate is magnetic resonance angiography in predicting neurovascular compression in patients with trigeminal neuralgia? A prospective, single-blinded comparative study. Br J Neurosurg 2003;17(February (1)):60–4. [13] Satoh T, Onoda K, Date I. Preoperative simulation for microvascular decompression in patients with idiopathic trigeminal neuralgia: visualization with three-dimensional magnetic resonance cisternogram and angiogram fusion imaging. Neurosurgery 2007;60:104–13. [14] Yousry I, Moriggi B, Holtmannspoetter M, Schmid UD, Naidich TP, Yousry TA. Detailed anatomy of the motor and sensory roots of the trigeminal nerve and their neurovascular relationships: a magnetic resonance imaging study. J Neurosurg 2004;101(September):427–34. [15] Lorenzoni J, Massager N, David F, et al. Neurovascular compression anatomy and pain outcome in patients with classic trigeminal neuralgia treated by radiosurgery. Neurosurgery 2008;62:368–76. [16] De Ridder D, Moller A, Verlooy J, Cornelissen M, De Ridder L. Is the root/exit zone important in microvascular compression syndromes? Neurosurgery 2002;51:427–33.