Clinical features and surgical treatment of trigeminal neuralgia caused solely by venous compression

Transcription

1 Acta Neurochir (2011) 153: DOI /s x CLINICAL ARTICLE Clinical features and surgical treatment of trigeminal neuralgia caused solely by venous compression Wenyao Hong & Xuesheng Zheng & Zhenghai Wu & Xinyuan Li & Xuhui Wang & Yi Li & Wenchuan Zhang & Jun Zhong & Xuming Hua & Shiting Li Received: 14 November 2010 / Accepted: 19 January 2011 / Published online: 6 February 2011 # Springer-Verlag 2011 Abstract Purpose To summarize our experience and lessons of microvascular decompression surgery for trigeminal neuralgia caused solely by venous compression. Methods Fifteen patients with idiopathic trigeminal neuralgia caused by venous compression only underwent microvascular decompression. The entire course of the trigeminal root was explored thoroughly; and coagulating and cutting techniques were preferred in decompressing the culprit veins. Their clinical features, outcomes and operative complications were analyzed. Results The compressing veins included the transverse pontine vein in five cases (33.3%), the transverse pontine vein and the vein of middle cerebellar peduncle in one (6.7%), the transverse pontine vein and the vein of cerebellopontine fissure in one (6.7%), the superior petrosal vein in three (20%), the pontotrigeminal vein in one (6.7%), the vein of the cerebellopontine fissure in two (13.3%), and the plexus venosus or venule in two (13.3%). After microvascular decompression, 11 cases (73.3%) had excellent or good pain relief. Four cases (26.7%) failed the first surgery; and two of them underwent re-operation and got excellent pain relief. Wenyao Hong and Xuesheng Zheng contributed equally to this study. W. Hong : X. Zheng (*) : Z. Wu : X. Li : X. Wang : Y. Li : J. Zhong : S. Li Department of Neurosurgery, XinHua Hospital, Affiliated to Shanghai JiaoTong University School of Medicine, 1665 Kongjiang Road, Yangpu District, Shanghai , China xueshengzheng@gmail.com W. Zhang : J. Zhong : X. Hua : S. Li (*) The Cranial Nerve Disease Center of Shanghai, 1665 Kongjiang Road, Yangpu District, Shanghai , China pheiphei@163.com Postoperative facial numbness appeared in four cases, due to injury to trigeminal nerve when coagulation. Conclusion The transverse pontine vein is the most common offending vein. For this type of trigeminal neuralgia, coagulating and cutting techniques are preferred in decompressing the culprit veins. The entire course of the trigeminal root should be explored and decompressed. Following these principles, excellent or good pain relief could be achieved in most cases; and recurrence is rare. However, sometimes injury to the nerve is unavoidable when coagulating the culprit vein. Keywords Microvascular decompression. Trigeminal neuralgia. Venous compression Abbreviations TN Trigeminal neuralgia MVD Microvascular decompression NVC Neuro-Vascular Conflict REZ Root Entry Zone Introduction Trigeminal neuralgia (TN) is a syndrome characterized by paroxysmal facial pain. Microvascular decompression (MVD) surgery, which is based on the neurovascular conflict (NVC) hypothesis, is widely accepted as the standard treatment for TN [4, 8, 10]. Many studies regarding the causes of TN and the surgical results of MVD have been reported, and arterial compression was believed to be the main cause in most cases of TN. There are few articles studying idiopathic TN caused by venous compression [7, 9, 11, 16]. However, we think venous

2 1038 Acta Neurochir (2011) 153: compression as the pathogenesis of TN deserves more attention of neurosurgeons, because treatment tactic to TN attributable to veins is special and the recurrence rate for this type is very high [11]. In this study, we report our findings on the surgical treatment for TN cases in which only the vein was found to conflict with the trigeminal nerve. Patients and methods Patient population During July 2008 to August 2009, 343 patients of idiopathic TN underwent MVD at The Cranial Nerve Disease Center of Shanghai, Xinhua Hospital (Table 1). Among them, 15 TN cases were caused by venous compression. Fourteen patients (93.3%) were women. The age of the patients at the surgery ranged from 38 to 78 years (mean age, 55.6 years). All patients had unilateral facial pain. Three divisions of the trigeminal nerve were affected in one (6.7%) case, two divisions were affected in nine (60%), and one division was affected in five (33.3%). The V2 division was the most commonly affected area in 13 cases (86.7%). Nine patients had a typical history of TN, while the other six had atypical symptoms. Typical TN is defined by four features: (1) lancinating and electrical pain in one or more trigeminal nerve distributions; (2) having definite trigger points; (3) having definite trigger stimuli or activities; (4) memorable onset of TN. Atypical TN denotes the following: (1) no definite trigger point; (2) having intermittent facial pain or persistent pain; (3) suffering from facial numbness or dysesthesia [13]. None were treated successfully by medical therapy. None had undergone surgical treatment before admission to our center. The duration of pain ranged from 2 to 125 months (mean, 62 months). As a clinical routine for TN, pre-operative threedimensional time-of-flight sequence magnetic resonance imaging ((3D-TOF MRI) examination was done in all cases to identifiy the relationships between the trigeminal nerve and the vessels, except for just one patient who had a metal stent placement. However, on the MRI films, paraneural vessels were found in none of the 14 cases (Fig. 1). In addition, paraneural tumor was also excluded by MRI. Surgical treatment For all the cases, we performed MVD using the suboccipital superior-lateral cerebellar approach as described by Li et al. [13]. Patients with caused by venous compression only were included, i.e. the entire course of the trigeminal root was carefully explored under the microscope to exclude any possibility of arterial contact and compression. Since the culprit vein usually adhered firmly to the brain stem and cranial nerves, it is almost impossible to transpose the course of the culprit vein. Coagulating and cutting techniques were preferred in decompressing the culprit veins (used in 12 cases). When we coagulated the offending vein, we tried the best to keep the tips of the bipolar Table 1 Patient population and anatomical observation and follow-up results ID Sex Age (years) duration (months) Division affected Typical Offending veins Site of conflicts severity of conflicts Outcome Facial numbness Follow-up duration (months) 1 F V2 Typical Transverse pontine Along Grade II Excellent No Lost 2 F V2 Atypical Transverse pontine Petrous Grade I Excellent No 22 3 F V2, V3 Typical Transverse pontine REZ Grade III Failed Yes 15 & middle cerebellar peduncle 4 F 38 2 V1, V2, V3 Atypical Plexus venosus Cisternal Grade II Failed Yes 20 5 F V1, V2 Typical Transverse pontine REZ Grade II Good (delayed) No 14 6 F V2 Atypical Superior petrosal Cisternal Grade I Good No 21 7 F V2, V3 Typical Cerebellopontine fissure REZ Grade II Failed (1st) Yes 21 Excellent (2nd) 8 F V1, V2 Typical Venule Petrous Grade III Excellent No 14 9 F V2, V3 Atypical Transverse pontine REZ Grade I Excellent Yes F V3 typical Cerebellopontine fissure Petrous Grade I Failed (1st) Yes 21 transverse pontine Excellent (2nd) 11 F V3 Typical Superior petrosal REZ Grade II Excellent No F V2, V3 Typical Superior petrosal Petrous Grade II Excellent No 26 (delayed) 13 F V2, V3 Atypical Pontotrigeminal REZ Grade II Excellent No lost 14 F V2, V3 Atypical Cerebellopontine fissure REZ Grade I Excellent Yes M V2, V3 Typical Transverse pontine Along Grade II Excellent Yes 22

3 Acta Neurochir (2011) 153: indentation of the root was present [17]. Nerve alterations and surrounding abnormalities were also noted down when present. Results Fig. 1 A typical 3D-TOF MRI image of the TN patients caused by venous compression. There is no NVC in the pre-operative MRI. N trigeminal nerve, A artery, REZ root entry zone coagulation forceps away from the trigeminal nerve and minimal power (3-5 W) was used to avoid injury to the nerve. While coagulation is dangerous, in the other three cases Teflon felts were interposed into the gap between the trigeminal nerve and the culprit vein, so as to protect the nerve and preserve the main drainage. Outcome evaluation Operative results, including pain relief and operative complications, were assessed at discharge, and also by periodic telephone surveys (every 3 months). Outcome was graded as excellent (free from facial pain, or at least 98% pain-free), good (pain was relieved by 75% or greater) and failure (pain relief was less than 75%). The follow-up period ranged from 14 to 26 months, with a mean of 19 months. Anatomical observation Under the microscope, it was possible to look at the entire trigeminal root, from the porus of Meckel s cave to the root entry zone (REZ). The main anatomical features collected for this study were: the presence or the absence of vessels in contact or conflict with the trigeminal nerve, the degree of severity of the conflict, and its site and relation to the nerve. In order to classify the location of NVC, the trigeminal nerve root was divided into three parts, i.e., trigeminal REZ, cisternal portion (at the midthird of the root) and the petrous segment (near Meckel s cave) [17]. The degree of neurovascular compression severity was graded as following. Grade I: when the vessel was in simple contact with the root but without any visible indentation. Grade II: when the vessel was in contact with and adhered to the root. Grade III: when a marked Presence of offending veins In 13 (86.7%) of the 15 cases, single venous compression was found during surgery. The remaining two cases (13.3%) had two veins compressing the nerve (Table 1). Based on the nomenclature that Rhoton previously reported [15], the compressing veins included the transverse pontine vein in five cases (33.3%), the transverse pontine vein and the vein of the middle cerebellar peduncle in one (6.7%), the transverse pontine vein and the vein of the cerebellopontine fissure in one (6.7%), the superior petrosal vein in three (20%), the pontotrigeminal vein in one (6.7%), the vein of the cerebellopontine fissure in two (13.3%), and the plexus venosus or venule in two (13.3%). An anatomical sketch was provided (Fig. 2) to illustrate these seven types of neurovascular relationships. In those three cases with superior petrosal vein compression, the interposing method was applied to preserve the superior petrosal vein. Site and severity of the conflicts In four patients (26.7%), the NVCs located at the trigeminal REZ; in two cases (13.3%) at the cisternal portion; in seven cases (46.7%) at the petrous segment of trigeminal root (near the Meckel s cave); and in the other two cases (13.3%) the offending vein was attached to the trigeminal root along the entire course. In five cases (33.3%) the severity of conflicts was Grade I; and in eight cases (53.3%) Grade II. In the other two cases, the severity was regarded as Grade III, since both patients had marked indentation and trigeminal nerve atrophy, i.e., the caliber diminished by one-third or more, and the root looked ribbon-shaped and grayish with marked angulation or indentation [13, 17]. Outcome The outcome was graded by personnel other than the operating surgeon. Nine cases (60%) were free of facial pain, achieving an excellent outcome, two cases (13.3%) had a good outcome, and four cases (26.7%) failed. Among the failed, one patient (no. 4) was atypical, having constant numbness and slight pain on the face rather than paroxysmal severe pain, so we did not recommend reoperation for her; therefore, the exact cause to the failure of the surgery is unknown. Another patient (no. 3) was cured by radiofrequency thermocoagulation of Gasser s ganglion 1 day after the first operation; thus, the cause of failure is again unknown. The other two patients (nos. 7 and 10) had a second operation for MVD 1 week and 1 month respectively after the first MVD, and both achieved

4 1040 Acta Neurochir (2011) 153: Fig. 2a g The spacious relationships between the trigeminal nerve and the offending veins. a The transverse pontine vein was medial to the nerve. b The superior petrosal vein ascended and crossed the nerve, and emptied into the superior petrosal sinus. c The vein of the cerebellopontine fissure crossed the fissue and inferiorly compressed the REZ of the nerve. d The transverse pontine vein and the vein of the middle cerebellar peduncle laterally compressed the nerve together. e The pontotrigeminal vein passed medial to the nerve. f The plexus venosus was attached and went through the nerve. g The venule tied the nerve and caused an evident indentation. N trigeminal nerve, BS brain stem, CL cerebellum, V vein excellent outcomes immediately after the re-operation. It turned out that another culprit vein (the transverse pontine vein) had been missed in the first operation of case no. 10. After that vein was coagulated and cut in the second operation, the pain relieved completely (Fig. 3a, b). As for case no. 7, in the first MVD we found that the offending vein ran between the trigeminal motor root and the sensory root, so we coagulated it partially and then tried to interpose it using a small piece of Teflon sponge. In the second MVD, we had to coagulate it completely and cut it (Fig. 3c, d); although the pain was relieved completely, facial numbness was inevitable. In this study, it is interesting to find that two cases showed delayed healing. One of them (no. 5) had pain relief 2 months after MVD, and the other one (no. 12) had pain relief 1 month after. Complications There were no serious operative complications in this population. Before the operation, two cases (nos. 4 and 9) already had facial numbness, which did not change significantly after surgery; therefore, they were not regarded as operative complications. In case no. 3, facial numbness appeared after the procedure of radiofrequency thermocoagulation of Gasser s ganglion; thus it was unavoidable. There were another four patients (nos. 7, 10, 14, and 15) who had facial numbness, because the trigeminal nerves were impaired significantly due to coagulation. Recurrence There was no recurrence during the follow-up. Discussion The role of pure venous compression as the cause of idiopathic TN is uncertain yet. In this study, nine cases achieved excellent pain relief and two cases were good. Based on the fact that the possibility of arterial compression was excluded intraoperatively and venous compression on the nerve was clear in these 11 cases, we would postulate venous compression to be one of the causes of TN, as some papers have reported [3, 20]. The veins that commonly compress the trigeminal nerve are tributaries of the superior petrosal vein; namely, the

5 Acta Neurochir (2011) 153: Fig. 3 Patients with TN caused by venous compression who underwent second microvascular decompression. a Case no. 10: the transverse pontine vein (arrow) had been missed in the first operation; b this vein was coagulated and cut in the second operation; the pain relieved completely, but the nerve was burned. c Case no. 7: the offending vein (arrow) runs between the trigeminal motor root and the sensory root, so it was coagulated partially and thenasmallpieceofteflon sponge was interposed in the first MVD; d in the second MVD, the offending vein was coagulated completely and cut; the pain was relieved completely, but facial numbness was inevitable. N trigeminal nerve, V vein transverse pontine vein, the pontotrigeminal vein, and the vein of cerebellopontine fissure and the vein of middle cerebellar peduncle. Some autopsy studies concluded that the transverse pontine vein was the most common offending vein [14]. Our findings support this statement. The transverse pontine vein, which is usually attached to the superior-medial or inferior surface of the trigeminal nerve just in front of the Meckel s cave, was neglected during the first procedure of MVD in one case (no. 10). Although other offending tributaries were less presented, they should not be ignored either, as multiple compression may exist, otherwise it might lead to surgical failure. A thorough understanding of the venous anatomy around the trigeminal nerve may help surgeons to identify the culprit veins accurately. REZ is considered to be the common site of pathology [5], as it is the locus of transition from central to peripheral myelin, a more vulnerable part of the nerve [8, 18, 23]. However, decompression of REZ was not enough. Thorough exploration and decompression of the whole trigeminal root from Meckel s cave to pons is recommended [6, 12, 17]. This was also applied to MVD of the trigeminal nerve caused by only venous compression, because NVCs are located not only at the REZ but also all along the root, and multiple offending vessels may occur in the same patient. Although we were very cautious to avoid burning the nerve when coagulating the culprit vein, there were four cases (26.7%) of postoperative facial numbness due to heat of coagulation. All these four patients achieved excellent pain relief (two of them got pain relief after re-operation). It seems that the pain relief was the result of postoperative facial numbness. However, the relation of pain relief with facial numbness after the MVD procedure is disputed. It has been suggested that MVD relieves pain because of trauma to the trigeminal nerve, like the ablative procedures [1, 19]. However, Barker et al. [2] found no evidence that postoperative facial numbness predicted pain relief after MVD, and he suggested that trauma to the trigeminal root during MVD should be avoided whenever possible, because facial numbness predicted postoperative burning and aching facial pain. John M. Tew commented: Although this conclusion may be true, there is incontrovertible evidence that partial injury to the trigeminal root will relieve TN [21, 22]. In our opinion, although sometimes injury to the trigeminal root and facial numbness is almost inevitable (Fig.3b), we should avoid it as far as possible, since the essence of MVD for TN is pain relief without complication. As Lee et al. reported [11], TN attributable to venous compression recurred in about 31% cases after MVD; and if pain recurs, it is likely to recur within 1 year after the initial operation. However, in this group there was no recurrence during the follow-up (14-26 months), because we preferred to coagulate and cut the offending veins, except the very big one. Most patients in this study are female, which indicates that women are more susceptible to TN caused by vein compression, in accordance with the report of Lee et al. [11]. Another interesting phenomenon is that two patients

6 1042 Acta Neurochir (2011) 153: showed delayed healing, which is very common in hemifacial spasm, but has been little reported in TN as yet. We do not know its mechanism. Conclusion Venous compression is one of the causes of idiopathic TN. Females are more susceptible to TN caused by vein compression. The transverse pontine vein is the most common offending vein. For this type of TN, coagulating and cutting techniques are preferred in decompressing the culprit veins. The entire course of the trigeminal root should be explored and decompressed. Following these principles, excellent or good pain relief could be achieved in most cases; and recurrence is rare. However, sometimes injury to the nerve is unavoidable when coagulating the culprit vein. Acknowledgements This study was supported by a grant from the National Science Foundation of China (NSFC, No ), and by the New Hundred Talents Program of Shanghai JiaoTong University School of Medicine (to Xuesheng Zheng). Conflicts of interest References None. 1. Adams CB (1989) Microvascular compression: an alternative view and hypothesis. J Neurosurg 70: Barker FG 2nd, Jannetta PJ, Bissonette DJ, Jho HD (1997) Trigeminal numbness and tic relief after microvascular decompression for typical trigeminal neuralgia. Neurosurgery 40: Barker FG 2nd, Jannetta PJ, Bissonette DJ, Larkins MV, Jho HD (1996) The long-term outcome of microvascular decompression for trigeminal neuralgia. N Engl J Med 334: Borucki L, Szyfter W, Wrobel M, Sosnowski P (2006) Neurovascular conflicts. Otolaryngol Pol 60: Calvin WH, Loeser JD, Howe JF (1977) A neurophysiological theory for the pain mechanism of tic douloureux. Pain 3: Hai J, Li ST, Pan QG (2006) Treatment of atypical trigeminal neuralgia with microvascular decompression. Neurol India 54:53 56, discussion Helbig GM, Callahan JD, Cohen-Gadol AA (2009) Variant intraneural vein-trigeminal nerve relationships: an observation during microvascular decompression surgery for trigeminal neuralgia. Neurosurgery 65: , discussion Jannetta PJ (1967) (2007) Arterial compression of the trigeminal nerve at the pons in patients with trigeminal neuralgia. J Neurosurg 107: Kimura T, Sako K, Tohyama Y, Yonemasu Y (1999) Trigeminal neuralgia caused by compression from petrosal vein transfixing the nerve. Acta Neurochir (Wien) 141: Kureshi SA, Wilkins RH (1998) Posterior fossa reexploration for persistent or recurrent trigeminal neuralgia or hemifacial spasm: surgical findings and therapeutic implications. Neurosurgery 43: Lee SH, Levy EI, Scarrow AM, Kassam A, Jannetta PJ (2000) Recurrent trigeminal neuralgia attributable to veins after microvascular decompression. Neurosurgery 46: , discussion Li ST, Pan Q, Liu N, Shen F, Liu Z, Guan Y (2004) Trigeminal neuralgia: what are the important factors for good operative outcomes with microvascular decompression. Surg Neurol 62: , discussion Li ST, Wang X, Pan Q, Hai J, Liu N, Shen F, Liu Z, Guan Y (2005) Studies on the operative outcomes and mechanisms of microvascular decompression in treating typical and atypical trigeminal neuralgia. Clin J Pain 21: Matsushima T, Huynh-Le P, Miyazono M (2004) Trigeminal neuralgia caused by venous compression. Neurosurgery 55: , discussion Rhoton AL Jr (2000) The cerebellopontine angle and posterior fossa cranial nerves by the retrosigmoid approach. Neurosurgery 47:S93 S Sato O, Kanazawa I, Kokunai T (1979) Trigeminal neuralgia caused by compression of trigeminal nerve by pontine vein. Surg Neurol 11: Sindou M, Howeidy T, Acevedo G (2002) 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) 144:1 12, discussion Sindou M, Leston J, Decullier E, Chapuis F (2007) Microvascular decompression for primary trigeminal neuralgia: long-term effectiveness and prognostic factors in a series of 362 consecutive patients with clear-cut neurovascular conflicts who underwent pure decompression. J Neurosurg 107: Steiger HJ (1991) Prognostic factors in the treatment of trigeminal neuralgia. Analysis of a differential therapeutic approach. Acta Neurochir (Wien) 113: Sun T, Saito S, Nakai O, Ando T (1994) Long-term results of microvascular decompression for trigeminal neuralgia with reference to probability of recurrence. Acta Neurochir (Wien) 126: Taha JM, Tew JM Jr (1996) Comparison of surgical treatments for trigeminal neuralgia: reevaluation of radiofrequency rhizotomy. Neurosurgery 38: Taha JM, Tew JM Jr, Buncher CR (1995) A prospective 15-year follow up of 154 consecutive patients with trigeminal neuralgia treated by percutaneous stereotactic radiofrequency thermal rhizotomy. J Neurosurg 83: van Loveren H, Tew JM Jr, Keller JT, Nurre MA (1982) A 10- year experience in the treatment of trigeminal neuralgia. Comparison of percutaneous stereotaxic rhizotomy and posterior fossa exploration. J Neurosurg 57: