Skull Base Institute Home Page
Skull Base Brain Tumor Research

Endoscope-assisted microvascular decompression of the trigeminal nerve
By Reza Jarrahy, M.D., George Berci, M.D., F.A.C.S., Hrayr K. Shahinian, M.D. FACS


Twenty-one patients with classical symptoms of trigeminal neuralgia underwent microvascular decompression of the trigeminal nerve via a retrosigmoid approach to the cerebellopontine angle. Endoscopy was used as an adjunctive imaging modality to microscopy. Specifically, endoscopes were used to confirm nerve-vessel conflicts identified by the microscope and to reveal others that escaped microscopic survey. Endoscopes were also used to assess the adequacy of the decompression performed microscopically. A total of 51 nerve-vessel conflicts were identified and treated, 14 of which were discovered only after endoscopy was employed. Additionally, in 5 patients endoscopic examination of the surgical intervention demonstrated that further maneuvers were required to completely decompress the nerve. These results highlight the value of endoscopy in the diagnosis and therapy of cranial nerves pathology in the posterior fossa.


The concept of using endoscopes to visualize the anatomy of the posterior fossa is not new. Until recently, however, the application of endoscopic surgical techniques in the region of the posterior skull base has been relatively rare. As early as 1917, Doyen1 described his technique for trigeminal root neurectomy under endoscopic guidance via an occipital craniotomy. Since then, published experience with endoscopy of the posterior fossa has been scarce. This may be contributed to a general unfamiliarity with the endoscopic surgical anatomy of the region, the lack of a reproducible animal model for training in endoscopic skull base surgery techniques, or to technical limitations of endoscopic equipment. In recent years these impediments have been addressed, lending increased momentum to the practice of minimally invasive surgery in the posterior fossa.

First, there has been renewed focus on endoscopic imaging of this region. O'Donoghue and O'Flynn2 described and classified the endoscopic anatomy of the cerebellopontine angle (CPA) in 1992. Since then, a number of neurotologists have presented their experiences with the use of CPA endoscopy, especially in acoustic neuroma surgery.3-6 Magnan and Sanna7 as well have recently published an atlas detailing the endoscopic surgical anatomy of the cranial nerves at the posterior skull base.

Second, the proliferation of teaching models necessary for the education of surgeons in this practice has begun. We have proposed a porcine animal model to train surgeons in the anatomy, methodology, and instrumentation involved in endoscopic surgery of the anterior and posterior skull base.8, 9 Development of such models will be essential to familiarizing practitioners with endoscopic perspectives of these intricate anatomical regions.

Third, the latest generation of endoscopic instrumentation offer devices that are specifically tailored to the needs of the skull base surgeon. These include endoscopes of varying lengths and diameters, optical properties that allow viewing over a wide range of angles, endoscope holding arms, anti-fog systems, irrigation sheaths, and sophisticated imaging and recording devices.7

Surgeons who practice microvascular decompression of the cranial nerves at the CPA stand to benefit greatly from this progress. The endoscope is particularly well suited to application in this region of the anatomy, where neurovascular structures are not always completely exposed under the microscope. Magnan10 has reported on his series of endoscope-assisted microvascular decompressions of the facial nerve for patients with hemifacial spasm. To date, observations drawn from the use of endoscopy in microvascular decompression of the trigeminal nerve have not been described.

In a series of 21 patients with trigeminal neuralgia who underwent endoscope-assisted microvascular decompression of the trigeminal nerve, we assessed the usefulness of endoscopy in identifying nerve-vessel conflicts and in determining the adequacy of therapeutic procedures performed under the microscope.

Method: Patient selection

A series of 21 patients who were referred to our service for surgical management of trigeminal neuralgia that was refractory to medical therapy underwent microvascular decompression of the trigeminal nerve. At presentation, all patients complained of classic symptoms of trigeminal neuralgia.11 All patients were found to have vascular structures associated with the trigeminal nerve at the base of the skull on preoperative MRI/MRA studies, suggesting vascular compression as an etiology of their symptoms.12, 13

Fully informed consent, including detailed explanations of the use of endoscopy as an adjunctive imaging modality, was obtained from each patient prior to surgery by the senior surgeon (HKS).

Operative technique:

A total of 21 patients received endoscope-assisted microvascular decompression of the trigeminal nerve between August 1997 and May 1999 at our institution. All procedures were performed by the same surgeon (HKS). The technique used in all cases was similar to the retrosigmoid approach to the posterior fossa described by Bremond and Garcin.14 Following the development of a c-shaped subperiosteal retroauricular skin flap, a 2-cm craniotomy is made at the confluence of the transverse and sigmoid sinuses, based upon external bony landmarks. The dura is incised in a curvilinear fashion and 4-0 neurolon sutures are used to reflect the sigmoid sinus anteriorly. As cerebrospinal fluid (CSF) is progressively drained from the pre-pontine cistern, the cerebellum begins to fall away without retraction10 revealing tentorial and emissary veins, the petrous temporal bone, the lower cranial nerves and jugular foramen, the acousticofacial bundle and internal auditory canal, and the trigeminal nerve and Meckel's cave.

An initial survey of the neurovascular anatomy is first performed in the standard fashion under the microscope. We attempt to identify all sites of compression during this survey. (Compression is defined as a grossly appreciable change in appearance of the nerve in the area of contact with a vessel, including indentation, discoloration, or evidence of trauma to the nerve.) Following this step, the microscope is removed and both 0° and 30° endoscopes (Karl Storz of America, Culver City, CA) are alternately inserted through the craniotomy under direct vision. Watching their progress directly on a monitor, they are used to conduct a panoramic inspection of Level 1 of the posterior fossa,12 where the trigeminal nerve and related vessels are viewed.

The conflicts originally identified under the microscope are confirmed endoscopically. Regions of the nerve that are not clearly visualized under the microscope are carefully examined for evidence of additional nerve-vessel conflicts. These areas include the medial and inferior faces of the nerve root entry zone, the superior, inferior, and medial aspects of the cisternal portion of the nerve, and the site of entry of the nerve into Meckel's cave. Proximal and distal mapping of all offending vascular structures is conducted under the endoscope.

Once a thorough examination of the nerve is completed and all conflicts are satisfactorily identified, the endoscopes are removed and the microscope is repositioned. Implicated venous segments are electrocoagulated and pledgets of insulating Teflon wool are interposed between the nerve and associated arteries.

After these procedures are performed, endoscopic inspection of Teflon placement is conducted and the thoroughness of the decompression is assessed. Again, special attention is given to those areas that are obscure under the microscope. Further manipulation is carried out as necessary based upon this final endoscopic survey.

At the completion of the operation, the areas of decompression are bathed in a fibrin-based sealant, the dura is tightly sutured, the bone flap is affixed, and the retroauricular skin flap is repositioned.

Data collection:

Intraoperative observations regarding the locations of nerve-vessel conflicts were noted. Specifically, the number and location of nerve-vessel conflicts observed in each patient was recorded. These were defined as either simple or complex based upon the involved neurovascular anatomy. Simple conflicts were defined as those involving a singular vascular structure (artery or vein) compressing a solitary site on the trigeminal nerve. By contrast, complex conflicts involved either more than one offending vascular component or more than one site on the nerve. Each nerve was divided topographically into three parts. Conflicts were described as involving either the nerve root entry zone, the cisternal portion of the nerve, or the site of nerve ingress to Meckel's cave. Conflict locations were further specified as being related either to the lateral, superior, medial, or inferior aspects of the nerve.

Cases where sites of conflict were not identified under microscopic examination but were revealed during endoscopic survey were noted. Cases where the final endoscopic survey revealed an inadequate decompression, prompting either repositioning of Teflon pads or placement of additional ones, were also documented.

Endoscopic techniques were deemed to have a significant impact on the operative procedure if the initial endoscopic examination identified nerve-vessel conflicts that were not seen under the microscope or if the final endoscopic survey revealed areas of the nerve not completely decompressed after microscopic placement of Teflon barriers.

All intraoperative observations were correlated with retrospective reviews of operative videotapes and dictated operation summaries.

Patient outcomes

Medical records of the follow-up visits of the 21 patients in the study were retrospectively reviewed with attention to the degree of pain relief achieved postoperatively. Patient outcomes were determined in a manner similar to that described by Benderson and Wilson.15 An excellent result was defined as complete relief of preoperative symptoms. Patients with a significant and well tolerated yet incomplete reduction in pain, including those requiring continued use of medication, were classified as having a good result. Patients with a poorly tolerated level of residual pain were deemed to have a poor result. Outcomes from this series were compared to published results from other series of patients receiving microvascular decompression of the trigeminal nerve without the use of endoscopy.15-18

Results: Patient parameters & surgical outcomes

The mean age of the patients in this series was 57 (range 32-86). Eight patients were male (38%) and 13 were female (62%). There were no perioperative mortalities. Three patients developed CSF leaks postoperatively; one of these cases was complicated by meningitis. All three were repaired surgically without subsequent complications. One patient in the series obtained a superficial wound infection that was successfully treated with a course of oral antibiotics. Another patient developed a wound abscess that required surgical drainage in addition to a course of parenteral antibiotics.

Mean postoperative follow-up in this series is 6 months (range 2-20 months). Fifteen patients (71%) had excellent relief of their preoperative trigeminal neuralgia, as defined above. Three patients (14%) reported a good overall result, and the remaining three patients (14%) reported a poor outcome.

Microscopic vs. endoscopic neurovascular anatomy

Nerve-vessel conflicts were simple in 5 of the 21 cases (24%). In the remaining 16 cases (76%), the pattern of compression of the trigeminal nerve was complex. Among the 21 patients studied, a total of 51 isolated sites of compression were observed. Table 1 shows the number of conflicts per nerve region. These are divided into those that were discovered microscopically and those that were not visible under the microscope but identified endoscopically. The nerve root entry zone was a site of compression in 19 patients (90% of total). In these patients 31 individual sites of compression were identified. Of these conflicts, 6 (19%) were not detected during microscopic survey of the trigeminal nerve. One was located superiorly, three were located inferiorly, and two were found medially.

The cisternal portion of the nerve was involved in 14 patients (67% of total); each case revealed a single conflict in this area. Of the 14 cisternal sites of compression, 5 (36%) were identified only after endoscopic survey. Two of these were on the superior aspect of the nerve, two were on the inferior aspect, and the fifth was found lying medially.

The entry of the nerve into Meckel's cave was a site of compression in 6 patients (28% of total). Again, conflicts were singular at this site. Three of the conflicts (50%) were endoscopically discovered. Two of these were superiorly oriented and the third was located lateral to the opening of the foramen.

In 5 patients (24%), endoscopic survey following microscopic placement of Teflon pledgets revealed areas of the nerve that were not adequately insulated. In these cases either the placement of additional barrier material was required (4 cases), or adjustment of the pads already in place was necessary (1 case) to fully complete the decompression.

Endoscopic procedures added an average time of approximately one-quarter to one-half hour to the overall duration of surgery and were not associated with any intraoperative complications.


Renewed interest in minimally invasive intracranial surgery coincides with advances in endoscopic engineering. Endoscopes with different optical ranges, stereoscopic endoscopes, and accessory instruments are now available to the surgeon, making endoscopy a very flexible mode of surgical imaging. As general clinical experience accumulates, the ultimate roles of intracranial endoscopic techniques will be determined. In some cases, the endoscope will serve as an important adjunct to the microscope; for some indications it will replace it.

In a series of three hundred eighty cases, Fries and Perneczky19, 20 describe the application of endoscopic techniques in various regions of the anatomy of the central nervous system. Less than 10% of these procedures involved vascular decompression of the cranial nerves at the cerebellopontine angle. This paucity underscores several salient points regarding endoscopy in this area. The neurovascular anatomy of the posterior fossa is quite intricate; exploration without thorough knowledge of its endoscopic surgical anatomy would be unwise. Cadaveric studies of the anatomy of the posterior fossa2 fall short of mimicking the intraoperative experience working in this space. Likewise, Magnan's7 operative photos elaborately depict the endoscopic perspective of posterior fossa structures but cannot duplicate vital conditions. Like our colleagues in the fields of laparoscopy,21, 22 thoracoscopy,22, 23 and arthroscopy24, 25 a reliable and reproducible animal training model is required. We have presented our experience with a porcine model as an initial step in this effort.8, 9

Nevertheless, function has followed form, and the capabilities of new endoscopes have provided for endoscopic application in surgeries that were once exclusively carried out under the microscope. Comparisons of still images obtained from microscopic and endoscopic examinations of the posterior fossa reveal striking differences. Even more impressive are the differences seen during in vivo surgical navigation of this space using these different modalities. The limitations to viewing depth and angles imposed upon the surgeon by the microscope give way to broad panoramic surveys when endoscopes are employed.

The qualitative differences between endoscopy and microscopy alone justify the use of the former in the posterior fossa. In experienced hands, endoscopic procedures pose no additional risk to the patient and add only a minimal amount of time to the total duration of surgery. The strength of the endoscope in this area is the ability it gives to surgeon to visualize areas that are typically hidden from the view of the microscope, especially when angled endoscopes are used.

These advantages are particularly useful in microvascular decompression of the cranial nerves, where pathological anatomy may be obscure under the microscope. Magnan10 concluded that the endoscope was a valuable tool in confirming microscopic observations, adding an additional 72% accuracy rate to that of the microscope in identifying nerve-vessel conflicts involving the facial nerve. Other than this series, there is a lack of published literature on experiences with endoscopy in microvascular decompression surgery.

Our observations on the relevance of endoscopy in defining trigeminal nerve conflicts are in accordance with those of Magnan. We have shown that of the fifty-one total nerve-vessel conflicts identified in our series, fourteen (27%) could only be detected endoscopically. These were in areas that were inaccessible to adequate microscopic visualization. Furthermore, we have specifically analyzed the impact of the endoscopic survey upon the actual decompression performed. In 24% of cases, an endoscopic survey of the decompression revealed areas where the nerve-vessel conflicts were not completely resolved. Again, these were in areas that were poorly exposed by the microscope.

Comparison of our patient outcomes with those of larger series is difficult, as there are no uniform criteria for categorizing surgical results. Barker and Jannetta17 quantified patients' degree of relief while Benderson and Wilson15 offered qualitative analysis. Regarding excellent, good, and poor outcomes our results are similar to those of Benderson and other series,18 perhaps suggesting that despite a subjectively better perspective and an objectively improved detection of pathology, overall surgical results are unchanged by the addition of endoscopy. Our small sample size, however, and relatively early follow-up interval allow only a preliminary commentary regarding the diagnostic and therapeutic efficacy of posterior fossa endoscopy for this indication. Final conclusions must be reserved pending completion of a greater number of cases and a longer duration from the time of surgery.

We are, however, able to conclude that in this series microscopic exploration of the surgical field failed to detect roughly one-quarter of the offending vessels. Likewise, it is evident that in one-quarter of cases what was deemed to be an adequate decompression of the trigeminal nerve under the microscope proved to be incomplete. These observations emphasize the value of controlled application of endoscopic techniques at the cerebellopontine angle and merit further attention.

  1. Doyen E. Surgical therapeutics and operative techniques. vol. I. London: Balliere, Tindall, and Cox; 1917. p. 599-602.
  2. O'Donoghue GM, O'Flynn P. Endoscopic anatomy of the cerebellopontine angle. Am J Otol 1993;14:122-5.
  3. Bottrill ID, Poe DS. Endoscope-assisted ear surgery. Am J Otol 1995;16:158-62.
  4. Jennings CR, O'Donoghue GM. Posterior fossa endoscopy. J Laryngol Otol 1998;112:227-9.
  5. Rosenberg SI. Endoscopic otologic surgery. Otolaryngol Clin N Am 1996;29:291-300.
  6. Tatagiba M, Matthies C, Samii M. Microendoscopy of the internal auditory canal in vestibular schwanoma surgery. Neurosurgery 1996;38:737-40.
  7. Magnan J, Sanna M, editors. Endoscopy in neuro-otology. New York: Thieme; 1999. p. 7-23.
  8. Jarrahy R, Young J, Berci G, et al. Endoscopic skull base surgery I: a new animal model for pituitary surgery. J Invest Surg 1999;12:1-6.
  9. Jarrahy R, Young J, Berci G, et al. Endoscopic skull base surgery II: a new animal model for surgery of the posterior fossa. J Invest Surg 1999;13:1-5.
  10. Magnan J, Caces F, Locatelli P, et al. Hemifacial spasm: endoscopic vascular decompression. Otolaryngol Head Neck Surg 1997;117:308-14.
  11. Wilkins RH. Trigeminal neuralgia: introduction. In: Wilkins RH, Rengachary SS, editors. Neurosurgery. 2nd ed. New York: McGraw-Hill; 1996. p. 3921-3.
  12. Meaney JFM, Eldridge PR, Dunn LT, et al. Demonstration of neurovascular compression in trigeminal neuralgia with magnetic resonance imaging. J Neurosurg 1995;83:799-805.
  13. Meaney JFM, Miles JB, Nixon TE, et al. Vascular contact with the fifth cranial nerve at the pons in patients with trigeminal neuralgia: detection with 3D FISP imaging. AJR 1994;163:1447-52.
  14. Bremond G, Garcin M. Microsurgical approach to the cerebellopontine angle. J Laryngol 1975;89:237-48.
  15. Benderson JB, Wilson CB. Evaluation of microvascular decompression and partial sensory rhizotomy in 252 cases of trigeminal neuralgia. J Neurosurg 1989;71:359-67.
  16. Klun B. Microvascular decompression and partial sensory rhizotomy in the treatment of trigeminal neuralgia: personal experience with 220 patients. Neurosurgery 1992;30:49-52.
  17. Barker FG, Jannetta PJ, Bissonette DJ, et al. The long term outcome of microvascular decompression for trigeminal neuralgia. NEJM 1996;334:1077-83.
  18. Sweet WH. Trigeminal neuralgia: problems as to cause and consequent conclusions regarding treatment. In: Wilkins RH, Rengachary SS, editors. Neurosurgery. 2nd ed. New York: McGraw-Hill; 1996. p. 3931-43.
  19. Perneczky A, Fries G. Endoscope-assisted brain surgery: part 1—evolution, basic concept, and current technique. Neurosurgery 1998;42:219-24.
  20. Fries G, Perneczky A. Endoscope-assisted brain surgery: part 2—analysis of 380 procedures. Neurosurgery 1998;42:226-31.
  21. Hill AB, MacKenzie KS, Steinmetz OK, et al. Videoendoscopic thoracic aorta-to-femoral artery bypass. A feasibility study in a canine model. J Vasc Surg 1998;27:948-54.
  22. Hendrickson DA, Wilson DG. Instrumentation and techniques for laparoscopic and thoracoscopic surgery in the horse. Vet Clin N Am Equ Prac 1996;12:235-59.
  23. Vachon AM, Fischer AT. Thoracoscopy in the horse: diagnostic and therapeutic indication in 28 cases. Equ Vet J 1998;30:467-75.
  24. Trotter GW, McIlwraith CW. Advances in equine arthroscopy. Vet Clin N Am Equ Prac 1996;12:261-81.
  25. Marshall KW, Chan AD. Arthroscopic anterior cruciate ligament transection induces canine osteoarthritis. J Rheum 1996;23:338-43.

Table 1. Efficacy of endoscopy vs. microscopy in detecting pathological lesions in different regions of the trigeminal nerve
Site of nerve compression Total conflicts per site Number detected microscopically (%) Number requiring endoscopy for detection (%)
root entry zone 31 25 (81%) 6 (19%)
cisternal portion 14 9 (64%) 5 (36%)
entry to Meckel's cave 6 3 (50%) 3 (50%)