Endonasal endoscopic approach to the pterygopalatine and infratemporal fossae.

The pterygopalatine fossa and infratemporal fossa are spaces located under the skull base, housing important neurovascular structures. Surgical access to these spaces is challenging because of their deep location and complex anatomy. Their surgical access has been classically carried out through multiple craniofacial approaches until the advent of endoscopic endonasal surgery at the end of the XXth century. Our goal is to describe the transmaxillary-transsphenoidal-transpterygoid approach to the pterygopalatine and infratemporal fossae through endonasal endoscopic surgery based on anatomo-surgical dissection and an illustrative clinical case. We conclude that after careful radiologic evaluation of the feasibility of this technique, the endonasal endoscopic access to these spaces for tumor resection is efficient with reduced surgical morbidities. The endonasal approach is versatile and can be fashioned according to the nature and extent of the lesion.

Novel application of computer-assisted cisternal endoscopy for the biopsy of pineal region tumors: cadaveric study.

BACKGROUND: Long-standing debate continues about the management and biopsy of pineal tumors because of their complex microsurgical anatomy and deep location. Inspired by the concept of biopsy under direct visualization in the absence of hydrocephalus, we explored the effectiveness of neuroendoscope outside of its traditional territory using a new minimally invasive technique, computer-assisted cisternal endoscopy (CACE), for the biopsy of pineal tumors. METHOD: Five cadaver heads were dissected to expose the pineal region through the posterior fossa. In the other 5 heads, a rigid endoscope-wand combination was introduced in the supracerebellar space lateral to the arachnoid of the superior cerebellar cistern in midline. Endoscopic exposure of the pineal gland was correlated with the real-time image of the localizing wand. After the wand was removed, arachnoid was further dissected from the deep veins and the pineal gland, and a four-quadrant biopsy was obtained. FINDINGS: The combination of technologies of frameless guided stereotaxy and neuroendoscopy enhanced our ability to navigate the ventriculoscope in narrow spaces (e.g., posterior fossa cisterns). Compared with transventricular and conventional stereotactic trajectories, application of CACE in supracerebellar infratentorial trajectory offered the shortest route to the pineal region, anatomical orientation, no violation of eloquent neurovascular structures, and adequate visibility to deep veins and arteries. CONCLUSIONS: CACE may be used to approach pineal lesions outside the cerebral ventricular system for biopsy or debulking. Continuous computer updates on the endoscope position allows its safe navigation in narrow spaces (e.g., cerebrospinal fluid cistern). Its success will await future surgical trials.

Evidence for the improved exposure of the ophthalmic segment of the internal carotid artery after anterior clinoidectomy: morphometric analysis.

BACKGROUND: Although resection of the anterior clinoid process (ACP) is valuable in the surgical treatment of aneurysms of the ophthalmic (C6) segment of the internal carotid artery (ICA), quantitative assessment of this adjunct is incomplete. Our morphometric study assesses the effectiveness of the anterior clinoidectomy for exposure of the C6 segment of the ICA. METHODS: Ten formalin-fixed adult cadaveric heads were dissected bilaterally and pterional craniotomies were performed bilaterally. Measurements before and after resection of the ACP included the length of C6 segment of the ICA on its lateral aspect; C6 segment length on its medial aspect; and medial length of the optic nerve from the optic chiasm to falciform ligament (before ACP resection) then to the annulus of Zinn (after ACP resection). FINDINGS: Height and width of the intradural ACP were 8.67 +/- 2.63 and 6.57 +/- 1.68 mm, respectively. After clinoidectomy, mean length of the lateral C6 segment of the ICA increased 60% and mean exposure of the medial C6 segment of the ICA increased 113% (p < 0.001). Exposure of the optic nerve increased 150% (p < 0.001) after clinoidectomy and sectioning of the falciform ligament. No correlations were found between the lengths of the ACP and entire C6 segment, or the ACP size and amount of the C6 segment covered by the clinoid. CONCLUSIONS: Exposure of the C6 segment of the ICA is markedly increased by increase of the mobility of the optic nerve with clinoidectomy and section of the falciform ligament.

The one-piece orbitozygomatic approach: the MacCarty burr hole and the inferior orbital fissure as keys to technique and application.

OBJECTIVE: Use of the MacCarty keyhole burr hole and the inferior orbital fissure provides simplicity and safety to perform the one-piece frontotemporal orbitozygomatic (FTOZ1) approach. METHODS: We performed the FTOZ1 approach with its three subtypes (i.e., total, temporal, and frontal) in cadaveric head specimens in the Goodyear Laboratory and subsequently in surgical cases. RESULTS: The orbitozygomatic osteotomy, when added to a frontotemporal craniotomy, comprises the frontotemporal orbitozygomatic (FTOZ) approach, provides an expanded exposure to the anterior and middle cranial fossae, and enables the surgeon to create a window to the posterior cranial fossa. The MacCarty burr hole is used to facilitate orbital cuts, and the anterolateral portion of the inferior orbital fissure connects the orbital cuts to the zygomatic cuts. This allows the FTOZ1 craniotomy flap to be "out-fractured" with ease. The three types of FTOZ1 approach, i.e., the total, the temporal, and the frontal, are described step by step. CONCLUSIONS: Understanding the MacCarty keyhole burr hole and the microsurgical anatomy of the inferior orbital fissure is essential to performing the FTOZ1 approach. The three types of FTOZ1 approach enable the surgeon to tailor the approach according to the surgical exposure needed for each lesion.

Expanding the boundaries of the transsphenoidal approach: a microanatomic study.

The anatomic features of a transsphenoidal approach are reviewed, focusing on the microsurgical anatomy of parasellar structures. Pertinent microsurgical anatomy is described in sufficient detail for the neurosurgeon to successfully extend a standard transsphenoidal approach for treatment of lesions involving the region of the tuberculum sellae, planum sphenoidale, supradiaphragmatic intradural space, and medial cavernous sinus. The parasellar region of 50 formalin-fixed cadaveric heads was examined by using magnification 3x to 40x. The arterial and venous systems of five cadaveric specimens were injected under pressure with colored silicone rubber. The sellar region of three specimens was examined histologically. Important anatomic landmarks identified in the roof of the sphenoid sinus include a carotid and trigeminal prominence, as well as a tubercular, clival, and opticocarotid recess. The diaphragma sella is actually comprised of two layers of dura, with a venous system (circular sinus) interposed between the layers. The dura mater of the pituitary gland separates the gland from the medial compartment of the cavernous sinus. The microanatomic detail necessary to extend the transsphenoidal approach to the supradiaphragmatic intradural space and medial compartment of the cavernous sinus is described. These data are presented to facilitate the clinical application of these extended approaches.

Petroclival meningiomas: predictive parameters for transpetrosal approaches.

OBJECTIVE: To determine parameters that influence the selection of the proper petrosal approach or combined approaches for the excision of petroclival meningiomas. METHODS: We dissected 15 cadaver heads, inspected the petroclival region in 50 dry human skulls, and performed a retrospective analysis of the cases of 35 patients with petroclival meningiomas who underwent surgery via transpetrosal approaches. RESULTS: The petroclival region was divided into three "zones" based on the extent of surgical exposure achieved via the petrosal approaches with microscopic dissection of 15 preserved and silicone-injected cadaveric heads and with the measurements of 50 dry skulls. Zone I, defined as the area from the dorsum sellae to the internal auditory canal, is accessible via the anterior petrosal approach. Zone II, defined as the area from the internal auditory canal to the upper border of the jugular tubercle, is easily accessible in its lateral portion via the posterior petrosal approach. The medial portion of Zone II, the "central clival depression," is accessible only with cochlear resection and posterior facial nerve transposition. Zone III, defined as the area from the upper border of the jugular tubercle to the lower edge of the foramen magnum, is accessible via a suboccipital/transcondylar approach. The retrospective analysis of the cases of 35 patients who underwent transpetrosal resection of petroclival meningiomas between 1991 and 1998 was used to determine the predictive value of these anatomic parameters. The degree of tumor resection was analyzed with a novel grading scale combining the percentage of resection and the percentage of brainstem reexpansion. Total excision was achieved in 37% of the patients and complete brainstem reexpansion was achieved in an additional 40%. Residual tumor was concentrated in the central clival depression in Zone II, as predicted by anatomic parameters, and around infiltrated neurovascular structures. New cranial nerve deficit occurred in 31% of the patients in the early postoperative period and improved to 17% at 6 months. Major morbidity occurred in 9% of the patients, and mortality was 0%. Early Karnofsky scores were reduced in 37% of the patients, but 6-month Karnofsky scores were equal to preoperative baseline scores or improved in 91%. CONCLUSION: Anatomic parameters can predict the resectability of petroclival meningiomas. Judicious application of cytoreductive surgery in selected patients maintains an acceptable morbidity and achieves adequate brainstem reexpansion.

Microsurgical anatomic features and nomenclature of the paraclinoid region.

OBJECTIVE: We describe the detailed microsurgical anatomic features of the clinoid (C5) segment of the internal carotid artery (ICA) and surrounding structures, clarify the anatomic relationships of structures in this region, and emphasize the clinical relevance of these observations. Furthermore, because the nomenclature of the paraclinoid region is confusing and lacks standardization, this report provides a glossary of terms that are commonly used to descibe the anatomic features of the paraclinoid region. METHODS: The region surrounding the anterior clinoid process was observed in 70 specimens from 35 formalin-fixed cadaveric heads. Detailed microanatomic dissections were performed in 10 specimens. Histological sections of this region were obtained from the formalin-fixed cadaveric specimens. RESULTS: The clinoid segment of the ICA is the portion that abuts the clinoid process. This portion of the ICA can be directly observed only after removal of the clinoid process. The dura of the cavernous sinus roof separates to enclose the clinoid process. The clinoid segment of the ICA exists only where this separation of dural layers is present. Because the clinoid process does not completely enclose the ICA in most cases, the clinoid segment is shaped more like a wedge than a cylinder. The outer layer of the dura (dura propria) is a thick membrane that fuses with the adventitia of the ICA to form a competent ring that separates the intradural ICA from the extradural ICA. The thin inner membranous layer of the dura loosely surrounds the ICA throughout the entire length of its clinoid segment. The most proximal aspect of this membrane defines the proximal dural ring. The proximal ring is incompetent and admits a variable number of veins from the cavernous plexus that accompany the ICA throughout its clinoid segment. CONCLUSION: The narrow space between the inner dural layer and the clinoid ICA is continuous with the cavernous sinus via an incompetent proximal dural ring. This space between the clinoid ICA and the inner dural layer contains a variable number of veins that directly communicate with the cavernous plexus. Given the inconstancy of the venous plexus surrounding the clinoid ICA, we think that categorical labeling of the clinoid ICA as intracavernous or extracavernous cannot be justified.

Colored silicone injection for use in neurosurgical dissections: anatomic technical note.

OBJECTIVE: The dissection of cadaveric specimens is very important for a more sophisticated understanding of neurosurgical anatomic features and approaches. Teaching known approaches to residents or learning new approaches is best performed in a cadaveric laboratory. The utility of neurosurgical cadaveric dissections can be improved by injecting the intracranial vascular tree with colored silicone. The vascular anatomic features, which are integral to neurosurgical procedures, are much more clearly defined in injected specimens. METHODS: Self-curing colored silicone rubber is used to inject the arteries and veins (red and blue, respectively) of the head. This process is described in a step-by-step format. Six steps are required and can be summarized as follows: 1) exposure of the great vessels, 2) cannulation of the great vessels, 3) irrigation of the head, 4) preparation of the colored silicone, 5) injection of the colored silicone, and 6) evaluation of the final specimen. CONCLUSION: Injection of colored silicone into the vascular tree can enhance the educational value of cadaveric head dissections. This report describes the technique of vascular injection that is used in the Goodyear Microsurgical Laboratory, the University of Cincinnati, and the Mayfield Clinic.

The enigmatic foramen lacerum.

OBJECTIVE: The microanatomic details of the foramen lacerum and surrounding region are described to clarify the relationship between the internal carotid artery and the foramen lacerum. The terminology related to these structures is reevaluated. Examples of pathological abnormalities restricted to the foramen lacerum region are presented to document the clinical relevance of this region. METHODS: Microanatomic dissections were performed in 12 formalin-fixed cadaveric specimens. Bony landmarks were examined in 50 dry skulls. Microscopic sections of the region were obtained from cadaveric specimens that were formalin-fixed, decalcified, and processed for histological examination. RESULTS: The foramen lacerum is not a true foramen. No significant structures traverse its fibrocartilage. In this region, the bony and fibrous structures surround the internal carotid artery to form an incomplete canal, which serves as the rostral extension of petrous canal. CONCLUSION: The term foramen lacerum should be restricted to that portion of the cranial base at the confluence of the petrous portion of the temporal, basioccipital, and basisphenoid bones that in vivo is filled with fibrocartilage. The region immediately above the foramen lacerum, occupied by the internal carotid artery and traditionally considered the upper portion of the foramen lacerum, should be considered, anatomically and functionally, to be the rostral extension of the petrous canal. We suggest calling this extension the lacerum portion of the carotid canal.

Meningeal architecture of the cavernous sinus: clinical and surgical implications.

OBJECTIVE: The meningeal structure of the cavernous sinus (lateral sellar compartment) was anatomically and histologically studied. We discuss the clinical and surgical significance and present clinical examples of cranial base tumors. METHODS: Ten adult cadaveric heads were used for microsurgical dissection or histological studies. Specimens of the cavernous sinus were continuously sectioned in three dimensions and stained by Masson's trichrome method. The findings are anatomically discussed as they pertain to presented clinical cases. RESULTS: The cavernous sinus, located in an interdural space between periosteal and meningeal dura, is properly accessed by detachment of the periosteal bridge between the superior orbital fissure and the middle fossa. The lateral meningeal dura is dissected under minimal hemorrhage from the sinus, with a surgically important cleaving plane between the "deep layer," a semitransparent meningeal sheath with which the cranial nerves are covered and protected. It has various degrees of meningeal pockets, of which Meckel's cave is the largest example. Adventitia of the carotid artery in the sinus, uncovered with protective meninges, is considered to contact directly with tumors of the sinus origin. The meningeal wall of the cavernous sinus anatomically has three weak points as far as tumor invasion and extension are concerned: the venous plexus around the superior orbital fissure, the loose texture of the medial wall around the pituitary body, and dural pockets of the IIIrd and Vth cranial nerves. The dural wall is extremely thin or missing at those points. CONCLUSION: A surgical technique based on the meningeal anatomy is important for cavernous sinus surgery. The cavernous apex and Meckel's cave, which are spaces of convergence of cranial nerves, however, are weak points for surgical dissection. The presence or absence of tumor invasion into those areas may influence the microsurgical results.

Segments of the internal carotid artery: a new classification.

This study proposes an anatomically based nomenclature for the internal carotid artery (ICA) that can be applied by all disciplines. In 1938, Fischer published a seminal paper describing five segments of the ICA that were designated C1 through C5. These segments were based on the angiographic course of the intracranial ICA rather than its arterial branches or anatomic compartments. Subsequent attempts to apply modern nomenclature to these numerical segments failed to recognize Fischer's original intent of describing patterns of arterial displacement by tumors and, therefore, resulted in a nomenclature that was anatomically inaccurate. Fischer's system was further limited, because segments were numbered opposite the direction of blood flow and the extracranial ICA was excluded. The authors propose a new classification, which includes the entire ICA, uses a numerical scale in the direction of blood flow, and describes the segments of the ICA according to a detailed understanding of the anatomy surrounding the ICA and the compartments through which it travels. Twenty cadaveric specimens with intravascular injection of silicone rubber were used for microscopic dissection and 20 dry skulls were inspected. Histological sections in critical areas were examined. The authors' classification has the following seven segments: C1, cervical; C2, petrous; C3, lacerum; C4 cavernous; C5, clinoid; C6, ophthalmic; and C7, communicating. This classification is practical, accounts for new anatomic information and clinical interests, and clarifies all segments of the ICA.

Innervation of the spinal dura. Myth or reality?

STUDY DESIGN: Cranial and spinal dura from nine Sprague Dawley male rats were examined immunocytochemically for the presence of nerve fibers and mast cells and for innervation. The posterior longitudinal ligament and peridural membrane were also examined for these elements. OBJECTIVE: To examine the pattern of sensory innervation and the presence of mast cells in rat spinal dura. SUMMARY OF BACKGROUND DATA: The cranial dura is richly innervated and has a robust population of mast cells, which have been implicated in the pathogenesis of vascular headache. Moskowitz's explanation for vascular headache focused on the dura mater and neurogenic inflammation. Essential to his model are dural trigeminovascular fibers and mast cells. Previous studies provide contradictory and inconclusive results regarding spinal dural innervation. METHODS: Immunocytochemical techniques using antibodies to calcitonin gene-related peptide and substance P were used to identify sensory nerve fibers and antibodies to serotonin were used to identify mast cells. Specimens dissected included dura of the cranial vault in continuity with the dorsal cervical dura, dura of the skull base in continuity with the ventral cervical dura, lumbar dura, and posterior longitudinal ligament from the cervical and lumbar regions. RESULTS: A rich neural network and an abundant mast cell population were identified in the supratentorial and infratentorial cranial dura, both dorsally and ventrally. A paucity of nerve fibers and mast cells was observed in the cervical and lumbar dura; in contrast, these elements were prominent in the posterior longitudinal ligament and peridural membrane. CONCLUSIONS: Spinal dura does not have a rich innervation of calcitonin gene-related peptide- and substance P-positive nerve fibers or a robust population of mast cells. The spinal dura may serve as a protective covering. Unlike the cranial dura, it may not be implicated in the pathogenesis of pain. Additional studies on primate and human spinal dura are warranted to corroborate findings that the spinal dura may be relatively insensitive.

Immunohistochemical localization of endothelial nitric oxide synthase in vessels of the dura mater of the Sprague-Dawley rat.

Nitric oxide (NO) and the dura mater are implicated in the pathogenesis of vascular headache. Many studies have demonstrated the participation of NO in headache; however, few studies have identified NO in the dura mater. In this study, nine Sprague-Dawley rats were examined with immunohistochemistry using two different endothelial nitric oxide synthase (eNOS) monoclonal antibodies, H32 and ECNOS. eNOS was successfully localized to the endothelium of the middle meningeal artery. To the best of our knowledge, this is the first study to report NOS immunopositive endothelial cells in the blood vessels of the rat dura mater. The authors propose that NO plays an active role in dural vasodilation, contributing to the pathogenesis of vascular headache; in the future, NO inhibitors could serve as pharmacological agents to treat vascular headache.

Comparison of conventional and skull base surgical approaches for the excision of trigeminal neurinomas.

Trigeminal neurinomas have traditionally been excised through conventional approaches. Because symptomatic tumor recurrence exceeds 50% after conventional procedures, the authors evaluated the use of skull base approaches to achieve complete resection and a lower rate of symptomatic recurrence. Comparisons of skull base with conventional approaches to trigeminal neurinomas have been limited to small series with short-term follow-up periods. The authors reviewed their experiences with conventional (frontotemporal transsylvian, subtemporal-intradural, subtemporal-transtentorial, and suboccipital) and skull base (frontotemporal extradural-intradural, frontoorbitozygomatic, subtemporal anterior petrosal, and presigmoid posterior petrosal) surgical approaches for the excision of trigeminal neurinomas. In this paper they report the results of 15 patients with trigeminal neurinoma who underwent 27 surgical procedures between 1980 and 1990. Seventeen of the procedures used conventional and 10 used skull base approaches. All patients had tumors arising from Meckel's cave and the porus trigeminus either initially or on recurrence. Tumors located in the cavernous sinus recurred most frequently (83%); other tumors that recurred frequently were those located in Meckel's cave and the porus trigeminus (67%), and the posterior fossa (17%). The tumor extended into the anterolateral wall of the cavernous sinus in 38% of patients with cavernous sinus involvement. Tumor exposure and ease of dissection were superior with skull base approaches. Residual or recurrent tumors were found in 65% of patients following conventional approaches compared with 10% of patients following skull base approaches. Using skull base approaches, the surgeon was more accurate (90%) in estimating tumor excision than when using conventional approaches (43%). Perioperative complications were similar with both. The authors discuss the indications, advantages, and limitations of each approach. Based on anatomical considerations, they propose a strategy to best resect these tumors.

The far lateral/combined supra- and infratentorial approach. A human cadaveric prosection model for routes of access to the petroclival region and ventral brain stem.

A far lateral approach to the ventral brain stem, lower clivus, and anterior foramen magnum is described. Methods for further exposure of the superior petroclival region by incorporating a subtemporal craniotomy and posterior petrosectomy are also demonstrated. Eight sequentially illustrated steps depict this technique. The far lateral/combined supra- and infratentorial exposure is a comprehensive surgical approach that provides direct access to the entire anterior and lateral brain stem and craniovertebral junction. It minimizes brain-stem retraction and maximizes visualization of the neurovascular structures.

Effect of subarachnoid haemorrhage on trigeminovascular calcitonin-gene-related peptide and substance P of the rat dura mater versus cerebral vasculature.

While the presence of a robust perivascular neural network accompanying cerebral and dural blood vessels that contain various neuropeptides is well documented, the functional significance of this innervation is unclear. Following experimentally induced subarachnoid haemorrhage (SAH) in animal models, immunocytochemical studies have revealed that changes occur in the staining intensity of some of these neuropeptides. This study compared the immunostaining intensity of calcitonin-gene-related peptide (CGRP) and substance P (SP) in cerebral and dural perivascular nerve fibers after SAH in the rat. Subarachnoid haemorrhage was produced by injecting 0.3 ml of autologous blood into the cisterna magna of male Sprague Dawley rats. Sham operated animals received an equal volume of buffered lactated Ringer's solution (pH 7.4). Changes in the immunostaining intensity of cerebral and dural vessels were evaluated by independent observers at 6, 24, and 48 hours after SAH. Immunostaining of CGRP was reduced in cerebral vessels at 6 hours and returned to normal by 48 hours. In contrast, CGRP immunostaining of dural perivascular nerve fibers was unchanged at all time periods examined. A marked decrease in SP immunostaining was documented at 6 hours in both the cerebral and dural vessels in all animals; at 48 hours, the staining intensity had returned to control levels. These results support the idea that several subpopulations of trigeminovascular neurons containing CGRP, SP, or both project to cerebral and dural vessels. Since these subpopulations may be differentially activated in pathologic conditions, such as SAH or vascular headache, the potential exists for pharmacologic intervention of specific neuropeptides with the resultant abatement of a pathologic process.

Nitric oxide synthase immunoreactivity in the rat dura mater.

The dura mater has been implicated as a tissue where vascular headache develops. Identification of the neural components of this tissue is a prerequisite for understanding the mechanisms of this pathological process. The nitric oxide molecule, a potent vasodilator, may contribute to the vascular headache process by dilating dural vasculature. Our immunohistochemical study using nitric oxide synthase (NOS) antibodies revealed NOS-positive nerve fibers and a prominent mast cell population in the rat dura. A majority of the immunopositive fibers were associated with the anterior meningeal artery and its branches and sparse innervation with the middle meningeal artery, its branches, and superior sagittal sinus. We propose that the NOS-positive nerve fibers and mast cells be considered as possible participants in the pathogenesis of vascular headache.

Serotonin (5-HT) fibers of the rat dura mater: 5-HT-positive, but not authentic serotoninergic, tryptophan hydroxylase-like fibers.

Serotonin (5-HT)-positive, but not tryptophan-5-hydroxylase (TPOH)-positive, authentic serotoninergic fibers were shown in the rat dura mater. 5-HT immunoreactive fibers in the dura are postulated to result from 5-HT uptake from circulating blood elements (e.g. platelets, mast cells) by perivascular sympathetic nerve fibers. A robust TPOH-immunoreactive mast cell population was identified in the dura; this result confirms the TPOH antibody specificity to cells known to synthesize 5-HT. While these results indicate that there are no authentic serotoninergic fibers in the dura mater, the mast cells, platelets and cerebrospinal fluid can serve as a source of 5-HT activating 5-HT receptors known to be present in this tissue.