Skull Base Institute Home Page
Skull Base Brain Tumor Research

Endoscopic Skull Base Surgery
Chapter 4: Interventional Neuroradiology

By Hrayr K. Shahinian, M.D., FACS


Interventional neuroradiology procedures have evolved and now have an established role in the treatment and diagnosis of various skull base pathological conditions. In this chapter the most common aspects of interventional neuroradiologic procedures related to skull base surgery will be discussed, namely embolization of skull base tumors, permanent carotid artery occlusion and inferior petrosal sinus sampling, with emphasis on indications, techniques currently used and potential complications of each procedure.

1. Introduction

In interventional neuroradiology, embolization refers to occlusion or obliteration of a vascular channel or obliteration of a vascular bed. Numerous factors determine the technique and material to be used for any embolization procedure. Differences in the location and structure of the target to be embolized mandates the use of different techniques and materials. The goal of the embolization procedure has to be clearly determined, whether we are embolizing the tumor preoperatively, as a palliation or as a permanent cure affects our therapeutic strategy.

2. Embolization of skull base tumors

2.1. Paragangliomas

Paragangliomas are vascular locally aggressive tumors with a low metastatic potential which arise from paraganglionic chemoreceptor cells of neural crest origin, they can secrete vasoactive substances such as catecholamines and serotonin, multicentricity is estimated to be present in 10% of cases. Secretory activity and multicentricity are associated with an increased incidence of malignancy. Grossly these tumors appear encapsulated and they do not tend to recruit arterial supply from adjacent territories until they are quite large. The angio-architecture of these lesions is usually formed of one or more distinct compartments of the tumor, each fed by one or more arteries. These compartments have distinct capillary beds and venous drainage, direct arteriovenous connections within the tumor may also be present. Large paraganglioma that show multiple arterial supply most likely represent the confluence of multiple lesions that have grown together. This is particularly seen with tympanic and jugular para-gangliomas.

Surgical removal is the treatment of choice for these lesions, and the goal of embolization is to dry the operative field thus reducing the operative bleeding, allowing a shorter operative time and decreasing the overall surgical morbidity.

Before embolization the feeding arteries of the tumors are carefully studied, tympanic and jugular para-gangliomas are supplied by tympanic branches and the jugular branch of the neuromeningeal division of the ascending pharyngeal artery respectively. Contributions from meningeal arteries such as the petrosal branch of middle meningeal artery, accessory meningeal artery, meningeal branches of occipital and vertebral arteries and clival branches of the internal carotid artey may also be present. Selective catheterization of the feeding vesels is achieved using a microcatheter and the embolization is typically performed using PVA (poly vinyl alcohol) particles in the 150- to 250-micron range without difficulty , but there have been reported cases of cranial nerve damage after this embolization. Smaller particles or liquids (ethanol or pure glue) should not be used (Figures 1 (a) and (b)).

For extensive tumors with known arterial invasion, elective presurgical internal carotid artery occlusion may be the procedure of choice prior to surgery, this technique will be described further in this chapter.

2.2. Juvenile Nasopharyngeal Angiofibromas

Juvenile Nasopharyngeal Angiofibromas are benign non-encapsulated fibrovascular mass lesions typically occupying the nasopharynx and posterior nasal cavity of males around the age of puberty. These tumors usually arise from the lateral margin of the posterior nasal cavity, but can show extensive local spread. These lesions are generally supplied by branches of the internal maxillary (sphenopalatine and descending palatine branches) and ascending pharyngeal arteries. As spread occurs, there is a tendency of recruitment of other blood supply, including branches of the facial, ophthalmic, and internal carotid arteries.

The goal of treatment of a JNA is eradication of the tumor. The definitive treatment is surgical resection which may be technically difficult owing to the size of the tumor and excessive bleeding. Bleeding may obscure the surgical field, resulting in an incomplete resection and increased risk of complications. In an attempt to reduce intraoperative bleeding to a minimum (thereby reducing the risks involved in the procedure and improving visibility), embolization of these tumors is often undertaken preoperatively .

Embolization of Juvenile Nasopharyngeal Angiofibromas needs to be intra-tumoral. Once one feeder is occluded, the tumor recruits supply from other vessels. These newly recruited collaterals may prove more difficult to reach endovascularly as well as to control at surgery. Intra-tumoral embolization prevents this recruitment. Selective catheterization of the feeding vessels is achieved followed by infiltrating the tumor mass with PVA particles in the 50-150 micron range. It is better to inject these very small particles from a very distal location, as close to the tumor as possible, in order to obtain true intratumoral occlusion and avoid embolization of non-target tissues. Care must be taken during injection and awarness of the dangerous collaterals to the intracranial and intraorbital contents is mandatory, simple random injection of the internal maxillary artery may prove disastrous .

2.3. Skull Base Meningiomas

Meningiomas in general represent about 15% of all primary central nervous system neoplasms. They originate from cellular elements that form the meninges, but also may come from pial or dural fibroblasts. They typically affect adults aged 25 to 65 with a peak age of about 45 years. Multiple meningiomas are present in approximately 1% to 2% of cases, but this is usually associated with the central form of neurofibromatosis. Meningiomas are typically benign but rarely can be malignant. They are well-circumscribed lesions that tend to be round, with or without lobulations. They are slow growing and do not invade, but compress local structures.

Skull base meningiomas arising from the olfactory groove or the planum sphenoidale may receive blood supply from the anterior and posterior ethmoidal branches of the ophthalmic artery. Cavernous sinus and parasellar meningiomas receive blood supply from branches of the cavernous internal carotid artery as well as the middle and accessory meningeal arteries. In general skull base meningiomas are less vascular when compared to meningiomas in other locations rendering them less adapted for preoperative embolization, and also show an increased risk for the procedure because the vessels are tortous and their origins are short, with an increased risk of reflux to non target tissues specially the retina. Frequently, the best option for tumors in this location is removal en bloc, necessitating permanent internal carotid artery occlusion, which can be performed preoperatively with minimal morbidity (Figures 2 (a) and (b)).

3. Test and permanent occlusion of the internal carotid artery

3.1. Indications

Indications for test and permanent occlusion of an internal carotid artery (ICA) include fusiform aneurysms, giant aneurysms, and wide neck aneurysms (especially of the cavernous segment of the ICA), which cannot be treated selectively by either endovascular technique or surgical technique, post-traumatic lesions of the ICA, test occlusion may be performed preoperatively for skull base tumors when intraoperative occlusion of the ICA may become necessary.

3.2. General considerations

Successful and safe test and permanent occlusion requires multidisciplinary collaboration among neurosurgeons, interventional neuroradiologists, and anesthetists. An angiographic suite, with digital substraction, is needed. Neuroleptic analgesia is used to enable clinical assessment of the patient during the test occlusion. Strict aseptic technique is essential. Before test (and permanent) occlusion, all patients are evaluated with contrast-enhanced computed tomography (CT), magnetic resonance imaging (MRI), or both, depending on the pathology.

3.3. Procedure of test occlusion

The procedure is done under local anesthesia, we start by perfrming a four-vessel cerebral arteriography via one femoral artery to evaluate the complete intracra-nial hemodynamics and supply. Our diagnostic catheter is left in the carotid artery that will not be tested to evaluate the effectiveness of clamping and of alternative blood flow. The other femoral artery is used to access the side to be tested ,which is accessed using a guiding catheter. Test occlusion is done using a microcatheter with a non-detachable balloon or a detachable balloon at it's tip if our strategy is to permanently occlude the vessel.

To prevent thromboembolic complications, 25 U/kg of heparin is injected as an intravenous bolus, then 20 U/kg/hr is constantly infused intravenously by electric syringe pump. The ICA is catheterized selectively with the guiding catheter. The entire assembly of the microcatheter and balloon is then advanced through a Y-type hemostatic valve into the guide catheter. A three-way stopcock is placed on the other arm of the hemostatic valve. All the catheters are flushed continuously with normal saline using pressure bags to prevent thromboembolic complications related to the catheters. After the balloon-tipped microcatheter has passed through the guide catheter, the blood flow carries it to the desired location.

The ideal location for placing the balloon is the petrosal segment of the internal carotid artery, alternatively the balloon can be placed just proximal to the lesion ( tumor, aneurysm or traumatic lesion). The occlusion test is performed by inflating the balloon using a prepared mixture of contrast and saline until there is occlusion of the artery. This test should not exceed 30 minutes duration, even under full heparinization (Figures 4 (a) and (b)).

3.4. Evaluation of Tolerance

Several parameters are evaluated during the entire duration of the test occlusion. Clinical tolerance is tested through higher function, sensory and motor function, and cranial nerves. In the case of clinical findings of any minor deficit, the balloon is immediately deflated, and the test is abandoned. The integrity of the circle of Willis is verified by injecting the contralateral ICA during occlusion via the diagnostic catheter previously left in place to evaluate the anterior communicating artery and its contribution to the circulation. This is evaluated by the rapidity of emptying of the middle cerebral artery flow and its venous phase on the side of the test occlusion in comparison with the opposite side. The dominant vertebral artery is then injected to evaluate the contribution of the posterior communicating artery on the side of the test occlusion.

A patient is said to have tolerated the test occlusion if he is clinically perfect after tolerance for 30 minutes, and when the cerebral perfusion time and venous phase appearance show no more than a 1 second delay on the side of occlusion compared with the opposite side. At the end of the test occlusion the balloon is deflated and removed, and if necessary, protamine sulphate can be used to reverse the effect of heparin.

3.5. Permanent Occlusion

After a properly performed, well-tolerated test occlusion, permanent occlusion can be performed. A detachable balloon is introduced to the desired location and inflated with the mixture of contrast media and saline until occlusion of the vessel occurs, the balloon is then released by simple traction. A second safety balloon is then released 1 or 2 cm below the first one.

3.5. Complications and follow-up of permanent occlusion

The complications of permanent occlusion comprise thromboembolic complications, difficulties in balloon detachment with the risk of it's migration, and (3 premature deflation of the balloon in situations in which only one balloon is released.

After permanent occlusion, the patients should be transferred to intensive care unit for 48 hours of complete bedrest in a horizontal position with the systolic arterial pressure maintained at a level 20 mm Hg higher than normal to maintain adequate distal cerebral perfusion and to prevent hemodynamic hypoperfusion. Patients are followed up clinically and with serial skull films to confirm balloon inflation.

4. Inferior petrosal sinus sampling

4.1. Rationale of the procedure

The most important recent advance in the diagnostic evaluation of patients with Cushing's syndrome has been the use of bilateral inferior petrosal sinus sampling (BIPSS) for ACTH. The inferior petrosal sinuses receive drainage from the pituitary gland without mixture of blood from other sources. Therefore, if the patient has pituitary Cushing's, the ACTH levels in the IPS are high compared to an ACTH drawn in the periphery. In contrast, in ectopic Cushing's, the ACTH in the IPS and the periphery should be equivalent because the tumor is located elsewhere. This test has been shown to reliably determine whether or not there is a pituitary tumor, a peripheral ectopic source of ACTH, or primary adrenal disease. The endocrinologist can then use this information to guide therapy. If a primary adenoma is determined to be present in the pituitary gland, the surgeon can use this information to decide which half of the pituitary gland contains the offending secretory tumor, and thus which half to remove, rather than removing the entire gland and potentially causing panhypopituitarism.

4.2. Technique of the procedure

The venous drainage of the pituitary gland is into the cavernous sinus. The cavernous sinus usually drains into the bilateral inferior petrosal sinuses, superior petrosal sinuses, and basilar venous plexus. For evaluation of the side of a possible microadenoma, accurate sampling is required for testing. This is achieved by obtaining simultaneous multiple venous samples, from each inferior petrosal sinus, along with a concurrent peripheral nonselective sample.

Bilateral venous access is required, and is achieved by a standard transfemoral approach. Guiding catheters are then introduced and placed in both internal jugular veins. Knowledge of the inferior petrosal venous anatomy is essential , since there are wide variations in the drainage of the sinus . Usually, the IPS enters the internal jugular vein right where the vein turns posteriorly at the skull base and starts its course to the sigmoid sinus. The IPS entrance is typically at the apex of this curve on the medial superior surface. Occasionally the sinus may be hypoplastic or there may be multiple small channels instead of a single inferior petrosal vein. A hand injection in the superior jugular bulb, with 10 ml of contrast, during valsalva or compression of both internal jugular veins at the root of the neck, will usually opacify all the tributaries and the inferior petrosal sinuses. A road map can be acquired at this stage to mark the ostium of the inferior petrosal sinus. Lateral and anteroposterior views are acquired. Occasionally it may prove necessary to catheterize the arterial side, select the internal carotid artery, and perform a standard angiogram to locate the cavernous sinus and visualize its drainage in the venous phase of the angiogram, sometimes this proves to be the easiest way of locating the inferior petrosal sinus.

After the IPS has been found a microcatheter is advanced gently into it up to the region just below the cavernous sinus. A contrast injection in this location confirms the correct position before venous sampling. In addition, a forceful contrast injection usually fills across the midline, clearly demonstrating the contralateral IPS, which helps localization of the entry point of the IPS into the opposite sigmoid sinus.

After the catheters are in position bilaterally, samples are obtained from each microcatheter and also from the guiding catheters serving as the peripheral vein samples, Four to five samples are obtained from each location at 5-minute intervals. A significant gradient between the pituitary (central) and peripheral venous values of plasma ACTH, obtained by simultaneous sampling, is indicative of Cushing's disease (Figures 5 (a) and (b)).

4.3. Complications

Inferior petrosal sinus sampling is a safe procedure with extremely low complications rates. The primary complications are related to direct damage to the sinus from forceful catheterization or from thrombosis of the sinus. Both of these conditions are preventable now with the use of newer more adapted material and proper periprocedural anticoagulation.


Figures 1:
(A) Paraganglioma pre-embolization
(B) Paraganglioma post-embolization

Figures 2:
(A) Meningioma pre-embolization
(B) Meningioma post-embolization

Figures 3:
(A) Inflated balloon in petrous segment
(B) Control angiogram from contra lateral internal carotid artery

Figures 4:
(A) Lateral view of microcatheter in inferior petrosal sinus
(B) Anteroposterior (AP) view of microcatheters in inferior petrosal sinuses


About SBI   |   Treatments   |   Press Room   |   Patient Resources   |   Programs   |   Research   |   Contact


Pituitary Tumor | Trigeminal Neuralgia | Acoustic Neuroma | Meningioma | Craniopharyngioma | Head & Neck Tumors | Hemifacial Spasm | Rathke's Cyst | Arachnoid Cyst | Craniosynostosis | Chordoma | Cerebral Aneurysm | Arteriovenous Malformations
Endoscopic Brain Surgery | Dr. Hrayr Shahinian | Skull Base Foundation
Research Papers Main Page | Research Papers Site Map