Title | Delayed Onset of Mixed Cranial Neuropathies and Cluster Headache After Embolization of Indirect Carotid-Cavernous Fistula |
Creator | Arthur Wang, MD, Van V. Halbach, MD, Christopher F. Dowd, MD, Matthew D. Alexander, MD, Danial K. Hallam, MD, Basavarj Ghodke, MD, Golnaz Moazami, MD, Grace K. Mandigo, MD, Sean D. Lavine, MD, Philip M. Meyers, MD |
Affiliation | Neurosurgery and Radiology (AW, GKM, SDL, PMM), Columbia University Medical Center, New York, New York; Radiology and Biomedical Imaging (VVH, CFD), UCSF, San Francisco, California; Radiology and Imaging Sciences (MDA), University of Utah, Salt Lake City, Utah; Radiology (DKH, BG), University of Washington, Seattle, Washington; and Department of Ophthalmology (GM), Columbia University Medical Center, New York, New York |
Abstract | Carotid cavernous fistulas (CCF) often present with diplopia secondary to cranial nerve palsy (CNP). Immediate development of postoperative CNP has been described in the literature. This study described delayed- onset of CNP after complete and reconfirmed obliteration of the CCF and resolution of initial CNP |
Subject | CNP; CCF; Endovascular Treatment |
OCR Text | Show Original Contribution Section Editors: Clare Fraser, MD Susan Mollan, MD Delayed-Onset Cranial Nerve Palsy After Transvenous Embolization of Indirect Carotid Cavernous Fistulas Arthur Wang, MD, Van V. Halbach, MD, Christopher F. Dowd, MD, Matthew D. Alexander, MD, Danial K. Hallam, MD, Basavarj Ghodke, MD, Golnaz Moazami, MD, Grace K. Mandigo, MD, Sean D. Lavine, MD, Philip M. Meyers, MD Background: Carotid cavernous fistulas (CCF) often present with diplopia secondary to cranial nerve palsy (CNP). Immediate development of postoperative CNP has been described in the literature. This study described delayedonset of CNP after complete and reconfirmed obliteration of the CCF and resolution of initial CNP. Methods: A retrospective analysis was performed on patients with indirect CCF between 1987 and 2006 at 4 academic endovascular centers. Details of the endovascular procedures, embolic agents used, and complications were studied. Partial or complete obliteration was determined. Immediate and delayed cranial nerve palsies were independently assessed. Results: A total of 267 patients with symptomatic indirect CCF underwent transvenous endovascular treatment. Four patients (1.5%) developed delayed abducens nerve (VI) palsy after complete resolution of presenting symptoms after embolization. Delayed presentation ranged between 3 and 13 months after complete resolution of initial double vision and cranial nerve palsies. Transvenous coil embolization through the inferior petrosal sinus was performed in all 4 affected patients. All had follow-up angiography confirming durable closure of their CCF. MRI did not show new mass lesions or abnormal soft tissue enhancement. In all 4 patients, their abducens nerve (VI) palsy remained. Conclusions: Delayed CNP can develop despite complete endovascular obliteration of the CCF. The cause of delayed CNP is not yet determined, but may represent fibrosis and ischemia. Long-term follow-up is needed even after com- Neurosurgery and Radiology (AW, GKM, SDL, PMM), Columbia University Medical Center, New York, New York; Radiology and Biomedical Imaging (VVH, CFD), UCSF, San Francisco, California; Radiology and Imaging Sciences (MDA), University of Utah, Salt Lake City, Utah; Radiology (DKH, BG), University of Washington, Seattle, Washington; and Department of Ophthalmology (GM), Columbia University Medical Center, New York, New York. The authors report no conflicts of interest. Address correspondence to Arthur Wang, MD, Department of Neurosurgery, Columbia University Medical Center, 710 West 168th Street, New York, NY 10032; E-mail: aw3201@cumc.columbia.edu Wang et al: J Neuro-Ophthalmol 2021; 41: e639-e643 plete neurological and radiological recovery is attained in the immediate perioperative period. Journal of Neuro-Ophthalmology 2021;41:e639–e643 doi: 10.1097/WNO.0000000000001067 © 2020 by North American Neuro-Ophthalmology Society I ndirect carotid cavernous fistula (CCF) is an acquired arteriovenous shunt between dural branches of the internal or external carotid arteries and the cavernous sinus (1). Because of the venous drainage pattern and location of cranial nerves within the cavernous sinus, the resulting increased arterialized pressure within the venous sinus often manifests in proptosis, chemosis, pulsatile tinnitus, and diplopia secondary to cranial nerve palsy (CNP) (1,2). Transvenous embolization (TVE) has become the preferred option for curative endovascular occlusion of indirect CCF because of its safety and efficacy of obliteration of the fistula (3). Complications of TVE include venous congestion, sinus perforation, and the development of cranial nerve palsy (3,4). Although immediate postoperative cranial nerve palsy is a known complication of endovascular coil embolization, delayed onset of CNP is less commonly described in the literature (5–8). In the largest series, we describe 4 patients who developed delayed-onset CNP after curative embolization of their CCF and prior resolution of presenting symptoms. In all 4 patients, follow-up angiography did not demonstrate evidence of fistulas recurrence. No MRI abnormality was identified. We also discuss possible mechanisms for the CNP recurrence. METHODS Patient Population Under an institutional review board-approved protocol, a retrospective analysis was performed on all patients with e639 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution symptomatic indirect CCF between January 1987 and December 2016 at 4 academic endovascular centers. Patients were identified using an existing quality assurance database, tracking all patients and all neurovascular procedures. All patients identified underwent endovascular treatment. Baseline demographics, clinical presentation, and imaging characteristics were recorded with special note of any high-risk features such as cortical venous reflux and ophthalmic vein reflux. Endovascular Treatment Endovascular treatment was performed on symptomatic patients. Trans-arterial embolization, TVE, or combined embolization was selected according to the angioarchitecture of the CCF. Embolization materials included detachable platinum and fibered platinum coils, liquid embolic agents such as N-butyl-2-cyanoacrylate (Cordis Neurovascular, Miami Lakes, FL), and ethylene vinyl copolymer (Onyx; Medtronic, Minneapolis, MN), and polyvinyl alcohol. Complications during and after the interventional procedure were recorded. Immediate postoperative angiographic results were classified as partial and complete obliterations on angiographic findings. Functional Outcomes and Data Analysis After endovascular treatment, cranial nerve function, ophthalmoscopy, and intra-ocular pressure were assessed by ophthalmologists or neuro-ophthalmologists and the neuroendovascular team. Improvement or deterioration in ocular movement or facial sensation were noted. Patients were followed with clinical evaluation. Any new change in symptoms or development of cranial neuropathies was followed up by the ophthalmologist. MRI without and with gadolinium contrast and cerebral angiography were performed to assess for recurrence of the fistula. New CNP was assessed by ophthalmology. RESULTS On immediate postoperative angiography, 231/267 (86.5%) patients had complete obliteration of their CCF. For fistulas that were considered too high risk for complete obliteration, cortical vein reflux was eliminated at the end of the treatment session. Technical Details of Transvenous Embolization in Cases of Delayed Cranial Nerve Palsy Transvenous coil embolization was performed through a transfemoral vein approach under general endotracheal anesthesia. The goal of the treatment was complete obliteration of the arteriovenous shunt to the cavernous sinus. After achieving systemic heparinization, an 8F guide catheter was placed at the level of the jugular bulb and a 2.3F microcatheter was advanced over a 0.014-in wire into the cavernous sinus through the inferior petrosal sinus e640 (IPS). In all 4 cases, the IPS was already occluded on control angiogram. The right IPS was accessed in 2 patients and the left IPS in the remaining 2 patients. In all 4 patients, there was cortical venous drainage on initial angiography. This feature was an indication for treatment of the CCF. The microcatheter was passed over micro-guidewire through the cavernous sinus, then positioned at the confluence between the superior and inferior ophthalmic veins within the anterior cavernous sinus. Coil embolization with care not to occlude the vortex, or vorticose, veins proceeded, in general, from a ventral to dorsal direction. A combination of 0.012–0.018-in nylon/cotton-fibered pushable coils (TruFil, Codman Cordis; VortX, Boston Scientific; Tornado, Cook; Concerto, eV3-Medtronic). Additional coils were added until arteriovenous shunt by digital subtraction angiography had ceased. In each case, the patient tolerated their procedure well and without immediate complication. Within 24 hours, chemosis and proptosis were diminishing. In each case, intra-ocular pressure started to decline toward normal. Vision was maintained. Ophthalmoscopy did not show intra-ocular hemorrhage. Incidence and Outcome of Delayed Cranial Nerve Palsy Delayed presentation of CNP was seen in 4/267 (1.5%) patients (Table 1). All 4 patients originally presented with diplopia and cranial neuropathies including abducens nerve (VI) palsy. In all 4 patients, a transvenous approach with coil embolization of the fistula was performed. All 4 patients harbored an indirect CCF with cortical venous reflux. After complete endovascular obliteration of their CCF, the initial cranial neuropathy resolved in all patients. On long-term clinical follow-up, these patients presented with a delayed abducens nerve (VI) palsy within 13 months. Ophthalmologic exam did not show any delayed intra-ocular hemorrhage or increase in intra-ocular pressure. Repeat angiography to evaluate recurrent CNP demonstrated durable closure of their CCF. MRI brain scan with and without gadolinium contrast was performed in all patients at the time of the delayed abducens nerve palsy. MRI brain did not demonstrate any new infarct, hemorrhage, hydrocephalus, mass or abnormal contrast enhancement suggestive of neoplasm, or edema to suggest venous congestion. The location of the coil mass was noted on repeat angiography (Figs. 1, 2). All CNP persisted throughout long-term follow-up. DISCUSSION There exist reports on the development of new CNP immediately after transvenous coil embolization of direct and indirect CCF (5). Lee et al (5) reported a series of 121 patients treated for indirect CCF. Their reported incidence of immediate post-treatment new CNP was 19.8% following TVE. The order of CNP prevalence is abducens palsy Wang et al: J Neuro-Ophthalmol 2021; 41: e639-e643 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution TABLE 1. Summary of 4 patients with delayed abducens nerve palsy after TVE of CCF. Age, Case yr Sex Presentation 1 63 F 2 60 F Right abducens (VI) nerve palsy, chemosis Right abducens (VI) nerve palsy 3 64 F Right abducens (VI) nerve palsy 4 58 F Right abducens (VI) nerve palsy Right trochlear (IV) nerve palsy Time Until Delayed CNP, mo TVE Complete obliteration Complete obliteration Complete obliteration Complete obliteration CNP on Follow-up 6 Persistent 7 Persistent 13 Persistent 3 Persistent Follow-up Angiogram Complete obliteration Complete obliteration Complete obliteration Complete obliteration CCF, carotid cavernous fistula; CNP, cranial nerve palsy; TVE, transvenous embolization. FIG. 1. Anteroposterior (A) and lateral (B) angiogram images for patient 1. Anteroposterior (C) and lateral (D) angiogram images for patient 2. Arrow shows location of coil mass in posterolateral compartment of the cavernous sinus. (VI) followed by oculomotor (III) nerve palsy and trigeminal (V) nerve palsy (5). The least common CNP is trochlear nerve (IV) palsy which is difficult to assess for on neuroophthalmologic examination (5,8). Many of these CNP improve and recover completely within 6–12 months (5). In addition to development of new CNP after TVE, there can also be a paradoxical worsening of existing CNP immediately after TVE (1,5). After coil deposition within the cavernous sinus, thrombosis of the coils and the sinus preWang et al: J Neuro-Ophthalmol 2021; 41: e639-e643 sumably leads to increased pressure within the sinus, which can in turn exert local mass effect on the cranial nerves running thru the cavernous sinus and the dural leaflets. This leads to acute worsening before resolution of the patient’s CNP (1). In our series, 4 patients developed delayed abducens nerve (VI) palsy after diplopia completely resolved after coil embolization of their CCF. Delayed presentation ranged between 3 and 13 months after complete resolution of e641 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 2. Anteroposterior (A) and lateral (B) angiogram images for patient 3. Anteroposterior (C) and lateral (D) angiogram images for patient 4. Arrow shows location of coil mass in posterolateral compartment of the cavernous sinus. double vision. Follow-up angiogram did not reveal any fistula recurrence. In all patients, the CNP persisted afterward on long-term follow-up. Compared with reports on the immediate development of CNP after CCF embolization, there are only a handful of reports on delayed CNP after initial resolution of all symptoms. Lee et al (5) reported 1 patient in 121 patients treated for CCF who developed a late abducens nerve (VI) palsy at 31 months after transvenous coil embolization. Ghannam et al (7) reported 2 patients with direct traumatic CCF who developed late recurrences of abducens nerve (VI) palsies after successful coil embolization of their fistula. In both cases, late recurrence occurred 48 months after disappearance of their initial palsies and persisted indefinitely afterward. Similarly, Kashiwazaki et al (6) described 4 patients who developed delayed abducens nerve (VI) palsy at 3–65 months after transvenous coil embolization. The delayed abducens nerve (VI) palsy persisted in all 4 patients. Although there have been previous case reports on delayed CNP after CCF, our case series remains the largest to date on the development of delayed CNP after CCF treatment with a prevalence 1.5%. We demonstrate that although rare, delayed CNP does remain a potential complication and patients undergoing endovascular treatment should be observed and deserve long-term follow-up. e642 Development of acute CNP after TVE occurs because of 2 possible reasons; (a) the anatomical location of the cranial nerves within the cavernous sinus and (b) thrombus formation after coil deposition within the cavernous sinus. The anatomy of the cavernous sinus can be divided into 4 compartments in relation to the internal carotid artery: (a) a medial compartment (between the internal carotid artery and the pituitary gland), (b) a anteroinferior compartment (below the posterior genu of the internal carotid artery), (c) a posterosuperior compartment (between the carotid artery and the posterior half of the roof of the sinus), and (d) a lateral compartment (between the carotid artery and lateral sinus wall) (1,9). Anatomical dissections show that the lateral compartment is the narrowest of the 4 compartments (9). Because the abducens nerve (VI) runs in this narrow lateral compartment and because it runs freely rather than within the dural leaves of the cavernous sinus, makes it more prone to injury from local mass effect such as stretch injuries or direct injuries to the vasa nervorum of the nerve (9). Evaluating patients in our series, it is noteworthy that there is dense coil mass in the posterolateral part of the cavernous sinus adjacent to the lateral surface of the posterior bend of the cavernous ICA segment where the abducens nerve (VI) traverses (Figs. 1, 2). Wang et al: J Neuro-Ophthalmol 2021; 41: e639-e643 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution The second hypothesis focuses on acute thrombus formation. Thrombus formation in the cavernous sinus becomes curative after deposition of fibered coils. However, thrombus formation as a cause for de novo CNP is supported by studies showing that over packing of the cavernous sinus with coils can lead to postoperative cranial nerve palsies, most of which are abducens nerve (VI) palsies (1,6,10). In their series of TVE, Kashiwazaki et al (6) demonstrated an association between delayed CNP and average total coil length used to pack the sinus. The average total coil length used for their patients who developed new CNP was 206 ± 43.1 cm while the average total coil length used for patients without CNP was 113 ± 38.8 cm. Similarly, Nishino et al (1) showed that a postembolization coil volume .0.2 cm3 was associated with CNP, postoperatively. In their series, patients who had denser coil packing in the lateral portion of the cavernous sinus were more likely to develop an abducens nerve (VI) palsy (1). This acute thrombus hypothesis more likely explains the new CN VI palsies described in many studies. Explanation for the additional CNP has considered further thrombosis in the cavernous sinus as a cause to cranial nerve dysfunction. It is interesting to note that all 4 of our patients with delayed CNP had cortical venous drainage on initial angiography. One could postulate that cortical venous drainage is a feature of either a larger arteriovenous shunt or more restricted venous outflow which could in effect produce a larger overall space of shunting requiring a larger volume of coils to be deposited to cure the fistula. In each patient in this series, the patients fully recovered cranial nerve function following treatment only to have diplopia due to CNP return in a delayed fashion. Initial considerations may include recrudescence of the CCF with the same physiology that originally caused CNP. However, catheter cerebral arteriography showed no evidence of arteriovenous shunting or abnormal opacification in the cavernous sinus or surrounding venous structures. Another consideration might include mass effect from hematoma, phlegmon, hypertrophic scar tissue or neoplasm. To evaluate for these issues, MRI brain scan without and with gadolinium contrast showed no evidence of mass, mass effect, abnormal edema, or enhancement. Furthermore, once CNP reoccurred, there was no progression as may be seen with infection, neoplasm, or tumefactive scar (keloid). To date, the cause remains unknown. A few limitations exist with our study. It is retrospective in nature. We did not specifically record the length or volume of fibered platinum coils used in these patients to delineate whether there is an association between coil packing density and recurrence of CNP. Yet, the point Wang et al: J Neuro-Ophthalmol 2021; 41: e639-e643 seems moot because CNP fully resolved before reoccurring in a delayed fashion. This, we believe, mitigates against the acute thrombus and coil mass effect arguments. CONCLUSIONS Transvenous coil embolization of indirect CCF achieves favorable results in obliterating the fistula. Neurointerventionalists should be aware of the development of new cranial nerve palsies both immediately after embolization and in delayed fashion on long-term follow-up. Any new neurological findings require further evaluation as causes such as CCF reoccurrence, other arteriovenous fistula occurrence, infection, or neoplasm could occur. Long-term follow-up is needed even after complete neurological and radiological recovery is attained in the immediate perioperative period. REFERENCES 1. Nishino K, Ito Y, Hasegawa H, Kikuchi B, Shimbo J, Kitazawa K, Fujii Y. Cranial nerve palsy following transvenous embolization for a cavernous sinus dural arteriovenous fistula: association with the volume and location of detachable coils. J Neurosurg. 2008;109:208–214. 2. Meyers PM, Halbach VV, Dowd CF, Lempert TE, Malek AM, Phatouros CC, Lefler JE, Higashida RT. Dural carotid cavernous fistula: definitive endovascular management and long-term follow-up. Am J Ophthalmol. 2002;134:85–92. 3. Alexander MD, Halbach VV, Hallam DK, Cooke DL, Ghodke B, Dowd CF, Amans MR, Hetts SW, Higashida R, Meyers PM. Relationship of clinical presentation and angiographic findings in patients with indirect cavernous carotid fistulas. J Neurointervent Surg. 2019;11:937–939. 4. Alexander MD, Halbach VV, Hallam DK, Cooke DL, Ghodke BV, Dowd CF, Amans MR, Hetts SW, Higashida RT, Meyers PM. Long-term outcomes of endovascular treatment of indirect carotid cavernous fistulas: superior efficacy, safety, and durability of transvenous coiling over other techniques. Neurosurgery. 2019;85:E94–E100. 5. Lee SH, Cho WS, Kang HS, Kim JE, Cho YD, Yoo DH, Han MH. Newly occurring cranial nerve palsy after endovascular treatment of cavernous sinus dural arteriovenous fistulas. J Neurointervent Surg. 2019;11:1168–1172. 6. Kashiwazaki D, Kuwayama N, Akioka N, Kuroda S. Delayed abducens nerve palsy after transvenous coil embolization for cavernous sinus dural arteriovenous fistulas. Acta Neurochir (Wien). 2014;156:97–101. 7. Ghannam AB, Subramanian P. Delayed onset cranial nerve palsies after endovascular coil embolization of direct carotidcavernous fistulas. J Neuroophthalmol. 2018;38:156–159. 8. Durgam A, Hussain H, Nakatsuka AS, Raghuram K. Isolated trochlear nerve (cranial nerve IV) palsy in a patient with carotidcavernous fistula. BJR Case Rep. 2018;5:20180028. 9. Rhoton AL Jr. The cavernous sinus, the cavernous venous plexus, and the carotid collar. Neurosurgery. 2002;51(4 suppl):S375–S410. 10. Fujita A, Kohta M, Sasayama T, Kohmura E et al. Impact of transvenous embolization via superior ophthalmic vein on reducing the total number of coils used for patients with cavernous sinus dural arteriovenous fistula. Neurosurg Rev. 2019. https://doi.org/10.1007/s10143-019-01227-9. e643 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. |
Date | 2021-12 |
Language | eng |
Format | application/pdf |
Type | Text |
Publication Type | Journal Article |
Source | Journal of Neuro-Ophthalmology, December 2021, Volume 41, Issue 4 |
Publisher | Lippincott, Williams & Wilkins |
Holding Institution | Spencer S. Eccles Health Sciences Library, University of Utah, 10 N 1900 E SLC, UT 84112-5890 |
Rights Management | © North American Neuro-Ophthalmology Society |
ARK | ark:/87278/s68xbyyf |
Setname | ehsl_novel_jno |
ID | 2116158 |
Reference URL | https://collections.lib.utah.edu/ark:/87278/s68xbyyf |