Talajic, Julia C. MDCM*; Assalian, Alfred MD, FRCSC†; Roy, Daniel MD, FRCSC‡; Harasymowycz, Paul J. MD, FRCSC§
*Resident in Ophthalmology, University of Montreal
Departments of †Ophthalmology
‡Radiology, Notre-Dame Hospital, CHUM
§Michel-Mathieu Institute of Excellence in Ophthalmology and Guy-Bernier Research Center, Maisonneuve-Rosemont Hospital, University of Montreal affiliated site, Montreal, QC, Canada
Adherence: None of the authors have any financial or material support to disclose. The authors take full responsibility for the integrity and accuracy of this report.
Reprints: Julia Talajic, MDCM, Maisonneuve-Rosemont Hospital, 5415 de l'Assomption Boulevard, Montreal, QC, Canada, H1T 2M4 (e-mail: firstname.lastname@example.org).
Received for publication April 16, 2008; accepted January 15, 2009
Statement of Authorship, Notifications of Conflicts of Interest and Ethical.
Acute angle closure glaucoma (AACG) is a rare complication of carotid-cavernous fistula (CCF). However, rarer is iatrogenic AACG subsequent to CCF's standard treatment, endovascular embolization. To our knowledge, this is the first case in which AACG after angiography (a failed attempt at embolization) is treated primarily with embolization, and the first case of AACG and CCF that has yielded such a successful visual recovery.
Acute angle closure glaucoma (AACG) is a rare complication of carotid-cavernous fistula (CCF).1,2 However, rarer still is iatrogenic AACG subsequent to CCF's standard treatment, endovascular embolization.3–5 To our knowledge, this is the first case in which AACG after angiography (a failed attempt at embolization) for CCF is treated primarily with embolization, and the first that has yielded such a successful visual recovery.
A 70-year-old female presented with a 1-week history of diplopia, and redness and burning in her left eye. Visual acuity (VA) was 20/50 Oculus Dexter (OD) and 20/40 Oculus Sinister (OS). Left-sided episcleral venous engorgement, resistance to retropulsion, 3 mm proptosis and ptosis were noted. Applanation tonometry measured 16 mm Hg OD and 20 mm Hg OS. Left-sided third and sixth nerve palsies were present. Gonioscopy demonstrated blood in Schlemm's canal on the left, with open angles in both eyes. Medical history was significant for only hypertension.
Spontaneous, indirect CCF was diagnosed. Computed tomography revealed signs of left-sided intraorbital vascular engorgement (Fig. 1). Throughout the next 6 weeks the patient developed left-sided pain, retroorbital headache, subjective bruit, and V1, V2 dysesthesia. Raised episcleral pressure glaucoma developed with a peak pressure of 33 mm Hg OS. A week's treatment of topical dorzolamide hydrochloride-timolol preceded attempted endovascular embolization of the fistula. An inferior petrosal sinus approach was not possible because the latter was not patent. Approach via the right internal carotid to gain access to the fistula lying between it and the left superior ophthalmic vein was technically limited and ultimately unsuccessful—the procedure was abandoned because of the potential for contrast-induced nephrotoxicity in this elderly woman.
The morning after this angiography and cannulation procedure, VA was counting fingers at 2 ft. OS, the left pupil was semi-dilated and fixed, and corneal edema was present. Intraocular pressure (IOP) read 15 mm Hg OD and 42 mm Hg OS. The left angle was closed on gonioscopy whereas the right remained open. Left AACG was diagnosed, likely secondary to choroidal edema. No ultrasound was performed to verify this at the time, as it was unavailable. The dorzolamide hydrochloride-timolol was changed for brimonidine tartrate, timolol maleate, and oral acetazolamide.
The following day, a second embolization attempt was successful (Fig. 2). The neuroradiologists noted that at the beginning of this second procedure, the fistula was patent and retrograde flow still present into the superior ophthalmic vein. Twenty-four hours after successful embolization, AACG had resolved, VA had improved to 20/40 OD and 20/100 OS, and episcleral venous congestion had diminished with IOP at 12 mm Hg OD, 20 mm Hg OS. There was residual left oculomotor ophthalmoplegia and 2+ cells in the left anterior chamber. The patient was discharged on topical brimonidine tartrate, timolol maleate, and prednisolone.
Over the long term, CCF, episcleral venous and AACG, including ophthalmoplegia and proptosis, resolved completely. VA was 20/20 bilaterally at 6 months. Left-sided uveitis resolved at 5-week follow-up, although synechiae, iris atrophy, and glaucomflecken remained at 6 months. IOP at 1 year was 15 mm Hg Oculus Uterque.
This is the fourth reported case of AACG associated with radiologic intervention of CCF. It is of note that whereas the 3 cases cited involved AACG after successful fistula embolization, our case of AACG occurred after angiography with endovascular manipulation adjacent to the fistula, ultimately without embolization. Golnik and colleagues3 reported AACG and large ipsilateral ciliochoroidal effusions developing hours after successful CCF embolization, which they treated with topical cycloplegics and repeat iridotomies. Fourman4 described AACG after embolized closure of arteriovenous fistula that was treated with neurosurgical fistula closure and resulted in a VA of 20/200. Komiyama and colleagues5 recounted traumatic CCF with AACG 2 weeks after embolized closure. They reasoned the angle closure was precipitated by postembolization oculomotor ophthalmoplegia—after iridotomy, AACG and CCF resolved.
We agree with Golnik et al that embolization likely caused an acute increase in superior ophthalmic vein pressure, which may have overcome the IOP sufficiently to precipitate choroidal transudation, thereby displacing the iris-lens diaphragm anteriorly and closing the angle. Angiography and endovascular cannulation remains the most likely cause of AAC, having occurred 24 hours before it, and involving manipulation adjacent to the superior ophthalmic vein. However, other confounding factors exist. AAC can be precipitated by arteriovenous fistulization in and of itself.1,2 Second, the dorzolamide treatment, a sulfa derivative, could have played a role.6 Furthermore, oculomotor palsy has been linked to angle closure in rare cases.
“Our patient's first radiological intervention, intended for embolization of CCF, ended in angiography and cannulation only; AACG ensued 24 hours later. Notably, the angle closure resolved with successful embolization.” The visual outcome was formidable. Ultimately, embolization could represent optimal treatment of AACG complicating CCF. Choroidal drainage, iridotomy, and medical management are additional options, especially if fistula embolization has already been successfully achieved.1–5
1. Talks SJ, Salmon JF, Elston JS, et al. Cavernous-dural fistula with secondary angle-closure glaucoma. Am J Ophthalmol. 1997;124:851–853.
2. Buus DR, Tse DT, Parrish RK II. Spontaneous carotid cavernous fistula presenting with acute angle closure glaucoma. Arch Ophthalmol. 1989;107:596–597.
3. Golnik KC, Newman SA, Ferguson R. Angle-closure glaucoma consequent to embolization of dural cavernous sinus fistula. AJNR Am J Neuroradiol. 1991;12:1074–1076.
4. Fourman S. Acute closed-angle glaucoma after arteriovenous fistulas. Am J Ophthalmol. 1989;107:156–159.
5. Komiyama M, Morikawa T, Matsusaka Y, et al. Acute angle-closure glaucoma after successful embolization of traumatic carotid-cavernous sinus fistula—case report. Neurol Med Chir (Tokyo). 2003;43:142–145.
6. Postel EA, Assalian A, Epstein DL. Drug-induced transient myopia and angle-closure glaucoma associated with supraciliary choroidal effusion. Am J Ophthalmol. 1996;122:110–112.
© 2010 Lippincott Williams & Wilkins, Inc.