TECHNICAL CASE REPORT
Development of Arteriovenous Fistula After Revascularization Bypass for Moyamoya Disease: Case Report Abdullah H. Feroze, BS* Jacob Kushkuley, BS‡ Omar Choudhri, MD* Jeremy J. Heit, MD, PhD* Gary K. Steinberg, MD, PhD* Huy M. Do, MD* *Department of Neurosurgery, Stanford University School of Medicine, Stanford, California; ‡Department of Neurosurgery, University of Massachusetts Medical School, Worcester, Massachusetts Correspondence: Abdullah H. Feroze, BS, Department of Neurosurgery, Stanford University School of Medicine, Edwards Bldg, Room 201, 300 Pasteur Dr, Stanford, CA 94305. E-mail: [email protected] Received, July 7, 2014. Accepted, August 28, 2014. Published Online, September 23, 2014. Copyright © 2014 by the Congress of Neurological Surgeons.
BACKGROUND AND IMPORTANCE: Moyamoya disease is a rare cerebrovascular disorder often treated by direct and indirect revascularization bypass techniques as a result of a typically devastating disease course and poor response to medical therapy. In this report, we describe the formation and subsequent management of a de novo arteriovenous fistula identified in the setting of a patient treated with direct bypass surgery, a previously unreported phenomenon. CLINICAL PRESENTATION: A 51-year-old woman presenting with Suzuki stage IV bilateral moyamoya disease underwent bilateral extracranial-to-intracranial superficial temporal artery–to–middle cerebral artery bypass without complication at our institution. At the 6-month follow-up, she demonstrated no evidence of residual neurological deficits or continued symptoms despite documentation of an arteriovenous fistula arising at the site of the right extracranial-to-intracranial bypass on routine follow-up cerebral angiography. CONCLUSION: We present the first reported case of de novo arteriovenous fistula formation after superficial temporal artery-to-middle cerebral artery bypass for the treatment of moyamoya disease. Treatment of such iatrogenic arteriovenous fistulae fed by a patent bypass vessel may prove challenging without associated compromise of the bypass, meriting careful evaluation of all potential therapeutic options. The fistula described herein most likely occurred secondary to recanalization of a previously thrombosed vein of Trolard. This case demonstrates the possibility of arteriovenous fistula formation as a potential sequela of revascularization bypass surgery and lends support to the previously described traumatic origin of fistula formation. KEY WORDS: Angiography, Arteriovenous fistula, Bypass, De novo, Moyamoya, Revascularization Operative Neurosurgery 11:202–206, 2015
M
oyamoya disease is a rare cerebrovascular disease characterized by progressive, idiopathic steno-occlusive changes in the distal internal carotid arteries.1,2 As an increasingly recognized cause of cerebrovascular accident in both adults and children refractory to medical therapy, surgical revascularization by direct or indirect bypass is recommended in most symptomatic patients to improve neurological outcomes and to reduce ischemic and hemorrhagic events.3,4 In this report, we describe an arteriovenous fistula (AVF) identified on a 6-month followABBREVIATIONS: AVF, arteriovenous fistula; EC-IC, extracranial-to-intracranial; MCA, middle cerebral artery; STA, superficial temporal artery
E202 | VOLUME 11 | NUMBER 1 | MARCH 2015
DOI: 10.1227/NEU.0000000000000558
up diagnostic cerebral angiogram of a patient treated with superficial temporal artery (SCA)-tomiddle cerebral artery (MCA) bypass surgery, a previously unreported phenomenon.
CLINICAL PRESENTATION A 51-year-old woman presented to medical attention with a 1-year history of transient episodes of expressive aphasia, progressive bilateral upperand lower-extremity weakness, and headaches. Cerebral angiography demonstrated nearly occlusive stenosis within the bilateral supraclinoid internal carotid arteries, in addition to severe narrowing of bilateral anterior cerebral artery (A1) and MCA (M1) segments (Figure 1). Magnetic resonance color perfusion imaging
www.neurosurgery-online.com
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited
AVF FORMATION AFTER REVASCULARIZATION BYPASS
FIGURE 1. Preoperative imaging demonstrating bilateral moyamoya disease. A, preoperative right internal carotid artery angiogram (anteroposterior view) showing nearly occlusive stenosis of the supraclinoid internal carotid, middle cerebral, and anterior cerebral arteries with moyamoya collaterals. B, right internal carotid artery angiogram (lateral projection) showing similar findings. C, right external carotid artery angiography identifying the right superficial temporal artery (arrow) with some contrast reflux into the right internal carotid artery (arrowheads). D, magnetic resonance color perfusion imaging demonstrating perfusion delays in the bilateral anterior cerebral and middle cerebral artery territories.
confirmed perfusion delays with bilateral anterior cerebral artery and MCA territories, and the patient subsequently underwent bilateral extracranial-to-intracranial (EC-IC) SCA-to-MCA bypass without complication. At the 6-month follow-up, she demonstrated no evidence of residual neurological deficits or continued symptoms. Neurological examination was within normal limits.
seen from the donor STA. Early venous filling of the vein of Trolard was noted after injection into the right external carotid artery. The arteriovenous shunting was attributed to a de novo AVF arising at the site of the right EC-IC bypass, supplied by multiple leptomeningeal collaterals from the right STA and draining into the vein of Trolard.
Investigations The patient underwent a routine follow-up cerebral angiogram 6 months after her bypass procedure, which demonstrated patent bilateral EC-IC bypass grafts with both direct and indirect STA collaterals supplying the MCA territories. The direct bypass on the right was found to be occluded, but robust indirect collaterals were
Differential Diagnoses Differential diagnoses considered for the angiographic abnormality included an AVF of dural or pial origin, arteriovenous malformation, developmental venous anomaly, or early venous filling of an early infarct. However, given the filling pattern of the abnormality as mentioned above, the findings were most consistent with an AVF,
OPERATIVE NEUROSURGERY
VOLUME 11 | NUMBER 1 | MARCH 2015 | E203
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited
FEROZE ET AL
ultimately found to be of pial nature, given the direct fistulous connection of the graft lying below the dura to the vein of Trolard. Treatment The management of AVFs may involve conservative management, radiosurgery, microsurgical resection, or endovascular therapy.5,6 In this instance, the decision was made to manage the lesion conservatively because the patient remained asymp-
tomatic. Furthermore, fistula disconnection by either microsurgery or transarterial glue embolization would have likely compromised the integrity of the bypass graft. Transvenous embolization was also considered, but the right vein of Trolard was found to be draining the respective hemisphere without sufficient evidence of ample alternative routes and thus could not be sacrificed. Radiosurgery was also deferred because of concern for compromise to the vein of Trolard.
FIGURE 2. Postoperative imaging demonstrating formation of a pial arteriovenous fistula. A and B (magnified), lateral projection of the right external carotid artery angiogram with donor superficial temporal artery (arrowhead) supplying leptomeningeal collaterals to the right hemisphere and an arteriovenous fistula with early venous drainage into the vein of Trolard (arrow). C, intraoperative microscopic view demonstrating the superficial temporal artery with its soft tissue cuff laid over the cortical surface (white arrowheads) in close proximity to cortical vein (dotted arrow). The dura is opened in a stellate pattern (asterisk) and closed over the graft. D and E (magnified), anteroposterior projection of the right external carotid artery angiogram demonstrating the donor superficial temporal artery (arrowhead) and the early cortical vein draining the pial fistula (arrow). F, magnified intraoperative microscope view identifying the direct anastomosis between the superficial temporal artery (arrowheads) and middle cerebral artery (arrow). Cortical veins (dotted arrow) and dural edge (asterisk) are also identified.
E204 | VOLUME 11 | NUMBER 1 | MARCH 2015
www.neurosurgery-online.com
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited
AVF FORMATION AFTER REVASCULARIZATION BYPASS
Outcome and Follow-up The patient continues to remain asymptomatic 13 months after her initial direct STA-to-MCA bypass and 7 months after angiography demonstrating the AVF. The plan was made to follow the maturation of the bypass and evolution of the AVF through serial angiography. Follow-up cerebral angiography at 12 months postoperatively demonstrated a stable appearance of the AVF with a patent indirect bypass.
DISCUSSION Surgical bypass and revascularization has been accepted as a mainstay of treatment for moyamoya disease.7 Large-scale retrospective case series of patients, both pediatric and adult, treated with direct or indirect revascularization have described postoperative sequelae, including nonfunctioning bypasses, ischemic infarction, hemorrhage, and death.8,9 The concomitant presence of dural AVFs in the setting of moyamoya disease has also been reported.10 Both dural and pial fistulas have previously been described in the setting of prior craniotomy and ventriculostomy.11-15 Kubo et al12 described the formation of a pial AVF in a patient 21 months after craniotomy for an aneurysm in whom bleeding from a sylvian vein was controlled with Surgicel (oxidized regenerated cellulose). However, this is the first description of de novo formation of an AVF, pial or otherwise, secondary to EC-IC bypass. Current evidence seems to suggest that dural and pial fistulas are caused secondary to recanalization of thrombosed dural sinuses and cortical veins, respectively, and our observations noted herein are no exception to such a purported origin. A review of preoperative
angiography demonstrated no filling of the vein of Trolard, which most likely had thrombosed secondary to vasculopathy. Given subsequent findings of premature visualization of the vein of Trolard on follow-up 6 months postoperatively, it was determined that the initially intended direct STA-to-MCA bypass had closed, replaced instead by functionally robust, indirect collateralization (Figure 2). The recanalization of the vein of Trolard may have been responsible for preventing any direct graft uptake and hence may be the most likely cause of pial AVF in this instance. The recanalized vein of Trolard can be seen draining the fistula and the right hemisphere on the postoperative cerebral angiogram (Figure 3). There was no evidence of such angiographic abnormality on preoperative cerebral angiography (Figure 1), suggesting the AVF development to be iatrogenic secondary to the EC-IC bypass. A review of intraoperative images demonstrated close association of cortical veins near the bypass graft, also suggesting the potential for fistula formation within the operative bed (Figure 2). However, other potential factors remain worthy of consideration in precipitating the formation of such a lesion, including inadvertent cortical venous occlusion or preexisting arteriovenous shunting in the setting of chronic cerebral ischemia secondary to moyamoya disease exacerbated by the bypass. We posit that the development of leptomeningeal collaterals may have created venous connections leading to such fistulization in a pressurized state. The resulting increase in cortical venous drainage carries with it an unknown future risk of hemorrhage. Given the presence of cortical venous reflux, an argument may be made for the associated hemorrhage risk of such a lesion to mimic that of type IIb dural AVFs, which often merit intervention. However, treatment of this
FIGURE 3. Angiographic imaging demonstrating recanalization of a previously thrombosed vein of Trolard after revascularization bypass. A, preoperative lateral projection angiogram of the right internal carotid artery demonstrating late parenchymal and venous phase, with no evidence of filling of the vein of Trolard. The internal cerebral vein (dashed arrow) and basal vein of Rosenthal (arrowhead) can be seen. B, postoperative 6-month follow-up lateral projection angiogram of the right internal carotid artery venous phase with recanalized vein of Trolard (arrow) and unchanged filling of the internal cerebral vein (dashed arrow) and basal vein of Rosenthal (arrowhead).
OPERATIVE NEUROSURGERY
VOLUME 11 | NUMBER 1 | MARCH 2015 | E205
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited
FEROZE ET AL
fistula would have occluded the maturing EC-IC bypass and placed the patient at a significant risk for ischemic infarction. Furthermore, it remains possible that the resulting flow dynamics created by this pial AVF will promote sustained patency of the bypass graft, but further angiographic follow-up remains necessary to dictate whether intervention will ultimately be merited. The most recent follow-up cerebral angiogram at 1 year after surgery demonstrated no change in the characteristics of the fistula.
CONCLUSION We describe the formation of a de novo AVF secondary to EC-IC bypass for the treatment of moyamoya disease. This case further suggests an iatrogenic, traumatic origin for AVFs, with the possibility that the incidence and progression of fistulization are influenced, at least in part, by intraoperative dissection and technique. Recanalization of a previously thrombosed vein of Trolard likely contributed to the development of the pial AVF in the case described here. Given the patient’s stable clinical status, the decision was made to manage this rare event conservatively with serial angiography, but continued follow-up to monitor lesion progression will ultimately dictate how such iatrogenic lesions should be managed on an individual basis. Disclosure The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article.
REFERENCES 1. Guzman R, Lee M, Achrol A, et al. Clinical outcome after 450 revascularization procedures for moyamoya disease: clinical article. J Neurosurg. 2009;111(5):927-935.
E206 | VOLUME 11 | NUMBER 1 | MARCH 2015
2. Suzuki J, Takaku A. Cerebrovascular “moyamoya” disease: disease showing abnormal net-like vessels in base of brain. Arch Neurol. 1969;20(3):288-299. 3. Golby AJ, Marks MP, Thompson RC, Steinberg GK. Direct and combined revascularization in pediatric moyamoya disease. Neurosurgery. 1999;45(1):50-58; discussion 58-60. 4. Scott RM. Surgery for moyamoya syndrome? Yes. Arch Neurol. 2001;58(1): 128-129. 5. Youssef PP, Schuette AJ, Cawley CM, Barrow DL. Advances in surgical approaches to dural fistulas. Neurosurgery. 2014;74(suppl 1):S32-S41. 6. Vanlandingham M, Fox B, Hoit D, Elijovich L, Arthur AS. Endovascular treatment of intracranial dural arteriovenous fistulas. Neurosurgery. 2014;74(suppl 1):S42-S49. 7. Pandey P, Steinberg GK. Neurosurgical advances in the treatment of moyamoya disease. Stroke. 2011;42(11):3304-3310. 8. Abla AA, Gandhoke G, Clark JC, et al. Surgical outcomes for moyamoya angiopathy at Barrow Neurological Institute with comparison of adult indirect encephaloduroarteriosynangiosis bypass, adult direct superficial temporal artery-tomiddle cerebral artery bypass, and pediatric bypass: 154 revascularization surgeries in 140 affected hemispheres. Neurosurgery. 2013;73(3):430-439. 9. Dusick JR, Gonzalez NR, Martin NA. Clinical and angiographic outcomes from indirect revascularization surgery for moyamoya disease in adults and children: a review of 63 procedures. Neurosurgery. 2011;68(1):34-43; discussion 43. 10. Killory BD, Gonzalez LF, Wait SD, Ponce FA, Albuquerque FC, Spetzler RF. Simultaneous unilateral moyamoya disease and ipsilateral dural arteriovenous fistula: case report. Neurosurgery. 2008;62(6):E1375-E1376; discussion E1376. 11. Davie JC, Hodges F. Arteriovenous fistula after removal of meningioma: case report. J Neurosurg. 1967;27(4):364-369. 12. Kubo Y, Ogasawara K, Kashimura H, et al. De novo intracranial pial arteriovenous fistula after craniotomy. Neurosurg Quart. 2010;20(4):277-279. 13. Sasaki T, Hoya K, Kinone K, Kirino T. Postsurgical development of dural arteriovenous malformations after transpetrosal and transtentorial operations: case report. Neurosurgery. 1995;37(4):820-824; discussion 824-825. 14. Schuette AJ, Blackburn SL, Barrow DL, Cawley CM. Pial arteriovenous fistula resulting from ventriculostomy. World Neurosurg. 2012;77(5-6):785. e1-785.e2. 15. Yassari R, Jahromi B, Macdonald R. Dural arteriovenous fistula after craniotomy for pilocytic astrocytoma in a patient with protein S deficiency. Surg Neurol. 2002; 58(1):59-64; discussion 64.
www.neurosurgery-online.com
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited
Development of Arteriovenous Fistula After Revascularization Bypass for Moyamoya Disease: Case Report Abdullah H. Feroze, BS* Jacob Kushkuley, BS‡ Omar Choudhri, MD* Jeremy J. Heit, MD, PhD* Gary K. Steinberg, MD, PhD* Huy M. Do, MD* *Department of Neurosurgery, Stanford University School of Medicine, Stanford, California; ‡Department of Neurosurgery, University of Massachusetts Medical School, Worcester, Massachusetts Correspondence: Abdullah H. Feroze, BS, Department of Neurosurgery, Stanford University School of Medicine, Edwards Bldg, Room 201, 300 Pasteur Dr, Stanford, CA 94305. E-mail: [email protected] Received, July 7, 2014. Accepted, August 28, 2014. Published Online, September 23, 2014. Copyright © 2014 by the Congress of Neurological Surgeons.
BACKGROUND AND IMPORTANCE: Moyamoya disease is a rare cerebrovascular disorder often treated by direct and indirect revascularization bypass techniques as a result of a typically devastating disease course and poor response to medical therapy. In this report, we describe the formation and subsequent management of a de novo arteriovenous fistula identified in the setting of a patient treated with direct bypass surgery, a previously unreported phenomenon. CLINICAL PRESENTATION: A 51-year-old woman presenting with Suzuki stage IV bilateral moyamoya disease underwent bilateral extracranial-to-intracranial superficial temporal artery–to–middle cerebral artery bypass without complication at our institution. At the 6-month follow-up, she demonstrated no evidence of residual neurological deficits or continued symptoms despite documentation of an arteriovenous fistula arising at the site of the right extracranial-to-intracranial bypass on routine follow-up cerebral angiography. CONCLUSION: We present the first reported case of de novo arteriovenous fistula formation after superficial temporal artery-to-middle cerebral artery bypass for the treatment of moyamoya disease. Treatment of such iatrogenic arteriovenous fistulae fed by a patent bypass vessel may prove challenging without associated compromise of the bypass, meriting careful evaluation of all potential therapeutic options. The fistula described herein most likely occurred secondary to recanalization of a previously thrombosed vein of Trolard. This case demonstrates the possibility of arteriovenous fistula formation as a potential sequela of revascularization bypass surgery and lends support to the previously described traumatic origin of fistula formation. KEY WORDS: Angiography, Arteriovenous fistula, Bypass, De novo, Moyamoya, Revascularization Operative Neurosurgery 11:202–206, 2015
M
oyamoya disease is a rare cerebrovascular disease characterized by progressive, idiopathic steno-occlusive changes in the distal internal carotid arteries.1,2 As an increasingly recognized cause of cerebrovascular accident in both adults and children refractory to medical therapy, surgical revascularization by direct or indirect bypass is recommended in most symptomatic patients to improve neurological outcomes and to reduce ischemic and hemorrhagic events.3,4 In this report, we describe an arteriovenous fistula (AVF) identified on a 6-month followABBREVIATIONS: AVF, arteriovenous fistula; EC-IC, extracranial-to-intracranial; MCA, middle cerebral artery; STA, superficial temporal artery
E202 | VOLUME 11 | NUMBER 1 | MARCH 2015
DOI: 10.1227/NEU.0000000000000558
up diagnostic cerebral angiogram of a patient treated with superficial temporal artery (SCA)-tomiddle cerebral artery (MCA) bypass surgery, a previously unreported phenomenon.
CLINICAL PRESENTATION A 51-year-old woman presented to medical attention with a 1-year history of transient episodes of expressive aphasia, progressive bilateral upperand lower-extremity weakness, and headaches. Cerebral angiography demonstrated nearly occlusive stenosis within the bilateral supraclinoid internal carotid arteries, in addition to severe narrowing of bilateral anterior cerebral artery (A1) and MCA (M1) segments (Figure 1). Magnetic resonance color perfusion imaging
www.neurosurgery-online.com
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited
AVF FORMATION AFTER REVASCULARIZATION BYPASS
FIGURE 1. Preoperative imaging demonstrating bilateral moyamoya disease. A, preoperative right internal carotid artery angiogram (anteroposterior view) showing nearly occlusive stenosis of the supraclinoid internal carotid, middle cerebral, and anterior cerebral arteries with moyamoya collaterals. B, right internal carotid artery angiogram (lateral projection) showing similar findings. C, right external carotid artery angiography identifying the right superficial temporal artery (arrow) with some contrast reflux into the right internal carotid artery (arrowheads). D, magnetic resonance color perfusion imaging demonstrating perfusion delays in the bilateral anterior cerebral and middle cerebral artery territories.
confirmed perfusion delays with bilateral anterior cerebral artery and MCA territories, and the patient subsequently underwent bilateral extracranial-to-intracranial (EC-IC) SCA-to-MCA bypass without complication. At the 6-month follow-up, she demonstrated no evidence of residual neurological deficits or continued symptoms. Neurological examination was within normal limits.
seen from the donor STA. Early venous filling of the vein of Trolard was noted after injection into the right external carotid artery. The arteriovenous shunting was attributed to a de novo AVF arising at the site of the right EC-IC bypass, supplied by multiple leptomeningeal collaterals from the right STA and draining into the vein of Trolard.
Investigations The patient underwent a routine follow-up cerebral angiogram 6 months after her bypass procedure, which demonstrated patent bilateral EC-IC bypass grafts with both direct and indirect STA collaterals supplying the MCA territories. The direct bypass on the right was found to be occluded, but robust indirect collaterals were
Differential Diagnoses Differential diagnoses considered for the angiographic abnormality included an AVF of dural or pial origin, arteriovenous malformation, developmental venous anomaly, or early venous filling of an early infarct. However, given the filling pattern of the abnormality as mentioned above, the findings were most consistent with an AVF,
OPERATIVE NEUROSURGERY
VOLUME 11 | NUMBER 1 | MARCH 2015 | E203
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited
FEROZE ET AL
ultimately found to be of pial nature, given the direct fistulous connection of the graft lying below the dura to the vein of Trolard. Treatment The management of AVFs may involve conservative management, radiosurgery, microsurgical resection, or endovascular therapy.5,6 In this instance, the decision was made to manage the lesion conservatively because the patient remained asymp-
tomatic. Furthermore, fistula disconnection by either microsurgery or transarterial glue embolization would have likely compromised the integrity of the bypass graft. Transvenous embolization was also considered, but the right vein of Trolard was found to be draining the respective hemisphere without sufficient evidence of ample alternative routes and thus could not be sacrificed. Radiosurgery was also deferred because of concern for compromise to the vein of Trolard.
FIGURE 2. Postoperative imaging demonstrating formation of a pial arteriovenous fistula. A and B (magnified), lateral projection of the right external carotid artery angiogram with donor superficial temporal artery (arrowhead) supplying leptomeningeal collaterals to the right hemisphere and an arteriovenous fistula with early venous drainage into the vein of Trolard (arrow). C, intraoperative microscopic view demonstrating the superficial temporal artery with its soft tissue cuff laid over the cortical surface (white arrowheads) in close proximity to cortical vein (dotted arrow). The dura is opened in a stellate pattern (asterisk) and closed over the graft. D and E (magnified), anteroposterior projection of the right external carotid artery angiogram demonstrating the donor superficial temporal artery (arrowhead) and the early cortical vein draining the pial fistula (arrow). F, magnified intraoperative microscope view identifying the direct anastomosis between the superficial temporal artery (arrowheads) and middle cerebral artery (arrow). Cortical veins (dotted arrow) and dural edge (asterisk) are also identified.
E204 | VOLUME 11 | NUMBER 1 | MARCH 2015
www.neurosurgery-online.com
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited
AVF FORMATION AFTER REVASCULARIZATION BYPASS
Outcome and Follow-up The patient continues to remain asymptomatic 13 months after her initial direct STA-to-MCA bypass and 7 months after angiography demonstrating the AVF. The plan was made to follow the maturation of the bypass and evolution of the AVF through serial angiography. Follow-up cerebral angiography at 12 months postoperatively demonstrated a stable appearance of the AVF with a patent indirect bypass.
DISCUSSION Surgical bypass and revascularization has been accepted as a mainstay of treatment for moyamoya disease.7 Large-scale retrospective case series of patients, both pediatric and adult, treated with direct or indirect revascularization have described postoperative sequelae, including nonfunctioning bypasses, ischemic infarction, hemorrhage, and death.8,9 The concomitant presence of dural AVFs in the setting of moyamoya disease has also been reported.10 Both dural and pial fistulas have previously been described in the setting of prior craniotomy and ventriculostomy.11-15 Kubo et al12 described the formation of a pial AVF in a patient 21 months after craniotomy for an aneurysm in whom bleeding from a sylvian vein was controlled with Surgicel (oxidized regenerated cellulose). However, this is the first description of de novo formation of an AVF, pial or otherwise, secondary to EC-IC bypass. Current evidence seems to suggest that dural and pial fistulas are caused secondary to recanalization of thrombosed dural sinuses and cortical veins, respectively, and our observations noted herein are no exception to such a purported origin. A review of preoperative
angiography demonstrated no filling of the vein of Trolard, which most likely had thrombosed secondary to vasculopathy. Given subsequent findings of premature visualization of the vein of Trolard on follow-up 6 months postoperatively, it was determined that the initially intended direct STA-to-MCA bypass had closed, replaced instead by functionally robust, indirect collateralization (Figure 2). The recanalization of the vein of Trolard may have been responsible for preventing any direct graft uptake and hence may be the most likely cause of pial AVF in this instance. The recanalized vein of Trolard can be seen draining the fistula and the right hemisphere on the postoperative cerebral angiogram (Figure 3). There was no evidence of such angiographic abnormality on preoperative cerebral angiography (Figure 1), suggesting the AVF development to be iatrogenic secondary to the EC-IC bypass. A review of intraoperative images demonstrated close association of cortical veins near the bypass graft, also suggesting the potential for fistula formation within the operative bed (Figure 2). However, other potential factors remain worthy of consideration in precipitating the formation of such a lesion, including inadvertent cortical venous occlusion or preexisting arteriovenous shunting in the setting of chronic cerebral ischemia secondary to moyamoya disease exacerbated by the bypass. We posit that the development of leptomeningeal collaterals may have created venous connections leading to such fistulization in a pressurized state. The resulting increase in cortical venous drainage carries with it an unknown future risk of hemorrhage. Given the presence of cortical venous reflux, an argument may be made for the associated hemorrhage risk of such a lesion to mimic that of type IIb dural AVFs, which often merit intervention. However, treatment of this
FIGURE 3. Angiographic imaging demonstrating recanalization of a previously thrombosed vein of Trolard after revascularization bypass. A, preoperative lateral projection angiogram of the right internal carotid artery demonstrating late parenchymal and venous phase, with no evidence of filling of the vein of Trolard. The internal cerebral vein (dashed arrow) and basal vein of Rosenthal (arrowhead) can be seen. B, postoperative 6-month follow-up lateral projection angiogram of the right internal carotid artery venous phase with recanalized vein of Trolard (arrow) and unchanged filling of the internal cerebral vein (dashed arrow) and basal vein of Rosenthal (arrowhead).
OPERATIVE NEUROSURGERY
VOLUME 11 | NUMBER 1 | MARCH 2015 | E205
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited
FEROZE ET AL
fistula would have occluded the maturing EC-IC bypass and placed the patient at a significant risk for ischemic infarction. Furthermore, it remains possible that the resulting flow dynamics created by this pial AVF will promote sustained patency of the bypass graft, but further angiographic follow-up remains necessary to dictate whether intervention will ultimately be merited. The most recent follow-up cerebral angiogram at 1 year after surgery demonstrated no change in the characteristics of the fistula.
CONCLUSION We describe the formation of a de novo AVF secondary to EC-IC bypass for the treatment of moyamoya disease. This case further suggests an iatrogenic, traumatic origin for AVFs, with the possibility that the incidence and progression of fistulization are influenced, at least in part, by intraoperative dissection and technique. Recanalization of a previously thrombosed vein of Trolard likely contributed to the development of the pial AVF in the case described here. Given the patient’s stable clinical status, the decision was made to manage this rare event conservatively with serial angiography, but continued follow-up to monitor lesion progression will ultimately dictate how such iatrogenic lesions should be managed on an individual basis. Disclosure The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article.
REFERENCES 1. Guzman R, Lee M, Achrol A, et al. Clinical outcome after 450 revascularization procedures for moyamoya disease: clinical article. J Neurosurg. 2009;111(5):927-935.
E206 | VOLUME 11 | NUMBER 1 | MARCH 2015
2. Suzuki J, Takaku A. Cerebrovascular “moyamoya” disease: disease showing abnormal net-like vessels in base of brain. Arch Neurol. 1969;20(3):288-299. 3. Golby AJ, Marks MP, Thompson RC, Steinberg GK. Direct and combined revascularization in pediatric moyamoya disease. Neurosurgery. 1999;45(1):50-58; discussion 58-60. 4. Scott RM. Surgery for moyamoya syndrome? Yes. Arch Neurol. 2001;58(1): 128-129. 5. Youssef PP, Schuette AJ, Cawley CM, Barrow DL. Advances in surgical approaches to dural fistulas. Neurosurgery. 2014;74(suppl 1):S32-S41. 6. Vanlandingham M, Fox B, Hoit D, Elijovich L, Arthur AS. Endovascular treatment of intracranial dural arteriovenous fistulas. Neurosurgery. 2014;74(suppl 1):S42-S49. 7. Pandey P, Steinberg GK. Neurosurgical advances in the treatment of moyamoya disease. Stroke. 2011;42(11):3304-3310. 8. Abla AA, Gandhoke G, Clark JC, et al. Surgical outcomes for moyamoya angiopathy at Barrow Neurological Institute with comparison of adult indirect encephaloduroarteriosynangiosis bypass, adult direct superficial temporal artery-tomiddle cerebral artery bypass, and pediatric bypass: 154 revascularization surgeries in 140 affected hemispheres. Neurosurgery. 2013;73(3):430-439. 9. Dusick JR, Gonzalez NR, Martin NA. Clinical and angiographic outcomes from indirect revascularization surgery for moyamoya disease in adults and children: a review of 63 procedures. Neurosurgery. 2011;68(1):34-43; discussion 43. 10. Killory BD, Gonzalez LF, Wait SD, Ponce FA, Albuquerque FC, Spetzler RF. Simultaneous unilateral moyamoya disease and ipsilateral dural arteriovenous fistula: case report. Neurosurgery. 2008;62(6):E1375-E1376; discussion E1376. 11. Davie JC, Hodges F. Arteriovenous fistula after removal of meningioma: case report. J Neurosurg. 1967;27(4):364-369. 12. Kubo Y, Ogasawara K, Kashimura H, et al. De novo intracranial pial arteriovenous fistula after craniotomy. Neurosurg Quart. 2010;20(4):277-279. 13. Sasaki T, Hoya K, Kinone K, Kirino T. Postsurgical development of dural arteriovenous malformations after transpetrosal and transtentorial operations: case report. Neurosurgery. 1995;37(4):820-824; discussion 824-825. 14. Schuette AJ, Blackburn SL, Barrow DL, Cawley CM. Pial arteriovenous fistula resulting from ventriculostomy. World Neurosurg. 2012;77(5-6):785. e1-785.e2. 15. Yassari R, Jahromi B, Macdonald R. Dural arteriovenous fistula after craniotomy for pilocytic astrocytoma in a patient with protein S deficiency. Surg Neurol. 2002; 58(1):59-64; discussion 64.
www.neurosurgery-online.com
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited
