- Email: [email protected]
Intracranial Dural Arteriovenous Malformations: Results of Stereotactic Radiosurgery in 17 Patients
Clinical Oncology (2002) 14: 97–102 doi:10.1053/clon.2002.0072, available online at http://www.idealibrary.com on
Stereotactic Radiotherapy: Original Article – RCR Undergraduate Prize Winner 2001 Intracranial Dural Arteriovenous Malformations: Results of Stereotactic Radiosurgery in 17 Patients SUZANNE O’LEARY*, TIMOTHY J. HODGSON*, STUART C. COLEY*, ANDREAS A. KEMENY†, MATTHIAS W. R. RADATZ† *Department of Neuroradiology and †National Centre for Stereotactic Radiosurgery, Royal Hallamshire Hospital, Glossop Road, Sheffield S10 2JF, U.K. ABSTRACT: Aim: To assess the efficacy of stereotactic radiosurgery (STRS) for the treatment of dural arteriovenous fistulae (DAVF). Materials and methods: Between November 1987 and December 1998, 17 patients with a total of 18 DAVF were treated with STRS at the National Centre for Stereotactic Radiosurgery, Sheffield. Clinical and radiological data was collected retrospectively from the case notes and radiological records. Two neuroradiologists classified the pre-STRS appearance of the DAVF using the Borden criteria, and reviewed the follow-up imaging. Up to date follow-up was requested from the GPs and referring consultants. Results: In retrospect one DAVF had been misdiagnosed and was excluded from the study. The remaining 17 DAVF were located at the tentorium (6), cavernous sinus (3), right parasellar region (1), floor of left middle cranial fossa (1), midline posterior fossa (1), petrous apex (1) and transverse sinus (4). Ten were Borden type I, four were type II, three were type III. Follow-up angiography was available for 13 patients; 10 DAVF were completely obliterated, two showed considerable reduction in size, one showed deterioration. Clinical follow-up was available for 14 of the 16 patients. Conclusion: Stereotactic radiosurgery can successfully obliterate DAVF with few side effects. O’Leary, S. et al. (2002). Clinical Oncology 14, 97–102. 2002 The Royal College of Radiologists. Published by Elsevier Science Ltd. All rights reserved. Key words: Stereotactic Radiosurgery (STRS), Dural Arteriovenous Fistulae (DAVF)
INTRODUCTION
DAVF are rare vascular abnormalities that constitute between 10–15% of all intracranial arteriovenous malformations. Awad defined a DAVF as ‘a nidus of arteriovenous shunting within the dura mater’ [1]. The nidus of shunting is between the two layers of dura. DAVF are abnormal connections between the dural arterial supply, and the pial or dural venous drainage. It is widely accepted that DAVF have an acquired aetiology, directly related to a damaged, or thrombosed dural venous sinus [2–5]. An increase in venous pressure, caused by thrombosis, leads to the opening of tiny arteriovenous communications, known to already exist in the dura [6,7]. The risk of haemorrhage from a DAVF varies widely. Awad proposed that the most important Author for correspondence: Dr T. J. Hodgson, Department of Neuroradiology, Royal Hallamshire Hospital, Glossop Road, Sheffield S10 2JF, U.K. E-mail: [email protected] 0936–6555/02/020097+06 $35.00/0
predisposing factors for an aggressive course were site, (e.g. tentorial incisura), or those DAVF with either leptomeningeal venous drainage, variceal or aneurysmal venous dilatations, or galenic drainage. The risk of haemorrhage was most closely tied to the location of the lesion, and whether cortical venous drainage was present [1]. The management of DAVF is complex, and varies widely. Curative treatment is currently advised for those patients that have had haemorrhage, have pial venous drainage, or who presented with neurological signs and symptoms [1,8]. Treatment options include endovascular therapy [9–16], STRS [17–20], and surgical resection of the DAVF [1,21]. These techniques are often used in combination [19,22]. There are however many reported problems with each of these techniques. Embolization may be successful initially, provided the DAVF is accessible, but if followed up for long enough, some DAVF were found to recanalize [13–16,23]. Conventional surgery, whilst effective, is not an option for all patients,
2002 The Royal College of Radiologists. Published by Elsevier Science Ltd. All rights reserved.
98
Table 1 – Presenting symptoms of the 16 patients treated with STRS Presenting symptom
Number of patients
Bruit Chemosis/opthalmoplegia Focal neurological deficit Haemorrhage
3 5 1 10
and can be associated with a significant morbidity, particularly in the elderly. The combination of treatment with STRS, and embolization, was used with good success by Link et al. [22] in their group of 29 patients. CLINICAL MATERIAL AND METHODS
age was 56 years (range 24–77). These patients presented with a variety of symptoms and signs as shown in Table 1. Four patients had received previous treatment for their DAVF. All four had undergone embolization. In one of these patients, the embolization had worked initially, but had then recanalized. Two patients had undergone open surgery, which was unsuccessful in completely treating the lesions. Characteristics of DAVF
Two neuroradiologists defined the DAVF in terms of site and Borden classification, on the basis of the preSTRS conventional angiograms. The Borden classification (Table 2) was used since it is the system that helps predict those at greatest risk of poor outcome (Table 3).
Patient Population
Between November 1987 and December 1998, the National Centre for Stereotactic Radiosurgery, Sheffield treated 18 DAVF in 17 patients. In retrospect, one DAVF had been misdiagnosed, and was excluded from the study. Of the 16 patients correctly treated with STRS, 10 were female and six were male. The average
Radiosurgical Dose Planning
Each patient had their radiation dose planning carried out on the day of their procedure, with the stereotactic head-frame in place. All the patients had diagnostic cerebral angiograms, except for one who had CT. The radiation dose was then calculated, and the patients
Table 2 – Borden classification Type 1 2 3
Venous drainage
Arterial supply
Retrograde flow
Directly into dural venous sinuses or meningeal veins Dural sinuses or meningeal veins Subarachnoid veins but does not have dural sinus or meningeal vein drainage
Meningeal arteries
No
Meningeal arteries Meningeal arteries
In subarachnoid veins
Table 3 – Location and Borden classification of treated DAVF Position Left transverse sinus Right parasellar Left transverse sinus Transverse sinus Free edge of left tentorial Carotico-cavernous Petrous apex Floor of left middle cranial fossa Carotico-cavernous Right carotico-cavernous Tentorial Tentorial Midline posterior fossa Left transverse sinus Right tentorium Right tentorium Tentorial
Arterial pedicles External carotid External+internal carotid External carotid External carotid External carotid+internal carotid External carotid Right external carotid+right AICA+superior cerebellar External carotid Bilateral external carotid External carotid External carotid External+internal carotid External carotid External carotid Internal carotid Internal carotid Internal and external carotid
Specific angiographic feature Venous varix
Multiple varices+venous stenosis Large varix
Varix Venous outflow obstruction
Venous varix Venous aneurysm
Type 1 1 2 1 2 3 1 1 3 3 1 2 1 1 1 1 2
Table 4 – Clinical follow-up of 14 patients treated with STRS for DAVF Clinical findings
Number of patients
Improved Stable Neurologically worse not related to treatment Cerebral infarct since treatment Death
7 2 1 3 1
underwent treatment at Weston Park Hospital, Sheffield. Each patient received 2500 cGy (50% isodose curve) except one patient, who was given 2083 cGy. The radiation dose field ranged from 1–7 and the collimators used ranged from 4–18 mm. All patients were kept in hospital for 24 h after the procedure for observation. None of the patients experienced any side effects at this stage, apart from mild headache and nausea. Post-STRS Surveillance
The patients were followed up every 6 months by their local hospital for at least the first 2 years. After 2 to 3 years, a repeat cerebral angiogram was performed, to see if obliteration had occurred. This was done at the referring institution, and the reports were sent to the Royal Hallamshire Hospital. RESULTS Clinical Follow-up
We have documented clinical follow on 14 of the 16 patients. The length of time of clinical follow-up ranged from 8–96 months. Six patients experienced symptomatic improvement; this included diminished orbital bruit in three and resolution of opthalmoplegia and/or chemosis in four. Three patients had stable deficits that had neither improved nor deteriorated following STRS. One patient became totally deaf in one ear 3 years after STRS, probably as a result of treatment. A total of three patients experienced cerebral infarcts many months after treatment and it is believed that these events were not related to treatment (Table 4). One patient died due to an unrelated pathology. Radiological Follow-up
Two-year post-STRS cerebral angiograms were performed for 13 patients. There was complete resolution of the DAVF in 10 patients. One patient showed 85% reduction in the size of the lesion. Another patient showed considerable reduction of the arteriovenous shunting, the feeding arteries had decreased in calibre and a venous varix had gone. As these angiograms were performed 2 years after STRS, we are awaiting the 3-year follow-up, as these DAVF may go on to resolve
99
completely. Only one patient showed angiographic evidence of deterioration. Of the remaining three patients, one was unable to undergo angiography, but clinical improvement was dramatic and a CT scan did not suggest that the DAVF was still there. Another patient was too heavy for the angiography table, but showed considerable clinical improvement. There was no follow-up angiogram for the final patient, although his GP believes that it has been performed, and therefore we are still waiting to hear from the referring consultant. On the basis of the angiographic follow-up we have a complete obliteration rate of 77%.
DISCUSSION
DAVF are difficult lesions to treat. The natural history of these rare acquired lesions is variable, but can be predicted by certain clinical and angiographic features. DAVF presenting with seizures, focal neurological deficit, progressive visual symptoms, intractable bruit, intracranial hypertension and haemorrhage require treatment. Awad reported a study linking particular angiographic characteristics with a high risk of haemorrhage [3]. These patients should be treated independent of symptoms [1,24–26]. In the study by Malik the risk of haemorrhage of a DAVF close to a dural venous sinus had a lifetime risk of haemorrhage of 75% and one located away from the major sinuses had a 5% risk of haemorrhage [26]. Borden’s classification attempts to identify those at high risk of haemorrhage by identifying those lesions with retrograde flow in the venous sinus, and reflux into cortical veins [27]. There have been a handful of international studies in the treatment of DAVF, but this is the first study to look at the treatment of DAVF using STRS in the United Kingdom. Link et al. [22] reported a study including 29 patients with DAVF, however they used a combination of STRS and embolization post-STRS, for patients with symptoms, and those at very high risk of haemorrhage. They reported an obliteration rate of 72% in the 18 patients that they followed up with angiograms at a mean period of 24 months. Our results show an obliteration rate of 77%, with a higher percentage of patients followed up with angiography. Lewis described seven patients with a tentorial DAVF that were treated with transarterial embolization and linear accelerated radiosurgery [19]. They reported an obliteration rate of 56%. There are a few reports of conventional radiation therapy for DAVF, but they do not include outcome data. The combination of STRS with embolization has been reported with varying degrees of completeness [15,17,18]. After a mean follow-up period of 24 months 10 of our DAVF showed complete obliteration on angiogram. Thirteen of our 16 patients had follow-up angiograms, which gave us an obliteration rate of 77%. We had very good results with two other patients whose angiograms showed enormous reduction in size. In one of these two
100
Fig. 1 – Patient 1. (a) (Left) illustrating a left external carotid injection. An extensive tentorial dural fistula is demonstrated with a varix (arrowhead) and venous shunting into the straight sinus (arrow). (b) (Right) demonstrating complete resolution of the fistula 2 years following radiosurgery.
Fig. 2 – Patient 1. (a) (Left) illustrating a left internal carotid injection. Further filling of the fistula is demonstrated via hypertrophied meningo-hypophyseal branches (arrowhead). (b) (Right) demonstrating complete resolution of the fistula 2 years following radiosurgery.
we are awaiting the 3-year angiogram later this year as we believe it will be completely resolved by then. In one patient the DAVF became bigger. This patient has a stable deficit and it has been decided that he will undergo no further treatment. The other four DAVFs, three patients, did not undergo angiogram follow-up. One patient was too heavy for the angiographic table and the other was unable to stay still for long enough for
angiography because of an unrelated medical condition. However both these patients showed dramatic improvement in their clinical symptoms related to their DAVF. The final patient with the two DAVFs has not undergone angiography. There has been no reported haemorrhage in any of our patients to date. There have been few treatment related complications so far reported both by this review
101
Fig. 3. – Patient 2. (a) (Left) illustrating a right external carotid artery injection. A carotico-cavernous fistula is demonstrated with filling of the cavernous sinus (arrow). Venous drainage into the orbit via a hypertrophied superior ophthalmic vein is shown (arrowhead). (b) (Right) demonstrating complete resolution of the fistula 2 years following stereotactic radiosurgery.
and Link et al. One of our patients went on to become totally deaf in the left ear. It is possible that the vestibulocochlear nerve damage 3-years post treatment was directly related to the treatment but it might also have occurred as a result of successful obliteration. There are two cases of sudden deafness after spontaneous occlusion of a DAVF [28]. Three of our patients had strokes after STRS. One patient had a parietal infarct 26 months post STRS; one was in the left occipital region and the other in the right tentorial region. One patient had a small cerebellar stroke 56 months post STRS of a DAVF in the petrous apex. This seems unlikely to be related to the treatment or effects of untreated DAVF as this malformation showed complete obliteration at 24 months. The third patient had a left-sided stroke 3 months post STRS and a further TIA 13 months later. Whilst DAVF take 12–36 months to obliterate this patient was still at risk from haemorrhage, but this does not appear to have occurred in this patient. The rapidity of symptom onset was not suggestive of radionecrosis. To date none of our patients have developed new or recurrent symptoms to suggest that an angiographically obliterated DAVF had recanalized, although 4-year angiogram have not been universally performed.
CONCLUSION
STRS provides an excellent treatment option for DAVF. It is especially relevant for those patients who
are unfit for surgery, or whose DAVF is inaccessible to either surgery or embolization, but are at high risk of haemorrhage. Acknowledgements. Dr M. Collins for helping me with the initial field of research. Dr S. Coley and Dr T. Hodgson for their constant support and guidance during this study. The Stereotactic Radiosurgery, and Neuroradiology Departments at the Royal Hallamshire Hospital, Sheffield for all their help and support with this study. REFERENCES 1 Awad IA, Little JR, Akrawi WP, Ahl J. Intracranial dural arteriovenous malformations: factors predisposing to an aggressive neurological course. J Neurosurg 1990;72:839–850. 2 Aminoff MJ. Vascular anomalies in the intracranial dura mater. Brain 1973;96:601–612. 3 Chaudhary MY, Sachdev VP, Cho SH, Weitzner I Jr, Purlyic S, Huang YP. Dural arteriovenous malformation of the major venous sinuses. An acquired lesion. AJNR 1982;3:13–19. 4 Houser OW, Baker HL Jr, Rhoton AL Jr, Okazaki H. Intracranial dural arteriovenous malformations. Radiology 1972;105:55–64. 5 Houser OW, Campbell JK, Campbell RJ, Sundt TM Jr. Arteriovenous malformations affecting the transverse dural venous — an acquired lesion. Mayo Clin Proc 1979;54:651–661. 6 Kerber CW, Newton TH. The macro and microvasculature of the dura mater. Neuroradiology 1973;6:175–179. 7 Vidyasagar C. Persistent embryonic veins in the arteriovenous malformation of the dura. Acta Neurochir(Wien) 1979;48:199–216. 8 Malik GM, Pearce JE, Ausman JI, et al. Dural arteriovenous malformations and intracranial haemorrhage. Neurosurgery 1984;15:332–339. 9 Halbach VV, Higashide RT, Hieshma GB, et al. Dural fistulae involving the cavernous sinus: results of treatment in 30 patients. Radiology 1987;163:437–442.
102
10 Halbach VV, Higashide RT, Hieshma GB, et al. Dural fistulae involving the transverous and sigmoid sinuses: results of treatment in 28 patients. Radiology 1987;163:443–447. 11 Costin JA, Weinstein MA, Berlin AJ, et al. Dural arteriovenous malformations involving the cavernous sinus: a case report. Br J Opthalmol 1978;62:478–482. 12 Courtheoux P, Labbe D, Hamel C, et al. Treatment of bilateral spontaneous dural carotid-cavernous fistulas by coils and sclerotherapy. Case report. J Neurosurg 1987;66:468–470. 13 Fermand M, Reizine D, Melki JP, et al. Long term follow up of 43 pure dural arteriovenous fistulae (AVF) of the lateral sinus. Neuroradiology 1987;29:348–353. 14 Gobin YP, Rogopoulos A, Aymad A, et al. Endovascular treatment of intracranial dural arteriovenous fistulae with spinal perimedullary venous drainage. J Neurosurg 1992;77:718–723. 15 Hardy RW Jr, Costin JA, Weinstein M, et al. External carotid cavernous fistula treated by transfemoral embolization. Surg Neurol 1978;9:255–256. 16 Vinuela F, Fox AJ, Debrum GM, et al. Spontaneous carotid cavernous fistulas. Clinical, radiological and therapeutic considerations. Experience with 20 cases. J Neurosurg 1984;60:976–984. 17 Chandler HC, Freidman WA. Successful radiosurgical treatment of a dural arteriovenous malformation: Case report. Neurosurgery 1993;33:139–142. 18 Fabrikant JI, Lyman JT, Hosobuchi Y. Stereotactic heavy ionbragg peak radiosurgery for intracranial vascular disorders: method for treatment of deep arteriovenous malformations. Br J Radiol 1984;57:479–490. 19 Lewis AI, Tomsick TA, Tew JM Jr. Management of tentorial arteriovenous malformations: transarterial embolization combined
20
21
22
23
24
25
26
27
28
with Stereotactic radiosurgery or surgery. J Neurosurg 1994; 81:851–859. Steiner L, Lindquist C, Adler JR, et al. Clinical outcome of radiosurgery for cerebral arteriovenous malformations. J Neurosurg 1992;77:1–8. Barnwell SL, Halbach VV, Higashida RT, et al. Complex dural arteriovenous fistulae. Results of combined endovascular and neurosurgical treatment in 16 patients. J Neurosurg 1989;71: 352–358. Link MJ, Coffey RJ, Nichols DA. The role of radiosurgery and particulate embolization in the treatment of dural arteriovenous fistulae. J Neurosurg 1995;84:804–809. Vinuela FV, Debrun GM, Fox AJ, et al. Detachable calibrated leak balloon for superselective angiograph and embolization of dural arteriovenous malformations. J Neurosurg 1983;58:817– 823. King WA, Martin NA. Intracerebral haemorrhage due to dural arteriovenous malformations and fistulae. Neurosurg Clin North Am 1992;3:577–590. Lasjaun P, Chui M, Ter Brugge K. Neurological manifestations of intracranial dural arteriovenous malformations. J Neurosurg 1986;64:724–730. Malik GM, Pearce JE, Ausman JI, et al. Dural arteriovenous malformations and intracranial haemorrhage. Neurosurgery 1984;15:332–339. Borden JA, Wu JK, Shucart WA. A proposed classification for spinal and cranial dural arteriovenous fistulous malformations and implications for treatment. J Neurosurg 1995;82:166–179. Landman JA, Braun IF. Spontaneous closure of a dural arteriovenous fistula associated with acute hearing loss. AJNR 1985; 6:448–449.
Stereotactic Radiotherapy: Original Article – RCR Undergraduate Prize Winner 2001 Intracranial Dural Arteriovenous Malformations: Results of Stereotactic Radiosurgery in 17 Patients SUZANNE O’LEARY*, TIMOTHY J. HODGSON*, STUART C. COLEY*, ANDREAS A. KEMENY†, MATTHIAS W. R. RADATZ† *Department of Neuroradiology and †National Centre for Stereotactic Radiosurgery, Royal Hallamshire Hospital, Glossop Road, Sheffield S10 2JF, U.K. ABSTRACT: Aim: To assess the efficacy of stereotactic radiosurgery (STRS) for the treatment of dural arteriovenous fistulae (DAVF). Materials and methods: Between November 1987 and December 1998, 17 patients with a total of 18 DAVF were treated with STRS at the National Centre for Stereotactic Radiosurgery, Sheffield. Clinical and radiological data was collected retrospectively from the case notes and radiological records. Two neuroradiologists classified the pre-STRS appearance of the DAVF using the Borden criteria, and reviewed the follow-up imaging. Up to date follow-up was requested from the GPs and referring consultants. Results: In retrospect one DAVF had been misdiagnosed and was excluded from the study. The remaining 17 DAVF were located at the tentorium (6), cavernous sinus (3), right parasellar region (1), floor of left middle cranial fossa (1), midline posterior fossa (1), petrous apex (1) and transverse sinus (4). Ten were Borden type I, four were type II, three were type III. Follow-up angiography was available for 13 patients; 10 DAVF were completely obliterated, two showed considerable reduction in size, one showed deterioration. Clinical follow-up was available for 14 of the 16 patients. Conclusion: Stereotactic radiosurgery can successfully obliterate DAVF with few side effects. O’Leary, S. et al. (2002). Clinical Oncology 14, 97–102. 2002 The Royal College of Radiologists. Published by Elsevier Science Ltd. All rights reserved. Key words: Stereotactic Radiosurgery (STRS), Dural Arteriovenous Fistulae (DAVF)
INTRODUCTION
DAVF are rare vascular abnormalities that constitute between 10–15% of all intracranial arteriovenous malformations. Awad defined a DAVF as ‘a nidus of arteriovenous shunting within the dura mater’ [1]. The nidus of shunting is between the two layers of dura. DAVF are abnormal connections between the dural arterial supply, and the pial or dural venous drainage. It is widely accepted that DAVF have an acquired aetiology, directly related to a damaged, or thrombosed dural venous sinus [2–5]. An increase in venous pressure, caused by thrombosis, leads to the opening of tiny arteriovenous communications, known to already exist in the dura [6,7]. The risk of haemorrhage from a DAVF varies widely. Awad proposed that the most important Author for correspondence: Dr T. J. Hodgson, Department of Neuroradiology, Royal Hallamshire Hospital, Glossop Road, Sheffield S10 2JF, U.K. E-mail: [email protected] 0936–6555/02/020097+06 $35.00/0
predisposing factors for an aggressive course were site, (e.g. tentorial incisura), or those DAVF with either leptomeningeal venous drainage, variceal or aneurysmal venous dilatations, or galenic drainage. The risk of haemorrhage was most closely tied to the location of the lesion, and whether cortical venous drainage was present [1]. The management of DAVF is complex, and varies widely. Curative treatment is currently advised for those patients that have had haemorrhage, have pial venous drainage, or who presented with neurological signs and symptoms [1,8]. Treatment options include endovascular therapy [9–16], STRS [17–20], and surgical resection of the DAVF [1,21]. These techniques are often used in combination [19,22]. There are however many reported problems with each of these techniques. Embolization may be successful initially, provided the DAVF is accessible, but if followed up for long enough, some DAVF were found to recanalize [13–16,23]. Conventional surgery, whilst effective, is not an option for all patients,
2002 The Royal College of Radiologists. Published by Elsevier Science Ltd. All rights reserved.
98
Table 1 – Presenting symptoms of the 16 patients treated with STRS Presenting symptom
Number of patients
Bruit Chemosis/opthalmoplegia Focal neurological deficit Haemorrhage
3 5 1 10
and can be associated with a significant morbidity, particularly in the elderly. The combination of treatment with STRS, and embolization, was used with good success by Link et al. [22] in their group of 29 patients. CLINICAL MATERIAL AND METHODS
age was 56 years (range 24–77). These patients presented with a variety of symptoms and signs as shown in Table 1. Four patients had received previous treatment for their DAVF. All four had undergone embolization. In one of these patients, the embolization had worked initially, but had then recanalized. Two patients had undergone open surgery, which was unsuccessful in completely treating the lesions. Characteristics of DAVF
Two neuroradiologists defined the DAVF in terms of site and Borden classification, on the basis of the preSTRS conventional angiograms. The Borden classification (Table 2) was used since it is the system that helps predict those at greatest risk of poor outcome (Table 3).
Patient Population
Between November 1987 and December 1998, the National Centre for Stereotactic Radiosurgery, Sheffield treated 18 DAVF in 17 patients. In retrospect, one DAVF had been misdiagnosed, and was excluded from the study. Of the 16 patients correctly treated with STRS, 10 were female and six were male. The average
Radiosurgical Dose Planning
Each patient had their radiation dose planning carried out on the day of their procedure, with the stereotactic head-frame in place. All the patients had diagnostic cerebral angiograms, except for one who had CT. The radiation dose was then calculated, and the patients
Table 2 – Borden classification Type 1 2 3
Venous drainage
Arterial supply
Retrograde flow
Directly into dural venous sinuses or meningeal veins Dural sinuses or meningeal veins Subarachnoid veins but does not have dural sinus or meningeal vein drainage
Meningeal arteries
No
Meningeal arteries Meningeal arteries
In subarachnoid veins
Table 3 – Location and Borden classification of treated DAVF Position Left transverse sinus Right parasellar Left transverse sinus Transverse sinus Free edge of left tentorial Carotico-cavernous Petrous apex Floor of left middle cranial fossa Carotico-cavernous Right carotico-cavernous Tentorial Tentorial Midline posterior fossa Left transverse sinus Right tentorium Right tentorium Tentorial
Arterial pedicles External carotid External+internal carotid External carotid External carotid External carotid+internal carotid External carotid Right external carotid+right AICA+superior cerebellar External carotid Bilateral external carotid External carotid External carotid External+internal carotid External carotid External carotid Internal carotid Internal carotid Internal and external carotid
Specific angiographic feature Venous varix
Multiple varices+venous stenosis Large varix
Varix Venous outflow obstruction
Venous varix Venous aneurysm
Type 1 1 2 1 2 3 1 1 3 3 1 2 1 1 1 1 2
Table 4 – Clinical follow-up of 14 patients treated with STRS for DAVF Clinical findings
Number of patients
Improved Stable Neurologically worse not related to treatment Cerebral infarct since treatment Death
7 2 1 3 1
underwent treatment at Weston Park Hospital, Sheffield. Each patient received 2500 cGy (50% isodose curve) except one patient, who was given 2083 cGy. The radiation dose field ranged from 1–7 and the collimators used ranged from 4–18 mm. All patients were kept in hospital for 24 h after the procedure for observation. None of the patients experienced any side effects at this stage, apart from mild headache and nausea. Post-STRS Surveillance
The patients were followed up every 6 months by their local hospital for at least the first 2 years. After 2 to 3 years, a repeat cerebral angiogram was performed, to see if obliteration had occurred. This was done at the referring institution, and the reports were sent to the Royal Hallamshire Hospital. RESULTS Clinical Follow-up
We have documented clinical follow on 14 of the 16 patients. The length of time of clinical follow-up ranged from 8–96 months. Six patients experienced symptomatic improvement; this included diminished orbital bruit in three and resolution of opthalmoplegia and/or chemosis in four. Three patients had stable deficits that had neither improved nor deteriorated following STRS. One patient became totally deaf in one ear 3 years after STRS, probably as a result of treatment. A total of three patients experienced cerebral infarcts many months after treatment and it is believed that these events were not related to treatment (Table 4). One patient died due to an unrelated pathology. Radiological Follow-up
Two-year post-STRS cerebral angiograms were performed for 13 patients. There was complete resolution of the DAVF in 10 patients. One patient showed 85% reduction in the size of the lesion. Another patient showed considerable reduction of the arteriovenous shunting, the feeding arteries had decreased in calibre and a venous varix had gone. As these angiograms were performed 2 years after STRS, we are awaiting the 3-year follow-up, as these DAVF may go on to resolve
99
completely. Only one patient showed angiographic evidence of deterioration. Of the remaining three patients, one was unable to undergo angiography, but clinical improvement was dramatic and a CT scan did not suggest that the DAVF was still there. Another patient was too heavy for the angiography table, but showed considerable clinical improvement. There was no follow-up angiogram for the final patient, although his GP believes that it has been performed, and therefore we are still waiting to hear from the referring consultant. On the basis of the angiographic follow-up we have a complete obliteration rate of 77%.
DISCUSSION
DAVF are difficult lesions to treat. The natural history of these rare acquired lesions is variable, but can be predicted by certain clinical and angiographic features. DAVF presenting with seizures, focal neurological deficit, progressive visual symptoms, intractable bruit, intracranial hypertension and haemorrhage require treatment. Awad reported a study linking particular angiographic characteristics with a high risk of haemorrhage [3]. These patients should be treated independent of symptoms [1,24–26]. In the study by Malik the risk of haemorrhage of a DAVF close to a dural venous sinus had a lifetime risk of haemorrhage of 75% and one located away from the major sinuses had a 5% risk of haemorrhage [26]. Borden’s classification attempts to identify those at high risk of haemorrhage by identifying those lesions with retrograde flow in the venous sinus, and reflux into cortical veins [27]. There have been a handful of international studies in the treatment of DAVF, but this is the first study to look at the treatment of DAVF using STRS in the United Kingdom. Link et al. [22] reported a study including 29 patients with DAVF, however they used a combination of STRS and embolization post-STRS, for patients with symptoms, and those at very high risk of haemorrhage. They reported an obliteration rate of 72% in the 18 patients that they followed up with angiograms at a mean period of 24 months. Our results show an obliteration rate of 77%, with a higher percentage of patients followed up with angiography. Lewis described seven patients with a tentorial DAVF that were treated with transarterial embolization and linear accelerated radiosurgery [19]. They reported an obliteration rate of 56%. There are a few reports of conventional radiation therapy for DAVF, but they do not include outcome data. The combination of STRS with embolization has been reported with varying degrees of completeness [15,17,18]. After a mean follow-up period of 24 months 10 of our DAVF showed complete obliteration on angiogram. Thirteen of our 16 patients had follow-up angiograms, which gave us an obliteration rate of 77%. We had very good results with two other patients whose angiograms showed enormous reduction in size. In one of these two
100
Fig. 1 – Patient 1. (a) (Left) illustrating a left external carotid injection. An extensive tentorial dural fistula is demonstrated with a varix (arrowhead) and venous shunting into the straight sinus (arrow). (b) (Right) demonstrating complete resolution of the fistula 2 years following radiosurgery.
Fig. 2 – Patient 1. (a) (Left) illustrating a left internal carotid injection. Further filling of the fistula is demonstrated via hypertrophied meningo-hypophyseal branches (arrowhead). (b) (Right) demonstrating complete resolution of the fistula 2 years following radiosurgery.
we are awaiting the 3-year angiogram later this year as we believe it will be completely resolved by then. In one patient the DAVF became bigger. This patient has a stable deficit and it has been decided that he will undergo no further treatment. The other four DAVFs, three patients, did not undergo angiogram follow-up. One patient was too heavy for the angiographic table and the other was unable to stay still for long enough for
angiography because of an unrelated medical condition. However both these patients showed dramatic improvement in their clinical symptoms related to their DAVF. The final patient with the two DAVFs has not undergone angiography. There has been no reported haemorrhage in any of our patients to date. There have been few treatment related complications so far reported both by this review
101
Fig. 3. – Patient 2. (a) (Left) illustrating a right external carotid artery injection. A carotico-cavernous fistula is demonstrated with filling of the cavernous sinus (arrow). Venous drainage into the orbit via a hypertrophied superior ophthalmic vein is shown (arrowhead). (b) (Right) demonstrating complete resolution of the fistula 2 years following stereotactic radiosurgery.
and Link et al. One of our patients went on to become totally deaf in the left ear. It is possible that the vestibulocochlear nerve damage 3-years post treatment was directly related to the treatment but it might also have occurred as a result of successful obliteration. There are two cases of sudden deafness after spontaneous occlusion of a DAVF [28]. Three of our patients had strokes after STRS. One patient had a parietal infarct 26 months post STRS; one was in the left occipital region and the other in the right tentorial region. One patient had a small cerebellar stroke 56 months post STRS of a DAVF in the petrous apex. This seems unlikely to be related to the treatment or effects of untreated DAVF as this malformation showed complete obliteration at 24 months. The third patient had a left-sided stroke 3 months post STRS and a further TIA 13 months later. Whilst DAVF take 12–36 months to obliterate this patient was still at risk from haemorrhage, but this does not appear to have occurred in this patient. The rapidity of symptom onset was not suggestive of radionecrosis. To date none of our patients have developed new or recurrent symptoms to suggest that an angiographically obliterated DAVF had recanalized, although 4-year angiogram have not been universally performed.
CONCLUSION
STRS provides an excellent treatment option for DAVF. It is especially relevant for those patients who
are unfit for surgery, or whose DAVF is inaccessible to either surgery or embolization, but are at high risk of haemorrhage. Acknowledgements. Dr M. Collins for helping me with the initial field of research. Dr S. Coley and Dr T. Hodgson for their constant support and guidance during this study. The Stereotactic Radiosurgery, and Neuroradiology Departments at the Royal Hallamshire Hospital, Sheffield for all their help and support with this study. REFERENCES 1 Awad IA, Little JR, Akrawi WP, Ahl J. Intracranial dural arteriovenous malformations: factors predisposing to an aggressive neurological course. J Neurosurg 1990;72:839–850. 2 Aminoff MJ. Vascular anomalies in the intracranial dura mater. Brain 1973;96:601–612. 3 Chaudhary MY, Sachdev VP, Cho SH, Weitzner I Jr, Purlyic S, Huang YP. Dural arteriovenous malformation of the major venous sinuses. An acquired lesion. AJNR 1982;3:13–19. 4 Houser OW, Baker HL Jr, Rhoton AL Jr, Okazaki H. Intracranial dural arteriovenous malformations. Radiology 1972;105:55–64. 5 Houser OW, Campbell JK, Campbell RJ, Sundt TM Jr. Arteriovenous malformations affecting the transverse dural venous — an acquired lesion. Mayo Clin Proc 1979;54:651–661. 6 Kerber CW, Newton TH. The macro and microvasculature of the dura mater. Neuroradiology 1973;6:175–179. 7 Vidyasagar C. Persistent embryonic veins in the arteriovenous malformation of the dura. Acta Neurochir(Wien) 1979;48:199–216. 8 Malik GM, Pearce JE, Ausman JI, et al. Dural arteriovenous malformations and intracranial haemorrhage. Neurosurgery 1984;15:332–339. 9 Halbach VV, Higashide RT, Hieshma GB, et al. Dural fistulae involving the cavernous sinus: results of treatment in 30 patients. Radiology 1987;163:437–442.
102
10 Halbach VV, Higashide RT, Hieshma GB, et al. Dural fistulae involving the transverous and sigmoid sinuses: results of treatment in 28 patients. Radiology 1987;163:443–447. 11 Costin JA, Weinstein MA, Berlin AJ, et al. Dural arteriovenous malformations involving the cavernous sinus: a case report. Br J Opthalmol 1978;62:478–482. 12 Courtheoux P, Labbe D, Hamel C, et al. Treatment of bilateral spontaneous dural carotid-cavernous fistulas by coils and sclerotherapy. Case report. J Neurosurg 1987;66:468–470. 13 Fermand M, Reizine D, Melki JP, et al. Long term follow up of 43 pure dural arteriovenous fistulae (AVF) of the lateral sinus. Neuroradiology 1987;29:348–353. 14 Gobin YP, Rogopoulos A, Aymad A, et al. Endovascular treatment of intracranial dural arteriovenous fistulae with spinal perimedullary venous drainage. J Neurosurg 1992;77:718–723. 15 Hardy RW Jr, Costin JA, Weinstein M, et al. External carotid cavernous fistula treated by transfemoral embolization. Surg Neurol 1978;9:255–256. 16 Vinuela F, Fox AJ, Debrum GM, et al. Spontaneous carotid cavernous fistulas. Clinical, radiological and therapeutic considerations. Experience with 20 cases. J Neurosurg 1984;60:976–984. 17 Chandler HC, Freidman WA. Successful radiosurgical treatment of a dural arteriovenous malformation: Case report. Neurosurgery 1993;33:139–142. 18 Fabrikant JI, Lyman JT, Hosobuchi Y. Stereotactic heavy ionbragg peak radiosurgery for intracranial vascular disorders: method for treatment of deep arteriovenous malformations. Br J Radiol 1984;57:479–490. 19 Lewis AI, Tomsick TA, Tew JM Jr. Management of tentorial arteriovenous malformations: transarterial embolization combined
20
21
22
23
24
25
26
27
28
with Stereotactic radiosurgery or surgery. J Neurosurg 1994; 81:851–859. Steiner L, Lindquist C, Adler JR, et al. Clinical outcome of radiosurgery for cerebral arteriovenous malformations. J Neurosurg 1992;77:1–8. Barnwell SL, Halbach VV, Higashida RT, et al. Complex dural arteriovenous fistulae. Results of combined endovascular and neurosurgical treatment in 16 patients. J Neurosurg 1989;71: 352–358. Link MJ, Coffey RJ, Nichols DA. The role of radiosurgery and particulate embolization in the treatment of dural arteriovenous fistulae. J Neurosurg 1995;84:804–809. Vinuela FV, Debrun GM, Fox AJ, et al. Detachable calibrated leak balloon for superselective angiograph and embolization of dural arteriovenous malformations. J Neurosurg 1983;58:817– 823. King WA, Martin NA. Intracerebral haemorrhage due to dural arteriovenous malformations and fistulae. Neurosurg Clin North Am 1992;3:577–590. Lasjaun P, Chui M, Ter Brugge K. Neurological manifestations of intracranial dural arteriovenous malformations. J Neurosurg 1986;64:724–730. Malik GM, Pearce JE, Ausman JI, et al. Dural arteriovenous malformations and intracranial haemorrhage. Neurosurgery 1984;15:332–339. Borden JA, Wu JK, Shucart WA. A proposed classification for spinal and cranial dural arteriovenous fistulous malformations and implications for treatment. J Neurosurg 1995;82:166–179. Landman JA, Braun IF. Spontaneous closure of a dural arteriovenous fistula associated with acute hearing loss. AJNR 1985; 6:448–449.