Microsurgical Resection of Incompletely Obliterated Intracranial Arteriovenous Malformations Following Stereotactic Radiosurgery

Transcription

1 II-2. Selection of Treatment Microsurgical Resection of Incompletely Obliterated Intracranial Arteriovenous Malformations Following Stereotactic Radiosurgery Steven D. CHANG*, Gary K. STEINBERG*, Richard P. LEVY*, Michael P. MARKS*,**, Ken A. FRANKEL*, Deborah L. SHUSTER***, and Mary L. MARCELLUS** Departments of *Neurosurgery, **Radiology, and ***Pathology (Neuropathology), and the Stanford Stroke Center, Stanford University School of Medicine, Stanford, California, U. S. A. Abstract Radiosurgery is effective in obliterating small arteriovenous malformations (AVMs), but less successful in thrombosing larger AVMs. This study reviewed patients who underwent surgical resection of their large AVMs following failed radiosurgical obliteration. AVMs from 36 patients (aged 7 to 64 years, mean 29.9) were surgically resected 1 to 11 years after radiosurgery. Initial AVM volumes were 0.7 to 117cm3 (mean 21.6cm 3), and radiosurgical doses ranged from 4.6 to 45 Gray equivalent (GyE) (mean 21.1 GyE). Thirty AVMs (83%) were located in eloquent tissue. Venous drainage was deep (14), superficial (13), or both (9). Spetzler grades were II (2), III (12), IV (18), and V (4). Nine patients suffered rehemorrhage after radiosurgery but prior to surgery, while three patients developed radiation necrosis. Twenty-seven patients underwent endovascular embolization prior to surgery. During microsurgical resection, the AVMs were found to be significantly less vascular and more easily resected, compared to AVMs in patients who had not received radiosurgery. Histology showed endothelial proliferation with hyaline and mineralization in vessel walls. Partial or complete thrombosis of some AVM vessels, and evidence of vessel and brain necrosis were noted in many cases. Clinical outcome was excellent or good in 34 cases, with two patients dying of rebleeding from residual AVM. Five patients were neurologically worse following microsurgical resection. Final outcome was largely related to the pretreatment grade. Radiosurgery several years prior to surgical resection appears useful in treating unusually large and complex AVMs. Key words: arteriovenous malformation, embolization, microsurgery, stereotactic radiosurgery Introduction Stereotactic radiosurgery has been used for several decades to treat intracranial arteriovenous malformations (AVMs), especially those located in surgically inaccessible regions or in patients who are poor surgical candidates. 3.8,10-12,29, 33-37, 41) This technique has proven to obliterate small to moderate sized AVMs with high success and low complication rates. 2, 4, 7, 8, 12, 13, 24-29, 37-41) Drawbacks to this treatment method include the potential for rehemorrhage during the latency period before complete obliteration occurs and a decreased rate of obliteration as well as higher risk of radiation-induced complications for larger AVMs. 8, 14, 25,,29, 26, 37,,41) Additionally, some neurosurgeons have noted increased difficulty operating on AVMs following stereotactic radiosurgery, due to radiation changes. This report reviews our results in 36 patients who underwent surgical resection of AVMs 1 to 11 years after radiosurgical treatment. Some of this data has been previously reported. 36) Methods Over a 7 year period ( ), 36 patients underwent surgical AVM resection by the senior author (G. K. S. ) 1 to 11 years following radiosurgery treatment for cases of incomplete AVM obliteration (34) or radiation necrosis (2). Mean patient age was 29.9 years (range 7 to 64 years). Patients were graded clinically both before and after treatment by the scale of 200 Neural Med Chir Suppl (Tokyo) 38, 200 `207, 1998

2 Resection of AVMs after Radiosurgery 201 Table 1 Presenting symptoms* *Some patients presented with more than one symptom. Table 2 Arteriovenous malformation location Table 3 Arteriovenous malformation characteristics specific locations are shown in Table 2. For each AVM, the size was measured, venous drainage was noted as superficial, deep, or both, and the Spetzler- Martin grade was determined 34) (Table 3). All patients were treated with stereotactic radiosurgery prior to microsurgical resection. The doses of radiation used ranged from 4.6 to 45 Gray equivalent (GyE) (mean 21.1 GyE). AVM volume ranged from 0.7 to 117.8cm3 (mean 21.6cm3). Clinical follow-up and magnetic resonance images were obtained at least every 6 months following treatment, and cerebral angiograms every 12 months post-treatment to document AVM obliteration. Prior to surgery, embolization with polyvinyl alcohol or N-butyl cyanoacrylate was performed in 27 patients (75%). Embolization was performed prior to radiosurgery (9 patients), after radiosurgery (11 patients), or both before and after radiosurgery (7 patients). Surgical removal of the AVM was performed on all patients following failed AVM obliteration and all resected AVMs were submitted for histological review. The histology team reviewed each AVM without knowledge of AVM location, volume, treatment dose, or clinical patient status. AVMs were graded for endothelial proliferation, hyaline and mineralization in vessel walls, partial or complete vessel thrombosis, and necrosis of vessels or adjacent brain tissue. A semi-quantitative scale (mild, moderate, severe) was used to classify the extent of radiation-induced change. Results Drake et al. 10, 11): excellent (normal lifestyle with no neurological handicaps), good (able to work and live independently despite having a neurological deficit), poor (dependent on family or nursing for help due to a severe neurological deficit), or dead. At initial presentation, 26 patients were graded excellent, and 10 good, with no patients graded poor. Presenting symptoms are shown in Table 1. Thirty patients had AVMs located in "eloquent" or critical areas while six were located in non-eloquent areas (Figs. 1-3); Nine patients (25%) suffered an AVM hemorrhage between radiosurgery and microsurgical resection. Seven patients (19%) suffered one rebleed. The dose of radiation received in these patients was 15, 17, 18, 20, 28, 45, and 20 GyE. Another patient suffered two rebleeds following a radiation dose of 15 GyE, and one patient suffered three rebleeds after a radiation dose of 28 GyE. Three patients who rehemorrhaged developed a worsened neurological status (two transient, one permanent). Four patients (11%) had complications directly related to radiosurgery (Table 4). The percent of AVM volume obliterated with each embolization session as measured on angiograms ranged from 10% to 60% with a mean reduction of 28%. The total percent of AVM volume obliterated through all sessions of embolization for each patient ranged from 15% to 90% with a mean reduction of 55%. Embolization complications are shown in Table 4. During surgery, AVMs were noted to be significantly less vascular, partially thrombosed, and Neurol Med Chir Suppl (Tokyo) 38, 1998

3 202 S. D. Chang et al. Table 4 Complications *15 Gray equivalent (GyE) to 18cm3. **25 GyE to 24 cm3, 35 GyE to 7.5cm3, and 25 GyE to 18cm3. AVM: arteriovenous malformation. more easily resected compared to non-radiated AVMs. Despite the persistent angiographic filling of these AVMs, much of the small vessel component was obliterated by the radiosurgery. Furthermore, some patients had partial volume reduction in their AVM from radiosurgery-induced thrombosis, allowing resection of smaller, residual patent AVM. In 31 cases (86%), complete surgical resection was achieved and confirmed by angiography; in five cases a partial resection was achieved. Surgical complications were modest, with only four patients experiencing permanent neurological deficits (Table 4). Fig. 1 A 23-year-old female presenting with hemorrhage from a deep right thalamic arteriovenous malformation (AVM). Coronal magnetic resonance image (A). Anteroposterior (B) and lateral views (C) demonstrate AVM blood supply from the posterior circulation. The areas of the AVM targeted for stereotactic radiosurgery are outlined. Anteroposterior (D) and lateral views (E) 2 years following radiosurgery (25 Gray equivalent to 7000mm3) demonstrate approximately 80% obliteration of AVM. Anteroposterior (F) and lateral views (G) following microsurgical resection show complete resection of the AVM. Reproduced with permission from Steinberg et al. Surgical resection of large incompletely treated intracranial arteriovenous malformations following stereotactic radiosurgery. J Neurosurg 84: , ) Neurol Med Chir Suppl (Tokyo) 38, 1998

4 Resection of AVMs after Radiosurgery 203 Fig. 2 A 28-year-old male status post-four hemorrhages from a large right basal ganglia/thalamic arteriovenous malformation (AVM). Anteroposterior (A) and lateral views (B, C) showing AVM filling from the anterior circulation (B), and posterior circulation (A, C) 9 years after treatment with proton beam radiosurgery (17.5 Gray equivalent to 93,000mm3). Axial magnetic resonance image (D) shows extent of AVM after embolization but prior to microsurgical resection. The patient had a moderate left hemiparesis and homonymous hemianopsia prior to the first surgery and remained unchanged through the three surgeries. Anteroposterior anterior circulation view (E), and anteroposterior posterior circulation view (F) following surgery show complete resection of AVM. Reproduced with permission from Steinberg et al. Surgical resection of large incompletely treated intracranial arteriovenous malformations following stereotactic radiosurgery. J Neurosurg 84: , ) Upon histological review, the majority of the AVMs in this study showed evidence of endothelial proliferation, hyaline and mineralization deposits in vascular walls, partial or complete thrombosis of vessels, and necrosis of AVM vessels as well as adjacent brain tissue (Table 5). These changes likely accounted for the ease of resection of these AVMs compared to our experience with non-radiosurgery-treated AVMs. There was a strong correlation (p<0.0005) between the extent of radiation changes as determined by the semi-quantitative scale used in histolog- Table 5 Histological findings AVM: arteriovenous malformation. Neural Med Chir Suppl (Tokyo) 38, 1998

5 204 S. D. Chang et al. grade III, excellent (11) and good (1); grade IV, excellent (11), good (6), and dead (1); grade V, good (3) and dead (1). Discussion Fig. 3 A 34-year-old male with a right temporo-occipital arteriovenous malformation (AVM) on axial T2-weighted magnetic resonance image (A). The patient was treated with 15 Gray equivalent to 47,500mm3. Anteroposterior (B) and lateral views (C) 2 years after radiosurgery show AVM filling from branches of the middle cerebral artery with no significant reduction in AVM size. The patient underwent embolization (resulting in 70% obliteration) and then microsurgical resection. Lateral view (D) following surgical resection shows complete AVM obliteration. Reproduced with permission from Steinberg et al. Surgical resection of large incompletely treated intracranial arteriovenous malformations following stereotactic radiosurgery. J Neurosurg 84: , ) ical grading and the dose of radiation received. All but one patient receiving>20 GyE to the AVM developed moderate to severe radiation changes and all three patients treated with>30 GyE had severe radiation-induced changes. Clinical follow-up after surgical resection ranged from 12 to 90 months. After all multimodality treatments, the outcome was excellent in 24 patients, good in 10 patients, with two patients dying, both secondary to hemorrhage from residual AVM. Patient outcome worsened with higher Spetzler grade: grade II resulted in an excellent outcome (2); Stereotactic radiosurgery is generally a successful treatment for selected small to moderate sized intracranial AVMs. AVMs less than 3 cm in diameter treated with GyE have a 3 year obliteration rate of 80-95% with only mild morbidity ( % permanent neurological deficits, % transient deficits). 2, 4, 7, 12, 8, 13, 24, 28, 29, 37, 41) However, larger AVMs have a much lower obliteration rate (33-58%), even after 3 years, and a higher complication rate (20-30%) using treatment doses of GyE. 8, 12, 13, 25, 26, 29, 37, 41) Higher obliteration rates can be achieved for larger AVMs using higher treatment doses (25-45 GyE) but the risk of significant radiation-induced complications increases to approximately 50%. 12, 21, 37) The risks of surgery or embolization plus surgery for treating these large AVMs in critical regions is also significant. 3, 10, 11, 17, 22, 23, 34, 35) This series reviews patients who failed radiosurgical obliteration of their AVMs. Reasons for incomplete obliteration include subtherapeutic doses, inaccurate stereotactic targeting, too large a treatment volume, or other unknown intrinsic AVM properties. There were a total of 12 AVM bleeds over 150 patients/yr following radiosurgery treatment, corresponding to an annual rebleed rate of 8%. Since the patients in this small study were selected for non-obliteration, it is not known whether this is an increase in the natural history for AVM bleeds (3-4%/yr). 5, 9, 14, 15, 32) A % annual rebleed rate for AVMs following radiosurgery has been previously reported. 8, 13, 41) The data from this series does not support a protective effect of radiosurgery for incompletely obliterated AVMs. 4, 18, 19, 21, 40) Radiation necrosis following radiosurgery was noted in three patients (8%). All three patients had worsening in their neurological status, but two improved with surgical resection of the necrosis. This rate of radionecrosis is similar to other radiosurgical series, 8, 12, 29, 38, 41) and confirms that symptomatic radiation-induced necrosis is rare when doses of less than 25 GyE are used. 18, 25, 27, 37) Surgical resection of these AVMs revealed two benefits of radiosurgery as a preoperative adjunct. First, some patients had partial AVM thrombosis which significantly reduced the volume of residual AVM that required resection. Second, the radiated (but patent) AVMs were less vascular, even in nonembolized areas compared to AVMs not previously radiated. Radiosurgically treated AVM vessels were Neural Med Chir Suppl (Tokyo) 38, 1998

6 Resection of AVMs after Radiosurgery 205 easier to coagulate with a bipolar, facilitating quicker and safer resections with less blood loss. Despite preoperative angiograms usually demonstrating significant residual AVM, the observations at surgery suggested that the previous radiosurgery had obliterated the small vessel component of the AVM not visible on the angiogram. The clinical results following radiosurgery, embolization, and microsurgery are encouraging in this series. Following multi-modality therapy 25 of the 36 patients were clinically unchanged, six were improved, and five were worse. Deterioration of a patient's clinical condition was due to rebleeding in two patients, embolization complications in one patient, and both embolization and microsurgical complications in two patients. The clinical outcome of patients in our series was generally predicted by each AVM Spetzler-Martin classification. 34) The majority of our patients had AVMs of Spetzler-Martin grade III (12), grade IV (18), or grade V (4). Eleven of the grade III patients (92%) had an excellent final outcome, confirming the good prognosis with these patients following surgical treatment alone. 1, 3, 10, 11, 16, 17, 22, 31, 35) The grade IV patients had successful results in 17 patients (94%) with 11 excellent results (61%) and six good results (33%). One grade IV patient died (6%). Three of the four grade V patients (75%) had good results while one (25%) died. In our view prior radiosurgery decreases the surgical morbidity and improves the clinical outcome compared with preoperative embolization alone. Microscopic examination of these AVMs confirms the intraoperative impressions of small AVM vessel thrombosis. Evidence of endothelial proliferation, hyaline and mineralization deposits in AVM vascular walls, partial or complete thrombosis of some AVM vessels, as well as necrosis of AVM vessels and adjacent brain tissue were noted, and likely caused decreased vascularity of the AVM and made surgical resection easier. Patients receiving>20 GyE developed more radiation-induced vascular changes compared to those patients receiving<20 GyE; all patients receiving>30 GyE had severe radiation-induced vascular changes. The histological changes noted in this series confirms the few previous histological reports of radiated AV Ms. 6, 20, 39) Patients whose AVMs are not obliterated completely 3 years following radiosurgery treatment have several treatment options. Data suggests that if no further treatment is provided, patients are not protected from future hemorrhage. 4, 18, 19, 21, 37, 40) Alternatively such patients could have a second course of stereotactic radiosurgery. The drawbacks of such an approach include a second latency period of 1 to 3 years before obliteration occurs, the possibility that a second radiosurgery treatment may still not obliterate the AVM, and the increased risk of radiation injury. Another strategy used in certain patients is to embolize portions of the non-obliterated AVM to reduce the amount of volume requiring a second radiosurgery treatment. 30) Long-term outcome will be required to determine the overall efficacy of repeat radiosurgery or embolization plus repeat radiosurgery for AVM patients. The approach utilized in this series of 36 patients was to use microsurgery or embolization plus microsurgery for AVMs which failed to obliterate after radiosurgery alone. This was found to be valuable in treating some of the AVMs, particularly when patients rebled from residual AVM and there was a greater urgency to cure the AVM, rather than waiting for a second radiosurgery treatment to mature. For larger AVMs, stereotactic radiosurgery prior to microsurgical resection can transform these AVMs into those which can be resected with a high success rate and only modest morbidity. Acknowledgments This work was supported in part by funding from Bernard and Ronni Lacroute (G. K. S. ) and from the William Randolph Hearst Foundation (G. K. S. ). References 1) Adelt D, Zeumer H, Wolters J: Surgical treatment of cerebral arteriovenous malformations. Follow-up study of 43 cases. Acta Neurochir (Wien) 76: 45-49, ) Alexander E III, Loeffler JS: Radiosurgery using a modified linear accelerator. Neurosurg Clin North Am 3: ,1992 3) Batjer H, Samson D: Arteriovenous malformations of the posterior fossa: clinical presentation, diagnostic evaluation, and surgical treatment. J Neurosurg 64: ,1986 4) Betti OO, Munari, C, Rosier R: Stereotactic radiosurgery with the linear accelerator: Treatment of arteriovenous malformations. Neurosurgery 24: ,1989 5) Brown RD Jr, Wiebers DO, Forbes G, O'Fallon WM, Piepgras DG, Marsh WR, Maciunas RJ: The natural history of unruptured intracranial arteriovenous malformations. J Neurosurg 68: , ) Chang SD, Shuster DL, Steinberg GK, Levy RP, Frankel KA: Stereotactic radiosurgery of arteriovenous malformations: Pathologic changes in resected tissue. Clinl Neuropathol 16: , ) Columbo F, Benedetti A, Pozza F, Marchetti C, Chierego G: Linear accelerator radiosurgery of cerebral arteriovenous malformations. Neurosurgery Neural Med Chir Suppl (Tokyo) 38, 1998

7 206 S. D. Chang et al. 24: ,1989 8) Columbo F, Pozza F, Chierego G, Casentini L, De Luca G, Francescon P: Linear accelerator radiosurgery of cerebral arteriovenous malformations: An update. Neurosurgery 34: 14-21, ) Crawford PM, West CR, Chadwick DW, Shaw MD: Arteriovenous malformations of brain: Natural history in unoperated patients. J Neurol Neurosurg Psychiatry 49: 1-10, ) Drake CG: Cerebral arteriovenous malformations: considerations for and experience with surgical treatment in 166 cases. Clin Neurosurg 26: , ) Drake CG, Friedman AH, Peerless SI: Posterior fossa arteriovenous malformations. J Neurosurg 64: 1-10, ) Fabrikant JI, Levy RP, Steinberg GK, Phillips MH, Frankel KA, Lyman IT, Marks MP, Silverberg GD: Charged particle radiosurgery for intracranial vascular malformations. Neurosurg Clin North Am 3: , ) Friedman WA, Bova FJ: Linear accelerator radiosurgery for arteriovenous malformations. J Neurosurg 77: , ) Fults D, Kelly DL Jr: Natural history of arteriovenous malformations of the brain: a clinical study. Neurosurgery 15: , ) Graf CJ, Perret GE, Torner JC: Bleeding from cerebral arteriovenous malformations as part of their natural history. J Neurosurg 58: , ) Hamilton MG, Spetzler RF: The prospective application of a grading system for arteriovenous malformations. Neurosurgery 34: 2-7, ) Heros RC, Korosue K, Diebold PM: Surgical excision of cerebral arteriovenous malformations: Late results. Neurosurgery 26: , ) Kjellberg RN: Proton beam therapy for arteriovenous malformations of the brain, in Schmidek HH, Sweet WH (eds): Operative Neurosurgical Techniques: Indications, Methods, Results, ed 2, vol 2. Philadelphia, WB Saunders, 1988, pp ) Kjellberg RN, Candia GJ: Stereotactic proton beam therapy for cerebral AVMs, in: Harvard Radiology Update. Chestnut Hill, Harvard Medical School, 1990, pp ) Kjellberg RN, Davis KR, Lyons S, Butler W, Adams RD: Bragg peak proton beam therapy for arteriovenous malformations of the brain. Clin Neurosurg 31: , ) Kjellberg RN, Hanamura T, Davis KR, Lyons SL, Adams RD: Bragg-peak proton-beam therapy for arteriovenous malformations of the brain. N Engl J Med 309: , ) Kondziolka D, Humphreys RP, Hoffman HJ, Hendrick EB, Drake JM: Arteriovenous malformations of the brain in children: A forty year experience. Can J Neural Sci 19: 40-45, ) Korosue K, Heros RC: Complications of complete surgical resection of cerebral AVMs, in Barrow DL (ed): Intracranial Vascular Malformations. Park Ridge, Ill., American Association of Neurological Surgeons, 1990, pp ) Leksell L: Stereotactic radiosurgery. J Neural Neurosurg Psychiatry 46: , ) Levy RP, Fabrikant JI, Frankel KA, Phillips MH, Lyman JT: Stereotactic heavy-charged particle Bragg peak radiosurgery for the treatment of intracranial arteriovenous malformations in childhood and adolescence. Neurosurgery 24: , ) Levy RP, Fabrikant JI, Frankel KA, Phillips MH, Lyman JT: Charged-particle radiosurgery of the brain. Neurosurg Clin North Am 1: , ) Levy RP, Fabrikant JI, Frankel KA, Phillips MH, Steinberg GK, Marks MP, DeLaPaz RL, Chuang FYS: Clinical radiologic evaluation of sequelae of stereotactic radiosurgery for intracranial arteriovenous malformations, in Steiner L (ed): Radiosurgery: Baselines and Trends. New York, Raven, 1992, pp ) Lunsford LD, Kondziolka D, Bissonette DJ, Maitz AH, Flickinger JC: Stereotactic radiosurgery of brain vascular malformations. Neurosurg Clin North Am 3: 79-98, ) Lunsford LD, Kondziolka D, Flickinger JC, Bissonette DJ, Jungreis CA, Maitz AH, Horton JA, Coffey RJ: Stereotactic radiosurgery for arteriovenous malformations of the brain. J Neurosurg 75: , ) Marks MP, Lane B, Steinberg GK, Fabrikant JI, Levy RP, Frankel KA, Phillips MH: Endovascular treatment of cerebral arteriovenous malformations following radiosurgery. AJNR Am J Neuroradiol 14: , ) Nornes H, Lundar T, Wikeby P: Cerebral arteriovenous malformations: Results of microsurgical management. Acta Neurochir (Wien) 50: , ) Ondra SL, Troupp H, George ED, Schwab K: The natural history of symptomatic arteriovenous malformations of the brain: A 24 year follow up assessment. J Neurosurg 73: , ) Pollock BE, Kondziolka D, Lunsford LD, Bissonette PA, Flickinger JC: Repeat stereotactic radiosurgery of arteriovenous malformations: Factors associated with incomplete obliteration. Neurosurgery 38: , ) Spetzler RF, Martin NA: A proposed grading system for arteriovenous malformations. J Neurosurg 65: , ) Stein BM: Surgical decisions in vascular malformations of the brain, in Barnett HIM, Stein BM, Mohr JP, Yatzu FM (eds): Stroke: Pathophysiology, Diagnosis, and Management, vol 2. New York, Churchill Livingstone, 1992, pp ) Steinberg GK, Chang SD, Levy RP, Marks MP, Frankel K, Marcellus M: Surgical resection of large incompletely treated intracranial arteriovenous malformations following stereotactic radiosurgery. J Neurosurg 84: , ) Steinberg GK, Fabrikant JI, Marks MP, Levy RP, Frankel KA, Phillips MH, Shuer LM, Silverberg GD: Stereotactic heavy-charged-particle Bragg-peak radia- Neurol Med Chir Suppl (Tokyo) 38, 1998

8 Resection of AVMs after Radiosurgery 207 tion for intracranial arteriovenous malformations. N Engl J Med 323: , ) Steinberg GK, Levy RP, Marks MP, Fabrikant JI: Vascular malformations: Charged-particle radiosurgery, in Alexander E, Loeffler JS, Lunsford LD (eds): Stereotactic Radiosurgery. New York, McGraw-Hill, 1993, pp ) Steiner L: Treatment of arteriovenous malformations by radiosurgery, in Wilson CB, Stein BM (eds): Intracranial Arteriovenous Malformations. Baltimore, Williams & Wilkins, 1984, pp ) Steiner L: Radiosurgery in cerebral arteriovenous malformations, in Fein JM, Flamm ES (eds): Cerebrovascular Surgery, vol 4. New York, Springer- Verlag, 1985, pp ) Steiner L, Lindquist C, Adler JR, Turner JC, Alves W, Steiner M: Clinical outcome of radiosurgery for cerebral arteriovenous malformations. J Neurosurg 77: 1-8, 1992 Address reprint requests to: G. K. Steinberg, M. D., Room R-281, Department of Neurosurgery, Stanford University Medical Center, 300 Pasteur Drive, Stanford, CA 94062, U. S. A. Neurol Med Chir Suppl (Tokyo) 38, 1998