0148-396X /82/1106-0776802.00/0 NEUROSURGERY Copyright © 1982 by the Congress of Neurological Surgeons Vol. 11. No. 6. 1982 Printed in U.S.A Spontaneous Regression of Arteriovenous Malformations Daniel G. Nehls, M.D., CPT MC USA and Hal W. Pittman, M.D. Neurosurgical Department, Barrow Neurological Institute. St. Joseph's Hospital and Medical Center, Phoenix, Arizona Spontaneous regression of an arteriovenous malformation (AVM) is quite rare. When a regression does occur, there is usually an associated factor, such as intracranial Twenty-one cases of AVM regression identified in hemorrhage. operation, radiation therapy, or new neurological deficit. the literature are reviewed, An additional case in which the resolution was totally spontaneous is presented. Regressions occurred over intervals ranging from 7 months to 21 years. Several mechanisms for regression are discussed. (Neurosurgery 11:776-780, 1982) Key words: Arteriovenous malformation, Cerebra The documented regression of an arteriovenous malforma- tion (AVM) without therapy is a rare event. Spontaneous total disappearance is exceedingly rare. Partial reduction in the size of AVMs has been reported to occur spontaneously (23) and in association with radiation therapy (4, 28), intracranial hemor- rhage (13. 18), and sudden occlusion of the middle cerebral artery (MCA). Total disappearance has been reported in asso- ciation with subarachnoid hemorrhage (3, 5, 6, 14, 20), subdural hematoma (16), intracerebral hematoma (1, 15), sudden hem- iplegia (10), and the use of oral contraceptives (8). Spontaneous regression has been presumed from the demonstration at op- eration of thrombosed AVMs in patients with normal arterio- grams (1, 19), The total spontaneous disappearance of dural AVMs has been documented in patients receiving no therapy (7, 17). We review the literature and present a patient with a cerebral AVM that disappeared spontaneously without opera- tion, radiation therapy, coagulopathy, documented or pre- sumed intracranial hemorrhage, trauma, or the development of a neurological deficit. The patient had an incidental right MCA aneurysm, and his mother had been treated in the past by one of the authors (H.W.P.) for a left MCA aneurysm. CASE REPORT The patient is a 59-year-old man who. 13 years before admission, had been evaluated for an episode of severe, stab- bing right-sided headache that occurred while he was walking across a parking lot. The pain lasted several minutes and recurred several times over the following half-hour, He had experienced generalized mild headaches for about 2 years before the episode, but he noted an increase in the severity and frequency of the headaches after the episode. He stated that the headaches were more often right-sided and were on occa- sion accompanied by bilateral arm and leg weakness. He also noted the onset of occasional transient left hand numbness and right eye “fuzziness” not associated with the headaches. The patient’s family medical history was notable for the fact that his mother had undergone clipping of a left MCA aneurysm. The results of physical examination, including a detailed neu- rological examination, were totally normal. A lumbar puncture | month after the episode revealed no xanthochromia of the cerebrospinal fluid. An arteriogram re- vealed a right parietal AVM and a 7-mm saccular aneurysm of the right MCA. The AVM was approximately 2 cm in diameter. It was fed by a single vessel. the posterior parietal branch of the MCA. A single draining vein coursed superiorly and me- dially to the superior sagittal sinus (Fig. |). The patient's symptoms persisted. and angiography | year later again dem- onstrated the AVM and aneurysm. both unchanged in angio- aneurysm, Spontaneous regression graphic appearance. Neither arteriogram gave any evidence of intracranial hematoma, spasm. or ventriculomegaly. The patient's symptoms subsided gradually until | month before admission, when he experienced sharp pain in his right occiput lasting about | week. A computed tomographic scan revealed no evidence of subarachnoid hemorrhage. intracranial calcification, or ventriculomegaly. but did demonstrate the aneurysm. Angiography defined the aneurysm, which appeared unchanged. No AVM was seen. even on subtracted films (Fig. 2). The patient underwent elective clipping of the aneurysm. No hemosiderin was observed during the procedure. Postoperative angiography showed occlusion of the aneurysm with flow in the MCA distal to the point of clipping. Once again, no AVM was seen. Therefore, an AVM demonstrated on two arterio- grams underwent spontaneous resolution and was absent on two later arteriograms. DISCUSSION There are 25 cases of partial or complete regression of AVMs reported in the literature. Table | summarizes the 21 cases of AYM regression that we reviewed. There were 2 cases noted by Kelly et al. (11) and 2 cases noted by Krayenbiihl and Yasargil (12), but the clinical data were not presented. The intervals between the documentation and regression of the AVMs ranged from 7 months (8, 16) to 21 years (10). In several cases the resolution was thought to have occurred suddenly (3, 10, 26). In others, serial arteriograms showed progressive decline in size of the AVM (13, 14). In our case, the resolution occurred over the course of 13 years. There was no evidence to support either a sudden or a gradual resolution of the AVM in our patient. There are several proposed mechanisms for the regression of AVMs. Svein and Peserico postulated that the abnormal vessels in AVMs are prone to thrombosis, although they also called attention to the relationship between intracranial hemorrhage and the regression of AVMs (28). They thought that radiation therapy had also played a role in the reduction of the size of the AVM in their patient. Eisenman et al. thought that intra- cranial hemorrhage could lead to the resolution of AVMs by exerting a compressive force on the AVM directly by the hematoma and indirectly through local edema (6). They thought that this compression would reduce flow and make the AVM vulnerable to thrombosis. Lakke reported a gradual reduction in the size of vessels observed on serial angiograms (14). This was accompanied by a change in their position that he attributed to intermittent bleeding. He postulated that or- ganization of the clot and gliosis caused kinking of the feeding December 1982 aes he REGRESSION OF AVMs 777 Fic. L. Angiographic appearance of the AVM at the time of its discovery. The AVM is fed by a single vessel, the posterior parietal branch of the MCA. A single draining vein is visualized coursing superiorly and medially. Also present is a 7-mm saccular aneurysm of the MCA. Fic. 2. Arteriograms at the time of aneurysm clipping. 13 years after the original arteriograms. reveal disappearance of the AVM. The arrows indicate the prior feeding vessel vessels, facilitating thrombosis of the AVM. Conforti suggested that the probability of thrombosis is greater when the AVM is small and fed by a single vessel (3). Pool and Potts proposed that arteriosclerosis in the AVM could be a possible cause for the spontaneous regression of AVMs (22). Magidson and Weinberg suggested an interesting mechanism for the resolution of the dural AVM that they reported (17). They proposed that the AVM increased pressure in the dural sinuses, which in turn led to elevation of intracranial pressure. This led to stasis and eventual thrombosis of the sinuses and resolution of the dural AVM. In the cases in which an AVM s demonstrated and subsequently resolved. there was always an additional factor. except in the case reported by Sasaki et al. (23). The most common factor was intracranial bleeding. This was present in 12 cases. There had been prior operation in 5 cases and 778 Authors (Year) Cushing & Bailey (1928) Norlén (1949) DeLange (1955) Paterson & Mc- Kissock (1956) Hook & Johanson (1958) Svien & Peserico (1960) Castaigne et al. (1961) Pecker et al. (1961) Kushner & Alex- ander (1970) Lakke (1970) Conforti (1971) Eisenman et al. (1972) Sukoff et al. (1972) Levine et al. (1973) Hansen & Sogaard (1976) Magidson & Wein- berg (1976) Sasaki et al. (1976) Mabe & Furuse (1977) Endo et al. (1979) Present case Age & Sex 64M ha = 64 F 30M 46 F 49 F 37 F 26 F 59M NEHLS AND PITTMAN Regression” P T T “ P. partial regression: T. total regression. XRT. radiation therapy: SAH. subarachnoid hemorrhage: R. right. radiation therapy in 3. In | patient. the AVM resolved 7 months after she started taking oral contraceptives. None of these factors was present in our patient. Examination of the arteriograms presented in Figures | and TABLE | Reported Cases of AVM Regression Documentation Size of AVM de- creased at second operation Thrombosed AVM at operation Serial angiography Normal angiogram, occluded AVM at operation Thrombosed AVM at operation Residual AVM post- operatively, gone on later angio- gram Serial angiography Normal angiogram x 3, thrombosed AVM at operation Residual AVM post- operatively, gone on later angio- gram Serial angiography Residual AVM post- operatively, serial angiography Serial angiography Serial angiography Thrombosed AVM at operation Serial angiography Residual AVM post- operatively, serial angiography Serial angiography Serial angiography Interval angiography Serial angiography Interval angiography Interval angiography Interval! from Dis- covery to Regression 3yr 4yr 4yr 7 mo 2yr 13 yr Neurosurgery, Vol. 11, No. 6 Presentation’ Seizure. hand pares- thesia Transient field de- fect. SAH. sei- zures SAH Seizures, atrophic R leg Temporal lobe sei- zures Headache, limb numbness, diplo- pia, tinnitis Headache, SAH Fluctuating hemi- paresis, focal sei- zures Seizures, hemiplegia, aphasia SAH Headache, SAH Repeated SAH Headache, SAH Seizures, SAH, hemiparesis Headache, dementia Intracerebral hema- toma Headache, bruit Tinnitis Hemiparesis, focal seizures Hydrocephalus, im- paired develop- ment Tinnitis, vertigo Headache Other Factors” XRT, operation SAH SAH, hematoma Sudden hemiplegia 7 yr postopera- tively, operation XRT. SAH Hematoma Hematoma, SAH, operation, XRT SAH SAH, operation SAH, sudden head- ache & loss of consciousness be- fore last angio- gram SAH SAH Sudden MCA occlu- sion Intracerebral hema- toma, operation Oral contraceptives Dural AVM Subdural hematoma after ventriculo- peritoneal shunt Dural AVM 2 reveals interim disappearance of the AVM. with irregularity and diminution in size of the feeding vessel with no cutoff noted. Minor shift in the position of the feeding vessel on the lateral view can be explained by differences in technique. and December 1982 the position in the anteroposterior view is unchanged. Repeated small hemorrhages with ensuing organization and gliosis, as proposed by Lakke (14), would be an unlikely explanation for the resolution of the AVM without a greater shift of the vessels. Embolization is unlikely without an arterial cutoff. The angio- graphic tapering and irregularity of the feeding vessel suggest partial thrombosis. Several authors have proposed that it is the propensity of the abnormal vessels in AVMs to undergo throm- bosis that accounts for their spontaneous resolution (3, 26, 28). Several other factors could also play a role. The tortuosity and altered mechanical properties of these vessels could contribute to the formation of thrombi by causing stasis, eddy currents, and nonlaminar flow. The tortuosity would also create turbu- lence, which would promote endothelial damage, providing a nidus for thrombi formation and facilitating the development of atherosclerosis. It seems apparent that there are several mechanisms by which AVMs can undergo spontaneous resolution. Regression is most probable in AVMs that are small and fed by a single vessel. It is also more likely after an episode of intracranial hemorrhage or an operation. However, the present case and that of Sasaki et al. (23) demonstrate that totally spontaneous Tegression can occur. Recent reports have examined the relationship between AVMs and aneurysms (9, 24, 27). Suzuki and Onuma reviewed the literature and found that 2.7 to 8.7% of patients with AVMs had associated aneurysms (27), a figure similar to the incidence of aneurysm in the general population determined by con- trolled autopsy studies (2, 25). Hayashi et al. presented a patient with three aneurysms on a feeding vessel of an AVM, two of which resolved without surgical therapy after excision of the AVM (9). They reviewed 73 other reported cases of aneurysm and AVM and made a strong case for an association between the two. They pointed out the importance of increased hemo- dynamic stresses imposed by the presence of an AVM and the possible effects of this on aneurysm development. The aneu- rysm in our patient was located on the MCA proximal to the vessel feeding the AVM. There was no change in the size of the aneurysm over 13 years. suggesting no causal relationship in this case. SUMMARY A case of the totally spontaneous disappearance of an AVM has been presented. Previously reported cases were reviewed. There are probably several mechanisms for the regression of AVMs. depending on the individual case and the factors involved. The rarity of spontaneous regressions and the severity of complications of AVMs (10, 19, 21) argue against conserva- tive therapy in the hope that a regression will occur. However, careful study of spontaneous regressions of AVMs may aid in understanding the natural history of AVMs and may help elucidate further modes of therapy. DISCLAIMER The opinions expressed in this article are the private views of the authors and are not to be construed as official or reflecting the views of the Department of the Army or the Department of Defense. ACKNOWLEDGMENT The authors express their thanks to Georgia Frederic for all of her help during the preparation of this manuscript. Received for publication. June 15. 1982: accepted, August 3. 1982. Reprint requests: Daniel G. Nehls. M.D.. Barrow Neurological REGRESSION OF AVMs 719 Institute, St. Joseph's Hospital and Medical Center. 350 West Thomas Road. Phoenix, Arizona 85013. REFERENCES I. Castaigne P, Buge A. Pertuiset B. LaPlane D, Corbin J: Angiome thrombosé radiologiquement exclu. Bull Mem Soc Med Hop Paris 77:923-930. 1961. Chason JL, Hindman WM: Berry aneurysms of the circle of Willis: Results of a planned autopsy study. Neurology (NY) 8:41-44, 1958. 3. Conforti P: Spontaneous disappearance of cerebral arteriovenous angioma: Case report. J Neurosurg 34:432-434, 1971. 4. Cushing H, Bailey P: Tumors Arising from the Blood- Vessels of the Brain: Angiomatous Malformations and Hemangioblastomas. Springfield. Illinois, Charles C Thomas, 1928. 5. DeLange E: Een geval van spontane thrombose van een arterio- veneus aneurysma in de hersenen, Ned Tijdschr Geneeskd 99:944— 946, 1955 6, Eisenman JL, Alekoumbides A. Pribam H: Spontaneous throm- bosis of vascular malformations of the brain. Acta Radio! [Diagn] (Stockh) 13:77-85. 1972. 7. Endo S, Koshu K, Suzuki J: Spontaneous regression of posterior fossa dural arteriovenous malformation. J Neurosurg 51:715-717. 1979, 8. Hansen JH, Segaard 1: Spontaneous regr intracranial arteriovenous malformation: C 45:338-341, 1976. 9. Hayashi S, Arimoto T, Itakura T, Fujii T, Nishiguchi T, Komai N: The association of intracranial aneurysms and arteriovenous mal- formation of the brain. J Neurosurg 55:971-975, 1981 Ho6k O, Johanson C: Intracranial arteriovenous aneurysms: A follow-up study with particular attention to their growth, Arch Neurol Psychiatry 80:39-54, 1958. ll. Kelly DL Jr, Alexander E, Davis CH, Maynard DC: Intracranial arteriovenous malformations; Clinical review and evaluation of brain scans. J Neurosurg 31:422-428, 1969. 12. Krayenbiih! H. Yasargil G: Das Hirnaneurysma. Basel, Documenta Geigy. 1958. Cited by Kushner and Alexander (ref 13). 13. Kushner J, Alexander E Jr: Partial spontaneous regressive arterio- venous malformation: Case report with angiographic evidence. J Neurosurg 32:360-366, 1970. 14, Lakke JPWF: Regression of an arteriovenous malformation of the brain. J Neurol Sci 11:489--496, 1970. 15. Levine J, Misko JC. Seres JL. Snodgrass RG: Spontaneous angio- graphic disappearance of a cerebral arteriovenous malformation: Third reported case. Arch Neurol 28:195-196, 1973 16. Mabe H, Furuse M: Spontaneous disappearance of a cerebral arteriovenous malformation in infancy: Case report. J Neurosurg 46:81 1-815, 1977. 17. Magidson MA, Weinberg PE: Spontaneous closure of a dural arteriovenous malformation, Surg Neurol 6:107-110, 1976. 18. Norlén G: Arteriovenous sms of the brain: Report of ten cases of total removal of the lesion, J Neurosurg 6:475-494, 1949. 19. Paterson JH, McKissock W: A clinical survey of intracranial angiomas with special reference to their mode of progression and surgical treatment: A report of 110 cases. Brain 79:233-266, 1956. 20. Pecker J, Sabouraud O, Javelli A: Les thromboses spontanées des anéurysmes artério-veineux cérébreux. Neurochirurgie 7:155—157. 1961, 21. Perret G, Nishioka H: Report on the Cooperative Study of Intra- cranial Aneurysms and Subarachnoid Hemorrhage: Section V1. Arteriovenous malformations: An analysis of 545 cases of cranio- cerebral arteriovenous malformations and fistulae reported to the Cooperative Study. J Neurosurg 25:467-490. 1966. 22. Pool JL. Potts DG: Aneurysms and Arteriovenous Anomalies of the Brain: Diagnosis and Treatment. New York. Hoeber Medical Di- vision, Harper and Row, 1965. 23. Sasaki U. Itagaki T. Shimizu H. Takahashi M: Almost complete regression of cerebral arteriovenous malformation: Report of a case. Neurol Surg 4:175-178, 1976. 24. Shenkin HA, Jenkins F. Kim K: Arteriovenous anomaly of the brain associated with cerebral aneurysm. J Neurosurg 34:225-228. 1971 Vv ssion of an extra- and se report. J Neurosurg 10. 780 NEHLS AND PITTMAN nN os . Stehbens WE: Pathology of the Cerebral Blood Vessels. St Louis. CV Mosby. 1972. 26. Sukoff MH, Barth B, Moran T: Spontaneous occlusion of a massive arteriovenous malformation—case report, Neuroradiology 4:121- 123. 1972. 27. Suzuki J, Onuma T: Intracranial aneurysms associated with arte- riovenous malformations. J Neurosurg 50:742-746, 1979. 28. Svien HJ. Peserico L: Regression in size of arteriovenous anomaly. J Neurosurg 17:493-496. 1960. COMMENTS Nehls and Pittman have described another case of the spon- taneous regression or disappearance of an arteriovenous mal- formation. Their review of the literature and their descriptions of the possible causes of this spontaneous regression are well presented, I believe that this work presents another true case of spontaneous regression and contributes to our appreciation of the natural history of arteriovenous malformations. I thank the authors for their publishing of this interesting case. David L. Kelly, Jr., M.D. Winston-Salem, North Carolina Although not presenting a unique case report, the authors do underscore the rarity of the spontaneous regression of Neurosurgery, Vol. 11, No. 6 arteriovenous malformation. Such cases are difficult to obtain because of the necessity of interval arteriography. In this case, we have arteriographic examinations that span the course of 13 years, a most remarkable situation in this period when arteriograms are usually destroyed or stored in places where they cannot be retrieved. I was able to review nine cases (published as part of a larger article) in which interval arteri- ography of approximately 10 years had been carried out. We had no case of total regression of a malformation, but found that approximately one-third of the malformations got some- what smaller, one-third remained the same, and one-third increased somewhat in size. So few cases are reported that no one person can document the anatomical factors of the mal- formation that might predict regression or spontaneous oblit- eration, The authors’ speculation is interesting, but must be considered theoretical. One would not want to leave the impres- sion that the natural obliteration of arteriovenous malforma- tions is common and that they should not be operated on. It is not indicated why neither of these lesions was operated on 13 years previously, when they were first discovered. Ob- viously, opinion about the prophylactic treatment of aneurysms changed during the 13 years that this patient was followed. Bennett M. Stein, M.D. New York, New York