J Nearosurg 73:859-863, 1990

Unruptured intracranial aneurysms and arteriovenous malformations: frequency of intracranial hemorrhage and relationship of lesions ROBERT D. BROWN, JR., M.D., DAVID O. WIEBERS, M.D., AND GLENN S. FORBES, M.D.

Departments of Neurology, Health Sciences Research, and Diagnostic Radiology, Mayo Clinic and Mayo Foundation, Rochester, Minnesota u," Among 91 patients with unruptured intracranial arteriovenous malformations (AVM's), 16 patients had 26 unruptured intracranial saccular aneurysms. An actuarial analysis showed the risk of intracranial hemorrhage among patients with coexisting aneurysm and AVM to be 7% per year at 5 years following diagnosis compared to 1.7% for patients with AVM alone. The difference in length of survival free of hemorrhage was significant (log-rank, p < 0.0007). Several angiographic and clinical parameters were investigated to better understand the relationship of these lesions. The aneurysms occurred in similar percentages in patients with small, medium, and large AVM's. Twenty-five aneurysms were on arteries feeding the malformation system, almost equally distributed proximally and distally. Eleven aneurysms were atypical in location, and all arose from primary or secondary branch feeders to the malformation; 24 were on enlarged feeding arteries. Eleven (16%) of the 67 patients with high-flow AVM's had associated aneurysms, compared with five (21%) of the 24 patients with low-flow AVM's. Four (16%) of 25 low-shunt malformations and 12 (18%) of 65 high-shunt malformations had associated aneurysms. All five aneurysms associated with low-shunt malformations were on a direct arterial feeder of the malformation. These data suggest that the intracranial AVM's predispose to aneurysm formation within AVM feeding systems and that the mechanism is not simply based upon the high blood flow or high arteriovenous shunt in these systems. KEY WORDS

"

arteriovenous malformation

S

INCE the coexistence of intracranial arteriovenous malformation (AVM) and saccular aneurysm was first described in 1942, 27 numerous similar cases have appeared in the literature. 1-3"5"9.1~ These series have all included patients diagnosed after intracranial hemorrhage, with a combination of ruptured and unruptured lesions visualized on cerebral angiography. The frequency of aneurysms associated with AVM's has ranged from 3 . 7 % 19 t o 8 . 7 % 5 of patients with AVM. Predilection for these aneurysms to locate on AVM feeding arteries5'~8"2~ has led to the theory that increased blood flow 19'26is a major factor in formation of the aneurysms. Other factors that have been implicated in the pathogenesis include decreased resistance in the AVM feeding arteries 12 and increased hemodynamic turbulence in these arteries. 23 However, the contribution of these factors remains unknown. We describe a series of patients with AVM's and coexisting aneurysms diagnosed prior to clinically evident intracranial hemorrhage. The purpose of this study

J. Neurosurg. / Volume 73/December, 1990

9 aneurysm

9 subarachnoid hemorrhage

was to better define the relationship of these two lesions and to add to existing knowledge about the pathogenesis of intracranial saccular aneurysms.

Clinical Material and Methods

Clinical Material The records of all patients seen at the Mayo Clinic between 1974 and 1985 with the diagnosis of intracranial AVM were reviewed. Patients with clinically evident intracranial hemorrhage before diagnosis of AVM and those treated surgically were excluded, as were those with purely extracerebral dural AVM's and aneurysms of the vein of Galen.

Classification of A VM's The cerebral angiograms in all of the cases identified were reviewed by a Mayo Clinic neuroradiologist and documentation of all intracranial AVM's and aneurysms was confirmed, The presence or absence of aneurysm was based entirely upon angiograms performed 859

R. D. Brown, Jr., D. O. Wiebers, and G. S. Forbes 100 c-

8o

~

oo ~

,'

[

'-'---'

arate spherical structures protruding from the aneurysm wall, in contradistinction to bilobed aneurysms which contained two lobes arising from a common stalk.

AVM alone n ~ 75

I I

AVM with aneurysm n=16

40

I

'~

20 0

I

1

I

I

I

I

I

1

I

2

3

4

5

6

7

8

Years after diagnosis of AVM/aneurysm

FIG. 1. Actuarial analysis of survival free of intracranial hemorrhage, comparing patients with arteriovenous malformations (AVM) alone (75 cases) and patients with AVM coexisting with saccular aneurysms ( 16 cases).

at the time of AVM diagnosis. All patients in the study had been examined by a neurologist at the Mayo Clinic. The AVM characteristics studied and their method of classification have been described previously. 4 Based on the maximum diameter (with correction for angiographic magnification), AVM's were categorized as "small" if the diameter was less than 3.5 cm, "medium" if 3.5 to 5.4 cm, and "large" if greater than 5.5 cm. The degree of arteriovenous shunt was determined by the length of time required for venous filling to occur after arterial filling. "High-shunt" AVM's were defined as those filling within 2 seconds and included those with simultaneous filling. Those with a time to venous filling after arterial filling of more than 2 seconds were considered "low-shunt" lesions. The AVM flow was broadly categorized as either high or low by the neuroradiologist after considering the number and size of the arterial feeders along with the AVM size and shunt characteristics. The relative enlargement of the arterial feeder on which the aneurysm originated was obtained by comparing its measured diameter in millimeters to a "normal" diameter. The normal value was the diameter of the same vessel on the opposite side (if it was not also an AVM feeder) or that seen on a normal angiogram performed at the Mayo Clinic on an age-matched person. The arteriographic classification of the artery harboring the intracranial aneurysm was as follows: Type 1, no involvement in AVM supply; Type 2, involved in AVM supply but removed by two or more branching points; Type 3, proximally removed from the AVM by a single branching site; and Type 4, direct feeder to the AVM. The aneurysm site was designated "typical" or "atypical" based on the usual locations of saccular aneurysms unassociated with AVM at intracranial bifurcations in and around the circle of Willis. Typical sites included the middle cerebral artery trifurcation, basilar artery caput, ophthalmic artery origin, posterior and anterior communicating artery origins, and intracavernous/suprasellar segments of the internal carotid artery. Aneurysmal daughter sacs were defined as sep860

Follow-Up Methods Follow-up information was obtained from medical records, by letter, and by telephone conversation with the patient, a close relative, or the patient's physician. In all patients, follow-up monitoring continued until death, interventional treatment, or for a minimum of 4 years after diagnosis. Statistical Analysis The Kaplan-Meier product limit method was used for estimating the probability of length of survival free of intracranial hemorrhage during the follow-up period. Log-rank tests were used to compare survival curves. Results

Clinical Aspects Among the 91 patients with intracranial AVM's and complete sets of angiograms available for review, 16 patients (18 %) were identified as having 26 unruptured intracranial saccular aneurysms (eight patients had two to four aneurysms each). There were 11 women and five men. The mean age at diagnosis was 38 years (range 17 to 63 years) compared with 37.7 years in the overall AVM group. 4 Symptoms secondary to the AVM were noted in 14 patients at presentation. The most common symptom was seizure (nine patients), followed by focal ischemia believed to be related to the AVM (three patients) and headache (three patients). No signs or symptoms referable to the intracranial aneurysm were present. I ntracranial hemorrhage occurred in six (38 %) of the 16 patients with aneurysms and AVM's (mean followup period 4.5 years), compared with 13 (17%) of 75 patients who had an AVM without aneurysm (mean follow-up period 7.1 years). Among the six patients with rupture, the source of hemorrhage was the AVM in one patient and the aneurysm in another; the source was uncertain in four. Excision of the AVM ended the follow-up period in six of the 16 patients. We estimated the risk of intracranial hemorrhage among the 16 patients with coexisting AVM and aneurysm to be 7% at 1 year compared with 3% among those with AVM alone (Fig. 1). At 5 years, the risk persisted at 7% per year, while it decreased to 1.7% per year in those with an AVM unassociated with aneurysm. The difference in length of survival free of intracranial hemorrhage was significant (log-rank, p < 0.0007). Of the 16 patients with AVM's and aneurysms, three remained free of intracranial hemorrhage at 8 years following diagnosis, with hemorrhage in one patient occurring beyond that time. By comparison, in the 75 patients with AVM alone, 12 hemorrhages occurred after 10 years of follow-up monitoring and six after 15 years. 14 J. Neurosurg. / Volume 73/December, 1990

Intracranial arteriovenous malformations TABLE 1 Type of artery and associated aneurysm in 16 patients (26 aneurysms) Type of Artery

No. of Aneurysms

No. of Aneurysms with AVM (by size of AVM)* Large

Medium

Small

4 3 2

10 8 7

4 2 3

2 5 4

4 1 0

1

1

l

0

0

* Size of arteriovenous malformation (AVM) is based on maximum diameter: large >_5.5 cm, medium 3.5 to 5.4 cm, and small < 3.5 cm. For type of artery see text.

Radiological Aspects o f the A VM's The sizes of the A V M ' s associated with aneurysms were c o m p a r e d with the sizes in the overall group o f A V M ' s . The percentage o f A V M ' s coexisting with aneurysms was similar to the overall frequency o f A V M ' s within the three A V M size classifications. O f the 16 patients with coexisting A V M ' s a n d aneurysms, four (25%) had aneurysms in association with a small A V M , seven (44%) with a m e d i u m - s i z e d A V M , a n d five (31%) with a large A V M . Eleven o f these 16 patients h a d a high-flow A V M with venous filling within 2 seconds (high shunt), four had a low-flow A V M with venous filling of m o r e than 2 seconds after arterial filling (low shunt), a n d one had a low-flow/high-shunt lesion. The shunt a n d flow characteristics were similar to those in the overall A V M group. All a n e u r y s m s associated with low-flow A V M ' s were located on a distal p o r t i o n o f the arterial feeding system; four arteries were classified as T y p e 4 a n d one as T y p e 3.

Radiotogical Aspects o f the Aneurysms The m e a n d i a m e t e r o f the a n e u r y s m s was 7.2 m m (range 2 to 18 m m ) . One a n e u r y s m had a daughter sac a n d a n o t h e r was bilobed. The probable presence o f a daughter sac was noted in two others, although the angiographic views available precluded a definite determination. O f the 26 aneurysms, 25 (96%) were located on an A V M arterial feeder a n d 18 (69%) were r e m o v e d from the A V M by no m o r e t h a n one arterial b r a n c h i n g site (Type 3 or 4 artery) (Table 1). The arterial site o f a n e u r y s m d e v e l o p m e n t was categorized as typical or atypical. O f the 26 aneurysms, 11 were atypical a n d 15 were typical in location. T h e sizes o f the aneurysms at the two sites were similar, with a m e a n d i a m e t e r o f 7.2 m m in b o t h groups. All o f the a n e u r y s m s in atypical sites were on a T y p e 3 or 4 feeding artery, c o m p a r e d with seven (47%) o f 15 lesions in typical locations. All 15 lesions at typical sites were associated with either m e d i u m or large A V M ' s ; 10 (91%) o f the 11 aneurysms at u n c o m m o n sites were associated with small or m e d i u m - s i z e d A V M ' s . F r o m a h e m o d y n a m i c standpoint, all a n e u r y s m s in

J. Neurosurg. / Volume 73/December, 1990

TABLE 2 Distribution of aneurysms by size of artery of origin and typicality of location* Arterial Size Ratio

No. of Aneurysms

AVM Location Typical

Atypical

normal 2 1 1 2:1 II 5 6 3:1 6 4 2 4:1 0 0 0 5:1 4 2 2 indeterminate 3 3 0 * Arterial size ratio is the ratio of the diameter of the artery on which the aneurysm is located to the normal diameter of that artery. AVM = arteriovenous malformation.

typical locations were associated with high-flow/highshunt A V M ' s . In contrast, those at atypical locations were seen in association with both high-flow (four aneurysms) and low-flow (seven) lesions, a n d high-shunt (five) and low-shunt (six) lesions. The two types o f a n e u r y s m location occurred with relatively similar frequency on a range o f arterial size ratios (Table 2), T h e ratio o f the d i a m e t e r o f the artery on which the a n e u r y s m was located to the n o r m a l d i a m e t e r o f that artery was determined, a n d the n u m b e r o f a n e u r y s m s associated with each size ratio was n o t e d (Table 2). A m o n g the a n e u r y s m s for which the relative p a r e n t artery size could be determined, 21 (91%) f o r m e d on a dilated artery. A n e u r y s m s were classified according to whether they were located at arterial bifurcations, a n d the relationship o f this factor to a n e u r y s m a n d A V M size was considered (Table 3). It was not possible to classify five aneurysms because the arteriographic views o f the aneurysmal neck were inadequate. O f the 15 a n e u r y s m s located at a bifurcation, nine were associated with a m e d i u m - s i z e d A V M a n d six with a large A V M ; n o n e was found with a small A V M . Discussion

A n actuarial analysis revealed the risk o f intracranial hemorrhage a m o n g patients with a coexisting saccular

TABLE 3 Relationship of aneurysm location to A VM size and aneurysm size * Location at Arterial Bifurcation

No. of AVM's (by size) Mean No. of Aneurysm Aneurysms Large Medium Small Diameter (mm) yes 15 6 9 0 8.4 no 6 3 l 2 4.5 indeterminate 5 2 0 3 6.6 * Sizeofarteriovenous malformation (AVM) is based on maximum diameter: large >_5.5 era, medium 3,5 to 5.4 cm, and small < 3.5 cm. 861

R. D. Brown, Jr., D. O. Wiebers, and G. S. Forbes aneurysm and unruptured AVM to be 7% per year at 5 years following diagnosis, compared to 1.7% per year for the group of patients with an AVM occurring alone. Although the number of cases in this series is relatively small, the difference in length of survival free of intracranial hemorrhage was significant (log-rank, p < 0.0007). The higher risk of hemorrhage in the group with coexisting aneurysms and AVM's compared to the group with AVM alone most likely reflects rupture from a considerable number of coexisting aneurysms. It is also possible that the same factors predisposing to aneurysm formation may increase the likelihood among patients with AVM's that unruptured aneurysms will enlarge and rupture. Three theories have been proposed for the association of intracranial AVM's and saccular aneurysms: I) purely coincidental occurrence; 3 2) two types of congenital lesions occurring in the setting of a more generalized vascular maldevelopment; 1'15 and 3) increased blood flow to the AVM leading to aneurysm development. ~9Over the past two decades, the third theory has become the most popular for several reasons, including a proposed simple increase in blood flow in certain AVM-feeding arteries on which aneurysm formation took place, 5"9"12"14"15"24"25 increased occurrence on the feeding artery indicating some type of hemodynamic stress, 5'14'15'~8 and resolution of the aneurysm after removal of the AVM. 9'12'24 Lending further support for the formation of aneurysms in areas of abnormal hemodynamics has been the association of aneurysms with moyamoya disease. 1~.22The current study is unique in that it is based entirely upon clinically unruptured AVM's and aneurysms. The potential changes in appearance following rupture 6'~3makes the study of these lesions prior to rupture more suitable for answering questions regarding the pathogenesis of aneurysm formation in conjunction with an AVM. This study suggests that aneurysms occur with increased frequency in patients with an intracranial AVM of any size. The vessel harboring the aneurysm usually is an enlarged AVM arterial feeder. Many aneurysms associated with AVM's occur at atypical sites and may be unassociated with an arterial bifurcation. Flow and shunt characteristics of AVM's with and without associated aneurysms were similar, with both low and high flows and low and high shunts noted. The relative sizes of the aneurysmal parent arteries closely approximated those of the overall group of AVM feeders with no linear relationship between the size of the feeding artery and the presence of an aneurysm. It follows that hemodynamic abnormalities in the arteries leading to an AVM may predispose to aneurysm formation. Although the nature of the abnormality is unclear, the current information suggests that the mechanism is not solely based on high-flow or high-shunt characteristics. Other factors to consider include those contributing to and resulting from enlargement of the feeding arteries. The tendency for higher flow velocities to occur with larger AVM feeding arteries has been 862

documented during intraoperative measurement. 16 Subsequently, this increased flow velocity may reach the velocity for liquid flow at which turbulence occurs. 7 The presence of turbulence in AVM feeders has been described.17 The flow velocity also affects hemodynamic stress, with past demonstration of increasing hydrostatic pressure at arterial apices with a rise in flow velocity? The pathogenesis of aneurysm formation in association with large, medium, and small AVM's may be different. Aneurysms found with medium or large AVM's formed at both typical and atypical sites, and most commonly developed at an arterial bifurcation. All aneurysms with small AVM's were located at atypical sites and none formed at an arterial bifurcation. Furthermore, all aneurysms in typical locations occurred with high-flow, high-shunt AVM's, whereas those atypical in location were associated with a range of flow and shunt characteristics. It is possible that hemodynamic factors may be much more important in causing aneurysms in typical locations (and saccular aneurysms unassociated with AVM's) and less so in the genesis of those in atypical locations. These lesions in atypical sites may form as a result of other factors, such as a more generalized defect in vascular tissue, ultimately manifesting near the AVM due to another AVM-induced change. Even among patients with aneurysms in atypical sites, the theory of the aneurysm relating to a generalized vascular maldevelopment alone seems untenable considering the locations of these aneurysms. References

1. Anderson RM, Blackwood W: The association of arteriovenous angioma and saccular aneurysm of the arteries of the brain. J Pathol Bacteriol 77:101-110, 1959 2. Batjer H, Suss RA, Samson D: Intracranial arteriovenous malformations associated with aneurysms. Neurosurgery 18:29-35, 1986 3. Boyd-Wilson JS: The association of cerebral angiomas with intracranial aneurysms. J Neurol Neurosurg Psychiatry 22:218-223, 1959 4. Brown RD Jr, Wiebcrs DO, Forbes GS, et al: The natural history of unruptured intracranial arteriovenous malformations. J Neurosurg 68:352-357, 1988 5. Cronqvist S, Troupp H: Intracranial arteriovenous malformation and arterial aneurysm in the same patient. Acta Neurol Scand 42:307-316, 1966 6. du Boulay GH: Some observations on the natural history of intracranial aneurysms. Br J Radiol 38:721-757, 1965 7. Ferguson GG: Turbulence in human intracranial saccular aneurysms. J Neurosurg 33:485-497, 1970 8. Forbus WD: On the origin of miliary aneurysms of the superficial cerebral arteries. Bull Johns Hopkins Hosp 47: 239-284, 1930 9. Hayashi S, Arimoto T, Itakura T, et al: The association of intracranial aneurysms and arteriovenous malformation of the brain. Case report. J Neurosurg 55:971-975, 1981 10. Higashi K, Hatano M, Yamashita T, et al: Coexistence of posterior inferior cerebellar artery aneurysm and arterioJ. Neurosurg. / Volume 73/December, 1990

Intracranial arteriovenous malformations

11. 12. 13.

14. 15. 16. 17.

18.

19.

20.

venous malformation fed by the same artery. Surg Neurol 12:405-408, 1979 Konishi Y, Kadowaki C, Hara M, et al: Aneurysms associated with moyamoya disease. Neurosurgery 16: 484-491, 1985 Koulouris S, Rizzoli HV: Coexisting intracranial aneurysm and arteriovenous malformation: case report. Neurosurgery 8:219-222, 1981 London D, Enzmann D: The changing angiographic appearance of an arteriovenous malformation after subarachnoid hemorrhage. Neuroradiology 21:281-284, 198l Miyasaka K, Wolpert SM, Prager R J: The association of cerebral aneurysms, infundibula, and intracranial arteriovenous malformations. Stroke 13:196-203, 1982 Nehls DG, Carter LP: Multiple unusual aneurysms and arteriovenous malformations in a single patient: a case report. Neurosurgery 17:97-100, 1985 Nornes H, Grip A: Hemodynamic aspects of cerebral arteriovenous malformations. J Neurosurg 53:456-464, 1980 Nornes H, Grip A, Wikeby P: Intraoperative evaluation of cerebral hemodynamics using directional Doppler technique. Part 1: Arteriovenous malformations. J Neurosurg 50:145-151, 1979 Okamoto S, Handa H, Hashimoto N: Location of intracranial aneurysms associated with cerebral arteriovenous malformation: statistical analysis. Surg Neurol 22: 335-340, 1984 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 Perret G, Nishioka H: Report on the Cooperative Study of Intracranial Aneurysms and Subarachnoid Hemor-

J. Neurosurg. / Volume 73/December, 1990

21.

22. 23. 24. 25. 26. 27.

rhage. 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 Santucci N, Gazzeri G, Tamorri M: Association of two saccular aneurysms of the posterior inferior cerebellar artery with a cerebellar arteriovenous malformation fed by the same artery. Case report. J Neurosurg Sci 29: 109-112, 1985 Satoh T, Tamamoto Y, Asari S, et al: Disappearance and development of cerebral aneurysms in moyamoya disease. Case report. J Neurosurg 58:949-953, 1983 Sekhar LN, Heros RC: Origin, growth, and rupture of saccular aneurysms: a review. Neurosurgery 8:248-260, 1981 Shenkin HA, Jenkins F, Kim K: Arteriovenous anomaly of the brain associated with cerebral aneurysm. Case report. J Neurosurg 34:225-228, 1971 Shenkin HA, Spitz EB, Grant FC, et al: Physiologic studies of arteriovenous anomalies of the brain. J Neurosurg 5:165-172, 1948 Suzuki J, Onuma T: Intracranial aneurysms associated with arteriovenous malformations. J Neurosurg 50: 742-746, 1979 Walsh FB, King AB: Ocular signs of intracranial saccular aneurysms. Experimental work on collateral circulation through the ophthalmic artery. Arch Ophthalmol 27: 1-33, 1942

Manuscript received February 6, 1990, Accepted in final form June 13, 1990. Address reprint requests to: David O. Wiebers, M.D., Mayo Clinic, 200 First Street SW, Rochester, Minnesota 55905.

863

Unruptured intracranial aneurysms and arteriovenous malformations: frequency of intracranial hemorrhage and relationship of lesions.

Among 91 patients with unruptured intracranial arteriovenous malformations (AVM's), 16 patients had 26 unruptured intracranial saccular aneurysms. An ...
946KB Sizes 0 Downloads 0 Views