This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Pollock, B. E. Articles by Kondziolka, D. Search for Related Content PubMed PubMed Citation Articles by Pollock, B. E. Articles by Kondziolka, D.

(Stroke. 1996;27:1-6.)
© 1996 American Heart Association, Inc.

Articles

Factors That Predict the Bleeding Risk of Cerebral Arteriovenous Malformations

Bruce E. Pollock, MD; John C. Flickinger, MD; L. Dade Lunsford, MD; David J. Bissonette, PA-C Douglas Kondziolka, MD

From the Departments of Neurological Surgery (B.E.P., J.C.F., L.D.L., D.J.B., D.K.), Radiation Oncology (J.C.F., L.D.L., D.K.), and Radiology (L.D.L.), University of Pittsburgh (Pa) Medical Center.

Correspondence to Bruce E. Pollock, MD, and reprint requests to L. Dade Lunsford, MD, FACS, Department of Neurological Surgery, Presbyterian University Hospital, 200 Lothrop St, Pittsburgh, PA 15213.

	Abstract

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Background and Purpose Arteriovenous malformations (AVMs) have an overall 2% to 4% annual risk of hemorrhage. The purpose of this study was to determine whether specific clinical and radiographic factors predispose AVMs to bleed and to predict the bleeding risk for individual AVM patients.

Methods We reviewed the clinical histories and cerebral angiograms of 315 AVM patients who underwent stereotactic radiosurgery at our center. One half of the patient data (analysis cohort) was used to determine risk factors for bleeding and to construct AVM hemorrhage risk groups. These risk groups were then tested with the second half of the patient data (test cohort).

Results The mean AVM volume was 4.0±3.4 mL (approximate maximum diameter of 2 cm). One hundred ninety-six initial hemorrhages occurred in 10 348 patient-years for an annual initial bleed rate of 1.89%; 44 of these 196 patients had a repeat bleed in 591 patient-years for an annual rebleed rate of 7.45%. The overall crude annual hemorrhage rate was 2.40%. Multivariate analysis revealed three factors associated with hemorrhage: history of a prior bleed (relative risk [RR], 9.09; 95% confidence interval [CI], 5.44 to 15.19; P<.001), a single draining vein (RR, 1.66; 95% CI, 1.13 to 2.38; P<.01), and a diffuse AVM morphology (RR, 1.64; 95% CI, 1.12 to 2.46; P<.01). Four AVM hemorrhage risk groups were constructed on the basis of the significant factors. The annual rate of bleeding was 0.99% for low-risk AVMs, 2.22% for intermediate-low–risk AVMs, 3.72% for intermediate-high–risk AVMs, and 8.94% for high-risk AVMs.

Conclusions Analysis of a large group of AVM patients who underwent stereotactic radiosurgery demonstrated that small AVMs have an annual hemorrhage risk similar to that of the general AVM population. AVM patients have a wide variability of bleeding risk that can be predicted from their clinical presentation and the angiographic characteristics of the AVM. The management of AVM patients should be based not only on the morbidity of the proposed treatment but also those factors that predispose individual patients to either a low or high hemorrhage risk.

Key Words: cerebral arteriovenous malformation • hemorrhage • natural history

	Introduction

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Cerebral arteriovenous malformations (AVMs) are congenital anomalies that consist of abnormal arteries and veins without an intervening capillary bed. Large autopsy series have shown the incidence of AVMs to be between 0.04% and 0.52%.^{1 2} Although the majority of AVMs are discovered after an intracerebral hemorrhage, others are diagnosed in patients with seizures, headaches, and progressive neurological deficits.³ Natural-history studies of untreated AVMs have demonstrated an overall 2% to 4% annual rate of hemorrhage,^{4 5 6 7 8 9 10} with a combined annual morbidity and mortality of approximately 3%.¹⁰ Risk factors believed to predispose AVM patients to hemorrhage include a history of a prior bleed,^{5 6 7 8 9} small AVM size,^{8 11 12} deep venous drainage,^{11 13 14} intranidal aneurysms,^{13 15 16} and higher feeding artery pressures.^{11 12} The relative importance of these factors is unknown.

Newly diagnosed AVM patients usually are advised to undergo surgical resection of the malformation if the anticipated morbidity of the operation is low.^{17 18 19} This management approach assumes that the risk of bleeding over a patient's lifetime is substantial and that all AVMs have a similar risk of hemorrhage. To better understand the risk factors associated with AVM hemorrhage, we retrospectively reviewed the clinical histories and radiographic studies of 315 AVM patients who underwent stereotactic radiosurgery at our center before 1992. Because the purpose of this report was to examine the natural history of untreated AVMs, the bleeding rate of these AVM patients after radiosurgery was not included in the present study.

	Subjects and Methods

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Patients
From August 1987 to January 1992, 313 patients with angiographically identifiable AVMs underwent stereotactic radiosurgery at our institution. Two patients had two AVMs, and these were treated as separate AVM patients in the data analysis for a total of 315 AVM patients. Of 315 AVMs, 274 (87%) were "small" (<3 cm in average diameter).¹⁸ All patients were prospectively entered into a database at the time of their procedure. Excluded from this analysis were patients with angiographically occult vascular malformations (cavernous malformations) and patients with dural AVMs.

Study Design
We reviewed the outpatient and hospital records of all patients to determine their clinical histories. We contacted the referring physician or the patient when further information was needed. To determine the annual hemorrhage rate for the study population before stereotactic radiosurgery, we recorded the time from each patient's birth date to all documented AVM hemorrhages (confirmed by either CT, lumbar puncture, or operative report) and to the date of radiosurgery. The time patients were at risk for an initial AVM hemorrhage was defined as the time from each patient's date of birth to either a documented bleed or until the date of radiosurgery. The time patients were at risk for rebleeding was defined as the time from a patient's initial AVM hemorrhage to either a rebleed or to the date of radiosurgery. To calculate the annual initial hemorrhage rate, the number of initial AVM hemorrhages observed was divided by the total number of patient-years that the study population was at risk for an initial hemorrhage. To calculate the annual rebleed rate, the number of repeat AVM hemorrhages observed was divided by the total number of patient-years that the study population was at risk for rebleeding.

The stereotactic angiograms and MRI studies of every patient were reviewed, and the following information for each AVM was recorded: location, hemispheric (cerebral or cerebellar lobes) versus deep (thalamus, basal ganglia, corpus callosum, or brain stem); volume (V=/6 ·X · Y · Z dimensions); morphology (compact versus diffuse) (Fig 1); proximity to a pial or ependymal surface (yes versus no); venous drainage (superficial versus deep versus both); number of draining veins (1 versus >1) (Fig 2); related aneurysms (proximal or intranidal versus none or unrelated); and presence of a varix (yes versus no).

View larger version (75K): [in this window] [in a new window]   Figure 1. Lateral stereotactic cerebral angiograms. A, Right parietal AVM with a compact nidus; B, right basal ganglia AVM with a diffuse nidus.

View larger version (83K): [in this window] [in a new window]   Figure 2. Lateral stereotactic cerebral angiograms. A, Right temporal AVM with a single draining vein (arrow); B, right temporal AVM with multiple draining veins.

Statistical Analysis
The patient population was randomly divided into two groups (A and B). Hemorrhage risk data for both groups were separated into initial hemorrhage risk data (A^I and B^I) and rebleed risk data (A^R and B^R). To prevent any bias that may have occurred in the grouping of the patients, a crossover of patients was incorporated at the time of initial AVM hemorrhage. The data set (analysis cohort) for the univariate and multivariate analyses of AVM hemorrhage risk factors consisted of the initial hemorrhage risk data from group A patients combined with the rebleed risk data from group B patients (A^I+B^R). AVM risk groups were then constructed on the basis of the results from the analysis-cohort data evaluation. The validity of the AVM risk groups was then verified on the remaining data set (test cohort) consisting of the initial hemorrhage risk data from group B patients combined with the rebleed risk data from group A patients (B^I+A^R).

Actuarial risks of hemorrhage were calculated with the Kaplan-Meier method.²⁰ Observations on all patients were censored at the time of their radiosurgical procedure. Univariate analysis of clinical and radiographic AVM factors was performed with the log-rank test,²¹ and a stepwise multivariate analysis was performed with the Cox proportional hazards model to identify independent risk factors.²² Patient age was tested as a time-dependent covariate by the proportional hazards model for both the univariate and multivariate analyses.

	Results

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Clinical and radiographic characteristics of the 315 patients are shown in Table 1. The mean patient age was 35.0±15.2 years. The mean AVM volume was 4.0±3.4 mL. Two hundred sixty-three AVM hemorrhages were observed in 10 939 patient-years before radiosurgery for a crude annual hemorrhage rate of 2.40%. One hundred fifty-two patients had a single bleeding episode. Forty-four patients had multiple episodes of bleeding: two bleeds (n=30), three bleeds (n=9), four bleeds (n=1), and five bleeds (n=4). The annual initial hemorrhage rate was 1.89% (196 hemorrhages per 10 348 patient-years) compared with the annual rebleed rate of 7.45% (44 hemorrhages in the 196 patients at risk per 591 patient-years) (P<.001).

View this table: [in this window] [in a new window]   Table 1. Clinical and Radiographic Characteristics of 315 AVM Patients

Risk Factors for AVM Hemorrhage
Initial univariate analysis of clinical and radiographic AVM factors revealed six significant risk factors for hemorrhage: history of a prior bleed (P<.001), deep location (P<.001), deep venous drainage (P<.001), increasing patient age (P<.001), diffuse AVM morphology (P<.001), and one draining vein (P=.046). Eighty-four of 97 patients (87%) with deeply located AVMs presented after a hemorrhage compared with 112 of 218 patients (51%) with hemispheric AVMs (P<.001). This was felt to represent a presentation bias rather than a true risk factor for AVM hemorrhage and was removed from further analyses. Deep venous drainage was also excluded from further analyses because it strongly correlated with AVM location; 92 of 97 patients (95%) with deeply located AVMs had some component of deep venous drainage compared with 131 of 218 patients (60%) with hemispheric AVMs (P<.001). Exclusion of location and deep venous drainage because of presentation bias was supported by subset univariate analysis of subsequent hemorrhage risk in which neither factor was significant (location, P=.40; deep venous drainage, P=.37).

The results of the univariate analysis for the remaining factors are shown in Table 2. Multivariate analysis revealed three significant risk factors associated with hemorrhage: history of a prior bleed (relative risk [RR], 9.09; 95% confidence interval [CI], 5.44 to 15.19; P<.001), a single draining vein (RR, 1.66; 95% CI, 1.13 to 2.38; P<.01), and diffuse AVM morphology (RR, 1.64; CI, 1.12 to 2.46; P<.01).

View this table: [in this window] [in a new window]   Table 2. Univariate Analysis of Arteriovenous Malformation Factors Predictive of Hemorrhage

Development and Testing of AVM Risk Groups
Four AVM risk groups were constructed on the basis of the three significant variables of the multivariate analysis. Low-risk AVMs had no history of a prior bleed, >1 draining vein, and a compact nidus. Intermediate-low–risk AVMs had no history of a prior bleed, 1 draining vein, and/or a diffuse nidus. Intermediate-high–risk AVMs had a history of a prior bleed, >1 draining vein, and a compact nidus. High-risk AVMs had a history of a prior bleed, 1 draining vein, and/or a diffuse nidus. The AVM risk groups were then included in a separate multivariate analysis with the previously tested factors. In that analysis the AVM risk group assigned to each AVM was the only variable predictive of hemorrhage (RR, 2.34; 95% CI, 1.79 to 3.06; P<.001).

The annual rates of hemorrhage when the AVM risk groups were applied to the test cohort were 1.31% for low-risk AVMs, 2.40% for both intermediate-risk AVM groups, and 8.99% for high-risk AVMs. Table 3 shows the annual rates of hemorrhage when the AVM risk groups were applied to the entire study population. Inclusion of the entire study population to determine the annual hemorrhage rates for the AVM risk groups was necessary due to the small number of intermediate-high–risk patients in the test cohort (n=13). The RR for each step of the model was 1.60 (95% CI, 1.36 to 1.87; P<.001). Fig 3 shows the actuarial hemorrhage rates for the AVM risk groups when applied to the entire study population.

View this table: [in this window] [in a new window]   Table 3. Annual Hemorrhage Rates of the AVM Risk Groups

View larger version (26K): [in this window] [in a new window]   Figure 3. Graph shows actuarial rates of hemorrhage for the study population based on the AVM risk groups.

	Discussion

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Intracranial hemorrhage represents the most devastating and potentially fatal complication of cerebral AVMs. Management strategies for AVM patients therefore should eliminate the bleeding risk with the lowest possible attendant morbidity. Historically, a conservative approach toward AVMs was advocated based on beliefs that the risks of surgical resection were prohibitive and that their natural history was benign.²³ Untreated AVMs have a widely reported annual hemorrhage rate of 2% to 4%.^{4 5 6 7 8 9 10} The cumulative lifetime bleeding risk for young patients is substantial, and observation only should be recommended rarely. Advancements in anesthetic and microsurgical techniques have resulted in substantial reductions in the operative morbidity of these lesions, and operative mortality is now less than 1% in carefully selected patients.^{17 18 19 24 25} An alternative management strategy for AVMs is stereotactic radiosurgery, which is the precise, single-session delivery of a high radiation dose to induce delayed AVM obliteration. Stereotactic radiosurgery is effective in obliterating approximately 80% of AVMs less than 3 cm in average diameter within a latency interval of 2 to 3 years.^{26 27 28 29} The major disadvantage of this approach is that patients remain at risk for hemorrhage during the latency interval until the AVM obliterates. Considerable controversy remains over the proper management of AVM patients.

Published reports on AVMs may be divided into two types: natural-history studies^{4 5 6 7 8 9 10} and descriptive studies.^{11 12 13 14 15} Natural-history studies follow a group of AVM patients over time to determine bleeding rates and patient outcomes. Descriptive studies generally are based on a surgical series and examine clinical, anatomic, and physiological factors predictive of bleeding. There are weaknesses associated with both study types. First, both types of studies are hampered by selection bias. Patients with large or critically located AVMs are poor candidates for surgical removal, and observation is generally recommended. Patients with small AVMs in surgically accessible brain regions are candidates for surgical resection, since little morbidity is associated with removal of these AVMs. Second, most natural-history studies enrolled patients before the modern era of neuroimaging.^{5 6 7 8 9 10} These studies provide little information on the angioarchitecture of the malformations. Third, the results of descriptive studies are often confounded because factors associated with a clinical presentation of hemorrhage are mistaken for factors predictive of bleeding (ie, small AVM size, deep location). Fourth, descriptive studies (including the present series) can only calculate the nonfatal hemorrhage rate, since patients with initially fatal AVM hemorrhages never receive treatment. In such studies, the overall hemorrhage rate would be underestimated by 10% to 29% (the reported mortality rate for each AVM hemorrhage).^{3 4 7 8 10} To date, no study has enabled physicians to accurately predict the bleeding risk for individual AVM patients on the basis of clinical presentation and the radiological characteristics of the malformation.

Three assumptions were made in the data analysis that may limit the findings of the present study. First, a potential selection bias was present because the study population consisted only of patients who underwent stereotactic radiosurgery. Such patients generally have AVMs <3 cm in average diameter (87% in the present series), and these AVMs are often located in critical brain locations (31% were "deep" AVMs in the present series). Published natural-history studies have contained from 17% to 30% small AVMs and from 9% to 32% "deep" AVMs.^{4 5 7 8 9} Therefore, caution should be used when applying the current results to the general AVM population. Second, it was assumed that all AVMs are congenital. This is a reasonable assumption based on large surgical and autopsy series.^{2 25} Third, it was assumed that an AVM has the same risk of hemorrhage throughout a patient's life. Although the natural history of pediatric AVMs is poorly understood, it is recognized that the majority of pediatric AVMs are discovered after a hemorrhage.^{3 7 30 31} It remains unclear whether the changes in AVM morphology that occur throughout childhood and adolescence affect the risk of hemorrhage. This important question has not been settled; therefore, our findings may be more appropriate for the adult AVM population.

The present study found that a history of prior hemorrhage was the factor most predictive of future bleeding. The annual initial hemorrhage rate was 1.89% compared with the annual rebleeding rate of 7.45%. Previously published AVM natural-history studies have reported an overall 2% to 4% annual hemorrhage rate.^{4 5 6 7 8 9 10} The rebleeding rate for the year after hemorrhage has been reported to vary between 6% and 18%.^{6 7 8 9} It then returns to an approximately 2% to 3% rate of bleeding in subsequent years. Brown et al⁴ reported 168 patients with unruptured AVMs and observed 14 patients with nonfatal AVM bleeds who did not undergo further treatment; no patient sustained a repeat bleed. The authors state that the lack of recurrent hemorrhage was likely due to the short follow-up of these patients (mean follow-up, 58 months). Ondra et al¹⁰ reported a population-based prospective study of all symptomatic AVM patients in Finland from 1942 to 1975. They found an annual hemorrhage rate of 4.0%, regardless of the clinical presentation of the patient. Despite the completeness and length of follow-up in this study (mean follow-up, 23.7 years), the frequency of new hemorrhagic events was analyzed in 5-year intervals. It is possible that a transient increase in the bleeding rate of patients with a previous hemorrhage was not detected by their statistical analysis.

Patients with small AVMs more commonly present with hemorrhage. Whether the risk of bleeding is affected by the size of the malformation remains a controversial issue.^{11 12} Small AVMs (<3 cm in diameter) may have the same annual hemorrhage rate as larger AVMs but may be less likely to cause seizures or headaches that might lead to their clinical detection. Graf et al⁸ found small AVM size to be a risk factor for initial AVM hemorrhage. However, only 12 of 71 patients with unruptured AVMs <3 cm in diameter were included in their study. Of 134 patients with a history of hemorrhage, 51 had AVMs <3 cm in diameter, and AVM size was not found to increase the risk of rebleeding. Most natural-history AVM studies have reported that small AVMs more commonly present with hemorrhage, but these studies have found no correlation between AVM size and subsequent bleeds.^{5 6 7 9} Brown et al⁴ did not find AVM size to be related to bleeding risk in a series of unruptured AVMs. Because the majority of AVMs in the present series were <3 cm in diameter (the upper size limit for optimal AVM radiosurgery), the expected annual hemorrhage rate in our series should be greater than the commonly quoted 2% to 4% rate for AVMs of all sizes if small AVMs have a higher annual hemorrhage rate.^{4 5 6 7 8 9 10} The overall hemorrhage rate in our series was 2.40%. Therefore, our results are supported by published AVM natural-history studies: small AVM size does not increase the risk of AVM bleeding.

A number of anatomic AVM characteristics have been reported to predispose AVM patients to bleeding. The present study found diffuse AVM morphology and a single draining vein to be independent angiographic predictors of hemorrhage. Miyasaka et al¹⁴ analyzed the venous drainage system as a factor in AVM hemorrhage and found three variables associated with hemorrhage: a single draining vein, impaired venous drainage, and deep venous drainage alone. Unfortunately, no multivariate analysis of these factors was performed, and the location of the AVMs in relation to their venous drainage was not specified. Kader et al¹¹ also found deep venous drainage to be an independent risk factor for AVM bleeding, but this characteristic may be confounded by its relation to AVM location (as in the present series). Deeply located AVMs are unlikely to cause seizures, and patients are usually diagnosed after an intracerebral hemorrhage.

Recent reports have studied the intravascular pressures of AVMs and found that feeding artery pressures were greater in ruptured versus unruptured AVMs.^{11 12} Spetzler et al¹² found an inverse relationship between feeding artery pressure size and AVM size, whereas Kader et al¹¹ reported that feeding artery pressure was only weakly related to the size of the lesion. It remains unclear whether the increase in feeding artery pressure is an important factor in the pathophysiology of AVM hemorrhage or the hemodynamic sequelae of the AVM rupture. Young et al³² examined AVM venous physiology and discovered that there was a direct relationship between feeding artery and draining vein pressures. Draining vein pressure was affected more by changes in the patient's central venous pressure than elevations in the mean arterial pressure. Draining vein pressure was neither predictive of AVM hemorrhage nor related to AVM size, AVM location, or the direction of the venous drainage. These results are contrary to the hypothesis that restriction of venous outflow from an AVM is the primary determinant of AVM rupture.³³ Clearly, more information is needed to determine which hemodynamic aberrations result in AVM hemorrhage.

The coexistence of AVMs and cerebral aneurysms has been documented in 6% to 19% of AVM patients.^{3 4 5 7 13 16 34 35} Aneurysms related to the AVM may occur on arteries proximal to the malformation or within the substance of the nidus. The pathogenesis of related aneurysms is believed to be either a consequence of the high flow through the AVM or due to a shared developmental abnormality that weakens the cerebral vasculature. Marks et al¹³ reviewed the cerebral angiograms of 65 patients and found that all patients with intranidal aneurysms (n=9) presented with hemorrhage. Turjman et al¹⁵ reported that 58 of 100 AVM patients studied with superselective angiography had related aneurysms. Presence of a related aneurysm and intranidal aneurysm location were found to correlate with a clinical presentation of hemorrhage. The percentage of related aneurysms (58%) reported in that series is much higher than previously reported and probably relates to the use of superselective angiography to study the AVMs.

We found related aneurysms in 24 of 315 AVM patients (8%), but their presence was not predictive of bleeding (P=.63). The discrepancy between our findings and prior reports has several possible explanations. First, we may not have detected the true number of intranidal aneurysms in our AVM patients, since our results were based on the stereotactic angiograms performed at the time of radiosurgery. This technique is unlikely to be as sensitive as superselective angiography in detailing the angioarchitecture of cerebral AVMs.¹⁵ Second, intranidal aneurysms may be pseudoaneurysms that form at the site of an AVM rupture and are repaired after the bleeding episode. Thus, their detection may relate to the timing of angiography in relation to the bleeding episode. Patients in our series were studied angiographically at a median of 6 months (mean, 16 months) after their last hemorrhage. Nonetheless, we remain concerned that the presence of an intranidal aneurysm may be a significant risk factor for AVM hemorrhage, and further investigation is warranted.

The results of this study show that AVM patients are a heterogeneous population with respect to their risk of hemorrhage. Such heterogeneity has significant clinical ramifications. AVMs discovered incidentally in elderly patients may be managed conservatively because the patient's lifetime hemorrhage risk is low. Patients with low-hemorrhage-risk AVMs should be considered ideal candidates for stereotactic radiosurgery because their risk of hemorrhage during the latency interval before AVM obliteration is less than 3%. Conversely, patients with high-hemorrhage-risk AVMs have a greater than 25% chance of bleeding during the latency interval after stereotactic radiosurgery and should undergo removal of their AVM if it is in an accessible brain location. Thus, the management of AVM patients should be based on both the bleeding risk for each individual malformation and the morbidity of the proposed treatment.

Received August 1, 1995; revision received September 29, 1995; accepted September 29, 1995.

	References

Top Abstract Introduction Subjects and Methods Results Discussion References

 

1. Courville CB. Pathology of the Central Nervous System: A Study Based Upon a Survey of Lesions Found in a Series of Forty Thousand Autopsies. 3rd ed. Mountain View, Calif: Pacific Press Publishing Association; 1950:142-152.
2. McCormick WF. Pathology of vascular malformations of the brain. In: Wilson CB, Stein BM, eds. Intracranial Arteriovenous Malformations. Baltimore, Md: Williams & Wilkins Co; 1984:44-63.
3. Perret G, Nishioka H. Report on the cooperative study of intracranial aneurysms and subarachnoid hemorrhage, section VI: arteriovenous malformations: an analysis of 545 cases of cranio-cerebral arteriovenous malformations and fistulae reported to the cooperative study. J Neurosurg. 1966;25:467-490. [Medline] [Order article via Infotrieve]
4. Brown RD, Wiebers DO, Forbes G, O'Fallon WM, Piepgras DG, Marsh WR, Maciunias RJ. The natural history of unruptured intracranial arteriovenous malformations. J Neurosurg. 1988;68:352-357. [Medline] [Order article via Infotrieve]
5. Crawford PM, West CR, Chadwick DW, Shaw MDM. Arteriovenous malformations of the brain: natural history in unoperated patients. J Neurol Neurosurg Psychiatry. 1986;49:1-10. [Abstract]
6. Forster DMC, Steiner L, Hakanson S. Arteriovenous malformations of the brain: a long-term clinical study. J Neurosurg. 1972;37:562-570. [Medline] [Order article via Infotrieve]
7. Fults D, Kelly DL Jr. Natural history of arteriovenous malformations of the brain: a clinical study. Neurosurgery. 1984;15:658-662. [Medline] [Order article via Infotrieve]
8. Graf CJ, Perret GE, Torner JC. Bleeding from cerebral arteriovenous malformations as part of their natural history. J Neurosurg. 1983;58:331-337. [Medline] [Order article via Infotrieve]
9. Itoyama Y, Uemura S, Ushio Y, Kuratsu J, Nonaka N, Wada H, Sano Y, Fukumura A, Yoshida K, Yano T. Natural course of unoperated intracranial arteriovenous malformations: study of 50 cases. J Neurosurg. 1989;71:805-809. [Medline] [Order article via Infotrieve]
10. 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. 1990;73:387-391. [Medline] [Order article via Infotrieve]
11. Kader A, Young WL, Pile-Spellman J, Mast H, Sciacca RR, Mohr JP, Stein BM. The influence of hemodynamic and anatomic factors on hemorrhage from cerebral arteriovenous malformations. Neurosurgery. 1994;34:801-808. [Medline] [Order article via Infotrieve]
12. Spetzler RF, Hargraves RW, McCormick PW, Zabramski JM, Flom RA, Zimmerman RA. Relationship of perfusion pressure and size to the risk of hemorrhage from arteriovenous malformations. J Neurosurg. 1992;76:918-923. [Medline] [Order article via Infotrieve]
13. Marks MP, Lane B, Steinberg GK, Chang PJ. Hemorrhage in intracerebral arteriovenous malformations: angiographic determinants. Radiology. 1990;176:807-813. [Abstract/Free Full Text]
14. Miyasaka Y, Yada K, Ohwada T, Kithara T, Kurata A, Irikura K. An analysis of the venous drainage system as a factor in hemorrhage from arteriovenous malformations. J Neurosurg. 1992;76:239-243. [Medline] [Order article via Infotrieve]
15. Turjman F, Massoud TF, Vinuela F, Sayre JW, Gugliemi G, Duckwiler G. Aneurysms related to cerebral arteriovenous malformations: superselective angiographic assessment in 58 patients. AJNR. 1994;15:1601-1605. [Abstract]
16. Marks MP, Lane B, Steinberg GK, Snipes GJ. Intranidal aneurysms in cerebral arteriovenous malformations: evaluation and endovascular treatment. Radiology. 1992;183:355-360. [Abstract/Free Full Text]
17. Heros RC, Korosue K, Diebold PM. Surgical excision of cerebral arteriovenous malformations: late results. Neurosurgery. 1990;26:570-578. [Medline] [Order article via Infotrieve]
18. Spetzler RF, Martin NA. A proposed grading system for arteriovenous malformations. J Neurosurg. 1986;65:476-483. [Medline] [Order article via Infotrieve]
19. Stein BM, Kader A. Intracranial arteriovenous malformations. Clin Neurosurg. 1992;39:76-113. [Medline] [Order article via Infotrieve]
20. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc. 1958;53:457-481.
21. Peto R, Pike MC, Armitage P, Breslow NE, Cox DR, Howard SV, Martel N, McPherson K, Peto J, Smith PG. Design and analysis of randomized clinical trials requiring prolonged observation of the patient, II: analysis and examples. Br J Cancer. 1977;35:1-39. [Medline] [Order article via Infotrieve]
22. Cox DR. Regression models and life-tables. J R Stat Soc [B]. 1972;34:187-220.
23. Svein HJ, McRae JA. Arteriovenous anomalies of the brain: fate of patients not having definitive surgery. J Neurosurg. 1965;23:23-28. [Medline] [Order article via Infotrieve]
24. Drake CG. Cerebral arteriovenous malformations: considerations for and experience with surgical treatment in 166 cases. Clin Neurosurg. 1979;26:145-208. [Medline] [Order article via Infotrieve]
25. Yasargil MG. Microneurosurgery, Vol IIIB: AVM of the Brain. Clinical Considerations, General and Specific Operative Techniques, Surgical Results, Nonoperated Cases, Cavernous and Venous Angiomas, Neuroanesthesia. New York, NY: Georg Thieme Verlag; 1988:370-383.
26. Colombo F, Pozza F, Chierego G, Casentini L, De Luca G, Francescon P. Linear accelerator radiosurgery of cerebral arteriovenous malformations: an update. Neurosurgery. 1994;34:14-21. [Medline] [Order article via Infotrieve]
27. Friedman WA, Bova FJ. Linear accelerator radiosurgery for arteriovenous malformations. J Neurosurg. 1992;77:832-841. [Medline] [Order article via Infotrieve]
28. 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. 1983;309:269-274. [Abstract]
29. Lunsford LD, Kondziolka D, Flickinger JC, Bissonette DJ, Jungreis CA, Maitz AM, Horton JA, Coffey RJ. Stereotactic radiosurgery for arteriovenous malformations of the brain. J Neurosurg. 1991;75:512-524. [Medline] [Order article via Infotrieve]
30. Celli P, Ferrante L, Palma L, Cavedon G. Cerebral arteriovenous malformations in children: clinical features and outcome of treatment in children and adults. Surg Neurol. 1984;22:43-49. [Medline] [Order article via Infotrieve]
31. Kondziolka D, Humphreys RP, Hoffman HJ, Hendrick EB, Drake JM. Arteriovenous malformations of the brain in children: a forty-year experience. Can J Neurol Sci. 1992;19:40-45. [Medline] [Order article via Infotrieve]
32. Young WL, Kader A, Pile-Spellman J, Ornstein E, Stein BM. Arteriovenous malformation draining vein physiology and determinants of transnidal pressure gradients. Neurosurgery. 1994;35:389-396. [Medline] [Order article via Infotrieve]
33. Wilson CB, Hieshima G. Occlusive hyperemia: a new way to think about an old problem. J Neurosurg. 1993;78:165-166. [Medline] [Order article via Infotrieve]
34. Batjer H, Suss RA, Samson D. Intracranial arteriovenous malformations associated with aneurysms. Neurosurgery. 1986;18:29-35. [Medline] [Order article via Infotrieve]
35. Kondziolka D, Nixon BJ, Lasjaunias P, Tucker WS, Terbrugge K, Spiegel SM. Cerebral arteriovenous malformations with associated aneurysms: hemodynamic and therapeutic considerations. Can J Neurol Sci. 1988;15:130-134.[Medline] [Order article via Infotrieve]

This article has been cited by other articles:

				M. V. Jayaraman, M. L. Marcellus, H. M. Do, S. D. Chang, J. K. Rosenberg, G. K. Steinberg, and M. P. Marks Hemorrhage Rate in Patients With Spetzler-Martin Grades IV and V Arteriovenous Malformations: Is Treatment Justified? Stroke, February 1, 2007; 38(2): 325 - 329. [Abstract] [Full Text] [PDF]

				C. Stapf, H. Mast, R. R. Sciacca, J. H. Choi, A. V. Khaw, E. S. Connolly, J. Pile-Spellman, and J. P. Mohr Predictors of hemorrhage in patients with untreated brain arteriovenous malformation Neurology, May 9, 2006; 66(9): 1350 - 1355. [Abstract] [Full Text] [PDF]

				J.P. Mohr Brain Arteriovenous Malformations: Children and Adults Stroke, October 1, 2005; 36(10): 2060 - 2061. [Full Text] [PDF]

				H. J. Fullerton, A. S. Achrol, S. C. Johnston, C. E. McCulloch, R. T. Higashida, M. T. Lawton, S. Sidney, W. L. Young, and For the UCSF BAVM Study Project Long-Term Hemorrhage Risk in Children Versus Adults With Brain Arteriovenous Malformations Stroke, October 1, 2005; 36(10): 2099 - 2104. [Abstract] [Full Text] [PDF]

				S Renowden Interventional neuroradiology J. Neurol. Neurosurg. Psychiatry, September 1, 2005; 76(suppl_3): iii48 - iii63. [Full Text] [PDF]

				T. Todaka, J.-i. Hamada, Y. Kai, M. Morioka, and Y. Ushio Analysis of Mean Transit Time of Contrast Medium in Ruptured and Unruptured Arteriovenous Malformations: A Digital Subtraction Angiographic Study Stroke, October 1, 2003; 34(10): 2410 - 2414. [Abstract] [Full Text] [PDF]

				M. J. Gounis, B. B. Lieber, A. K. Wakhloo, R. Siekmann, and L.N. Hopkins Effect of Glacial Acetic Acid and Ethiodized Oil Concentration on Embolization with N-Butyl 2-Cyanoacrylate: An in Vivo Investigation AJNR Am. J. Neuroradiol., June 1, 2002; 23(6): 938 - 944. [Abstract] [Full Text] [PDF]

				M. A. Stefani, P. J. Porter, K. G. terBrugge, W. Montanera, R. A. Willinsky, and M. C. Wallace Large and Deep Brain Arteriovenous Malformations Are Associated With Risk of Future Hemorrhage Stroke, May 1, 2002; 33(5): 1220 - 1224. [Abstract] [Full Text] [PDF]

				M. A. Stefani, P. J. Porter, K. G. terBrugge, W. Montanera, R. A. Willinsky, and M. C. Wallace Angioarchitectural Factors Present in Brain Arteriovenous Malformations Associated With Hemorrhagic Presentation Stroke, April 1, 2002; 33(4): 920 - 924. [Abstract] [Full Text] [PDF]

				A. X. Halim, V. Singh, S. C. Johnston, R. T. Higashida, C. F. Dowd, V. V. Halbach, M. T. Lawton, D. R. Gress, C. E. McCulloch, and W. L. Young Characteristics of Brain Arteriovenous Malformations With Coexisting Aneurysms: A Comparison of Two Referral Centers Stroke, March 1, 2002; 33(3): 675 - 679. [Abstract] [Full Text] [PDF]

				Reporting Terminology for Brain Arteriovenous Malformation Clinical and Radiographic Features for Use in Clinical Trials Stroke, June 1, 2001; 32(6): 1430 - 1442. [Abstract] [Full Text] [PDF]

				C. S. Ogilvy, P. E. Stieg, I. Awad, R. D. Brown Jr, D. Kondziolka, R. Rosenwasser, W. L. Young, and G. Hademenos Recommendations for the Management of Intracranial Arteriovenous Malformations : A Statement for Healthcare Professionals From a Special Writing Group of the Stroke Council, American Stroke Association Stroke, June 1, 2001; 32(6): 1458 - 1471. [Full Text] [PDF]

				C. S. Ogilvy, P. E. Stieg, I. Awad, R. D. Brown Jr, D. Kondziolka, R. Rosenwasser, W. L. Young, and G. Hademenos Recommendations for the Management of Intracranial Arteriovenous Malformations : A Statement for Healthcare Professionals From a Special Writing Group of the Stroke Council, American Stroke Association Circulation, May 29, 2001; 103(21): 2644 - 2657. [Full Text] [PDF]

				M. BRADA and N. KITCHEN How effective is radiosurgery for arteriovenous malformations? J. Neurol. Neurosurg. Psychiatry, May 1, 2000; 68(5): 548 - 549. [Full Text]

				K. Tsuchiya, S. Katase, A. Yoshino, and J. Hachiya MR Digital Subtraction Angiography of Cerebral Arteriovenous Malformations AJNR Am. J. Neuroradiol., April 1, 2000; 21(4): 707 - 711. [Abstract] [Full Text]

				The Arteriovenous Malformation Study Group Arteriovenous Malformations of the Brain in Adults N. Engl. J. Med., June 10, 1999; 340(23): 1812 - 1818. [Full Text] [PDF]

				L. B. Morgenstern, R. F. Frankowski, P. Shedden, W. Pasteur, and J. C. Grotta Surgical treatment for intracerebral hemorrhage (STICH): A single-center, randomized clinical trial Neurology, November 1, 1998; 51(5): 1359 - 1363. [Abstract] [Full Text] [PDF]

				G. E. Swan, C. DeCarli, B. L. Miller, T. Reed, P. A. Wolf, L. M. Jack, and D. Carmelli Association of midlife blood pressure to late-life cognitive decline and brain morphology Neurology, October 1, 1998; 51(4): 986 - 993. [Abstract] [Full Text] [PDF]

				M. Diomedi, F. Placidi, L. M. Cupini, G. Bernardi, and M. Silvestrini Cerebral hemodynamic changes in sleep apnea syndrome and effect of continuous positive airway pressure treatment Neurology, October 1, 1998; 51(4): 1051 - 1056. [Abstract] [Full Text] [PDF]

				R. G. Hart, D. G. Sherman, J. D. Easton, and J. A. Cairns Prevention of stroke in patients with nonvalvular atrial fibrillation Neurology, September 1, 1998; 51(3): 674 - 681. [Abstract] [Full Text] [PDF]

				S. M. Greenberg Cerebral amyloid angiopathy: Prospects for clinical diagnosis and treatment Neurology, September 1, 1998; 51(3): 690 - 694. [Abstract] [Full Text] [PDF]

				M. O. McCarron and J. A.R. Nicoll Recurrent hemorrhage in cerebral amyloid angiopathy Neurology, September 1, 1998; 51(3): 924 - 924. [Full Text] [PDF]

				J-Ph. Neau and R. Gil Recurrent hemorrhage in cerebral amyloid angiopathy Neurology, September 1, 1998; 51(3): 924 - 925. [Full Text] [PDF]

				J. van Gijn Leukoaraiosis and vascular dementia Neurology, September 1, 1998; 51(3_Suppl_3): S3 - S8. [Abstract] [Full Text] [PDF]

				D. B. Matchar The value of stroke prevention and treatment Neurology, September 1, 1998; 51(3_Suppl_3): S31 - S35. [Abstract] [Full Text] [PDF]

M. J. Alberts tPA in acute ischemic str