Brainstem arteriovenous malformations (BSAVMs) are uncommon. The recommendations in this chapter have been derived from an analysis of small cohort studies of BSAVM shaped by outcomes reported from the larger pool of AVMs in other locations. Clinical, demographic, angioarchitectural, and anatomical considerations influence these management recommendations. BSAVMs most commonly present with hemorrhage. However, there is no evidence to suggest that further hemorrhage will occur at a higher rate than AVM in other locations. Therefore, a starting point in a decision-making algorithm commences with mode of presentation (further hemorrhage of 30% over the next 10 years after diagnosis) and the age of the patient (i.e., good-quality life-years that may be threatened by the BSAVM). An estimate of unfavorable outcome from treatment is next estimated. Greater evidence exists for successful management of BSAVM by surgical or radiosurgical techniques rather than endovascular techniques. The features important in predicting an unfavorable outcome differ for surgery and radiosurgery. For radiosurgery, the volume of BSAVM is critical in predicting an unfavorable outcome. The likelihood of an unfavorable outcome from surgery is influenced by features of BSAVM from both more cephalad and caudally based AVMs. From cephaladly located brain AVM, the features that constitute the Spetzler–Martin grades (with the addition of the Lawton–Young modification for Spetzler–Martin grade III AVM) contribute to the analysis, and from spinal cord AVM, location on the surface and supply (favorable for surgery) from paramedian arteries (unfavorable for surgery) contribute to the decision. Surgery requires competence in a number of approaches. These approaches are discussed in this chapter.
Introduction
Brainstem arteriovenous malformations (BSAVMs) are annually detected in no more than 1 per 1,000,000 population and thus evidence for their management is based upon either very small cohort series or extrapolation from more generalized knowledge of brain arteriovenous malformations (bAVM) behavior and management. Therefore, this chapter must be regarded as an interpretation of limited evidence as well as my limited experience. When there is limited experience, interinstitutional application of recommendations may only result in similar outcomes if the context of bAVM management is similar. As such, it is important for the readership to understand the context on which the author derives his or her experience. To provide the context from which the author has derived his recommendations, ▶Table 56.1details the author′s limited experience for BSAVM.
Table 56.1 Cases managed
Factors
All
Operate
Not operate
P comparing operate versus not operate
Complication of microsurgery leading to mRS > 1
Complication of microsurgery leading to mRS > 2
P comparing complications leading to mRS > 1
Residual or recurrence after surgery
Total
32
23
9
5
3
2
Length of follow-up, mo (median, range) (post-op for surgery and postreferral for nonoperated)
10, 0–265
23, 0.3–145
Female
9
7
2
2
0
0
Age, y (SD)
37 (17.7)
40 (18.4)
29 (12.9)
< 0.01
Hemorrhage
24
20
4
4
3
2
Nonhemorrhagic neurological deficits
6
2
4
1
0
0
Maximum diameter, cm (SD)
2.2 (1.1)
1.8 (0.7)
3.1 (1.4)
< 0.01
Midbrain ventral
1
1
0
0
0
0
Midbrain lateral
4
3
1
2
2
2
Midbrain dorsal
14
12
2
0
0
0
Midbrain extensive
1
0
1
0
0
0
Pons ventral
5
2
3
2
0
0
Pons lateral
1
0
1
0
0
0
Pons dorsal
1
1
0
1
1
0
Pons extensive
1
0
1
0
0
0
Medulla ventral
2
2
0
0
0
0
Medulla lateral
1
1
0
0
0
0
Medulla dorsal
1
1
0
0
0
0
Medulla extensive
0
0
0
0
0
0
Spetzler–Ponce class A
3
3
0
< 0.01
0
0
0.5
0
Spetzler–Ponce class B
20
18
2
4
2
1
Spetzler–Ponce class C
9
2
7
1
1
1
mRS < 2 at referral and mRS < 2 at 12 mo
11
7
4
0
0
0
mRS < 2 at referral and mRS = 2 at 12 mo
1
1
0
1
0
0
mRS < 2 at referral and mRS > 2 at 12 mo
1
1
0
1
1
1
mRS < 2 at referral and dead at 12 mo
0
0
0
0
0
0
mRS = 2 at referral and mRS < 2 at 12 mo
5
5
0
0
0
0
mRS = 2 at referral and mRS = 2 at 12 mo
4
3
1
1
0
0
mRS = 2 at referral and mRS > 2 at 12 mo
0
0
0
0
0
0
mRS = 2 at referral and dead at 12 mo
1
0
1
0
0
0
mRS >2 at referral and mRS < 2 at 12 mo
0
0
0
0
0
0
mRS > 2 at referral and mRS = 2 at 12 mo
2
1
1
0
0
0
mRS> 2 at referral and mRS > 2 at 12 months
4
3
1
0
0
0
mRS > 2 at referral and dead at 12 mo
3
2
1
2
2
1
Hemorrhage after referral not treatment related
3
0
3
0
0
0
Ventral paramedian basilar perforators
18
11
7
0.93
3
3
0.32
2
Superior circumferential arteries
18
13
5
2
0
1
Anterior inferior cerebellar artery (AICA)
6
3
3
2
0
0
Lower circumferential arteries
4
3
1
0
0
0
Persistent or residual at last follow-up
2
2
9
2
2
11
Abbreviation: mRS, modified Rankin Scale.
Although the majority of bAVM present with intraparenchymal hemorrhage (▶Table 56.2), location in the brainstem increases the likelihood of presentation by subarachnoid hemorrhage (SAH). This is likely due to the greater preponderance of pial and extrapial locations. This is in keeping with the brainstem transitioning from the central nervous system (CNS) of the spinal cord (where pial-based arteriovenous fistulas are more common than glomus-type lesions within the cord) to the prosencephalic CNS where the bAVMs are more often located within deeply placed sulci.
Table 56.2 Outcome of management
Series
Management
Follow- up (mo)
Total bAVM cohort from which BSAVM cases selected
Proportion presenting with hemorrhage
Total cases (including those untreated)
Number treated
Proportion located in the midbrain/medulla (of all BSAVM cases if known*)
Pretreatment embolization
Proportion of treated cases obliterated
Total deaths in series following treatment
Complication of treatment leading to new permanent neurological deficit
