Falcotentorial meningioma.pdf

J Neurosurg 104: 47–53, 2006

Falcotentorial meningioma: surgical outcome in 14 patients

T AKEO G OTO , M.D., K ENJI O HATA , M.D., M ICHIHARU M ORINO , M.D.,

T OSHIHIRO T AKAMI , M.D., N AOHIRO T SUYUGUCHI , M.D., A KIMASA N ISHIO , M.D.,

AND M ITSUHIRO H ARA , M.D.

Department of Neurosurgery, Osaka City University Graduate School of Medicine, Osaka, Japan

Object. The authors evaluated their surgical experience over 20 years with 14 treated falcotentorial meningiomas.

Methods. In the past 20 years, 14 patients with falcotentorial junction meningiomas were surgically treated. There

were seven men and seven women, whose ages ranged from 34 to 79 years. On the basis of neuroimaging studies, the

authors analyzed the influence of the anatomical relationship of the tumor to the vein of Galen, patency of the vein of

Galen, tumor size, and the signal intensities on the magnetic resonance images to determine possible difficulties that

might be encountered during surgery and to prognosticate the outcome of surgery. Depending on the relationship with

the vein of Galen, tumors were labeled as either a superior or an inferior type. All tumors were resected via an occip-

ital transtentorial approach.

The surgical outcome in eight patients was excellent; in the remaining six patients, it was fair. Of the prognostic fac-

tors, tumor location especially seemed to be the most important (p

K EY W ORDS

• meningioma • falcotentorial junction • galenic system

ENINGIOMAS arising from the falcotentorial junction

are relatively rare, and only isolated case reports or

small series related to surgical technique are avail-

able in the literature. 1,5,7,8,11–13,16 Because of the lesion’s depth

from the surface and its anatomical proximity with critical

neural and vascular structures, surgical access and tech-

nique are complex issues. A variety of factors influence

surgery and outcome. In this paper, on the basis of our expe-

rience, we analyze a set of factors to determine which prob-

ably influence surgical outcome.

achieved in 11 patients, and subtotal tumor removal in three.

Among the 11 patients, eight had no postoperative neuro-

logical worsening except for transient hemianopia; the oth-

er three patients experienced postoperative permanent neu-

rological deterioration, including memory disturbance and

hemianopia. Among the three patients who had undergone

a subtotal tumor removal, no neurological deterioration oc-

curred postoperatively. Surgical outcome was graded as sat-

isfactory in eight patients (Cases 1–8) given that the tumors

had been totally removed and there was no permanent neu-

rological deterioration. Outcome was unsatisfactory in six

patients (Cases 9–14) because either the lesions were not to-

tally removed or the patients had permanent functional dete-

rioration following total tumor removal.

Clinical Material and Methods

Patient Population and Tumor Characteristics

During the years between 1984 and 2004, 14 patients

with falcotentorial junction meningiomas were surgically

treated at our institution by the two senior surgeons (A. Ha-

kuba and K.O.). Pineal region meningiomas that did not

have a dural base in the falcotentorial region were not in-

cluded in this series. There were seven men and seven wom-

en whose ages ranged from 34 to 79 years (mean 57 years;

Table 1). The chief presenting symptoms included severe

headache in five, memory disturbance in five, gait distur-

bance in two, and tinnitus in two patients. All tumors were

exposed via an occipital transtentorial approach. The side of

the surgical approach was dependent on the predominant

side of tumor extension. Four tumors were approached on

the right side and five on the left. In five tumors the ap-

proach was bilateral because the tumor extended widely on

both sides of the midline. Total resection of the tumor was

Analysis of Prognostic Factors

Preoperative neuroimaging investigations included MR

imaging, MR venography, and angiography. Apart from

evaluating the physical characteristics of the tumor, we

evaluated the relationship of the tumor to the great vein of

Galen, the patency of the vein of Galen, and the straight si-

nus. Depending on the relationship of the tumor to the great

vein of Galen, tumors were classified into two types: tumors

located superior to the vein and compressing it downward

were labeled as the superior type, whereas those displacing

it superiorly were labeled as the inferior type (Fig. 1).

To evaluate prognostic factors, preoperative neuroimag-

ing findings and surgical findings were analyzed in eight

patients (Cases 1–8) with satisfactory outcomes and six

(Cases 9–14) with unsatisfactory outcomes. Neuroimag-

ing findings were statistically compared between these two

groups.

Abbreviation used in this paper: MR = magnetic resonance.

J. Neurosurg. / Volume 104 / January, 2006

0.01, Fisher exact test). The outcome associated

with the inferior type of tumor was significantly less optimal probably due to the relationship to the deep veins and the

brainstem. In this series, the occlusion of deep veins did not significantly influence outcome.

Conclusions. Classification of the tumor location by preoperative neuroimaging studies can be helpful in estimat-

ing the surgical difficulty that might be encountered in treating the falcotentorial junction meningioma.

T. Goto, et al.

TABLE 1

Summary of characteristics in 14 patients with falcotentorial junction meningiomas

Case

Age (yrs),

Chief

Extent of

Follow Up

Outcome

No.

Sex

Symptom

Tumor Removal

Complications

(mos)

Group

49, F

headache

total

none

238

satisfactory

50, F

headache

total

none

221

satisfactory

55, F

memory disturbance

total

none

220

satisfactory

54, M

memory disturbance

total

none

217

satisfactory

47, M

headache

total

transient hemianopia

187

satisfactory

63, M

memory disturbance

total

none

satisfactory

34, M

tinnitus

total

none

satisfactory

69, M

headache

total

none

satisfactory

69, F

memory disturbance

total

deterioration of memory

126

unsatisfactory

& hemianopia

49, F

headache

total

hemianopia

unsatisfactory

62, F

gait disturbance

total

memory disturbance

unsatisfactory

79, M

gait disturbance

subtotal

none

unsatisfactory

57, F

tinnitus

subtotal

none

unsatisfactory

61, M

memory disturbance

subtotal

none

unsatisfactory

* The MR imaging unit was not used before surgery in this case.

Results

Neuroimaging characteristics and surgical results are

summarized in Table 2. In the satisfactory-outcome group

(Cases 1–8), the tumor did not tightly adhere to the straight

sinus, deep veins (great vein of Galen and basal vein), col-

lateral venous channels, or brainstem. Only one patient

(Case 3) required suturing of the injured venous wall at a

single point during surgical procedures. This situation en-

abled the surgeons to remove the tumor totally without sur-

gical complication. On the other hand, in the three cases

(Cases 9–11) in which there was an unsatisfactory outcome,

there was tight adhesion or proximity between the tumor

and collateral venous channels of the deep veins. In all of

these three cases, there was intraoperative injury to the deep

veins or collateral veins during resection of the tumor. All

of these patients suffered postoperative venous infarction

associated with memory disturbance and hemianopia. The

cause of memory disturbance was unclear, although it was

presumed to be due to postoperative dysfunction of the left

mesial temporal lobe structures, which was related to com-

promise of the left basal vein of Rosenthal or the left inter-

nal cerebral vein.

Based on the lessons learned from our experience with

the initial three cases, a small amount of the tumor was left

behind to avoid injury to the deep veins, collateral veins,

and brainstem in the subsequent three cases (Cases 12–14).

Tumor Location Related to the Great Vein of Galen

In the satisfactory-outcome group, seven tumors were

classified as the superior type and one tumor as the inferi-

or type. On the other hand, in the unsatisfactory-outcome

group all six tumors were classified as the inferior type. The

location type was significantly different between the two

outcome groups (p

0.01, Fisher exact test).

Occlusion of the Great Vein of Galen and the

Straight Sinus

In the satisfactory-outcome group, preoperative inves-

tigations suggested occlusion of the great vein of Galen in

one case and of the straight sinus in another case. In the

unsatisfactory-outcome group, there were three cases of ve-

nous occlusion, one of which involved occlusion of the

straight sinus and two of which involved occlusion of both

the great vein and the straight sinus. Preoperative inves-

tigations in all cases with venous occlusion in both out-

come groups demonstrated well-developed collateral ve-

nous channels. Occlusion of the great vein of Galen and the

straight sinus did not significantly influence surgical out-

come (p = 0.34, Fisher exact test).

Tumor Size

Tumor size ranged from 22 to 58 mm (mean 38.6 mm).

In the satisfactory-outcome group, tumor size was signifi-

cantly larger than that in the unsatisfactory-outcome group

(mean

standard deviation, 44

9.4 mm compared with

31.3

6.1 mm; p

0.05, Welch t-test).

F IG . 1. Ilustration demonstrating the location type of the tumor.

A tumor located over the vein of Galen and compressing it down-

ward was classified as the superior type, and one situated under the

vein of Galen and dislocating it upward was the inferior type.

Magnetic Resonance Imaging Intensity of the Tumor

Eleven patients underwent MR imaging studies before

J. Neurosurg. / Volume 104 / January, 2006

Falcotentorial meningioma

TABLE 2

Neuroimaging characteristics and surgical findings in 14 patients with falcotentorial junction meningiomas*

Occlusion of Deep Veins

Surgical Findings

Vein

Tumor MRI Signal Intensity

Injuy to Deep

Suturing

Outcome

Case

Tumor

Straight

Size

or Collateral

of Vein

Group

No.

Location†

Galen

Sinus

(mm)†

T 1

T 2 †

Veins

Wall

satisfactory

superior

satisfactory

superior

satisfactory

superior

satisfactory

4‡

superior

iso

high

satisfactory

superior

iso

high

satisfactory

superior

iso

high

satisfactory

inferior

iso

high

satisfactory

superior

iso

high

unsatisfactory

inferior

iso

low

unsatisfactory

inferior

iso

unsatisfactory

inferior

iso

low

unsatisfactory

inferior

iso

low

unsatisfactory

inferior

iso

unsatisfactory

inferior

low

high

= patent.

† This prognostic factor significantly influenced surgical outcome.

‡ The MR imaging unit was not used before surgery in this case.

= occluded;

the operation; three patients were treated before the era of

MR imaging. On T 1 -weighted MR images, all tumors in the

satisfactory-outcome group and five of six tumors in the

unsatisfactory-outcome group appeared isointense. Signal

intensity on T 1 -weighted images was not significantly dif-

ferent between the two outcome groups (p = 0.68, Mann–

Whitney U-test). On T 2 -weighted MR images, four tumors

were hyperintense and one was isointense in the satisfacto-

ry-outcome patients. In the unsatisfactory-outcome group,

the tumor was hyperintense in one case, isointense in two

cases, and hypointense in three cases. The T 2 -weighted sig-

nal was significantly greater in the satisfactory-outcome

group than that in the unsatisfactory-outcome group (p

the falcotentorial junction. The size of the tumor was 53

mm in its maximal diameter. On an angiogram and a MR

venogram, the tumor was located over the vein of Galen and

dislocated it downward and was classified as the superi-

or type. The vein of Galen was occluded at the point of en-

try into the straight sinus, and the deep venous flow drained

through the collateral venous channels into the petrosal vein

and the transverse sinus. The tumor was excised via an oc-

cipital transtentorial approach. The tumor had a well-de-

fined plane of dissection from the brainstem, vein of Galen,

straight sinus, and collateral veins. The tumor could be to-

tally removed and there was no postoperative complication

(Fig. 2).

Case 13. This 57-year-old woman, who had presented

with a 2-month history of tinnitus, harbored an inferior type

of falcotentorial meningioma. Although the size of the tu-

mor was not very large (22 mm in its maximal diameter), it

was tightly adhered to the great vein of Galen and basal

veins. A small portion of the tumor was left behind around

these veins to avoid venous damage (Fig. 3). Postoperative-

ly, there was no neurological worsening.

Illustrative Cases

Case 8. This 69-year-old man presented with a 3-month

history of gradually progressing severe headache. There

was no neurological deficit at the time of admission.

Magnetic resonance imaging demonstrated a large mass at

F IG . 2. Case 8. a: Preoperative sagittal contrast-enhanced T 1 -weighted MR image demonstrating a superior type tu-

mor. b: An MR venogram demonstrating the anatomical relationship between the vein of Galen and the tumor. c: Post-

operative sagittal contrast-enhanced T 1 -weighted MR image showing total removal of the tumor.

J. Neurosurg. / Volume 104 / January, 2006

* Iso = isointense; NE = not evaluated;

0.05, Mann–Whitney U-test).

T. Goto, et al.

F IG . 3. Case 13. a: Preoperative sagittal contrast-enhanced T 1 -weighted MR image revealing an inferior type tumor.

b: An MR venogram showing the anatomical relationship between the vein of Galen and the tumor. c: Postoperative sag-

ittal contrast-enhanced T 1 -weighted MR image demonstrating partial removal of the tumor.

Discussion

Our study data suggested that the tumor location in rela-

tion to the great vein of Galen was the most significant

prognostic factor in surgery for the falcotentorial junction

meningioma. Occlusion of the galenic system, however,

was not an indicator of excellent surgical outcome. The size

of the tumor was not necessarily a prognostic factor given

that all of the large tumors in our series were the superior

type, could be totally resected, and were associated with a

satisfactory outcome. Concerning the MR imaging signal

intensity, it is clear that the softer meningiomas can be more

easily removed than the firmer ones.

A review of the literature demonstrates that preexisting

occlusion of the galenic system and the subsequent de-

velopment of collateral venous circulation is an important

factor when considering surgery on pineal region tumors,

including falcotentorial junction meningiomas. 1,3,9,10,14 In

such a situation, however, the collateral venous flow must

be preserved. 1,7 In the present study in cases of the inferior

type tumor with an occluded galenic system, dissecting the

tumor from surrounding collateral veins and brainstem was

relatively difficult technically. Resection of the tumor in

such cases led to damage of the surrounding structures with

additional neurological deficits. It appeared that in cases in

which the galenic venous system was patent, the surgical

procedure was more difficult because the tumor was tightly

adhered to the venous system. To prevent injury to the deep

vein, a small amount of the tumor can be left behind around

the deep veins to avoid their injury in the context of an infe-

rior type tumor. It was observed that in the case of the supe-

rior type tumor, even when the vein of Galen was patent, a

careful surgical technique enabled the surgeon to separate

the tumor from the vein of Galen. As a result, complete sur-

gical removal could be relatively safely performed in the

case of superior type tumors.

Several types of collateral venous channels developed as

a result of occlusions of the galenic system. Based on our

study data, it appears that collateral channels under the thick

arachnoid septum might be the most functionally and sur-

gically important collateral veins in surgery for the infe-

rior type tumor, because the tumor tends to adhere to all of

the galenic system including possible collateral veins in the

same cistern.

To evaluate the importance of tumor location as a prog-

nostic factor, we reviewed the literature on falcotentorial

junction meningiomas from the past 20 years and found 24

cases to which we could apply our classification system of

tumor location on the basis of neuroimages and lesion de-

scriptions (Table 3). 1,5,7,8,11–13,16 Of the 38 cases—that is, 14

from the present study and 24 from the literature—18 were

included in the satisfactory-outcome group and the other 20

TABLE 3

Relationship between tumor location and surgical

outcome in 38 cases

No. of Patients

Authors &

Tumor

factory

Unsatis-

Year

Location

Outcome

Suzuki, et al., 1984

superior

inferior

Odake & Goel, 1992

superior

inferior

Sekhar & Goel, 1992

superior

inferior

Asari, et al., 1995

superior

inferior

Matsuda & Inagawa,

superior

1995

inferior

Samii, et al., 1996

superior

inferior

Ziyal, et al., 1998

superior

inferior

Okami, et al., 2001

superior

inferior

present study

superior

F IG . 4. Graph showing the relationship between the tumor loca-

tion type and surgical outcome. The location of the tumor was sig-

nificantly different between the two outcome groups (p

inferior

total

0.01).

J. Neurosurg. / Volume 104 / January, 2006

Satis-

factory

Falcotentorial meningioma

F IG . 5. Drawings depicting the growing process of the superior type falcotentorial junction meningioma. a: The supe-

rior type tumor is originally located in the posterior pericallosal cistern. b: As the tumor increases in size, it pushes down

the quadrigeminal cistern, including the vein of the Galen and basal veins. The tumor and deep veins are in different cis-

terns. There is a thick arachnoid membrane between them.

in the unsatisfactory-outcome group. In the satisfactory-out-

come group, 13 tumors were classified as the superior type

and the other five as the inferior type. In the unsatisfactory-

outcome group, four tumors were categorized as the superi-

or type and the other 16 as the inferior type. Based on our

experience with the cases in our study and those reported in

the literature, it appears that there is a significant difference

in surgical outcome between superior and inferior types of

tumor locations (p

omy, the thick broad arachnoid membrane, extending from

the tentorial ridge to the splenium, clearly separates the pos-

terior pericallosal cistern from the quadrigeminal cistern,

which includes the great vein of Galen, basal veins, and dor-

sal midbrain. 2,4,6,15 It is most likely that the superior type tu-

mor, growing inside the posterior pericallosal cistern, might

compress deep veins over the arachnoid membrane. In this

situation, thick arachnoid membrane septum between two

cisterns protects the deep veins from direct tumor invasion,

which enables the surgeon to dissect the lesion from the

deep veins (Fig. 5). On the other hand, the inferior type tu-

mor, growing in the quadrigeminal cistern, might compress

the deep veins and dorsal midbrain in direct contact with

J. Neurosurg. / Volume 104 / January, 2006

0.01, Fisher exact test; Fig. 4).

The reason for the relative difficulty in surgical removal

and outcome between superior and inferior types of tumors

could be due to the relationship of the arachnoid membrane

to the deep veins, brainstem, and tumor. In the normal anat-

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika: