Revascularization Surgery in Childhood Associated with a Low Incidence of Microbleeds in Adult Patients with Moyamoya

Journal Pre-proof Revascularization surgery in childhood associated with a low incidence of microbleeds in adult moyamoya patients Yukihiro Yamao, M.D., Ph.D., Jun C. Takahashi, M.D., Ph.D., Takeshi Funaki, M.D., Ph.D., Yohei Mineharu, M.D., Ph.D., Takyuki Kikuchi, M.D., Ph.D., Tomohisa Okada, M.D., Ph.D., Kaori Togashi, M.D., Ph.D., Susumu Miyamoto, M.D., Ph.D. PII:

S1878-8750(19)32588-4

DOI:

https://doi.org/10.1016/j.wneu.2019.09.144

Reference:

WNEU 13444

To appear in:

World Neurosurgery

Received Date: 14 August 2019 Revised Date:

25 September 2019

Accepted Date: 26 September 2019

Please cite this article as: Yamao Y, Takahashi JC, Funaki T, Mineharu Y, Kikuchi T, Okada T, Togashi K, Miyamoto S, Revascularization surgery in childhood associated with a low incidence of microbleeds in adult moyamoya patients, World Neurosurgery (2019), doi: https://doi.org/10.1016/j.wneu.2019.09.144. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Elsevier Inc. All rights reserved.

Revascularization surgery in childhood associated with a low incidence of microbleeds in adult moyamoya patients

Yukihiro Yamao, M.D., Ph.D.,1 Jun C Takahashi, M.D., Ph.D.,2 Takeshi Funaki, M.D., Ph.D.,1 Yohei Mineharu, M.D., Ph.D.,1 Takyuki Kikuchi, M.D., Ph.D.,1 Tomohisa Okada, M.D., Ph.D.,3 Kaori Togashi, M.D., Ph.D.,4 Susumu Miyamoto, M.D., Ph.D.1

1

Department of Neurosurgery, Kyoto University Graduate School of Medicine

2

Department of Neurosurgery, National Cerebral and Cardiovascular Research Center

Hospital 3

Human Brain Research Center, Kyoto University Graduate School of Medicine

4

Department of Diagnostic Imaging and Nuclear Medicine, Kyoto University Graduate

School of Medicine

Corresponding to: Yukihiro Yamao, M.D., Ph.D. Department of Neurosurgery, Kyoto University Graduate School of Medicine, Kyoto, Japan 54, Shogoin Kawahara-cho, Sakyo-ku, Kyoto, 606-8507, Japan Tel & Fax: +81-75-751-3459 & +81-75-752-9501 E-mail: [email protected]

Key words: moyamoya disease, microbleeds, bypass surgery, hemorrhage Short title: Bypass surgery in childhood decreases MBs in MMD

1

Revascularization surgery in childhood associated with a low incidence of microbleeds in adult moyamoya patients

1

Abstract

2

Background

3

The clinical significance of asymptomatic microbleeds in moyamoya disease remains unclear.

4

The purpose of this study was to clarify the relationship between bypass surgery and the

5

incidence of asymptomatic microbleeds.

6

Methods

7

This retrospective study included 142 adult patients (mean age: 37.7 ± 13.5) with moyamoya

8

disease, of which 36 patients (25.3%) underwent bypass surgery in childhood. Hemorrhagic

9

onset was diagnosed in 31 patients (21.8%). The incidence of microbleeds was evaluated on

10

T2*- or susceptibility-weighted imaging from 3-Tesla magnetic resonance imaging. The patients

11

were subsequently categorized into “MBs” or “non-MBs” group. Since previous microbleeds

12

potentially lead to hemorrhage, MBs group was defined as patients with radiographic evidence of

13

bleeding, including asymptomatic microbleeds and/or hemorrhagic onset. The association of

14

baseline characteristics was evaluated.

2

1

Results

2

Asymptomatic microbleeds were detected in 38 patients (26.8%). Of 31 patients with

3

hemorrhagic onset, 18 had microbleeds, while 13 had no microbleeds. Therefore, 51 patients

4

(35.9%) were classified into MBs group. Bypass surgery in childhood (MBs: 7.8% versus

5

non-MBs: 35.2%, P < 0.01) and mean age (MBs: 42.9 ± 1.8 versus non-MBs: 34.7 ± 1.4 years, P

6

< 0.01) were statistically significant factors associated with microbleeds, but only bypass surgery

7

in childhood remained statistically significant after multivariable adjustment (odds ratio: 0.25,

8

95% confidence interval: 0.07–0.87, P = 0.03).

9

Conclusions

10

This study demonstrates the clinical significance of revascularization surgery in childhood

11

associated with a low incidence of asymptomatic microbleeds in adult patients with moyamoya

12

disease. This indicates that a newly established bypass can reduce hemodynamic overstress.

13

3

1

Introduction

2

Moyamoya disease is a unique cerebrovascular disease characterized by progressive occlusion of

3

the bilateral internal carotid arteries at their terminal portions and unusual secondarily formed

4

vascular networks (moyamoya vessels) that act as collateral pathways. Cerebral hemorrhage is

5

considered the most significant event contributing to a poor prognosis or death, and long-term

6

hemodynamic stress to the collateral vessels is thought to induce vascular pathologies leading to

7

hemorrhage.1

8

9

Extracranial-intracranial bypass surgery is often used for the treatment of ischemic moyamoya disease, and angiographic diminishment of moyamoya vessels can be observed after

10

surgery, which is regarded as decreased hemodynamic stress to these vessels. The Japan Adult

11

Moyamoya (JAM) trial, a multicentered, prospective, randomized, controlled study, has

12

demonstrated that direct bypass surgery significantly decreases the rate of rebleeding attacks

13

during the 5 years following it.1 The “posterior hemorrhage” group benefitted more from the

14

surgery than the “anterior hemorrhage” group.2 Involvement of a steno-occlusive lesion in the

15

posterior cerebral artery (PCA) is a relatively specific finding in juvenile-onset moyamoya

16

disease.3 The PCA usually provides high collateral flow to the anterior circulation in moyamoya

4

1

disease, and PCA involvement leads to the development of the collateral vessels or the so-called

2

“posterior moyamoya vessels”.4 Recent studies revealed that PCA involvement and choroidal

3

anastomosis were risk factors associated with posterior hemorrhage.2,5 In other studies, no or very

4

low incidences of intracranial hemorrhage occurred in pediatric patients after revascularization

5

surgery.6-8 These studies further indicate that a newly established bypass can reduce

6

hemodynamic overstress and intracranial hemorrhage.

7

Microbleeds are known to display low signal intensity on T2*-weighted (T2*WI) or

8

susceptibility-weighted imaging (SWI), resulting from magnetic inhomogeneity due to

9

hemosiderin deposition around angiopathic arterioles.9 Microbleeds are frequently detected in

10

patients with hypertension, lacunar infarction, or cerebral hemorrhage, and even in a small

11

number of healthy individuals.10-12 In patients with moyamoya disease, a higher frequency of

12

asymptomatic microbleeds has been reported, compared with normal controls;13 these are

13

primarily detected in the periventricular region, followed by the basal ganglia and the thalamus,

14

where intracranial bleeding often occurs.13-16 The presence of asymptomatic microbleeds may

15

thus be a predictor of subsequent hemorrhage.17,18 The clinical information on microbleeds is still

16

limited. In previous studies,16,19 no or very low incidences of asymptomatic microbleeds were

5

1

observed in pediatric patients, probably due to shorter disease durations. However, in adult

2

patients, the factors associated with low incidences of asymptomatic microbleeds remain unclear.

3

4

We retrospectively investigated the relationship between bypass surgery and the incidence of asymptomatic microbleeds in adult patients with moyamoya disease.

5

6

Material and methods

7

Patient selection

8

In this cross-sectional study, a total of 210 consecutive patients who underwent 3-Tesla magnetic

9

resonance imaging (MRI) scans between April 2010 and December 2013 were included. These

10

patients were referred to our institution for diagnosis and treatment. Moyamoya disease was

11

diagnosed according to the criteria proposed by the Research Committee on Moyamoya Disease

12

in Japan.20 We excluded patients < 18 years of age, as well as patients diagnosed with

13

quasi-moyamoya disease. Fifty-seven patients in total were excluded for age, and 11 patients

14

because of underlying diseases such as hyperthyroidism. A total of 142 patients were included in

15

further analyses. This study was approved by the Ethics Committee of Kyoto University

16

Graduate School of Medicine (IRB-E2247) and qualified for waiver of individual consent.

6

1

Variables

2

Baseline variables, including the age at the time of MRI scan, sex, timing of surgery, primary

3

clinical manifestations, and known risk factors for microbleeds or hemorrhage (PCA

4

involvement5 and hypertension10), were retrospectively obtained from the patient records. As for

5

the timing of surgery, bypass surgery in childhood was defined as “bypass surgery performed

6

before 18 years of age”, and other cases as “no bypass surgery before 18 years of age”; the latter

7

category included patients with bypass surgery performed at 18 years of age or later, candidates

8

for bypass surgery, and patients receiving conservative therapy. We employed a direct bypass

9

method, such as superficial temporal artery to middle cerebral artery bypass, as a first-line

10

treatment even for pediatric patients.21 Primary clinical manifestations were classified as either

11

hemorrhagic or non-hemorrhagic onset. Hemorrhagic onset was defined as the subtype causing

12

any symptomatic intracranial hemorrhage diagnosed with computed tomography (CT).

13

Non-hemorrhagic onset was defined as transient ischemic attack, seizure, headache, or

14

symptomatic ischemic stroke diagnosed with CT or MRI. Patients were diagnosed with

15

hypertension when they had a systolic blood pressure of ≥140 mmHg or a diastolic blood

7

1

pressure of ≥90 mmHg at admission, at the outpatient clinic, or when they were receiving

2

medical treatment for hypertension.

3

MRI protocol

4

All patients underwent imaging with a 3-Tesla MRI system (Magnetom TIM Trio or Skyra;

5

Siemens Healthcare, Erlangen, Germany) with a 32-channel head coil. T2*WI was acquired

6

using a gradient-echo sequence under the following conditions: time of repetition = 500 ms; echo

7

time = 20 ms; flip angle = 20°; matrix = 224 × 256; field of view = 193 × 220 mm; matrix size

8

= 0.86 × 0.86; 35 slices of 3-mm thickness with a 1-mm gap. SWI was acquired with a

9

3-dimensional fully flow-compensated gradient echo sequence using the following parameters:

10

time of repetition = 28 ms; echo time = 20 ms; flip angle = 15°; matrix = 230 × 320; field of

11

view = 179 × 230 mm; matrix size = 0.78 × 0.72 mm. The slab size was 128 mm, partitioned into

12

64 slices of 2 mm. MR angiography was acquired using a gradient-echo sequence under the

13

following conditions: time of repetition = 20 ms; echo time = 3.69 ms; flip angle = 20°; matrix

14

= 328 × 384; field of view = 188 × 220 mm; matrix size = 0.573 × 0.573 mm; 156 slices of 0.7

15

mm thickness acquired in four slabs.

16

8

1

Definition of MRI findings

2

The presence of microbleeds was assessed with either T2*WI or SWI. Asymptomatic

3

microbleeds were defined as small hypointense areas (<10 mm in diameter) with well-defined

4

margins,13 except those located in the cerebellum and brainstem (Figure 1). We also excluded

5

lesions in pre- and postcentral gyri because of possible artifacts due to craniotomy. Patients were

6

classified into the “MBs” or “non-MBs” group. In some cases with hemorrhagic onset, the

7

hemorrhage originated from the site where microbleeds were detected.17 Since chronic

8

hematoma on MRI scans potentially include previous microbleeds, the MBs group was defined

9

as patients with radiographic evidence of bleeding, including asymptomatic microbleeds and/or

10

11

hemorrhagic onset. Locations of asymptomatic microbleeds were divided into an anterior and a posterior

12

region, based on the JAM trial.2 Briefly, microbleeds in the anterior region were defined as those

13

located in the putamen, caudate head, frontal lobe, anterior half of the temporal lobe,

14

subependymal area of the anterior part of the lateral ventricle, or anterior half of the corpus

15

callosum. Microbleeds in the posterior region were defined as those located in the thalamus,

16

posterior half of the temporal lobe, parietal lobe, occipital lobe, subependymal area of the

9

1

posterior part of the lateral ventricle including the trigon, or posterior half of the corpus

2

callosum.

3

4

5

PCA involvement was defined as the presence of occlusion or stenosis greater than 50% in the P1–P3 segments of the PCA with decreased delineation of the cortical arteries.3,5 Repeated MRI scans were not performed during study registration in all patients due to

6

clinical limitations. Thus, in candidates for bypass surgery, preoperative MRI scans were

7

evaluated, while in other patients, the initial MRI scans during the registration were evaluated.

8

All MRI findings were evaluated by three coauthors (Y.Y., J.C.T., and S.M.) who were blinded

9

to the clinical information at the time of evaluation.

10

11

Statistical analyses

12

For comparisons of baseline characteristics and microbleeds locations, t-tests or the Fisher exact

13

test were used as appropriate. Variables with P < 0.05 on univariate analyses were selected for

14

further multivariable analyses. Multiple logistic regression analysis was used for multivariable

15

analysis. Two-sided P values < 0.05 were considered statistically significant. All statistical

16

analyses were performed with JMP software (version 14, SAS Institute Inc., Cary, NC, USA).

10

1

2

Results

3

Baseline variables

4

Of the 142 patients, 85 (59.8%) were female. The mean age at the time of the MRI scan was 37.7

5

years (standard deviation [SD] 13.5 years). Thirty-six patients (25.4%) underwent direct bypass

6

surgery before 18 years of age (surgery in childhood), and 106 patients (74.6%) did not undergo

7

bypass surgery before 18 years of age; 77 patients (54.2%) underwent direct bypass surgery at 18

8

years of age or later, and 29 patients (20.4%) had no bypass surgery. Thirty-one patients (21.8%)

9

were diagnosed with hemorrhagic moyamoya disease, while 111 patients (78.2%) with

10

non-hemorrhagic moyamoya disease. PCA involvement was identified in 48 patients (33.8%).

11

Twenty-eight patients (19.7%) were diagnosed with hypertension.

12

13

Comparison between the MBs and non-MBs groups

14

Asymptomatic microbleeds were detected in 38 patients (26.8%) in this cohort. Of the 31

15

patients with hemorrhagic onset, 18 had asymptomatic microbleeds, while 13 had no

16

microbleeds. Thus, 51 (38+13) patients (35.9%) were classified into the MBs group (patients

11

1

with radiographic evidence of bleeding) and 91 patients (64.1%) into the non-MBs group. The

2

mean age at the time of the MRI scan was significantly higher in the MBs group than in the

3

non-MBs group (42.9 ± 1.8 versus 34.7 ± 1.4, P < 0.01). A statistically significant difference was

4

also observed in the number of patients who underwent surgery in childhood (MBs group: 7.8%

5

versus non-MBs group: 35.2%, P < 0.01). No significant difference was observed in sex (MBs

6

group: 56.9% female versus non-MBs group: 61.5% female, P = 0.59), PCA involvement (MBs

7

group: 31.4% versus non-MBs group: 35.2%, P = 0.65), or hypertension (MBs group: 19.6%

8

versus non-MBs group: 19.8%, P = 0.98). The results are shown in Table 1.

9

Variables including mean age at the time of MRI scan and surgery in childhood were

10

incorporated into a further multivariable analysis. In this analysis, surgery in childhood (odds

11

ratio [OR] 0.25, 95% confidence interval [CI] 0.07–0.87, P = 0.03) was identified as a significant

12

factor, while mean age at the time of MRI scan was not (OR 1.03, 95% CI 0.99–1.06, P = 0.09).

13

The results of the multivariable analysis are shown in Table 2.

14

15

Locations of asymptomatic microbleeds

12

1

An illustrative case is shown in Figure 1, and the distribution of asymptomatic microbleeds (total

2

number: 60) is described in Figure 2 and Table 3. Microbleeds were predominantly observed in

3

the periventricular region (35 microbleeds; 58.3%) and the basal ganglia/thalami (13

4

microbleeds; 21.7%). Forty-eight of the 60 asymptomatic microbleeds (80.0%) were

5

predominantly located in the posterior region. In patients with bypass surgery in childhood, 3 out of 5 microbleeds (60.0%) were

6

7

detected in the anterior region, whereas in patients without bypass surgery in childhood (bypass

8

surgery as adults or conservative therapy), 40 out of 55 microbleeds (72.7%) were located in the

9

posterior region. There was no statistically significant correlation between bypass surgery and

10

the location of microbleeds (P = 0.13). Of the 22 asymptomatic microbleeds in the 14 patients with PCA involvement, 15

11

12

microbleeds (68.2%) were predominantly detected in the ipsilateral hemisphere of the involved

13

PCA.

14

15

Discussion

13

1

The major finding of the present retrospective study is that revascularization surgery in

2

childhood significantly reduces the incidence of microbleeds. To the best of our knowledge, this

3

retrospective study is the first to confirm the association between the timing of revascularization

4

surgery and incidence of asymptomatic microbleeds in moyamoya disease.

5

Moyamoya vessels are formed by the development and dilatation of branches of

6

lenticulostriate arteries or choroidal arteries that frequently supply the basal ganglia, thalamus, or

7

periventricular region, which are the most common sites of bleeding in moyamoya disease.2,18,22

8

Asymptomatic microbleeds in patients with moyamoya disease are also predominantly observed

9

in the periventricular region and the basal ganglia/thalami.13-16 These observations are in line

10

with our results. The precise pathophysiology of microbleeds or hemorrhage in patients with

11

moyamoya disease is still unclear. It is likely, from a recent in vivo study,23 that microbleeds

12

reflect fragile microvessels with an altered blood brain barrier and that microbleeds are the

13

results of blood product extravasation; long-term hemodynamic overstress is thought to induce a

14

rupture/extravasation of the dilated, fragile moyamoya vessels.1,18 Our multivariable analysis

15

results show that revascularization surgery in childhood is a significant factor associated with a

16

low incidence of asymptomatic microbleeds. The JAM trial has demonstrated that newly

14

1

established bypass flow can reduce hemodynamic overstress of the collateral vessels in

2

hemorrhagic moyamoya disease.1 Our results, therefore, suggest that revascularization surgery at

3

younger ages reduces the hemodynamic overstress of the collateral vessels in moyamoya disease

4

as well as the incidence of asymptomatic microbleeds.

5

In a previous study, the “posterior hemorrhage” group was found to be at a higher risk

6

of rebleeding, and PCA involvement was identified as a risk factor associated with posterior

7

hemorrhage.5 The present study reveals that, although asymptomatic microbleeds were not

8

significantly associated with PCA involvement, microbleeds are predominantly located in

9

posterior regions and in the same hemisphere as the involved PCA. These findings also suggest

10

that microbleeds might predict subsequent hemorrhage. In previous studies using digital

11

subtraction angiography, asymptomatic microbleeds were found to be associated with the

12

dilation and extension of anterior choroidal and posterior communicating arteries.10,15 Because

13

we evaluated intracranial arteries only by MR angiography, we could not fully evaluate collateral

14

arteries, including the anterior or posterior choroidal, the posterior communicating arteries, or the

15

PCA. Further studies will be necessary to evaluate how genuine collateral vessels are associated

16

with microbleeds.

15

1

Limitations

2

Due to the retrospective design of the assessment, there are several limitations to be considered.

3

First, the onset of moyamoya disease was not clear in all patients. The disease duration from the

4

onset and the duration after bypass surgery were not consistent across patients. In addition, the

5

repeated MRI scans were not evaluated; these might have biased the findings. In previous

6

studies,16,19 new cerebral microbleeds were detected in 4.2−6.9% of adult patients with

7

moyamoya disease during a mean follow-up period of 17.0−43.1 months. No microbleeds were

8

reported to be detected in pediatric patients, and the possibility of shorter disease periods in

9

pediatric patients leading to no or very low incidences of asymptomatic microbleeds were also

10

reported.16 In our retrospective study, MRI data at the disease onset were not available in all

11

patients; thus, it remains unclear when microbleeds occurred in each patient. Second,

12

angiography was not available for all patients; thus, we could not evaluate the stages of

13

moyamoya disease, such as Suzuki’s angiographic staging.24 In a recent review, Suzuki’s

14

angiographic staging was significantly higher in hemorrhagic onset patients than in ischemic

15

onset patients.25 Since microbleeds potentially lead to hemorrhage,17,18 microbleeds might occur

16

more frequently in patients with higher Suzuki’s staging.

16

1

2

Conclusions

3

The present retrospective study demonstrates the clinical significance of revascularization

4

surgery in childhood, which is associated with a low incidence of asymptomatic microbleeds in

5

adult patients with moyamoya disease. Further prospective studies are required to identify the

6

genuine effect of revascularization surgery on the incidence of asymptomatic microbleeds.

7

8

Funding

9

This research did not receive any specific grant from funding agencies in the public, commercial,

10

11

12

or not-for-profit sectors.

17

1

Revascularization surgery in childhood associated with a low incidence of microbleeds in

2

adult moyamoya patients

3 4

Abstract

5

Background

6

The clinical significance of asymptomatic microbleeds in moyamoya disease remains unclear.

7

The purpose of this study was to clarify the relationship between bypass surgery and the

8

incidence of asymptomatic microbleeds.

9

Methods

10

This retrospective study included 142 adult patients (mean age: 37.7 ± 13.5) with moyamoya

11

disease, of which 36 patients (25.3%) underwent bypass surgery in childhood. Hemorrhagic

12

onset was diagnosed in 31 patients (21.8%). The incidence of microbleeds was evaluated on

13

T2*- or susceptibility-weighted imaging from 3-Tesla magnetic resonance imaging. The patients

14

were subsequently categorized into “MBs” or “non-MBs” group. Since previous microbleeds

15

potentially lead to hemorrhage, MBs group was defined as patients with radiographic evidence of

16

bleeding, including asymptomatic microbleeds and/or hemorrhagic onset. The association of

17

baseline characteristics was evaluated.

2

1

Results

2

Asymptomatic microbleeds were detected in 38 patients (26.8%). Of 31 patients with

3

hemorrhagic onset, 18 had microbleeds, while 13 had no microbleeds. Therefore, 51 patients

4

(35.9%) were classified into MBs group. Bypass surgery in childhood (MBs: 7.8% versus

5

non-MBs: 35.2%, P < 0.01) and mean age (MBs: 42.9 ± 1.8 versus non-MBs: 34.7 ± 1.4 years, P

6

< 0.01) were statistically significant factors associated with microbleeds, but only bypass surgery

7

in childhood remained statistically significant after multivariable adjustment (odds ratio: 0.25,

8

95% confidence interval: 0.07–0.87, P = 0.03).

9

Conclusions

10

This study demonstrates the clinical significance of revascularization surgery in childhood

11

associated with a low incidence of asymptomatic microbleeds in adult patients with moyamoya

12

disease. This indicates that a newly established bypass can reduce hemodynamic overstress.

13

3

1

Introduction

2

Moyamoya disease is a unique cerebrovascular disease characterized by progressive occlusion of

3

the bilateral internal carotid arteries at their terminal portions and unusual secondarily formed

4

vascular networks (moyamoya vessels) that act as collateral pathways. Cerebral hemorrhage is

5

considered the most significant event contributing to a poor prognosis or death, and long-term

6

hemodynamic stress to the collateral vessels is thought to induce vascular pathologies leading to

7

hemorrhage.1

8

9

Extracranial-intracranial bypass surgery is often used for the treatment of ischemic moyamoya disease, and angiographic diminishment of moyamoya vessels can be observed after

10

surgery, which is regarded as decreased hemodynamic stress to these vessels. The Japan Adult

11

Moyamoya (JAM) trial, a multicentered, prospective, randomized, controlled study, has

12

demonstrated that direct bypass surgery significantly decreases the rate of rebleeding attacks

13

during the 5 years following it.1 The “posterior hemorrhage” group benefitted more from the

14

surgery than the “anterior hemorrhage” group.2 Involvement of a steno-occlusive lesion in the

15

posterior cerebral artery (PCA) is a relatively specific finding in juvenile-onset moyamoya

16

disease.3 The PCA usually provides high collateral flow to the anterior circulation in moyamoya

4

1

disease, and PCA involvement leads to the development of the collateral vessels or the so-called

2

“posterior moyamoya vessels”.4 Recent studies revealed that PCA involvement and choroidal

3

anastomosis were risk factors associated with posterior hemorrhage.2,5 In other studies, no or very

4

low incidences of intracranial hemorrhage occurred in pediatric patients after revascularization

5

surgery.6-8 These studies further indicate that a newly established bypass can reduce

6

hemodynamic overstress and intracranial hemorrhage.

7

Microbleeds are known to display low signal intensity on T2*-weighted (T2*WI) or

8

susceptibility-weighted imaging (SWI), resulting from magnetic inhomogeneity due to

9

hemosiderin deposition around angiopathic arterioles.9 Microbleeds are frequently detected in

10

patients with hypertension, lacunar infarction, or cerebral hemorrhage, and even in a small

11

number of healthy individuals.10-12 In patients with moyamoya disease, a higher frequency of

12

asymptomatic microbleeds has been reported, compared with normal controls;13 these are

13

primarily detected in the periventricular region, followed by the basal ganglia and the thalamus,

14

where intracranial bleeding often occurs.13-16 The presence of asymptomatic microbleeds may

15

thus be a predictor of subsequent hemorrhage.17,18 The clinical information on microbleeds is still

16

limited. In previous studies,16,19 no or very low incidences of asymptomatic microbleeds were

5

1

observed in pediatric patients, probably due to shorter disease durations. However, in adult

2

patients, the factors associated with low incidences of asymptomatic microbleeds remain unclear.

3

4

We retrospectively investigated the relationship between bypass surgery and the incidence of asymptomatic microbleeds in adult patients with moyamoya disease.

5

6

Material and methods

7

Patient selection

8

In this cross-sectional study, a total of 210 consecutive patients who underwent 3-Tesla magnetic

9

resonance imaging (MRI) scans between April 2010 and December 2013 were included. These

10

patients were referred to our institution for diagnosis and treatment. Moyamoya disease was

11

diagnosed according to the criteria proposed by the Research Committee on Moyamoya Disease

12

in Japan.20 We excluded patients < 18 years of age, as well as patients diagnosed with

13

quasi-moyamoya disease. Fifty-seven patients in total were excluded for age, and 11 patients

14

because of underlying diseases such as hyperthyroidism. A total of 142 patients were included in

15

further analyses. This study was approved by the Ethics Committee of Kyoto University

16

Graduate School of Medicine (IRB-E2247) and qualified for waiver of individual consent.

6

1

Variables

2

Baseline variables, including the age at the time of MRI scan, sex, timing of surgery, primary

3

clinical manifestations, and known risk factors for microbleeds or hemorrhage (PCA

4

involvement5 and hypertension10), were retrospectively obtained from the patient records. As for

5

the timing of surgery, bypass surgery in childhood was defined as “bypass surgery performed

6

before 18 years of age”, and other cases as “no bypass surgery before 18 years of age”; the latter

7

category included patients with bypass surgery performed at 18 years of age or later, candidates

8

for bypass surgery, and patients receiving conservative therapy. We employed a direct bypass

9

method, such as superficial temporal artery to middle cerebral artery bypass, as a first-line

10

treatment even for pediatric patients.21 Primary clinical manifestations were classified as either

11

hemorrhagic or non-hemorrhagic onset. Hemorrhagic onset was defined as the subtype causing

12

any symptomatic intracranial hemorrhage diagnosed with computed tomography (CT).

13

Non-hemorrhagic onset was defined as transient ischemic attack, seizure, headache, or

14

symptomatic ischemic stroke diagnosed with CT or MRI. Patients were diagnosed with

15

hypertension when they had a systolic blood pressure of ≥140 mmHg or a diastolic blood

7

1

pressure of ≥90 mmHg at admission, at the outpatient clinic, or when they were receiving

2

medical treatment for hypertension.

3

MRI protocol

4

All patients underwent imaging with a 3-Tesla MRI system (Magnetom TIM Trio or Skyra;

5

Siemens Healthcare, Erlangen, Germany) with a 32-channel head coil. T2*WI was acquired

6

using a gradient-echo sequence under the following conditions: time of repetition = 500 ms; echo

7

time = 20 ms; flip angle = 20°; matrix = 224 × 256; field of view = 193 × 220 mm; matrix size

8

= 0.86 × 0.86; 35 slices of 3-mm thickness with a 1-mm gap. SWI was acquired with a

9

3-dimensional fully flow-compensated gradient echo sequence using the following parameters:

10

time of repetition = 28 ms; echo time = 20 ms; flip angle = 15°; matrix = 230 × 320; field of

11

view = 179 × 230 mm; matrix size = 0.78 × 0.72 mm. The slab size was 128 mm, partitioned into

12

64 slices of 2 mm. MR angiography was acquired using a gradient-echo sequence under the

13

following conditions: time of repetition = 20 ms; echo time = 3.69 ms; flip angle = 20°; matrix

14

= 328 × 384; field of view = 188 × 220 mm; matrix size = 0.573 × 0.573 mm; 156 slices of 0.7

15

mm thickness acquired in four slabs.

16

8

1

Definition of MRI findings

2

The presence of microbleeds was assessed with either T2*WI or SWI. Asymptomatic

3

microbleeds were defined as small hypointense areas (<10 mm in diameter) with well-defined

4

margins,13 except those located in the cerebellum and brainstem (Figure 1). We also excluded

5

lesions in pre- and postcentral gyri because of possible artifacts due to craniotomy. Patients were

6

classified into the “MBs” or “non-MBs” group. In some cases with hemorrhagic onset, the

7

hemorrhage originated from the site where microbleeds were detected.17 Since chronic

8

hematoma on MRI scans potentially include previous microbleeds, the MBs group was defined

9

as patients with radiographic evidence of bleeding, including asymptomatic microbleeds and/or

10

11

hemorrhagic onset. Locations of asymptomatic microbleeds were divided into an anterior and a posterior

12

region, based on the JAM trial.2 Briefly, microbleeds in the anterior region were defined as those

13

located in the putamen, caudate head, frontal lobe, anterior half of the temporal lobe,

14

subependymal area of the anterior part of the lateral ventricle, or anterior half of the corpus

15

callosum. Microbleeds in the posterior region were defined as those located in the thalamus,

16

posterior half of the temporal lobe, parietal lobe, occipital lobe, subependymal area of the

9

1

posterior part of the lateral ventricle including the trigon, or posterior half of the corpus

2

callosum.

3

4

5

PCA involvement was defined as the presence of occlusion or stenosis greater than 50% in the P1–P3 segments of the PCA with decreased delineation of the cortical arteries.3,5 Repeated MRI scans were not performed during study registration in all patients due to

6

clinical limitations. Thus, in candidates for bypass surgery, preoperative MRI scans were

7

evaluated, while in other patients, the initial MRI scans during the registration were evaluated.

8

All MRI findings were evaluated by three coauthors (Y.Y., J.C.T., and S.M.) who were blinded

9

to the clinical information at the time of evaluation.

10

11

Statistical analyses

12

For comparisons of baseline characteristics and microbleeds locations, t-tests or the Fisher exact

13

test were used as appropriate. Variables with P < 0.05 on univariate analyses were selected for

14

further multivariable analyses. Multiple logistic regression analysis was used for multivariable

15

analysis. Two-sided P values < 0.05 were considered statistically significant. All statistical

16

analyses were performed with JMP software (version 14, SAS Institute Inc., Cary, NC, USA).

10

1

2

Results

3

Baseline variables

4

Of the 142 patients, 85 (59.8%) were female. The mean age at the time of the MRI scan was 37.7

5

years (standard deviation [SD] 13.5 years). Thirty-six patients (25.4%) underwent direct bypass

6

surgery before 18 years of age (surgery in childhood), and 106 patients (74.6%) did not undergo

7

bypass surgery before 18 years of age; 77 patients (54.2%) underwent direct bypass surgery at 18

8

years of age or later, and 29 patients (20.4%) had no bypass surgery. Thirty-one patients (21.8%)

9

were diagnosed with hemorrhagic moyamoya disease, while 111 patients (78.2%) with

10

non-hemorrhagic moyamoya disease. PCA involvement was identified in 48 patients (33.8%).

11

Twenty-eight patients (19.7%) were diagnosed with hypertension.

12

13

Comparison between the MBs and non-MBs groups

14

Asymptomatic microbleeds were detected in 38 patients (26.8%) in this cohort. Of the 31

15

patients with hemorrhagic onset, 18 had asymptomatic microbleeds, while 13 had no

16

microbleeds. Thus, 51 (38+13) patients (35.9%) were classified into the MBs group (patients

11

1

with radiographic evidence of bleeding) and 91 patients (64.1%) into the non-MBs group. The

2

mean age at the time of the MRI scan was significantly higher in the MBs group than in the

3

non-MBs group (42.9 ± 1.8 versus 34.7 ± 1.4, P < 0.01). A statistically significant difference was

4

also observed in the number of patients who underwent surgery in childhood (MBs group: 7.8%

5

versus non-MBs group: 35.2%, P < 0.01). No significant difference was observed in sex (MBs

6

group: 56.9% female versus non-MBs group: 61.5% female, P = 0.59), PCA involvement (MBs

7

group: 31.4% versus non-MBs group: 35.2%, P = 0.65), or hypertension (MBs group: 19.6%

8

versus non-MBs group: 19.8%, P = 0.98). The results are shown in Table 1.

9

Variables including mean age at the time of MRI scan and surgery in childhood were

10

incorporated into a further multivariable analysis. In this analysis, surgery in childhood (odds

11

ratio [OR] 0.25, 95% confidence interval [CI] 0.07–0.87, P = 0.03) was identified as a significant

12

factor, while mean age at the time of MRI scan was not (OR 1.03, 95% CI 0.99–1.06, P = 0.09).

13

The results of the multivariable analysis are shown in Table 2.

14

15

Locations of asymptomatic microbleeds

12

1

An illustrative case is shown in Figure 1, and the distribution of asymptomatic microbleeds (total

2

number: 60) is described in Figure 2 and Table 3. Microbleeds were predominantly observed in

3

the periventricular region (35 microbleeds; 58.3%) and the basal ganglia/thalami (13

4

microbleeds; 21.7%). Forty-eight of the 60 asymptomatic microbleeds (80.0%) were

5

predominantly located in the posterior region. In patients with bypass surgery in childhood, 3 out of 5 microbleeds (60.0%) were

6

7

detected in the anterior region, whereas in patients without bypass surgery in childhood (bypass

8

surgery as adults or conservative therapy), 40 out of 55 microbleeds (72.7%) were located in the

9

posterior region. There was no statistically significant correlation between bypass surgery and

10

the location of microbleeds (P = 0.13). Of the 22 asymptomatic microbleeds in the 14 patients with PCA involvement, 15

11

12

microbleeds (68.2%) were predominantly detected in the ipsilateral hemisphere of the involved

13

PCA.

14

15

Discussion

13

1

The major finding of the present retrospective study is that revascularization surgery in

2

childhood significantly reduces the incidence of microbleeds. To the best of our knowledge, this

3

retrospective study is the first to confirm the association between the timing of revascularization

4

surgery and incidence of asymptomatic microbleeds in moyamoya disease.

5

Moyamoya vessels are formed by the development and dilatation of branches of

6

lenticulostriate arteries or choroidal arteries that frequently supply the basal ganglia, thalamus, or

7

periventricular region, which are the most common sites of bleeding in moyamoya disease.2,18,22

8

Asymptomatic microbleeds in patients with moyamoya disease are also predominantly observed

9

in the periventricular region and the basal ganglia/thalami.13-16 These observations are in line

10

with our results. The precise pathophysiology of microbleeds or hemorrhage in patients with

11

moyamoya disease is still unclear. It is likely, from a recent in vivo study,23 that microbleeds

12

reflect fragile microvessels with an altered blood brain barrier and that microbleeds are the

13

results of blood product extravasation; long-term hemodynamic overstress is thought to induce a

14

rupture/extravasation of the dilated, fragile moyamoya vessels.1,18 Our multivariable analysis

15

results show that revascularization surgery in childhood is a significant factor associated with a

16

low incidence of asymptomatic microbleeds. The JAM trial has demonstrated that newly

14

1

established bypass flow can reduce hemodynamic overstress of the collateral vessels in

2

hemorrhagic moyamoya disease.1 Our results, therefore, suggest that revascularization surgery at

3

younger ages reduces the hemodynamic overstress of the collateral vessels in moyamoya disease

4

as well as the incidence of asymptomatic microbleeds.

5

In a previous study, the “posterior hemorrhage” group was found to be at a higher risk

6

of rebleeding, and PCA involvement was identified as a risk factor associated with posterior

7

hemorrhage.5 The present study reveals that, although asymptomatic microbleeds were not

8

significantly associated with PCA involvement, microbleeds are predominantly located in

9

posterior regions and in the same hemisphere as the involved PCA. These findings also suggest

10

that microbleeds might predict subsequent hemorrhage. In previous studies using digital

11

subtraction angiography, asymptomatic microbleeds were found to be associated with the

12

dilation and extension of anterior choroidal and posterior communicating arteries.10,15 Because

13

we evaluated intracranial arteries only by MR angiography, we could not fully evaluate collateral

14

arteries, including the anterior or posterior choroidal, the posterior communicating arteries, or the

15

PCA. Further studies will be necessary to evaluate how genuine collateral vessels are associated

16

with microbleeds.

15

1

Limitations

2

Due to the retrospective design of the assessment, there are several limitations to be considered.

3

First, the onset of moyamoya disease was not clear in all patients. The disease duration from the

4

onset and the duration after bypass surgery were not consistent across patients. In addition, the

5

repeated MRI scans were not evaluated; these might have biased the findings. In previous

6

studies,16,19 new cerebral microbleeds were detected in 4.2−6.9% of adult patients with

7

moyamoya disease during a mean follow-up period of 17.0−43.1 months. No microbleeds were

8

reported to be detected in pediatric patients, and the possibility of shorter disease periods in

9

pediatric patients leading to no or very low incidences of asymptomatic microbleeds were also

10

reported.16 In our retrospective study, MRI data at the disease onset were not available in all

11

patients; thus, it remains unclear when microbleeds occurred in each patient. Second,

12

angiography was not available for all patients; thus, we could not evaluate the stages of

13

moyamoya disease, such as Suzuki’s angiographic staging.24 In a recent review, Suzuki’s

14

angiographic staging was significantly higher in hemorrhagic onset patients than in ischemic

15

onset patients.25 Since microbleeds potentially lead to hemorrhage,17,18 microbleeds might occur

16

more frequently in patients with higher Suzuki’s staging.

16

1

2

Conclusions

3

The present retrospective study demonstrates the clinical significance of revascularization

4

surgery in childhood, which is associated with a low incidence of asymptomatic microbleeds in

5

adult patients with moyamoya disease. Further prospective studies are required to identify the

6

genuine effect of revascularization surgery on the incidence of asymptomatic microbleeds.

7

8

Funding

9

This research did not receive any specific grant from funding agencies in the public, commercial,

10 11

or not-for-profit sectors.

17

1

References

2

1.

Miyamoto S, Yoshimoto T, Hashimoto N, et al. Effects of extracranial-intracranial bypass

3

for patients with hemorrhagic moyamoya disease: results of the Japan adult moyamoya

4

trial. Stroke. 2014;45:1415-1421.

5

2.

Takahashi JC, Funaki T, Houkin K, et al. Significance of the Hemorrhagic Site for

6

Recurrent Bleeding: Prespecified Analysis in the Japan Adult Moyamoya Trial. Stroke.

7

2016;47:37-43.

8

3.

on long-term clinical and social outcome of pediatric moyamoya disease. Journal of

9

neurosurgery Pediatrics. 2013;12:626-632.

10

11

Funaki T, Takahashi JC, Takagi Y, et al. Impact of posterior cerebral artery involvement

4.

Miyamoto S, Kikuchi H, Karasawa J, Nagata I, Ikota T, Takeuchi S. Study of the

12

posterior circulation in moyamoya disease. Clinical and neuroradiological evaluation. J

13

Neurosurg. 1984;61:1032-1037.

14

5.

Funaki T, Takahashi JC, Houkin K, et al. Angiographic features of hemorrhagic

15

moyamoya disease with high recurrence risk: a supplementary analysis of the Japan Adult

16

Moyamoya Trial. J Neurosurg. 2017;14:1-8.

18

1

6.

anastomosis for moyamoya disease. Neurosurgical focus. 1998;5:e5.

2

3

Miyamoto S, Akiyama Y, Nagata I, et al. Long-term outcome after STA-MCA

7.

Funaki T, Takahashi JC, Miyamoto S. Late Cerebrovascular Events and Social Outcome

4

after Adolescence: Long-term Outcome of Pediatric Moyamoya Disease. Neurologia

5

medico-chirurgica. 2018;58:240-246.

6

8.

Funaki T, Takahashi JC, Takagi Y, et al. Incidence of late cerebrovascular events after

7

direct bypass among children with moyamoya disease: a descriptive longitudinal study at

8

a single center. Acta Neurochir (Wien). 2014;156:551-559; discussion 559.

9

9.

Fazekas F, Kleinert R, Roob G, et al. Histopathologic analysis of foci of signal loss on

10

gradient-echo T2*-weighted MR images in patients with spontaneous intracerebral

11

hemorrhage: evidence of microangiopathy-related microbleeds. AJNR Am J Neuroradiol.

12

1999;20:637-642.

13

10.

Schrag M, Greer DM. Clinical Associations of Cerebral Microbleeds on Magnetic

14

Resonance Neuroimaging. Journal of stroke and cerebrovascular diseases : the official

15

journal of National Stroke Association. 2014;4:00332-00332.

16

11.

Kato H, Izumiyama M, Izumiyama K, Takahashi A, Itoyama Y. Silent cerebral

19

1

microbleeds on T2*-weighted MRI: correlation with stroke subtype, stroke recurrence,

2

and leukoaraiosis. Stroke. 2002;33:1536-1540.

3

12.

Roob G, Lechner A, Schmidt R, Flooh E, Hartung HP, Fazekas F. Frequency and location

4

of

5

2000;31:2665-2669.

6

13.

microbleeds

in

patients

with

primary

intracerebral

hemorrhage.

Stroke.

Kikuta K, Takagi Y, Nozaki K, et al. Asymptomatic microbleeds in moyamoya disease:

7

T2*-weighted gradient-echo magnetic resonance imaging study. J Neurosurg.

8

2005;102:470-475.

9

14.

Kazumata K, Shinbo D, Ito M, et al. Spatial Relationship between Cerebral Microbleeds,

10

Moyamoya Vessels, and Hematoma in Moyamoya Disease. Journal of stroke and

11

cerebrovascular diseases : the official journal of National Stroke Association. 2014.

12

15.

Sun W, Yuan C, Liu W, et al. Asymptomatic cerebral microbleeds in adult patients with

13

moyamoya disease: a prospective cohort study with 2 years of follow-up.

14

Cerebrovascular diseases (Basel, Switzerland). 2013;35:469-475.

15

16

16.

Kuroda S, Kashiwazaki D, Ishikawa T, Nakayama N, Houkin K. Incidence, locations, and longitudinal course of silent microbleeds in moyamoya disease: a prospective

20

T2*-weighted MRI study. Stroke. 2013;44:516-518.

1

2

17.

Kikuta K, Takagi Y, Nozaki K, Sawamoto N, Fukuyama H, Hashimoto N. The presence

3

of multiple microbleeds as a predictor of subsequent cerebral hemorrhage in patients with

4

moyamoya disease. Neurosurgery. 2008;62:104-111, discussion 111-102.

5

18.

Lancet Neurol. 2008;7:1056-1066.

6

7

19.

20.

Guidelines for diagnosis and treatment of moyamoya disease (spontaneous occlusion of the circle of Willis). Neurologia medico-chirurgica. 2012;52:245-266.

10

11

Wenz H, Wenz R, Maros M, et al. Incidence, Locations, and Longitudinal Course of Cerebral Microbleeds in European Moyamoya. Stroke. 2017;48:307-313.

8

9

Kuroda S, Houkin K. Moyamoya disease: current concepts and future perspectives.

21.

Karasawa J, Touho H, Ohnishi H, Miyamoto S, Kikuchi H. Long-term follow-up study

12

after extracranial-intracranial bypass surgery for anterior circulation ischemia in

13

childhood moyamoya disease. J Neurosurg. 1992;77:84-89.

14

22.

Ryan RW, Chowdhary A, Britz GW. Hemorrhage and risk of further hemorrhagic strokes

15

following cerebral revascularization in Moyamoya disease: A review of the literature.

16

Surg Neurol Int. 2012;3:72.

21

1

23.

Narducci A, Yasuyuki K, Onken J, Blecharz K, Vajkoczy P. In vivo demonstration of

2

blood-brain barrier impairment in Moyamoya disease. Acta Neurochir (Wien).

3

2019;161:371-378.

4

24.

net-like vessels in base of brain. Archives of neurology. 1969;20:288-299.

5

6

7

8 9

Suzuki J, Takaku A. Cerebrovascular "moyamoya" disease. Disease showing abnormal

25.

Fujimura M, Tominaga T. Hemorrhagic Moyamoya Disease : A Recent Update. J Korean Neurosurg Soc. 2019;62:136-143.

22

1

Figure legends

2

Figure 1.

3

An illustrative case. A. Susceptibility-weighted imaging with 3-Tesla MRI reveals asymptomatic

4

microbleeds in the left thalamus (white arrowhead). B. MR angiography demonstrates bilateral

5

anterior, middle cerebral, and left posterior cerebral artery occlusion (white arrows). The location

6

of microbleeds in the posterior region and in the same hemisphere as the involved posterior

7

cerebral artery is of note.

8

9

Figure 2. Schematic illustration of the distribution of asymptomatic microbleeds

Table 1. Characteristics between the MBs and non-MBs group Total

MBs group

Non-MBs group

P value

Number of patients

142

51

91

Female (%)

85 (59.8)

29 (56.9)

56 (61.5)

0.59

Age (years ± SD)

37.7 ± 13.5

42.9 ± 1.8

34.7 ± 1.4

<0.01

Bypass surgery in

36 (25.3)

4 (7.8)

32 (35.2)

<0.01

PCA involvement (%)

48 (33.8)

16 (31.4)

32 (35.2)

0.65

Hypertension (%)

28 (19. 7)

10 (19.6)

18 (19.8)

0.98

childhood (%)

MBs = microbleeds; PCA = posterior cerebral artery; SD = standard deviation

Table 2. Multivariable analysis for factors associated with asymptomatic microbleeds Multivariable analysis Adjusted OR (95% CI) for microbleeds

P value

Age

1.03 (1.00–1.06)

0.09

Bypass surgery in childhood

0.25 (0.07–0.87)

0.03

CI = confidence interval; OR = odds ratio

Table 3. Location of asymptomatic microbleeds Number of microbleeds Bypass surgery in childhood

No bypass surgery in childhood

Anterior region (%)

3 (60.0)

15 (27.3)

Posterior region (%)

2 (40.0)

40 (72.7)

A

BB

Line between anterior and posterior region Microbleeds in patients with bypass surgery in childhood Microbleeds in patients without bypass surgery in childhood

Abbreviations CI = confidence interval; CT = computed tomography; JAM = Japan Adult Moyamoya; MRI = magnetic resonance imaging; OR = odds ratio; PCA = posterior cerebral artery; SD = standard deviation; SWI = susceptibility-weighted imaging; T2*WI = T2*-weighted imaging

Declaration of interest Conflicts of interest: none