close

Вход

Забыли?

вход по аккаунту

?

Pediatric moyamoya disease An analysis of 410 consecutive cases.

код для вставкиСкачать
ORIGINAL ARTICLE
Pediatric Moyamoya Disease:
An Analysis of 410 Consecutive Cases
Seung-Ki Kim, MD, PhD,1 Byung-Kyu Cho, MD, PhD,1
Ji Hoon Phi, MD,1 Ji Yeoun Lee, MD,1 Jong Hee Chae, MD, PhD,2
Ki Joong Kim, MD, PhD,2 Yong-Seung Hwang, MD, PhD,2
In-One Kim, MD, PhD,3 Dong Soo Lee, MD, PhD,4,5
Joongyub Lee, MD,6 and Kyu-Chang Wang, MD, PhD1
Objective: Moyamoya disease (MMD) is a cerebrovascular occlusive disease of the bilateral internal carotid arteries
that causes a compensatory abnormal vascular network at the base of brain. The rare incidence and various
surgical techniques applied have limited the clinical research on MMD.
Methods: We conducted a retrospective analysis of the surgical outcome of 410 pediatric MMD patients. All
patients were treated in a relatively uniform scheme at a single institution. The surgical procedures consisted of
bilateral encephaloduroarteriosynangiosis augmented by bifrontal encephalogaleo-/periosteal synangiosis. Logistic
regression analyses were applied to reveal the prognostic factors for surgical outcome.
Results: The overall clinical outcome was excellent in 66%, good in 15%, fair in 15%, and poor in 4% of the
patients. Therefore, 81% of the patients had a favorable clinical outcome (excellent and good). Multivariate
analyses revealed that infarction on presentation was associated with unfavorable clinical outcome (odds ratio [OR],
2.85; 95% confidence interval [CI], 1.49 –5.46; p ⬍ 0.01) and decreased vascular reserve only on single-photon
emission computerized tomography (OR, 0.07; 95% CI, 0.01– 0.52; p ⬍ 0.01), with favorable clinical outcome.
Interpretation: Our results indicate that an early diagnosis and active intervention before establishment of irreversible hemodynamic change are essential to achieve a favorable clinical outcome in children with MMD.
ANN NEUROL 2010;68:92–101
M
oyamoya disease (MMD) is a cerebrovascular occlusive disease of the bilateral internal carotid arteries that causes a compensatory abnormal vascular network
at the base of the brain.1 Although the incidence of
MMD is not high (the annual incidence, 0.54 per
100,000 population2), it is the most common surgically
treated pediatric cerebrovascular disease in East Asia, particularly in Korea and Japan.3
The clinical presentation of MMD usually includes
repeated transient ischemic attacks (TIAs) in children and
intracranial hemorrhage in adults.4 The benefit of surgery
for the ischemic type of MMD has been established.4 –11
Because of its rarity and the various surgical techniques to establish adequate collateral circulation in the
ischemic brain, small patient numbers and a lack of multivariate analysis have limited prior surgical series of pediatric MMD.
We previously compared the surgical results of simple
encephaloduroarteriosynangiosis (EDAS) and EDAS with
bifrontal encephalogaleo-/periosteal synangiosis (EGPS) for
pediatric MMD in 159 children.4 Here we report our
assessment of the long-term outcomes in a larger group of
children with MMD who underwent similar presurgical
evaluation and relatively uniform surgical procedures at a
single institute. We conducted a multivariate analysis to
provide more details of the clinical characteristics and the
independent prognostic factors for clinical outcome following bypass surgery. We also demonstrated correlations
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/ana.21981
Received Aug 9, 2009, and in revised form Dec 25. Accepted for publication Jan 7, 2010.
Address correspondence to Dr Wang, Division of Pediatric Neurosurgery, Seoul National University Children’s Hospital, 101 Daehangno, Jongnogu, Seoul 110-744, Republic of Korea. E-mail: kcwang@snu.ac.kr
From the 1Division of Pediatric Neurosurgery, 2Division of Pediatric Neurology, and 3Division of Pediatric Radiology, Seoul National University
Children’s Hospital; and 4Department of Nuclear Medicine, 5Department of Internal Medicine, and 6Medical Research Collaborating Center,
Seoul National University Hospital, Seoul National University College of Medicine, Seoul, Republic of Korea.
92
© 2010 American Neurological Association
Kim et al: Pediatric Moyamoya Disease
between the results from postoperative evaluative modalities and clinical outcome.
Patients and Methods
Our series consisted of 410 patients younger than 18 years who
underwent surgery for MMD at the Division of Pediatric Neurosurgery, Seoul National University Children’s Hospital between 1988 and 2006.
We enrolled those with a confirmed diagnosis of MMD
and who underwent surgery for bilateral MMD (n ⫽ 357) or
unilateral MMD (n ⫽ 53). Patients who associated with another
condition such as Down syndrome or neurofibromatosis (moyamoya syndrome) were excluded from this study. This study was
approved by the Seoul National University Hospital Institutional Review Board.
All children had their diagnoses confirmed preoperatively
with the use of either computed tomography (CT) (n ⫽ 5 patients) or magnetic resonance imaging (MRI) (n ⫽ 405 patients), and all but 1 patient underwent conventional angiography to confirm the diagnosis of MMD. In 1 patient,
angiography was omitted due to a hypersensitive reaction to the
contrast agent, so the diagnosis of MMD was made based on
MRI and magnetic resonance angiography (MRA). The preoperative angiographic stage was evaluated according to Suzuki’s
classification (n ⫽ 818 hemispheres).1 Patients with definite
moyamoya vessel change in 1 hemisphere whose contralateral
hemisphere showed normal (n ⫽ 33) or mild focal narrowing of
the anterior cerebral artery (ACA) only (n ⫽ 20) were considered as unilateral MMD.
Perfusion MRI was performed using a gradient echo
single-shot echo-planar image sequence, following an intravenous bolus injection of 0.1mmol/kg gadopentetate dimeglumine
(Magnevist; Berlex Laboratories, Wayne, NJ), as previously described.12 The acquired data were analyzed in 2 ways: by regional cerebral blood volume (rCBV) and by time to peak
(TTP) images.
Basal (n ⫽ 396 patients) and acetazolamide stress (n ⫽ 346
patients) brain perfusion single-photon emission computerized tomography (SPECT) with 99mTc-hexamethylpropyleneaminoxime
was performed to evaluate hemodynamic status, as previously described.13 Images were analyzed for the severity and extent of hemodynamic abnormality. Severity of perfusion abnormality was
classified as a defect (perfusion same as or below that of white
matter) or mild to severe decrease (compared with that of contralateral gray matter). Perfusion defect is suggestive of cortical
infarction. Vascular reserve was assessed by comparison of basal
and acetazolamide SPECT.
Electroencephalography (EEG) was performed in 333 patients using the International 10-20 electrode system, and activation waking records produced by hyperventilation for 3 minutes were analyzed (n ⫽ 232 patients). We evaluated
neurocognitive function in 258 patients and intelligence quotient (IQ) in 276 patients.
Before October 1995, most patients had been managed by
simple EDAS (n ⫽ 91 patients). Subsequently, EDAS with bi-
July, 2010
frontal EGPS was the main surgical treatment (n ⫽ 319 patients). EDAS using the occipital artery was performed selectively for patients with posterior cerebral artery (PCA)
involvement (n ⫽ 44 patients). The follow-up period ranged
from 5 days to 236 months (mean, 61 months).
The overall clinical outcomes were divided into 4 categories4: (1) excellent, where the preoperative symptoms (such as
TIAs or seizures) had totally disappeared without fixed neurological deficits; (2) good, where the symptoms had totally disappeared but the mild neurological deficits remained; (3) fair,
where the symptoms persisted albeit less frequently; and (4)
poor, where the symptoms remained unchanged or worsened.
The postoperative CT/MRI (n ⫽ 393 patients; mean, 44
months; range, 1–205 months) findings were compared with the
preoperative findings and classified into 2 categories.4 In the favorable category, the CT or MRI showed no infarctions in either the pre- or postoperative imaging or no further increase of
the infarction. In the unfavorable category, neuroimaging
showed an increase in the infarct size or the formation of a new
lesion.
The follow-up angiogram was done after the EDAS operation in 526 hemispheres, after the bifrontal EGPS operation in
494 hemispheres, and after the occipital EDAS in 29 hemispheres (Fig). The development of collateral circulation of the
middle cerebral artery (MCA) territory (n ⫽ 526 hemispheres;
mean, 25 months; range, 1–192 months) through the bypass
was graded according to the system described by Matsushima et
2
al,14 where a good score represented revascularization of 3 of the
MCA distribution, a fair score represented revascularization of
1
2
between 3 and 3 of the MCA distribution, and a poor score
represented slight or no revascularization.
The collateral circulation of the ACA territory (n ⫽ 494
hemispheres; mean, 26 months; range, 1–168 months) was evaluated in a similar fashion.15 A good score represented revascu2
larization of 3 of the ACA distribution, a fair score represented
1
2
revascularization of between 3 and 3 of the ACA distribution,
and a poor score represented slight or no revascularization.
The collateral circulation of the PCA territory (n ⫽ 29
hemispheres; mean, 28 months; range, 3– 83 months) was
graded with 2 scores. A good score represented revascularization
1
of ⬎2 of the PCA distribution, and a poor score represented
1
revascularization of ⬍2 of the PCA distribution, including no
revascularization.
Postoperative perfusion MRI (n ⫽ 185 patients; mean, 36
months; range, 6 –106 months) was compared with the preoperative findings and classified into 2 groups.4 The favorable
group included those patients whose perfusion showed no hemodynamic abnormalities either pre- or postoperatively, or a decrease in size or number of the hemodynamic abnormalities.
The unfavorable classification included those patients whose perfusion MRI showed no change in size or number of hemodynamic abnormalities or new hemodynamic abnormalities. The
hemodynamic abnormalities included increased or decreased
rCBV and delayed TTP.
Postoperative SPECT findings of the whole brain (n ⫽
355 patients; mean, 50 months; range, 2–205 months) were
93
ANNALS
of Neurology
FIGURE: Representative angiographies of the external
carotid artery. (A, B) Good revascularization of 32 of the anterior cerebral artery distribution after bifrontal encephalogaleo-/periosteal synangiosis. (C, D) Good revascularization of 32
of the middle cerebral artery distribution after encephaloduroarteriosynangiosis (EDAS) using the superficial temporal artery. (E, F) Good revascularization of >21 of the posterior cerebral artery distribution after EDAS using the occipital
artery. Left panel (A, C, E): preoperative angiographies; right
panel (B, D, F): postoperative angiographies.
compared with the preoperative findings and classified into 1 of
2 groups.4 The favorable classification included those patients
whose SPECT showed no hemodynamic abnormalities either
pre- or postoperatively, or a decrease in the severity or extent of
hemodynamic abnormalities. The unfavorable classification included those patients whose SPECT showed no change in the
severity or extent of hemodynamic abnormalities or new hemodynamic abnormalities. The hemodynamic abnormalities included perfusion defect, decreased perfusion, and decreased vascular reserve on acetazolamide SPECT.
Postoperative neuropsychological (NP) tests were performed in 106 patients (mean, 44 months; range, 2–176
months), and the patients were categorized into 2 groups: ⬎90
and ⬍90 full IQ.
Operative Technique
The surgery was usually performed in 2 stages, initially in the
symptomatic and hemodynamically affected hemisphere. The
94
average interval between the first and subsequent operations was
6 months (range, 1–50 months). A third operation was performed selectively to improve PCA or ACA territory hemodynamics (n ⫽ 21 patients; mean, 42 months; range, 3–122
months).
To obtain the collateral formation in the MCA territory,
EDAS was performed.4,16 A sigmoid scalp incision was made
along the course of the superficial temporal artery (STA), usually
along the parietal branch of the STA (STAp). The STAp with
the strip of galea was freed from the subcutaneous fat and pericranium. After incision of the temporal muscle and periosteum
and creation of a bone flap, a linear dural incision was made
along the course of the STAp. The entire dural layer was folded
into the subdural space, leaving the major meningeal artery intact. The arachnoid membrane over the cortical sulci was dissected as widely as possible to promote the ingrowth of the neovasculature. The STAp galeal flap laid on the exposed cortex was
sutured to the folded ridge of the dura mater.
To obtain the collateral formation in the ACA territory,
bifrontal EGPS was performed and combined with EDAS surgery.4 The scalp was incised separately for EDAS and EGPS. At
the EGPS site, an S-shaped incision was made in the scalp 2cm
anterior to the coronal suture. The galea or the pericranium was
dissected and incised in a 傺-shape based laterally. A 4 ⫻ 8cm
craniotomy was made crossing the superior sagittal sinus. The
dura was incised separately on both hemispheres, and the arachnoid membrane was incised. The prepared galeo-/periosteal flap
laid on the exposed cortex was sutured to the incised edge of the
dura mater.
To obtain the collateral formation in the PCA territory,
EDAS using the occipital artery was performed.16 A sigmoid
scalp incision was made along the course of the occipital artery.
The occipital artery with a strip of galea was freed from the
subcutaneous fat and pericranium. After creation of occipital
bone flap, a linear dural incision was made along the middle of
the long axis. The entire dural layer was folded into the subdural
space. The arachnoid membrane over the cortical sulci was dissected. The occipital artery/galeal flap laid on the exposed cortex
was sutured to the folded ridge of the dura mater. All surgical
procedures were performed by 4 neurosurgeons.
Statistical Analysis
The statistical analysis was done on the patient base, but characteristics of angiographic studies were described on the hemisphere base. Clinical features of the study participants were presented as the proportion of patients for the given category of
characteristics. The clinical outcome was dichotomized into favorable (excellent and good outcomes) and unfavorable (fair and
poor outcomes). We used the logistic regression model to estimate the impact of preoperative and operative clinical factors on
the surgical outcome. In the multiple logistic regression model,
we included variables with p ⬍ 0.1 in the univariate analysis,
but perfusion MRI, EEG, and NP data were excluded from
analysis, because they were only available in a limited number of
patients. To control the problem of multicollinearity, we used
the forward stepwise selection method for the preoperative pa-
Volume 68, No. 1
Kim et al: Pediatric Moyamoya Disease
rameters. We calculated the Spearman correlation coefficients
between clinical outcome and other postoperative examination
results. All p values are 2-sided, and p values ⬍0.05 are considered statistically significant. SAS 9.1.3 (SAS Institute Inc., Cary,
NC) was used for the statistical analysis.
TABLE 1: Clinical Features of 410 Children with
Moyamoya Disease at Presentation
Clinical Features
Number of Cases (%)
Age, yr (n ⴝ 410 patients)
0-3
63 (15%)
4-6
109 (27%)
Results
7-18
238 (58%)
Clinical Features
Table 1 shows the clinical features of our patients at their
presentation. Their age at the first operation was between
7 months and 18 years (mean, 7.3 years). When patients
were categorized into 3 groups according to their age at
surgery, 63 patients (15%) were younger than 3 years,
109 patients (27%) were between the ages of 4 and 6
years, and 238 patients (58%) were older than 6 years.
Sex ratio (girl to boy) of the patients was 1:3. The incidence of familial occurrence was 12% (51/410). Twentyfour of 53 (45%) unilateral MMD cases showed progression to bilateral involvement (mean, 23 months; range,
0.5–150 months).
TIA was the most common clinical manifestation
(56%, Table 2). The others were infarction (39%), headache (33%), seizure (19%), involuntary movement (4%),
and hemorrhage (3%). A headache was defined as the
presence of severe headache that disturbed daily activity
requiring rest or medication, and which occurred at least
once a month.17 The interval between the onset of clinical symptoms and surgery averaged 21 months (range, 1
week to 108 months).
The CT/MRI showed evidence of infarction in 53%
of the patients (major lobar infarction, 39%; border zone
infarction, 14%) and hemorrhage in 3%. Forty-four percent of the patients showed neither infarction nor hemorrhage. Preoperative angiography demonstrated various
stages, but most of the hemispheres were between stages 2
and 4. In patients with unilateral involvement, the unaffected hemispheres demonstrated normal or mild ACA
narrowing only.
Perfusion MRI demonstrated areas of increased
rCBV in 53% of the patients, decreased rCBV in 16%,
and delayed TTP in 89%. SPECT revealed areas of perfusion defect in 23% of the patients, and areas of decreased perfusion in 60%. Acetazolamide SPECT demonstrated a decreased vascular reserve in 84% of the patients.
In patients with unilateral involvement, the unaffected
hemispheres showed normal hemodynamic status.
The re-buildup phenomenon was observed in 32%
of the patients who underwent activation waking records
by hyperventilation (74/232).
Sex (n ⴝ 410 patients)
July, 2010
Male
178 (43%)
Female
232 (57%)
Clinical manifestation (n ⴝ 410 patients)
Transient ischemic attacks
229 (56%)
Infarction
160 (39%)
Headache
134 (33%)
Seizure
76 (19%)
Involuntary movement
17 (4%)
Hemorrhage
11 (3%)
CT/MRI (n ⴝ 410 patients)
No infarction
182 (44%)
Infarction
Border zone infarct
59 (14%)
Major infarct
158 (39%)
Hemorrhage
11 (3%)
Angiographic stagesa (n ⴝ 818 hemispheres)
I
54 (7%)
II
180 (22%)
III
371 (44%)
IV
120 (15%)
V
58 (7%)
VI
2 (1%)
Normal
33 (4%)
Perfusion MRI (n ⴝ 237 patients)
Normal
18 (8%)
Increased rCBV
126 (53%)
Decreased rCBV
39 (16%)
Delayed TTP
212 (89%)
SPECT (n ⴝ 396 patients)
Perfusion defect
92/396 (23%)
Decreased perfusion
239/396 (60%)
Decreased reserve
291/346 (84%)
Posthyperventilation EEG (n ⴝ 333 patients)
Re-buildup phenomenon (⫹)
74 (22%)
Re-buildup phenomenon (⫺)
158 (48%)
No hyperventilation trial
101 (30%)
Neurocognitive function test (n ⴝ 258 patients)
Normal
19 (7%)
Abnormal
239 (93%)
IQ (n ⴝ 276 patients)
ⱖ90
209 (76%)
⬍90
67 (24%)
a
Angiographic stages were evaluated according to the
criteria of Suzuki angiographic stage.1 CT ⫽ computed
tomography; MRI ⫽ magnetic resonance imaging;
rCBV ⫽ regional cerebral blood volume; TTP ⫽ time to
peak; SPECT ⫽ single-photon emission computerized
tomography; EEG ⫽ electroencephalography; IQ ⫽
intelligence quotient.
95
ANNALS
of Neurology
TABLE 2: Summary of Ischemic Symptoms
Territory
Middle cerebral artery
Anterior cerebral artery
Posterior cerebral artery
Typical Symptoms
Motor-sensory, speech dysfunction
Paraparesis, urinary incontinence
Visual symptoms
% of Patients
At Presentation
Newly Developed
after the First
Operation
92
52
10
3
7
8
TABLE 3: Surgical Outcomes of 410 Children with Moyamoya Disease
Parameters
Overall clinical outcome (n ⴝ 410 patients)
Excellent
Good
Fair
Poor
MRI data (n ⴝ 393 patients)
Favorable
Unfavorable
Extent of revascularization of MCA territory (n ⴝ 526 hemispheres)
Good
Fair
Poor
Extent of revascularization of ACA territory (n ⴝ 494 hemispheres)
Good
Fair
Poor
Extent of revascularization of PCA territory (n ⴝ 29 hemispheres)
Good
Poor
Perfusion MRI data (n ⴝ 185 patients)
Favorable
Unfavorable
SPECT changes (n ⴝ 355 patients)
Favorable
Unfavorable
Neuropsychological test (n ⴝ 106 patients)
IQ ⱖ90
IQ ⬍90
Number of Cases (%)
270 (66)
63 (15)
62 (15)
15 (4)
295 (75)
98 (25)
177 (34)
294 (56)
55 (10)
216 (44)
219 (44)
59 (12)
20 (69)
9 (31)
121 (65)
64 (35)
214 (60)
141 (40)
71 (67)
35 (33)
MRI ⫽ magnetic resonance imaging; MCA ⫽ middle cerebral artery; ACA ⫽ anterior cerebral artery; PCA ⫽ posterior cerebral
artery; SPECT ⫽ single-photon emission computerized tomography; IQ ⫽ intelligence quotient.
96
Volume 68, No. 1
Kim et al: Pediatric Moyamoya Disease
TABLE 4: Univariate Logistic Regression Analysis
for Predictive Factors of Unfavorable Surgical
Outcome
Characteristics
OR
95% CI
pa
Age at onsetb
0.89
0.83–0.96
0.003
TABLE 4: Continued.
Characteristics
95% CI
pa
Increased rCBV
Sex
No
1.00
Yes
0.46
0.05
0.21–1.01
Delayed TTPc
Female
1.00
Male
1.04
0.88
0.63–1.71
Presentation data
No
1.00
Yes
—
—
Angiographic data
Transient ischemic attacks
Suzuki grade
Absent
1.00
Present
0.24
⬍0.001
0.14–0.42
ⱕ3
1.00
ⱖ4
1.10
0.79
0.56–2.16
SPECT data
Infarction
Absent
1.00
Present
4.33
⬍0.001
Perfusion defect
2.55–7.36
Headache
Absent
1.00
Present
0.45
Absent
1.00
Present
2.02
0.24–0.82
1.00
Present
1.35
1.14–3.59
1.00
Present
2.55
0.43–4.26
1.00
Yes
0.74
1.00
Yes
1.30
0.46
1.00
Yes
0.89
0.30
1.00
4.14
0.66
0.52–1.50
1.00
Present
0.20
⬍0.001
2.45–7.01
0.008
0.06–0.66
Hemorrhage
Absent
1.00
Present
2.55
0.14
Decreased rCBV
No
1.00
6.96
1.00
Yes
0.76
0.48
0.36–1.63
No
1.00
Yes
0.04
0.002
0.01–0.31
No
1.00
Yes
0.34
0.04
0.15–0.79
Normal
1.00
Abnormal
1.46
0.62
0.32–6.59
No
1.00
Yes
4.09
⬍0.001
2.06–8.11
p Values are calculated from the model that excluded
unmeasured observations. bAge of onset is included in the
model as a continuous variable, whereas all the other
variables are categorized considering clinical relevance. cIn
patients without delayed TTP, none had unfavorable
outcome. Therefore, odds ratio and p value could not be
evaluated. OR ⫽ odds ratio; CI ⫽ confidence interval;
MCA ⫽ middle cerebral artery; MRI ⫽ magnetic
resonance imaging; rCBV ⫽ regional cerebral blood
volume; TTP ⫽ time to peak; SPECT ⫽ single-photon
emission computerized tomography; EEG ⫽
electroencephalography; IQ ⫽ intelligence quotient.
0.73–8.95
Perfusion MRI data
Yes
No
a
Border zone infarction
Absent
0.09
0.93–2.79
IQ <90
0.79–2.15
Major infarction
Absent
1.62
Cerebral function test
MRI data
Present
1.00
Yes
Neuropsychological test
0.34–1.65
MCA symptoms only
No
No
Re-buildup
0.73–8.95
Symptoms >12 months
No
⬍0.001
1.76–5.22
EEG data
0.14
Unilateral symptoms
No
3.03
Decreased reserve only
0.61
Hemorrhage
Absent
1.00
Yes
Decreased reserve
0.02
Involuntary movement
Absent
No
Decreased perfusion
0.009
Seizures
July, 2010
OR
⬍0.001
3.01–16.09
Of 258 patients who were evaluated for neurocognitive function, 93% had cognitive dysfunction. On the
other hand, 76% of the patients were mentally normal
(ⱖ90) by IQ determination (n ⫽ 276 patients).
97
ANNALS
of Neurology
TABLE 5: Multiple Logistic Regression for Unfavorable Clinical Outcome from Preoperative Characteristics of
345 Patients
Characteristics
Favorable
Outcome
No.
%
Unfavorable
Outcome
No.
%
Onset agea
Sex
F
M
Infarction
No
Yes
Decreased reserve only
No
Yes
OR
0.97
168
123
57.7
42.3
32
22
59.3
40.7
95% CI
p
LL
UL
0.88
0.44
1.07
1.53
0.54
0.53
1.49
5.46
0.002
0.01
0.52
0.009
1
0.82
206
85
70.8
29.2
19
35
35.2
64.8
1
2.85
201
90
69.1
30.9
53
1
98.1
1.9
1
0.07
a
Onset age is included to the model as a continuous variable, all the other variables are categorized considering clinical relevance
OR ⫽ odds ratio; CI ⫽ confidence interval; LL ⫽ lower level; UL ⫽ upper level.
Surgical Outcomes
The surgical outcomes are summarized in Table 3.
The overall mortality rate was 0.5% (2/410). A
2-year-old girl died 5 days after EDAS of severe intraoperative brain swelling and extensive cerebral infarction immediately after the operation. This case must be considered a surgery-related failure and emphasized the
importance of perioperative management to avoid hemodynamic instability. A 20-year-old boy died 124 months
after EDAS and bifrontal EGPS of acute infarction in the
left temporoparieto-occipital area secondary to PCA territory insufficiency. This case strongly suggests that careful
and long-term follow-up is essential in pediatric MMD.
Postoperative infarctions of variable size were the
most clinically relevant complication and occurred in 54
cases (13% per patient; 6% per operation) after 845 operations (mean, 2 days; range, 0 –14 days). Notably, 67%
of them improved to normal neurological status during
follow-up. The other complications were: wound problems (6%), epidural hematoma (4%), subdural hematoma
(2%), angiography-associated morbidity (1%), seizure
(1%), and intraparenchymal hemorrhage (0.5%). Ninetytwo percent of the patients had a Karnofsky performance
status (KPS) score of ⬎70 on their most recent follow-up
examination.
The overall clinical outcome was excellent in 66%,
good in 15%, fair in 15%, and poor in 4% of the patients. Therefore, 81% of the patients had a favorable
clinical outcome (excellent and good). The period from
98
surgery to the disappearance of the TIA symptoms was
between 0 and 87 months (mean, 4.3 months).
Seizure control was excellent. Ninety-seven percent
(74/76 patients) of the patients were seizure free (antiepileptic drugs were successfully withdrawn in 91% 67⁄74 of
them), 1 had rare seizures, and only 1 experienced persistent seizures. Preoperative headache disappeared in 72%
of the patients (97/134 patients; mean, 16 months; range,
0 – 87 months). However, headache newly developed in
16% of the 276 patients without preoperative headache
during follow-up.
After surgery, 75% of the patients had favorable
MRI changes. Revascularization after indirect bypass surgery was excellent; good or fair filling of the MCA territory was achieved in 90% of the patients after EDAS surgery, of the ACA territory in 88% after bifrontal EGPS,
and of the PCA territory in 69% after occipital EDAS.
Improved cerebral hemodynamics was confirmed using perfusion MRI and SPECT; perfusion MRI showed a
favorable outcome in 65% of the patients, and SPECT
demonstrated the same outcome in 60% of the patients.
Follow-up NP testing showed that 67% of the patients were mentally normal (ⱖ90) by IQ examination.
Predictors of Surgical Outcome
Univariate logistic regression analyses of the preoperative
clinical variables showed that older age of symptom onset
(odds ratio [OR], 0.89/year; 95% confidence interval
[CI], 0.83– 0.96; p ⬍ 0.01), presence of TIA (OR, 0.24;
Volume 68, No. 1
Kim et al: Pediatric Moyamoya Disease
95% CI, 0.14 – 0.42; p ⬍ 0.001) or headache (OR, 0.45;
95% CI, 0.24 – 0.82; p ⬍ 0.01), border zone infarction
on MRI (OR, 0.20; 95% CI, 0.06 – 0.66; p ⬍ 0.01), increased rCBV on perfusion MRI (OR, 0.46; 95% CI,
0.21–1.01; p ⫽ 0.05), decreased vascular reserve only
with normal basal perfusion on SPECT (OR, 0.04; 95%
CI, 0.01– 0.31; p ⬍ 0.01), and re-buildup phenomenon
on EEG (OR, 0.34; 95% CI, 0.15– 0.79; p ⫽ 0.04) were
associated with a favorable clinical outcome. On the contrary, the presence of infarction (OR, 4.33; 95% CI,
2.55–7.36; p ⬍ 0.001), seizure (OR, 2.02; 95% CI,
1.14 –3.59; p ⫽ 0.02), major infarction on MRI (OR,
4.14; 95% CI, 2.45–7.01; p ⬍ 0.001), decreased rCBV
on perfusion MRI (OR, 6.96; 95% CI, 3.01–16.09; p ⬍
0.001), perfusion defect on SPECT (OR, 3.03; 95% CI,
1.76 –5.22; p ⬍ 0.001), and IQ ⬍90 (OR, 4.09; 95%
CI, 2.06 – 8.11; p ⬍ 0.001) were identified as predictors
of unfavorable clinical outcome (Table 4).
Multivariate logistic regression analyses (Table 5) revealed that infarction on presentation was associated with
unfavorable clinical outcome (OR, 2.85; 95% CI, 1.49 –
5.46; p ⬍ 0.01), and decreased vascular reserve only on
SPECT (OR, 0.07; 95% CI, 0.01– 0.52; p ⬍ 0.01) was
associated with favorable clinical outcome.
We separately analyzed surgery-related variables on
the clinical outcome. The occurrence of postoperative infarction as a complication was associated with unfavorable
clinical outcome (OR, 5.66; 95% CI, 3.10 –10.35; p ⬍
0.01). The addition of bifrontal EGPS showed a protective effect toward favorable clinical outcome (OR, 0.48;
95% CI, 0.28 – 0.82; p ⫽ 0.01).
Evaluative Modalities During Follow-up
The clinical outcome and various evaluative modalities
showed high degrees of correlation. The KPS score is an
equivalent of clinical examination and showed the highest
correlation with the clinical outcome (␳ ⫽ ⫺0.62, p ⬍
0.01). MRI change (␳ ⫽ 0.35, p ⬍ 0.01), perfusion MRI
outcome (␳ ⫽ 0.35, p ⬍ 0.01), and SPECT outcome
(␳ ⫽ 0.25, p ⬍ 0.01) showed high degrees of correlation
with the clinical outcome. Progressive PCA narrowing on
postoperative angiography was also strongly correlated
with an unfavorable clinical outcome (␳ ⫽ 0.25, p ⬍
0.01). However, degree of MCA revascularization on
postoperative angiography was only modestly correlated
with an unfavorable clinical outcome (␳ ⫽ ⫺0.11, p ⫽
0.03), and degree of ACA revascularization on postoperative angiography was not correlated with an unfavorable
clinical outcome (␳ ⫽ ⫺0.04, p ⫽ 0.57).
July, 2010
Discussion
This study is among the largest series focused on digitized
information of clinical manifestation and preoperative or
surgically related variables predictive of clinical outcome
in children with MMD. Correlations between results
from postoperative evaluative modalities and clinical outcome were also verified.
The large number of cases and long-term follow-up
data in this series allow us to establish management guidelines for MMD. This study is also useful for counseling
families with MMD children.
Digitized Information of Clinical Manifestation
Most cases of MMD appear to be sporadic, but 12% of
cases in this series are familial. Therefore, MRI and MRA
appear to be mandatory to screen family members of
MMD patients who show suspicious clinical symptoms
such as TIA, headache, and seizure. Although unilateral
MMD has been recognized as stable in adults,18 the occlusive lesions in the carotid forks frequently progress in
pediatric patients. Indeed, some cases of unilateral MMD
in children might be an early form of definite MMD. In
45% of patients, the unilateral MMD at the initial diagnosis progressed to bilateral involvement 2 years later.
The time interval from unilateral to bilateral MMD is
noteworthy: disease progression can occur 150 months after the initial diagnosis. Therefore, careful long-term
follow-up is essential to monitor potential progression of
unilateral MMD.
In addition to hemodynamically associated symptoms such as TIA and infarction, a substantial portion of
patients suffered from headaches (33%) and seizures
(19%). Each symptom showed striking differences in response to bypass surgery. Compared with favorable seizure outcome (97% seizure free) after surgery, headache
remained a challenging problem to neurosurgeons17,19;
headache persisted in 26% of the patients who had a preoperative headache and newly developed in 16% of the
patients. Not only cerebral ischemia, but also progressive
recruitment and redistribution of blood flow might be a
cause of headaches in MMD patients.17
MCA symptoms were most common (95%), but
ACA symptoms were identified in 52% and PCA symptoms in 18% of patients. Based on the progressive nature
of ischemia, the number of affected patients, and risk of
irreversible ischemic change, revascularization of the ACA
or PCA territory should be considered in pediatric patients with MMD who have symptoms or radiological evidence of hemodynamic insufficiency in those territories.
99
ANNALS
of Neurology
Preoperative Variables Predictive of
Clinical Outcome
Known prognostic factors for pediatric MMD outcome
after revascularization operation are preoperative multiple
cerebral infarctions, early onset at a young age, high Suzuki stages on cerebral angiography, the surgical procedure itself, and perioperative complications such as ischemic events.10,20 –23
Although many clinical factors were related with clinical outcome on the univariate analysis, those factors, including age, clinical symptoms, neuroimaging findings, and
electrophysiological and neuropsychological data, are essentially focused on a single factor, that is, reversibility of hemodynamic ischemia. Patients with reversible ischemia
achieved a favorable clinical outcome. They shared the following common features: ⬎3 years old (OR, 0.89 per
year),23 TIA on presentation (OR, 0.24), increased rCBV
on perfusion MRI (OR, 0.46),12 decreased vascular reserve
(OR, 0.76) or decreased vascular reserve only with normal
basal perfusion (OR, 0.04) on SPECT (OR, 0.76),24,25 and
re-buildup phenomenon on EEG (OR, 0.34).26,27 On the
other hand, we identified characteristics suggesting that irreversible infarcts showed a greater risk of unfavorable clinical outcome. In addition to infarction on presentation
(OR, 4.33), seizure, which might be more common in patients with infarcts, demonstrated increased risk for unfavorable clinical outcome (OR, 2.02). Not only major infarction on MRI (OR, 4.14) for structural imaging, but
also hemodynamic compromises on perfusion MRI or
SPECT indicating infarction, harbored increased risk for an
unfavorable clinical outcome (decreased rCBV [OR, 6.96],
perfusion defect [OR, 3.03]). Furthermore, patients with
an IQ ⬍90, suggesting cognitive end results of infarction,
revealed an unfavorable clinical outcome (OR, 4.09).
Multivariate analyses confirmed the importance of
the reversibility of hemodynamic ischemia; infarction on
presentation had an approximately 3-fold increased risk
for unfavorable clinical outcome, whereas decreased vascular reserve only with normal basal perfusion on SPECT
harbored a 14-fold increased protective effect for favorable
clinical outcome. The present result, therefore, strongly
suggests that diagnosis and therapeutic intervention as
early as possible would reduce the incidence of completed
stroke and may improve clinical outcome.
Surgery-Related Variables Predictive of
Clinical Outcome
This series suggests that the extent of surgical procedure
may have some influence on clinical outcome. The favorable clinical outcome by the addition of bifrontal EGPS
might be related to epoch, because it was commenced at a
later date, and general perioperative management may have
100
been improved. On the other hand, it has been known that
indirect procedures such as EDAS and encephalomyosynangiosis, widely performed in patients with MMD, have
the disadvantages of limited revascularization confined to
the craniotomy field.28 In addition, chronic blood flow reduction in the frontal lobe may be responsible for poor
intellectual outcome.28 Therefore, the revascularization procedure over as wide an area as possible, including ACA territory, may be essential to improve not only clinical outcome, but also intellectual outcome.
Surgery-related infarctions were the most common
complication, and occurred within 2 weeks of the operation. They were closely related to an unfavorable outcome
(OR, 5.66). Therefore, careful attention should be paid to
maintaining adequate hemodynamic status (normotensive,
normocapnic, and normal hemoglobin level) and pain control to avoid hyperventilation during the perioperative period.29 –31
Correlations between Postoperative Evaluative
Modalities and Clinical Outcome
The time interval between surgery and the disappearance
of symptoms is noteworthy. The main surgical procedure
in our institute is indirect bypass surgery; therefore, symptoms could persist. Similar to previous reports,31 symptoms usually disappeared at an average time of 4.3
months. Interestingly, some patients showed a beneficial
effect immediately after operation.
MRI, perfusion MRI, and SPECT were proven to
have a close relationship with clinical outcome in this
study. Favorable changes in neuroimaging reflected favorable clinical outcome. Patients with a decreased cerebrovascular reserve will have remaining neurological deficit
and ischemic attacks on follow-up.13
The effect of revascularization surgeries on the reduction in anterior circulation of abnormal collateral vessels
seemed to be definite.32 However, their effect on the reduction in abnormal posterior circulation collateral vessels
remains uncertain.32 Notably, progression of PCA narrowing with insufficient reduction of the abnormal posterior
circulation collateral formations on follow-up angiogram
were highly correlated with an unfavorable outcome. PCA
territory, usually uncovered by initial operations, might be
involved during natural disease progression. Therefore,
close observation of PCA status and active intervention appears to be necessary after revascularization surgery on anterior circulation.33
There are certain limitations to this study that
should be noted. First, the present study is a nonrandomized historical cohort study. However, it is ethically undesirable to perform randomized clinical trials to confirm
the beneficial effects of surgical revascularization on subVolume 68, No. 1
Kim et al: Pediatric Moyamoya Disease
13.
So Y, Lee HY, Kim SK, et al. Prediction of the clinical outcome
of pediatric moyamoya disease with postoperative basal/
acetazolamide stress brain perfusion SPECT after revascularization surgery. Stroke 2005;36:1485–1489.
14.
Matsushima T, Inoue T, Suzuki SO, et al. Surgical treatment of
moyamoya disease in pediatric patients— comparison between
the results of indirect and direct revascularization procedures.
Neurosurgery 1992;31:401– 405.
15.
Robertson RL, Burrows PE, Barnes PD, et al. Angiographic
changes after pial synangiosis in childhood moyamoya disease.
AJNR Am J Neuroradiol 1997;18:837– 845.
16.
Matsushima Y, Inaba Y. Moyamoya disease in children and its
surgical treatment. Introduction of a new surgical procedure and
its follow-up angiograms. Childs Brain 1984;11:155–170.
This study was supported by the Korea Healthcare Technology R&D Project, Ministry for Health, Welfare and
Family Affairs, Republic of Korea (grant A080588-8 to
S-K.K.) and an Seoul Broadcasting System Grant-in-Aid
for Seoul National University Children’s Hospital research (grant 06-2009-018-0 to S-K.K.).
17.
Seol HJ, Wang KC, Kim SK, et al. Headache in pediatric moyamoya disease: review of 204 consecutive cases. J Neurosurg
2005;103(5 suppl):439 – 442.
18.
Kuroda S, Ishikawa T, Houkin K, et al. Incidence and clinical
features of disease progression in adult moyamoya disease.
Stroke 2005;36:2148 –2153.
19.
Scott RM, Smith JL, Robertson RL, et al. Long-term outcome in
children with moyamoya syndrome after cranial revascularization by
pial synangiosis. J Neurosurg 2004;100(2 Suppl Pediatrics):142–149.
Potential Conflicts of Interest
20.
Kurokawa T, Tomita S, Ueda K, et al. Prognosis of occlusive
disease of the circle of Willis (moyamoya disease) in children.
Pediatr Neurol 1985;1:274 –277.
21.
Fukuyama Y, Umezu R. Clinical and cerebral angiographic evolutions of idiopathic progressive occlusive disease of the circle of
Willis (“moyamoya” disease) in children. Brain Dev 1985;7:21–37.
22.
Karasawa J, Touho H, Ohnishi H, et al. Long-term follow-up study after
extracranial-intracranial bypass surgery for anterior circulation ischemia
in childhood moyamoya disease. J Neurosurg 1992;77:84–89.
sequent ischemic stroke in MMD, because a benefit of
surgery for the ischemic type of MMD has been established.4 –11 Second, because of the insufficient postoperative neuropsychological assessment and quality-of-life
evaluation, we have no definite data on those topics in
our patients. Future studies using standardized postoperative neuropsychological testing and quality-of-life assessment will be necessary to clarify these issues.
Acknowledgments
Nothing to report.
References
1.
Suzuki J, Kodama N. Moyamoya disease—a review. Stroke 1983;
14:104 –109.
2.
Kuriyama S, Kusaka Y, Fujimura M, et al. Prevalence and clinicoepidemiological features of moyamoya disease in Japan: findings from
a nationwide epidemiological survey. Stroke 2008;39:42– 47.
23.
Kim SK, Seol HJ, Cho BK, et al. Moyamoya disease among young
patients: its aggressive clinical course and the role of active surgical
treatment. Neurosurgery 2004;54:840 – 844; discussion 844 – 846.
3.
Kim SK, Wang KC, Kim DG, et al. Clinical feature and outcome
of pediatric cerebrovascular disease: a neurosurgical series.
Childs Nerv Syst 2000;16:421– 428.
24.
4.
Kim SK, Wang KC, Kim IO, et al. Combined encephaloduroarteriosynangiosis and bifrontal encephalogaleo(periosteal)synangiosis in
pediatric moyamoya disease. Neurosurgery 2002;50:88 –96.
Touho H, Karasawa J, Ohnishi H. Preoperative and postoperative
evaluation of cerebral perfusion and vasodilatory capacity with
99mTc-HMPAO SPECT and acetazolamide in childhood moyamoya disease. Stroke 1996;27:282–289.
25.
Ikezaki K, Han DH, Kawano T, et al. A clinical comparison of
definite moyamoya disease between South Korea and Japan.
Stroke 1997;28:2513–2517.
Mountz JM, Foster NL, Ackermann RJ, et al. SPECT imaging of
moyamoya disease using 99mTc-HM-PAO. Comparison with computed tomography findings. J Comput Tomogr 1988;12:247–250.
26.
Imaizumi T, Hayashi K, Saito K, et al. Long-term outcomes of
pediatric moyamoya disease monitored to adulthood. Pediatr
Neurol 1998;18:321–325.
Kuroda S, Kamiyama H, Isobe M, et al. Cerebral hemodynamics
and “re-build-up” phenomenon on electroencephalogram in children with moyamoya disease. Childs Nerv Syst 1995;11:214 –219.
27.
Kodama N, Aoki Y, Hiraga H, et al. Electroencephalographic findings
in children with moyamoya disease. Arch Neurol 1979;36:16–19.
5.
6.
7.
Hirotsune N, Meguro T, Kawada S, et al. Long-term follow-up
study of patients with unilateral moyamoya disease. Clin Neurol
Neurosurg 1997;99(suppl 2):S178 –S181.
28.
Kuroda S, Houkin K, Ishikawa T, et al. Determinants of intellectual
outcome after surgical revascularization in pediatric moyamoya
disease: a multivariate analysis. Childs Nerv Syst 2004;20:302–308.
8.
Ueki K, Meyer FB, Mellinger JF. Moyamoya disease: the disorder
and surgical treatment. Mayo Clin Proc 1994;69:749 –757.
29.
9.
Ozgur BM, Aryan HE, Levy ML. Indirect revascularisation for paediatric moyamoya disease: the EDAMS technique. J Clin Neurosci 2006;13:105–108.
Kansha M, Irita K, Takahashi S, Matsushima T. Anesthetic management of children with moyamoya disease. Clin Neurol Neurosurg 1997;99(suppl 2):S110 –S113.
30.
Sainte-Rose C, Oliveira R, Puget S, et al. Multiple bur hole surgery for the treatment of moyamoya disease in children. J Neurosurg 2006;105(6 suppl):437– 443.
Nomura S, Kashiwagi S, Uetsuka S, et al. Perioperative management protocols for children with moyamoya disease. Childs Nerv
Syst 2001;17:270 –274.
31.
Kuroda S, Houkin K. Moyamoya disease: current concepts and
future perspectives. Lancet Neurol 2008;7:1056 –1066.
10.
11.
Golby AJ, Marks MP, Thompson RC, Steinberg GK. Direct and
combined revascularization in pediatric moyamoya disease. Neurosurgery 1999;45:50 –58.
32.
Goda M, Isono M, Ishii K, et al. Long-term effects of indirect bypass
surgery on collateral vessel formation in pediatric moyamoya disease. J Neurosurg 2004;100(2 Suppl Pediatrics):156 –162.
12.
Kim SK, Wang KC, Oh CW, et al. Evaluation of cerebral hemodynamics with perfusion MRI in childhood moyamoya disease.
Pediatr Neurosurg 2003;38:68 –75.
33.
Choi IJ, Hong SH, Cho BK, et al. Encephalo-duro-arteriosynangiosis (EDAS) using occipital artery in children with moyamoya disease. J Kor Neurosurg Soc 2005;38:413– 418.
July, 2010
101
Документ
Категория
Без категории
Просмотров
2
Размер файла
162 Кб
Теги
410, moyamoya, consecutive, analysis, case, disease, pediatrics
1/--страниц
Пожаловаться на содержимое документа