Keywords
moyamoya - direct bypass - indirect bypass - combined - revascularization
Introduction
Moyamoya vasculopathy is a rare cerebrovascular disease with a prevalence ranging
from 3.92 to 16.1 cases per 100,000 individuals and an incidence of 0.09 to 2.3 cases
per 100,000 individuals per year.[1] It is characterized by the progressive steno-occlusion of large intracranial arteries
along with the secondary development of small vascular collaterals with a characteristic
smoky appearance on angiography.[2]
[3]
[4] According to the latest guidelines from the Research Committee on Moyamoya Disease
(RCMD),[5] moyamoya disease (MMD) is idiopathic moyamoya vasculopathy in the absence of any
other risk factors for the arterial stenosis or occlusion, while moyamoya syndrome
(MMS) represents the same diagnostic criteria of MMD but linked with other comorbidities
that are associated with the vasculopathy, such as autoimmune disease, meningitis,
brain tumors including meningioma, craniopharyngioma, hemangioblastoma, and glioma,
head irradiation, neurofibromatosis type 1, Down syndrome, and sickle cell disease.[6] MMD or syndrome might present with a wide variety of heterogeneous manifestations
such as ischemic stroke, intracranial hemorrhage (ICH), seizures, and cognitive deterioration.[7] Neither neurointerventional approaches nor medications showed promising therapeutic
effects. Medical therapy and supportive treatment might help reducing the risk of
complications. Still, surgery is required in the vast majority of patients, especially
those with a high risk of future vascular events.[7] Neurosurgeons acknowledged the advantages of surgical revascularization. Currently,
studies suggest that combined bypass (direct and indirect) is better for moyamoya
patients as it increases blood flow from direct anastomosis and collaterals ingrowth
from indirect bypass.[7]
[8]
[9] However, postoperative complications, such as ischemia, infarction, hemorrhage,
and seizures, are frequent.[10]
[11]
[12]
[13] This article describes our experience in Hamad Medical Corporation center with combined
surgical revascularization of moyamoya cases over the past 7 years.
Materials and Methods
Study Design, Settings, and Eligibility Criteria
This was a retrospective record review study involving all adult patients with MMD
or MMS who were operated in Hamad Medical Corporation center in Qatar between 2015
and 2022. The Hamad Medical Corporation center is the largest and only neurosurgery
center in the Middle East and North Africa region that operates this disease on regular
patients. The inclusion criteria included adult patients, a definite diagnosis of
MMD or MMS, surgically operated cases, and who had follow-up data in their medical
records. Patients were candidates for surgery depending on the presence of symptoms
and radiological findings of magnetic resonance imaging/magnetic resonance angiography,
computed tomography (CT) angiography, and digital subtraction angiography (DSA) according
to the diagnostic criteria for MMD[5] ([Supplementary Table S1] [available in the online version). Patients who were managed conservatively (i.e.,
medical therapy only) and those with incomplete data or lost to follow-up were not
included in the study.
Surgical Management
A single surgeon and the same surgical techniques operated on all the patients. Prior
to surgery, all patients underwent DSA to document occlusion of the internal carotid
artery ([Fig. 1]). During the surgical procedure, both direct (external carotid internal carotid)
and indirect (encephaloduroarteriosynangiosis via placing the temporalis muscle over
the brain surface) revascularization techniques were used ([Fig. 2]). CT with volume rendering technique was conducted postoperatively to assess the
surgical results ([Fig. 3).]
Fig. 1 Diagnostic cerebral angiogram of the bilateral internal carotid, superficial temporal,
and external carotid arteries (ECAs). (A) Left internal carotid artery (ICA) occlusion; (B) fetal circulation from the right (Rt) vertebral artery; (C) total occlusion of the Rt ICA immediately distal to the posterior communicating
artery (PCOM) with an adequate filling of the Rt posterior cerebral artery (PCA) through
PCOM artery. Multiple hairline collaterals at the skull base that are filling the
Rt middle cerebral artery (MCA). Pial collaterals are seen contributing to the Rt
MCA territories from the Rt PCA branches. The per-callosal anastomosis refilling of
the Rt anterior cerebral artery (ACA) is noted from the PCA territories. There is
also markedly narrowing, and irregularity of the left supraclinoidal left ICA as well
as the origin of both the MCA and ACA, mounting to occlusion at the later. There is
total occlusion of Rt A1/A2 segments with pial collateral refilling noted during the
left PCA territories; (D and E) average caliber and smooth outline of both superficial temporal arteries during
ECA injections.
Fig. 2 (A) Intraoperative indocyanine green image showing the patency of the anastomosis. (B) Intraoperative image showing the revascularization technique.
Fig. 3 (A and B) Postoperative angiogram of external carotid artery demonstrating bilateral superficial
temporal artery to middle cerebral artery (STA-MCA) bypass with moderate focal stenosis
at the right (Rt) STA-MCA anastomosis. The left STA-MCA bypass appears patent. (C) No internal carotid artery (ICA) contribution from the radiotherapy or left ICA
is seen. (D) Postoperative computed tomography and a volume rendering technique showing opacification
of the left superficial temporal artery with anastomosis with the left MCA branches,
well opacification of the Rt superficial temporal artery and Rt MCA branches, significant
narrowing of the intracavernous and intracanalicular part bilateral ICAs.
Surgical Technique
The surgical technique that was used involves a combined bypass using the frontal
and parietal branches of the superficial temporal artery. After patient intubation
and shaving the surgical site, Doppler ultrasound is used to track the course of the
artery branches. A curvilinear incision is made to harvest maximum length, with considerations
for anatomical variations. Blunt and sharp dissection techniques are employed, and
early identification of the main artery stem is crucial. The parietal branch is dissected
first to prevent stretching, followed by the frontal branch. Coagulation is used for
tiny branches. A 2 mm galeoadventitial cuff is harvested with donor vessels, followed
by dissection and flushing. The temporalis muscle is split, and a small craniotomy
exposes the M4 branches while protecting the middle meningeal artery. Suitable branches
are identified, and arachnoid dissection is performed, sacrificing limited branches.
The donor vessel is prepared, and an anastomosis is performed using sutures ([Fig. 2B]). Doppler ultrasound and indocyanine green (ICG) confirm blood flow through the
bypass ([Fig. 2A]). The procedure is repeated for the second anastomosis, followed by the application
of fibrin glue. After completing the direct bypass, the dura is addressed by inverting
the proximal leaflets over the brain surface and obtaining two flaps from the temporalis
muscle. The distal dural leaflets are approximated and stay sutures are used to secure
the muscle in place, ensuring good hemostasis. The bone flap is cut in half to avoid
pressure on the doner artery and fixed with miniplates.
Data Collection
The patients' sociodemographic and clinical characteristics were obtained from the
medical records. Preoperative modified Rankin score (mRS) was also obtained. The intraoperative
indocyanine green (ICG) fluorescence, Doppler findings, postoperative mRS, and complications
were collected.
Statistical Analysis
All data obtained were fed into a computer and analyzed using International Business
Machines Statistical Package for the Social Sciences software version 26.0. Mean and
standard deviation were utilized to describe quantitative parametric variables. Frequency
and percentage were utilized to describe qualitative variables. The association between
initial clinical presentation and different outcome variables was studied using Fisher's
exact test. The change between pre- and postoperative mRS was assessed using Wilcoxon
signed rank test. p-values less than 0.05 were considered statistically significant.
Results
Demographics and Clinical Characteristics of the Included Patients
Twenty adults with moyamoya were operated in this study. Their mean age was 37.4 ± 10.26
years and the mean follow-up period was 13.6 months. Almost two-thirds were males
(60%), and most cases (85%) were Asian. None of the patients had a positive family
history of MMD, but a single patient had a family history of subarachnoid hemorrhage.
Transient ischemic attack (TIA) was the main presentation occurring in approximately
55% of the cases. Headache and ICH were the next two most common presentations occurring
in 25 and 20% of patients, respectively. The disease involved both hemispheres in
the vast majority of the cases (80%). Of all patients, 85% (n = 17) had MMD, and only 15% (n = 3) patients were diagnosed with MMS. The mean preoperative mRS score of the patients
was 2.9 ± 1.119. [Table 1] summarizes the demographics and clinical characteristics of the patients.
Table 1
Demographics and baseline characteristics of included patients
|
Variables
|
Total (N = 20)
|
|
Age (mean ± SD)
|
37.4 ± 10.262
|
|
Gender (n, %)
|
|
|
Male
|
12 (60%)
|
|
Female
|
8 (40%)
|
|
Region (n, %)
|
|
|
Asia
|
17 (85%)
|
|
Africa
|
3 (15%)
|
|
Family history (n, %)
|
|
|
No
|
19 (95%)
|
|
SAH
|
1 (5%)
|
|
Presentation (n, %)
|
|
|
TIA
|
11 (55%)
|
|
ICH
|
4 (20%)
|
|
IVH
|
3 (15%)
|
|
SAH
|
2 (10%)
|
|
Stroke
|
2 (10%)
|
|
Headache
|
5 (25%)
|
|
Dysphasia
|
2 (10%)
|
|
Confusion
|
2 (10%)
|
|
Dizziness
|
2 (10%)
|
|
Unsteadiness
|
1 (5%)
|
|
Dysarthria
|
2 (10%)
|
|
Vomiting
|
1 (5%)
|
|
Comorbidities (n, %)
|
|
|
No
|
13 (65%)
|
|
HTN
|
6 (30%)
|
|
HTN/DM
|
1 (5%)
|
|
D/S (n, %)
|
|
|
Disease
|
17 (85%)
|
|
Syndrome
|
3 (15%)
|
|
Preoperative mRS (n, %)
|
2.9 ± 1.119
|
|
Bilaterality (n, %)
|
|
|
Rt/Lt
|
16 (80%)
|
|
Rt
|
2 (10%)
|
|
Lt
|
2 (10%)
|
Abbreviations: D/S, disease/syndrome; DM, diabetes mellitus; HTN, hypertension; ICH,
intracerebral hemorrhage; IVH, intraventricular hemorrhage; Lt, left; mRS, modified
Rankin Scale; Rt, right; SAH, subarachnoid hemorrhage; TIA, transient ischemic attack.
Outcomes
Intraoperatively, all the patients (100%) had a patent artery on ICG fluorescence
and intraoperative Doppler. The mRS improved postoperatively to a mean of 1 ± 1.78
with a significant mean change of −1.9 ± 2.1, p-value = 0.001. Around 80% (n = 16) of the patients had no deficits postoperatively. While three patients (15%)
had left-sided weakness (same as preoperative), and one patient developed ICH during
surgery and died. After a mean of 13.6 months of follow-up, one patient developed
a hemorrhagic stroke, and another showed right-side numbness ([Table 2).]
Table 2
Surgical outcomes of the included patients
|
Variables
|
Total (N = 20)
|
|
Intraoperative ICG (n, %)
|
|
|
Patent
|
20 (100%)
|
|
Intraoperative Doppler (n, %)
|
|
|
Patent
|
20 (100%)
|
|
Postoperative status (n, %)
|
|
|
Improved
|
16 (80%)
|
|
Not improved
|
3 (15%)
|
|
Died
|
1 (5%)
|
|
Postoperative mRS, (mean ± SD)
|
1 ± 1.777
|
|
Follow-up stroke/TIA (n, %)
|
|
|
Hemorrhagic stroke
|
1 (5%)
|
|
Rt-side numbness
|
1 (5%)
|
Abbreviations: ICG, indocyanine green fluorescence; mRS, modified Rankin Scale; Rt,
right; SD, standard deviation; TIA, transient ischemic attack.
On analyzing the potential impact of patients' characteristics on outcomes, it was
noted that the initial clinical presentation was the only variable significantly correlated
with the postoperative status (p = 0.004). No significant different was found between MMD and MMS patients (p = 1; [Table 3).]
Table 3
Associations between postoperative status and baseline data using Fisher's exact test
|
Variables
|
Postoperative status
|
|
Gender
|
0.462
|
|
Region
|
1
|
|
Family history
|
1
|
|
Presentation
|
0.004[a]
|
|
Comorbidities
|
0.09
|
|
D/S
|
1
|
|
Bilaterality
|
0.393
|
Abbreviation: D/S, disease/syndrome.
a Statistically significant.
Discussion
Surgical revascularization is the mainstay treatment in MMD to enhance cerebral perfusion,
prevent ischemic stroke, reduce blood pressure in the small vascular collaterals,
and prevent hemorrhage.[7] Both direct and indirect surgical techniques are used to connect the external carotid
artery with the cerebral hemispheres.[7] In the past decade, the popularity of the combined bypass has increased. They take
advantage of the quick perfusion of direct bypass and collaterals ingrowth from indirect
bypass.[14] In this research, we described our experience with implementing a combined surgical
revascularization technique in a sample of 20 patients in our center. We found that
combined bypass improved the mRS score, and most patients showed no deficits. On follow-up,
5% had hemorrhagic stroke, and 5% had right-side numbness. There was a significant
association between the preoperative presentation and postoperative status.
A recent large meta-analysis based on 143 articles, with a total of 11,454 patients,
showed that combined bypass was significantly associated with more favorable outcomes
than indirect bypass. Furthermore, combined and direct bypasses are favored over indirect
bypass in lowering the risk of late stroke and hemorrhage.[15] Moreover, a meta-analysis of 18 trials proved that combined and direct bypasses
reduced the recurrence stroke rate.[11] Further studies showed that MMD patients who underwent combined bypass had considerable
functional and angiographic outcomes.[16]
[17]
[18]
[19] These results are consistent with our results that revealed that combined bypass
was effective in MMD patients by reducing the mRS score. In comparison to conservative
management, multiple meta-analyses have demonstrated that bypass surgery is favored
in patients with MMD.[20]
[21]
[22] On the other hand, when Zhao et al compared direst to combined bypass, they revealed
that there was no statistically significant difference between the two surgical groups
in the effect of revascularization for MMD.[23]
In this study, 16 patients (80%) had favorable outcomes, three patients (15%) showed
no improvement, and one patient (5%) developed ICH during surgery and died. On the
last follow-up, one patient (5%) developed a hemorrhagic stroke, and another (5%)
showed right-side numbness. In a prior prospective study, Cho et al have reported
that, out of 60 patients who underwent combined bypass, 96.6% improved, 1.7% remained
the same, and 1.7% worsened. Postoperative complications included symptomatic cerebral
hyper-perfusion syndrome (29.9%), hemorrhage (5.2%), infarction (13%), seizures (2.6%),
and wound infection (1.3%). However, their mortality rate was zero compared with 5%
in our study, and the permanent residual deficits were seen in 6.7% of cases in comparison
to 20% of our cases.[24] Sum et al reported no mortality and 9.8% with deficit in 61 patients who underwent
direct/combined bypass.[19] Moreover, a study of 13 adult patients undergoing combined surgery for MMD has suggested
that combined bypass can provide temporally complementary revascularization, resulting
in no mortality; one case (7.7%) suffered from minor stroke with transient symptoms,
two cases (15.3%) showed transient symptoms consistent with seizure, and one case
(7.7%) developed TIA.[25] In Zhao et al study, of the 71 moyamoya patients, 54 underwent combined bypass.
The noted complications in the combined bypass group were ischemic events (4.4%),
hemorrhage (4.2%), TIA (2.8%), and wound infection (1.4%).[23]
This was in agreement with several reports in the literature, especially in the pediatric
population.[10]
[26] Further 518 direct and combined procedures were conducted by Chen et al, of which
11 (2.1%) cases have developed ICH or ICH with intraventricular hemorrhage. They reported
that preoperative hypertension, posterior circulation involvement, and CT perfusion
stage more than III were risk factors for developing ICH after direct/combined bypass,[13] while the only factor in our records that was significantly correlated with the
postoperative status was the initial clinical presentation. Previous studies suggested
that other factors were found to significantly impact the postoperative outcome of
moyamoya patients, such as age, history of diabetes mellitus or infarction, the preoperative
Karnofsky performance scale score, and higher Suzuki stage.[14]
[26]
[27]
Our study's limitations include the limited sample size and the variation in follow-up
periods among the included patients, in addition to the retrospective nature of our
research and the data obtained over 7 years. However, the consistency of evaluation
and surgical approach by single-surgeon may mitigate such point. Another limitation
is the lack of comparison to alternative therapy options. Therefore, prospective comparative
studies with a larger sample size are required.
Conclusion
Combined direct and indirect surgical revascularization methods had favorable outcomes
in patients with MMD with low risk of postoperative complications and good long-term
follow-up.
