Effects of Surgical Revascularization on Outcome of Patients With Pediatric Moyamoya Disease
Background and Purpose We reviewed surgically treated patients with pediatric moyamoya disease and examined whether vasoreconstructive surgeries reduced the risk of recurrent ischemic attacks and changed overall outcomes in terms of the patients’ performance and intellectual status.
Methods Sixty-four hemispheric sides in 34 pediatric moyamoya disease patients who received surgical treatment were examined. We performed superficial temporal artery to middle cerebral artery (STA-MCA) bypass and encephalo-duro-arterio-myo-synangiosis (EDAMS) on 48 sides (combined group) and indirect bypass surgery such as EDAMS on 16 sides (indirect group). These 34 patients were observed postoperatively from 1 to 14 years (mean±SD, 6.6±3.8 years) and were examined for the incidence of recurrent ischemic attack. Of the 34 patients, 23 were followed up for >5 postoperative years, and their overall outcomes in terms of their performance and intellectual status were determined.
Results Perioperative ischemic events (≤2 weeks after surgery) occurred in 5 surgeries (31%) of the indirect group and in 6 (13%) of the combined group (P=NS). The incidence of postoperative ischemic events (>2 weeks after surgery) was significantly reduced in the combined group (10%) compared with the indirect group (56%; P<.01). Of the 23 patients observed >5 years, 7 patients (30%) were mentally retarded and regarded as having a fair outcome.
Conclusions Combined surgery (STA-MCA bypass with EDAMS) for pediatric moyamoya disease was effective in reducing the risk of postoperative ischemic attacks compared with indirect surgery. Surgical revascularization may be effective in preventing intellectual deterioration and improving overall outcome.
Vasoreconstructive surgery has been accepted as treatment for moyamoya disease. Various kinds of surgeries have been proposed: direct bypass such as STA-MCA bypass1 ; indirect bypass such as EDAS,2 3 4 EDAMS,5 and omentum transplantation;6 and a combination of direct and indirect bypasses.7 8 9 10 11 These surgeries have been shown to provide good neovascularization through the extracranial arteries and reduced risk of cerebral ischemia after the surgery.1 2 3 4 5 6 7 8 9 10 11 It has not been determined, however, whether such vasoreconstructive surgeries changed the overall outcomes of the patients with pediatric moyamoya disease in terms of their performance and intellectual status.
Since 1988, we have surgically treated 34 patients with pediatric moyamoya disease. We examined their incidence of recurrent ischemic attacks after surgery. In addition, we observed them over long periods of time and analyzed their clinical symptoms and final outcomes.
Subjects and Methods
Prevention of Recurrent Ischemic Attack
Between 1988 and 1995, 64 hemispheric sides in 34 patients (15 males and 19 females) with pediatric moyamoya disease who underwent surgical treatment were included in this study. The ages of the patients when the initial symptoms appeared ranged from 0 to 11 years (mean±SD, 5.7±2.8 years). Ages at the time of the surgery were 0 to 19 years (mean±SD, 7.6±3.6). All symptoms were manifested as ischemic symptoms. According to the preoperative status, we classified the ischemic symptoms of those patients into two types: (1) TIAs without any lesions of cerebral infarction on CT scanning or MRI (in 25 patients) (TIA type) or (2) completed ischemic strokes that were detected with CT scan or MRI (in the other 9 patients) (CS type).
We have usually performed a combined surgery of STA-MCA bypass and EDAMS for these patients. The details of the operative procedure are described elsewhere.11 We performed such combined surgery on 48 hemispheric sides (combined group). Only indirect bypass surgery such as EDAS or EDAMS was performed on the 16 hemispheric sides (indirect group) because the patients received surgical treatments in the other institutes or the appropriate recipient vessels were not found on the surface of the brain at the time of surgery. All 34 patients were followed up for 1 to 14 years (mean±SD, 6.6±3.8 years). The incidences of perioperative ischemic attacks (≤2 weeks after surgery) and postoperative ischemic attacks (>2 weeks after surgery) were examined.
Outcomes After Long-term Observation
Of the 34 patients, we focused on 23 patients who were followed up >5 years after the surgical treatment; 8 were male and 15 were female. Their ages at onset ranged from 2 to 10 years (mean±SD, 5.3±2.2 years). Their ages at the time of the surgery were 3 to 19 years (mean±SD, 7.4±3.5 years). The patients were observed from 5 to 14 years postoperatively (mean±SD, 8.7±2.8 years). Six were CS-type patients, and the other 17 patients were categorized as TIA type; their ages at onset ranged from 4 to 7 years (mean±SD, 4.5±1.5 years) and from 2 to 10 years (mean±SD, 5.6±2.4 years), respectively. Five patients (4 TIA type, 1 CS type) were treated with bilateral indirect bypass surgery. Fourteen patients (10 TIA type, 4 CS type) underwent combined surgery on both sides. The other 4 patients (3 TIA type, 1 CS type) received indirect bypass surgery on one side and combined bypass surgery on the other side.
The overall outcome in terms of activities of daily living and final performance status was examined with interviews of the patients and/or their families. According to these findings, the patients were divided into three groups. The patients who were completely dependent and not able to go to school were assigned to the “poor” outcome group. The patients who had neurological deficits that made them dependent in daily life and/or had intellectual problems were grouped into the “fair” outcome group. The patients who had a normal daily life for their age were assigned to the “good” outcome group. In 16 patients, intellectual ability in the postoperative period was estimated with the WISC-R.
The incidences of perioperative and postoperative ischemic strokes were compared between the indirect group and the combined group with the χ2 test. The overall outcome of the patients was compared not only among the groups classified on the basis of the type of surgery they received but also between TIA-type patients and CS-type patients with the χ2 test. The postoperative WISC-R scores were compared between TIA-type patients and CS-type patients with Student’s t test.
Recurrent Ischemic Attack
Perioperative Ischemic Attack
As shown in Table 1⇓, perioperative ischemic complications occurred in 11 of 64 surgeries (17%). In the indirect group, ischemic complications appeared as 3 TIAs and 2 RINDs. In the combined group, TIA occurred in 4 surgeries, RIND in 1, and CS in 1. The CS occurred in the left temporal lobe on the 11th postoperative day when the patient had been crying, although related symptoms disappeared in a few weeks. In total, perioperative ischemic events occurred in 5 surgeries (31%) in the indirect group and in 6 (13%) in the combined group, but the difference in incidence of perioperative ischemic events was not significant between the two groups.
Postoperative Ischemic Attack
As shown in Table 2⇓, after the surgery ischemic events such as motor weakness after hyperventilation persisted in 9 hemispheric sides (56%) of the indirect group and in 5 hemispheric sides (10%) of the combined group. In the indirect group TIAs occurred in 6 hemispheric sides and RINDs in 3 sides, and in the combined group TIAs occurred in 4 sides and RIND in 1 side. Ischemic events that developed to a CS were not observed in either group. The incidence of postoperative ischemic events was significantly reduced in the combined group compared with the indirect group (P<.01). The intervals from surgery to the disappearance of the ischemic attacks were 0 to 60 (mean, 17.3) months in the indirect group and 0 to 48 (mean, 3.8) months in the combined group. In 13 of the 14 patients who suffered postoperative persistent ischemic symptoms, the symptoms were usually manifested as weakness of the lower extremities.
The present status of the 23 patients who were observed >5 years is shown in Table 3⇓. Of these patients, 16 patients with a good outcome were living normal daily lives as students or employees. However, 4 of the students were doing very poorly intellectually at school. Another student is mentally retarded and is going to a high school for handicapped people. Two adolescents who had already graduated from school had simple jobs because of their mental retardation. These 7 patients were regarded as having had a fair outcome. None of the patients were completely dependent and considered to have had a poor outcome. The type of surgery the patients received was not significantly related to the final outcome (Table 3⇓). On the other hand, only 2 of the 17 TIA-type patients (12%) were in the fair outcome group, although 5 of the 6 CS-type patients (83%) were in the fair outcome group (Table 3⇓). The incidence of a fair outcome was significantly higher in the CS-type patients than in the TIA-type patients (P<.001).
The WISC-R scores ranged from 124 to 75 (mean±SD, 93±14) among the TIA-type patients; this mean score was significantly higher than that of the CS-type patients (range, 52 to 79; mean±SD, 63±11; P<.05).
Moyamoya disease is characterized by progressive arterial narrowing or obstruction of major intracranial cerebral arteries, which is associated with an extensive development of collateral vessels. In childhood, most of the patients develop ischemic symptoms such as TIA, RIND, and CS. Although the natural course of this disease has been controversial, ischemic symptoms usually develop from reversible ischemia to cerebral infarction associated with permanent neurological deficits and intellectual impairment.12 13 According to a report by Fukuyama and Umezu14 in which they treated patients nonsurgically, ischemic attacks were still observed after a mean of 6 years and 11 months in half of the TIA-type patients examined; in the other half of the TIA-type group, ischemic symptoms disappeared within 5 years after onset. Those authors also found that the TIA-type patients had a good intellectual prognosis.15 However, other studies have suggested that even in TIA-type patients, intellectual impairment would progress if they did not undergo appropriate surgical treatment.2 3 12
Revascularization surgery for pediatric moyamoya disease has been consistently effective in eliminating ischemic neurological symptoms irrespective of the use of direct or indirect bypass surgery.1 3 4 5 8 10 16 According to a report regarding encephalomyosynangiosis by Karasawa et al1 and another report regarding EDAMS by Kinugasa et al,5 revascularization from indirect bypass begins to develop 2 weeks after surgery and becomes well developed 3 months postoperatively. However, as Miyamoto et al9 pointed out, some patients had persistent ischemic attacks after receiving indirect bypass surgery and needed to have an additional direct bypass surgery. According to a study by Matsushima et al,8 STA-MCA anastomosis with EMS was superior to EDAS in both the development of collateral circulation and postoperative clinical improvement at 1 year after surgery. In their reports, TIA disappeared completely in the sides where both direct and indirect bypass surgeries were performed, whereas it disappeared in 3 of 13 sides (23%) receiving indirect bypass only.8 The amount of cerebral blood flow that was brought in by a branch of STA may be very limited and might not be sufficient to adequately restore ischemic brain. In addition, it is conceivable that the STA-MCA bypass was cumbersome to perform in pediatric moyamoya disease patients and may increase the risk of perioperative ischemia. However, cerebral hemodynamics were severely impaired even in the TIA-type patients,17 and blood flow that is directly brought in by STA-MCA bypass, even if it is a very small amount, can be helpful in preventing the perioperative risk of stroke and terminating postoperative ischemic symptoms. In fact, we observed a tendency for a lower incidence of perioperative ischemic symptoms when we performed STA-MCA bypass in addition to EDAMS. Moreover, our study revealed that combined surgery (STA-MCA bypass and EDAMS) was effective in terminating ischemic symptoms significantly earlier than indirect surgery only. Indeed, after direct and indirect bypass surgery, TIA disappeared immediately postoperatively in 43 hemispheric sides (90%), whereas indirect bypass alone had an immediate effect on ischemic symptoms in only 7 hemispheric sides (44%). Our results support the benefit of direct bypass surgery combined with indirect bypass for pediatric moyamoya disease. Sole STA-MCA anastomosis has not been selected for patients with pediatric moyamoya disease because it can be easily combined with the indirect bypass procedure, which results in an additional and rich neovascularization.
We ranked the patients’ outcomes according to their performance status. All patients were independent in their daily lives. However, 7 of 23 patients (30%) were not performing very well, mostly because of intellectual impairment. Five of the 6 CS-type patients had only a fair outcome. In contrast, only 2 of the 17 TIA-type patients had a fair outcome. It has been shown that intellectual prognosis is usually poor in patients with CS-type ischemic symptoms and/or in those whose age at onset is <5 years.2 12 13 14 15 According to Kurokawa et al,12 the natural course of intellectual ability is not always favorable in TIA-type patients. Intellectual impairment developed with time, and mild to severe intellectual impairments were present in 62% of patients who had been observed for >5 years. In contrast, according to a report by Matsushima et al,2 an IQ score >86 is considered normal, and patients who received EDAS had better IQ scores than those in the natural course group. We measured the intellectual ability of our patients by WISC-R and found that scores were significantly lower in the CS-type patients than in the TIA-type patients, in keeping with the results of previous reports. When we examined only our 12 TIA-type patients whose WISC-R score could be obtained, the mean score was 93.0, and the scores were >86 in 9 of the 12 (75%). This result was not fully satisfactory, but >80% of the TIA-type patients were living normal daily lives. Therefore, surgical revascularization for moyamoya disease may prevent intellectual deterioration and improve overall outcome. In contrast to the immediate postoperative effect observed by elimination of ischemic events, the type of surgery the patients received was not related to final outcome. After a certain period of time when revascularization through indirect bypass had developed, we observed no differences irrespective of the use of direct bypass surgery.
In view of the fact that CS-type onset is common in patients aged <5 years and that such patients have had a relatively poor prognosis, diagnosis and surgical intervention as early as possible are of paramount importance.3 4 Screening high-risk groups, such as those with a family history of moyamoya disease,18 and awareness of pediatricians regarding this disease would improve prognosis.
In conclusion, we found that combined surgery (STA-MCA bypass with EDAMS) for pediatric moyamoya disease was more effective in reducing the risk of postoperative ischemic attacks than indirect surgery. Surgical revascularization for moyamoya disease may be effective in preventing intellectual deterioration and improving overall outcome.
Selected Abbreviations and Acronyms
|RIND||=||reversible ischemic neurological deficit|
|STA-MCA||=||superficial temporal artery to middle cerebral artery|
|TIA||=||transient ischemic attack|
|WISC-R||=||Wechsler Intelligence Scale for Children, Revised|
This study was supported in part by a grant from the Research Committee on Spontaneous Occlusion of the Circle of Willis of the Ministry of Health and Welfare (Japan).
Reprint requests to Tatsuya Ishikawa, MD, Department of Neurosurgery, Hokkaido University School of Medicine, North 15, West 7, Kitaku, Sapporo, 060 Japan.
- Received December 16, 1996.
- Revision received April 11, 1997.
- Accepted April 11, 1997.
- Copyright © 1997 by American Heart Association
Houkin K, Kamiyama H, Abe H, Takahashi A, Kuroda S. Surgical therapy for adult moyamoya disease: can surgical revascularization prevent the recurrence of intracerebral hemorrhage? Stroke. 1996;27:1342-1346.
Fukuyama Y, Umezu R. Prognosis of occlusive disease of the circle of Willis in childhood: in terms of EEG, CT scan, and intellectual development [in Japanese]. In: Annual Report 1983 of the Research Committee on Spontaneous Occlusion of the Circle of Willis. Tokyo, Japan: Ministry of Health and Welfare; 1983:81-88.