Background and Purpose We previously reported that the level of basic fibroblast growth factor (bFGF) is high in cerebrospinal fluid (CSF) taken from patients with moyamoya disease. The present study investigated the levels of other angiogenic growth factors in the CSF of moyamoya patients and the clinical significance of bFGF in moyamoya disease.
Methods The levels of bFGF, interleukin-8, platelet-derived growth factor, transforming growth factor-β, endothelial growth factor, and vascular endothelial cell growth factor in CSF, taken from 38 patients with moyamoya disease and 16 patients with atherosclerotic occlusive disease (control group), were measured by an enzyme-linked immunosorbent assay. We analyzed the correlation between the level of bFGF and the clinical factors of age, onset pattern, development of neovascularization, and cerebral circulation.
Results The CSF of moyamoya patients contained a high concentration of bFGF to a significant (P<.05) extent. The bFGF level was apparently elevated in the patients in whom neovascularization from indirect revascularization, such as encephaloduroarteriosynangiosis, was well developed (P<.01). A linear correlation between the values of bFGF and cerebral vascular response to acetazolamide (r=.7; P<.05) was revealed. The other angiogenic factors were not significantly high compared with the control group.
Conclusions The elevation of bFGF in moyamoya disease seems to be specific and is not related simply to cerebral ischemia. Clinically, the bFGF level is a useful indicator to predict the efficacy of indirect revascularization after surgery.
The cause of spontaneous occlusion of the circle of Willis (moyamoya disease) has been investigated from several vantage points, such as environmental factors and viral and bacterial infections. Its true cause is still unknown, although some clinical and basic features have gradually been clarified. Moyamoya disease is distinct from other cerebrovascular occlusive diseases in three aspects. The first is that a progressive stenosis takes place predominantly at the bilateral carotid forks in moyamoya disease, which is not seen in other cerebrovascular occlusive diseases. The second is a characteristic development of neovascularization from various collateral pathways, ie, transdural anastomosis and a fine-vessel network at various locations such as the frontal skull base (ethmoidal moyamoya), the skull base (basal moyamoya), and the posterior fossa (posterior moyamoya). The third difference is a forming of neovascularization by indirect revascularization from a simply placed vascularized tissue such as the temporal muscle and meningeal arteries by surgery. These features are not generally seen in other cerebrovascular occlusive diseases. Anatomically, this formation of collaterals is difficult to induce because of the interruption of CSF between the vascularized tissue and the brain surface. Angiogenic factors in the CSF are therefore thought to be related to the formation of collaterals.
We previously reported that the bFGF level was elevated in the CSF of patients with moyamoya disease.1 The neovascularization that is characteristic in moyamoya disease can be directly or indirectly regulated by other growth factors, however.2 3 4 5 Direct angiogenic growth factors including bFGF, VEGF, and IL-8 are mitogenic for endothelial cells in vitro and stimulate angiogenesis in vivo.3 6 7 8 Other indirect factors, including PDGF and TGF-β, which do not have mitogenic functions in endothelial cells in vitro, may play some role in moyamoya disease.2 3 9 10 EGF can regulate proliferation of smooth muscle cells.11 In the present study we investigated these growth factors in CSF taken from patients with moyamoya disease. The clinical significance of bFGF in the CSF of moyamoya patients was also examined.
Subjects and Methods
Of the patients who underwent surgical treatments for moyamoya disease since 1992, there were 38 (15 males, 23 females) from whom more than 1 mL of CSF was taken from the subarachnoid space of the cerebral cortex (frontal cortex or superficial sylvian fissure) during the bypass surgery. The average age was 29.8 years, and the age distribution showed two peaks, one of children (younger than 15 years) and the other of adults (older than 20 years) (Fig 1⇓). All of the children (n=15) and 16 of the adults (n=23) had an initial onset of cerebral ischemia. The other adult patients had an onset of cerebral hemorrhage. We have ordinarily performed both an STA-MCA anastomosis and EDAMS. EDAMS is an indirect revascularization procedure in which the vascularized tissue, such as the temporal muscle and middle meningeal artery, is placed on the brain surface.12 The patients were classified as angiographic stage III based on the criteria of Suzuki and Takaku,13 except for 2 patients (1 at stage IV and 1 at stage V).
For a control group, CSF from 16 patients with cerebral ischemia due to atherosclerotic stenosis and/or occlusion of the cerebral major arteries was investigated. CSF was also obtained from the brain surface during bypass surgery. Extracranial to intracranial bypass surgery was performed in selected cases that met the entry criteria described elsewhere (briefly, those in which the CBF was reduced [low CBF] with poor response to the acetazolamide challenge test [poor CVRa] confirmed by SPECT without any clinically significant infarction).14 15 The mean age of the control subjects was 48.7 years. The bypass surgery for moyamoya disease and ischemic diseases due to atherosclerotic changes was performed at least 4 weeks after the last insult to eliminate the influence of acute brain damage caused by cerebral ischemia.
Measurement of Angiogenic Growth Factors
The CSF samples obtained from the patients and control subjects were filtered through a 0.22-mm filter (Millipore Co) and stored at −80°C until assay. The growth factors concerned with angiogenesis—bFGF, TGF-β, PDGF, IL-8, EGF, and VEGF—were measured with an enzyme-linked immunosorbent assay kit (Quantikine, R&D Systems) according to the manufacturer's instructions. The results were statistically analyzed with the Mann-Whitney U test. Values of P<.05 were considered significant.
Postoperative angiography was performed in 21 patients at least 4 months after the revascularization surgery. The patency of all the STA-MCA anastomoses was satisfactory. We evaluated the differences in the levels of growth factors between a “good result” and a “poor result” group after indirect revascularization. The effect of the indirect revascularization was evaluated as “good” when the collateral from the deep temporal muscle artery and the middle meningeal artery was clearly seen in the postoperative angiography and as “poor” when it was not.
We performed the surgical revascularization for the patients with low perfusion detected by SPECT. Of the 38 patients, 10 patients were examined by 133Xe inhalation SPECT (Shimazu Headtome Set 031) in the same hospital. The absolute values of the ipsilateral mCBF, at rest and after the intravenous injection of acetazolamide (10 mg/kg), were measured. We used cerebral vascular response to acetazolamide (CVRa) for the quantitative analysis of cerebral vascular reserve. CVRa was calculated as follows:CVRa|<|=|>|(Acetazolamide mCBF|<|-|>|Rest mCBF)/Rest mCBF|<|\times|>|100The correlation between the level of bFGF and preoperative cerebral circulation was analyzed. The radiological findings of angiography and SPECT were blindly checked by three authors, T.Y., K.H., and A.T.
Concentration of Angiogenic Factors
The mean value of bFGF in the CSF of moyamoya patients was 64.0 pg/mL (range, 0 to 355 pg/mL). The value of bFGF in the control subjects was 6.5 pg/mL (range, 0 to 26.8 pg/mL). bFGF was present significantly more often in the CBF of the moyamoya patients than in that of the control subjects (P<.05) (Fig 2⇓). The mean values of bFGF in children and adults with moyamoya disease were 76.6 pg/mL and 52.8 pg/mL, respectively (higher in the children, but not significantly different). The bFGF levels in the patients with ischemic and hemorrhagic onsets were 60.1 pg/mL and 72.9 pg/mL, respectively (not significantly different).
IL-8, PDGF, TGF-β, EGF, and VEGF
The mean level of IL-8 was 82.5 pg/mL (range, 0 to 413 pg/mL) in the CSF of the moyamoya patients and 46.8 pg/mL (range, 18 to 115 pg/mL) in the control subjects. The IL-8 levels in children and adult moyamoya patients were 108.0 pg/mL and 60.9 pg/mL, respectively. The values of IL-8 in the ischemic and hemorrhagic onsets were 90.7 pg/mL and 79.0 pg/mL, respectively. However, there were no significant differences in IL-8 among any of the groups (Fig 3⇓).
The values of PDGF and TGF-β were 19.2 pg/mL (range, 10.7 to 43.1 pg/mL) and 9.3 ng/mL (range, 0 to 15.9 ng/mL), respectively, in the moyamoya patients and 16.7 pg/mL (range, 0 to 27.3 pg/mL) and 14.6 ng/mL (range, 1.9 to 24.3 ng/mL) in the control subjects. There were no significant differences. The EGF level in CSF was below the limits for measurement. The value of VEGF was 13.1 ng/mL (range, 9.6 to 15.5 ng/mL) in the moyamoya patients and 12.1 ng/mL (range, 13.0 to 16.5 ng/mL) in the control subjects, which was not a significant difference (Fig 4⇓).
Effect of Indirect Revascularization
Postoperative angiography revealed good neovascularization from EDAMS in 15 of the 21 patients examined by this method. However, little vascularization was seen in 6 patients examined by angiography, regardless of good patencies in the STA-MCA anastomoses. The ages of these patients were 47, 39, 67, 32, 62, and 5 years; the mean age was 42 years, which was higher than that of the group with good results. Therefore, adults tended to belong to the group with poor results, although the difference was not significant. The value of bFGF in the good result group was 109.5 pg/mL, which was significantly higher than the 16.7 pg/mL of the poor result group (P<.01) (Fig 5⇓). The IL-8 level in the good result group was 134.9 pg/mL, which was higher than that in the poor result group (40.1 pg/mL), but not significantly. The values of PDGF and TGF-β were 22.5 pg/mL and 12.6 ng/mL, respectively, in the good result group and 18.6 pg/mL and 9.0 ng/mL in the poor result group (not significantly different). The values of VEGF in the good and poor result groups were 12.8 ng/mL and 13.5 ng/mL, respectively, which was also not a significant difference.
Preoperative Cerebral Circulation
The values of bFGF in the CSF of the moyamoya patients ranged from 0 to 355 pg/mL. There was not a close correlation between the values of bFGF and mCBF (r=.09) at rest. However, there was a linear correlation between the values of bFGF and CVRa (r=.7; P<.05; n=10) (Fig 6⇓).
Our previous study demonstrated that bFGF is high in the CSF of moyamoya patients, although the origin of the high bFGF in moyamoya disease is unknown. This elevation of bFGF can be interpreted in two ways. First, bFGF may be a causative factor in inducing progressive arterial occlusion in moyamoya disease. Second, bFGF may not be a causative factor but rather a result of the ischemic brain damage in moyamoya disease. Basically, bFGF exists in neurons, glia, endothelial cells, and vascular smooth muscle cells.16 17 18 19 20 It was reported that astrocytes in ischemic rat brain are positive for bFGF-like immunoreactivity19 and that lethal cell injury and cell death are important mechanisms of bFGF release.11 This implies that bFGF is closely related to ischemic brain damage. However, as shown in the present study, the direct influence of ischemic brain damage can be eliminated, because (1) this study examined chronic-stage moyamoya patients (CSF was taken at least 4 weeks after the last ischemic event), (2) the control group with arteriosclerotic occlusive disease had a low level of bFGF, and (3) as shown in the present study, other angiogenic factors in moyamoya patients that would be elevated by simple ischemic brain damage are not significantly high, as is bFGF. Therefore, the high level of bFGF seen in moyamoya disease is thought to be related to the pathogenesis of moyamoya disease and specific to this disease, although the influence of cerebral ischemia cannot be completely ruled out.
bFGF can induce the proliferation of vascular endothelial cells, smooth muscle cells, and fibroblastic cells, as well as angiogenesis.1 3 8 11 16 17 19 20 bFGF was localized in the extracellular matrix of smooth muscle cells in the media and in the thickened intima in an autopsy study of moyamoya patients.21 Pathologically, thickened intima composed predominantly of smooth muscle cells has been disclosed in the carotid fork in moyamoya disease.22 23 24 25 An immunohistochemical study revealed the presence of bFGF on the endothelium and vascular smooth muscle in the STA obtained from patients with moyamoya disease.26
bFGF is believed to have two different primary effects: to induce endothelial cell proliferation, which may result in stenotic changes of the carotid fork, and to induce angiogenesis, which may lead to collateral formation. The difference in bFGF concentration and/or the interval for exposure between the paracarotid cistern and the peripheral subarachnoid space may explain the two apparently distinct phenomena seen in moyamoya disease (steno-occlusive changes of the major arteries and rich formation of collateral circulation in the peripheral arterial system). Consequently, we can speculate about a role of bFGF in the pathogenesis of moyamoya disease as follows. bFGF, exuded and/or released into CSF by some stimulation (promoted by some genetic abnormality), is first gathered into the basal cistern. High-concentration bFGF infiltrates into the wall of the cerebral major trunks for a long interval and stimulates progressive thickening of the intima, resulting in stenosis of the bilateral cerebral major trunks. The posterior circulation, which is usually normal in moyamoya disease, may be free from this concentration of bFGF since the Lilieqvist membrane separates the carotid-basal cistern into two compartments. Subsequently, bFGF diffuses to the subarachnoid space of the cerebral cortex and its concentration decreases, resulting in the development of collateral pathways such as transdural anastomosis and indirect revascularization.
This hypothesis, which could be termed the “bFGF–moyamoya disease theory,” may be a reasonable explanation of the elevation of bFGF in moyamoya disease, although not all the specific phenomena seen in moyamoya disease can necessarily be explained by this hypothesis. Moyamoya disease is currently thought to be related to a genetic mutation, although no study has demonstrated a chromosomal abnormality in moyamoya disease.27 The genetic change may result in enhanced transcription of the bFGF gene. Further precise genetic study is necessary to verify this very interesting hypothesis.
This study revealed that cytokines other than bFGF have little relation to moyamoya disease. However, in the present study IL-8 levels were high, although not significantly, in the CSF of moyamoya patients. Masuda et al23 showed that some macrophages and T cells were localized at the thickened intima with proliferation of the smooth muscle cells. Macrophages can become angiogenic in a hypoxic state. They can degrade the local extracellular matrix where bFGF is stored by being bound to heparin like glycosaminoglycans.28 In addition, IL-8 released from macrophages was chemotactic for endothelial cells, neutrophils, and macrophages and could induce their proliferation.6 28 29 These findings suggest that a subsequent response of inflammatory cells, especially macrophages, may assist intimal thickening of the major cerebral arteries and angiogenesis in moyamoya disease.
The elevation of bFGF in the CSF in moyamoya disease is clinically important because it is closely related to the effect of indirect revascularization. Transdural anastomosis is a form of collateral circulation that is frequently well developed in patients with moyamoya disease. This collateral circulation is rarely seen in other general cerebral arteriosclerotic steno-occlusive diseases. Indirect revascularization such as EDAMS, EDAS (encephaloduroarteriosynangiosis), and EMS (encephalomyosynangiosis) performed in patients with moyamoya disease results in excellent neovascularization and improvement of cerebral hemodynamics after surgery.12 30 31 In contrast, it is well known that this indirect revascularization is not effective for other general cerebral arteriosclerotic steno-occlusive diseases. However, it has been reported that indirect revascularization failed in some patients with moyamoya disease,31 as was also seen in our six patients who had poor results. Interestingly, the present study revealed that bFGF was significantly lower after revascularization in patients with poor results than in patients with good results. Moreover, our study revealed that the disturbance of cerebral vascular reserve detected by the acetazolamide challenge test is closely related to the level of bFGF. Surgical outcome after indirect revascularization depends on the age of patients and the onset of clinical signs.32 However, it is conceivable that good neovascularization can be expected in a patient with highly disturbed cerebral vascular reserve with a high level of bFGF.
In conclusion, while the mechanism of its initial expression in moyamoya disease is unexplored, a high level of bFGF in the CSF of patients with moyamoya disease is believed to be a major factor in the development of stenosis of intracranial major arteries and angiogenesis of collateral circulation. However, ischemic brain damage may also play a role in the increase of bFGF. Further study is necessary to reveal the true role of bFGF in moyamoya disease. Clinically, the bFGF level can be a useful indicator because we can predict the efficacy of indirect revascularization after surgery by measurement of the bFGF level in CSF.
Selected Abbreviations and Acronyms
|bFGF||=||basic fibroblast growth factor|
|CBF||=||cerebral blood flow|
|CVRa||=||cerebral vascular response to acetazolamide|
|EGF||=||endothelial growth factor|
|MCA||=||middle cerebral artery|
|mCBF||=||mean cerebral blood flow|
|PDGF||=||platelet-derived growth factor|
|SPECT||=||single-photon emission computed tomography|
|STA||=||superficial temporal artery|
|TGF||=||transforming growth factor|
|VEGF||=||vascular endothelial cell growth factor|
- Received March 11, 1996.
- Revision received August 18, 1996.
- Accepted August 19, 1996.
- Copyright © 1996 by American Heart Association
Brogi E, Wu T, Namiki A, Isner JM. Indirect angiogenetic cytokines upregulate VEGF and bFGF gene expression in vascular smooth muscle cells, whereas hypoxia upregulates VEGF expression only. Circulation. 1994;90:649-652.
Folkman J, Shing Y. Angiogenesis. J Biol Chem. 1992;267:10931-10934.
Koch AE, Polverini PJ, Kunkel SL, Harlow LA, DiPietro LA, Elner WM, Elner SG, Strieter RM. Interleukin-8 as a macrophage-derived mediator of angiogenesis. Science. 1992;258:1798-1801.
Pertovaara L, Kaipainen A, Mustonen T, Orpana A, Ferrara N, Saksela R, Alitalo K. Vascular endothelial growth factor is induced in response to transforming growth factor-β in fibroblastic and epithelial cells. J Biol Chem. 1994;269:6271-6274.
Winkles LA, Gay CG. Serum phorbol ester, and polypeptide mitogens increase class 1 and 2 heparin binding (acid and basic fibroblast growth factor) gene expression in human vascular smooth muscle cells. Cell Growth Differ. 1991;2:531-540.
Takahashi A, Kamiyama H, Houkin K, Abe H. Surgical treatment of childhood moyamoya disease: comparison between the reconstructive surgery centered on the frontal region and that centered on the parietal region. Neurol Med Chir (Tokyo). 1995;35:234-237.
Ishikawa T, Houkin K, Abe H, Isobe M, Kamiyama H. Cerebral haemodynamics and long term prognosis after extra-cranial-intracranial bypass surgery. J Neurol Neurosurg Psychiatry. 1995;59:625-628.
Baird A, Waliche PA. Fibroblast growth factors. Br Med Bull. 1989;45:438-452.
Gospodarowicz D. Isolation and characterization of acidic and basic growth factor. In: Methods in Enzymology, Peptide Growth Factors. Vol 147B. Orlando, Fla: Academic Press Inc; 1987:106-117.
Schweigerer L, Neufeld G, Freidman J, Arabaham JA, Fiddes JC, Gospodarowicz D. Capillary endothelial cells express basic fibroblast growth factor, a mitogen that promotes their own growth. Nature. 1989;325:209-221.
Sueishi K, Kono S. Immunohistochemical localization of basic fibroblast growth factor in the narrowed arteries of the arteries of Willis in moyamoya disease. In: Fukui H, ed. 1992 Proceedings of the Research Committee on Spontaneous Occlusion of the Circle of Willis [in Japanese]. Tokyo, Japan: Ministry of Health and Welfare Japan. 1993:131-133.
Kono S, Oka K, Sueishi K. Histopathological and morphometric studies of leptomeningeal vessels in moyamoya disease. Stroke. 1990;21:1044-1050.
Masuda J, Ogata J, Yutani C. Smooth muscle cell proliferation and localization of macrophages and T cells in the occlusive intracranial major arteries in moyamoya disease. Stroke. 1993;24:1960-1967.
Ogata J, Masuda J. Phenotype changes and proliferation of smooth muscle cells in occlusive arterial lesions in moyamoya disease. In: 1992 Proceedings of the Research Committee on Spontaneous Occlusion of the Circle of Willis [in Japanese]. Tokyo, Japan: Ministry of Health and Welfare Japan. 1993:131-133.
Oka K, Yamashita M, Sadoshima S, Tanaka K. Cerebral haemorrhage in moyamoya disease at autopsy. Virchows Arch. 1981;392A:247-261.
Sunderketter C, Steinbrink K, Goebeler M, Bhardwaj R, Sorg C. Macrophages and angiogenesis. J Leukoc Biol.. 1994;55:410-422.
Kubo H. Angiogenesis on encephalo-myo-synangiosis. Arch Jpn Chir. 1993;62:82-91.
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.