Child s Nerv Syst (1997) 13: 454– 464 © Springer-Verlag 1997

PERSPECTIVES

Koreaki Mori

Pediatric neurosurgery and myself

K. Mori Department of Neurosurgery, Kochi Medical School, Kohasu, Okoh-cho, Nankoku-City, Kochi, 783 Japan

Abstract The career of one neurosurgeon interested in pediatric neurosurgery is described, with reference to the people who influenced him and his own publications.

Prologue – why pediatric neurosurgery?

When I graduated from Kyoto University Medical School in 1961, the standard of medicine in Japan was not so high as in the United States. Many graduates took ECFMG tests in order to have postgraduate training in the United States. I was one of the students who qualified, but instead of going to the U.S., I entered the postgraduate course in Neurosurgery after a 1-year internship. At the beginning of my training I studied to be a general surgeon, and then specialized in Neurosurgery under the guidance of the late Professor emeritus Chisato Araki (Fig. 1) and Professor emeritus Hajime Handa (Fig. 2). I did research work on electrically induced thrombosis [5] along with the clinical training, and I obtained the degree of Doctor of Medical Science (DMS, PhD) in 1967 after writing a dissertation entitled “Experimental studies on electrically induced arterial thrombosis in dogs, with special reference to the treatment of intracranial aneurysms and arteriovenous malformations” [16]. I qualified for the Japanese board of Neurosurgery in 1968. During my training in neurosurgery, most neurosurgeons were interested in surgical treatment of cerebrovascular disease, especially cerebral aneurysms. Owing to the wide use of cerebral angiography and the introduction of the operative microscope, surgical treatment of cerebrovascular disease became of major interest to Japanese neurosurgeons.

Key words Pediatric neurosurgery · Career · Publications · Moyamoya · Periventricular lucencies

Around the latter half of the 1960s, a campus struggle took place at medical schools in Japan, initially because of an inadequate internship system. Most university hospitals ceased to have patients, because all facilities were occupied by radical students and could not be used. My clinical activity reached its lowest point. At that time, I met another graduate of Kyoto University, Dr. Satoshi Matsumoto, who was 7 years ahead of me (Fig. 3.). He had just come back from Chicago after finishing his residency in Neurosurgery. He kindly introduced me to Dr. Anthony J. Raimondi (Fig. 4), who gave me the chance of acquiring some experience specifically in pediatric neurosurgery in 1971. I decided to train in pediatric neurosurgery, because many people were studying cerebrovascular surgery as a neurosurgical specialty, whereas interest in pediatric neurosurgery was only just dawning. Dr. Raimondi created special terms for my residency. After he had accepted me as a fourth-year resident at Northwestern University, I was able to spend 2 years at the Children’s Memorial Hospital under his special program. I had the shortest training program of his many residents, but I believe that I gained the most benefit even though I stayed a shorter time than the others. I did a residency program for pediatric neurosurgery with research on experimental hydrocephalus for 6 months, followed by pediatric neurology for another 6 months. Then I spent 1 year in pediatric neurosurgery with Dr. David McLone (Fig. 5). We two residents were on call every other night.

455

Fig. 1 The late Professor emeritus Chisato Araki Fig. 2 Professor emeritus Hajime Handa Fig. 3 Professor emeritus Satoshi Matsumoto Fig. 4 Professor Anthony J. Raimondi Fig. 5 Professor David G. McLone

In 1997, the International Society for Pediatric Neurosurgery will celebrate its 25th Anniversary. On this occasion, I would like to trace the quarter century of my career in pediatric neurosurgery as the basis of this autobiography.

Electron microscopic findings of the periventricular white matter and “periventricular lucency” on the CT scan

I was given a chance to study experimental hydrocephalus using the electron microscope at the begining of my resi-

456

dency program. At that time the research laboratory had two types of hydrocephalic mice: ch and hy-3. I was taught the basic techniques of electron microscopy by the excellent technicians at the Children’s Memorial Hospital. I concentrated my study on the periventricular white matter of the hydrocephalus ch mouse. As in other experimental hydrocephalic animals, the extracellular space in the periventricular white matter was increased in the hydrocephalus ch mouse [36]. This finding was thought to be due to increased periventricular water content as a result of increased permeability of the ependyma caused by high pressure within the ventricles and structural changes in the ependyma. Based on the electron microscopic findings in the hydrocephalus ch mouse and the “double density” pattern in human hydrocephalus as revealed by isotope ventriculography [12], I focused my study on the periventricular area on the computed tomogram (CT) immediately after its introduction to clinical use, because CT was very sensitive in detecting water. I expected a markedly increased water content in the periventricular white matter in advanced hydrocephalus, because the more advanced the hydrocephalus becomes, the more fluid penetrates the ependyma and gets into the subependymal white matter. Interestingly enough, “periventricular lucency” (PVL) was minimal in long-standing hydrocephalus, and the PVL possibly represented acute periventricular edema, rather than an alterna-

Fig. 6 a, b Schematic drawings of the cerebral angiogram of hydrocephalus. Left A-P view, right lateral view. a Communicating hydrocephalus, b non-communicating hydrocephalus

tive pathway for cerebrospinal fluid absorption in chronic compensated hydrocephalus [32, 40, 47]. I continued to study PVL in experimental hydrocephalus [2, 74]. NMR study suggested that the tissue fluid changes might be depicted more clearly than by CT alone if NMR was also used. PVL on a CT scan can now be seen definitely on T2-weighted MRI, as periventricular high intensity (PVH or PVHI). PVL or PVH is thought to be an indication for hydrocephalus surgery.

Strong impact of the angiographic diagnosis of hydrocephalus on my career in neurosurgery

Before I started taking special training in pediatric neurosurgery, cerebral angiography had been used to diagnose cerebrovascular diseases, to demonstrate tumor stain of brain tumors, or to detect an avascular area for the diagnosis of intracranial hematomas. Dr. Raimondi was such a genius that he could correlate the ventricular system and brain structures with the cerebral vessels. This approach allowed us to diagnose hydrocephalus in newborn children by angiographic evaluation of the size of each part of the ventricular system [80, 81]. I was deeply impressed, especially by the angiographic differentiation between communicating and noncommunicating hydrocephalus or aqueduct stenosis (Fig. 6).

457

I had a chance of doing cerebral angiography in children every other week and mastered the diagnosis not only of hydrocephalus, but also of common pediatric neurosurgical disorders. During my stay with Dr. Raimondi, he published his masterpiece Pediatric Neuroradiology (1972) [82]. This book influenced me so greatly that I began to be much more interested in neuroradiology from that time on.

this condition might be called “benign subdural effusion in infants” [48]. After the introduction of MRI, however, this condition was found to be located in the subarachnoid space rather than the subdural space. The condition can be complicated by subdural hematoma if the patient sustains a head injury during the process of its resolution. If care is taken clinically to avoid head trauma during the follow-up period, the course is usually benign without any surgical intervention [75].

“Shunt business” – an unexciting but important part of pediatric neurosurgery

A shunting operation is not as attractive to neurosurgeons as the surgical treatment of an aneurysm or an AVM. For this reason, in most cases of hydrocephalus shunts have been inserted by young neurosurgeons who have not finished their basic training. The more neurosurgeons do these operations, the more definitely they are convinced that a shunting operation is the most difficult. The shunt system must be functioning as a whole from the proximal to the distal end. Otherwise it falls into malfunction. For a shunting operation, an appropriately selected shunt system [18, 19] and a meticulous technique are needed. We are still waiting for an ideal shunt system. The incidence of infection is high in shunting operations, because foreign material is inserted into the body. I have encountered many complications related to shunting and have combated them [35, 37, 38, 72]. A shunting operation should not be considered easy.

Mystery of subdural fluid collections

Subdural pathology is still mysterious even in the MRI era. Subdural fluid collection may be given different names according to the nature of the fluid. Subdural hematoma (effusion) may be present together with hydrocephalus [43]. Subdural hematomas or effusions can have various kinds of etioloy and have been treated in different ways, such as burr hole irrigation, repeated subdural taps, subdural-peritoneal shunt and membranectomy. Subdural pathology has been demonstrated as an avascular area or clear space in a cerebral angiogram and as air filling space in pneumosubdurography. Since CT scan and MRI have been in use as diagnostic tools, small subdural fluid collections have become a new radiological entity, which cannot be shown by cerebral angioraphy [53], and patients with subdural pathology can now be followed easily. Infants whose developmental milestones were slightly delayed because of subdural fluid collection had been followed up by means of CT scans, and it was found that the fluid collection decreased as time went by and the head circumference and developmental milestones reached the normal ranges. Therefore, I thought that

Two main subjects in pediatric neurosurgery: hydrocephalus and meningomyelocele

Hydrocephalus is a multifactorial disease and is caused by various etiologies. Its pathogenesis has a wide spectrum. Different types of hydrocephalus have been known from the fetus to the aged. Each type of hydrocephalus has its own charateristics. Even in congenital hydrocephalus, hydrocephalus associated with meningomyelocele is different from congenital infantile hydrocephalus in pathogenesis [61]. Regardless of the causes and etiologies, shunting has revolutionized the treatment of hydrocephalus although it is not a radical treatment. Classification of hydrocephalus may reflect the progress in research on hydrocephalus. Revision of the classification and the evolution of a new classification system may be necessary for better understanding of the pathogenesis and treatment of hydrocephalus [22, 23, 26, 28, 30, 70]. We may encounter hydrocephalus that cannot be treated by a shunting operation. In many cases of “intractable hydrocephalus” severe brain dysgenesis or brain damage is present. In Japan, a research committee has been formed to investigate intractable hydrocephalus, sponsored by the Ministry of Health and Welfare of Japan. This committee is dedicated to clarifying the factors leading to intractability, and to preventing and treating intractable hydrocephalus [29, 31]. To clarify the pathogenesis and improve the prognosis of hydrocephalus, the application of neurological science to fundamental research on experimental hydrocephalus is needed [13–15, 57, 64, 66]. I believe that research of this kind will lead to the radical treatment of hydrocephalus in the near future. At present, only non-communicating hydrocephalus can be treated by III ventriculostomy using endoscopy without the insertion of a shunt. However, radical treatments will probably be devised using pharmacological or molecular biological approaches. Although the incidence of neural tube defect has declined recently, it is still a common problem in pediatric neurosurgery [21]. Most patients with meningomyelocele have Chiari type II malformation. It is clearly shown on the MR sagittal image. The natural history of brain stem function in meningomyelocele is not known. Some patients

458

manifest clinical symptoms of brain stem dysfunction immediately after birth and some present delayed symptoms. Fortunately, most patients have a favorable clinical course after early closure of the meningomyelocele and a shunting operation for hydrocephalus. In attempts to find whether prediction of the occurrence of symptoms might be possible, we have followed up patients with meningomyelocele by means of MRI and evoked potentials that have their tracts in the brain stem, such as auditory evoked potentials (AEP) or auditory brain stem response (ABR), somatosensory evoked potentials (SEP) and the blink reflex [34, 63, 76–78]. Because of the limited number of patients, no definite evidence predicting the manifestation of symptoms could be obtained. However, the following results on functional changes of the brain stem in meningomyelocele were obtained. Electrophysiological studies were found to be useful for the functional evaluation of the brain stem and lower cranial nerves in patients with meningomyelocele. Electrophysiological abnormalities of the brain stem did not always correlate with those of neuroimaging. The multiple recordings indicated gradual latency shortening of the brain stem components and latency prolongation of the peripheral ones. Interestingly, the brain stem function was found to be abnormal even in cases without Chiari malformation. Symptoms caused by brain stem dysfunction, such as apneic spells, can develop in cases with no MRI findings of Chiari malformation [79].

Children are not little adults

How is pediatric neurosurgery different from adult neurosurgery? The answer is that the patients we pediatric neurosurgeons treat have brains that are still developing. Common brain tumors in childhood are different from ones found in adults. According to a statistical analysis of cases from the Brain Tumor Registry of Japan, the three most common brain tumors in children are astrocytoma, medulloblastoma and craniopharyngioma [33]. Compared with the treatment results for craniopharyngioma in adults, the results of treatment are less favorable in pediatric cases. From the prognostic point of view, growth retardation in the pediatric cases is a problem that still remains to be solved [50]. Of the cases of craniopharyngioma that underwent radiotherapy, there was one in which radiation-induced vasculopathy [45] was experienced as a rare complication. The potential hazards of radiotherapy – radiation-induced occlusive changes in the circle of Willis – must be considered when benign basal brain tumors are treated in children. The results of our clinical analysis of arteriovenous malformations in children compared with adult cases showed that (1) intracerebral hematoma attributable to rupture of AVMs and focal neurological deficits were more frequent

in children. (2) Operative mortality was higher in children than in adults in cases where total removal was not feasible. (3) Follow-up results suggested that the prognosis in children was less favorable than in adults. (4) When dealing with spontaneous intracerebral hematoma in children, care should be taken to search for underlying cryptic AVMs [49].

How to treat pediatric cystic lesions

Arachnoid cysts are the most common benign cystic lesions in children. Arachnoid cysts can develop in any locality in which the arachnoid is present, but there are predictions for such sites as the sylvian fissure, the convexity, the interhemispheric fissure [25], suprasellar, paracollicular, cerebellopontine angle, retrocerebellar regions [41, 42], and spinal cord [58]. The treatment of arachnoid cysts still remains controversial. To operate or not to operate? To shunt or to fenestrate: what is the best surgical treatment for arachnoid cysts? Direct surgery on the cyst has the following merits: (1) it leaves the patient shunt-independent, (2) sufficient decompression and communication between the cyst and the subarachnoid space and/or ventricle can be obtained, (3) in patients without hydrocephalus, cyst fenestration is the treatment of choice, (4) the placement of a ventriculo-peritoneal (V-P) shunt followed by resection of the cyst wall may also give good results, especially in the presence of hydrocephalus, and (5) it facilitates inspection of the underlying brain. On the other hand, placement of a cystperitoneal (C-P) shunt for arachnoid cysts has the following advantages: (1) in the case of a huge cyst or one associated with marked hydrocephalus, a shunt avoids the kind of extreme brain shift that can follow sudden decompression, and (2) the shunting operation is less invasive. Treatment for arachnoid cysts associated with hydrocephalus may be problematic. Considering the complications of craniotomy, and the large number of patients who need shunts even after fenestration, a shunting operation may be the preferred initial surgical treatment for arachnoid cysts associated with hydrocephalus [25].

Characteristics of pediatric head injury

Head injury differs in children and adults. Children, especially infants, are susceptible to injuring their heads even in only slight falls. As the infant brain grows rapidly, metabolic disturbances of the brain occur easily, as does brain edema. Moreover, because cerebral function is still developing, localizing symptoms due to cerebral compression are delayed. Vomiting and generalized convulsions are common. When symptoms do occur, rapid aggravation is frequently observed. However, the prognosis is generally

459

more favorable than in adults. This is thought to be due to cerebral plasticity. Because the skull is soft and pliable, localized skull fractures such as Derby-hat-type depressed fracture (pingpong ball fracture) are common and coup injury often occurs. Moreover, even when the fracture develops, it is often asymptomatic. Diastatic fractures develop easily because the sutures are not closed. The dura mater is easily injured by a linear fracture of the cranial bones, which at times is the focus for a growing skull fracture. Subcutaneous hematoma, subgaleal hematoma, and subperiosteal hematoma develop easily because the adhesion between the periosteum and bone is loose. Subepicranial varix may develop [39]. However, extradural hematoma does not develop easily, because the adhesion between the dura mater and the internal surface of the bone is strong. In contrast, subdural hematoma (effusion) does develop easily. The onset of symptoms of epidural hematoma of the posterior fossa in children is not necessarily acute. Epidural hematomas of the posterior fossa in children have a tendency to liquefy earlier than supratentorial hematomas. The possibility of the presence of an epidural hematoma of the posterior fossa should be borne in mind in the treatment of head injury in children, even in the case of a minor head injury [56]. There is also a relatively long interval between the head injury and the onset of symptoms caused by an epidural hematoma in patients with long-standing arrested hydrocephalus. Such a large hematoma is accommodated by the decrease in size of the markedly dilated ventricles. Therefore, the possible presence of an epidural hematoma should also be borne in mind during the management of head injury in patients with long-standing arrested hydrocephalus [60].

Vascular malformations of the central nervous system: diagnosis and treatment

Besides arteriovenous malformations, cavernous and venous angiomas (malformations) are common types of vascular malformations. Cavernous angiomas can occur in any region of the brain. Most are located in the cerebral hemisphere. Intracranial cavernous angiomas can be diagnosed most accurately by CT scanning [9] and MRI. Extra-axial cavernous angiomas (cavernomas) may be encountered in the middle fossa [52]. Intracerebral cavernomas can be removed easily. Cavernomas in the middle fossa (“Japanese cavernomas”) must be considered separately, from both diagnostic and operative standpoints, from those in other locations. They are difficult to remove because of massive hemorrhage [52]. Venous angiomas are benign lesions and usually asymptomatic: they rarely cause seizures or hemorrhage. No specific findings can be seen on CT scans. The definite diag-

nosis should be made by cerebral angiography [73]. Venous angiomas in the posterior fossa may be different from those in the supratentorial compartment, in that the anatomy of the lesion seems to suggest a venous drainage anomaly that includes the absence of normal venous drainage and the presence of a compensating, anatomically anomalous, enlarged vein. Venous angiomas in the posterior fossa may represent anatomically disordered but physiologically essential venous drainage for the surrounding brain. Because venous angiomas in the posterior fossa are often associated with agenesis of the dural sinus and may provide a functional or potential route for venous drainage in the posterior fossa, they should be treated conservatively unless they are associated with cavernous or arteriovenous malformations, which tend to bleed [68, 71].

Pediatric stroke

Cerebral vascular accident is a rare condition in children. The most common cause of pediatric stroke in Japan is moyamoya disease. Moyamoya disease is a nonatherosclerotic, idiopathic occlusive basal vasculopathy leading to progressive parenchymal infarction. This disease is a radiological entity and involves occlusion of moderate-sized cerebral arteries at or near the circle of Willis. The age at onset varies, but the condition is most prevalent in young female persons. Familial occurrence is sometimes reported. Symptoms and types of progression vary, with an asymptomatic type and a type with transient or persistent neurological deficits of slight or severe degree. Cerebral ischemia is observed, usually in juvenile patients, and hemorrhage in adult patients. In juveniles, hemiparesis, monoparesis, sensory disturbance, involuntary movement, headache, and convulsion recur, sometimes on alternating sides. Mental retardation or persistent neurological deficits are also observed. The hemorrhagic type is rare in children [6, 27]. On cerebral angiograms, a typical case shows bilateral occlusion of the internal carotid artery at the bifurcation and a fine vascular network called a moyamoya (this name was coined by the late Professor Suzuki), which means “puff of smoke” in English, at the base of the brain. In addition to the deep collaterals, transdural collateral channels are seen as “natural” EC-IC bypasses. Such pathologic collaterals are seldom sufficient to prevent progressive parenchymal infarction. Multiple infarctions are often detected on CT or MRI. Unilateral occlusion of the internal carotid artery may progress to bilateral occlusion. Angiographic findings are different in each stage. In the late stage, cerebral blood may be supplied from extracranial arteries. When we discuss the indications for surgery in the treatment of moyamoya disease in children, we should consider its natural course. The natural history of moyamoya disease is influenced by three factors: (1) the blood flow nec-

460

essary to maintain normal functioning of the brain, (2) the extent and speed of the occlusive process, and (3) development of collateral channels. The course and type of this disease depend on a combination of these factors. According to a CBF study on children, the brains of younger children need a larger amount of blood flow than do adult brains. Therefore, patients under the age of 2 years often develop infarction. The total amount of circulating blood in the brain is the sum of the blood flow through the decreased natural route and that developing via collateral channels. Most patients with moyamoya disease present symptoms of ischemia because blood flow via collateral channels cannot compensate for the decrease of blood flow via natural routes. If the decrease in the total amount of CBF is slight the patient may remain asymptomatic or may develop TIA or RIND on taking exercise that would normally be adequate to increase the CBF. Infarcted areas increase in size as the disease progresses and the amount of circulating blood flow eventually reaches the level needed to maintain brain function both by development of collateral channels and infarction. This compensated condition can be obtained by sacrificing part of the normal brain. The patients who have attained apparent cure often show delayed mental development. They often present with convulsive seizures. Revascularization by surgical treatment may shorten the period of ischemic stroke leading to infarction. Early revascularization immediately after the onset before development of infarction may increase CBF and hence improve mental development. Moyamoya disease has been extensively investigated by the Research Committee on Spontaneous Occlusion of the Circle of Willis sponsored by the Ministry of Health and Welfare of Japan since 1977. The results of treatment were evaluated in 595 patients who had been registered from 1983 through 1990 and intensively followed up [84]. Of these 595 cases, 541 were definite cases of moyamoya disease. The mean duration of follow-up for these cases was 4.3 years. Patients with moyamoya disease generally have a good prognosis: more than 80% of them are in good health or living independently, irrespective of the treatment received, but many patients are not well accommodated in school or in their social life owing to lower IQ, psychological impairment and/or personality changes. A detailed analysis of the patients who underwent surgical treatment revealed that the ADL score was improved in patients who received both direct and indirect revascularization together. The proportion of patients with good scores on ADL and in good health increased during the follow-up period. By contrast, patients who only underwent indirect revascularization or received medication did not show this tendency. Further assessment is necessary to find how effective surgical treatment can be in improving IQ. With the exception of moyamoya disease, cerebral artery occlusion is rare in childhood. Basilar artery occlu-

sion is very rare [46]. The main clinical manifestations of basilar artery occlusion are disturbed consciousness, hemiplegia or quadriplegia, and pupillary abnormalities. The prognosis is better in children than in adults.

What are congenital brain tumors?

Congenital tumors of the central nervous system are frequent. Although many congenital brain tumors do not cause symptoms until later in life, they develop from congenitally misplaced or abnormally developed tissue. The mechanism of development is not clear in some tumors of prenatal origin. Consequently, different neuropathologists classify congenital brain tumors differently. We classified congenital brain tumors according to their characteristics [17, 32]: (1) embryonic tumors – epidermoid [55], dermoid and teratoma [83], (2) germ-cell tumors, (3) neuroblastic tumors, (4) tumors related to embryonal remnant tissues – craniopharyngioma [50, 51], Rathke cleft cyst [1] and chordoma, (5) tumors affected by genetic factors – tumors associated with neurocutaneous syndrome, (6) colloid cyst of the III ventricle, (7) heterotopia and hamartoma [10, 20, 54], (8) lipoma [3, 4, 62] and (9) vascular tumors. Congenital brain tumors grow slowly and are relatively benign in most cases. They may threaten life, however, when they occur in certain locations. The word congenital has its origin in the latin congenitus (born together) and means “present at the time of birth and usually before birth.” Clinical diagnosis of congenital brain tumors is not always simple; the following three groups are generally included in classification systems for such tumors: (1) tumors already causing symptoms at birth or during the neonatal period (definitely congenital tumors); (2) tumors causing symptoms and diagnosed during infancy (probably congenital tumors); and (3) tumors diagnosed after infancy, with the onset of symptoms during infancy (possibly congenital tumors). Some investigators stress age at diagnosis, rather than at symptom onset, as a diagnostic criterion of congenital brain tumors.

Anomalies of the central nervous system – neurosurgical malformations

Malformations of the central nervous system are common disorders in the pediatric neurosurgical field. I have been collecting cases with anomalies of the central nervous system for the past quarter of a century, starting during my time in Chicago, and have published the collected cases in a monograph [32]. The English edition of this volume [17] was published with the help of Dr. Derek Harwood-Nash

461

Fig. 7 The late Professor Derek C. Harwood-Nash

(Fig. 7). The use of CT has been adopted with great enthusiasm for the diagnosis of anomalies of the central nervous system. However, about 20 years have elapsed since its introduction, and it is becoming increasingly obvious that CT is not always the most effective method of investigation. As a result of progress in computer technology, various methods of neuro-imaging have come into existence. One type in current use is magnetic resonance imaging (MRI). Soon after its introduction, it was discovered that alterations in pulse sequences resulted in significant changes in the images, and this was considered to be a disadvantage at the time. The difficulties in obtaining more information from MRI than from CT delayed its becoming well established as a clinical procedure. Recently, various new and effective MRI techniques have emerged: with the introduction of pulse sequences it is now possible to obtain information from MRI that it was previously impossible to obtain by CT scan alone. Realization of this has led to the establishment of this new technology in many institutions. Changes in images resulting from alterations in the pulse sequences facilitate tissue characterization, which in turn enhances the differentiation of lesions that could not previously be diagnosed. In the past, CT was the first choice among diagnostic measures for anomalies of the central nervous system, and MRI was regarded as a supplementary test, filling in what was lacking in CT. However, with the introduction of an everincreasing number of devices, MRI is being used routinely, and it is possible that eventually CT will become supplementary to MRI [24].

Because the stages of development of the central nervous system (CNS) are of long duration, lasting from the early stage of formation of the neural tube to the perinatal period, a disorder of organogenesis can lead to a wide variety of cerebral malformations. Most morphological anomalies are produced during the 8 weeks of the embryonic stage, and in general the earlier the disorder occurs the more severe is the malformation. Anomalies of the CNS may be classified arbitrarily into the following three groups according to the development of the CNS: (1) genetic disorders, (2) disorders in the embryo and fetus, and (3) perinatal disorders. Genetic disorders are brought about during the course of inheritance and fertilization. An example of a genetic disorder is Down syndrome. In general, genetic disorders cause metabolic abnormalities, and rarely malformations. Chromosomal disorders often cause mental abnormalities. Microcephaly caused by chromosomal abnormalities does not involve any specific brain malformations. The stage up to 8 weeks after fertilization, during which the bone marrow in the humeral bone develops, is called the embryonic stage. In this stage, disorders mainly cause abnormalities in the development of tissues. Disorders such as neurocutaneous syndrome [11] are common in this stage. Disorders during organ formation cause malformation, defective formation, or deformation. Most morphological abnormalities occur during this stage. Common anomalies such as holoprosencephaly [8] and Chiari malformation [44] are kinds of embryopathy. After growth, the embryo becomes a fetus. Disorders in the fetal stage are referred to as fetopathies.

Difficulty in treating craniosynostosis

Craniosynostosis can be divided into primary and secondary types. Primary craniosynostosis results from an intrinsic abnormality of the cranial sutures. There are two types of secondary craniosynostosis: (1) craniosynostosis as a part of other known syndromes (syndromic craniosynostosis) and (2) craniosynostosis in association with other conditions. The pathogenesis of craniosynostosis is still not clear. Familial occurrence is often seen, and genetic factors may play a part, especially in coronal synostosis. Craniosynostosis is often associated with skull base deformity. The combined deformities are called craniofacial dysostosis. Treatment of craniosynostosis has been directed toward creating sutures surgically to facilitate the expansion of the skull. Different types of reconstructive cranioplasty have been reported for the treatment of craniosynostosis. However, no single method has been enough to treat it. During the long-term follow-up most patients need reoperation. We tried to treat different types of craniosynostosis by the following strategy: for early correction of craniosynos-

462

tosis in the newborn or in early infancy, strip craniectomy may produce satisfactory results. For late correction, however, more radical cranial reconstruction procedures are essential to achieve adequate remodeling of cranial deformities and normalization of intracranial pressure. Craniosynostosis presents various types of cranial deformities, and different procedures for its correction have been reported. Expanding cranioplasty is presented for the increase in cranial volume and immediate correction of the cranial deformity [65, 69]. The results of treatment are not always appreciated by the patients. Many patients with secondary craniosynostosis, especially of the sydromic type, are not good candidates for operation. Even if staged operations are performed, the results of operation for craniofacial deformities are not usually satisfactory. Methods of treatment of craniosynostosis should be considered on a case-by-case basis. Most cases need staged reconstruction: primary cranio-orbital decompression and reshaping, followed by craniotomy and reshaping again if necessary, correction of midface deformity at school age, and correction of orthognathic deformity at puberty. At present, surgical correction of craniosynostosis cannot solve its biological problems. Therefore, multiple operations are necessary when it is treated.

Epilogue – unsolved problems in pediatric neurosurgery: Quo vadis pediatric neurosurgery?

A shunting operation is an excellent method of treatment for hydrocephalus, but it is not a radical treatment. Re-

search should be undertaken to devise a radical treatment for hydrocephalus without shunt placement. III Ventriculostomy using endoscopy is accepted as an excellent treatment for hydrocephalus. However, this method is applied only for non-communicating hydrocephalus. For radical treatment of hydrocephalus, research at the molecular level may be necessary. Congenital anomalies of the CNS can be detected in utero by ultrasound. Fetal surgery for CNS anomalies is not promising at the present time. Genetic counseling should be practiced widely in order to prevent CNS anomalies. New measures should be devised to repair neuronal damage caused by various pathologies and brain damage due to trauma. For this purpose, measures allowing regrowth of damaged neurons should be sought. In the near future, great progress in neuroimaging can be expected; magnetic resonance angiography (MRA) may take the place of conventional cerebral angiography. Functional imaging will be available in the daily practice of pediatric neurosurgery. Magnetic resonance spectroscopy will be refined and used more widely. In the future, pediatric neurosurgery will move on to less invasive treatments. Radiosurgery and intravascular surgery will occupy more space than microsurgery. Brain tumors may be treated by gene therapy combined with other treatment modalities. Disorders in pediatric neurosurgery will be treated by minimally invasive surgery in conjunction with molecular neurosurgery. Acknowledgements The author wishes to express his gratitude to Mr. Daniel B. Ribble, Kochi Medical School, for his review of this manuscript, and to Miss Miki Tamaki for secretarial assistance.

References 1. Arisawa M, Kurisaka M, Mori K, Yagyu K (1984) Ultrastructural studies of Rathke’s cleft cyst. J Clin Electron Microsc 17:5–6 2. Asato R, Murata T, Mori K, Handa H (1981) NMR: its application to the experimental study of hydrocephalus and brain edema. Brain Nerv (Jpn) 33:603–609 3. Eghwrudjakpor PO, Kurisaka M, Fukuoka M, Mori K (1991) Intracranial lipomas. Acta Neurochir (Wien) 110:124–128 4. Eghwrudjakpor PO, Kurisaka M, Fukuoka M, Mori K (1992) Intracranial lipomas: current perspectives in their diagnosis and treatment. Br J Neurosurg 6:139–144

5. Handa H, Mori K (1968) Large varix of the superior ophthalmic vein: demonstration by angular phlebography and removal by electrically induced thrombosis. J Neurosurg 29:202–205 6. Handa H, Mori K (1978) The “moyamoya” syndrome. Seara Medica Neurocirurgica 7:249–264 7. Ishikawa M, Handa H, Mori K, Matsuda I (1979) “Moyamoya” vessels on the tumor in the sellar region. Arch Jpn Chir 48:639–644 8. Ishikawa M, Handa H, Osaka K, Mori K, Matsuda I (1980) Holoprosencephaly with severe hydrocephalus. Arch Jpn Chir 49:119–128 9. Ishikawa M, Handa H, Moritake K, Mori K, Nakano Y, Aii H (1980) Computed tomography of cerebral cavernous hemangiomas. J Comput Assist Tomogr 4:587–591

10. Kurisaka M, Eghwurdjakpor PO, Mori K (1987) Ultrastructural studies of hypothalamic hamartoma. J Clin Electron Microsc 20:5–6 11. Kurisaka M, Mori K, Moriki T (1989) A case of meningioma in the anterior part of third ventricle with tuberous sclerosis. J Clin Electron Microsc 22:807–880 12. Milhorat TH, Hammock MK (1971) Isotope ventriculography. Interpretation of ventricular size and configuration in hydrocephalus. Arch Neurol 25:1–8

463

13. Miyake H, Eghwrudjakpor PO, Sakamoto T, Kurisaka M, Mori K (1991) Neurotransmitter changes in hydrocephalus: effects of cerebral metabolic activator on kaolin-induced hydrocephalus. In: Matsumoto S, Tamaki N (eds) Hydrocephalus – pathogenesis and treatment. Springer, Tokyo Berlin Heidelberg, pp 68–74 14. Miyake H, Eghwrudjakpor PO, Sakamoto T, Mori K (1992) Catecholamine alterations in experimental hydrocephalus. Child’s Nerv Syst 8:243–246 15. Miyake H, Sakamoto T, Mori K (1992) Changes of neurotransmitters in experimental hydrocephalus. Curr Treatment Hydrocephalus (Tokyo) 2:5–10 16. Mori K (1967) Experimental studies on electrically induced arterial thrombosis in dogs, with special reference to the treatment of intracranial aneurysms and arteriovenous malformations. Arch Jpn Chir 36:35–62 17. Mori K (1985) Anomalies of the central nervous system: neuroradiology and neurosurgery. Thieme Stratton, New York / Thieme, Stuttgart 18. Mori K (ed) (1986) CSF shunt system. Excerpta Medica, Tokyo 19. Mori K (1987) Shunt characteristics. In: Raimondi AJ (ed) Pediatric neurosurgery, theoretic principles, art of surgical techniques. Springer, New York Berlin Heidelberg, pp 483–489 20. Mori K (1987) Is an operation of hypothalamic hamartoma for the treatment of precocious puberty feasible? Proceedings of the First International Symposium of Pediatric Neurooncology, pp 184–187 21. Mori K (1988) Management and prognosis of neural tube defects. Congen Anom 28 [Suppl]:193–203 22. Mori K (1990) Hydrocephalus – revision of its definition and classification with special reference to “intractable infantile hydrocephalus”. Child’s Nerv Syst 6:198–204 23. Mori K (1991) Hydrocephalus – revision of its classification. In: Matsumoto S, Tamaki N (eds) Hydrocephalus – pathogenesis and treatment. Springer, Tokyo Berlin Heidelberg, pp 362–368 24. Mori K (ed) (1991) MRI of the central nervous system. A pathology atlas. Springer, Tokyo Berlin Heidelberg 25. Mori K (1992) Giant interhemispheric cysts associated with agenesis of the corpus callosum. J Neurosurg 76:224–230

26. Mori K (ed) (1993) Proceeding of the Symposium on Hydrocephalus: Current Concepts by the Research Committee on Intractable Hydrocephalus sponsored by the Ministry of Health and Welfare of Japan, Tokyo (in Japanese with English abstract) 27. Mori K (1993) Treatment of moyamoya disease. The First Postgraduate Course in Pediatric Neurosurgery, 9–10 April, Seoul 28. Mori K (1994) Definition, classification, and diagnostic criteria of hydrocephalus. Journal of the Citizen Ambasador Program: Teratology Delegation to the People’s Republic of China 29. Mori K (ed) (1994) Proceedings of the Symposium on Intractable Factors in Hydrocephalus by the Research Committee on Intractable Hydrocephalus sponsored by the Ministry of Health and Welfare of Japan (in Japanese with English abstract) 30. Mori K (1995) Current concept of hydrocephalus: evolution of new classification. Child’s Nerv Syst 11:523–532 31. Mori K (ed) (1993–1996) Annual Report by the Research Committee on Intractable Hydrocephalus sponsored by the Ministry of Health and Welfare of Japan, Tokyo (in Japanese with English abstract) 32. Mori K, Handa H (1979) Clinics and CT of congenital anomalies (in Japanese). Neuron Press, Tokyo 33. Mori K, Kurisaka M (1986) Brain tumors in childhood – statistical analysis of cases from the brain tumor registry of Japan. Child’s Nerv Syst 2:233–237 34. Mori K, Nishimura T (1995) Electrophysiological studies on brainstem function in patients with myelomeningocele. Pediatr Neurosurg 22:120–131 35. Mori K, Raimondi AJ (1975) An analysis of external ventricular drainage as a treatment for infected shunts. Child’s Brain 1:243–250 36. Mori K, Raimondi AJ (1975) Submicroscopic changes in the periventricular white matter of hydrocephalus ch mouse. Arch Jpn Chir 44:159–168 37. Mori K, Yamashita J, Handa H (1975) „Missing tube“ of peritoneal shunt – migration of the whole system into the ventricle. Surg Neurol 4:57–59 38. Mori K, Handa H, Muroya H (1976) Treatment of shunt infection: conversion of external ventricular drainage to gastric shunt for persistent ventriculitis. Arch Jpn Chir 45:398–400 39. Mori K, Yoneda S, Handa H (1976) Posttraumatic subepicranial varix. Surg Neurol 5:337–339 40. Mori K, Murata T, Nakano Y, Handa H (1977) Periventricular lucency in hydrocephalus on computerized tomography. Surg Neurol 8:337–340

41. Mori K, Hayashi T, Handa H (1977) Radiological manifestation of infratentorial retrocerebellar cysts. Neuroradiology 13:201–207 42. Mori K, Hayashi T, Handa H (1977) Infratentorial retrocerebellar cysts. Surg Neurol 7:135–142 43. Mori K, Hayashi T, Handa H (1977) Subdural haematoma (effusion) and internal hydrocephalus. Neurochirurgia 20:154–161 44. Mori K, Handa H, Okuno T, Hazama F (1978) Arnold-Chiari type II malformation. A clinicopathological study with special reference to its surgical treatment. Neurochirurgia 21:9–14 45. Mori K, Takeuchi J, Ishikawa M, Handa H, Toyama M, Yamaki T (1978) Occlusive arteriopathy and brain tumor. J Neurosurg 49:22–35 46. Mori K, Miwa S, Murata T, Okumura A, Handa H (1979) Basilar artery occlusion in children: report of a case. Arch Neurol 36:100–102 47. Mori K, Handa H, Murata T, Nakano Y (1980) Periventricular lucency in computed tomography of hydrocephalus and cerebral atrophy. J Comput Assist Tomogr 4:204–209 48. Mori K, Handa H, Itoh M, Okuno T (1980) Benign subdural effusion in infants. J Comput Assist Tomogr 4:466–471 49. Mori K, Murata T, Hashimoto N, Handa H (1980) Clinical analysis of arteriovenous malformations in children. Child’s Brain 6:13–25 50. Mori K, Handa H, Murata T, Takeuchi J, Miwa S, Osaka K (1980) Results of treatment for craniopharyngioma. Child’s Brain 6:303–312 51. Mori K, Handa H, Ishikawa M, Takeuchi J, Osaka K (1980) Craniopharyngioma with unusual topography and associated with vascular pathology. Acta Neurochir (Wien) 53:53–68 52. Mori K, Handa H, Gi H, Mori K (1980) Cavernomas in the middle fossa. Surg Neurol 14:21–31 53. Mori K, Handa H, Itoh M, Okuno T (1981) Differential diagnosis and treatment of small subdural effusion in children. Neurol Med Chir (Tokyo) 21:121–125 54. Mori K, Handa H, Takeuchi J, Hanakita J, Nakano Y (1981) Hypothalamic hamartoma. J Comput Assist Tomogr 5:519–521 55. Mori K, Handa H, Moritake K, Takauchi J, Nakano Y (1982) Suprasellar epidermoid. Neurochirurgia 25:138–142

464

56. Mori K, Handa H, Munemitsu H, Oda Y, Hashimoto N, Kojima M (1983) Epidural hematomas of the posterior fossa in children. Child’s Brain 10:130–140 57. Mori K, Fujito K, Kamimura Y (1984) Binding assay for muscarinic cholinergic receptors in kaolin induced hydrocephalus. Arch Jpn Chir 53:695–702 58. Mori K, Oda Y, Lee YS, Handa H (1985) Symptomatic intradural spinal arachnoid diverticulum: report of a case and review of the literature. Arch Jpn Chir 54:372–377 59. Mori K, Seike M, Kurisaka M, Sada Y (1985) CT documentation of the growth of craniopharyngioma. J Pediatr Neurosci 1:210–214 60. Mori K, Morimoto M, Kamimura Y (1986) Post-traumatic epidural hematoma in two patients with long-standing “arrested” hydrocephalus – report of two cases. Child’s Nerv Syst 1:288–290 61. Mori K, Kamimura Y, Kurisaka M Uchida Y, Eguchi S (1986) ICP monitoring in long-standing hydrocephalus associated with myelomeningocele. J Pediatr Neurosci 2:195–204 62. Mori K, Kamimura Y, Uchida Y, Eguchi S (1986) Large intramedullary lipoma of the cervical cord and postrior fossa. J Neurosurg 64:974–976 63. Mori K, Uchida Y, Nishimura T, Eghwrudjakpor PO (1988) Brainstem auditory evoked potentials in Chiari II malformation. Child’s Nerv Syst 4:154–157 64. Mori K, Miyake H, Sakamoto T (1991) Changes of neurotransmitter metabolites in the cerebrospinal fluid and prediction of the prognosis of shunted hydrocephalus (in Japanese with English abstract). Mod Neurosurg 2:421–429

65. Mori K, Sakamoto T, Nakai K (1992) Expanding cranioplasty for craniosynostosis and allied disorders. Child’s Nerv Syst 8:399–405 66. Mori K, Miyake H, Kurisaka M, Sakamoto T (1993) Immunohistochemical localization of superoxide dismutase in congenital hydrocephalic rat brain. Child’s Nerv Syst 9:136–141 67. Mori K, Sakamoto T, Nishimura K, Fujiwara K (1993) Subarachnoid fluid collection in infants complicated by subdural hematoma. Child’s Nerv Syst 9:282–284 68. Mori K, Seike M, Kurisaka M, Kamimura Y, Morimoto M (1994) Venous malformation in the posterior fossa: guidline for treaatment. Acta Neurochir (Wien) 126:107–112 69. Mori K, Sakamoto T, Nakai K (1995) The surgical management of scaphocephaly. In: Goodrich JT, Hall CD (eds) Craniofacial anomalies: growth and development from surgical perspective. Thieme, Stuttgart New York, pp 23–32 70. Mori K, Shimada J, Kurisaka M, Sato K, Watanabe K (1995) Classification of hydrocephalus and outcome of treatment. Brain Dev 17:338–348 71. Mori K, Seike M, Kurisaka M, Kamimura Y, Morimoto M (1996) Treatment of venous malformation in the posterior fossa. In: Hakuba A (ed) Surgery of the intracranial venous system. Springer, Tokyo, pp 533–537 72. Mori T, Arisawa M, Fukuoka M, Tamura K, Kurisaka M, Mori K (1993) Management of a broken arterial catheter migrated into the heart: a rare complication of ventriculo-atrial shunt – case report. Neurol Med Chir (Tokyo) 33:713–715 73. Moritake K, Handa H, Mori K, Ishikawa M, Morimoto M, Takebe Y (1980) Venous angiomas of the brain. Surg Neurol 14:95-105 74. Murata T, Handa H, Mori K, Nakano Y (1981) The significance of periventricular lucency on computed tomography: experimental study with canine hydrocephalus. Neuroradiology 20:221–227

75. Nishimura K, Mori K, Sakamoto T, Fujiwara K (1996) Management of subarachnoid fluid collection in infants based on a long-term follow-up study. Acta Neurochir (Wien) 138:179–184 76. Nishimura T, Mori K (1996) Blink reflex in meningomyelocele, with special reference to its usefulness in the evaluation of brainstem dysfunction. Child’s Nerv Syst 12:2–12 77. Nishimura T, Mori K (1996) Somatosensory evoked potentials to median nerve stimulation in meningomyelocele: what is occurring in the hindbrain and its connections during growth? Child’s Nerv Syst 12:13–26 78. Nishimura T, Mori K, Uchida Y, Ohira T, Tamura K (1991) Brainstem auditory-evoked potentials in meningomyelocele. Natural history of Chiari II malformations. Child’s Nerv Syst 7:316–326 79. Nishimura T, Mori K, Sada Y, Fujii M (1995) Apneic spells in a patient with myelomeningocele without Chiari type II malformation. Case report. Neurol Med Chir (Tokyo) 35:876–881 80. Raimondi AJ (1969) Angiographic diagnosis of hydrocephalus in the newborn. J Neurosurg 31:550–560 81. Raimondi AJ (1971) The angiographic evaluation of ventricular size in the hydrocephalic newborn. Prog Neurol Surg 4:1–53 82. Raimondi AJ (1972) Pediatric neuroradiology. Saunders, Philadelphia 83. Takeuchi J, Mori K, Moritake K, Tani F, Waga S, Handa H (1975) Teratomas in the suprasellar region: report of five cases. Surg Neurol 3:247-255 84. Yonekawa Y, Kawano T (1991) Follow-up study of 595 cases in spontaneous occlusion of the circle of Willis registered from 1983 to 1990. In: Yonekawa Y (ed) Annual Report (1990) by the Research Committee on Spontaneous Occlusion of the Circle of Willis. Ministry of Health and Welfare, Tokyo, pp 23–29