ISSN (print): 2068-8040

prepress / press / postpress

The Romanian Journal of Child and Adolescent Neurology and Psychiatry

< Return to the archive of articles


Authors: A. V. Ciurea, Claudiu Palade


Embryonic tumors account for approximately 25 per cent of all childhood brain tumors.1 They originate from pluripotent progenitor cells of the central nervous system (CNS) and share similar characteristic features of morphology and biology. They are classified by the World Health Organization (WHO) as highly malignant (grade IV) tumors.2 

Primitive neuroectodermal tumor (PNET) is the most frequent embryonic tumor.About 85 per cent of PNETs arise in the cerebellum, where they are referred to as medulloblastomas.



Medulloblastoma is the most frequent malignant brain tumor of childhood, with an annual incidence of 0.5–0.7 cases per 100 000 children under the age of 15 years (peak five to seven years) and a slight male predominance of 1.1–1.7 to one.3 The classical medulloblastoma usually arises from the cerebellar vermis. The desmoplastic typearises more often from the cerebellar hemispheres and occurs more frequently in adolescents and adults.4 Towards the end of the twentieth century, the population survival rate at ten years remained below 50 per cent for all children with medulloblastoma. 5–7 

With modern techniques of neurosurgery and radiotherapy, between 50 and 60 per cent of these children are expected to be alive and free of progression five years following diagnosis.8,9 The addition of a three-drug adjuvant chemotherapy regimen has resulted in five-year progression-free survival (PFS) rates of approximately 80 per cent or higher, in particular in children with localized disease.10,11 Unfortunately, children who survive medulloblastoma are at high risk of developing serious neurocognitive, endocrine, and neuropsychological deficits. This is related to a variety of factors, including the direct and indirect effects of the tumor and the effects of surgery. However, one of the most significant factors is the brain irradiation received.12–14 The long-term sequelae are more dramatic in young children who received high doses of wholebrain irradiation.15–17



By definition, medulloblastoma is a PNET arising in the cerebellum. The majority (80 per cent) arise in the vermis and the median part of the cerebellum; only 20 per cent arise in the cerebellar hemispheres. The hemispheric location is proportionally more frequent in adults, and the desmoplastic histological form is more frequent in this situation. Locally, dissemination of medulloblastoma very often involves the fourth ventricle, sometimes reaching the supratentorial area through the aqueduct of Sylvius. It may involve the brainstem, the most frequent area of involvement being the floor of the fourth ventricle. Involvement deep within the brainstem is less frequent. Laterally, cerebellar peduncles, and rarely the cerebellopontine angles, may also be involved. 

Meningeal and cerebrospinal fluid (CSF) dissemination is frequent in this disease. Local meningeal involvement may be observed within the posterior fossa, and CSF dissemination may also occur through the involvement of the fourth ventricle. The most frequent distant metastatic sites are in the subarachnoid space, either in the spinal canal and/or in the supratentorial area. Parenchymal metastases within the brain or the spinal cord are possible but less frequent. Metastases outside the CNS are possible in medulloblastoma but are observed very rarely at diagnosis. The less rare sites of hematogenous metastasis are bone and bone marrow. Dissemination within the peritoneal cavity following ventriculoperitoneal shunting has been described but is rare.



Correct staging in medulloblastoma is very important in order to make appropriate therapeutic decisions. Postoperative imaging of the tumor bed is often difficult to interpret because of the frequent uncertainty regarding the significance of postsurgical abnormalities on imaging. The best early postoperative imaging is magnetic resonance imaging (MRI)-Fig. 1, but computed tomography (CT) scanning is often employed. 

It is important to have an early postoperative exam with and without contrast in order to attempt to differentiate non-pathological postoperative changes from residual tumor. Residual disease is best demonstrated by comparing the patient’s preoperative MRI imaging with that obtained postoperatively. It is accepted that the postoperative scan is best performed between 24 and 72 hours after surgery, after which postoperative changes render interpretation of residual disease difficult.

Figure. 1. MRI (T1 sequence). Horizontal plane. 4 year old patient (personal case)


With regard to local disease, several recent series have demonstrated the prognostic importance of achieving a gross total or near gross total surgical excision. 18 This was demonstrated clearly by the North American Children’s Cancer Group CCG-921 study, which showed a survival advantage for patients having less than 1.5cm2 residual disease on postoperative imaging compared with those patients with 1.5cm2 or more of residual disease.19 

Thus, presently the Children’s Oncology Group (COG) defines standard-risk patients in respect of local disease as those having 1.5cm2 or less of residual disease after surgery. However, the definition of local residual disease has been heterogeneous in the literature, which makes it still more difficult to know the prognostic value of this parameter. It is possible that prognosis may be affected adversely by residual tumor volume of more than 1.5 cm 2, 19 more than 1.5 cm 3, 20 more than 50 per cent of the preoperative tumor volume more 11, or no identifiable or measurable residual tumor on early postoperative imaging without and with contrast enhancement.21 The Chang staging classification is the most widely used system for medulloblastoma: 

Chang staging system for medulloblastoma

-M1: positive CSF cytology.

-M2: meningeal metastases within the posterior fossa or supratentorial area.

-M3: spinal-canal metastases.

-M4: metastases outside the CNS. 

With regard to the extent of disease, the presence of metastatic disease at presentation as diagnosed by the presence of meningeal enhancement on MRI of the brain (Chang stage M2) or spine (Chang stage M3) clearly carries a poor prognosis.22 In the MRI era, metastases are now looked for using the craniospinal MRI images. 

The use of preoperative spinal axis MRI scanning in the case of a posterior fossa tumor has been advocated in order to avoid postoperative artifacts in the cervical region. Goodquality imaging avoiding movement artifacts and incomplete staging (especially the missing of cervical or lower end of the thecal sac) is mandatory in order to determine the appropriate risk group. 

The prognostic significance of Chang stage M1 disease, in which tumor cells are found within the CSF and without radiological evidence of metastasis, is lessclear, although several studies have shown that patients with M1 disease do have a worse prognosis than those without evidence of such tumor spread.19,23,24 

Other imaging examinations need not be used at diagnosis in the usual clinical presentation of medulloblastoma. Indeed, bone or bone marrow metastases at diagnosis are rare, thus isotopic bone scanning and bone marrow examination should not be systematically studied initially.

Summary of diagnostic procedures for children with medulloblastoma:

-Pre- and postoperative brain MRI, without andwith contrast.

-Full spinal axis MRI (if possible preoperatively).

-Postoperative CSF cytology, through lumbar puncture (usually about 15 days after surgery).



Collaborative groups now generally classify patients according to the presence or absence of prognostic factors: 

Medulloblastoma risk groups 

-Standard risk: _1.5cm2 postsurgical residual disease on MRI scan 24–72 hours after surgery, no metastase, and negative CSF cytology (M0).

-High risk: _1.5cm2 postsurgical residual disease on MRI scan 24–72 hours after surgery, and/or metastases (M1–4).



Headaches, morning vomiting on an empty stomach resulting in transient improvement of the headaches, and lethargy are often the earliest clinical signs and are caused by increased intracranial pressure. Unfortunately, these symptoms are frequently not recognized early, and less than half of children with medulloblastoma are diagnosed within four weeks, compared with about 80 per cent of children with acute lymphoblastic leukemia (ALL) or nephroblastoma.25 Some children present with only declining academic performance at school or personality changes. 

Further typical signs and symptoms are ataxic gait, unsteadiness, and squint due to a sixth cranial nerve palsy. Symptoms due to local tumor invasion within the posterior fossa include dysarthria, dysphonia, dysphagia (deficits of cranial nerves), and gait disturbance (invasion of long tracts). In infants, developmental delay, irritability, increasing head circumference, and the “setting sun sign” are characteristic features.



Good clinical practice includes prospective evaluation of sensorineural and neurocognitive impacts of the disease. Ophthalmic assessment, including looking for papilledema and visual acuity, should be performed initially, at least in the postoperative period, and repeated later because of the frequency of raised intracranial pressure at diagnosis. Furthermore, because of treatment morbidity (radiation therapy as well as chemotherapy),26 hearing should be studied initially and repeated.27 Assessment of neurocognitive function initially and prospectively during follow-up appears necessary in medulloblastoma, although optimal timing is not yet determined. Because of the usual delay of cognitive function alteration, and because of the difficulty in assessingcognition in the preoperative or early postoperative period, it is often proposed that these evaluations should be performed at least a few weeks after surgery.However, a better understanding of these alterations in cognitive function necessitates a thorough prospective evaluation with longterm follow- up.15,17 

Health Utility Index (HUI) methodology has also been proposed to prospectively evaluate neurocognitive sequelae28,29 and to screen for the most important late effects.15 Finally, although prospective neuroendocrine assessment is part of the good clinical practice in children followed up after treatment for medulloblastoma, it does not appear necessary to have an early postoperative evaluation. Long-term endocrine follow-up starting at one or two years after surgery is usually sufficient to detect neuroendocrine damage, and a baseline evaluation soon after diagnosis does not seem to be justified.30–32



Medulloblastoma is a highly malignant tumor that cannot be cured by neurosurgical resection.33 Even a so-called total resection is not a radical resection in terms of general oncological standards. However, the disease is characterized by its radiosensitivity and chemosensitivity, and adjuvant non-surgical treatment is essential. 

Over the past three or four decades, it has become standard for children to receive craniospinal radiotherapy (CSRT). The omission of CSRT results in a dramatic increase of relapses and a poor outcome, because in almost all patients at least occult microscopic dissemination along the CSF pathway has to be assumed. 34,35 Although medulloblastoma is radiosensitive, the potential for long-term sequelae associated with craniospinal irradiation is a limiting factor, particularly in young children. Medulloblastoma is also a chemosensitive tumor.36–38 

However, the blood–brain barrier in the tumor area adjacent to the normal brain and the blood– CSF barrier remains a therapeutic problem, when craniospinal irradiation should be reduced, delayed, or even omitted. The major targets of our efforts are not only to increase the cure rates of our patients but also to ensure that children cured of a medulloblastoma are healthy and will have a place within our society and not on its fringe. Supportive care and individual rehabilitation have to be part of the modern management of these children to compensate for the neurological, intellectual, endocrine, psychological, and social deficits of long-term survivors for a better quality of life.



1)Treatment of hydrocephalus

Raised intracranial pressure is by far the most common presentation for children with brain tumors. 

This is usually secondary to hydrocephalus rather than a primary effect of the size of the tumor itself. The majority of pediatric tumors occur in the midline and obstruct the fourth ventricle, the aqueduct of Sylvius, or the third ventricle. This type of hydrocephalus is therefore termed “obstructive.” In contrast, blood or infection in the cerebrospinal fluid (CSF) pathways can interfere with the absorption of the CSF; this type of hydrocephalus is termed “communicating.” Before the advent of modern imaging, it was not uncommon for children to present late with a long history of headache and vomiting and to be dehydrated on admission. 

When shunts became available in the 1950s, it became routine practice to treat the hydrocephalus with a shunt before tumor surgery. This practice is no longer necessary as, in general, tumors are diagnosed much earlier and the child is therefore usually in a better clinical state. 

Overall, only about one-third of posterior fossa tumors now require a permanent shunt, almost always receiving this at some point in the postoperative period. Modern management consists of commencing the child on steroids and early surgery. At the time of tumor surgery, many surgeons will either place an external ventricular drain (a silastic catheter passing through the brain into the lateral ventricle) or place a burr hole, so that if there is an urgent requirement in the postoperative period to drain CSF, access is available. Occasionally, children still present with marked hydrocephalus and are drowsy and require external ventricular drain (EVD) before tumor surgery. The main risk associated with EVD insertion is infection. By ten days, virtually all EVDs will have been colonized. The other risk (albeit very rare) of treating the hydrocephalus before removal of a posterior fossa tumor is that of upward herniation.



A shunt usually consists of a ventricular catheter connected to a valve and reservoir, which allows CSF to be tapped through the skin. On to this is attached a catheter, which is tunneled subcutaneously to a distal site. Although most cavities in the body have been tried, the abdomen, the atrium, and the pleura are used most widely.Atrial shunts were commonplace in the 1960s and 1970s, but problems with endocarditis and glomerulonephritis have resulted in the abdominal cavity becoming the site of preference. Before the advent of shunts in the 1950s, hydrocephalus was frequently fatal.While shunts were therefore considered a huge advance, it quickly became apparent that shunts had problems of their own. In particularly, virtually all studies to date have shown that approximately 40 per cent of shunts will malfunction within the first year of implantation1–3 and that the exponential decline in shunt function then continues at approximately five per cent per year. This malfunction may be the result of one of the following: 

-Mechanical failure: complete, partial, or intermittent obstruction, fracture, migration, or disconnection of the shunt system. Obstruction is the cause of 50 per cent of primary shunt malfunctions, and the vast majority of these are blockages of the ventricular catheter.

-Functional failure: most commonly overdrainage, causing symptoms of low-pressure headaches, slit-ventricle syndrome, and subdural hematomas.

-Infection: the most common infecting organisms are Staphylococcus epidermidis (40 per cent) and Staphylococcus aureus (20 per cent). 

Although some single-institution studies have quoted infection rates of less than one per cent, the generally accepted rate is of the order of ten per cent (with far higher figures being seen in the neonatal period). 2,3 Furthermore, concurrent surgical procedures and hydrocephalus secondary to obstruction of the CSF pathways by tumor have been shown to be independent risk factors for shunt failure. 4 It is not, therefore, surprising that much time and research have been invested in trying to improve shunt function (flow-regulated valves, anti-siphon devices, programmable valves). However, the fundamental problem is how to design a shunt that functions adequately in the horizontal position but that does not then overdrain (due to siphoning) in the vertical position. Despite the advertising propaganda pedaled by shunt manufacturers, there is no evidence to suggest that any one shunt is superior.3.



The posterior fossa is the most common site for both malignant and low-grade tumors in children. Most of these tumors are midline and are best approached suboccipitally via a midlin incision. 

The patient may be placed prone (the most commonly used position), in the “Concorde position,”or in the sitting position. The advantages of the sitting position include the fact that there is no pooling of blood and that it gives a good view to tumors that extend high up towards the tentorium. The disadvantages of this position include the risks of air embolism and the fact that the surgeon is left in an uncomfortable position with their hands outstretched. By monitoring cardiac Doppler and end tidal carbon dioxide, it is possible to detect the consequences of air embolism early and to take steps to seal the source of the embolism. The degree of tonsillar herniation and the inferior extent of the tumor will determine the number of cervical laminae that are exposed. 

As mentioned earlier, many surgeons place an occipital burr hole and insert an external ventricular drain as the initial step in the operation. The actual bone removal can be undertaken piecemeal (craniectomy) or, more commonly, en bloc (craniotomy).With the latter approach, the bone is replaced at the end of the procedure. Most posterior fossa tumors are located within the fourth ventricle, and the approach consists of dividing the vermis to expose the tumor. Although the mechanism is not understood, it is generally accepted that the cerebellar mutism seen occasionally after posterior fossa surgery is related to the splitting of the vermis.


Figure 2. Intraoperative aspect (same personal case as in fig.1)


Although the surgeon will usually claim a “watertight closure” of the dura in the operation notes, CSF leakage and pseudomeningocele formation (collection of CSF under the skin) are relatively common.This is an indication of ongoing abnormal CSF dynamics, which may take many days or even weeks to settle. A mixture of suturing, pressure bandages, and lumbar puncture/lumbar drain insertion are usually sufficient to tide the patient over until a new equilibrium between CSF production and absorption is reached.

Laterally placed tumors, in particular cerebellar pontine angle (CPA) tumors, are best approached directly, with the patient in the park-bench position. The patient is placed on their side with a sandbag under the axilla to avoid brachial plexus injury, and with the dependant arm either hanging in a sling or strapped across the chest. The incision is placed just medial to the mastoid process and continues down into the neck. The main risks of this approach include damage to the vertebral artery, problems with CSF rhinorrhea if entry into the air cells is not dealt with, and damage to the lower cranial nerves that run through the CPA.


Figure 3. MRI aspect (horizontal incidence). Same case as in fig. 1. (postoperative aspect)



CSRT is one of the most complex techniques delivered in the majority of radiotherapy departments. The target volume for CSRT has an irregular shape that includes the whole of the CNS and the meninges. It is generally delivered using a technique in which the lower borders of lateral whole-brain fields are matched to the upper border of the spinal field. 

Technical accuracy of planning and delivery of CSRT is essential for optimal results. Careful attention to coverage of the entire target volume is essential. In the most recent North American and French cooperative group studies, the frequency of major deviations from protocol is still approximately 30 per cent; such deviations have been shown to correlate with outcome. In a series of French medulloblastoma studies, the risk of recurrence has been demonstrated to relate to the accuracy of planning, particularly in the region of the cribriform fossa.39,40 

Traditionally, shielding blocks have been used in the lateral fields to shield not only the nasal and oral structures and teeth but also the lens in order to minimize the risk of cataract. However, the cribriform plate lies between the eyes in young children. It may be preferable to risk the development of a cataract rather than underdose the cribriform fossa, with the increased risk of relapse.41 

Another major issue in treatment planning for medulloblastoma is the choice of the volume and technique for the posterior fossa boost. Currently, it is standard to irradiate the entire posterior fossa. Using posterior oblique fields planned with CT scanning, it is possible to reduce the dose to the inner ear. This may be of benefit for patients also receiving cisplatincontaining adjuvant chemotherapy.42 

From several studies, it has emerged that it is important to avoid unnecessary gaps in treatment due to machine servicing, public holidays, etc. The Société Internationale d’Oncologie Pédiatrique (SIOP) PNET-3 study has shown a significantly worse outcome when the duration of treatment exceeds 50 days as compared with the results for children treated as planned over 45–47 days.4 The “standard” dose of CSRT for medulloblastoma has been 35–36 Gy with a boost to the posterior fossa, giving a total dose to this area of 54–56 Gy.



Since the 1950s, the standard treatment of medulloblastoma has been surgery followed by craniospinal radiotherapy. The introduction of chemotherapy in the treatment of medulloblastoma was justified to try to improve prognosis. The drugs chosen were those usually used for brain tumors in adults, especially the nitrosoureas. 5,7 The justification for using lipophilic drugs was their ability to cross the blood–brain barrier better. Although the blood–brain barrier is often disrupted by meningeal involvement by medulloblastoma, this is still a potential problem in normal brain, where it is necessary to treat occult distant locally invasive or metastatic tumor cells. 

The use of “sandwich chemotherapy” between surgery and radiotherapy has been advocated to allow better drug penetration in the tumor bed as well as better hematological tolerance and lower neurotoxi-city and ototoxicity compared with giving following craniospinal irradiation.44–46 Until recently, the benefit of such timing of chemotherapy was not demonstra-ted in prospective randomized trials.19,47 However, the PNET-3 study has reported a significant advantage in eventfree survival for the use of intensive pre-radiotherapy chemotherapy compared with radiotherapy alone.43 In recent years, the role of chemotherapy has become increasingly important for several reasons, including:

-the tumor response rates observed in phase II studies;

-the therapeutic benefit demonstrated in metastatic medulloblastoma;

-the aim to decrease late effects by decreasing the dose of radiation to the CNS in standardrisk medulloblastoma;

-attempts to postpone or avoid CNS radiotherapy in very young children.



Medulloblastoma is the most frequent malignant brain tumor of childhood, with an annual incidence of 0.5–0.7 cases per 100 000 children under the age of 15 years 

-With modern techniques of neurosurgery and radiotherapy, between 50 and 60 per cent of these children are expected to be alive and free of progression five years following diagnosis. 

-The addition of a three-drug adjuvant chemotherapy regimen has resulted in five-year progression- free survival (PFS) rates of approximately 80 per cent or higher, in particular in children with localized disease. 

-Children who survive medulloblastoma are at high risk of developing serious neurocognitive, endocrine, and neuropsychological deficits; The long-term sequelae are more dramatic in young children who received high doses of wholebrain irradiation.



  1. Kaatsch P, Spix C, Michaelis J. 20 years German Childhood Cancer Registry. Annual Report 1999. Mainz: Institute for Medical Statistics and Documentation of the University, 2000, 
  2. Kleihues P, Cavenee WK. World Health Organization Classification of Tumours of the Nervous System: Pathology and Genetics. Lyon: IARC Press, 2000. 
  3. Parkin DM, Kramarova E, Draper GJ, et al. International Incidence of Childhood Cancer, vol. II. IARC scientific publication no. 144. Lyon: IARC Press, 1998. 
  4. Prados, MD, Warnick RE, Wara WM, Larson DA, Lamborn K, Wilson CB. Medulloblastoma in adults. Int J Radiat Oncol Biol Phys 1995; 32:1145–52. 
  5. Evans AE, Jenkin RDT, Sposto R, et al. The treatment of medulloblastoma: results of a prospective randomized trial of radiation therapy with and without CCNU, vincristine and prednisone. J Neurosurg 1990; 72:572–82. 
  6. Kaatsch P, Rickert CH, Kühl J, Schütz J, Michaelis J. Populationbased epidemiologic data on brain tumors in German children. Cancer 2001; 92:3155–64. 
  7. Tait DM, Thornton-Jones H, Bloom HJG, Lemerle J, Morris- Jones P. Adjuvant chemotherapy for medulloblastoma: the first multi-centre control trial of the International Society of Paediatric Oncology (SIOP I). Eur J Cancer 1990; 26:464–9. 
  8. Jenkin DK, Goddard D, Armstrong L, et al. Posterior fossa medullo-blastoma in childhood: treatment results and a proposal for a new staging system. Int J Radiat Oncol Biol Phys 1990; 19:265–74. 
  9. Krischer JP, Ragab AH, Kun L, et al. Nitrogen mustard, vincristine, procarbazine and prednisone as adjuvant chemotherapy in the treatment of medulloblastoma. J Neurosurg 1991; 74:905–9. 
  10. Packer RJ, Sutton LN, Elterman R, Lange B, Goldwein J, Nicholson HS. Outcome for children with medulloblastoma treated with radiation and cisplatin, CCNU and vincristine chemotherapy. J Neurosurg 1994; 81:690–8. 
  11. Packer RJ, Goldwein J, Nicholson HS, et al. Treatment of children with medulloblastomas with reduced-dose craniospinal radiation therapy and adjuvant chemotherapy: a Children’s Cancer Group study. J Clin Oncol 1999; 17:2127–36.
  12. Hoppe-Hirsch E, Brunet L, Laroussinie F, et al. Intellectual outcome in children with malignant tumors of the posterior fossa: influence of the field of irradiation and quality of surgery. Childs Nerv Syst 1995; 11:340–6. 
  13. Roman DD, Sperduto PW. Neuropsychological effects of cranial radiation: current knowledge and future directions. Int J Radiat Oncol Biol Phys 1995; 31:983–98. 
  14. Marx M, Beck JD, Müller H, Kühl J, Langer T, Dörr HG. Endocrine late-effects of brain tumour therapy in childhood and adolescence: concept of a prospective endocrinological follow-up. Klin Paediatr 2000; 212:224–8. 
  15. Mulhern RK, Kepner JL, Thomas PR, Armstrong FD, Friedman HS, Kun LE. Neuropsychologic functioning of survivors of childhood medulloblastoma randomized to receive conventional or reduced-dose craniospinal irradiation: a pediatric oncology group study. J Clin Oncol 1998; 16:1723–8. 
  16. Radcliffe J, Bunin GR, Sutton LN, Goldwein JW, Phillips PC. Cognitive deficits in long-term survivors of childhood medulloblastoma and other noncortical tumors: age-dependent effects of whole brain radiation. Int J Dev Neurosci 1994; 12:327–34. 
  17. Grill J, Kieffer-Renaux V, Bulteau C, et al. Long-term intellectual outcome in children with posterior fossa tumors according to radiation doses and volumes. Int J Radiat Oncol Biol Phys 1999; 45:137–45. 
  18. Albright AL, Wisoff JH, Zeltzer PM, Boyett JM, Rorke LB, Stanley P. Effects of medulloblastoma resections on outcome in children: a report from the Children’s Cancer Group. Neurosurgery 1996; 38:265–71. 
  19. Zeltzer PM, Boyett JM, Finlay JL, et al. Metastasis stage, adjuvant treatment and residual treatment are prognostic factors for medulloblastomas in children: conclusions from the Children’s Cancer Group 921 randomized phase III study. J Clin Oncol 1999; 17:832–45. 
  20. Thomas PR, Deutsch M, Kepner JL, et al. Low-stage medulloblastoma: final analysis of trial comparing standarddose with reduced-dose neuraxis irradiation. J Clin Oncol 2000; 18:3004–11. 
  21. Gnekow A. Recommendations of the Brain Tumor Subcommittee for the reporting of trials. SIOP Brain Tumor Subcommittee. International Society of Pediatric Oncology Med Pediatr Oncol 1995; 24:104–8.
  22. Chang CH, Housepian EM, Herbert C, Jr. An operative staging system and a megavoltage radiotherapeutic technique for cerebellar medulloblastomas. Radiology 1969; 93:1351–9. 
  23. Fouladi M, Heideman R, Langston JW, Kun LE, Thompson SJ, Gajjar A. Infectious meningitis mimicking recurrent medulloblastoma on magnetic resonance imaging. J Neurosurg 1999; 91:499–502. 
  24. Miralbell R, Bieri S, Huguenin P, et al. Prognostic value of cerebrospinal fluid cytology in pediatric medulloblastoma. Swiss Pediatric Oncology Group. Ann Oncol 1999; 10:239–41. 
  25. Flores LE, Williams DL, Bell AB, O’Brien M, Ragab AH. Delay in the diagnosis of pediatric brain tumors. Am J Dis Child 1986; 140:684–6. 
  26. Schell MJ, McHaney VA, Green AA, et al. Hearing loss in children and young adults receiving cisplatin with or without prior cranial irradiation. J Clin Oncol 1989; 7:754–60. 
  27. Brock PR, Bellman SC, Yeomans EC, Pinkerton CR, Pritchard J. Cisplatin ototoxicity in children: a practical grading system. Med Pediatr Oncol 1991; 19:295–300. 
  28. Glaser AW, Furlong W, Walker DA, et al. Applicability of the Health Utilities Index to a population of childhood survivors of central nervous system tumours in the UK. Eur J Cancer 1999; 35:256–61.
  29. Le Gales C, Costet N, Gentet JC, et al. Cross-cultural adaptation of a health status classification system in children with cancer. First results of the French adaptation of the Health Utilities Index Marks 2 and 3. Int J Cancer Suppl 1999; 12:112–18. 
  30. Spoudeas HA, Hindmarsh PC, Matthews DR, Brook CG. Evolution of growth hormone neurosecretory disturbance after cranial irradiation for childhood brain tumours: a prospective study. J Endocrinol 1996; 150:329–42. 
  31. Lannering B, Marky I, Lundberg A, Olsson E. Long-term sequelae after pediatric brain tumors: their effect on disability and quality of life. Med Pediatr Oncol 1990; 18:304–10.
  32. Adan L, Sainte-Rose C, Souberbielle JC, Zucker JM, Kalifa C, Brauner R. Adult height after growth hormone (GH) treatment for GH deficiency due to cranial irradiation. Med Pediatr Oncol 2000; 34:14–19. 
  33. Rutka JT, Hoffman HJ. Medulloblastoma: a historical perspective and overview. J Neuro-Oncol 1996; 29:1–7. 
  34. Bouffet E, Bernard JL, Frappaz D, et al. M4 protocol for cerebellar medulloblastoma: Supratentorial radiotherapy may not be avoided. Int J Radiat Oncol Biol Phys 1992; 24:79–85. 
  35. Landberg TG, Lindgren ML, Cavallin-Stahl EK, et al. Improvements in the radiotherapy of medulloblastoma. Cancer 1980; 45:670–8. 
  36. Kramer ED, Packer RJ. Chemotherapy of malignant brain tumors in children. Clin Pharmacol 1992; 15:163–85. 
  37. Kühl J, Müller HL, Berthold F, et al. Pre-radiation chemotherapy of children and young adults with malignant brain tumors: results of the German pilot trial HIT ’88/’89. Klin Paediatr 1998; 210:227–33. 
  38. Kühl J. Modern treatment strategies in medulloblastoma. Childs Nerv Syst 1998; 14:2–5. 
  39. Carrie C, Alapetite C, Mere P, et al. Quality control of radiotherapeutic treatment of medulloblastoma in a multicentric study: the contribution of radiotherapy technique to tumour relapse. The French Medulloblastoma Group. Radiother Oncol 1992; 24:77–81. 
  40. Carrie C, Hoffstetter S, Gomez F, et al. Impact of targeting deviations on outcome in medulloblastoma: study of the French Society of Pediatric Oncology (SFOP). Int J Radiat Oncol Biol Phys 1999; 45:435–9. 
  41. Taylor RE. UKCCSG radiotherapy and brain tumour groups. Medulloblastoma/PNET and craniospinal radiotherapy (CSRT). Report of a workshop held in Leeds 30 June 99. Clin Oncol 2001; 13:58–64. 
  42. Fukunaga-Johnson N, Sandler HM, Marsh R, Martel MK. The use of 3D conformal radiotherapy (3D CRT) to spare the cochlea in patients with medulloblastoma. Int J Radiat Oncol Biol Phys 1998; 41:77–82. 
  43. Taylor RE, Bailey CC, Robinson K, et al. Results of a randomized study of pre-radiotherapy chemotherapy vs radiotherapy alone for non-metastatic (M0-1) medulloblastoma. The SIOP/ UKCCSG PNET-3 study. J Clin Oncol 2003; 21:1581–91. 
  44. Pendergrass TW, Milstein JM, Geyer JR, et al. Eight drugs in one day chemotherapy for brain tumors: experience in 107 children and rationale for preirradiation chemotherapy. J Clin Oncol 1987; 5:1221–31. 
  45. Gentet JC, Bouffet E, Doz F, et al. Preirradiation chemotherapy including “eight drugs in 1 day” regimen and high dose methotrexate in childhood medulloblastoma: results of the M7 French cooperative study. J Neurosurg 1995; 82:608–14. 
  46. Kovnar EH, Kellie SJ, Horowitz ME, et al. Preirradiation cisplatin and etoposide in the treatment of high-risk medulloblastoma and other malignant embryonal tumors of the central nervous system: a phase II study. J Clin Oncol 1990; 8:330–6. 328 Embryonic tumors.