-
[show abstract]
[hide abstract]
ABSTRACT: A simple and fair benchmarking system or financial indicators for use on the clinical department level have been lacking to evaluate the management efficiency and activity of each clinical department or division of a hospital. New financial indicators have therefore been developed based on personnel costs. Indicator 1: The ratio of marginal profit after personnel cost per personnel cost (RMP). Indicator 2: The ratio of investment (=indirect cost) per personnel cost (RIP). The difference between RMP and RIP demonstrates the operation profit in US Dollars for
personnel cost (OPP). A turning point in profitability similar to the break-even point (BEP) and break-even ratio (BER) could be also defined by the combination of the RMP and RIP. The merits of these two indicators are not only the ability to indicate the relationship between the
medical profit and the investments in the hospital, but also the capability to demonstrate such indicators as BEP, BER and OPP on a single graph. The two indicators were applied to the hospitals in the National Hospital Organization and to the clinical department in one hospital. Using these two indicators, it was possible to evaluate the management efficiency and medical activity not only in the whole hospital but also in each department and DPC/DRG group.
This will be of use to a manager of a hospital in checking the management efficiency of his/her hospital despite the variations among hospitals, departments and divisions.
-
[show abstract]
[hide abstract]
ABSTRACT: The authors report a case of choroid plexus papilloma in a girl with hypomelanosis of Ito, and they review the literature in brief. Hypomelanosis of Ito is a rare neurocutaneous syndrome characterized by cutaneous hypopigmented whorls, streaks, and patches along lines of Blaschko. Most patients exhibit CNS manifestations, including psychomotor retardation, seizures, hypotonia, and ataxia. A 6-year-old girl with hypomelanosis of Ito was referred to the authors' hospital with bilateral tumors in the lateral ventricles. The right lateral ventricle tumor was surgically removed. Immunohistochemical investigations revealed the tumor to be a choroid plexus papilloma (WHO Grade I). A chromosomal investigation revealed that the tumor tissue demonstrated a large loss of heterozygosity at chromosome 10. The case reported here serves as a reminder that de novo brain tumors may arise in patients with chromosomal mosaicism.
Journal of Neurosurgery Pediatrics 07/2012; 10(3):182-5. · 1.53 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: A simple and fair benchmarking system or financial indicators for use on the clinical department level have been lacking to evaluate the management efficiency and activity of each clinical department or division of a hospital. New financial indicators have therefore been developed based on personnel costs. Indicator 1: The ratio of marginal profit after personnel cost per personnel cost (RMP). Indicator 2: The ratio of investment (=indirect cost) per personnel cost (RIP). The difference between RMP and RIP demonstrates the operation profit in US Dollars for personnel cost (OPP). A turning point in profitability similar to the break-even point (BEP) and break-even ratio (BER) could be also defined by the combination of the RMP and RIP. The merits of these two indicators are not only the ability to indicate the relationship between the medical profit and the investments in the hospital, but also the capability to demonstrate such indicators as BEP, BER and OPP on a single graph. The two indicators were applied to the hospitals in the National Hospital Organization and to the clinical department in one hospital. Using these two indicators, it was possible to evaluate the management efficiency and medical activity not only in the whole hospital but also in each department and DPC/DRG group. This will be of use to a manager of a hospital in checking the management efficiency of his/her hospital despite the variations among hospitals, departments and divisions.
Journal of Medical Systems 08/2011; 35(4):625-37. · 1.13 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: A hospital director must estimate the revenues and expenses not only in a hospital but also in each clinical division to determine the proper management strategy. A new prospective payment system based on the Diagnosis Procedure Combination (DPC/PPS) introduced in 2003 has made the attribution of revenues and expenses for each clinical department very complicated because of the intricate involvement between the overall or blanket component and a fee-for service (FFS). Few reports have so far presented a programmatic method for the calculation of medical costs and financial balance. A simple method has been devised, based on personnel cost, for calculating medical costs and financial balance. Using this method, one individual was able to complete the calculations for a hospital which contains 535 beds and 16 clinics, without using the central hospital computer system.
Journal of Medical Systems 04/2011; 35(2):251-64. · 1.13 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: The authors report the case of an infant with a giant glioependymal cyst. Although it has been suggested that these cysts originate from the tela choroidea, their origin remains controversial. This 35-month-old girl with truncal ataxia was referred to the authors' hospital. Magnetic resonance imaging revealed a giant cystic mass extending from the anterior to the posterior cranial fossa. Hydrocephalus was caused by obstruction of the sylvian aqueduct. Endoscopic fenestration of the cyst wall was performed. Histochemical and immunohistochemical staining identified the lesion as a glioependymal cyst. Magnetic resonance imaging performed 8 months later suggested that the cyst originated from the tela choroidea. At 5-year follow-up, there was no tumor recurrence and she had fully recovered. The origin of glioependymal cysts is discussed, and the authors suggest that their origin is the tela choroidea.
Journal of Neurosurgery Pediatrics 02/2011; 7(2):175-8. · 1.53 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: A neonate with hydrocephalus associated with Dandy-Walker malformation was successfully treated with an endoscopic placement of a transaqueductal ventricular single catheter. The modified catheter was provided with additional fenestration on its proximal side to allow simultaneous drainage from both the supra- and infratentorial compartments. This technique is well known for isolated fourth ventricles, but has not been applied to hydrocephalus associated with Dandy-Walker malformation. The cyst-ventriculoperitoneal shunt effectively drained both compartments. The patient was doing well 18 months after the surgical procedure. Endoscopic transaqueductal shunt placement can be considered, especially in patients with aqueductal patency.
Neurologia medico-chirurgica 01/2011; 51(3):256-9. · 0.61 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: Women with ventriculoperitoneal (VP) shunt require special care during pregnancy, although they develop few complications related to hydrocephalus.
We recently encountered a pregnant patient with hydrocephalus, which was caused by VP shunt malfunction induced by increased intra-abdominal pressure associated with pregnancy. She was treated by replacement of the VP shunt with a ventriculoatrial (VA) shunt. After treatment, she achieved vaginal delivery.
We recommend VA shunt implantation as a treatment option for the management of VP shunt malfunction during pregnancy. We further propose that patients in whom the VP shunt is replaced with a VA shunt can achieve spontaneous vaginal delivery.
Archives of Gynecology 02/2010; 282(3):339-42. · 0.91 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: It is very difficult to treat the patients with malignant brain tumor in children, especially under 3 years, because the conventional irradiation cannot be applied due to the damage of normal brain tissue. However, boron neutron capture therapy (BNCT) has tumor selectivity such that it can make damage only in tumor cells. We evaluated the clinical results and courses in patients with malignant glioma under 15 years. Among 183 patients with brain tumors treated by our group using BSH-based intra-operative BNCT, 23 patients were under 15 years. They included 4 patients under 3 years. There were 3 glioblastomas (GBM), 6 anaplastic astrocytomas(AAS), 7 primitive neuroectodermal tumors (PNET), 6 pontine gliomas and 1 anaplastic ependymoma. All GBM and PNET patients died due to CSF and/or CNS dissemination without local tumor regrowth. All pontine glioma patients died due to regrowth of the tumor. Four of 6 anaplastic astrocytoma and 1 anaplastic ependymoma patients alive without tumor recurrence. BNCT can be applied to malignant brain tumors in children, especially under 3 years instead of conventional radiation. Although it can achieve the local control in the primary site, it cannot prevent CSF dissemination in patients with glioblastoma.
Applied radiation and isotopes: including data, instrumentation and methods for use in agriculture, industry and medicine 04/2009; 67(7-8 Suppl):S27-30. · 1.09 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: This study demonstrates the characterization of proton spot scanning on a Li target assembly for accelerator-based BNCT from the viewpoint of heat removal and material strength. These characteristics are investigated as to their dependence on the Li target thickness, considering that the Cu backing plate has more suitable heat removal properties than Li. Two situations are considered in this paper, i.e. the cyclic operation of the spot scanning, and a stalled spot scanning cycle where the proton beam stays focused on a single position on the Li target. It was found that the maximum of the Li temperature and the strain of the Cu backing increase as the cycle period increases. A cycle period less than 120 ms (over 8.3 Hz of frequency) enables the Li temperature to be kept below 150 degrees C and a cycle of less than 115 ms (8.7 Hz) keeps the Cu strain below the critical value for a 230 microm thick Li target, though the values are evaluated conservatively. Against expectation, the Li temperature and Cu strain are larger for a 100 microm thick target than for a 230 microm target. The required cycle period in this case is 23 ms (43 Hz) for maintaining a reasonable Li temperature and 9 ms (110 Hz) to prevent Cu fatigue fracture. For a stall in the spot scanning cycle, the Cu temperature increases as the beam shutdown time increases. The time for Cu to reach its melting point is estimated to be 4.2 ms at the surface, 20 ms at 1mm depth, for both of 100 and 230 microm thick targets. At least 34 ms is estimated to be enough to make a hole on Cu backing plate. A beam shutdown mechanism with a response time of about 20 ms is therefore required.
Applied radiation and isotopes: including data, instrumentation and methods for use in agriculture, industry and medicine 11/2008; 67(2):259-65. · 1.09 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: The usable range of thickness for the solid lithium target in the accelerator-based neutron production for BNCT via the near-threshold (7)Li(p,n)(7)Be reaction was investigated. While the feasibility of using a (7)Li-target with thickness equal to that which is required to slow down a mono-energetic 1.900 MeV incident proton to the 1.881 MeV threshold of the (7)Li(p,n)(7)Be reaction (i.e., t(min) = 2.33 microm) has already been demonstrated, dosimetric properties of neutron fields from targets greater than t(min) were assessed as thicker targets would last longer and offer more stable neutron production. Additionally, the characteristics of neutron fields generated by (7)Li(p,n)(7)Be for Gaussian incident protons with mean energy of 1.900 MeV were evaluated at a (7)Li-target thickness t(min). The main evaluation index applied in this study was the treatable protocol depth (TPD) which corresponds to the depth in an irradiated medium that satisfies the requirements of the adapted dose protocol. A maximum TPD (TPD(max)) was obtained for each irradiation condition from the relationship between the TPD and the thickness of boron dose enhancer (BDE) used. For a mono-energetic 1.900 MeV proton beam, the deepest TPD(max) of 3.88 cm was attained at the (7)Li-target thickness of t(min) and a polyethylene BDE of 1.10 cm. When the intended TPD for a BNCT clinical treatment is shallower than the deepest TPD(max), the usable (7)Li-target thickness would be between t(min) and an upper limit t(upper) whose value depends on the BDE thickness used. In terms of the effect of stability of the incident proton energy, Gaussian incident proton energies stable to within +/-10 keV of 1.900 MeV were found to be feasible for the neutron production via the near-threshold (7)Li(p,n)(7)Be reaction for BNCT provided that a suitable BDE is used.
Physics in Medicine and Biology 03/2007; 52(3):645-58. · 2.83 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: Clinical and electroencephalographic features of epilepsy and the prognosis of intelligence were investigated in 156 children with hydrocephalus. Of these 53 (34.0%) had epilepsy. The incidence and outcome of epilepsy were determined by the etiology of hydrocephalus. The incidence was high in children with hydrocephalus caused by intra-cranial infection and dysgenetic hydrocephalus without dysraphism. Furthermore, it was difficult to control the epileptic seizures in these patients. In contrast, children with dysgenetic hydrocephalus caused by dysraphism and simple hydrocephalus had lower incidence of epilepsy and epileptic seizures had been well controlled in these cases. In hydrocephalic children, localization-related epilepsy was the most common. Complex partial seizures and focal motor seizures were frequently observed. Common electroencephalographic findings were focal spikes or multi-focal spikes. The intelligence quotient was significantly lower in children with epilepsy than in those without epilepsy. All these findings suggest that the incidence of epilepsy may be a crucial prognostic factor in children with hydrocephalus.
No to hattatsu. Brain and development 10/2006; 38(5):353-8.
-
[show abstract]
[hide abstract]
ABSTRACT: An evaluation of mono-energetic proton energies ranging from 1.885 MeV to 1.920 MeV was carried out to determine the viability of these near threshold energies in producing neutrons for BNCT via the (7)Li(p,n)(7)Be reaction. Neutron fields generated at these proton energies were assessed using the treatable protocol depth (TPD) and the maximum TPD (TPD(max)) as evaluation indices. The heavy charged particle (HCP) dose rate to tumour was likewise applied as a figure of merit in order to account for irradiation time and required proton current. Incident proton energies closer to the reaction threshold generated deeper TPDs compared to higher energy protons when no boron dose enhancers (BDE) were placed in the irradiation field. Introducing a BDE resulted in improved TPDs for high proton energies but their achievable TPD(max) were comparatively lower than that obtained for lower proton energies. In terms of the HCP dose rate to tumour, higher proton energies generated neutron fields that yielded higher dose rates both at TPD(max) and at fixed depths of comparison. This infers that higher currents are required to deliver the prescribed treatment dose to tumours for proton beams with energies closer to the (7)Li(p,n)(7)Be reaction threshold and more achievable proton currents of around 10 mA or less for proton energies from 1.900 MeV and above. The dependence on incident proton energy of the TPD, TPD(max) and the HCP dose rate to tumour with respect to the (10)B concentration in tumour and healthy tissues were also clarified in this study. Increasing the (10)B concentration in tumour while maintaining a constant T/N ratio resulted in deeper TPD(max) where a greater change in TPD(max) was obtained for proton energies closer to the (7)Li(p,n)(7)Be reaction threshold. The HCP dose rates to tumour for all proton energies also went up, with the higher proton energies benefiting more from the increased (10)B concentration.
Physics in Medicine and Biology 09/2006; 51(16):4095-109. · 2.83 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: To clarify the correlation between the radiation dose and clinical outcome of sodium borocaptate-based intraoperative boron neutron capture therapy in patients with malignant glioma.
The first protocol (P1998, n = 8) prescribed a maximal gross tumor volume (GTV) dose of 15 Gy. In 2001, a dose-escalated protocol was introduced (P2001, n = 11), which prescribed a maximal vascular volume dose of 15 Gy or, alternatively, a clinical target volume (CTV) dose of 18 Gy.
The GTV and CTV doses in P2001 were 1.1-1.3 times greater than those in P1998. The maximal vascular volume dose of those with acute radiation injury was 15.8 Gy. The mean GTV and CTV dose in long-term survivors with glioblastoma was 26.4 and 16.5 Gy, respectively. A statistically significant correlation between the GTV dose and median survival time was found. In the 11 glioblastoma patients in P2001, the median survival time was 19.5 months and 1- and 2-year survival rate was 60.6% and 37.9%, respectively.
Dose escalation contributed to the improvement in clinical outcome. To avoid radiation injury, the maximal vascular volume dose should be <12 Gy. For long-term survival in patients with glioblastoma after boron neutron capture therapy, the optimal mean dose of the GTV and CTV was 26 and 16 Gy, respectively.
International Journal of Radiation OncologyBiologyPhysics 08/2006; 65(5):1446-55. · 4.11 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: The characteristics of moderator assembly dimension are investigated for the usage of 7Li(p,n) neutrons by 2.5 MeV protons in boron newtron capture therapy (BNCT) of brain tumors in the present study. The indexes checked are treatable protocol depth (TPD), which is the greatest depth of the region satisfying the dose requirements in BNCT protocol, proton current necessary to complete BNCT by 1 h irradiation, and the heat flux deposited in the Li target which should be removed. Assumed materials are D2O for moderator, and mixture of polyethylene and LiF with 50 wt % for collimator. Dose distributions have been computed with MCNP 4B and 4C codes. Consequently, realized TPD does not show a monotonical tendency for the Li target diameter. However, the necessary proton current and heat flux in the Li target decreases as the Li target diameter increases, while this trend reverses at around 10 cm of the Li target diameter for the necessary proton current in the condition of this study. As to the moderator diameter, TPD does not exhibit an apparent dependence. On the other hand, necessary proton current and heat flux decrease as the moderator diameter increases, and this tendency saturates at around 60 cm of the moderator diameter in this study. As to the collimator, increase in inner diameter is suitable from the viewpoint of increasing TPD and decreasing necessary proton current and heat flux, while these indexes do not show apparent difference for collimator inner diameters over 14 cm for the parameters treated here. The practical viewpoint in selecting the parameters of moderator assembly dimension is to increase TPD, within the technically possible condition of accelerated proton current and heat removal from the Li target. In this process, the values for which the resultant characteristics mentioned above saturate or reverse would be important factors.
Medical Physics 07/2006; 33(6):1688-94. · 2.83 Impact Factor
-
Nippon rinsho. Japanese journal of clinical medicine 10/2005; 63 Suppl 9:442-6.
-
[show abstract]
[hide abstract]
ABSTRACT: The dependence of boron-dose enhancer (BDE) characteristics on dose protocol and 10B concentration was evaluated for BNCT using near-threshold 7Li(p,n)7Be direct neutrons. The treatable protocol depth (TPD) was utilized as an evaluation index. MCNP calculations were performed for near-threshold 7Li(p,n)7Be at a proton energy of 1.900 MeV and for a polyethylene BDE. The effect of dose protocol on BDE characteristics was reflected in terms of the optimum BDE thickness needed for maximum TPD which was found to be independent of the treatable dose but was observed to vary for different combinations of the tolerance doses for heavy charged particles and gamma rays. For the 10B concentration dependence, the TPD was increased by increasing the T/N ratio, i.e., the ratio of the 10B concentration in the tumour (10B(Tumour)) to that in the normal tissue (10B(Normal)), and by increasing 10B(Tumour) and 10B(Normal) at constant T/N ratio. It was found that the use of BDE becomes unnecessary from the viewpoint of increasing the TPD, when 10B(Tumour) is over a certain level which is decided by the conditions of the dose protocol.
Physics in Medicine and Biology 02/2005; 50(1):167-77. · 2.83 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: The dose contribution of (10)B(n,alpha)(7)Li reaction in BNCT using near threshold (7)Li(p,n)(7)Be direct neutrons can be increased through the use of materials referred to as boron-dose enhancers (BDE). In this paper, possible BDE optimization criteria were determined from the characteristics of candidate BDE materials namely (C(2)H(4))(n), (C(2)H(3)F)(n), (C(2)H(2)F(2))(n), (C(2)HF(3))(n), (C(2)D(4))(n), (C(2)F(4))(n), beryllium metal, graphite, D(2)O and (7)LiF. The treatable protocol depth (TPD) was used as the assessment index for evaluating the effect of these materials on the dose distribution in a medium undergoing BNCT using near threshold (7)Li(p,n)(7)Be direct neutrons. The maximum TPD (TPD(max)) did not exhibit an explicit dependence on material type as evidenced by its small range and arbitrary variations. The dependence of TPD on BDE thickness was influenced by the BDE material used as indicated by the sharply peaked TPD versus BDE thickness curves for materials with hydrogen compared to the broader curves obtained for those without hydrogen. The BDE thickness required to achieve TPD(max) (BDE(TPD(max))) were also found to be thinner for materials with hydrogen. The TPD(max), the dependence of TPD on BDE thickness, and the BDE(TPD(max)) were established as appropriate BDE optimization parameters. Based on these criteria and other practical considerations, the suitable choice as BDE among the candidate materials considered in this study for treatments involving tumors located at shallow depths would be (C(2)H(4))(n) while beryllium metal was judged as more appropriate for treatment of deep-seated tumors.
Applied Radiation and Isotopes 12/2004; 61(5):1003-8. · 1.17 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: Since 1998, we have introduced a mixed epithermal- and thermal neutron beam for boron neutron capture therapy (BNCT) to improve the neutron beam distribution. Sixteen patients with malignant glioma (glioblastoma, n = 14; anaplastic ependymoma, n = 1; PNET, n = 1) were treated by BNCT in Japan. Of these, 9 died; 3 due to cerebrospinal fluid (CSF) dissemination, 1 each of tumor invasion, meningitis, pneumonia, and unknown causes, and 2 patients died of local recurrence or radiation necrosis. The current postmortem study is comprised of 3 patients with glioblastoma who were treated with BNCT employing an epithermal neutron beam and sodium borocaptate (BSH: Na2B12H11SH). None of the patients manifested local regrowth at the primary site. However, in 2 patients there was CSF dissemination; tumor cells were recognized throughout the subarachnoid space. In the other patient, tumor cells had massively invaded the ipsilateral- and contralateral hemisphere and brain stem from the bottom of the tumor cavity via the corpus callosum and cerebral peduncle. Our findings indicate that BNCT can achieve local control of glioblastoma at the primary site. However, to further improve the clinical outcome after BNCT, steps must be taken to prevent CSF dissemination.
Journal of Neuro-Oncology 06/2004; 68(1):25-32. · 3.21 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: The characteristics of a number of candidate boron-dose enhancer (BDE) materials for boron neutron capture therapy (BNCT) using near threshold 7Li(p,n)7Be direct neutrons were evaluated based on the treatable protocol depth (TPD), defined in this paper. Simulation calculations were carried out by means of MCNP-4B transport code for candidate BDE materials, namely, (C2H4)n, (C2H3F)n, (C2H2F2)n, (C2HF3)n, (C2D4)n, (C2F4)n, beryllium metal, graphite, D2O and 7LiF. Dose protocols applied were those used for intra-operative BNCT treatment for brain tumour currently used in Japan. The maximum TPD (TPDmax) for each BDE material was found to be between 4 cm and 5 cm in the order of (C2H4)n < (C2H3F)n < (C2H2F2)n < (C2HF3)n < beryllium metal < (C2D4)n < graphite < (C2F4)n < D2O < 7LiF. Based on the small and arbitrary variations in the TPDmax for these materials, an explicit advantage of a candidate BDE material could not be established from the TPDmax alone. The dependence of TPD on BDE thickness was found to be influenced by the type of BDE material. For materials with hydrogen, sharp variations in TPD were observed, while those without hydrogen exhibited more moderate fluctuations in TPD as the BDE thickness was varied. The BDE thickness corresponding to TPDmax (BDE(TPDmax)) was also found to depend on the type of BDE material used. Thicker BDE(TPDmax), obtained mostly for BDE materials without hydrogen, significantly reduced the dose rates within the phantom. The TPDmax, the dependence of TPD on BDE thickness and the BDE (TPDmax) were ascertained as appropriate optimization criteria in choosing suitable BDE materials for BNCT. Among the candidate BDE materials considered in this study. (C2H4)n was judged as the suitable material for near-surface tumours and beryllium metal for deeper tumours based on these optimization criteria and other practical considerations.
Physics in Medicine and Biology 03/2004; 49(5):819-31. · 2.83 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: Mercaptoundecahydrododecaborate (Na2B12H111SH, sodium borocaptate or 'BSH') has been used clinically as a boron compound for boron neutron capture therapy (BNCT) in patients with malignant glioma in Japan and Europe. Boron-10 is known to accumulate selectively only in brain tumor cells. This work was aimed to clarify the subcellular biodistribution of BSH in a rat glioma model using immunohistochemical approach. Wistar rats were used for this experiment. An intracerebral injection of 5.0 x 10(6) C6 glioma cells was introduced into the region of cerebral hemisphere. Fifty milligrams of "'B/kg BSH was infused intravenously two weeks after implantation. Host rats were divided into six groups according to the sampling time: 1, 4, 8, 16, 24 and 48 h after the start of BSH infusion. Immunohistochemical study was carried out using anti-BSH antibody. Boron was already found in a whole cell 1 h after BSH infusion, and then seemed to collect in a cell nuclei around 8-16 h after infusion. It was still recognized in tumor cell 48 h after infusion. This study supports the following hypothesis on selective boron uptake in a tumor. BSH can pass through the disrupted blood-brain barrier (BBB) easily and can come in contact with tumor cells; there, BSH can bind on the extracellular surface of plasma membrane to choline residues. After binding to the plasma membrane, boron with choline residues may be internalized into the cell by endocytic pathways and eventually travel to cell nuclei, and then stay there for a long time.
Journal of Neuro-Oncology 10/2002; 59(2):135-42. · 3.21 Impact Factor
