Case report and review of the literature
Stephen J. Hentschel, Kimberly G. Yen and Frederick F. Lang
✓The authors describe a rare case of a Chiari I malformation presenting with acute acquired comitant esotropia (AACE) in a 5-year-old boy. A posterior fossa decompression with duraplasty and a C1–2 laminectomy were performed. There was an immediate postoperative improvement in the esotropia, which completely resolved by 7 months following surgery. The pertinent literature is discussed and reasons are presented for recommending posterior fossa decompression in certain patients, rather than strabismus surgery, as the initial treatment for esotropia. The authors suggest that in patients with AACE, even subtle symptoms and signs of Chiari I malformation should prompt imaging of the posterior fossa. Strong consideration should be given to performing posterior fossa decompression in patients with Chiari I malformation and AACE before strabismus surgery because the esotropia may completely resolve with decompression.
Vikram V. Nayar, Ronald J. Benveniste and Frederick F. Lang
The infratentorial supracerebellar approach to the pineal region presents special challenges during patient positioning. The head must be flexed and the body positioned to allow an operative trajectory under the straight sinus. Image guidance is not useful during positioning because registration and navigation take place after the head is fixed in its final position. Therefore, a reliable method of positioning based on external, easily identifiable landmarks to estimate the surgical trajectory along the straight sinus toward the pineal region is needed. Based on observation, the authors hypothesized that a line between 2 palpable external landmarks, the inion and the bregma, often approximates the surgical trajectory along the straight sinus. They tested this hypothesis by quantifying the relationship between the straight sinus and the bregma, and describe a method for estimating the working angle during patient positioning.
The midsagittal, Gd-enhanced, T1-weighted MR images of 102 patients were analyzed. Demographic data and the presence or absence of tentorial pathological entities was recorded. The slant of the straight sinus was classified as common, high, or low, based on a previously described classification system. A line along the bottom of the straight sinus (that is, the straight-sinus line) was extended superiorly to its intersection with the calvaria, and the distance from this intersection point to the bregma was measured.
The intersection point of the straight-sinus line and the calvaria was on average 2 ± 8.2 mm (these values are expressed as the mean ± SD throughout) anterior to the bregma (range 19.9 mm anterior to 19.1 mm posterior). The distance from the intersection point to the bregma was not statistically significantly different in younger or older patients, or in patients with or without tumors involving the pineal region. In patients with a low slant of the straight sinus, the intersection point was 5.3 ± 6.3 mm anterior to the bregma, whereas in patients with a high slant of the straight sinus, the intersection point was 0.21 ± 9.1 mm posterior to the bregma (p = 0.015).
The straight-sinus line, which defines the working angle for the supracerebellar infratentorial approach, intersects the calvaria very close to the bregma in the majority of patients. Therefore, ideal patient positioning can be achieved by flexing the patient's head to optimize the working angle defined by an imaginary line connecting the torcula (inion) to the bregma.
Part 1: Pathology
Douglas C. Miller, Frederick F. Lang and Fred J. Epstein
Histopathological features that suggest the diagnosis of ganglioglioma require, in most cases, confirmation by special stains to distinguish these tumors from other gliomas. For this purpose, immunostaining for synaptophysin, which has previously been shown to selectively label the cell surface of neoplastic ganglion cells, was used to retrospectively examine glioma tumor specimens. Sixty-three cases of ganglioglioma were identified. The files of the Division of Neuropathology of New York University Medical Center contained 45 tumors that had been diagnosed as ganglioglioma, of which 42 were verified by synaptophysin; three cases were reclassified, two as astrocytomas and one as a gangliocytic paraganglioma. Thus, a tumor identified as ganglioglioma based on other criteria was likely to be a ganglioglioma. The other 21 cases of gangliogliomas were originally diagnosed as astrocytoma or mixed glioma, but were shown by synaptophysin staining to be gangliogliomas. In some cases the ultimate diagnosis was obtained after radical surgery provided relatively abundant amounts of tissue, thereby limiting sampling errors, in contrast to the biopsies from which the original diagnoses were made.
Histopathological review of these cases demonstrated that four features represent important clues to the correct diagnosis: 1) clusters of large cells potentially representing neurons (without such cells the tumor cannot be classified as a ganglioglioma); 2) no perineuronal clustering of the glial cells around the alleged neoplastic neurons; 3) fibrosis (desmoplasia); and 4) calcification. Binucleate neurons, previously suggested to be common in gangliogliomas, were not frequently found in this series, and lymphocytic infiltrates, while common, are so often found in other tumors that they gave no specific hint that any single neoplasm was a ganglioglioma. The glial elements were astrocytic in all cases, except that one tumor also had oligodendroglial and ependymal patterns. Four tumors also had small mature neurons, as seen in neurocytomas. Cells from one tumor were successfully grown in short-term tissue culture; the culture contained large dividing neurons with synaptophysin immunoreactivity as well as smaller dividing cells, demonstrating that the neuronal cells are a proliferating element in gangliogliomas.
Fadi Hanbali, Peyman Tabrizi, Frederick F. Lang and Franco DeMonte
Object. Published data obtained in children with tumors of the skull base are sparse. In the majority of the available reports, the authors focus on the technical application of skull base approaches, but they contribute a paucity of information on the management of specific tumors, especially malignant skull base lesions. The purposes of this report are to increase the collective experience with the treatment of these tumors and to identify successful management paradigms.
Methods. The authors retrospectively reviewed the clinical records, pathological reports, and diagnostic images obtained in 24 children (≤ 19 years of age) with tumors arising from the cranial base in whom resection was part of their management between 1992 and 2002. Surgery-related complications and outcomes were analyzed with regard to tumor type and surgical approach.
The median age of the group was 14 years. Tumors involved the anterior skull base in eight (33%), the middle skull base in 10 (42%), both the anterior and middle skull base in four (17%), and the posterior skull base in two patients (8%). Benign lesions were discovered in 11 patients (46%) and malignant neoplasms in 13 (54%). The tumors were most commonly of mesenchymal origin (21 [87.5%] of 24 tumors). Thirty surgical procedures were performed using a number of skull base approaches. A gross-total resection was achieved in 23 procedures (77%) and a subtotal resection in five (17%); a biopsy procedure was performed in one case; and the disease process could not be accessed in one case. One patient died in the perioperative period. Minor complications (Karnofsky Performance Scale score ≥ 90, no prolongation of hospital stay, and no further surgery needed) occurred following 10 (33%) of the 30 surgical procedures. These affected 10 (42%) of the 24 patients and resulted in persistent or prolonged deficits in only five patients (21%). In nine patients (38%) the tumor recurred after a mean duration of 23 months.
Conclusions. Skull base tumors in children affect mainly the anterior and middle cranial fossa. Sarcomas account for the majority of malignant tumors. Treatment of skull base tumors in children and adolescents needs to be tailored to patient age, tumor location, and tumor type.
Frederick F. Lang, O. Kenneth Macdonald, Gregory N. Fuller and Franco DeMonte
Object. Primary meningiomas arising outside the intracranial compartment (primary extradural meningiomas [PEMs]) are rare tumors. To develop a better understanding of these tumors and to establish a comprehensive classification scheme for them, the authors analyzed a series of patients treated at the M. D. Anderson Cancer Center (MDACC) and reviewed all cases reported in the English-language literature since the inception of the use of computerized tomography (CT) scanning.
Methods. Clinical records, results of radiographic studies, and histological slides were reviewed for all cases of PEM at MDACC. Demographic features, symptoms, tumor location, histological grade, and patient outcome were assessed in all cases. A comprehensive literature search identified 168 PEMs in 142 patients reported during the CT era. These reports were also analyzed for common features. Tumors for both data sets were classified as purely extracalvarial (Type I), purely calvarial (Type II), and calvarial with extracalvarial extension (Type III). Type II and Type III tumors were further categorized as convexity (C) or skull base (B) lesions.
The incidence of PEMs at MDACC was 1.6%, which was consistent with the rate reported in the literature. In both data sets, the male/female ratio was nearly 1:1. The most common presenting symptom was a gradually expanding mass. The age of patients at diagnosis of PEM was bimodal, peaking during the second decade and during the fifth to seventh decades. In all MDACC cases and in 90% of those reported in the literature the PEMs were located in the head and neck. The majority of tumors originated in the skull (70%).
In the MDACC series and in the literature review, the majority (67% and 89%, respectively) of tumors were histologically benign. Although fewer PEMs were malignant or atypical (33% at MDACC and 11% in the literature), their incidence was higher than that observed for primary intracranial meningiomas. Distant metastasis was not a common feature reported for patients with PEMs (6% in the literature).
Outcome data were available in 96 of the cases culled from the CT-era literature. The combination of the MDACC data and the data obtained from the literature demonstrated that patients with benign Type IIB or Type IIIB lesions were more likely to experience recurrence than patients with benign Type IIC or Type IIIC tumors (26% compared with 0%, p < 0.05). The more aggressive atypical and malignant tumors were associated with a statistically significant higher death rate (29%) relative to benign tumors (4.8% death rate, p < 0.004).
Conclusions. Defining a tumor as a PEM is dependent on the tumor's relation to the dura mater and the extent and direction of its growth. Classification of PEMs as calvarial or extracalvarial and as convexity or skull base lesions correlates well with clinical outcome.
Frederick F. Lang, Douglas C. Miller, Maxim Koslow and Elizabeth W. Newcomb
✓ To characterize some of the genetic events underlying the development of glioblastoma multiforme, the authors analyzed 65 astrocytic tumors (seven pilocytic astrocytomas, eight astrocytomas, 16 anaplastic astrocytomas, and 34 glioblastomas multiforme) for loss of heterozygosity for chromosome 17p, loss of heterozygosity for chromosomes 10p and 10q, amplification of the epidermal growth factor receptor (EGFR) gene, and amplification of the oncogenes N-myc, c-myc, and N-ras using Southern blot analysis. Alterations of the p53 gene (positive immunostaining for p53 protein in tumors with or without p53 gene mutations) in these 65 tumors were analyzed previously. None of the 65 tumors showed amplification or rearrangement of N-myc, c-myc, or N-ras oncogenes.
The molecular analysis presented here demonstrates distinct variants of astrocytic tumors, with at least three genetic pathways leading to glioblastoma multiforme. One pathway was characterized by 43 astrocytomas with alterations in p53. Glioblastomas with p53 alterations may represent tumors that progress from lower-grade astrocytomas. This variant was more likely to show loss of chromosome 17p than tumors without p53 alterations (p < 0.04). Seventy-five percent of tumors with loss of one 17p allele demonstrated mutations in the p53 gene. Loss of chromosome 10 was associated with progression from anaplastic astrocytoma (13%) to glioblastoma (38%) (p < 0.04). Amplification of the EGFR gene was a rare (7%) but late event in tumor progression (p < 0.03). A second pathway was characterized by six astrocytomas without p53 alterations and may represent clinically de novo high-grade tumors. These tumors were more likely to show amplification of the EGFR gene (83%) than tumors with p53 alterations. Sixty percent of tumors with EGFR amplification also showed loss of chromosome 10; loss of chromosome 17p was infrequent in this variant. One or more alternative pathways were characterized by 16 astrocytomas without p53 alterations and with none of the genetic changes analyzed in this study. Glioblastomas are a heterogeneous group of tumors that may arise via multiple genetic pathways.
Daniel K. Fahim, Claudio E. Tatsui, Dima Suki, Joy Gumin, Frederick F. Lang and Laurence D. Rhines
There is currently no reproducible animal model of human primary malignant bone tumors in the spine to permit laboratory investigation of the human disease. Therefore, the authors sought to adapt their previously developed orthotopic model of spinal metastasis to a model for primary malignant bone tumors of the spine.
A transperitoneal surgical approach was used to implant osteosarcoma (Krib-1) into the L-3 vertebral body of nude mice via a drill hole. Motor function was evaluated daily using the previously validated qualitative key milestones of tail dragging, dorsal stepping, hindlimb sweeping, and paralysis. A subset of these animals was euthanized upon reaching the various milestones, and the spines were removed, sectioned, and stained. The degree of spinal cord compression was correlated with the occurrence of milestones and assessed by a ratio between the neural elements divided by the area of the spinal canal. Another subset of animals received stably transfected Krib-1 cells with the luciferase gene, and bioluminescence was measured at 10, 20, and 30 days postimplantation.
Osteosarcoma xenografts grew in all animals according to a reliable and reproducible time course; the mean time for development of behavioral milestones was noted in relation to the day of implantation (Day 1). Tail dragging (Milestone 1) occurred on Day 19.06 (95% CI 16.11–22.01), dorsal stepping (Milestone 2) occurred on Day 28.78 (95% CI 26.79–30.77), hindlimb sweeping (Milestone 3) occurred on Day 35.61 (95% CI 32.9–38.32), and paralysis of the hindlimb (Milestone 4) occurred on Day 41.78 (95% CI 39.31–44.25). These clinically observed milestones correlated with increasing compression of the spinal cord on histological sections. The authors observed a progressive increase in the local bioluminescence (in photons/cm2/sec) of the implanted level over time with a mean of 2.17 (range 0.0–8.61) at Day 10, mean 4.68 (range 1.17–8.52) at Day 20, and mean 5.54 (range 1.22–9.99) at Day 30.
The authors have developed the first orthotopic murine model of a primary malignant bone tumor in the spine, in which neurological decline reproducibly correlates with tumor progression as evidenced by pathological confirmation and noninvasive bioluminescence measurements. Although developed for osteosarcoma, this model can be expanded to study other types of primary malignant bone tumors in the spine. This model will potentially allow animal testing of targeted therapies against specific primary malignant tumor types.
Joshua J. Chern, Andrew J. Tsung, William Humphries, Raymond Sawaya and Frederick F. Lang
Intracranial hemorrhage (ICH) is a frequent complication found in leukemia patients with thrombocytopenia. At the University of Texas MD Anderson Cancer Center, when a leukemia patient is found to have ICH, a platelet transfusion is generally recommended until 50,000/μl is reached. The authors examine the feasibility and outcome of their intervention strategy in this study.
Records were reviewed from 76 consecutive leukemia patients with newly diagnosed ICH at the University of Texas MD Anderson Cancer Center from January 1, 2007, to December 31, 2009. Variables of interest included age, platelet count at presentation, leukemia subtype, history of trauma, Glasgow Coma Scale score at presentation, whether the 50,000/μl goal was reached after transfusion, and whether the patient was a transfusion responder (platelet count increase > 2000/μl/unit transfused). Outcome parameters were mortality rates at 72 hours and 30 days and imaging-documented hemorrhage progression.
Thrombocytopenia was prevalent at the time of presentation (68 of 76 patients had platelet levels < 50,000/μl at presentation). Despite an aggressive transfusion protocol, only 24 patients reached the 50,000/μl target after an average of 16 units of transfusion. Death due to ICH occurred in 15 patients within the first 72 hours (mortality rate 19.7%). Death correlated with the presenting Glasgow Coma Scale score (p = 0.0075) but not with other transfusion-related parameters. A significant mortality rate was again observed after 30 days (32.7%). The 30-day mortality rate, however, was largely attributable to non-ICH related causes and correlated with patient age (p = 0.032) and whether the patient was a transfusion responder (p = 0.022). Reaching and maintaining a platelet count > 50,000/μl did not positively correlate with the 30-day mortality rate (p = 0.392 and 0.475, respectively).
Platelet transfusion in the setting of ICH in leukemia patients is undoubtedly necessary, but whether the transfusion threshold should be 50,000/μl remains unclear. Factors other than thrombocytopenia likely contribute to the overall poor prognosis.
Claudio E. Tatsui, Frederick F. Lang, Joy Gumin, Dima Suki, Naoki Shinojima and Laurence D. Rhines
There is currently no reproducible animal model of human spinal metastasis that allows for laboratory study of the human disease. Consequently, the authors sought to develop an orthotopic model of spinal metastasis by using a human lung cancer cell line, and to correlate neurological decline with tumor growth.
To establish a model of spinal metastasis, the authors used a transperitoneal surgical approach to implant PC-14 lung tumors into the L-3 vertebral body of nude mice via a drill hole. In 24 animals, motor function was scored daily by using the validated semiquantitative Basso-Beattie-Bresnahan (BBB) scale. A second group of 26 animals (6 or 7 per time point) were sacrificed at specific times, and the spines were removed, sectioned, and stained. Canal compromise was analyzed quantitatively by determining the ratio of the area of the neural elements to the area of the spinal canal on histological sections (neural/canal ratio). Correlations between BBB score and histological evaluation of tumor growth were assessed.
Lung cancer xenografts grew in all animals undergoing functional evaluation (24 mice) according to a reliable and reproducible time course, with paraplegia occurring at a median interval of 30 days following tumor implantation (95% CI 28.1–31.9 days). Importantly, the analysis defined 4 key milestones based on components of the BBB score; these were observed in all animals, were consistent, and correlated with histological progression of tumor. From Days 1 to 14, the mean BBB score declined from 21 to 19. The animals progressed from normal walking with the tail up to walking with the tail constantly touching the ground (milestone 1). The median time to tail dragging was 12 days (95% CI 10.8–13.2). Histological studies on Day 14 demonstrated that tumor had progressed from partial to complete VB infiltration, with initial compression of the neural elements and epidural tumor extension to adjacent levels (mean neural/canal ratio 0.32 ± 0.05, 7 mice). From Days 15 to 20/21 (left/right leg), the mean BBB score declined from 19 to 14. Animals showed gait deterioration, with the development of dorsal stepping (milestone 2). The median time to dorsal stepping was 21 days (95% CI 19.4–22.6) in the left hindlimb and 23 days (95% CI 20.6–25.4) in the right hindlimb. Histological studies on Day 21 demonstrated an increase in the severity of the neural element compression, with tumor extending to adjacent epidural and osseous levels (mean neural/canal ratio 0.19 ± 0.05, 6 mice). From Days 22 to 26/27 (left/right leg), the mean BBB score declined from 14 to 8. Animals had progressive difficulty ambulating, to the point where they showed only sweeping movements of the hindlimb (milestone 3). The median time to hindlimb sweeping was 26 days (95% CI 23.6–28.4) and 28 days (95% CI 27.1–28.9) in the left and right hindlimbs, respectively. Histological studies on Day 28 revealed progressive obliteration of the spinal canal (mean neural/canal ratio 0.09 ± 0.01, 7 mice). From Days 29 to 36, the animals progressed to paralysis (milestone 4). The median time to paralysis was 29 days (95% CI 27.6–30.4) and 30 days (95% CI 28.1–31.9) in the left and right hindlimbs, respectively.
The authors have developed an orthotopic murine model of human spinal metastasis in which neurological decline reproducibly correlates with severity of tumor progression. Although developed for lung cancer, this model can be expanded to study other types of metastatic or primary spinal tumors. Ultimately, this will allow testing of targeted therapies against specific tumor types.