Ependymomas: Clinical and Pathological Aspects

Full access

The propriety of grading ependymomas by a system based upon histological anaplasia, and the effectiveness of irradiation as an adjunctive measure in treatment, are two of the more controversial topics related to this tumor.

Utilizing material from the Armed Forces Institute of Pathology (AFIP) supplemented by cases from the Medical College of Georgia (MCG), we will discuss certain aspects of the classification, pathology, diagnosis, and therapy of ependymomas.

<bold>Source of Material</bold>

One hundred and eighty cases were available for study: 167 from the AFIP; 13 from the MCG. The AFIP material consisted of histological slides and in many

The propriety of grading ependymomas by a system based upon histological anaplasia, and the effectiveness of irradiation as an adjunctive measure in treatment, are two of the more controversial topics related to this tumor.

Utilizing material from the Armed Forces Institute of Pathology (AFIP) supplemented by cases from the Medical College of Georgia (MCG), we will discuss certain aspects of the classification, pathology, diagnosis, and therapy of ependymomas.

Source of Material

One hundred and eighty cases were available for study: 167 from the AFIP; 13 from the MCG. The AFIP material consisted of histological slides and in many cases gross material from which additional slides could be made along with a clinical summary. The patient population from which this material was taken consisted primarily of military personnel and their dependents, veterans, and occasional civilians where slides had been sent to AFIP for consultation. Talmadge Memorial Hospital at the MCG, on the other hand, is a general referral hospital for the State of Georgia.

Cases Excluded

Of the 180 cases, 47 were excluded for one or more of the following reasons:

  1. Inadequate pathological material to permit accurate diagnosis.

  2. Disagreement with the original diagnosis. Most of these cases were seen before 1940 and originally classified as “spongioblastoma ependymale,” they are now considered “glioblastoma multiforme” and so have been omitted. All cases in which the degree of anaplasia was such that the basic ependymal origins were not recognizable were also excluded.

  3. A few cases in the series had been classified as “subependymal glioma,” “subependymal spongioblastoma,” or “subependymal spongioneuroblastoma,” after Kuhlenbeck.6,7,13 He, of course, described this neoplasm as consisting primarily of spongioblastic elements, with only occasional tubule or rosette formations as evidence of any ependymal component. We have not included these tumors in the present series for we feel that one of the problems in the past has been the tendency of some to call a tumor an ependymoma in a supratentorial location with relatively little support, such as the presence of a single perivascular pseudorosette. Such tumors are probably glioblastomas from the outset and simply happen to imitate the ependymoma pattern in a few areas. The converse is probably also true, that in some cases where an ependymoma pattern is present it has gone unrecognized. This is well illustrated in the proceedings of the Cancer Seminar on Intracranial Tumors;5 histopathologic diagnoses submitted by mail on two cases of ependymoma included malignant choroid plexus tumor, chromophobe adenoma, angiosarcoma, medulloblastoma, and 20 “others.” Again, tumors that did not primarily show the patterns described below were excluded.

  4. Scanty or no clinical data available.

Cases Included

One hundred and thirty-three cases remained for this study. Because clinical information was not available for all of these 133 cases, however, small differences in totals will be noted in the accompanying statistics.

Tumor Location

The most frequent site of the tumors in this series as in others was the fourth ventricle.3,9 The location of the tumor was known in 124 cases. Of this number, 86 (69%) were in the posterior fossa; 32 cases (26%) were supratentorial; and 6 cases (5%) arose from the spinal cord or cauda equina.

In most cases of ependymoma of the fourth ventricle, the point of origin was probably the floor, occasionally the cerebellopontine angle. At surgery, however, the exact point from which the tumor arose could not always be determined due to the size of the lesion, its impingement upon the cerebellum, its spread out from the lateral recesses, and extension down over the cervical cord. The surgeon may have the impression that the tumor readily shells away from all adjacent structures.

In 19 of 32 supratentorial tumors, proximity to the ependymal lining of the ventricle and filling or partial filling of the ventricular system were noted either at operation or postmortem examination. Four cases (Table 1) are particularly noteworthy examples. In these, the tumor arose from the septum pellucidum or the ventricular system; one case also had a “tongue” of tumor protruding into the aqueduct with a 5 cm tumor implant in the cerebellum. Three of these tumors were mixed ependymomas and the fourth a subependymoma. Ringertz and Reymond17 have previously pointed out that relation to the ventricles is infrequent in supratentorial tumors. In the present series, however, this relationship was observed in slightly more than one-half of the cases.

TABLE 1

Tumors filling the ventricular system

Patient's Age (yrs)Tumor TypeFindingsDuration of Symptoms
20mixedfilled 3rd ventricle; extended into rt. lat. vent.; filled lf. lat. vent. to temporal hornBroad-based gait for 6 mos, with tendency to fall to rt., headaches
41mixedfilling to tips of both temporal horns and all of 3rd ventricleDizziness, seizures, incontinence, lf. side weakness; 6 mos duration
44mixedpartially filled 3rd and lat. vent. extending into aqueduct; cerebellar implant (5 cm)Weakness in legs, diplopia, dizziness dysphasia, ataxia, but not lateralizing; 10 yrs duration
28subependymomafilled lateral and 3rd ventricleFrontal headaches, episodic loss of coordination; 5 yrs duration. Nausea and vomiting; 5 months duration. Severe rt. side headache before admission

Five tumors were found in the cauda equina and one in the dorsal spinal cord.

Age of Patients

The ages of 113 patients varied from 4 weeks to 64 years (Table 2). The mean age of patients with supratentorial lesions was 27.8 years (30 patients), and that of patients with infratentorial lesions was 18.7 years (83 patients). This is in contrast to Ringertz and Reymond's data which showed no age difference in patients with the two tumors, although there was a predominance of infratentorial tumors in the first decade, as in our series. As these authors pointed out, this lack of age differential is in disagreement with Zulch23 who has commented upon the predominance of cerebral ependymomas in childhood. We found only three in the first decade that were supratentorial as compared to 33 that were infratentorial.

TABLE 2

Patient's age at time of diagnosis of supratentorial or infratentorial tumor

SupratentorialInfratentorial
DecadesNo. CasesDecadesNo. Cases
13133
2326
36316
47412
5457
664
7475
Total3083

The six patients with cord and cauda equina tumors averaged 38.4 years of age, with extremes of 17 and 59 years. This series is mostly made up of military personnel, and therefore sex incidence will not be commented upon.

Diagnosis

The diagnosis of a neoplasm was clinically correct and localization rarely in error in most of the material reviewed. However, the fact that the signs and symptoms may be confusing is evidenced by a consideration of the erroneous diagnoses made. These included multiple sclerosis, senile dementia, lead intoxication, acute demyelinating disease, tuberculous meningitis, stroke, and von Recklinghausen's neurofibromatosis.

Radiologic Diagnosis

Plain skull films of the 10 patients from the MCG showed localizing evidence (calcification) of a mass lesion in only one case. Of the two cases of posterior fossa ependymoma which underwent brain scan, the scan was positive in both, while the only brain scan of a proven supratentorial subependymoma near the midline was negative. In diagnosing supratentorial ependymomas we have found arteriography and air contrast studies to be mutually complementary.

Pantopaque Ventriculography

In our hands, the single most useful contrast study leading to the diagnosis of a posterior fossa ependymoma has been Pantopaque ventriculography. Using positive contrast media, we have been able to outline the anterior and usually the superior aspects of the tumor itself, with limited manipulation of the patient. On the other hand, when air has been introduced from above, we have usually been able to identify little more than a dilated ventricular system above the tumor.

Classification and Pathology

The histological types of ependymomas have been thoroughly described2,8,20 and will be very briefly reviewed here. In agreement with Russell and Rubinstein,18 we have included “subependymal astrocytomas” in this group. In addition we found a significant number of tumors in which the mixture of cellular plus epithelial or papillary types or even neoplastic cells of another glial origin was large enough to warrant a separate category. This group has been labeled as “mixed.” The classification used in this study is as follows (Table 3):

  1. Cellular. This type is made up of uniform cells arranged in a repetitious pattern about blood vessels, leaving “perivascular clear spaces” or “gliovascular structures” devoid of nuclei. This pattern, usually very vascular and varying from densely to sparsely cellular, often in combination with occasional tubules or papillary structures, is the most common. Mitotic figures are rare (Fig. 1).

  2. Papillary-Myxopapillary. The papillary type consists of ependymal cells arranged about a vascular core or papilla. Myxomatous degeneration is often present to a greater or lesser degree, producing the second variety under this heading. Tumors with pronounced degeneration were almost exclusively found in the cauda equina (Fig. 2).

  3. Subependymoma. This is a densely fibrillary tumor containing unevenly scattered clusters of ependymal cells and astrocytes. This pattern was often seen about the periphery of other ependymomas as well as occasionally being the predominant pattern (Fig. 3).

  4. Epithelioid. This type is composed of orderly, well-formed canals and tubules, the cells of which abut upon a basement membrane. This tumor type was not encountered as a pure type in our series but was seen in combination with another (usually the “cellular”) pattern (Fig. 4, left).

  5. Mixed. Tumors were seen containing mixtures of the types above, most commonly the “cellular” pattern in combination with one of the others. Where this mixture was great and not just a random canal in an otherwise cellular ependymoma, the tumor was placed in this category. In addition, other glial types, often oligodendroglia but frequently astrocytes, were seen in tumors that were otherwise typical ependymomass (Fig. 4 right).

Fig. 1.
Fig. 1.

Photomicrograph of a cellular ependymoma. H. & E., ×200.

Fig. 2.
Fig. 2.

Left: Photomicrograph of a papillary ependymoma. H. & E., ×200. Right: Photomicrograph of a myxopapillary ependymoma. H. & E., ×100.

Fig. 3.
Fig. 3.

Left: Subependymoma (subependymal glomerate astrocytoma). H. & E., ×200. Right: Subependymoma and cellular ependymoma occurring in adjacent areas of the same neoplasm. H. & E., ×100.

Fig. 4.
Fig. 4.

Left: Epithelial ependymoma. H. & E., ×200. Right: Mixed ependymoma. Typical gliovascular structures and larger “ependymal” cells were interspersed with small cells surrounded by haloes. Calcosphenites were prominent among the oligodendroglia. H. & E., ×100.

TABLE 3

Classification of tumors

Tumor TypeNo.
Cellular64
Mixed34
Myxopapillary-Papillary5
Subependymoma14
Epithelioid1

In agreement with Kernohan and Fletcher8 and Russell and Rubinstein18 we found the “cellular” type to be most common (Fig. 3 left); in combination with other patterns it constituted approximately three fourths of the tumors studied. These patterns were all readily identified with the usual hematoxylin-and-eosin stains, so that special staining techniques were not needed. In an attempt to provide greater uniformity of thought between clinicians and pathologists, Kernohan and his associates9,10,14 introduced a system of grading ependymomas. Based upon histological evidence of anaplasia this system graded ependymomas from I to IV, in increasing order of malignancy. In discussing the criteria of a grade I ependymoma, the authors pointed out: “These tumors, although composed entirely of normal-appearing cells, are considered to be malignant because they are tumefactions located in tissue which is particularly vulnerable to the effects of space-occupying lesions.” Since the “tissue” is usually the vital brain stem or equally sensitive nuclear groups in the region of the third ventricle, the prognostic value of the system is confused by the unusual hazards of surgery in this region regardless of the “grade” of the tumor.22

In a later publication Kricheff, et al.,12 referred to the ependymomas as relatively benign tumors and related their malignancy to compression of surrounding vital structures.

A review of operative notes accompanying our material revealed that the operator often had the impression that he was dealing with a well-circumscribed or “encapsulated” lesion.

Change in Tumor Characteristics

In 21 cases both biopsy and autopsy material was available (Table 4). Twelve of these patients survived more than 30 postoperative days while nine died within 3 days of operation. With one exception, tumors from patients surviving surgery showed no greater degree of histological malignancy at postmortem examination than at the time of surgery. The single exception occurred in a 34-year-old man whose frontoparietal tumor was considered a grade III cellular ependymoma at the time of surgery; at the time of death, 26 months later, it was unquestionably a glioblastoma. Of 15 tumors occurring in the fourth ventricle, none showed greater malignancy at postmortem than at surgery regardless of survival time; one was graded III both times and the remainder were histologically benign. Russell and Rubinstein have pointed out that cytologically malignant tumors are uncommon even in recurrences. Except as noted, in all cases so studied the cytoarchitectural pattern at surgery was the same as that at autopsy.

TABLE 4

Comparison of biopsy and autopsy diagnoses in 21 cases

Tumor LocationBiopsyAutopsyPostop Death
frontoparietalcellular ependymoma IIIglioblastoma IV20 mos
parietalmixed ependymoma IVmixed ependymoma IV7 mos
IV ventriclecellular Icellular ependymoma I11 mos
IV ventriclecellular IIcellular II3 mos
cerebellumpapillary Ipapillary I10 mos
IV ventriclecellular IIIcellular III12 mos
IV ventriclecellular Icellular I24 mos
T-6cellular malignant Icellular IV5 wks
IV ventricle seededpapillary Ipapillary I58 mos
temporalcellular IVcellular IV4 mos
dorsal cordcellular Icellular I6 mos
IV ventriclecellular Icellular I48 days
IV ventriclemixed IImixed II24 hrs
IV ventriclecellular IIcellular IIat surgery
IV ventriclecellular Icellular I24 hrs
IV ventriclecellular IIcellular II24 hrs
IV ventriclemixed Imixed I24 hrs
IV ventriclecellular Icellular I24 hrs
septum pellucidummixed Imixed I24 hrs
IV ventriclecellular Icellular I48 hrs
IV ventriclecellular Icellular I72 hr

These observations seem to indicate, at least so far as infratentorial tumors are concerned, that persistent histological benignancy is the rule rather than the exception. There are exceptions, however; in addition to the case given above, one of us has made the definite diagnosis of ependymoma and glioblastoma in different sections of the same irradiated tumor of the fourth ventricle.

Seeding of Ependymomas

The frequency of seeding from ependymomas has not been clarified. One of our benign tumors had seeded the subarachnoid space at the time of death, 58 months following operation. Svien, et al.,20 likewise noted seeding from histologically benign tumors; none of their patients developed symptoms from the implants. Kricheff, et al.,11 examined the spinal cord in 12 cases of ependymoma without giving the location of the primary tumor and found seeding in none. (The time following surgery was not stated.) One patient in their series of 65 had clinical evidence of spinal cord implantation. Polmeteer and Kernohan16 had previously reported their findings in 42 cases of gliomas in which implants were present; in this group ependymomas ranked third in order of frequency of implantation, behind medulloblastoma and glioblastoma. In a later article Svien, et al.,19 further emphasized the frequency of implantation by reporting six cases of posterior fossa ependymoma with associated cord implantation. This was an incidence of 50% association with implantation from ependymomas of the posterior fossa. None was found associated with tumors above the tentorium. None of their cases showed clinical evidence of spinal cord metastases; this point is worth remembering when one is considering whether or not to remove the spinal cord at an autopsy on a patient with an ependymoma. Phillips, et al.,15 reported two cases of “clinically significant” seeding in 42 cases of ependymoma. In contrast to these cases in which seeding was asymptomatic, Tarlov and Davidoff21 and others have reported instances in which spinal cord implants have been responsible for symptoms, and they found a 37.5% incidence of seeding.

That seeding may occur prior to surgery, that the symptoms produced by the implant may be quite late in appearing, and that the initial surgery may be for symptoms produced by the implant and not the primary lesions, have all been previously shown.19 With one exception all cases of seeding were seen in postoperative patients. The single exception occurred in a patient with a tumor described as filling the third and lateral ventricle with a 5 cm implant in the cerebellar hemisphere. One patient was operated on for symptoms due to seeding into the spinal subarachnoid space, the primary tumor being in the fourth ventricle. Of 62 autopsied cases, there were 16 instances of seeding. However, it is not known whether the spinal cord was examined in all cases and, further, if the clinical histories were taken into consideration; there were a number of cases in which the patient had symptoms that could easily have been related to cord or cauda equina involvement, and the spinal cord was not taken for examination. In nine cases in which seeding occurred, the primary site was in the posterior fossa; in five, the primary was supratentorial, and in two, the spinal cord.

One patient with an ependymoma of the fourth ventricle survived 12 years following surgery and at autopsy no tumor was found at the primary site, although there was tumor in the spinal cord subarachnoid space. This had not produced symptoms and the patient had died of unrelated causes.

Therapy

The patients in our series were treated at several different hospitals by different surgeons and radiation therapists. This obviously unstandardized approach must of course limit the evaluation of results. However, we still feel that certain points may be made.

To evaluate methods of therapy and further appraise some of the clinicopathological correlates pertaining to ependymomas, we studied 52 AFIP and MCG patients for whom follow-up data were available and who survived more than 30 days postoperatively. Patients who died in less than 30 days we considered operative deaths and excluded from the group. We divided the group of surviving patients according to location of the lesion and method of treatment; 36 had been treated with surgery followed by radiation and 16 by surgery alone (Table 5).

TABLE 5

Results of treatment

Tumor LocationSurgery AloneCombined Therapy 
PatientsAv. Surv. (mos)PatientsAv. Surv. (mos) 
Posterior Fossa832.924 
Supratentorial741.38 
Spinal Cord & Cauda Equina11.34 

Surgery Plus Radiation

Of the 36 patients receiving both forms of therapy, 24 had posterior fossa lesions; 8 were supratentorial and in 4 the tumor involved either the spinal cord or the cauda equina. None of the 8 tumors located supratentorially was considered histologically benign, all being graded from III to IV. Seven patients in this group have now died; the eighth was still living 2 months after treatment. Excluding the one patient still surviving, the average post-treatment survival time of the remaining seven was 16.1 months, and there were no 5-year cures. The duration of preoperative symptoms in all patients with supratentorial tumors averaged 11.5 months (known in 12 patients), with extremes of 1 to 60 months.

Of the 24 patients with posterior fossa lesions receiving both forms of treatment, the tumor was graded I in 15, II in six, and III in only one. No tumor was graded IV. Two of the neoplasms could not be graded because of scanty material. The average duration of preoperative symptomatology in patients with infratentorial tumors (known in 41 patients) was 5.6 months, with extremes of 2 weeks to 4 years. Two patients were lost to follow-up at 16 and 15 years at which time they were reported gainfully employed. A third was discharged from the hospital after completion of treatment and no further follow-up is available. Five patients are living with minimal to no neurological deficit, at 12, 17, 18, 24, and 36 months following completion of treatment. Sixteen patients have died. The average survival time of these 16 patients was 46.5 months, and of all cases in this group was 53.2 months.

Four patients had tumors involving the cord (one patient) or cauda equina (three) and received both forms of therapy. One tumor was grade II, and three were grade I. The duration of preoperative symptomatology was 5 months in one case, 2 years in another, and unknown in two. One patient had a grade I cellular ependymoma (spinal cord) and one a mixed ependymoma grade I (cauda equina); they died at 24 and 6 months respectively, both having widespread seeding throughout the neuroaxis. The other two patients with tumor involving the cauda equina were alive and asymptomatic 31 months (grade I tumor) and 5 years (grade II tumor) post surgery; the latter was re-explored and no tumor found.

Ayres,1 in his excellent review of cauda equina ependymomas, pointed out that cellular and epithelial types have a somewhat worse prognosis than the myxopapillary. He also felt that the immediate prognosis for all types was good but the long-term prognosis poor.

Surgery Alone

Sixteen patients were treated with surgery alone. Seven of these patients had supratentorial tumors, two of which were grade I, one grade II, two grade III, and two grade IV. One patient was lost to follow-up at 14 years (grade II) and another was reportedly asymptomatic at 5 years (grade I). Patients with grade III and IV tumors had survival times of 4 to 24 months; patients with grade I and II tumors had survival times of 6 weeks to 14 years. Five patients have died after survival times ranging from 6 weeks to 2 years. The patient who died 6 weeks postoperatively had a histologically benign tumor that involved the region about the third ventricle; again, a benign tumor in a malignant location. The long-term survivors had essentially hemispheric involvement.

Eight patients in the non-irradiated group had infratentorial lesions. Of the patients about whom follow-up information is available (seven), the average postoperative survival time was 32.9 months, but with wide individual variation and no significant correlation with grade of tumor. Three patients with grade IV ependymomas were in this group and were the only three with grade IV tumors in the entire group of 32 posterior fossa tumors studied regarding therapy. Two of these grade IV tumors later developed extensive subarachnoid and cerebral implants. The other five tumors in the non-irradiated group were grade I. One (grade I, cellular) survived 12 years post-surgery and died of unrelated causes with no tumor being found at postmortem; one was lost to follow-up 1 month after discharge from the hospital, and six died after survival times of 7 to 36 months. Two of these patients, both with grade IV lesions, had extensive subarachnoid and cerebral implants.

There was one cord tumor in the non-irradiated group, a cellular ependymoma at the T-6 spinal level, treated with decompression; this patient died 5 weeks after surgery.

Little can be said regarding the efficacy of irradiation in the small group of patients with supratentorial tumors. The fact that one case in the non-irradiated group of supratentorial tumors survived 14 years produced tremendous bias. If this case were excluded, the results of surgical vs combined therapy would be essentially the same.

However, when results of treatment of infratentorial tumors are compared, the average survival time of the irradiated group is approximately double that of the non-irradiated group. Among patients with infratentorial tumors not receiving radiation there was only one 5-year survivor (12 years) and two patients still living. Among those receiving radiation there were four 5-year survivors and seven patients still living. It appears, therefore, that radiation may provide effective palliation, but we have no indication that it is curative. Little can be said regarding cord tumors because of the small number in our series.

Evaluation of “Grading” Systems

A correlation between grade of tumor and survival time could not be made. The number in each group of non-irradiated tumors was small; among the irradiated infratentorial tumors there was only one histologically malignant tumor while there were no histologically benign tumors among the irradiated supratentorial lesions. One gains the impression that among posterior fossa ependymomas histological benignancy is the rule and “malignancy” is primarily a reflection of damage to surrounding structures, but that among supratentorial ependymomas histological malignancy is more common.

Taveras' group,11,12 in studying a series of ependymal tumors treated both by surgery and irradiation, arrived at essentially the same conclusion, reporting that, “Patients with supratentorial tumors have a poorer prognosis and generally have greater neurologic deficit than do patients with infratentorial ependymomas.” However, the high operative mortality rate, the pronounced variation in length of survival times regardless of grade, and the frequency of seeding associated with histologically “benign” tumors all seem to rule against effective utilization of a system of histological grading.

The histological pattern was also studied as a means of prognosticating and found to be unreliable, with widely varied survival times noted for all subtypes. There was no tendency for neoplasms with well-developed canals, for example, to have any more favorable prognosis than essentially cellular tumors, a group that included the longest survivors. With regard to cauda equina and cord tumors, our series is too small to be significant other than to say that results in two MCG cases do not disagree with Ayres' conclusion that patients with myxopapillary ependymomas of the cauda have a relatively favorable prognosis.

Floor of Fourth Ventricle

The question of mode of involvement of the structures beneath the floor of the fourth ventricle is a pertinent one. A review of the pathological material reveals that this is most often by compression. Extensive infiltration as is seen with the astrocytomas is the exception, although a very narrow zone of infiltration is seen occasionally. This has also been the impression of the operating neurosurgeon when the dissection was carried to the floor of the ventricle and the tumor found to “shell out.” However, even this bit of manipulation may be attended by a high mortality; the dissection appears best stopped short of this point.

Summary

Our review of 133 verified cases of ependymoma derived from the Armed Forces Institute of Pathology and the Medical College of Georgia, has led to the following conclusions:

  1. Ependymomas are most frequently found in the posterior fossa and most arise from the floor of the 4th ventricle.

  2. Supratentorial ependymomas in infancy and early childhood are uncommon; infratentorial ependymomas occur with about ten times greater frequency.

  3. Proximity to and partial or complete filling of the lateral and 3rd ventricles are often characteristic.

  4. Histological subclassification of ependymomas is useful for categorizing these tumors but with the possible exception of myxopapillary ependymomas of the cauda equina it appears there is no correlation between clinical prognosis and tumor subtype.

  5. Accurate prognosis based on a system of grading anaplasticity is thwarted by the characteristic proximity of ependymomas to vital CNS structures and consequent damage to these structures even by histologically “benign tumors.”

  6. Supratentorial ependymomas tend to show greater anaplasia and pleomorphism, and in general appear less well organized than ependymomas of the posterior fossa. Supratentorial ependymomas seem to metastasize throughout the neural axis with greater frequency than has been previously appreciated.

  7. Seeding of ependymomas occurs, although the exact frequency is uncertain and the symptoms secondary to the seeding may be quite late in appearing.

  8. Among ependymomas of the posterior fossa, there is a pronounced tendency for both the cytoarchitectural pattern and the degree of histological malignancy to remain constant throughout the course of the disease regardless of duration.

  9. It is difficult to assess the effectiveness of treatment because we are unable to correlate the histological appearance of an ependymoma with its clinical behavior. Adequate surgical decompression without disturbance of vital structures combined with postoperative radiation seems to be most effective; with rare exceptions this method of treatment is not curative.

References

  • 1.

    AyresW. W. Ependymoma of the cauda equina. A report of the clinicopathologic aspects and follow-up studies of eighteen cases. Milit. Med.1958122:1035.AyresMilit. Med.122:10–35.

  • 2.

    BaileyP. A study of tumors arising from ependymal cells. 19242:127.Bailey2:1–27.

  • 3.

    CraigW. McK.KernohanJ. W. Tumors of the fourth ventricle. J. Am. med. Ass.1938111:23702377.CraigKernohanJ. Am. med. Ass.111:2370–2377.

  • 4.

    DavisL.DavisR. A. Principles of neurological surgery. Philadelphia: W. B. Saunders Co.19631st ed.608 pp. (See pp. 121.)DavisDavisPrinciples of neurological surgery

  • 5.

    Del RegatoJ. A. Cancer seminar on intracranial tumors. Proceedings. vol. 32Oct. 1963.Del RegatoCancer seminar on intracranial tumors. Proceedings

  • 6.

    GlobusJ. H.KuhlenbeckH. Tumors of the striatothalamic and related regions: their probable source of origin and more common forms. Archs Path.194234:674734.GlobusKuhlenbeckArchs Path.34:674–734.

  • 7.

    GlobusJ. H.KuhlenbeckH. The subependymal cell plate (matrix) and its relationship to brain tumors of the ependymal type. J. Neuropath. exp. Neurol.19443:135.GlobusKuhlenbeckJ. Neuropath. exp. Neurol.3:1–35.

  • 8.

    KernohanJ. W.FletcherE. M. Ependymomas. A study of 109 cases. Proc. Ass. Res. nerv. ment. Dis.193716:182209.KernohanFletcherProc. Ass. Res. nerv. ment. Dis.16:182–209.

  • 9.

    KernohanJ. W.WoltmanH. W.AdsonA. W. Gliomas of the cerebellopontine angle. J. Neuropath. exp. Neurol.19487:349367.KernohanWoltmanAdsonJ. Neuropath. exp. Neurol.7:349–367.

  • 10.

    KernohanJ. W.SayreG. P. Tumors of the central nervous system. In: Atlas of tumor pathology. Washington, D.C.: Armed Forces Institute of Pathology19521st ed. (See pp. 4359.)KernohanSayreAtlas of tumor pathology

  • 11.

    KricheffI. I.BeckerM.SchneckS. A.TaverasJ. M. Intracranial ependymomas: a study of survival in 65 cases treated by surgery and irradiation. Am. J. Roentg.196491:166175.KricheffBeckerSchneckTaverasAm. J. Roentg.91:166–175.

  • 12.

    KricheffI. I.BeckerM.SchneckS. A.TaverasJ. M. Intracranial ependymomas: factors influencing prognosis. J. Neurosurg.196421:714.KricheffBeckerSchneckTaverasJ. Neurosurg.21:7–14.

  • 13.

    KuhlenbeckH. Neoplastic transformation of the subependymal cell plate in the floor of the fourth ventricle (subependymal spongioblastoma): with a clinicopathologic case report. J. Neuropath. exp. Neurol.19476:139151.KuhlenbeckJ. Neuropath. exp. Neurol.6:139–151.

  • 14.

    MabonR. F.SvienH. J.KernohanJ. W.CraigW. McK. Ependymomas. Proc. Staff Meet. Mayo Clin.194924:6571.MabonSvienKernohanCraigProc. Staff Meet. Mayo Clin.24:65–71.

  • 15.

    PhillipsT. L.ShelineG. E.BoldreyE. Therapeutic considerations in tumors affecting the central nervous system: ependymomas. Radiology196483:98105.PhillipsShelineBoldreyRadiology83:98–105.

  • 16.

    PolmeteerF. E.KernohanJ. W. Meningeal gliomatosis. A study of forty-two cases. 194757:593616.PolmeteerKernohan57:593–616.

  • 17.

    RingertzN.ReymondA. Ependymomas and choroid plexus papillomas. J. Neuropath. exp. Neurol.19498:355380.RingertzReymondJ. Neuropath. exp. Neurol.8:355–380.

  • 18.

    RussellD. S.RubinsteinL. J. Pathology of tumors of the nervous system. London: Edward Arnold Ltd.19591st ed.318 pp. (See pp. 132142.)RussellRubinsteinPathology of tumors of the nervous system

  • 19.

    SvienH. J.GatesE. M.KernohanJ. W. Spinal subarachnoid implantation associated with ependymoma. 194962:847856.SvienGatesKernohan62:847–856.

  • 20.

    SvienH. J.MabonR. F.KernohanJ. W.CraigW. McK. Ependymoma of the brain: pathologic aspects. 19533:115.SvienMabonKernohanCraig3:1–15.

  • 21.

    TarlovI. M.DavidoffL. M. Subarachnoid and ventricular implants in ependymal and other gliomas. J. Neuropath. exp. Neurol.19465:213224.TarlovDavidoffJ. Neuropath. exp. Neurol.5:213–224.

  • 22.

    TrolardC. E.KendrickJ. F.SahyounP. F. Ependymoma: a critical re-evaluation of classification with report of cases. J. Neuropath. exp. Neurol.195110:295308.TrolardKendrickSahyounJ. Neuropath. exp. Neurol.10:295–308.

  • 23.

    ZulchK. J. . New York: Springer Publishing Co.19652nd ed.326 pp. (See pp. 178189.)Zulch

Article Information

© AANS, except where prohibited by US copyright law.

Headings

Figures

  • View in gallery

    Photomicrograph of a cellular ependymoma. H. & E., ×200.

  • View in gallery

    Left: Photomicrograph of a papillary ependymoma. H. & E., ×200. Right: Photomicrograph of a myxopapillary ependymoma. H. & E., ×100.

  • View in gallery

    Left: Subependymoma (subependymal glomerate astrocytoma). H. & E., ×200. Right: Subependymoma and cellular ependymoma occurring in adjacent areas of the same neoplasm. H. & E., ×100.

  • View in gallery

    Left: Epithelial ependymoma. H. & E., ×200. Right: Mixed ependymoma. Typical gliovascular structures and larger “ependymal” cells were interspersed with small cells surrounded by haloes. Calcosphenites were prominent among the oligodendroglia. H. & E., ×100.

References

1.

AyresW. W. Ependymoma of the cauda equina. A report of the clinicopathologic aspects and follow-up studies of eighteen cases. Milit. Med.1958122:1035.AyresMilit. Med.122:10–35.

2.

BaileyP. A study of tumors arising from ependymal cells. 19242:127.Bailey2:1–27.

3.

CraigW. McK.KernohanJ. W. Tumors of the fourth ventricle. J. Am. med. Ass.1938111:23702377.CraigKernohanJ. Am. med. Ass.111:2370–2377.

4.

DavisL.DavisR. A. Principles of neurological surgery. Philadelphia: W. B. Saunders Co.19631st ed.608 pp. (See pp. 121.)DavisDavisPrinciples of neurological surgery

5.

Del RegatoJ. A. Cancer seminar on intracranial tumors. Proceedings. vol. 32Oct. 1963.Del RegatoCancer seminar on intracranial tumors. Proceedings

6.

GlobusJ. H.KuhlenbeckH. Tumors of the striatothalamic and related regions: their probable source of origin and more common forms. Archs Path.194234:674734.GlobusKuhlenbeckArchs Path.34:674–734.

7.

GlobusJ. H.KuhlenbeckH. The subependymal cell plate (matrix) and its relationship to brain tumors of the ependymal type. J. Neuropath. exp. Neurol.19443:135.GlobusKuhlenbeckJ. Neuropath. exp. Neurol.3:1–35.

8.

KernohanJ. W.FletcherE. M. Ependymomas. A study of 109 cases. Proc. Ass. Res. nerv. ment. Dis.193716:182209.KernohanFletcherProc. Ass. Res. nerv. ment. Dis.16:182–209.

9.

KernohanJ. W.WoltmanH. W.AdsonA. W. Gliomas of the cerebellopontine angle. J. Neuropath. exp. Neurol.19487:349367.KernohanWoltmanAdsonJ. Neuropath. exp. Neurol.7:349–367.

10.

KernohanJ. W.SayreG. P. Tumors of the central nervous system. In: Atlas of tumor pathology. Washington, D.C.: Armed Forces Institute of Pathology19521st ed. (See pp. 4359.)KernohanSayreAtlas of tumor pathology

11.

KricheffI. I.BeckerM.SchneckS. A.TaverasJ. M. Intracranial ependymomas: a study of survival in 65 cases treated by surgery and irradiation. Am. J. Roentg.196491:166175.KricheffBeckerSchneckTaverasAm. J. Roentg.91:166–175.

12.

KricheffI. I.BeckerM.SchneckS. A.TaverasJ. M. Intracranial ependymomas: factors influencing prognosis. J. Neurosurg.196421:714.KricheffBeckerSchneckTaverasJ. Neurosurg.21:7–14.

13.

KuhlenbeckH. Neoplastic transformation of the subependymal cell plate in the floor of the fourth ventricle (subependymal spongioblastoma): with a clinicopathologic case report. J. Neuropath. exp. Neurol.19476:139151.KuhlenbeckJ. Neuropath. exp. Neurol.6:139–151.

14.

MabonR. F.SvienH. J.KernohanJ. W.CraigW. McK. Ependymomas. Proc. Staff Meet. Mayo Clin.194924:6571.MabonSvienKernohanCraigProc. Staff Meet. Mayo Clin.24:65–71.

15.

PhillipsT. L.ShelineG. E.BoldreyE. Therapeutic considerations in tumors affecting the central nervous system: ependymomas. Radiology196483:98105.PhillipsShelineBoldreyRadiology83:98–105.

16.

PolmeteerF. E.KernohanJ. W. Meningeal gliomatosis. A study of forty-two cases. 194757:593616.PolmeteerKernohan57:593–616.

17.

RingertzN.ReymondA. Ependymomas and choroid plexus papillomas. J. Neuropath. exp. Neurol.19498:355380.RingertzReymondJ. Neuropath. exp. Neurol.8:355–380.

18.

RussellD. S.RubinsteinL. J. Pathology of tumors of the nervous system. London: Edward Arnold Ltd.19591st ed.318 pp. (See pp. 132142.)RussellRubinsteinPathology of tumors of the nervous system

19.

SvienH. J.GatesE. M.KernohanJ. W. Spinal subarachnoid implantation associated with ependymoma. 194962:847856.SvienGatesKernohan62:847–856.

20.

SvienH. J.MabonR. F.KernohanJ. W.CraigW. McK. Ependymoma of the brain: pathologic aspects. 19533:115.SvienMabonKernohanCraig3:1–15.

21.

TarlovI. M.DavidoffL. M. Subarachnoid and ventricular implants in ependymal and other gliomas. J. Neuropath. exp. Neurol.19465:213224.TarlovDavidoffJ. Neuropath. exp. Neurol.5:213–224.

22.

TrolardC. E.KendrickJ. F.SahyounP. F. Ependymoma: a critical re-evaluation of classification with report of cases. J. Neuropath. exp. Neurol.195110:295308.TrolardKendrickSahyounJ. Neuropath. exp. Neurol.10:295–308.

23.

ZulchK. J. . New York: Springer Publishing Co.19652nd ed.326 pp. (See pp. 178189.)Zulch

TrendMD

Metrics

Metrics

All Time Past Year Past 30 Days
Abstract Views 4 4 4
Full Text Views 90 90 30
PDF Downloads 70 70 46
EPUB Downloads 0 0 0

PubMed

Google Scholar