:Acta N&?ochirurgica

Acta Neurochir (Wien) (1992) 116:23-32

9 Springer-Verlag 1992

Printed in Austria

Neurosurgical Experience with Tumours of the Pineal Region at Clinica Puerta de Hierro J. Vaquero, J. Ramiro, R. Martinez, and G. Bravo Department of Neurosurgery, Clinica Puerta de Hierro, Autonomous University, Madrid, Spain

Summary The clinicopathological experience with 50 cases of pineal region tumours at Clinica Puerta de Hierro is presented. In this series, 88% of the patients were evaluated by CT-scan. Pineal region turnouts make up approximately 0.7 % of the intracranial expansive processes in the Spanish population. The largest group of lesions appearing in this localization is that of the germinomas (38%), followed by nontumoural lesions (20%) and tumours generally considered to be of the vicinity, such as meningiomas, gliomas and metastases (18 %), tumours of the pineal parenchyma (14%), and non-germinoma germinal tumours (10%). In our series, in addition to an intracranial hypertension syndrome, an ophthalmological and, to a minor degree, an endocrinological syndrome predominate in germ-cell tumours, with a cerebellar syndrome appearing in gliomas of the pineal region. All the patients in the series diagnosed as having a germinoma and treated by irradiation are alive, and free of disease, after followup ranging from 2 to 20 years (mean: 8 years). The experience obtained with the present series supports the opinion that, in radiosensitive tumours, surgical resection adds no therapeutic benefit to treatment with radiotherapy alone. We suggest that when dealing with a turnout of the pineal region, CT-scan and clinical assessment now permit an initial selection of patients susceptible to surgery as a first therapeutic option, indicating those patients who, because they are considered to have either a "probable germinoma" or a "tumour of uncertain diagnosis", should undergo stereotaxic biopsy or trial radiotherapy and, only when this has proved a failure, should be subjected to open surgery.

used in this series. We have intentionally excluded suprasellar germinomas as we consider that, given their anatomical location, their clinical picture is different. On the other hand, we have included nontumoural expansive processes to assess their relative incidence among the Spanish population with respect to the totality of pineal region masses. Clinical Material and Methods The present series consists of 37 male and 13 female patients, with a mean age of 23 years. Figure 1 shows the distribution of the patients within the different decades of life.

Types of Lesions In order to group them histogenetically, and taking into account Teilum's concept regarding the differential 25-

Males: 37 Females: 13 20

r O;

16-

13.

Keywords: Brain tumour; pineal tumour, radiation therapy.

o 10"

JO

Introduction Over the last 20 years, 50 patients with tumour of the pineal region were studied in the Service of Neurosurgery of Clinica Puerta de Hierro in Madrid. This series consists of 31 verified tumours and 19 unverified. The purpose of this work is to report our clinicopathological experience with these lesions and to analyze the different therapeutic options which have been

E

0-10

10-20

20-30

30-40

40-50

50-60

Age Fig. 1. Age distribution of the patients of the series

60-70

24

J. Vaquero et aL: Neurosurgical Experience with Tumours of the Pineal Region at Clinica Puerta de Hierro

Table 1. Classification and Number of Pineal Region Lesions Seen at the Clinica Puerta de Hierro from 1970 to 1989 Gerrninoma (19) Embryonal carcinoma (2) Ohoriocarcinoma

4%

48%

--Endoderma[ sinus tumour

- Germ-cell origin (24)

--Benign teratoma (1)

4%

/ E p i d e r m o i d cyst --Teratoid( \ Dermoid cyst -- Malignant teratoma (2)

- Pineocytoma (4)

TUMOURS (40)

._~ -- originNeur~176

b G L

Pineoblastoma (3) licmas (5) Others

- - Meningioma (2) ~_ Others

Mesodermic - - o r i g i n (2)

-- Other tumours

TUMOURSOF GERM-CELLORIGIN

[]

NEUROECTODERMICTUMOURS

[]

MESODERMICTUMOURS

[]

OTHERTUMOURS

[]

NON-TUMOURALLESIONS

Fig. 2. Graph showing the relative percentage of patients in each group of lesions in the series

(2)

~

Glial cysts (2) Arachnold cysts (3)

NON-TUMOURAL LESIONS (10)

[]

Parasitic cysts (1)

Vascular hamartomas (4) Others

potentiality of the germ-cell42, a specific classification for pineal region tumour has been applied. This classification and the incidence of the different lesions in our series are shown in the Table 1. Figure 2 shows the relative percentage of each subgroup within the series. Tumours of Germ-Cell Origin They represented the largest group of lesions in this series (48%) and the most numerous with respect to other tumours (60%). The age of this patients ranged from 4 to 33 years (mean: 16 years), and the predominance of males was noteworthy (22/2). Germinomas: They are generally accepted to be the most common tumour of the pineal region. Our series includes 19 cases. Nevertheless, only 7 of them have been verified histologically. The rest were considered to be germinomas on the basis of clinical data, response to radiotherapy and evolution. The age of these patients ranged between 4 and 33 years (mean: 16 years), and

most of them were in the second decade of life (63%); the male/female ratio was 17/2. Embryonal carcinomas: Two patients of the series were classified as having an embryonal carcinoma. The first, a 13-year-old boy, underwent surgery in 1973 for a pineal mass pathologically diagnosed as a benign teratoma. Nevertheless, eight months later, the symptoms reappeared and a second craniotomy revealed a typical embryonal carcinoma. The patient died one year later. The second case was a 15-year-old boy with an elevated alpha-fetoprotein (AFP) level (17,769 ng/ml) in his cerebrospinal fluid (CSF) and a large, extraordinarily vascularized tumour in the pineal region. The tumour disappeared with radiotherapy and the patient is asymptomatic 6 years later. Although the possibility of an AFP-secreting germinoma or an endodermal sinus tumour can not be ruled out, and the precise nature of this tumour remains speculative, it was presumed to be an embryonal carcinoma. Benign teratomas: In this series, there was only one case of well-differentiated teratoma with benign outcome. The subject was a 20-year-old male who underwent resection of a partially calcified pineal mass in November of 1987. A correct presumptive diagnosis was pre-operatively established on the basis of computerized tomography (CT)-scan and magnetic resonance imaging (MRI).

J, Vaquero et al.: Neurosurgical Experience with Tumours of the Pineal Region at Clinica Puerta de Hierro

Malignant teratomas: Two such cases are reported in our series. The first was a 32-year-old male with an AFP-secreting pineal tumour, that was treated with chemotherapy and radiotherapy after a ventriculoperitonal shunt had been performed in another hospital. The patient died two years later due to bone marrow aplasia and sepsis, and necropsy revealed a teratomatous lesion mainly composed of epidermoid cystic tissue. The second case was a 14-year-old male whose tumour was surgically verified and irradiated. He died one year later.

25

Other Tumours In this subgroup were two patients with metastatic tumours. One was a 64-year-old man with a confirmed bronchial carcinoma, who developed a single brain metastasis restricted to the pineal origin. He underwent a ventriculoperitoneal shunt as a palliative measure, and died one month later. The other case was a 37-yearold female, subjected to surgery for a pineal region tumour with brain stem involvement, which was found to be a melanoma. A few weeks after surgery, systemic dissemination from an alleged primary skin melanoma was disclosed, and she died two months after diagnosis.

Tumours of Neuroectodermal Origin This group represented 24% of the cases in the series and 30% when only tumoural lesions are considered. The age range was 18 months to 63 years (mean: 24 years), and the male/female ration was 7/5. Pineocytomas." Four patients of the present series were surgically treated for pineocytoma. These patients ranged in age between 12 and 44 years (mean: 22,5 years) with a male/female ration of 3/1. All the cases showed signs of neuronal differentiation in the pathological study, according to the criteria of Rubinstein 31, and they have been included in a previously reported series47. Pineoblastomas." Three patients of the series, aged 18 months and 6 and 48 years, underwent removal of a pineoblastoma. All were females and they died, because of their diseases, at ages of 4.9 and 55 years, respectively. Gliomas: This subgroup consisted of five patients. Two male patients, with tumours pathologically diagnosed as astrocytomas, were 3 and 2l years old, respectively, at operation, and they are alive, 12 and 4 years after surgery. A 6-year-old female underwent surgery for a malignant astrocytoma, and she died four months later due to progression of residual tumour. Two males were 50 and 63 years when a diagnosis of glioblastoma multiforme was established. One of these cases was surgically verified, and both patients died six months after diagnosis.

Nontumoural Lesions Ten patients of the series (20%) had expansive nontumoural lesions: 2 glial cysts, 3 arachnoid cysts, 1 case of cysticerosis, 1 case of arteriovenous malformation, and 3 cavernous angiomas. Some of these cases have been described in detail previously4.-46.

Clinical Syndromes Four different types of clinical syndrome were clearly distinguished in the patients of this series (Fig. 3-7). A syndrome of intracranial hypertension was recorded in 84% of the germinomas, 80% of the nongerminoma germ-cell tumours, 86% of the parenchyreal pineal tumours, 80% of the gliomas, and 70% of the nontumoural lesions. An ophthalmological syndrome defined by one or more ocular manifestations, such as upward-gaze palsy,

~

bellar Syndrome

os~:~jnoli~oal ynd me

Tumour of Mesodermic Origin In this subgroup, we include two patients successfully operated on for meningioma affecting the pineal region. One was a 36-year-old female with seizures, and the other was a 60-year-old male with progressive loss of intellectual function. Both patients are symptomfree, after 6 and 4 years of follow-up.

.

0

.

.

.

.

.

I

i

I

I

20

40

60

80

1 O0

Percentage of Cases Fig. 3. Percentage of patients with germinoma showing the different clinical syndromes

26

J. Vaquero eta/.: Neurosurgical Experience with Tumours of the Pineal Region at Clinica Puerta de Hierro

light accomodation palsy, convergence nystagmus, complete Parinaud's syndrome, or oculomotor palsy, not interpreted as secondary to intracranial hypertension, was recorded in 68% of the germinomas; 80% of the non-germinoma germ-cell tumours, 28% of the pineal parenchymal tumurs, 20% of the gliomas, and 20% of the non-tumoural lesions. An endocrinologieal syndrome was recorded in 26% of the germinomas, 40% of the non-germinoma germcell tumours, 20% of the gliomas, and 10% of the nontumoural lesions. This syndrome was defined by the presence of one or more of the following manifestations: precocious puberty (3 germinomas, 1 embryonal carcinoma), diabetes insipidus (4 germinomas, 1 glioblastoma), panhypopituitarism (1 germinoma),

0

20

40

60

80

100

Percentage of Cases Fig. 6. Percentage of patients with gliomas showing the different clinical syndromes

~

ndrome

ilii~:E:n:d~

~176 g::!c:a:~:,S Y:nd!~

llar Syndrome

al Syndrome

iiiiiiiii ;;im'XCo;c' " ; ;o;;iiiii ii',N i iiiiNiiiii!i!i iiii#~iiii~ii~iiii~ii~iiiiiiii~ii~ii~ii~i~N~iiii~ii~iiiiiiiiii~iii~i~ii~iiiiiii~iiii~iii~iiiiiiii~ii~i~iiii~ ~il

0

20

40

60

80

N:~i%!i~i ;;i~:i!;::;ii i i!~i~i N.:i~i~i 100

Percentage of cases

L

I~i;il Intracranial Hypertens on Syndrome ~i~ii~il;;~ii!~iiiiii~i~!i [g"

0

Fig. 4. Percentage of patients with non-germinoma germ-cell tumours showing the different clinical syndromes

2;

,

,

40

60

J

80

Percentage of Cases Fig. 7. Percentage of patients with non-tumoural lesion showing the different clinical syndromes

Cerebellar Syndrome

I

iNiNNiiiiiiii

hypogonadism (4 germinomas, 1 cavernous angioma), or isolated GH deficiency (1 germinoma). Occasionally, two or more endocrinological dysfunctions occurred in the same patient. A cerebellar syndrome characterized by ataxia or dysmetria was recorded in 11% of the germinomas, 20% of the non-germinoma germ-cell tumours, in 14% of the parenchymal pineal tumours, 60% of the gliomas, and 10% of the nontumoural lesions.

Syndrome

/

;g 7;;n;;i%;; eg;',o % ;o 'g:N!iiliiii i:iiii!iiiiiiii 20

40

Percentage

60

80

100

of C a s e s

Fig. 5. Percentage of patients with parenchymal pineal tumours showing the different clinical syndromes

Diagnostic Procedures Tumoural Markers Alpha-fetoprotein (AFP) and human chorionic gonadotropin (HCG) were determined in CSF in 17 cases

J. Vaquero etal.: Neurosurgical Experience with Turnours of the Pineal Region at Clinica Puerta de Hierro

of the series, in which the final diagnoses were germinoma (9 cases), embryonal carcinoma (1 case), benign teratoma (1 case), pineocytoma (2 cases), astrocytoma (2 cases) and glioblastoma (1 case). Positivity was found in only four cases. The first was a 7-yearold male with unverified tumour, undergoing precocious puberty and with elevated levels of H C G in CSF (400 mU/cc). This patient was initially presumptively diagnosed as having a choriocarcinoma, but his dramatic response to radiotherapy and the benign outcome (symptom-free and no sign of disease 7 years later) led us to consider this case in the present series as possible HCG-secreting germinoma. Nevertheless, the true histological nature of this tumour remains a subject of speculation. Another case, a 33-year-old male with diabetes insipidus and hypopituitarism, was found by CT-scan to have tumoural masses in the pineal and suprasellar regions, and the level of AFP in CSF was 320 ng/cc. His lesions disappeared rapidly with radiotherapy, and the patient is symptom-free and shows no evidence of disease 10 years later. On the basis of these data, a diagnosis of AFP-secreting germinoma has been established. The third case corresponded to a 32-year-old male, treated with chemotherapy and radiotherapy, whose CSF AFP value was 1,845 ng/cc. The patient died two years after diagnosis and a presumptive diagnosis of malignant teratoma with areas of endodermal sinus tumour was established. The fourth case was a 15-year old male with a highly vascularized tumour and AFP values in CSF of 17,769 ng/cc. The lesion disappeared with radiotherapy and, six years later, the patient is symptom-free and shows no signs of disease. Although in this case, as we reported earlier, the final diagnosis remains speculative, it was classified as a supposed embryonal carcinoma. CSF Cytology This was performed in 9 patients (6 germinomas, 1 benign teratoma, 1 malignant teratoma and 1 astrocytoma), but did not reveal positivity in any of these cases.

CT-Scan This investigation was carried out in 44 patients of the series (17 germinomas, 3 pineocytomas, 3 pineoblastomas, 3 astrocytomas, 3 arachnoid cysts, 3 cavernous angiomas, 2 glioblastomas, 2 meningiomas, 2

27

metastases, 2 glial cysts, 1 benign teratoma, 1 malignant teratoma, 1 cystic cysticerosis, and 1 arteriovenous malformation). All the cases showed moderate to significant hydrocephalus at the time of CT-scan. In the largest group of germinomas, the characteristic CT-scan pattern was that of a hyperdense mass enhanced after contract administration (100%) and frequently enveloping a rounded pineal calcification (40%). CT-scan in pineocytomas showed calcification and cystic features, with an isodense tumoural zone that became enhanced after contrast administration. In the CT-scan study of the 3 pineoblastomas of the series, 2 were found to be hyperdense and the third was isodense. All showed enhancement after contrast administration. Magnetic Resonance Imaging (MRI) This was performed in 6 of the most recent cases of the series (2 pineocytomas, 1 benign teratoma, 1 glial cyst, 1 astroeytoma and 1 metastatic tumour). The two cystic pineocytomas and the glial cyst showed hypointense signal on T1 and hyperintense signal on T2 weighted images. In the benign teratoma, MRI was useful in identifying areas of fatty tissue, blood an necrosis in the turnout.

Analysis of Treatment Shunting Procedures In 33 cases of the series (66%), a shunting procedure was performed to correct the hydrocephalus as a step prior to surgery of radiotherapy. Surgical Approaches and Related Mortality Surgery was performed in 29 cases (58%) of the series, either as an initial therapeutic attempt (23 cases) or after a lack of response to radiotherapy (6 cases). In 16 cases, a transcallosal approach was used, in 9 the approach was supracerebellar, and in 4, it was occipital trans-tentorial. Surgical mortality was 11% (1 cavernous angioma operated on by the supracerebellar approach and 2 germinomas with the transcallosal approach). The cases treated surgically, after radiotherapy had failed, were 2 pineocytomas, 1 astrocytoma, 1 glioblastoma, 1 cavernous angioma and l glial cyst. In the germinomas, the most frequent lesions in this series, surgery was the initial therapeutic strategy, excepting shunting procedures, in 6 cases (31.5%), with a surgical mortality of 2 cases (33%).

28

J. Vaquero etai.: NeurosurgicalExperiencewith Tumoursof the Pineal Regionat ClinicaPuerta de Hierro

Radiotherapy Thirty patients of the series received radiotherapy at on time or another. Because a radiosensitive turnout was suspected, a trial dose of 2,000 cGy was administered to 22 patients, 15 of whom showed a sensitive response (13 supposed germinomas, 1 unverified embryonal carcinoma and 1 supposed malignant teratoma). In these patients, radiotherapy was completed up to a total of 4,000 cGy holocranially, which was boosted to up to 5,000-5,500cGy over the tumoural region, with fractioned doses of 200 cGy/day, 5 days a week. Only 5 patients received prophylactic spinal radiotherapy (2 supposed germinomas, one of them AFPsecreting, 1 supposed embryonal carcinoma, and 2 pineoblastomas, one of whom developed arachnoid lumbar seeding two years after spinal radiation). In all the cases, the dose covering the spine was 1,000-1,200 cGy. Six of the seven patients that did not respond to trial radiotherapy underwent surgery. Their diagnosis, on the basis of pathological study, were pineocytoma in 2 cases and 1 case each of glioblastoma, cavernous angioma, glial cyst and astrocytoma. Radiotherapy was used as the only treatment, excepting shunting procedures, in 14 cases of the series (12 supposed germinomas, 1 unverified embryonal carcinoma and 1 unverified glioblastoma), and as adjuvant to surgical resection in 11 verified tumours (4 germinomas, 3 pineoblastomas, 2 astrocytomas, 1 glioblastoma and 1 malignant teratoma). Chemotherapy Only one subject in the series received chemotherapy with cisplatin, bleomycin and vinblastine, according to Einhorn's protocol 9. This case corresponds to the malignant AFP-secreting teratoma mentioned above, but the exact diagnosis of which could be considered to be in doubt. Patient Survival

Fifteen patients (30%) of the series died during follow-up. One subject, with a germinoma verified at necrospy, died of acute hydrocephalus prior to receiving any treatment. Another 2 with germinomas and 1 with cavernous angioma died of surgical complications, and the remaining 11 deaths were due to the patient's disease (2 metastatic tumours, 2 glioblastomas, 1 embryonal carcinoma, 2 malignant teratomas, 3 pineoblastomas and 1 malignant astrocytoma). In our series, except for the case in whom death was

due to acute hydrocephalus and the two patients who died after surgery, the remaining 16 germinoma patients are presently alive and symptom-free, after follow-up ranging between 2 and 20 years (mean: 8 years). All were treated with radiotherapy, as the only treatment (12 cases), or as adjuvant treatment after surgery (4 cases). The four pineocytomas of the present series, all surgically removed and without radiotherapy after surgery, are presently alive and symptom-free, after follow-up ranging between 3 and 7 years (mean: 4 years). Although the criteria for choosing surgery as the initial therapeutic approach (elective surgery) or only once the ineffectiveness of radiotherapy had been demonstrated (selective surgery) were not established until the final years of the series, and after having gathered experience in the interpretation of clinical and CT-scan data, it is now possible to analyze retrospectively the consequences, in terms of survival, of having chosen one therapeutic alternative or another. Thus, if we exclude the cases with obvious diagnosis (meningiomas, arachnoid cysts, arteriovenous malformations, metastasis in patients with known carcinoma) in whom radiotherapy was not considered to be indicated as an initial option, as well as the patient who died of acute hydrocephalus immediately after admission, and a supposed cavernous angioma where the patient refused treatment, the present series is reduced to 40 cases consisting of lesions which today would be interpreted as "probable germinomas" or of"uncertain diagnosis". Of these 40 cases, 19 were managed according to a criterion elective surgery, with a mortality of 10.5%, corresponding to 2 patients with germinoma (whose deaths may have been avoided if they had been treated according to the criterion of selective surgery), and 21 with criteria of selective surgery. In the group that underwent elective surgery, only 6 patients required operation after confirmation of a lack of response to radiotherapy, with 17% mortality, corresponding to the cavernous angioma involving the brain stem mentioned above. A comparative analysis of survival among the patients managed according to one criterion or another has not revealed significant differences (Fig. 8). Discussion Incidence

Over the last 20 years, some 5,000 cases of intracranial expansive processes have been studied in our hospital. Those of the pineal region represent approx-

J. Vaquero et al.: Neurosurgical Experience with Tumours of the Pineal Region at Clinica Puerta de Hierro

29

100

90

80'

A

................................................. ---._..

70-

n-'21

60

-g>

n: 19 50

40a} o.

30-

I'-"i-'-

20

SELECT/VE SURGERY

10

ELECTIVE SURGERY 0

I

12

2~4

Follow-up

I

I

I

36

48

60

(Months)

imately 10% of these lesions. Nevertheless, we feel that this percentage does not reflect the true incidence of the expansive processes of the pineal region in the Spanish population since Clinica Puerta de Hierro is a hospital to which patients are referred from other neurosurgical services around the country. For this reason, we have studied the incidence of these disorders in two other Madrid hospitals which do not receive patients from other centers, and in both have found that the relative frequency of these processes is approximately 0.7% of all intracranial expansive lesions. This agrees with the incidence reported for these tumours in other series studied in our country and with the majority of the reports found in the Occident 15' 27 On the other hand, it is obvious that the diagnosis of pineal region tumour has increased in recent years, owing mainly to the availability of CT-scan. In our series, between 1970 and 1974, only 5 patients were diagnosed; between 1975 and 1979, using CT-scan, there were 13; between 1980 and 1984, 16 and from 1985 to 1989, another 16. In our present series, 48 % of the cases were tumours of germ-cell origin. Germinomas represented 79% of germ-cell tumours and 38% of all lesions in the series, which agrees with other series reported in the literature7, 25, 28, 39, 48. In our geographic area, nontumoural lesions represented the second largest group of lesions (20%), followed by tumours considered to be in the vicinity of the pineal gland (meningiomas, gliomas and metastases) that constitute 18~ by the tumours of the pineal parenchyma (14%) and, finally, by non-germinoma germ-cell tumours (10%).

Fig. 8. Survival graphs of the 40 patients of this series who today would be diagnosed as "probable germinoma" or as "uncertain diagnosis" on the basis of clinical and radiological data. When compared the patients who were treated on criteria of "elective surgery" (19 cases) and the patients who were treated on criteria of"selective surgery" (21 cases) a significant difference was not found

Clinical Data Age." The ages of our patients ranged between 18 and 64 years (mean: 23 years) and 72% of the cases were within the three decades of life, which agrees with previous observations that the percentage of patients in the first three decades of life falls between 66.5% (30%) and 100% 4. Our experience also confirms that the germinomas appeared in adolescents and young adults7' ~6, ~8, 19, 32, 49

With respect to tumours of neuro-ectodermic origin (parenchymal pineal tumours and gliomas), our experience supports the fact that these tumours can appear at any age in life22' 3~, 33, a clinical argument favouring the integration of all these tumours into a single subgroup.

Sex." The analysis of our series confirms the predominance of males when pineal region tumours are considered overall. The predominance is much more pronounced if only germinomas are considered (89.4% in our series). Nevertheless, the fact that only two female patients were 4 and 11 years old supports the opinion that under 15 years of age, there is not a clear sex-related predominance in these tumours4~. Symptomatology: In the last 15 years of the series, with the availability of CT-scan, the mean interval between the onset of symptoms and diagnosis has been 3 months. This mean interval is half that of the patients assessed prior to CT-scan and that reported in other series34, 37 It is evident that intracranial hypertension syn-

30

J. Vaquero et aL: Neurosurgical Experience with Tumours of the Pineal Region at Clinica Puerta de Hierro

drome, related to hydrocephalus, predominates in our patients. In the present series, an ophthalmological syndrome is frequent in germ-cell tumours, and a cerebellar syndrome in gliomas, possibly because of the infiltration of the cerebellar peduncles. In our present experience, although an endocrinological syndrome is relatively frequent in germ-cell neoplasms, its presence can not be considered a significant indication of the nature of a pineal region turnout.

Diagnostic Procedures and Management Criteria In contrast to reports by other authors 8' 33, 43 CSF cytology has been of no diagnostic value in the management of the tumours dealt with in this series. Concerning the value of tumoural markers, our concept of their specificity has evolved in recent years. At the beginning of this study, we thought that the positivity of AFP and HCG would rule out germinoma and point to a diagnosis of malignant germ-cell tumour, with tissue of endodermal sinus tumour or choriocarcinoma, respectively2' 18, 24, 35, 36, 50. Nevertheless, the observations of typical germinomas with AFP and HCG secretion 3' 12, 17, 20, 24, 26, 52 and our experience with two unverified tumours with radiological data, response to radiotherapy and follow-up characteristics of typical germinomas, suggest that tumoural markers doubtfully can be considered specific for the establishment of the exact histological nature or the biological prognosis of a germ-cell tumour. Although stereotactic biopsy has been advocated as useful in the management of pineal region tumours 22, this procedure has not been employed in the present series. Disadvantages and risks related to stereotactic biopsy of pineal masses include: the difficulty of obtaining enough tissue in order to get an accurate histological diagnosis, considering that in germinal turnouts there appear frequently areas with different degrees of malignancy and variable histology; danger of injury to the surrounding venous structures; and eventually the possibility of a tumoural seeding after the procedure 4. Nevertheless, the actual image-directed techniques permit the accurate stereotactic biopsy of pineal tumours in a safer way. For this reason, we now consider this procedure as one of the possible diagnostic techniques for to establish a therapeutic decision. We agree with Jooma and Kendall 18 that nowadays CT-scan is very reliable in the initial assessment of a patient with a tumour of the pineal region, and when the pattern of a well-defined lesion, moderately hyperdense, with homogenous enhancement after con-

trast administration 1~ 11,14, 51, 53, is added to the clinical data, negativity in the tumoural markers, etc., and the patient is a young male, a diagnosis of "probable germinoma" can be established and the subject should be selected for trial radiotherapy. When CT-scan points to a well defined surgical lesion (benign teratoma, meningioma, benign astrocytoma, etc.), it is obvious that surgery should be considered the most logical initial therapeutic option. Nevertheless, when on the basis of CT-san and clinical evaluation, the diagnosis remains uncertain, which can happen in approximately 30% of cases, we think that the option of surgical verification is no better, in terms of prognosis and survival, than a therapeutic management consisting of the administration of trial radiotherapy, the so-called "biological biopsy ''15 or "diagnostic radiotherapy ''4~ and only when failure of radiotherapy is confirmed, should surgery be performed (Table 2). In our experience, this regimen of therapeutic management permits a high rate of survival and obviously is the most innocuous and effective mode of treatment of the germinomas, which contitutes the most numerous group of tumours. Although it is true that the present surgical techniques afford an approach to tumours of the pineal region with acceptables rates of morbidity and mortalityS, 13,26, the good results can not be generalized and are obviously related to an experience and an ability not necessarily within reach of all the neurosurgeons who have to face the management of tumours of this region. Thus, we feel that before proposing surgery in

Table 2. Diagram Showing our Present Management of Pineal Region lesions

I CLINICAL AND CT-SCAN EVALUATION

Probable germinoma

1

Probable surgical lesion (rneningiema, glioma, benign teratoma, pineocytoma, cyst)

Uncertain diagnosis

I RADIOTHERAPY

Response

RADIOTHERAPY

No Response

F IRAD,OTHERA

Response

?

\

[RAD,OHERPYi

J. Vaquero et al.: Neurosurgical Experience with Tumours of the Pineal Region at Clinica Puerta de Hierro these lesions, it m a y be useful to remember a phrase expressed by Davidoff6: " . . . I t is better to have a living

patient, restored to good health, whose diagnosis lacks the stamp of certainty the microscope provides, than it is to have beautiful microphotographs from postmortem specimens". It is obvious that with the therapeutic regimen that we propose there could be diagnostic errors, since some tumours, such as pineoblastomas, which initially respond well to radiotherapy, can be misdiagnosed as germinomas, as also can occur with some non-germin o m a germ-cell t u m o u r s showing partial response to radiotherapy. Nevertheless, we consider that these diagnostic errors do not lead to significant therapeutic errors since histological verification in these cases does not substantially m o d i f y the indication for radiotherapy, the results o f which do not appear to be clearly enhanced by previous surgical resection 29. Furthermore, the a r g u m e n t that surgery o f pineal region t u m o u r s is hindered by prior radiotherapy 2I' 23, 38 does n o t agree with our own experience nor with that o f other authors. This measure has even come to be r e c o m m e n d e d prior to surgery to reduce the tumoural vascularization and facilitate surgical resection 1, 43 With respect to CT-scan evaluation and management o f pineocytomas, their cystic nature and calcifications appear to be data to be taken into a c c o u n t in our cases; all o f them show signs o f neuronal differentiation in the pathological study. The supposed benign behaviour o f pineocytomas with neuronal differentiation 31 correlates with our experience, and suggests that postoperative radiotherapy is not indicated in these t u m o u r s 47. O u r study shows that, in selected cases o f pineal region tumours, radiotherapy alone can achieve a high rate o f cures, and support the opinion that surgical resection in malignant t u m o u r s does not enhance the benefits obtained with irradiation 5' 16. In our series, all patients with supposed germinomas subjected to radiotherapy are alive and symptom-free, with a followup that ranged between 2 and 20 years (mean: 8 years). On the other hand, we consider that prophylactic spinal irradiation is not necessary in radiosensitive tum o u r s o f the pineal region, in agreement with the opinion o f other authors 5 and with reported data that in unoperated patients with pineal germ-cell tumour, spinal seeding is relatively u n c o m m o n 32. In our series, only 5 patients received radiotherapy o f the spine, and one o f these subjects was the only case (a pineoblastoma) that developed spinal seeding two years later.

31

References 1. Abay II EO, Laws ER, Grado GL, Bruckman JE, Forbes GS, Grmez MR, Scott M (1981) Pineal tumours in children and adolescents. J Neurosurg 55:889-895 2. Allen JS, Nisselbaum J, Epstein F, Rosen G, Schwartz MK (1979) Alphafetoprotein and human chorionic gonadotropin determination in cerebrospinal fluid. An aid to the diagnosis and management of intracranial germ cell tumours. J Neurosurg 51: 368-374 3. Arita N, Bitoh S, Ushio Y, Hayakawa T, Hasegawa H, Fujiwara M, Ozaki K, Parkhen L, Mori T (1980) Primary endodermal sinus tumour with elevated serum and CSF alphafetoprotein levels. J Neurosurg 53:244-248 4. Chang CG, Kageyama N, Kobayashi T, Yoshida J, Negoro M (1981) Pineal tumours: clinical diagnosis, with special emphasis on the significance of pineal calcification. Neurosurgery 8: 656667 5. Danoff B, Sheline GE (1984) Radiotherapy of pineal tumours. In: Neuwelt EA (ed) Diagnosis and treatment of pineal region turnouts. Williams and Wilkins, Baltimore 6. DavidoffLM (1967) Some considerations in the therapy ofpineal tumours. Rudolf Virchow lecture. Bull N Y Acad Med 43: 537561 7. De Girolami U, Schmidek H (1973) Clinicopathological study of 53 tumours of the pineal region. J Neurosurg 39:455-462 8. Edwards MSB, Hudgins RJ, Wilson CHB, Levin VA, Wara WM (1988) Pineal region tumours in children. J Neurosurg 68: 689-697 9. Einhorn LH, Donohue JP (1977) Cis-diamine dichloroplatinum, vinblastine and Neomycin combination chemotherapy in disseminated testicular cancer. Ann Intern Med 87:293-298 10. Futrell NN, Osborn AG, Cheson BD (1981) Pineal region tumours:'computed tomographic-pathologic spectrum. AJR 137: 951-956 11. Ganti SR, Hilal SK, Stein BM, Silver AJ, Mawad M, Sane P (1986) CT of pineal region tumours. AJR 146:451-458 12. Handa H, Yamashita J (1981) Current treatment of pineal tumours. Neurol Med Chir (Tokio) 21:147-154 13. Hoffman HJ (1984) Transcallosal approach to pineal tumours and the Hospital for sick children series of pineal region tumours. In: Neuwelt EA (ed) Diagnosis and treatment of pineal region tumours. Williams and Wilkins, Baltimore 14. Inoue Y, Takeuchi T, Tamaki M, Nin K, Hakuba A, Nishimura S (1979) Sequential CT observations of irradiated intracranial germinomas. AJR 132:361-365 15. Isamat F (1979) Tumours of the posterior part of the third ventricle: Neurosurgical criteria.: In: Advances and technical standards in neurosurgery, Vol 6. Springer, Wien New York 16. Jenkin RDT, Simpson WJK, Keen CW (1978) Pineal and suprasellar germinomas. J Neurosurg 48:99-107 17. Jennings MT, Gelman R, Hochberg F (1984) Intracranial germcell tumours: natural history and pathogenesis. In: Neuwelt EA (ed) Diagnosis and treatment of pineal region tumours. Williams and Willdns, Baltimore 18. Jooma R, Kendall BE (1983) Diagnosis and management of pineal tumours. J Neurosurg 58:654-665 19. Koide O, Watanabe Y, Sato (1980) A pathologic survey of intracranial germinoma and pinealoma in Japan. Cancer 5: 2119-2130

32

J. Vaquero el al.: Neurosurgical Experience with Tumours of the Pineal Region at Clinica Puerta de Hierro

20. Kubo O, Yamashi N, Kamijo Y, Amano K, Kitamura K, De~ mura R (1977) Human chorionic gonadotropin produced by ectopic pinealoma in a girl with precocious puberty. J Neurosurg 7:101-105 21. Lapras C (1984) Surgical therapy of pineal region tumours. In: Neuwelt EA (ed) Diagnosis and treatment of pineal region tumours. Williams and Wilkins, Baltimore 22. Moser RP, Backlund EO (1984) Stereotactic techniques in the diagnosis and treatment of pineal region tumours. In: Neuwelt EA (ed) Williams and Wilkins, Baltimore 23. Neuwelt EA, Batjer HH (1984) Pre and postoperative management of pineal region turnouts and the occipital transtentorial approach. In: Neuwelt EA (ed) Diagnosis and treatment of the pineal region tumours. Williams and Wilkins, Baltimore 24. Neuwelt EA, Frenkel EP (1984) Germinoma and other pineal tumours: chemotherapeutic responses. In: Neuwelt EA (ed) Diagnosis and treatment of pineal region turnouts. Williams and Wilkins, Baltimore 25. Neuwelt EA, Glasberg M, Frenkel E, Clark WK (1979) Malignant pineal region tumours. J Neurosurg 51:597-607 26. Neuwelt EA (ed) (1984) Diagnosis and treatment of pineal region tumours. Williams and Wilkins, Baltimore 27. Obrador S, Soto M, Gutierrez-Diaz JA (1976) Surgical management of tumours of the pineal region. Acta Neurochir (Wien) 34:159-171 28. Onoyama Y, Ono K, Nakajima T, Hiraoka M, Abe M (1979) Radiation therapy of pineal tumours. Radiology 130:757-760 29. Pluchino F, Broggi G, Fornari M, Franzini A, Solero CL, A1legranza A (1989) Surgical approach to pineal turnouts. Acta Neurochir (Wien) 96:26-31 30. Poppen JL, Marino R (1968) Pinealomas and tumours of the posterior portion of the third ventricle. J Neurosurg 28: 357364 31. Rubinstein JL (1982) Tumours of the central nervous system (Suppl). Atlas of tumour pathology. Armed Forces Institute of Pathology, Washington DC, pp 15-20 32. Sano K, Matsutani M (1981) Pinealoma (germinoma) treated by direct surgery and postoperative irradiation. A long-term follow-up. Child's Brain 8:81-97 33. Sano K (1976) Diagnosis and treatment of tumours in the pineal region. Acta Neurochir (Wien) 34:153-157 34. Schmidek HH (1977) Pineal tumours. Masson, New York 35. Shinoda J, Miwa Y, Sakai N, Yamada H, Shima H, Kato K, Takahashi M, Shimokawa K (1985) Immunohistochemical study of placental alkaline phosphatase in primary intracranial germ-cell tumours. J Neurosurg 63:733-739 36. Shinomiya Y, Toya S, Iwata T, Hosada Y (1979) Radioimmunoassay of alpha-fetoprotein in children with primary intracranial tumour. Childs Brain 5:450-458 37. Smith NJ, El-Mahdi AM, Constable WC (1976) Results of irradiation of tumours in the region of the pineal body. Acta Radiol Ther Phys Biol 25:17-22

38. Stein BM (1984) Surgical therapy of benign pineal tumours. In: Neuwelt EA (ed) Diagnosis and treatment of pineal region tumours. Williams and Wilkins, Baltimore 39. Sung DI, Harisiadis L, Chang CH (1978) Midline pineal tumours and suprasellar germinomas: highly curable by irradiation. Radiology 128:745-751 40. Takakura K(1984)Intracranial germ celltumours. Clinical neurosurgery. Proceedings of the Congress of Neurological Surgeons, New York N Y. Williams and Wilkins, Baltimore 41. Takakura K (1984) Nonsurgical pineal tumour therapy. The japanese experience. In: Neuwelt EA (ed) Diagnosis and treatment of pineal region tumours. Williams and Wilkins, Baltimore 42. Teilum G (1965) Classification of endodermal sinus tumour (mesoblastoma vitellinum) and so-called embryonal carcinoma of the ovary. APMIS 64:407-429 43. Ueki T, Tanaka R (1980) Treatment of prognoses of pineal tumours. Experience of 1I0 cases. Neurol Med Chir (Tokyo)20: 1-26 44. Vaquero J, Carrillo R, Cabezudo J, Leunda G, Villoria F, Bravo G (1980) Cavernous angiomas of the pineal region. Report of two cases. J Neurosurg 53:833-835 45. Vaquero J, Carrillo R, Cabezudo JM, Nombela L, Bravo G (1981) Arachnoid cysts of the posterior fossa. Surg Neurol 16: 117-121 46. Vaquero J, Martinez R, Escand6n J, Bravo G (1988) Symptomatic glial cysts of the pineal gland. Surg Neurol 30:468-470 47. Vaquero J, Ramiro J, Martinez R, Coca S, Bravo G (1990) Clinicopathological experience with pineocytomas: report of five surgically treated cases. Neurosurgery. 27:612-619 48. Ventureyra ECG (1981) Pineal region. Surgical management of tumours and vascular malformations. Surg Neurol 16:77-84 49. Waga S, Handa H, Yamashita J (1979) Intracranial germinomas: treatment and results. Surg Neurol 11: 167-172 50. Wilson ER, Takei K, Bikoff WT, O'Briens MS, Tindall GT, Boehm WM (1979) Abdominal metastase of primary intraeranial yolk sac tumours through ventriculoperitoneal shunts: report of three cases. Neurosurgery 5:356-364 51. Wood JH, Zimmerman RA, Bruce DA, Bilaniuk LT, Norris DG, Schut L (1981) Assessment and management of pinealregion and related turnouts. Surg Neurol 16:192-210 52. Yamagami T, Handa H, Yamashita J, Okumura T, Paine J, Hachara H, Furukawa F (1987) An immunohistochemical study of intracranial germ cell tumours. Acta Neurochir (Wien) 86: 33-41 53. Zimmerman RA, Bilaniuk LT, Wood JH, Bruce DA, Schut L (1980) Computed tomography of pineal, parapineal and histologically related tumours. Radiology 137:669-677

Correspondence and Reprints: Dr. Jesfls Vaquero Crespo, Servicio de Neurocirugia, Clinica Puerto de Hierro, San Martin de Porres, 4, 28035-Madrid, Spain.