Cavernous sinus invasion by pituitary macroadenomas
A. Cukiert, M. Andrioli, J. Goldman, M. Nery, L. Salgado, M. Knoepfelmacher and B. Liberman
Departments of Endocrinology and Neurosurgery, Hospital Brigadeiro, Sao Paulo SP, Brazil.
Cavernous sinus invasion by pituitary adenomas has been noted since the early days of pituitary surgery (1). On the other hand, the pre-operative diagnosis of cavernous sinus invasion was rarely done before the introduction of CT scanning in the clinical practice and correctly performed only in the more obvious cases by means of CT (2). The availability of high resolution MR imaging (3) has made it possible to study the sellar region in a detail not possible using CT scans (4).
Different criteria have been suggested to standardize the neuroradiological diagnosis of cavernous sinus invasion by pituitary adenomas (5,6). Unfortunately, the pituitary / medial wall of the cavernous sinus interface is poorly visualized even with MRI and that explains why no definite criteria have been accepted so far.
The prevalence of radiological and surgical findings compatible with cavernous sinus invasion has been increasing over the last decades. This plays an important role while treating these patients since cavernous sinus invasion is usually related to uncomplete surgical removal of the tumour, poor endocrinological results and the need for adjuvant therapy, especially the different types of radiation therapy.
This study reports the MRI findings in patients with pituitary macroadenomas and try to establish guidelines for the neuroradiological pre-operative diagnosis of cavernous sinus invasion in such patients.
Seventeen patients with pituitary macroadenomas (2 prolactinomas, 4 with Cushing’s disease, 7 with acromegaly and 4 with non-secreting tumors) that were submitted to transesphenoidal tumour resection and had intraoperative confirmation of cavernous sinus invasion were included in this study. Intraoperative diagnosis of cavernous sinus invasion was based on the visual inspection of this structure and clear signs of invasion. Discoloured or rough medial walls were not rated as “invasion”.
Pre-operative MRI scanning consisted of multiple T1 coronal slices through the sellar region with and without the administration of paramagnetic contrast. Post-contrast thin T1 axial slices of the sellar region were also obtained. A sagittal sequence was acquired for diagnostic purposes but was not used to define cavernous sinus invasion.
Different MR features were especifically studied in coronal (presence of tumor lateral to the carotid artery, within the carotid syphon and loss of the ring enhancement of the medial wall of the cavernous sinus) and axial (oppening of the double posterior leaflets of the cavernous sinus) slices.
In 8 patients, MRI disclosed tumor lateral to the carotid artery; 13 patients had tumor located within the carotid syphon and all had loss of the ring enhancement of the medial wall of the cavernous sinus as defined by MRI . In 10 patients, opening of the double posterior leaflets of the cavernous sinus could be seen in axial slices.
Only 1 (GH-secreting) of the 13 patients with secreting tumors with cavernous sinus invasion was cured by surgery alone. This patient had a progressive decline in GH levels over the first week and probably had a subacute necrosis of intracavernous tumour. All patients with secreting tumours not cured by surgery alone (n=12) were submitted to post-operative radiation therapy.
The presence of tumor lateral to the carotid artery in patients with pituitary adenomas has been considered the classic gold-standard for the pre-operative radiological diagnosis of cavernous sinus invasion (7). It is present in the bigger tumours (8,9). On the other hand, it is conceivable that if tumour can be often seen lateral to the carotid artery, that it could have been seen medial to it or within the carotid syphon in an earlier phase. Actually, the challange today is to define guidelines to identify early invasion of the cavernous sinus by pituitary tumors.
We have studied patients with pituitary macroadenomas. In these patients, lack of the ring enhancement of the medial wall of the cavernous sinus seems to be the first sign of cavernous sinus invasion and was seen in all patients. Tumor within the carotid syphon was also seen in the majority of the cases and seems to be related to a more advanced stage of invasiveness. The posterior leaflets of the cavernous sinus were also frequently widened, as could be seen in axial slices. Axial slices are not routinely acquired in MR imaging of the sellar region but it should be included in the protocol if cavernous sinus involvement is suspected. The prevalence of the reported cavernous sinus invasion by pituitary tumours have risen in recent years. It is probably even higher then the presently reported (10). It should be noted that microscopic dural invasion can be seen in patients in whom the surgeon reported no intraoperative invasiveness (11).
This study did not include microadenomas but patients with microadenomas can also have cavernous sinus invasion (12). This situation might correlate to the 10% failure rate for endocrinological cure in patients with microadenomas. We are presently carrying out the evaluation of the meaning of early signs of cavernous sinus invasion as defined by MRI in the prognosis of patients with microadenomas.
Interestingly, the only (acromegalic) patient with cavernous sinus invasion cured by surgery had a sub-acute cure which is quite different from what is seen conventionally in patients cured by surgery, especially in tumours which secrete hormones with a very small half-life ( e.g., GH). We postulated that this patient has had a progressive post-manipulation necrosis of the intracavernous portion of the tumour and achieved cure.
The multiple radiological patterns of cavernous sinus involvement requires an individualized analysis of each patient. Imaginary lines and compartments are not useful to standardize this diagnosis (13).
Only 3 patients disclosing the presently analyzed MRI criteria for cavernous sinus (CS) invasion and with intraoperative findings not confirming CS invasion were found in a series of 250 patients. Taken together with the data from this study, it seems that the analised MR criteria in macroadenomas could provide useful guidelines for the pre-operative diagnosis of cavernous sinus invasion and patients’ counselling. How these features apply to microadenomas has yet to be seen.
The endocrinological results obtained in patients with cavernous sinus invasion (only 7% of remission) by macroadenomas are in marked contrast to those obtained in non-invasive adenomas. Our data showed that endocrine remission can be expected in 90% of non-invasive microadenoma and in 70% of non-invasive macroadenomas.
- Martins AN, Hayes GJ and Kempe LG. Invasive pituitary adenomas. J Neurosurg 1965;22:268-276.
- Nichols DA, Laws ER, Houser OW and Abboud CF. Comparison of MRI and CT in the preoperative evaluation of pituitary adenomas. Neurosurgery 1988;22:380-385.
- Daniels DL, Czervionke LF, Bonneville JF, Cattin F, Mark LP, Pech P, Hendrix LE, Smith DF, Haughton VM and Williams AL. MR imaging of the cavernous sinus: value of spin-echo and gradient recalled echo images. A J Roentgenol 1988;151:1009-1014.
- Davis PC, Hoffman JC, Spencer T, Tindall GT and Braun IF. MR imaging of pituitary adenoma: CT, clinical and surgical correlation. A J Roentgenol 1987;148:797-802.
- Komiyama M. MRI of the cavernous sinus. Radiat Med 1990;8:136-144.
- Komiyama M, Hakuba A, Yasui T, Yagura H, Fu Y, Baba M and Nishimura S. MRI of intracavernous pathology. Neurol Med Chir Tokyo 1989;29:573-578
- Ahmadi J, North CM, Segall HD, Zee CS and Weiss MH. Cavernous sinus invasion by pituitary adenomas. A J Roentgenol 1986;146:257-262.
- Kaufman B, Kaufman BA, Arafah BM, Roessmann U and Selman WR. Large pituitary gland adenomas evaluated with MRI. Neurosurgery 1987;21:540-546.
- Scotti G, Yu CY, Dillon WP, Norman D, Colombo N, Newton TH, Groot J and Wilson CB. MR imaging of cavernous sinus involvement by pituitary adenomas. A J Roentgenol 1988;151:799-806.
- Selman WR, Laws E, Scheithauer E and Carpenter S. The occurrence of dural invasion in pituitary adenomas. J Neurosurgery 1986;64:402-407.
- Daita G, Yonemasu Y, Nakai H, Takei H and Ogawa K. Cavernous sinus invasion by pituitary adenomas-relationship between MRI findings and histologically verified dural invasion. Neurol Med Chir Tokyo 1995;35:17-21.
- Hirohata T, Uozumi T, Mukada K, Arita K, Kurisu K, Sumida M, Yano T, Takechi A and Onda J. Microprolactinoma invading the cavernous sinus. Report of 3 cases. Neurol Med Chir Tokyo 1992;32:10-12.
- Knosp E, Steiner E, Kitz K and Matula C. Pituitary adenomas with invasion of the cavernous sinus space: a magnetic resonance imaging classsification compared to surgical findings. Neurosurgery 1993;33:610-618.