Analysis of Secondary Empty Sella Syndrome Due to Brain Tumors by Brain MRI: A Preliminary Report

Abstract

We experienced a case of delayed postoperative CSF rhinorrhea after the removal of a huge falx meningioma by the parieto-occipital interhemispheric approach. Retrograde review of the MRI revealed empty sella and subsequent Isovist CT cisternography disclosed the site of leakage, which was managed by a trans-sphenoidal approach. In order to avoid such cases we have made an analysis of brain MRIs to assist in detecting secondary empty sella syndromes. The authors retrospectively reviewed the preoperative MRIs of 53 cases of large brain tumors regardless of pathology, exempting pituitary tumors. Retrospective analysis of empty sella and review of brain MRI was done. The pituitary fossa was enlarged and filled partially or fully with CSF. Empty sella was evident in 60.3% of cases (32/53) and negative in 39.7% of cases (21/53). The incidence of secondary empty sella by pathology was as follows: (1) 20 out of 24 cases (83.3%) in meningiomas excluding 4 cases where the tumor had invaded the pituitary fossa (out of total 28 cases); (2) 55.6% (5/9 cases) in gliomas; (3) 0 cases out of 1 case in ganglioma; (4) 2/3 cases in hemangioblastomas; (5) 1/3 cases in acoustic schwannoma; (6) 2/2 cases in NF2; (7) 1/1 case of hemangiopericytoma; (8) 0/1 case lymphoma; (9) 1/2 cases lymphoma; (10) 0/1 case of PNET; (11) 0/1 case in AT & RT. Increased ICP attributed to long-standing, relatively large or huge benign tumor, causes compression of the CSF pathway, resulting in the herniation of the basal cisternal arachnoid membrane through a weakening in the diaphragm sellae into the pituitary fossa. This is generally regarded as the pathogenesis of secondary empty sella syndrome. The analysis showed that relatively slow-growing meningiomas with the exception of cavernous sinus, tuberculum sella, and clinoidal meningioma that directly invaded the pituitary fossa had a high incidence of secondary empty sella (83.3%), while rapid-growing glial tumors whose pathomechanism is invasion rather than ICP elevation showed lower incidence (55.6%). The cases of other pathologies were too few to analyze their respective incidences. Careful review of brain MRI with periodic follow-up is necessary for the detection of secondary empty sella in patients with brain tumor. In patients with confirmed empty sella, follow-up is mandatory because management of hypopituitarism, CSF rhinorrhea, visual disturbance, and increased ICP is necessary. Further accumulation of cases and analysis is required.