Transsphenoidal treatment of non-neoplastic intrasellar cysts

Transcription

1 J Neurosurg 60:8-13, 1984 Transsphenoidal treatment of non-neoplastic intrasellar cysts A report of 38 cases DAVID S. BASKIN, M.D., AND CHARLES B. WILSON, M.D. Department of Neurological Surgery, University of Cal~rnia School of Medicine, San Francisco, Calffbrnia ~," Thirty-eight patients underwent transsphenoidal microsurgical treatment of non-neoplastic intrasellar cysts: 36 had cyst drainage and biopsy of the cyst wall, and in two the cyst was totally removed. Surgical morbidity was 8%. The mean follow-up time was 46.3 months; 100% patient follow-up evaluation was achieved. Sixteen female patients (mean age 24.6 years) had pars intermedia cysts: 88% had menstrual irregularities, 63% had galactorrhea, 31% had headache, and 56% had hyperprolactinemia. Within these groups, menstrual cycles returned in 86%, galactorrhea ceased in 90%, headaches resolved in 80%, and serum prolactin levels were restored to normal in 66%. Eight females and three males had Rathke s cleft cysts (mean age 34.0 years): of these 11 patients, 91% had headaches and 18% had hyperprolactinemia; of the eight females, 63% had amenorrhea and 63% had galactorrhea. Within these groups, serum prolactin levels normalized in 50%, and 80% noted reduced headache. Of the females, 80% had return of menses and 50% noted cessation of galactorrhea. Six males and two females had arachnoid cysts (mean age 42.2 years): 50% had headaches: 50% were asymptomatic. Preoperatively, 50% of these patients had hypothyroidism and 25% had adrenal hypofunction. Postoperatively, 75% of patients with headache noted improvement, and 33% of patients with abnormal thyroid function had normal function. Adrenal function did not improve. Three patients had an intrasellar cysticercosis cyst, epidermoid cyst, and postoperative cyst, respectively. All had evidence of partial hypopituitarism: none improved postoperatively. The results indicate that different types of pituitary cysts produce different clinical syndromes, and suggest that simple transsphenoidal drainage and partial removal of the cyst wall can provide safe and effective therapy. KEY WORDS arachnoid cyst non-neoplastic intrasellar cyst pituitary cyst pars intermedia cyst Rathke s cleft cyst 9 transsphenoidal microsurgery C YSTS within the pituitary fossa are a common observation in anatomical studies. Shanklin ~9 noted 13 cases of pituitary cysts in an examination of 100 human pituitary glands from consecutive patients ranging in age from newborn to 86 years. McGrath ~4 noted an even higher incidence: 33% of 83 necropsy specimens had cysts within the sellar hypophysis. In the clinical situation, however, distinguishing patients with a non-neoplastic intrasellar cyst from those who have nonsecreting pituitary adenomas or craniopharyngiomas is difficult because the symptoms, signs, and radiological characteristics of these cysts are similar. As adenomas and craniopharyngiomas are more common than cysts, the cyst is often misdiagnosed preoperatively. Previous clinical reports concerned with the evaluation and treatment of intrasellar cysts consist of between one and five patients, 4-~~176 most of whom were treated by craniotomy for removal of a presumed adenoma. It is difficult to compare one report with another because of the various terminologies, a paucity of detailed preoperative and postoperative endocrinological data, and, in most cases, the lack of consistent and detailed long-term follow-up evaluations. Excluding craniopharyngiomas and cystic pituitary adenomas, we analyzed the preoperative and postoperative data from 38 patients undergoing transsphenoidal surgery for the treatment of intrasellar cysts at the University of California, San Francisco (UCSF), during 8 J. Neurosurg. / Volume 60 / January, 1984

2 Transsphenoidal treatment of intrasellar cysts the past 8 years. In this review we have attempted to clearly delineate the types of cysts that were found, the preoperative symptoms and signs, and the kind and degree of improvement that can be expected postoperatively. Clinical Material and Methods Between January, 1974, and December, 1982, 38 patients underwent transsphenoidal microsurgical treatment of non-neoplastic intrasellar cysts at UCSF. Thirty-six patients had simple drainage of the cyst and biopsy of the cyst wall; two had total removal of the cyst. Excluded from consideration were patients with cystic neoplasms, including craniopharyngiomas and pituitary adenomas, and any patient with a partially or totally empty sella, as demonstrated by pneumoencephalography (PEG), computerized tomography (CT) scanning, or metrizamide CT cisternography. Clinical Evaluation All patients were interviewed to obtain a detailed history of their illness, and complete physical and neurological examinations were performed. Specific notations were made regarding the presence or absence of headache, loss of vision, menstrual irregularities, diabetes insipidus, galactorrhea, hypothyroidism, impotency, infertility, and other symptoms and signs of endocrine disease. Endocrinological Evaluation Neuroendocrinological examinations of anterior pituitary function were performed in all patients preoperatively, approximately 6 weeks postoperatively, and at varying times thereafter. These included determinations of the following hormone levels: serum prolactin, growth hormone (GH), follicle-stimulating hormone (FSH), luteinizing hormone (LH), and in selected patients, thyroid-stimulating hormone (TSH). and adrenocorticotrophic hormone (ACTH). All patients also underwent thyroid function tests, including determinations of serum thyroxine levels and triiodothyronine (T3) uptake, and tests of adrenal function that included morning and evening serum cortisol levels. Neuroradiological Evaluation Neuroradiological studies included hypocycloidal sellar polytomography in the anteroposterior and lateral planes, polytomographic PEG, CT, and in some cases, selective carotid angiography. Over the last 3 years of this study, magnified high-resolution CT reconstructions of the sellar and suprasellar regions in the coronal and sagittal planes were obtained instead of polytomographic PEG s, because of the greater precision afforded by this newly available CT procedure. Carotid angiography was performed routinely during the first 4 years of the study, but thereafter it was restricted to patients in whom a parasellar aneurysm was suspected. All cysts examined by CT scanning had smooth borders, an upward convexity, and homogeneous low-density contents with attenuation characteristics resembling cerebrospinal fluid (CSF). Surgical Procedures All operations were performed by one of us (C.B.W.). In all but two cases, the cyst cavity was identified and drained, and the cyst wall was biopsed if it was readily accessible. In the remaining two cases (a cysticercosis and an epidermoid cyst), the lesions were totally removed. Saline was then injected into a previously placed indwelling subarachnoid catheter to determine if there was communication with the subarachnoid space. The cyst cavity was packed with fat and the floor of the sella was reconstructed with nasal cartilage. Postoperative Status Most of the patients were able to have a clear liquid diet and sit up in bed on the day of operation. By the 1st day postoperatively, all were ambulatory and able to eat regular meals. Patients were usually discharged on the 5th to 7th postoperative day, and most of them were able to resume normal activity within 4 weeks. Follow- Up Review All patients underwent detailed endocrinological reevaluation 6 weeks after the operation and at various times thereafter. From September to December, 1982, all patients and their referring physicians were contacted and interviewed by telephone. All available information regarding preoperative symptoms and signs, preoperative and postoperative endocrine testing, and any possible complications was obtained. Classification of Cysts Patients were divided into four groups, distinguished on the basis of intraoperative observations made through the surgical microscope and histological analysis of the wall of the cysts (Table 1). Small cysts containing clear fluid that were located between the anterior and posterior lobe were placed in the category of "pars intermedia cysts." As observed through the surgical microscope, these cysts had no communication with the subarachnoid space: once drained, they did not refill with fluid. In the biopsy specimens obtained, the wall of the cyst cavity was made up of only fibrous tissue. "Rathke s cleft cysts" were defined by the presence of either yellowish fluid containing crystals or thick mucoid material in the cyst cavity. The biopsied cyst wall was composed of cuboidal, columnar, or pseudostratified columnar epithelium. "Arachnoid cysts" were distinguished by the surgical observation of a clear colorless fluid in a cavity that, in most cases, had a pinhole communication with the subarachnoid space. The communication was demonstrated by injection of sterile saline into a previously placed lumbar subarach- J. Neurosurg. / Volume 60 / January,

3 D. S. Baskin and C. B. Wilson TABLE 1 Classification of non-neoplastic intrasellar cysts Communi- Cyst Type Size Location Description of Cyst Fluid cation* Histology pars intermedia small (2-3 mm) Rathke s cleft moderate (3-10 mm) arachnoid moderate (3-10 mm) miscellaneous large (10-20 mm) cysts * Communication with subarachnoid space. intermediate lobe clear fluid none intermediate lobe yellowish fluid on mucoid mate- none rial; may contain crystals pars distalis or subarachnoid clear fluid yes space pituitary parenchyma variable, depending on cyst type none fibrous tissue capsule cuboidal, columnar, or pseudostratified columnar epithelium fibrous tissue capsule variable, depending on cyst type noid catheter that caused the cavity to refill very slowly with clear colorless fluid. These cysts were not located in the intermediate lobe, but rather in the pars distalis or immediately beneath the dura, outside of the pituitary parenchyma. Arachnoid cysts did not fill with air during PEG, as evidenced on all projections, including the hanging-head views; and the intracranial surface of the cyst typically had a convex contour. The biopsied cyst wall showed only fibrous tissue. The category of "miscellaneous cysts" included cysticercosis and epidermoid cysts that were diagnosed by the usual histological criteria, as well as the symptomatic intrasellar cyst (which we termed "posttherapy cyst") that developed after craniotomy and irradiation of a pituitary adenoma. Results Complete follow-up data were obtained for all patients in all groups. The results of this study were analyzed with two objectives. The first was to examine the patient profile in each of the groups in order to compare and contrast age and sex distribution and preoperative symptoms and signs (Table 2). The second objective was to examine the efficacy of therapy with respect to each preoperative abnormality in each group (Table 3). Pars Interrnedia Cysts Sixteen patients, all female, had pars intermedia cysts. The mean age was 24.6 years (range 18 to 35 years). The mean period of follow-up evaluation was 45.6 months. Most of these patients presented with menstrual irregularities and/or galactorrhea. Headache was an uncommon symptom in this group, as compared to the others. Surgical morbidity (Table 4) occurred in three cases. One patient developed a postoperative CSF fistula with resultant meningitis, and required a second operation to repack the sella with fat. The other two patients had minor complications: one developed a nasal septal perforation, and the other a slight nasal deformity. Rathke s CleJ~ Cysts Eight female and two male patients had Rathke s cleft cysts. The mean age in this group was 34 years (range 21 to 69 years). The mean follow-up period was Abnormality TABLE 2 Preoperative symptoms and signs Pars Intermedia Cysts Rathke s Cleft Cysts Arachnoid Cysts Other Cysts Percent No. Percent No. Percent No. Percent No. headache 31 5/ /11 menstrual irregularities* 88 14/ /8 galactorrhea* 63 10/ /8 impotencet /3 infertility* 6 1 / visual field cut hyperprolactinemia / /11 2/11 hypothyroidism 6 1 / /11 adrenal abnormality 6 1 / /11 decreased testosteronet /3 discovered incidentally -- * Females only. t Males only. 50 4/8 33 1/ /8 4/ /3 3/3 25 2/8 66 2/ / /1 10 J. Neurosurg. / Volume 60 / January, 1984

4 - - o Transsphenoidal treatment of intrasellar cysts Abnormality TABLE 3 Postoperative improvement in dy~/imction Pars Intermedia Cysts Rathke s Cleft Cysts Arachnoid Cysts Other Cysts Percent No. Percent No. Percent No. Percent No. headache 80 4/5 80 8/10 menstrual irregularities* 86 12/ /5 galactorrhea* 90 9/ /2 impotence /2 infertility* 0 0/1 -- visual field cut /3 hyperprolactinemia 66 6/9 50 1/2 hypothyroidism / 1 0 0/2 adrenal abnormality / 1 50 l/2 decreased testosterone+ * Females only. r Males only. 75 3/ /1 0/1 -- o o/1 O/l 0 0/ /1 25 1/4 33 1/3 0 0/2 0 0/2 -- o o/1 34 months. The majority of these patients presented with headache and, among the females, menstrual irregularities. Three patients had visual field deficits that were directly related to suprasellar extension and chiasmal compression. These deficits rapidly resolved after surgical therapy. This group included one patient with minor morbidity: a nasal septal perforation. A rachnoid Cysts Six male and two female patients had arachnoid cysts. The mean age was 42.4 years (range 17 to 59 years). The mean follow-up period was 70 months; 100% follow-up review was achieved. In 50% of these patients, the cyst had been discovered incidentally when sellar abnormalities were noted on x-ray films obtained for the evaluation of minor head trauma. Headaches became less frequent or less severe in all but one patient who had them preoperatively. Hypopituitarism did not improve in most cases. Surgical morbidity in this group included one instance of a postoperative CSF fistula that closed after 4 days of lumbar subarachnoid drainage, and one case of CSF fistula and meningitis, in which repacking of the sella with fat and lumboperitoneal shunting were required to close the fistula. Cyst Group TABLE 4 Surgical morbidity related to each cyst group pars intermedia 16 cysts Rathke s cleft cysts 10 arachnoid cysts 6 other cysts 3 No. of Cases * CSF = cerebrospinal fluid. Total Compli- Complication* cations 3 CSF fistula with meningitis nasal deformity nasal septal perforation 1 nasal septal perforation 2 CSF fistula with meningitis CSF fistula without meningitis 0 none J. Neurosurg. / Volume 60 / January, 1984 Miscellaneous Cysts The group with miscellaneous other cysts consisted of three patients. One, a 41-year-old man, had an intrasellar epidermoid cyst, and another, a 42-year-old woman, had systemic cysticercosis and an intrasellar cysticercosis cyst. Both of these cysts, including the cyst wall, were removed totally. The third patient in this group had previously undergone craniotomy and irradiation for a pituitary adenoma at another institution. She developed a new symptomatic cyst that was treated at UCSF by transsphenoidal drainage; at the time of surgery, no residual adenoma was seen either grossly or on microscopic examination of the cyst wall. There was no morbidity in this group. Discussion Pituitary cysts presently are classified according to a variety of systems, and the nomenclature differs among endocrinologists, anatomists, and neurosurgeons. In these systems, classifications are based on histological, developmental, or endocrinological characteristics, each more or less to the exclusion of the other. It appears, however, that these cysts are not easily distinguished on the basis of such criteria. In the past, many authors believed that all pituitary cysts arise from Rathke s cleft. ~s Anatomical and zoological evidence ~-3 ~ ~,2o now suggests that this is not the case. lntrasellar cysts may develop when colloidal substances accumulate in a follicle or pseudofollicle. These cysts occur in the zona intermedia and pars distalis; 3 such cysts in organs other than the central nervous system are referred to as distention cysts. Horvath, et al., j~ suggested that the follicles develop in the human pars distalis around foci of ruptured and degenerating endocrine cells. In fact, several authors have suggested that these cysts may be a consequence of cellular degeneration;j-3,~, adenohypophyseal cells, particularly basophils, have been implicated in this phenomenon. Intrasellar cysts may arise also from vestigial remnants of the third ventricle or subarachnoid space that extend 11

5 D. S. Baskin and C. B. Wilson into the pituitary fossa. As vascular insufficiency can lead to infarction with subsequent cyst formation, cysts in the sparsely perfused pars intermedia 3 may occur as a result of ischemia. Considering these diverse possibilities, we propose that a definitive identification of non-neoplastic intrasellar cysts can be made only if the morphological configuration of these cysts is considered in conjunction with their in vivo characteristics and consequences. Our system of classification represents an attempt to formulate such a comprehensive scheme, and thereby to simplify the nomenclature. A number of factors warrant further evaluation, however. The pars intermedia cysts in our series may have resulted from any of several pathophysiological mechanisms. A high percentage of patients with pars intermedia cysts and Rathke s cleft cysts had resolution of headache, menstrual irregularities, and galactorrhea. In contrast, the patients with arachnoid cysts did not fare well endocrinologically, although their headaches resolved; this was also the case in the group of patients with miscellaneous cysts. Perhaps the long-standing pulsatile pressure in the arachnoid cyst causes more intrinsic adenohypophyseal damage than does the simple expansile pressure produced by the other groups of cysts. The lack of postoperative improvement in patients in the group with miscellaneous cysts is not surprising, as these lesions were more extensive and had destroyed most of the normal pituitary gland. The exclusive occurrence of pars intermedia cysts and high incidence of Rathke s cleft cysts in females has not been noted previously. Perhaps the endocrinological surges in FSH, LH, and prolactin secretion associated with menses and pregnancy in women increase the likelihood of degeneration of the adenohypophyseal cells, which leads subsequently to cyst formation. Although many authors have reported good results with the treatment of intrasellar cysts by means of craniotomy and cyst drainage, 4 6"8"9"j5"17"2~ this procedure should be undertaken only in cases in which the risks ofcraniotomy can be justified. During the past 4 years, several authors have described the treatment of these cysts with transsphenoidal methods--an approach that involves fewer risks and complications, and therefore permits judicious surgical intervention in a greater number of cases. 7 10"12"13"16"21"24 In almost all of our cases, the cyst was discovered during the transsphenoidal operation. As increasing numbers of surgeons use the transsphenoidal approach for lesions in the intrasellar region, it is likely that more patients will be found to harbor these cysts. As a result of the development of high-resolution CT scanners and of computer software for obtaining specialized scans of the sellar region, smaller and more subtle intrasellar abnormalities are being detected. Despite the sensitivity of these techniques, our experience indicates that there are no specific radiographic features that distinguish between different types of cysts; in fact, there are no radiographic features that reliably distin- guish cysts from neoplasms, even when scans are examined retrospectively. Clinical signs and symptoms usually do not clarify the diagnosis: in most of our patients, the combination of symptoms and signs (headache, visual disturbances, menstrual irregularities, and hyperprolactinemia) and the abnormality noted on CT scanning misled us to a preoperative diagnosis of pituitary tumor. These factors have influenced us to develop the following guidelines to the management of presumed intrasellar lesions. The surgeon must adhere to clear neurological and endocrinological indications for operation. Bearing in mind that 20% of individuals who have no clinical manifestations of intrasellar pathology have a focal intrasellar area of hypodensity on CT scanningy 5 the surgeon must be as certain as possible that the patient s clinical manifestations can be attributed to the lesion, whether neoplastic or non-neoplastic, represented by such an area on the scan. Without going into the details of the evaluation process, the matter can be stated simply: If the endocrine data indicate the presence of hyperprolactinemia that can be ascribed to a prolactinsecreting tumor, and if the patient fulfills the criteria for transsphenoidal removal, then the surgeon should proceed with the operation. If the results of conventional and dynamic testing indicate that endocrine function is normal, and if compression of extrasellar structures is excluded, then an operation is not indicated-- even for patients who may have been referred because of headache or because of normoprolactinemic amenorrhea, galactorrhea, or both. Simple drainage, together with biopsy and histological analysis of the cyst wall to exclude the possibility of neoplasm, is adequate therapy. As total removal of the cyst wall was not necessary to prevent recurrence in all but three of the reported cases, J2"J~ further dissection of the cyst wall would appear to be unwarranted, particularly as it subjects the patient to an increased risk of morbidity, including partial or total pituitary insufficiency. The results of our study suggest that different types of pituitary cysts produce different clinical syndromes that vary with respect to symptoms, signs, and predilection for a particular age group or sex. Except in the case of large cysts with suprasellar extension and chiasmal compression, the size of the cyst did not influence the dividual patient s clinical syndrome in our series, nor did it correlate with eventual outcome. Simple transsphenoidal drainage appears to provide safe and effective therapy for the majority of patients. Acknowledgments We thank Shalawn Weitzel for preparing the manuscript and Susan Eastwood-Berry for editorial assistance. References 1. Baar HS: Dysontogenetic pituitary cysts (pituitary cachexia in childhood). Arch Dis Child 22: , 1947 ~2 J, Neurosurg. / Volume 60 / January, 1984

6 Transsphenoidal treatment of intrasellar cysts 2. Benjamin M: Review. Cysts (large follicles) and colloid in pituitary glands. Gen Comp Endocrino145: , Benjamin M, Williams JG: Pituitary cysts in the ninespined stickleback, Pungitius pungitius L. II. Light and electron microscopy. Acta Zool 60: , Berry RG, Schlezinger NS: Rathke-cleft cysts. Arch Neurol 1:48-58, Chambers EF, Turski PA, LaMasters D, et al: Regions of low density in the contrast-enhanced pituitary gland: normal and pathologic processes. Radiology 144: , Concha S, Hamilton BPM, Millan JC, et al: Symptomatic Rathke s cleft cyst with amyloid stroma. J Neurol Neurosurg Psychiatry 38: , Dobbs ME: Amenorrhea-galactorrhea syndrome caused by a benign pituitary cyst: case report. Fertil Steril 33: , Fager CA, Carter H: Intrasellar epithelial cysts. J Neurosurg 24:77-81, Fairburn B, Larkin IM: A cyst of Rathke s cleft, d Neurosurg 21: , Hatter LP, Silverberg GD, Brant-Zawadski M: Intrasellar arachnoid cyst: case report. Neurosurgery 7: , Horvath E, Kovacs K, Penz G, et al: Origin, possible function and fate of "follicular cells" in the anterior lobe of the human pituitary. An electron microscopic study. Am J Pathol 77: , Marcincin RP, Gennarelli TA: Recurrence of symptomatic pituitary cysts following transsphenoidal drainage. Surg Neurol 18: , Martinez L J, Osterholm JL, Berry RG, et al: Transsphenoidal removal ofa Rathke s cleft cyst. Neurosurgery 4:63-65, McGrath P: Cysts of sellar and pharyngeal hypophyses. Pathology 3: , Raskind R, Brown HA, Mathis J: Recurrent cyst of the pituitary: 26-year follow-up from first decompression. Case report. J Neurosurg 28: , Rout D, Das L, Rao VRK, et al: Symptomatic Rathke s cleft cysts. Surg Neuroi 19:42-45, Rowbotham GF, Clarke PRR: Colloid cyst of the pituitary gland causing chiasmal compression. Br J Surg 44: , Shanklin WM: The incidence and distribution of cilia in the human pituitary with a description of micro-follicular cysts derived from Rathke s cleft. Aeta Anat 1! : , Shanklin WM: On the presence of cysts in the human pituitary. Anat Ree 104: , Shuangshoti S, Netsky MG, Nashold BS Jr: Epithelial cysts related to sella turcica. Proposed origin from neuroepithelium. Arch Pathol 90: , Spaziante R, de Divitiis E, Stella L, et al: Benign intrasellar cysts. Surg Neurol 15: , Smith RA III, Bucy PC: Pituitary cyst lined with a single layer of columnar epithelium. J Neurosurg 10: , Steinberg GK, Koenig GH, Golden JB: Symptomatic Rathke s cleft cysts. Report of two cases. J Neurosurg 56: , Trokoudes KM, Walfish PG, Holgate RC, et al: Sellar enlargement with hyperprolactinemia and a Rathke s pouch cyst. JAMA 240: , Turski PA, Newton TH, Horten BH: Sellar contour: anatomic-polytomographic correlation. AJNR 2: , Weber EL, Vogel FS, Odom GL: Cysts on the sella turcica. J Neurosurg 33:48-53, Yoshida J, Kobayashi T, Kageyama N, et al: Symptomatic Rathke s cleft cyst. Morphological study with light and electron microscopy and tissue culture. J Neurosurg 47: , 1977 Manuscript received March 21, Accepted in final form August 9, Address reprint requests to: David S. Baskin, M.D., Department of Neurological Surgery, % The Editorial Office, 350 Parnassus Avenue, Suite 807, San Francisco, California J. Neurosurg. / Volume 60 / January,