March 2007 Vol 7 Isuue 1 FINAL without suicide .indd ABSTRACT Objective: From a recently instituted web-based pituitary tumour registry at Sultan Qaboos University Hospital, Oman, this study explores the results of comprehensive clinical evaluation, hormonal levels, radiological evidence of pituitary mass lesion using magnetic resonance (MRI) and the different treatment modalities. Methods: All patients who were diagnosed with pituitary mass tumours in our tertiary care endocrinology clinic between January 998 and February 2006 were registered in the Oman pitui- tary tumour registry. Two physicians performed hospital chart review and data entry. Results: A total of 60 entries were made into the pituitary tumour registry. The overall mean age of the cohort was 32 ±2 years (age range 8-73 years). The majority of registrations were female (n=4; 7%). There were 8 patients with non-functioning adenomas (50.6%), 59 with prolactinoma (36.9%) eight with acromegaly (5%), seven with craniopharyngioma (4.4%), four with Cushing’s disease (2.5%) and one with sarcoidosis (0.6%). Sub-group analyses were done only for the subjects with the 3 most prevalent pituitary tumours (non-functioning adenomas, prolactinomas, and acromegaly). The most prevalent symptoms are amenorrhea-galactorrhea (n=55; 37%), headache (n=3; 2%) and fatigue (n=23; 6%). The most common treatment modality was medical (n=58; 39%), followed by observation (n=56; 38%), surgery (n=3; 2%) and surgery plus medical (n=3; 2%). None of the patients in this registry are recorded to have died. Conclusion: To our knowledge, this is the first pituitary tumour registry in the Arabian Gulf countries using a web-based programme. This tumour registry will enable us to Clinical and Epidemiological Characteristics of Pituitary Tumours using a Web-based Pituitary Tumour Registry in Oman *Abdullah Al-Futaisi 1, Al-Yaarubi Saif1, Ibrahim Al-Zakwani 2, Salim Al-Qassabi 3, Shaden Al-Riyami 4, Yasser Wali 4 1Department of Medicine, Sultan Qaboos University, P. O. Box 35, Al-Khod 123, Sultanate of Oman, 2Department of Pharmacy, Sultan Qaboos University Hospital, P. O. Box 35, Al-Khod 123, Sultanate of Oman, 3Department of Medicine, Royal Hospital, Muscat, Oman, 4Child Health De- partment, College of Medicine & Health Sciences, Sultan Qaboos University, P. O. Box 35, Al-Khod 123, Sultanate of Oman. *To whom correspondence should be addressed. Email: alfutaisi@squ.edu.om الشبكة باستعمال الغدة النخامية ألورام الصفات السريرية والوبائية عمان في سلطنة العاملية والي وياسر الريامي شادن القصابي، سالم الزكواني، إبراهيم اليعربي، سيف الفطيسي، عبداهللا السلطان مستشفى جامعة في النخامية في الغدة بأورام املصابني املرضى املعلومات عن جميع بتجميع قمنا الدراسة هذه امللخص: الهدف: في العاملية. ــبكة الش في النخامية الغدة ــأورام اخلاص ب ماني العُ ــجل الس في وإدخالها بتحليلها ــابقة وقمنا الس ــنوات الس الثماني خالل ــوس في قاب 2006 فبراير 1998 إلى يناير من للفترة قابوس السلطان جامعة ــفى مستش في النخامية الغدة بأورام املصابني املرضى ــجيل كل تس الطريقة: مت تتعلق من معلومات السجل يتكون املرضى. عن هؤالء املعلومات كل بتجميع طبيبني قام العاملية. بالشبكة النخامية الغدة بأورام خاص سجل في 160 حالة تسجيل طرق العالج. النتائج: مت ومختلف األورام تُبني التي املغناطيسي الرنني وصور الدم في الهرمونات مستوى ، املريض الصحية بحالة .(71% - 114) اإلناث من املرضى معظم (8 الى73 سنة). من واملدى ، ــنة) املرضى(12±32 س أعمار ــط متوس بلغ النخامية. الغدة بتورم مصابة ، (36.9%) البروالكتني بورم يفرز هرمون مصابا 59 مريضا و ، (50.6%) ــبة بنس هرمون أي ال يفرز النخامية بورم الغدة مصابا 81 مريضا ــجيل مت تس يفرز بورم 4 مرضى مصابني و ،(4.4%) النخامية الغدة في يّ يٌّ بُلْعومِ فِ قِحْ بوَرَمٌ مصابني 7 مرضى و ،(5%) النمو هرمون يفرز بورم مصابني 8 مرضى و األورام من 3 أنواع انتشار تبني الفرعية وقد اموعات حتليل يتم أيضاً .(0.6%) بساركويدوسيس مصاب ومريض واحد ر ظْ الكُ رِ شْ هُ لقِ ُوَجِّ ونُ امل رْمُ الهُ الشهرية الدورة انقطاع من تعاني كانت احلاالت أكثر النمو). هرمون تفرز التي واألورام البروالكتني هرمون تفرز التي واألورام هرمونات تفرز ال التي (األورام األدوية بواسطة العالج كان الدراسة هذه في شيوعاُ العالج طرق أكثر .(16% - (23 حالة واالرهاق (21% - (31 حالة الصداع و (37% - (55 حالة يتوفى لم .(2% - (3 مرضى مع األدوية اجلراحي والتدخل (21% - (31 حالة والتدخل اجلراحي (38% - (56 مريضا واملالحظة (39% - (58 مريضا ــجالت في الس أوائل من ــبكة العاملية الش في النخامية الغدة ألورام ماني العُ ــجل الس اخلالصة: يعد هذا ــة. الدراس فترة املرضى خالل هؤالء من أي البحوث إجراء على يساعد مما املرضى، هؤالء عن جتميع املعلومات سهولة على والباحثني يساعد األطباء سوف السجل هذا تكوين أن اخلليج العربي. ماني. العُ اتمع في الصحية الكفاءة ورفع عالجهم طرق تسهيل إلى وهذاسيؤدي األورام، هذه عن علم االوبئة. سريري، مان، عُ تسجيل، ، يّ نُخامِ يٌّ دِّ غُ الكلمات: وَرَمٌ مفتاح SULTAN QABOOS UNIVERSITY MEDICAL JOURNAL APRIL 2007 VOL 7, NO. 1 SULTAN QABOOS UNIVERSITY© C L I N I C A L & B A S I C R E S E A R C H A B D U L L A H A L - F U TA I S I , A L -YA A R U B I S A I F, I B R A H I M A L - Z A K WA N I , S A L I M A L - Q A S S A B I , S H A D E N A L - R I YA M I , YA S S E R WA L I 26 PITUITARY ADENOMAS ARE TUMOURS OF THE anterior pituitary that are relatively common neoplasms representing 10% to 15% of all in- tracranial tumours.1, 2 Pituitary adenomas are the most common cause of a mass in the sella constituting over 90% of pituitary lesions.3, 4 The prevalence of silent pituitary microadenomas detected by high-resolution computer-tomography (HRCT) or magnetic-reso- nance imaging (MRI) is approximately 20%.5, 6, 7, 8, 9 At autopsy, careful histological assessment identifies pi- tuitary adenomas in 10% to 20% of unselected series.10 The incidence peaks in the third and fourth decades of life; children and adolescents account for about 10% of total patients.10,11 For clinical purposes, pituitary adenomas are ar- bitrarily divided by size into microadenomas (<1.0 cm in diameter) or macroadenomas (≥1.0 cm in di- ameter). When tumours erode the dura or bone, they are considered invasive and may infiltrate surround- ing structures, such as the cavernous sinus, cranial nerves, blood vessels, sphenoid bone and sinus or brain. Locally invasive pituitary adenomas are nearly always histologicaly benign.12 They can also be classi- fied on the basis of excessive hormonal secretion into secreting and non-secreting types.2 Pituitary tumours with no enhanced biological activity of the anterior pituitary hormones are called non-functioning pitui- tary adenomas (NFPA). However, approximately 90% of these tumours will produce or secrete low amounts of follicle-stimulating hormone (FSH) and luteinizing hormone (LH) and or their α and β subunits 2,13 Us- ing immunoassay and in-situ hybridization, a minority of the NFPAs shows no evidence of hormone produc- tion. These tumours are called null cell adenomas. The true incidence of pituitary tumours is still not known and accurate tumour type-specific epidemio- logical information is unreliable.14 This is probably due to the lack of an international pituitary tumour regis- try, the long natural history of tumours and the low annual incidence of pituitary adenoma diagnosis. In an effort to register these patients in Oman, we cre- ated a retrospective and prospective pituitary tumour registry in the year 2004 using an Access 2000 Mi- crosoft web-based programme. This programme was used to collect demographic data, biochemical results, therapeutic options and clinical outcomes from pa- tients with different pituitary mass lesions referred to our endocrinology clinic at Sultan Qaboos University Hospital (SQUH), the main referral hospital for these patients, or submitted by other tertiary institutions since January 1998. The web-address for the registry (http://www.omands.org:2082/frontend/rvblue/files/ showfile.html?dir=/home/omands&file=list.htm) is password-protected to prevent any unauthorized access and preserve the privacy and confidentiality of the patients. The purpose of this study is to use this pituitary tumour registry to characterize the clinical and epidemiological features of pituitary tumours in the Sultanate of Oman. M E T H O D S All patients who were suspected to have a pituitary mass tumour and were referred to our tertiary care endocrinology clinic at SQUH, between January 1998 Characteristic NFPA (n=81) Prolactinomas (n=59) Acromegaly (n=8) Total (n=148) p-value Age, mean±SD 33±14 32 ±8 42 ±15 33 ±12 0.058 Male gender, n (%) 23 (28%) 14 (24%) 5 (63%) 42 (28%) 0.087 Microadenoma, n (%) 54 (67%) 51 (86%) 2 (25%) 107 (72%) <0.001 Macroadenoma, n (%) 27 (33%) 8 (14%) 6 (75%) 41 (28%) With extension, n (%) 24 (30%) 4 (7%) 4 (50%) 32 (22%) <0.001 Without extension, n (%) 57 (70%) 55 (93%) 4 (50%) 116 (78%) Non-functioning Adenomas; SD=Standard Deviation; Percents are total percents; p-values were generated using univariate statistics; The 0.058 represents the p-value for the overall F-statistic, derived from univariate regression analysis. However, group-wise differences for age did exist, especially between the cohorts, acromegaly and NFPA (42 vs. 33 years; p=0.045) and acromegaly and prolactinomas (42 vs. 32 years; p=0.018). Table 1 : Demographic and tumour characteristics of the three most prevalent pituitary tumours characterize clinical and the epidemiological features of pituitary tumours in the Sultanate of Oman. Keywords: Pituitary adenomas, Registry, Oman, Clinical, Epidemiology. C L I N I C A L A N D E P I D E M I O L O G I C A L C H A R A C T E R I S T I C S O F P I T U I TA R Y TU M O U R S U S I N G A WE B - B A S E D P I T U I TA R Y TU M O U R R E G I S T R Y I N O M A N 27 and February 2006, were included in the newly de- veloped Oman pituitary tumour registry. Other ter- tiary hospitals in the country were invited to submit their data. The analysis performed was retrospective and did not in anyway interfere with the manage- ment of the patients. Futhermore, those involved in the data analysis did not have access to patient details. Therefore, present study was not deemed to contra- vene ethical dimensions or patient confidentiality. After confirming the diagnosis for each new patient, his/her data were entered in the software programme, a web-based pituitary tumour registry computer pro- gramme, Access 2000 from Microsoft Corporation, 2000. Up to February 2006, the study had enrolled 160 patients. The data include demographic informa- tion (age, gender, area of residence), a comprehensive clinical evaluation, hormonal levels, medications and radiological evidence of pituitary mass lesion using MRI. Pituitary tumour type was confirmed by the clinical presentation of patients, hormones levels, ra- diographic presentations, and immuno-histochemical detection of hormone content whenever available. The standard protocol for the evaluation of pituitary gland with MRI was used to visualize intrasellar pituitary mass lesion. It consists of obtaining T1-weighted im- ages and then repeating the T-1 images after admin- istration of an intravenous contrast agent containing Gadolinium-based media.12 Not all of the MRI’s were done at SQUH. Others were performed elsewhere and brought in by patients themselves. The size of the ad- enomas was classified by radiographic analysis. Pitui- tary adenomas were divided by size into microadeno- mas (<1.0 cm in diameter) or macroadenomas (≥1.0 cm in diameter).12 The hormonal samples were ana- lyzed using Beckman-Coulter Immunochemistry Sys- tem Analyzer for Prolactin, FSH, LH, growth hormone (GH), thyroid-stimulatory hormone (TSH), cortisol and insulin-like growth factor (IGF-I). All variables were updated and tracked on each visit. Patient con- fidentiality was maintained by entering a pre-assigned arbitrary identification number known only to the in- vestigators. Clinical and mortality outcomes were also collected and tracked on flow worksheets. STATI STI C A L A N A LY SI S Descriptive statistics were used to describe the data. For categorical variables, frequencies and percentages were reported. Differences between groups were ana- lyzed using Pearson’s X2 tests (or Fisher’s exact tests for cells less than 5). For continuous variables, means and standard deviations (±SD) were presented. Mean differences between groups were analyzed using Stu- dent’s t-tests. An a priori two-tailed level of signifi- cance was set at the 0.05 level. R E S U L T S D E MO GR A PH I C A N D T UMO UR C H A R A C TE R I S - TI C S Over an 8-year period (1998-2006), a total of 160 en- tries representing a heterogeneous mixture of the Om- ani population were made into the pituitary tumour registry. The overall mean age of the cohort was 32±12 years with an age range from 8 to 73 years. The ma- jority of the registrations were females (n=114; 71%). Just over half of the pituitary tumours were non-func- tioning adenomas (n=81; 50.6%), followed by prolac- tinomas (n=59; 36.9%), acromegaly (n=8; 5.0%), crani- opharyngiomas (n=7; 4.4%), Cushing’s disease (n=4; 2.5%), and sarcoids (n=1; 0.6%). Sub-group analyses Feature Column percentages Non-functioning Adenomas (n=81) Prolactinomas (n=59) Acromegaly (n=8) =148 Amenorrhea-Galactorrhea 4 (4.9%) 50 (84.7%) 1 (12.5%) Headache 29 (35.8%) 2 (3.4%) 0 Visual Field Defects 4 (4.9%) 0 0 Diabetes Insipidus 3 (3.7%) 0 0 Hypogonadal symptoms 11 (13.6%) 4 (6.8%) 0 Delayed Growth 2 (2.5%) 0 0 Hypopituitary symptoms 7 (8.6%) 1 (1.7%) 1 (12.5%) Fatigue 21 (25.9%) 2 (3.4%) 0 Acral Enlargement 0 0 6 (75.0%) Table 2: Clinical features of the three most prevalent pituitary tumours A B D U L L A H A L - F U TA I S I , A L -YA A R U B I S A I F, I B R A H I M A L - Z A K WA N I , S A L I M A L - Q A S S A B I , S H A D E N A L - R I YA M I , YA S S E R WA L I 28 were performed only for those subjects with the three most prevalent pituitary tumours (non-functioning adenomas, prolactinomas and acromegaly). There were significant age differences in patients with the three most prevalent pituitary tumours [Ta- ble 1]. The subjects with acromegaly were significantly older compared to those with non-functioning adeno- mas (42 versus 33 years; p=0.045) or prolactinomas (42 versus 32 years; p=0.018). Those with acromegaly were more likely to be males than females (63% versus 37%; p=0.042). Furthermore, subjects with acromegaly were significantly more likely to have macroadenomas than those with either non-functioning adenomas or prolactinomas (75% versus 33% versus 14%; p<0.001). Additionally, those with acromegaly were also more likely to have tumours with extension than those with either non-functioning adenomas or prolactinomas (50% versus 30% versus 7%; p<0.001). Those with ex- tension were significantly more likely to be males than females (38% versus 15%; p=0.002). In addition, males were also significantly more likely to have macroad- enomas as compared to females (45% versus 21%; p=0.003). Tumour-specific presenting clinical features are presented in Table 2. The most prevalent symptoms were amenorrhea-galactohorrhea (n=55; 37.2%), headache (n=31; 21%) and fatigue (n=23; 16%). The symptom, amenorrhea-galactohorrhea, was mostly associated with those patients with prolactinomas rather than those with non-functioning adenomas or acromegaly (84.7% versus 4.9% versus 12.5%; p<0.001). Headache was most associated with those with non- functioning adenomas rather than those with either prolactinomas or acromegaly (35.8% versus 3.4% ver- sus 0%; p<0.001). Furthermore, fatigue was also most- ly associated with non-functioning adenomas rather than prolactinomas or acromegaly (25.9% versus 3.4% versus 0%; p=0.001). TH E R A P Y A N D C L I N I C A L O U TC OME Several therapeutic modalities were employed for the treatment of the different pituitary tumours includ- ing observation, medical, surgery, and medical plus surgery [Table 3]. Out of the three most prevalent pituitary tumour types, the most common treatment modality was medical (n=58; 39%) followed by obser- vation (n=56; 38%), surgery (n=31; 21%) and surgery plus medical (n=3; 2%). Those with prolactinomas were all medically treated compared with those with either non-functioning adenomas or acromegaly (98% versus 0% versus 0%; p<0.001). Observation was the main treatment modality for those with non-func- tioning adenomas, compared with those with either prolactinomas or acromegaly (69% versus 0% versus 0%; p<0.001). Surgical intervention was mostly associ- ated with those that have acromegaly than those with either non-functioning adenomas or prolactinomas (75% versus 31% versus 0%; p<0.001). Up to February 2006 none of the patients were recorded to have died. D I S C U S S I O N During the past 3 decades, the Sultanate of Oman has undergone a remarkable transformation and mod- ernization in all fields, including education and health care. It has progressively established a nationwide network of modern health services that are accessible even at the village level and have ranked Oman with the developed world.15 World wide, tumour registries have become one of the strongest tools used to help health workers in the clinical and epidemiological studies of cancer. The true incidence of pituitary tu- mours is still not known and accurate tumour type- specific epidemiological information is lacking, due to the non-existence of any web-based international pi- tuitary tumour registry. In addition, accurate records of outcomes and mortality are not available. For all these reasons, we developed a web-based programme in which all patients in Oman with pituitary tumours are registered and followed. To our knowledge, this is Non-functioning Adenomas (n=81) Prolactinomas (n=59) Acromegaly (n=8) Medical 0 98% 0 Observation 69% 0 0 Surgery 31% 0 75% Surgery+Medical 0 2% 25% Total 100% 100% 100% Table 3: Therapeutic modalities of the three most prevalent pituitary tumours C L I N I C A L A N D E P I D E M I O L O G I C A L C H A R A C T E R I S T I C S O F P I T U I TA R Y TU M O U R S U S I N G A WE B - B A S E D P I T U I TA R Y TU M O U R R E G I S T R Y I N O M A N 29 the first pituitary tumour registry in the Arabian Gulf countries and the Middle East. Of maximum interest is the unexpectedly large number of new patients (160) registered in a country with a population of a 2.3 million15 over the past 8-year period, giving one case per 14,375 of the population. The majority of patients were symptomatic and almost half of them had functioning adenomas. In a recent ar- ticle by Drange et al., who reviewed the hospital charts of patients with pituitary tumours between 1982 to 1999 in California State (with a population of more than 35 million), USA, they found 404 patients,14 an incidence of 1 per 86,634 of the population. There are many factors that might explain the higher numbers in the Omani population. Genetic, ethnic and environ- mental differences and the extended use of traditional medicine may contribute. These hypotheses need fur- ther studies. The mean age of acromegalic patients at diagnosis (42 years) did not differ from those reported in previ- ous series.14,16 There was a significant male predomi- nance 63% versus 37% (p=0.042). Beauregard et al. reported a similar male predominance, however sev- eral other studies reported female predominance.17 The female predominance may be due to the fact that females usually present earlier with amenorrhea.2, 3, 11, 12, 13, 15 Additionally, those with acromegaly were also more likely to have tumours with extension than those with either non-functioning adenomas or prolactino- mas (50% versus 30% versus 7%; p<0.001). Those with extension were significantly more likely to be males than females (38% versus 15%; p=0.002). In addition, males were also significantly more likely to have mac- roadenomas as compared to females (45% versus 21%; p=0.003). This gender difference could be due to the fact that more men feel shy to complain about their sexual problems (decreased libido and impotence). In our study, these patients appeared to be less sympto- matic in comparison to patients with other tumour types. This can be explained by the fact that growth harmone GH secreting tumours grow slowly and re- sult in subtle changes in body features. All of them were treated surgically and 25% of them received med- ical treatment as well GH receptor antagonist. Fifty-nine patients in the database were confirmed as having prolactinomas; females (44) constituted 76% of them. They were predominantly microadenomas (86%). These results are consistent with the results of previously published figures.18 In our patients, the most frequent clinical presentations were amenorrhea and galactorrhea followed by hypogonadal symptoms and headache. The tumour showed evidence of exten- sion in 2 males (14.3%) and 2 females (4.6%). Males were more likely to have tumour extensions due to their late presentation. Non-functioning pituitary tumours were the most frequent tumour (n=81) and most commonly affected the females (72%). They were mostly microadenoma with no extension (70%). The most common present- ing symptoms were headache (35.8%) followed by fatigue (25.9%), hypogonadal symptoms (13.6%) and amenorrhea (4.9%). In our study we found that 8.7% of patients with non-functioning adenomas had hypo- pituitary symptoms. Pituitary insufficiency can be en- countered in patients with any size or type of intrasel- lar mass and, thus, serves as an important diagnostic indicator of a pituitary lesion.19 C O N C L U S I O N A tumour registry systematically collects information about the occurrence of cancer, the types of cancer that occur (site, morphology and behaviour), extent of disease at the time of diagnosis (stage), the kinds of treatment received by cancer patients and the out- comes of treatment (survival). The ultimate goal of the analysis of cancer data is to prevent and control cancer. Data is also used to respond to local questions about referral patterns and trends by disease site, by treatment modality and even by clinician. The regis- try’s role of evaluating clinical outcomes and quality of medical care includes assessing treatment patterns, complications, cancer recurrence and survival. The goal of the pituitary tumour registry is improvement in early identification of these patients, tumour man- agement strategies, and long-term outcomes. In conclusion, we have developed a novel compu- terized pituitary tumour registry programme in the Sultanate of Oman designed to collect comprehensive demographic, therapeutic, pathologic, and outcome information on patients harbouring these rare tu- mours and who are referred for tertiary consultation. This registry can be used to improve surveillance pro- tocols and long-term therapeutic outcomes. A C K N OW L E D GE ME N TS The authors would like to acknowledge the work of all the healthcare professionals who were involved in the management of the patients. A B D U L L A H A L - F U TA I S I , A L -YA A R U B I S A I F, I B R A H I M A L - Z A K WA N I , S A L I M A L - Q A S S A B I , S H A D E N A L - R I YA M I , YA S S E R WA L I 30 R E F E R E N C E S 1. Gsponer J, De Tribolet N, Deruaz JP, et al. Diagnosis, treatment, and outcome of pituitary tumours and other abnormal intrasellar masses. Retrospective analysis of 353 patients. Medicine 1999; 78:236-269. 2. Freda PU, Post KD. Differential diagnosis of sellar mass- es. Endocrinol Metab Clin North Am 1999; 28:81-117. 3. Molitch ME. Pituitary incidentalomas. Endocrinol Me- tab Clin North Am 1997; 26:725-740. 4. Kovacs K, Scheithauer BW, Horvath G, Lloyd RV. The World Health Organization classification of adenohy- pophysial neoplasms. Cancer 1996; 78:502-510. 5. Auer RN, Alakija P, Sutherland GR. Asymptomatic large pituitary adenomas discovered at autopsy. Surg Neurol 1996; 46:28-31. 6. Burrow GN, Wortzman G, Rowcastle NB, Holgate RC, Kovacs K. Microadenomas of the pituitary and abnor- mal sellar tomograms in an unselected autopsy series. N Engl J Med 1981; 304:156-158. 7. Hall WA, Luciano MG, Doppman JL, Patronas NJ, Old- field EH. Pituitary magnetic resonance imaging in nor- mal human volunteers: occult adenomas in the general population. Ann Intern Med 1994; 112:817-820. 8. Freda PU, Wardlaw SL, Post KD. Unusual causes of sell- ar/parasellarmasses in a large transsphenoidal surgical series. J Clin Endocrinol Metab 1996; 81:3455-3459. 9. Molitch ME, Russell EJ. The pituitary incidentaloma. Ann Intern Med 1990; 112:925-931. 10. Annegers JF, Coulam CB, Abboud CF, Laws ER, Kurland LT. Pituitary adenoma in Olmstead County, Minnesota, 1935–1977. A report of an increasing incidence of di- agnosis in women of childbearing age. Mayo Clinic Proc 1978; 53:641-643. 11. Kurland LT, Schoenber BS, Annegers JF, Okazaki H, Molgaard CA. The incidence of primary intracranial neoplasms in Rochester, Minnesota, 1935–1977. Ann NY Acad Sci 1982; 381:6-16. 12. Pressman, BD. Pituitary imaging. In: Melmed S. ed. The Pituitary. 2nd ed. , Cambridge: Blackwell Scientific Pub- lications, 2002. p. 690-730 13. Katnelson L, Alexander JM, Klibanski A. Clinically non- functioning pituitary adenomas. J Clin Endocrinol Me- tab 1993; 76:1089-1094. 14. Drange MR, Fram NR, Herman-Bonert V, Melmed S. Pituitary tumour registry: A Novel Clinical Resource.J Clin Endocrinol Metab. 2000; 85:168-174.