| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
BRIEF REPORT |
Department of Internal Medicine (A.A.S., E.V.D., A.L.B.), Division of Endocrinology and Metabolism, and Department of Neurosurgery (W.F.C.), University of Michigan Medical Center, and Department of Veterans Affairs Medical Center (A.L.B.), Ann Arbor, Michigan 48109
Address all correspondence and requests for reprints to: Ariel Barkan, M.D., University of Michigan Medical Center, 3920 Taubman Center, Box 0354, Ann Arbor, Michigan 48109. E-mail: abarkan{at}umich.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsDiscussionReferences |
|---|
We report three female patients with pituitary tumors who presented without any signs or symptoms of acromegaly but with elevated IGF-I levels. Plasma GH was measured every 10 min for 24 h, and an oral glucose tolerance test was performed. All patients had abnormally elevated mean and trough plasma GH levels as well as post-glucose nadir GH concentrations. All patients had magnetic resonance imaging scans revealing pituitary tumors and underwent transsphenoidal surgery. Histologically, they had GH-producing pituitary tumors. Plasma IGF-I levels returned to normal in two patients after surgery.
Some pituitary adenomas are true GH-secreting tumors despite not being accompanied by obvious clinical stigmata of acromegaly. Natural history of this disease is unknown because of the small number of reported patients and inconsistent results of biochemical testing. Based on the results of this and previous reports, we propose that all patients with known pituitary tumors, especially younger women with normal or mildly elevated prolactin level, be evaluated for GH excess.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Case 1. A 44-yr-old woman presented to our clinic complaining of headaches. She had a history of a 1.5-cm pituitary adenoma and had pituitary biopsy 8 yr earlier at a different institution. At that time, she had amenorrhea and galactorrhea and had mildly elevated prolactin level at 35 ng/ml but no symptoms of GH excess. Pituitary tumor biopsy was positive for GH by immunohistochemistry, and postoperative IGF-I level was elevated. She was treated with octreotide acetate up to 200 µg three times daily with no change in IGF-I levels. At the time of her presentation to our clinic, she had no history of acral growth, arthropathy, increased sweating, glucose intolerance, soft tissue growth, or skin tags. Examination of her old photographs failed to disclose any facial changes. Three months after discontinuation of octreotide, her random GH was 3.2 µg/liter, IGF-I 332 µg/liter (normal 100–300 µg/liter), prolactin 12.1 ng/ml, and T4 6.9 µg/dl. Pituitary magnetic resonance imaging (MRI) showed a 1.5-cm right-sided tumor with displacement of the cavernous sinus.
Case 2. A 28-yr-old woman presented with amenorrhea of 3.5-yr duration. She had no change in facial appearance or voice deepening, no increase in hand or shoe size, and no joint aches, headaches, or diaphoresis. Her skin was normal, with no evidence of excessive soft tissue volume, and there was no evidence of facial coarsening, macroglossia, or diabetes mellitus. Basal fasting plasma GH was 8.3 µg/liter, and serum IGF-I was 652 µg/liter (normal 130–480 µg/liter). Prolactin level was slightly elevated at 24.5 ng/ml, and T4 was normal at 5.8 µg/dl. Her MRI revealed a 1.1-cm right-sided adenoma with displacement of the pituitary stalk to the left.
Case 3. A 31-yr-old woman initially presented elsewhere with headache and mild hyperprolactinemia and was found to have a 2-cm pituitary tumor. Initial surgery was limited to tumor biopsy, and immunohistochemistry was positive for the presence of GH-producing cells. Postsurgical serum IGF-I level was elevated at 415 µg/liter, and random GH was 3.5 µg/liter. Her follow-up IGF-I levels remained elevated above 400 µg/liter. She had no facial changes, no symptoms of excessive acral growth, diaphoresis, arthropathy, or impaired glucose tolerance. On physical examination, no clinical stigmata of acromegaly could be detected. Her serum prolactin was 58 ng/ml, and serum IGF-I was elevated at 465 µg/liter (normal 100–390 µg/liter) at the time of presentation. Her repeat MRI continued to demonstrate a 2-cm adenoma in the left side of the sella with no change from the previous study.
At presentation, all patients had amenorrhea with low gonadotropins, were not taking any estrogen preparations, and had no history or current evidence of liver, renal, or cardiac disease or diabetes mellitus.
Control group
Normative plasma GH profile data were obtained from a group of six healthy female volunteers studied during early follicular phase of the cycle. These data were reported previously (5).
Study protocol
The protocol was approved by the Institutional Review Board of the University of Michigan. Written informed consent was obtained from all patients and healthy subjects before any protocol procedures.
Subjects were admitted into the Clinical Research Center, and an iv catheter was placed in a forearm vein for blood sampling. Meals were served at 0700, 1200, and 1800 h. Water was allowed ad libitum. Blood sampling for GH was performed every 10 min from 0700 h on d 1 until 0700 h the next day. Subjects were fasting on the second day from 2400 h. At 0800 h next morning they received a 75-g glucose load, and blood sampling for GH and glucose was performed every 10 min for the next 2 h.
Assays
Plasma GH was measured in duplicate using a chemiluminometric assay (Nichols Institute Diagnostics, San Juan Capistrano, CA) with an assay sensitivity of 0.01 µg/liter as previously described (5). Mean intraassay coefficient of variation of replicate samples was less than 8% in each assay. Assays (immunoradiometric analysis) for plasma IGF-I were performed by the Pathology Laboratories of the University of Michigan Medical Center using commercially available kits (Diagnostic Systems Laboratories, Webster, TX). The sensitivity of the assay is 2.06 µg/liter. Mean intraassay precision is less than 7%. Normal ranges for IGF-I are age and sex dependent. In female subjects, the normal IGF-I ranges are as follows: 20–30 yr old, 130–480 µg/liter; 30–40 yr old, 100–390 µg/liter; 40–50 yr old, 100–300 µg/liter.
Data analysis
Mean 24-h plasma GH was calculated as the average of all plasma GH values over the 24-h sampling period. Trough 24-h plasma GH was calculated as the mean of the lowest 5% of the GH values over the 24-h sampling period. The maximum plasma GH values are defined as the highest GH value during the 24-h sampling protocol. The nadir plasma GH during an OGTT is the lowest GH value during the 2-h period after the administration of oral glucose.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
In all patients, mean, trough, and glucose-suppressed nadir GH levels were elevated. Similarly, their plasma IGF-I levels were all above the upper limit of the age/sex-adjusted normal range. (See data in Table 1
and Figs. 1 and 2).
|
|
|
A GH-producing pituitary adenoma was histologically and immunohistochemically identified in all three patients.
Case 1. Immunohistochemical examination revealed scattered adenoma cells positive for GH, prolactin, and chromogranin A.
Case 2. Immunohistochemical examination revealed scant foci with a faint GH immunoreactivity compressed between the bands of fibrous tissue. Cells were negative for prolactin, ACTH, FSH, and TSH but positive for chromogranin A.
Case 3. Immunohistochemical examination revealed that GH-reactive cells varied from scarce to approximately 50% in different regions of the tumor. Prolactin-positive cells varied from scarce to 35% in different regions.
Postsurgical results
Patient 1. Postoperatively, her random plasma GH was 0.9 µg/liter, and plasma IGF-I level decreased to normal at 230 µg/liter and remained normal during the subsequent 4 yr. Her serum prolactin level decreased to 2.5 ng/ml, and free T4 remained normal at 1.7 ng/dl after the surgery. Her headaches improved markedly, but she remained amenorrheic. Follow-up MRI 2 yr after surgery revealed postsurgical changes.
Patient 2. IGF-I level remained elevated at 374 µg/liter, and prolactin remained slightly elevated at 40.4 ng/ml postoperatively. She was treated with cabergoline for persistent amenorrhea and hyperprolactinemia and had two uneventful pregnancies. Repeat pituitary MRI scan 3 yr after surgery revealed postsurgical changes and no evidence of the tumor. Despite normalization of serum prolactin, her plasma IGF-I remained elevated between 397 and 618 µg/liter for the next 5 yr of follow-up with no development of any symptoms or signs of acromegaly.
Patient 3. Serum IGF-I level returned to normal postoperatively at 230 µg/liter, and random GH decreased to less than 0.5 µg/liter. These values remained essentially unchanged for the next 3 yr of follow-up. Prolactin remained slightly elevated at 29.3 ng/ml, and free T4 was normal at 1.28 ng/dl. Follow-up MRI scan 1 yr after surgery failed to reveal any residual tumor. She had resolution of headaches, and her menstrual periods normalized.
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
This is the first study where GH secretion was assessed with the use of a highly sensitive assay and frequent blood sampling. Our results clearly indicate the existence of abnormal GH secretion in these patients as evidenced by elevated mean and trough GH levels as well as by the lack of GH suppression with glucose load. Although it is conceivable that these patients did in fact have somatic stigmata of GH/IGF-I excess, the degree of clinical manifestations must have been so exceedingly mild as to escape detection by highly experienced pituitary surgeons and endocrinologists.
Several theories have been proposed to explain the absence of clinical signs of GH excess in this entity. Absent GH release and defects in one or multiple sites in the synthesis or posttranslational processing of the GH gene product could explain normal plasma GH values in some of the previously reported patients (4, 6, 12). Increased serum GH levels and abnormal pattern of GH secretion in our patients confirm active GH secretion from the tumors. A lack of biological activity of secreted GH has been proposed (6, 13) as a reason for the lack of signs and symptoms in these patients. This theory is contradicted by the elevation of serum IGF-I level, which is the result of high GH milieu and the main mediator of the growth-promoting GH effects on peripheral tissues. It is possible that silent somatotroph adenomas represent an early stage in somatotrophic adenoma development. The lack of clinical signs then could be related to short duration of GH hypersecretion (2, 6, 7, 14). However, our patients had no clinical symptoms and signs of acromegaly despite a longstanding history of GH hypersecretion. Patient 1 had at least 8 yr of documented IGF-I excess before the final diagnosis, and patient 2 remained under our follow-up for 5 yr, during which time her IGF-I levels also remained persistently high. Neither of them had any visible clinical stigmata of acromegaly. Last, the degree of GH hypersecretion might not have been sufficiently high to cause a full-blown clinical syndrome. Indeed, the degree of GH immunoreactivity within the tumor tissue in our patients was rather mild, and the magnitude of GH and IGF-I elevations was also lower than in the majority of patients with classic acromegaly. Thus, this explanation seems to be the most convincing. On the other hand, we have recently reported a series of patients with full-blown active acromegaly who had normal mean 24-h GH concentrations and only modestly increased trough GH concentrations (5). Thus, a significant heterogeneity of clinical responses must exist between patients with only mild to modest GH increase.
Nonfunctioning pituitary tumors that do not cause cranial nerve compression and do not cause loss of pituitary function do not necessarily require surgical intervention. Periodic documentation of a stable tumor size is needed to support this conservative approach. Biochemical evidence of GH hypersecretion introduces an additional complicating factor in the medical decision making. Because of the small number of reported patients with clinically silent somatotroph adenomas and lack of long-term follow-up, the natural history of this condition is unknown. Epidemiological studies of acromegaly have attributed a 2- to 3-fold higher risk of premature mortality to plasma GH above 2.5 µg/liter and to elevated IGF-I levels (15, 16, 17, 18, 19). Thus, although there is no clear evidence of benefit, the epidemiological data coupled with the low risk of transsphenoidal surgery make a compelling case for the removal of clinically (albeit, not biochemically) silent somatotropinomas. Indeed, in our small group, two of three patients achieved normal postoperative IGF-I levels and had resolution of headaches, and one had normalization of menstrual periods.
Two of the three studied patients also had positive tumor staining for prolactin in addition to GH. Mixed GH/prolactin-secreting tumors may consist of two separate cell populations, or both hormones may be synthesized and secreted by the same cell (20, 21, 22). Recently, Andersen et al. (22) reported that during several years of follow-up, three of 78 patients with prolactinomas developed clinical and biochemical acromegaly, and an additional five developed elevated IGF-I levels with no clinical signs of acromegaly. All of the reported patients had normal IGF-I levels at the time of presentation. The authors postulated a process of acquisition of GH-producing characteristics by the neoplastic prolactin-secreting cells and proposed annual measurement of IGF-I level in patients with prolactinoma as a screening test for development of acromegaly. Detection of elevated IGF-I levels in these patients may lead to a change in management and early surgical intervention.
In conclusion, clinically silent somatotroph adenoma is an infrequent clinical entity that should be actively searched for in otherwise asymptomatic patients with pituitary tumors, especially prolactinomas. IGF-I level is an appropriate screening test in these circumstances. Patients with known prolactinomas should have their IGF-I levels checked intermittently to detect mild GH overproduction. This approach will allow earlier identification and treatment of GH-secreting tumors, thus, hopefully, preventing the long-term consequences of acromegaly.
|
Footnotes |
|---|
First Published Online January 25, 2005
Abbreviations: MRI, Magnetic resonance imaging; OGTT, oral glucose tolerance test.
Received May 11, 2004.
Accepted January 14, 2005.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  S. Melmed Acromegaly N. Engl. J. Med., December 14, 2006; 355(24): 2558 - 2573. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
