| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Endocrine Care |
Service dEndocrinologie et des Maladies de la Reproduction (P.C., F.D., J.Y., G.S.), and Service de Neuroradiologie (A.B., D.D.), Centre Hospitalier Universitaire de Bicêtre, Assistance Publique-Hôpitaux de Paris, and Faculté de Médecine Paris-Sud, F-94275 Le Kremlin-Bicêtre, and Laboratoire dHistologie-Embryologie (M.K.), Faculté de Médecine Pitié-Salpêtrière, F-75013 Paris, all in France
Address all correspondence and requests for reprints to: Dr. Philippe Chanson, Service dEndocrinologie et des Maladies de la Reproduction, Hôpital Bicêtre, 78 rue du Général Leclerc, F-94275 Le Kremlin-Bicêtre, France. E-mail: pchanson{at}club-internet.fr
Abstract
Enlargement of the pituitary gland is a frequent cause of incidentaloma and of referrals to endocrinologists for hormonal evaluation and therapeutic advice. In neuroradiological series, 2550% of healthy women who are 1835 yr old have a convex superior pituitary contour, but pituitary height exceeds 9 mm in less than 0.5% of cases.
This study was performed to provide thorough clinical and hormonal data and long-term endocrinological and imaging follow-up data on subjects with incidentally discovered pituitary hypertrophy (height > 9 mm). Seven eugonadal nulliparous women, 1527 yr old, referred between 1989 and 1998 with incidentally diagnosed pituitary gland enlargement (height > 9 mm) and a suspected pituitary tumor, were studied. At presentation and at yearly intervals, PRL plasma levels and corticotropic, somatotropic, and thyrotropic pituitary function were measured; and pituitary dimensions and signal on magnetic resonance imaging (MRI), before and after iv gadolinium-diethylene-triamine-pentaacetic acid injection, were assessed.
PRL plasma levels were normal; and corticotropic, somatotropic, and thyrotropic pituitary function was considered normal in all cases. In all the women, the upper boundary of the pituitary was convex, on MRI, and touched the optic chiasm in four cases. The width and anteroposterior diameter of the gland were normal. The pituitary itself seemed normal, with a homogeneous signal, on plain and dynamic studies with iv contrast injection. Despite normal initial hormone values, two women underwent surgery, by the transsphenoidal approach, in another center. During surgery, the pituitary seemed normal in both cases, with no evidence of tumoral or inflammatory processes. Biopsy specimens showed the morphologic characteristics of a normal, nonhyperplastic pituitary gland. All seven women were seen at yearly intervals for 28 yr (median, 4 yr). Clinical and hormonal status remained stable, as did the structure and size of pituitary, on serial MRI. No tumor formation occurred, supporting the diagnosis of physiologic hypertrophy of the pituitary gland.
In conclusion, these observations suggest that careful examination of MRI results may help to distinguish physiologic pituitary hypertrophy from pituitary tumors and infiltrating lesions. The former diagnosis is confirmed by normal baseline pituitary function in extensive hormonal tests. Correct identification of such patients is important to avoid unnecessary pituitary surgery and costly MRI surveillance.
IMPROVEMENTS IN DIAGNOSTIC imaging techniques over the last several years have led to an increasing recognition of asymptomatic lesions in the pituitary (1, 2). The management of such pituitary incidentalomas is controversial (3, 4, 5, 6, 7, 8, 9, 10). Some lesions may increase in size, owing to impaired pituitary hormone production or compressed optic chiasm, whereas others will remain unchanged in size and will never produce tumoral or hormonal symptoms. The type of initial endocrinological assessment and the required frequency and length of follow-up for pituitary incidentaloma need to be carefully determined, taking cost effectiveness into account. In this setting, it is crucial to obtain precise information about the natural history of each type of pituitary lesion.
Besides pituitary masses, physiological enlargement of the pituitary gland is a frequent cause of incidentaloma and of referrals to endocrinologists for hormonal evaluation and therapeutic advice. Sex- and age-dependent variations in size and shape of the pituitary have been reported. In neuroradiological series, 2550% of healthy women, 1835 yr old, had a convex superior pituitary contour, but pituitary height exceeds 9 mm in less than 0.5% of cases (11, 12, 13, 14). No thorough clinical and hormonal data or long-term endocrinological and imaging follow-up data are available on these subjects.
The present study was thus designed to obtain more information, at the
time of initial referral and during follow-up, on the clinical,
hormonal, and neuroradiological characteristics of young women with
incidentally diagnosed pituitary gland enlargement (height
9
mm) and a suspected pituitary tumor, with the aim of proposing
practical guidelines.
Subjects and Methods
Subjects
Seven nulliparous young women, 1527 yr old, were referred to
our endocrinology unit, between 1989 and 1999, with incidentally
diagnosed pituitary gland enlargement (height
9 mm) and a
suspected pituitary tumor. The initial neuroradiological diagnosis was
always that of a pituitary mass, found during either computed
tomography (CT) or magnetic resonance imaging (MRI) of the head. None
of these women had signs of pituitary disease before the
examination.
The reasons that led to MRI or CT were syncope or suspected epilepsy (n = 3) or nonspecific headaches (n = 4). In all but one case, the first neuroradiological examination was by CT, which was followed by MRI. Two women (nos. 2 and 3) were directly referred to a neurosurgeon and underwent surgery, by the transsphenoidal approach, in another center, after a minimal hormonal workup. They were subsequently followed in our center.
These 7 women were part of a total series of 41 incidental pituitary masses, corresponding to clinically nonfunctioning pituitary adenomas (n = 20) but also, less frequently, by other pituitary tumors or cysts (Rathke cleft cysts, prolactinomas, craniopharyngiomas, and others).
Baseline and follow-up assessment
At baseline, every 6 months during 2 yr and yearly thereafter, each woman underwent an extensive workup including physical examination, hormonal studies, and pituitary MRI examination. The protocol was approved by our local institutional review board. Informed consent was obtained from all patients.
Neuroradiological studies. High-resolution CT of the pituitary fossa, with 1.2-mm sections, was performed, after iv administration of contrast material, with third-generation scanning machines. Reformatting of multiple axial images and thin-section coronal scans was obtained in all subjects undergoing CT scan.
MRI studies were performed on a 1.5 Tesla Magnetom Vision-Siemens device (Siemens SA, Saint-Denis, France). Thin (3-mm) slices through the sella in the coronal and sagittal views were obtained in T1-weighted spin-echo pulse sequences, 200 x 256 matrix and a 13-cm field of view; and coronal images were obtained in T2-weighted turbo-spin-echo pulse sequences with a 238 x 512 matrix and a 26-cm of field of view, before and after administration of gadolinium-diethylene-triamine-pentaacetic acid. Dynamic sequential imaging was performed in three cases, with images acquired every 30 sec, both in the coronal and sagittal views. T1-weighted 3-mm coronal and sagittal slices were acquired after administration of paramagnetic contrast medium.
Hormonal studies. Hormonal studies consisted of the tests usually performed in our center when a patient with a clinically nonfunctional pituitary mass is referred. For each pituitary function, this hormonal assessment aims to rule out a pituitary hormone excess attributable to a clinically silent adenoma, and a pituitary hormone deficiency attributable to a pituitary tumor (or pituitary infiltrative disease) with mass-effects hormonal consequences.
Somatotropic axis.GH hypersecretion was evaluated by measuring plasma GH levels during an 8-h diurnal profile, and/or oral 75-g glucose tolerance test; insulin-like growth factor-I (IGFI) levels were also determined. Acromegaly was ruled out when at least one of the nine GH serum levels measured during the profile was less than 0.4 µg/L and/or when GH levels during OGTT fell below 1 µg/L while sex- and age-adjusted IGFI levels were normal (15).
GH deficiency was investigated using stimulation tests with insulin or arginine or ornithine. Somatotropic function was considered normal when peak GH levels during the stimulation test exceeded 10 µg/L (16, 17).
PRL.PRL levels were assessed at baseline (normal < 23 µg/L in women) and after iv administration of 200 µg TRH.
Gonadotropic axis.All the women were free of estrogen and
progestin treatment during these investigations, and all were evaluated
during the early follicular phase. LH and FSH were measured at baseline
and after a GnRH test (100 µg) iv. Free
-subunit was also
measured at baseline and after TRH and GnRH administration. Plasma
estradiol, testosterone, and
-4-androstenediol were also
measured (18). To determine whether normal cycles were
ovulatory, estradiol and progesterone levels were measured once between
days 21 and 23 of the cycle.
Corticotropic axis.Excess ACTH secretion was evaluated by 4-hourly cortisol and ACTH measurements for 24 h and/or by a low-dose dexamethasone test. Cortisol secretion was considered normal when the midnight cortisol level was less than 1.8 µg/dL (50 nmol/L) (19) and/or when plasma cortisol was suppressed to less than 1.8 µg/dL (50 nmol/L) after low-dose dexamethasone (20). Corticotropic deficiency was studied by means of a metyrapone test or an insulin tolerance test. The response to the metyrapone test was considered normal when 11-desoxycortisol levels increased above 10 µg/L; the response to the insulin tolerance test was considered normal when plasma cortisol levels increased to above 20 µg/dL (550 nmol/L) (16).
Thyrotropic function.Thyrotropic function was assessed by measuring free T4, free T3, and TSH plasma levels. A TRH test (100 µg iv) was also performed.
Histological studies
Pituitary specimens were fixed in buffered formol (10%) and embedded in paraffin wax. Four-micrometer sections were performed and stained according to the following methods used in routine: Herlants tetrachrome, periodic acid-Schiff orange, and Wilder silver technique. Immunocytochemical studies of the samples were done as previously described (18, 21, 22).
Assays
Plasma GH levels were assayed with a commercial
immunoradiometric assay kit (23). Plasma IGFI levels were
measured as described elsewhere (24). Cortisol,
11-desoxycortisol, and ACTH were measured using commercial kits. LH and
FSH were measured using commercial kits. Free
-subunits were
measured with an immunoradiometric assay using two monoclonal
antibodies (Immunotech, Marseille, France)
(18). PRL levels were measured using commercial kits.
Results
Baseline clinical and hormonal studies (Table 1
)
The seven young, nulliparous women were eugonadal; five had
ovulatory cycles and two had oligospaniomenorrhea (nos. 5 and 7). These
latter two women had a polycystic ovary syndrome as shown by mild
hirsutism, increased basal plasma levels of LH with an ample response
to GnRH administration, moderately increased levels of testosterone
(0.49 ng/mL and 1.1 ng/mL, respectively) and
-4 androstenedione (2.2
ng/mL and 3.9 ng/mL, respectively), and enlarged ovaries on
ultrasonography (50 and 49 mm largest diameter in patient no. 5 and
45 x 41 mm in patient no. 7); both women had multiple follicles
less than 4 mm in diameter) (25, 26). Progestin
administration for 10 days induced normal menstruation in both women.
Estradiol and PRL plasma levels were in the normal range in every
case.
|
Neuroradiological studies
In all the women, pituitary height exceeded 9 mm (Figs. 1
and 2
).
The superior edge of the gland showed convexity in every case and
touched the optic chiasm in four cases. Visual fields studied with
a Goldmann perimeter in these four women were normal in two cases or
showed very partial defects in two (both normalized their visual fields
during follow-up: one postoperatively, the other spontaneously, as
assessed by repeated examination). The width and anteroposterior
diameter of the gland were normal (1014 mm) in six of seven patients
and increased (16 mm) in one patient. The pituitary itself appeared
normal, with a homogeneous signal that was isointense to gray matter on
plain sequences (Figs. 1
and 2
). Dynamic studies with iv contrast agent
showed homogenous uptake with a normal pattern and sequence (Fig. 3
). No abnormal foci were found. The
posterior pituitary always yielded the typical focal area of high
signal intensity posteriorly on T1-weighted MRI sequences.
|
|
|
Despite normal initial hormone values, the two women (nos. 2 and 3) who had been referred directly to a neurosurgeon with a false diagnosis of clinically nonfunctioning pituitary adenoma underwent surgery, by the transsphenoidal approach, in another center. During surgery, the pituitary seemed normal in both cases, with no evidence of tumoral or inflammatory processes.
At optical examination in both cases biopsy specimens showed the
morphologic characteristics of a normal, nonhyperplastic pituitary
gland. The cordonal pattern was perfectly preserved (Wilder technique),
with a well developed reticulin fiber network around each acinus (Fig. 4
). There was no striking increase in
diameter of the pituitary cords so that no hyperplastic modification
could be evoked. All the different cellular phenotypes were also
present (Herlants tetrachrome, APS-orange G), more or less numerous,
according to the observed area. These staining results were also
confirmed by immunocytochemical studies using all the available
pituitary hormone antibodies.
|
|
All seven women were seen at yearly intervals for 28 yr. Clinical and hormonal status remained stable, as did the structure and size of pituitary, on repeat MRI. No tumor formation occurred, supporting the diagnosis of physiologic hypertrophy of the pituitary gland.
Discussion
With the increasing use of modern imaging techniques, more and more abnormalities of the pituitary gland are being found in asymptomatic patients (so-called pituitary incidentalomas) (8, 29). This is not surprising, given the high frequency of pituitary lesions (particularly pituitary microadenomas but also Rathke cleft cysts) in unselected autopsy specimens (1, 2, 5). However, pituitary lesions are not limited to microlesions less than 10-mm diameter. Several studies have reported incidental discovery of macroadenomas and other pituitary tumor masses in asymptomatic patients (3, 4, 7). In our experience, another reason for referring patients with pituitary incidentalomas to endocrinologists is the discovery of pituitary enlargement, which ultimately proves to be caused by physiologic hypertrophy. The peculiar neuroradiologic aspect of this entity differs from that of tumoral masses and infiltrative processes; when repeated hormone assays are normal in such patients, unnecessary surgery can be avoided, and a final diagnosis of normal pituitary variation can be made. In two of our subjects, who were initially referred to a neurosurgeon for suspected pituitary tumors, histological documentation of pituitary normality was provided by biopsy performed via transsphenoidal route. In both cases, the pituitary seemed normal, with a normal distribution of adenohypophyseal cell types and normal electron microscopic features, ruling out tumoral or infiltrative processes (30). It also must be emphasized that neither of these patients had histological characteristics of pituitary hyperplasia (28, 31). It is important to stress, however, that in such small fragments of surgical specimens, subtle lesions could have been missed.
The height of the normal pituitary in neuroradiologic series ranges from 39 mm in healthy women between 15 and 30 yr old (11, 12, 13, 14, 32, 33, 34, 35, 36, 37). In 10% of normal women, pituitary height exceeds 7 mm. By combining the data of three large neuroradiological series (more than 600 healthy subjects), pituitary height was more than 7 mm in 6.2%, more than 8 mm in 1.1%, and more than 9 mm in 0.5% (11, 12, 14). Such images are generally seen in adolescent girls (between 10 and 19 yr old) and in menopausal women. Thus, the first conclusion that may be drawn from our data is that such a neuroradiological aspect may be encountered not only in adolescent girls but also in older women, because the pituitary measured 12 mm in height and 12 mm in width in a 24-yr-old woman. Careful measurement of the sella turcica ruled out an artifactual increase in pituitary size, relative to the sella turcica [the normal-sized pituitary extends beyond the borders of the sella turcica (33, 38)], in every case.
Pituitary hyperplasia has been reported in physiological circumstances such as pregnancy (39, 40) and also in pathological conditions such as hypothyroidism and CRH or GHRH hypersecretion (27, 41, 42). In our seven subjects, however, serial hormonal evaluations and long-term clinical outcome were both normal. Moreover, the histological aspect of pituitary specimens obtained at surgery in two women were not consistent with hyperplasia (28, 31, 43).
This increase in pituitary size, which, in this series, was limited to females, is puzzling. As regards the age of the patients, it was tempting to suggest an involvement of gonadotroph cells whose activity is supposed to be increased in the peripubertal period. However, from a histological point of view, the size, number, and aspect of gonadotroph cells was normal. Importantly, serial MRI examination during follow-up showed no further change in pituitary size. This points to the existence of normal variations in pituitary size, which persist over time and are not attributable to transient hormonal events such as puberty (12, 13, 44).
In conclusion, physiological pituitary hypertrophy seems to be a frequent cause of incidentaloma and of referrals to endocrinologists for hormonal evaluation and therapeutic advice. Careful examination of MRI images may help to distinguish physiologic pituitary hypertrophy from pituitary tumors and infiltrating lesions. The former diagnosis is confirmed by normal baseline pituitary function in extensive hormonal tests. Correct identification of such patients is important to avoid unnecessary pituitary surgery and costly MRI surveillance. We consider that pituitary enlargement (pituitary height > 9 mm), incidentally diagnosed in a young woman or adolescent girl, should be considered as normal pituitary hypertrophy when the MRI aspect is normal (gland homogeneity on plain and contrast-enhanced images) and when extensive hormonal tests are also normal. Hormonal and frequent neuroradiological follow-up seems to be unnecessary in such cases.
Acknowledgments
We thank Prof. Jacqueline Mikol for performing electron microscopy studies and Dr. Clément Iffenecker for help in the preparation of the manuscript.
Received January 19, 2001.
Revised March 19, 2001.
Accepted March 26, 2001.
References
This article has been cited by other articles:
![]() |
O M Dekkers, S Hammer, R J W de Keizer, F Roelfsema, P J Schutte, J W A Smit, J A Romijn, and A M Pereira The natural course of non-functioning pituitary macroadenomas Eur. J. Endocrinol., February 1, 2007; 156(2): 217 - 224. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Salenave, B. Gatta, S. Pecheur, F. San-Galli, A. Visot, P. Lasjaunias, P. Roger, J. Berge, J. Young, A. Tabarin, et al. Pituitary Magnetic Resonance Imaging Findings Do Not Influence Surgical Outcome in Adrenocorticotropin-Secreting Microadenomas J. Clin. Endocrinol. Metab., July 1, 2004; 89(7): 3371 - 3376. [Abstract] [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 |