| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Division of Clinical and Molecular Endocrinology (D.H.Z., B.M.A.) and Department of Neurological Surgery (W.R.S.), University Hospitals of Cleveland and Case Western Reserve University, Cleveland, Ohio 44106
Address all correspondence and requests for reprints to: Dr. Baha M. Arafah, Division of Clinical and Molecular Endocrinology, University Hospitals of Cleveland, 11100 Euclid Avenue, Cleveland, Ohio 44106. E-mail: baha.arafah{at}case.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
A recent study conducted at our institution provided additional evidence supporting the postulated pathophysiological mechanism for hypopituitarism in patients with macroadenomas (4). The study elucidated the dominant role of increased intrasellar pressure in the pathophysiology of hypopituitarism in patients with pituitary macroadenomas (4). More specifically, it was demonstrated that intrasellar pressure (ISP) was increased, particularly in patients with macroadenomas, hypopituitarism, and headaches (4). The strong correlation between serum prolactin (PRL) levels and the measured ISP provided additional support for the dominant role of increased ISP in the pathophysiology of hypopituitarism in patients with macroadenomas (4). Thus, tumor growth within the confines of the rigid sellar wall results in increased ISP, decreased blood flow through the portal vessels, and consequently decreased tissue perfusion pressure. Diminished blood flow through the portal vessels leads to decreased delivery of hypothalamic hormones to the anterior pituitary and also to areas of ischemic necrosis in regions where blood supply has reached critical limits because of limited arterial sources.
Pituitary tumor growth is generally slow and is associated with continuous modulation of the sella turcica resulting in a progressive increase in its size. The slow rate of tumor growth would not only allow the sella turcica to expand gradually, but would also probably lead to increased arterial blood supply to the anterior pituitary (8, 9). These postulated mechanisms explain the maintenance of pituitary cell viability despite a reduction in portal blood flow. However, the previously described adaptive mechanisms that were postulated to be present in patients with slowly growing macroadenomas would not be applicable if there was a rapid increase in intrasellar contents and presumably a sudden rise in ISP. Pituitary tumor apoplexy represents an example where hemorrhagic infarction of a pituitary adenoma would result in a rapid increase in intrasellar contents (10). Neurosurgeons often describe extrusion of intrasellar contents under pressure in patients who had surgery shortly after an apoplectic event (10). There are no studies in the literature investigating ISP measurements in patients with pituitary tumor apoplexy shortly after the onset of symptoms.
In the current investigation we determined ISP measurements in patients with pituitary tumor apoplexy who had surgical exploration within 1 wk of the onset of the episode. We postulated that ISP would be markedly elevated shortly after pituitary tumor apoplexy. We also postulated that the sudden rise in ISP would often result in hypopituitarism and variable degrees of ischemic necrosis of the anterior pituitary. Using serum PRL concentrations as a marker of anterior pituitary cell viability, we postulated that a normal or elevated serum PRL level at presentation would predict recovery or maintenance of pituitary function in patients with pituitary tumor apoplexy.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Hypopituitarism was defined as partial or complete loss of the hypothalamic-pituitary-gonadal, adrenal, or thyroidal function. GH secretion was evaluated in some, but not all, patients and was not used in defining hypopituitarism. Testing procedures and definitions used at our institution in the evaluation of hypothalamic pituitary function were previously described (1, 3, 4, 10, 11). However, because of the critical nature of the patients’ presentation and the urgent need to administer glucocorticoids, only baseline hormone levels were obtained on admission. At presentation, none of the patients was receiving hormone replacement therapy or receiving medications known to influence PRL levels or pituitary-thyroidal, adrenal, or gonadal function. Two patients who had no clinical or biochemical evidence of preoperative ACTH deficiency (defined arbitrarily as a baseline serum cortisol at presentation of more than 15.0 µg/dl or 413.8 nmol/liter) were each given one dose of hydrocortisone (50 mg, iv) just before surgery. Thereafter, they did not receive glucocorticoid replacement at any time. These two patients had surgical decompression within a few hours after admission. Preoperative treatment with glucocorticoids (hydrocortisone, 25 mg, iv, every 6 h) was started after baseline hormone levels were obtained in the remaining 11 patients. All 13 patients underwent transsphenoidal decompressive surgery using standard procedures within 1–2 d of admission. After measurements of ISP were made, necrotic tissue and residual tumor specimens were removed and sent for pathological confirmation. The institutional review board approved the study, and informed consent was obtained from all patients.
Surgical/pathological findings
The diagnosis of pituitary tumor apoplexy was confirmed at surgery in all patients. Pathological specimens were available for nine patients, whereas only necrotic tissues, which could not be further characterized, were available for the remaining four. Immunostaining of the nine adenomas that had viable tissues revealed cells with variable degrees of positive immunostaining for FSH/LH/
-subunit in three, GH/FSH/LH in two, and negative staining in the remaining four tumors.
Management in the perioperative period
Preoperatively, the following serum hormone levels were determined before glucocorticoids were administered: cortisol, free T4, TSH, FSH, LH, testosterone (in males), estradiol (in premenopausal women), and PRL. In addition, plasma IGF-I levels were determined preoperatively in five patients. After their initial evaluation, patients received supportive measures including iv fluids, glucocorticoids, and close monitoring. As stated above, each of the two patients, who were presumed to have normal adrenal function at presentation, received a single dose of hydrocortisone just before surgery. Thereafter, no glucocorticoids were administered to these two patients. In the remaining 11 patients, glucocorticoid therapy was administered preoperatively (25 mg hydrocortisone, iv, every 6 h) and continued for 24 h after surgery. At that time hydrocortisone therapy was withdrawn while patients were carefully monitored in the intensive care unit for signs and symptoms of adrenal insufficiency (10, 11, 12). After glucocorticoid therapy was discontinued, blood samples were drawn three times daily for serum cortisol measurements. The timing of these measurements included the morning (0600–0800 h), midafternoon (1400–1700 h), and early evening (1800–2000 h) hours. Glucocorticoids were reinstituted in symptomatic patients and/or those with documented low (<5.0 µg/dl or 138.0 nmol/liter) serum cortisol levels measured 24 h or more after glucocorticoids were withdrawn. The safety of this approach has been demonstrated repeatedly since the original report was published (3, 4, 10, 11, 12). A serum cortisol level of more than 15.0 µg/dl (413.8 nmol/liter) measured 24 h or more after hydrocortisone therapy was discontinued was considered an indication of normal hypothalamic-pituitary-adrenal axis, whereas a level less than 5.0 µg/dl (138.0 nmol/liter) indicated persistent ACTH deficiency. Patients with serum cortisol levels between 5.0 and 15.0 µg/dl (138.0 and 413.8 nmol/liter) were given hydrocortisone replacement if they had symptoms, whereas those who were asymptomatic were discharged without replacement. In the latter group of patients, reassessment of pituitary adrenal function was performed 6–12 wk after surgery, or sooner if they had symptoms.
Postoperative evaluation of pituitary function
Postoperative evaluation of pituitary function was performed 6–12 wk after surgery as previously outlined (1, 3, 4, 11, 12). A normal pituitary adrenal function was defined as a baseline (not stimulated) serum cortisol level of 15.0 µg/dl (413.8 nmol/liter) or more and/or a normal age- and gender-adjusted serum dehydroepiandrosterone sulfate level (13). Additional testing was performed in patients whose baseline levels did not fulfill the latter criteria. Such testing included insulin-induced hypoglycemia and/or low dose (1 µg) Cortrosyn (Amphastar Pharmaceuticals, Rancho Cucamonga, CA) stimulation test (13). A serum cortisol level of more than 21.0 µg/dl (579.4 nmol/liter) was considered a normal response to the low dose Cortrosyn test. The latter test was not used as the sole determinant of the normality of the axis. The pituitary-thyroid axis was evaluated by measurements of free T4 and TSH serum levels. Baseline serum concentrations of gonadotropins (FSH and LH), total and free testosterone (in men) and estradiol (in premenopausal women) were determined to assess pituitary-gonadal function. Baseline serum PRL levels were determined in all patients. Plasma IGF-I levels were determined in nine patients, whereas serum GH levels were measured only in the five patients who had insulin-induced hypoglycemia postoperatively.
ISP measurements
As previously described (4), we used the OLM-Intracranial Pressure Monitoring Kit (Camino Laboratories, San Diego, CA) to determine ISP. The kit uses a fiber optic transducer connected to a pressure monitor. The transducer is located at the tip of a catheter that is 1.3 mm in diameter. At transsphenoidal surgery, a portion of the floor of the sella was removed, and an approximately 2-mm dural opening was made to allow placement of the catheter without extravasation of intrasellar contents. The tip of the transducer was inserted into the tissues of the pituitary fossa. Thirty to 60 sec later and after a stable waveform was obtained, the mean ISP was recorded, and the transducer was removed. ISP measurements could not be determined in three patients, in whom necrotic tissue and blood issued forth immediately after the dura was opened.
Statistical analysis
Baseline demographic and clinical variables between groups were analyzed using the Mann-Whitney test for continuous outcomes and exact Fisher tests for categorical outcomes. Statistical differences in serum PRL levels and ISP measurements between groups were determined using the Mann-Whitney test. The data are presented as the mean ± SD unless stated otherwise.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
The 13 patients included in the study were 25–80 yr of age (mean ± SD. 52.5 ± 17; median, 50 yr). There were seven males and six females in our study group. All patients had macroadenomas (mean size, 2.3 ± 0.4 cm; range, 1.5–3.5 cm; median, 2.5 cm).
Baseline serum cortisol concentrations were less than 5.0 µg/dl (138.0 nmol/liter) in seven patients, between 5.0–15.0 µg/dl (138.0–413.8 nmol/liter) in four patients, and greater than 15.0 µg/dl (413.8 nmol/liter) in two patients (17 and 25 µg/dl or 469.0 and 689.8 nmol/liter). The latter two patients were considered to have a normally functioning pituitary-adrenal axis even though dynamic tests were not performed. Five patients had low serum free T4 levels and clinical features suggestive of hypothyroidism; indicating that hypothyroidism preceded the apoplectic episode. All 13 patients had clinical and/or biochemical evidence of impairment of pituitary-gonadal function. It is possible, though unlikely, that the stress of the presentation and the acute illness might have contributed to the observed impairment in gonadal function in some patients. Plasma IGF-I levels were low (2.1–3.3 SD below age- and gender-adjusted mean values) in all five patients who had the measurements preoperatively.
Serum PRL levels in these 13 patients were generally low at presentation. The levels ranged from 0.8–22.0 µg/liter, with a mean of 4.1 ± 7.3 µg/liter and a median of 3.5 µg/liter. Serum PRL levels were less than 5.0 µg/liter in eight of 13 patients, 5.0–15.0 µg/liter in two of 13, and more than 15.0 µg/liter in three of 13.
ISP measurements and their correlation with serum PRL levels
Mean ISP measurements were markedly elevated. The measurements raged from 25–58 mm Hg, with a mean (±SD) of 44.1 ± 11.1 and a median of 47 mm Hg. In these 13 patients, there was a significant inverse correlation (r = –0.76; P < 0.01) between ISP measurements and the serum PRL levels obtained at presentation (Fig. 1
).
|
The ISP and serum PRL levels data obtained in the current investigation were compared with similar data reported previously (4) in a group of patients who did not have tumor apoplexy, but had hypopituitarism and macroadenomas of similar sizes. The comparison data are shown in Table 1. The mean ISP in patients with pituitary tumor apoplexy (44.1 ± 11.1 mm Hg) was higher (P = 0.004) than that in patients with hypopituitarism without tumor apoplexy (33.6 ± 13.4 mm Hg).
|
) than those observed in the 13 patients with apoplexy (4.1 ± 7.3 µg/liter) reported herein.
The correlation between mean ISP measurements and serum PRL levels in these two groups of patients were distinctly different. Whereas mean ISP correlated positively with serum PRL concentrations in patients with macroadenomas and hypopituitarism, there was an inverse correlation between these variables in patients with tumor apoplexy (Fig. 1).
Postoperative pituitary function in patients with tumor apoplexy
Pituitary function in the 13 patients with pituitary tumor apoplexy was evaluated 6–12 wk after surgery. Seven of the 13 patients had normal adrenal function, as defined above (Table 2). The remaining six patients had persistent ACTH deficiency. Serum free T4 levels were normal in six patients, including two of the five with preoperative hypothyroidism and four of the eight who had normal serum free T4 levels before surgery. Gonadal function normalized in two of the seven men and one of the six women with abnormal preoperative function.
|
The seven patients who had partially maintained or normalized pituitary function postoperatively had serum PRL levels that were greater than 3.5 µg/liter before surgery and had a median ISP of 37 mm Hg (Table 3). In contrast, the remaining six patients with persistent or worsening hypopituitarism had lower (P = 0.013) serum PRL levels and higher (P < 0.001) ISP measurements than the respective values in the seven patients who partially recovered or maintained normal pituitary function (Table 3). Serum PRL levels in the six patients with persistent hypopituitarism were all less than 2.5 µg/liter, and their median ISP was 57 mm Hg.
|
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Normal intrasellar pressure is not known, but is believed to be similar to or less than the normal intracranial pressure of 7–15 mm Hg (15, 16). Although direct measurements of intrasellar pressure in normal subjects have not been reported, indirect evidence suggests that the predicted values are likely to be accurate. In a study by Lees et al. (17), six patients with the empty sella syndrome had a mean ISP of 13.5 ± 3 mm Hg. Similarly, the same study reported that seven patients with microadenomas (<5 mm) had a mean ISP of 12 ± 3 mm Hg.
Patients with pituitary macroadenomas and hypopituitarism have higher ISP than those without hypopituitarism (4). In our previous study involving patients who underwent transsphenoidal surgery for pituitary adenomas, ISP measurements were higher in the subgroup of patients who had hypopituitarism than in those without hypopituitarism (33.6 ± 13.4 vs. 19 ± 7.3 mm Hg, respectively) (4). Using slightly different techniques, similar ISP measurements were previously reported in hypopituitary patients (14, 15, 16, 17, 18). There are no studies in the literature that specifically addressed intrasellar pressure measurements in patients with pituitary tumor apoplexy. Interestingly, in their study of 48 patients with pituitary tumors, Kruse et al. (16) included one patient with pituitary tumor apoplexy who had a mean ISP of 60 mm Hg. This value is remarkably similar to those observed in our patients.
In an earlier study, Lees et al. (19) studied blood flow in relation to tumor growth in the estrogen-induced pituitary tumor model in rats. The researchers found that there was a decrease in blood flow that was associated with progressive tumor growth until alternative routes of blood supply were developed (19). Other studies demonstrated that the pituitary portal circulation could be compromised by increases in surrounding pressure (16, 20). In that respect, Antunes et al. (20) demonstrated that complete cessation of blood flow through the portal vessels of rhesus monkey was achieved by a Valsalva maneuver, where venous pressure was raised to 30 cm of H2O. Similar observations were made by Kruse et al. (16), who measured blood flow to the pituitary in humans during surgical removal of pituitary adenomas. The researchers demonstrated a rise in ISP associated with dramatic decrease or even complete arrest of blood flow to the pituitary when hemorrhage within the adenoma was inadvertently induced intraoperatively (16). In the latter study by Kruse et al. (16), an increase in ISP to 32, 40, and 50 mm Hg in three patients with pituitary tumors resulted in a marked decrease or even arrest of anterior pituitary blood flow. The dramatic decrease in blood flow to the anterior pituitary leads to ischemia and cell necrosis. Because these cells cannot regenerate, their loss is likely to lead to permanent hypopituitarism, as was postulated previously and observed in six of the patients in the current investigation.
Unlike other anterior pituitary hormones, PRL is under dominant negative hypothalamic control. Mild hyperprolactinemia is frequently observed in patients with non-PRL-secreting adenomas and hypopituitarism (1, 2, 3, 4, 5). This phenomenon was postulated to reflect the role of compression of portal vessels in the pathophysiology of hypopituitarism. Furthermore, the presence of normal or increased serum PRL level in patients with non-PRL-secreting adenomas had been considered a clear indication for the presence of viable anterior pituitary tissue (1, 4, 21). Thus, it is not surprising to note that in patients with non-PRL-secreting adenomas and hypopituitarism, higher serum PRL levels predicted recovery of pituitary function after surgery (1, 3). The current investigation indicates that similar findings were observed in patients with pituitary tumor apoplexy. Specifically, patients with apoplexy whose serum PRL levels were less than 2.5 µg/liter on admission were unlikely to recover or maintain any pituitary function after urgent surgery, whereas those who had PRL levels greater than 3.5 µg/liter did partially recover pituitary function.
In contrast to the gradual increase in intrasellar pressure caused by the slow, but steady, growth of pituitary adenomas, tumor apoplexy results in a rapid increase in intrasellar contents and, consequently, a sudden rise in ISP. The rapid and sharp increases in ISP compromise anterior pituitary cell viability and result in a greater likelihood of ischemic necrosis. Although hyperprolactinemia was caused by interruption of the delivery of hypothalamic inhibitory factors to viable pituitary cells, we believe that cell viability in that setting was probably maintained by arterial blood supply. In contrast, the prevalence of hypoprolactinemia in patients with tumor apoplexy reflected the greater role played by ischemic necrosis of anterior pituitary cells. The sudden and extreme elevation in ISP without sufficient arterial blood supply in many patients with tumor apoplexy compromised cell viability and limited the potential for recovery of pituitary function after surgical decompression.
In summary, our study demonstrates that ISP measurements were markedly elevated in patients with pituitary tumor apoplexy and correlated negatively with serum PRL levels measured at presentation. These findings suggest that the rapid increase in ISP may lead to ischemic necrosis of portions of the anterior pituitary in many patients. Furthermore, a very low serum PRL level at presentation reflects significant necrosis of the normal pituitary tissue and predicts permanent hypopituitarism.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: ISP, Intrasellar pressure; PRL, prolactin.
Received May 11, 2004.
Accepted July 26, 2004.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  R. N. Nawar, D. AbdelMannan, W. R. Selman, and B. M. Arafah Analytic Review: Pituitary Tumor Apoplexy: A Review J Intensive Care Med, March 1, 2008; 23(2): 75 - 90. [Abstract] [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 |
). The other cohort represents hypopituitary patients who had no tumor apoplexy (
) and who were included in a previous report (