0021-972X/90/7102-0323$02.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1990 by The Endocrine Society
Vol. 71, No. 2 Printed in U.S.A.
Improvement of Pituitary Function after Surgical Decompression for Pituitary Tumor Apoplexy* BAHA M. ARAFAH, J. FREDRICK HARRINGTON, ZUHAYR T. MADHOUN, AND WARREN R. SELMAN Divisions of Endocrinology and Neurological Surgery, Case Western Reserve University School of Medicine and University Hospitals of Cleveland, Cleveland, Ohio 44106
ABSTRACT. Hypopituitarism is a major manifestation of pituitary adenoma apoplexy. We postulated that the acute deterioration in pituitary function may be caused by compression of portal vessels and the pituitary stalk, secondary to a sudden increase in intrasellar contents. If this were the case, one would predict improvement in pituitary function to occur after surgical decompression. We studied pituitary function in eight patients presenting with the clinical syndrome of pituitary adenoma apoplexy before and on multiple occasions after urgent surgical decompression. Partial or complete hypopituitarism was seen in all subjects at the time of presentation. Serum cortisol levels were inappropriately low (5.9 ± 1.4 /xg/dL; 162.8 ± 38 nmol/L) for the degree of stress in seven patients and appropriately elevated in only one subject (55.3 fig/dh; 1525.7 nmol/L). High normal increments in cortisol levels were noted in three subjects given test doses of cosyntropin. Patients were given glucocorticoids before, during, and for 2 days after surgery. Serum cortisol concentrations measured on or after the third day when gluco-
corticoids had already been stopped were normal in seven subjects and consistently low in one. These seven subjects were discharged on no replacement and were subsequently documented by dynamic testing to have normal pituitary-adrenal function. Gonadal function improved in two of four men and in one of two women who had hypogonadism on presentation. Improvement in thyroid function was documented in two of three subjects with preoperative hypothyroidism. Gradual improvement and almost complete resolution of the neuroophthalmological abnormalities occurred days to weeks after decompression. These observations demonstrate that urgent surgical decompression after pituitary tumor apoplexy was associated with improvement not only in neurological defects but also in pituitary function. The rapid improvement in pituitary function indicates not only that the hypopituitarism was reversible, but also that it might be caused by compression of the portal circulation and pituitary stalk by the sudden increase in intrasellar contents. (J Clin Endocrinol Metab 7 1 : 323-328,1990).
P
ITUITARY adenoma apoplexy is a rare clinical syndrome characterized by hemorrhagic infarction of a preexisting adenoma, followed by the acute onset of headaches, vomiting, visual disturbances, and possible alteration in the level of consciousness (1-7). Published reports suggest that 1-10% of patients with pituitary adenomas may present with the acute syndrome (4-7). Although necrosis and/or hemorrhagic changes can be seen in 10-28% of pathologically examined pituitary adenomas (4), the term apoplexy is reserved for instances where symptomatic hemorrhagic infarction occurs acutely, resulting in the clinical syndrome described previously (1-7). The clinical manifestations of the syndrome relate to compression of structures in the sella turcica and the perisellar region. Hypopituitarism is a prominent finding in patients with apoplexy and may contribute to the morbidity and
mortality from this disease (8-14). The pathophysiology of hypopituitarism in this setting is not well understood, but is felt to be secondary to infarction and/or necrosis of the normal gland. However, a few case reports showed that weeks or months after the apoplectic episode, some of the normal pituitary hormone secretions can be demonstrated (11-13). The documentation of hormone secretion by the gland, months after apoplexy, suggested that the normal pituitary was not completely destroyed during the apoplectic episode. It would be difficult to discern from these reports whether normal hormone secretion was maintained during and after apoplexy or whether these subjects developed hypopituitarism that subsequently improved. Pituitary adenoma apoplexy occurs primarily in patients with preexisting macroadenomas (5, 7,9,15) where partial or complete hypopituitarism is not uncommon (16). We postulated that the hypopituitarism seen in patients with large adenomas was due largely to compression of the pituitary stalk and/or portal vessels by the expanding adenoma (16). The recovery of pituitary hormone function after selective adenomectomy in most
Received December 6, 1989. Address all correspondence and requests for reprints to: Baha M. Arafah, M.D., Division of Endocrinology and Hypertension, University Hospitals of Cleveland, 2074 Abington Road, Cleveland, Ohio 44106. * Presented in part at the 71st Annual Meeting of The Endocrine Society, June 1989, in Seattle, WA. 323
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 08 September 2015. at 03:14 For personal use only. No other uses without permission. . All rights reserved.
324
ARAFAH ET AL.
patients supported the hypothesis (16). Sudden expansion of hemorrhagic tumors may contribute to worsening of normal pituitary function. This may occur over and above the already compromised or impaired pituitary function seen in at least some patients. We postulate that hypopituitarism, which is a frequent and prominent part of the presenting symptoms of apoplexy, might be reversed by prompt removal of the necrotic tissue. To investigate this hypothesis, we studied pituitary function before and serially after urgent surgical decompression in eight patients with pituitary tumor apoplexy. Our findings indicate that improvement in pituitary function could be demonstrated in the majority of patients days after surgical decompression. Subjects and Methods Eight consecutive patients who presented with pituitary apoplexy and underwent surgical decompression were included in the study. There were four males and four females whose average ages were 47 and 41 years, respectively (Table 1). Pituitary adenoma apoplexy was precipitated by the iv injection of a test dose of GnRH in one, previously reported patient (17). In another patient, pituitary tumor apoplexy occurred 3 weeks after initiating bromocriptine therapy. In the remaining six subjects, adenomas were previously unrecognized, and apoplexy was their first clinical presentation. Patients presented 4 h to 19 days (Table 1) after the onset of symptoms. Seven complained of headaches, while one obtunded patient provided no such history. Seven patients had sudden visual field loss, and four had one or more cranial nerve palsies. Two patients had nausea, vomiting, lethargy, and hypotension as their major presenting symptoms, while three had postural dizziness, fatigue, and generalized weakness. The diagnosis of pituitary adenoma apoplexy in this series was based on the clinical presentation and the demonstration of hemorrhagic infarction of a pituitary macroadenoma by computed tomography or magnetic resonance imaging. This was verified intraoperatively and later documented by pathological examination of the resected tissues. Patients underwent transsphenoidal surgery within 4 days of admission (mean, 2.2 days). All visible tumor (hemorrhagic and nonhemorrhagic) was resected by an experienced surgeon (W.R.S.), who attempted to perform near-complete adenomectomy in all patients. There was no significant surgical morbidity or mortality. Immunocytochemical staining (GH, PRL, ACTH, TSH, FSH, and LH) was performed on multiple sections by the peroxidase-antiperoxidase method (18), as previously described (16). The results are included in Table 1. Endocrine studies Preoperatively, the following hormone levels were determined before glucocorticoids were administered: cortisol, T4, free T4 index, TSH, FSH, LH, testosterone (in males), PRL, and GH. In one patient clinically suspected and biochemically documented to have acromegaly, the somatomedin-C level was
JCE & M • 1990 Vol 71 • No 2
also measured. Three subjects had a cosyntropin stimulation test (0.25 mg, iv) performed before glucocorticoids were administered. Postoperative pituitary dynamic studies were performed 4-8 weeks after surgery. At the time of testing, patients were receiving no replacement therapy. Evaluation of the pituitary-adrenal axis was performed using the cortisol response to insulin-induced hypoglycemia (16,19). In patients who could not tolerate hypoglycemia, metyrapone (750 mg, orally, every 4 h for 24 h) was given, and the levels of cortisol and 11deoxycortisol were measured (16, 19). The pituitary-thyroid axis was evaluated by measurement of T4, free T4 index, TSH, and the rise in TSH after TRH administration (16). Basal and LHRH (100 Mg» iv)-stimulated gonadotropin (FSH and LH) concentrations were determined in addition to serum testosterone (for men) or estradiol (in women). Basal and stimulated (TRH and insulin-induced hypoglycemia) serum PRL levels were also determined (16, 20). The rise in serum GH during hypoglycemia was determined as an index of GH secretion. All pituitary dynamic tests were performed at the Clinical Research Center. The study was approved by the Institutional Review Board, and informed consent was obtained from each patient. Analytical methods All hormones were measured using established and specific RIAs at our institution (16, 17, 19, 20). Standards and antisera for PRL, GH, TSH, FSH, and LH were kindly supplied by the NIDDK. The interassay coefficient of variation for these assays was consistently below 15%. Measurements of T4, free T4 index, cortisol, testosterone, and estradiol were performed using commercially available kits. Perioperatiue management After their initial evaluation, patients received iv fluids and other supportive measures. Once the diagnosis was suspected and after blood samples were drawn, glucocorticoids [200 mg/ day hydrocortisone (n = 5) or 16 mg/day dexamethasone (n = 3)] were administered iv, and surgical decompression was performed within 3 days. Glucocorticoids were discontinued 2-4 days after decompressive adenomectomy while the patients were carefully monitored in the intensive care unit for signs and symptoms of adrenal insufficiency (19). Morning (0800 h) and evening (1800 h) serum cortisol levels were obtained for 46 days after glucocorticoids were stopped. Cortisol levels drawn 36 h or more after the last steroid dose were considered to represent endogenous cortisol production. Glucocorticoids were reinstituted in symptomatic patients and/or those with documented low serum cortisol levels (see Results). Postoperative cortisol levels in our patients were evaluated in comparison to a large number of patients treated at our institution and documented to have a normal pituitary-adrenal axis before and after adenomectomy (19). Morning or evening cortisol levels of more than 10 ng/dL (>275.9 nmol/L) were considered normal, while those below 5 Mg/dL (138 nmol/L) were low (19). Patients with equivocal values were formally tested (insulin-induced hypoglycemia) before discharge to ensure adequate cortisol secretion. Of three patients with documented hypothyroidism
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 08 September 2015. at 03:14 For personal use only. No other uses without permission. . All rights reserved.
PITUITARY TUMOR APOPLEXY at presentation, one was treated preoperatively with L-T4, and the treatment was stopped 2 weeks after surgery. Results The characteristics of the patients included in this study are shown in Table 1. Immunostaining of the resected tissue paralleled the clinical findings in five of eight patients. However, in the remaining three, discrepancies between tumor immunocytochemical staining and the clinical history were evident (patients 1, 5, and 7). This might be caused by the difficulties encountered in instances where necrosis and hemorrhage involved most of the tumor. The clinical history in all three patients was consistent with the presumptive diagnosis of prolactinomas, although the possibility of hyperprolactinemia in association with nonsecreting tumors could not be completely excluded. Preoperatiue endocrine studies Baseline hormone levels obtained before the administration of glucocorticoids showed a wide range of abnormalities in pituitary function. In only one patient was the serum cortisol level appropriately elevated (55.3 fig/dh; 1525.7 nmol/L) for the degree of stress with which they presented. In the remaining seven patients, serum cortisol ranged from 2.3-11.9 (63.5-328.3 nmol/L), with a mean ± SE of 5.9 ± 1.4 (162.8 ± 38 nmol/L). The two patients with the most severe symptoms on admission had the highest cortisol levels (8.0 and 11.9 Mg/dL; 237.3 and 328.3 nmol/L). The remaining five patients had low or low normal cortisol levels (2.3-6.0 /j.g/dL; 63.5-165.4 nmol/L). One patient who presented 19 days after the episode had a clearly low (2.7 Mg/dL; 74.5 nmol/L) morning cortisol level. All seven patients had signs and symptoms of adrenal insufficiency at the time their blood samples were drawn. The administration of cosyntropin (0.25 mg, iv) to three of these seven subjects resulted in normal increments in the serum cortisol level (22, 29, and 39 Mg/dL; 607, 800, and 1076 nmol/L, respectively). Clinical and biochemical evidence for hypothyroidism was
325
noted in three subjects who had low levels of T4 (3.2, 4.1, and 4.8 Mg/dL; 41.2, 52.8, and 61.8 nmol/L, respectively) and free T4 index (3.6, 4.4, and 4.7; normal, 5-12). Serum TSH levels in these patients were in the normal range for euthyroid subjects (0.5-3.5 mU/L). Serum testosterone and gonadotropin levels were low in all four men (Table 2) and consistent with the clinical history of impotence and diminished libido. Serum gonadotropin levels were clearly low in two women and within the normal range in the remaining two female subjects. Serum GH levels were normal (2 and 3 Mg/L) in one patient with acromegaly even though his somatomedin-C levels were evaluated (4.5 and 3.9 U/ml; normal, 0.5-2.0 U/mL). Serum PRL levels were elevated (150 Mg/dL) in only one subject (no. 6) with documented macroprolactinoma. The level was high normal in two subjects (no. 2 and 3) and low in the remaining five subjects (0.5-3 ng/dL). Galactorrhea was noted on examination in three of the five subjects with low serum PRL levels. None of the patients had signs or symptoms to suggest impairment of antidiuretic hormone secretion. Thus, at the time of their initial presentation, seven patients had glucocorticoid deficiency, three had hypothyroidism, and six had hypogonadism (Table 3). There was no correlation between the duration of symptoms and the severity or extent of the neuroophthalmological abnormalities or pituitary dysfunction. Endocrine studies in the immediate postoperative period Glucocorticoid therapy was discontinued 2-4 days after surgical decompression. Consistently low serum cortisol levels (
Vol. 71, No. 2 Printed in U.S.A.
Improvement of Pituitary Function after Surgical Decompression for Pituitary Tumor Apoplexy* BAHA M. ARAFAH, J. FREDRICK HARRINGTON, ZUHAYR T. MADHOUN, AND WARREN R. SELMAN Divisions of Endocrinology and Neurological Surgery, Case Western Reserve University School of Medicine and University Hospitals of Cleveland, Cleveland, Ohio 44106
ABSTRACT. Hypopituitarism is a major manifestation of pituitary adenoma apoplexy. We postulated that the acute deterioration in pituitary function may be caused by compression of portal vessels and the pituitary stalk, secondary to a sudden increase in intrasellar contents. If this were the case, one would predict improvement in pituitary function to occur after surgical decompression. We studied pituitary function in eight patients presenting with the clinical syndrome of pituitary adenoma apoplexy before and on multiple occasions after urgent surgical decompression. Partial or complete hypopituitarism was seen in all subjects at the time of presentation. Serum cortisol levels were inappropriately low (5.9 ± 1.4 /xg/dL; 162.8 ± 38 nmol/L) for the degree of stress in seven patients and appropriately elevated in only one subject (55.3 fig/dh; 1525.7 nmol/L). High normal increments in cortisol levels were noted in three subjects given test doses of cosyntropin. Patients were given glucocorticoids before, during, and for 2 days after surgery. Serum cortisol concentrations measured on or after the third day when gluco-
corticoids had already been stopped were normal in seven subjects and consistently low in one. These seven subjects were discharged on no replacement and were subsequently documented by dynamic testing to have normal pituitary-adrenal function. Gonadal function improved in two of four men and in one of two women who had hypogonadism on presentation. Improvement in thyroid function was documented in two of three subjects with preoperative hypothyroidism. Gradual improvement and almost complete resolution of the neuroophthalmological abnormalities occurred days to weeks after decompression. These observations demonstrate that urgent surgical decompression after pituitary tumor apoplexy was associated with improvement not only in neurological defects but also in pituitary function. The rapid improvement in pituitary function indicates not only that the hypopituitarism was reversible, but also that it might be caused by compression of the portal circulation and pituitary stalk by the sudden increase in intrasellar contents. (J Clin Endocrinol Metab 7 1 : 323-328,1990).
P
ITUITARY adenoma apoplexy is a rare clinical syndrome characterized by hemorrhagic infarction of a preexisting adenoma, followed by the acute onset of headaches, vomiting, visual disturbances, and possible alteration in the level of consciousness (1-7). Published reports suggest that 1-10% of patients with pituitary adenomas may present with the acute syndrome (4-7). Although necrosis and/or hemorrhagic changes can be seen in 10-28% of pathologically examined pituitary adenomas (4), the term apoplexy is reserved for instances where symptomatic hemorrhagic infarction occurs acutely, resulting in the clinical syndrome described previously (1-7). The clinical manifestations of the syndrome relate to compression of structures in the sella turcica and the perisellar region. Hypopituitarism is a prominent finding in patients with apoplexy and may contribute to the morbidity and
mortality from this disease (8-14). The pathophysiology of hypopituitarism in this setting is not well understood, but is felt to be secondary to infarction and/or necrosis of the normal gland. However, a few case reports showed that weeks or months after the apoplectic episode, some of the normal pituitary hormone secretions can be demonstrated (11-13). The documentation of hormone secretion by the gland, months after apoplexy, suggested that the normal pituitary was not completely destroyed during the apoplectic episode. It would be difficult to discern from these reports whether normal hormone secretion was maintained during and after apoplexy or whether these subjects developed hypopituitarism that subsequently improved. Pituitary adenoma apoplexy occurs primarily in patients with preexisting macroadenomas (5, 7,9,15) where partial or complete hypopituitarism is not uncommon (16). We postulated that the hypopituitarism seen in patients with large adenomas was due largely to compression of the pituitary stalk and/or portal vessels by the expanding adenoma (16). The recovery of pituitary hormone function after selective adenomectomy in most
Received December 6, 1989. Address all correspondence and requests for reprints to: Baha M. Arafah, M.D., Division of Endocrinology and Hypertension, University Hospitals of Cleveland, 2074 Abington Road, Cleveland, Ohio 44106. * Presented in part at the 71st Annual Meeting of The Endocrine Society, June 1989, in Seattle, WA. 323
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 08 September 2015. at 03:14 For personal use only. No other uses without permission. . All rights reserved.
324
ARAFAH ET AL.
patients supported the hypothesis (16). Sudden expansion of hemorrhagic tumors may contribute to worsening of normal pituitary function. This may occur over and above the already compromised or impaired pituitary function seen in at least some patients. We postulate that hypopituitarism, which is a frequent and prominent part of the presenting symptoms of apoplexy, might be reversed by prompt removal of the necrotic tissue. To investigate this hypothesis, we studied pituitary function before and serially after urgent surgical decompression in eight patients with pituitary tumor apoplexy. Our findings indicate that improvement in pituitary function could be demonstrated in the majority of patients days after surgical decompression. Subjects and Methods Eight consecutive patients who presented with pituitary apoplexy and underwent surgical decompression were included in the study. There were four males and four females whose average ages were 47 and 41 years, respectively (Table 1). Pituitary adenoma apoplexy was precipitated by the iv injection of a test dose of GnRH in one, previously reported patient (17). In another patient, pituitary tumor apoplexy occurred 3 weeks after initiating bromocriptine therapy. In the remaining six subjects, adenomas were previously unrecognized, and apoplexy was their first clinical presentation. Patients presented 4 h to 19 days (Table 1) after the onset of symptoms. Seven complained of headaches, while one obtunded patient provided no such history. Seven patients had sudden visual field loss, and four had one or more cranial nerve palsies. Two patients had nausea, vomiting, lethargy, and hypotension as their major presenting symptoms, while three had postural dizziness, fatigue, and generalized weakness. The diagnosis of pituitary adenoma apoplexy in this series was based on the clinical presentation and the demonstration of hemorrhagic infarction of a pituitary macroadenoma by computed tomography or magnetic resonance imaging. This was verified intraoperatively and later documented by pathological examination of the resected tissues. Patients underwent transsphenoidal surgery within 4 days of admission (mean, 2.2 days). All visible tumor (hemorrhagic and nonhemorrhagic) was resected by an experienced surgeon (W.R.S.), who attempted to perform near-complete adenomectomy in all patients. There was no significant surgical morbidity or mortality. Immunocytochemical staining (GH, PRL, ACTH, TSH, FSH, and LH) was performed on multiple sections by the peroxidase-antiperoxidase method (18), as previously described (16). The results are included in Table 1. Endocrine studies Preoperatively, the following hormone levels were determined before glucocorticoids were administered: cortisol, T4, free T4 index, TSH, FSH, LH, testosterone (in males), PRL, and GH. In one patient clinically suspected and biochemically documented to have acromegaly, the somatomedin-C level was
JCE & M • 1990 Vol 71 • No 2
also measured. Three subjects had a cosyntropin stimulation test (0.25 mg, iv) performed before glucocorticoids were administered. Postoperative pituitary dynamic studies were performed 4-8 weeks after surgery. At the time of testing, patients were receiving no replacement therapy. Evaluation of the pituitary-adrenal axis was performed using the cortisol response to insulin-induced hypoglycemia (16,19). In patients who could not tolerate hypoglycemia, metyrapone (750 mg, orally, every 4 h for 24 h) was given, and the levels of cortisol and 11deoxycortisol were measured (16, 19). The pituitary-thyroid axis was evaluated by measurement of T4, free T4 index, TSH, and the rise in TSH after TRH administration (16). Basal and LHRH (100 Mg» iv)-stimulated gonadotropin (FSH and LH) concentrations were determined in addition to serum testosterone (for men) or estradiol (in women). Basal and stimulated (TRH and insulin-induced hypoglycemia) serum PRL levels were also determined (16, 20). The rise in serum GH during hypoglycemia was determined as an index of GH secretion. All pituitary dynamic tests were performed at the Clinical Research Center. The study was approved by the Institutional Review Board, and informed consent was obtained from each patient. Analytical methods All hormones were measured using established and specific RIAs at our institution (16, 17, 19, 20). Standards and antisera for PRL, GH, TSH, FSH, and LH were kindly supplied by the NIDDK. The interassay coefficient of variation for these assays was consistently below 15%. Measurements of T4, free T4 index, cortisol, testosterone, and estradiol were performed using commercially available kits. Perioperatiue management After their initial evaluation, patients received iv fluids and other supportive measures. Once the diagnosis was suspected and after blood samples were drawn, glucocorticoids [200 mg/ day hydrocortisone (n = 5) or 16 mg/day dexamethasone (n = 3)] were administered iv, and surgical decompression was performed within 3 days. Glucocorticoids were discontinued 2-4 days after decompressive adenomectomy while the patients were carefully monitored in the intensive care unit for signs and symptoms of adrenal insufficiency (19). Morning (0800 h) and evening (1800 h) serum cortisol levels were obtained for 46 days after glucocorticoids were stopped. Cortisol levels drawn 36 h or more after the last steroid dose were considered to represent endogenous cortisol production. Glucocorticoids were reinstituted in symptomatic patients and/or those with documented low serum cortisol levels (see Results). Postoperative cortisol levels in our patients were evaluated in comparison to a large number of patients treated at our institution and documented to have a normal pituitary-adrenal axis before and after adenomectomy (19). Morning or evening cortisol levels of more than 10 ng/dL (>275.9 nmol/L) were considered normal, while those below 5 Mg/dL (138 nmol/L) were low (19). Patients with equivocal values were formally tested (insulin-induced hypoglycemia) before discharge to ensure adequate cortisol secretion. Of three patients with documented hypothyroidism
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 08 September 2015. at 03:14 For personal use only. No other uses without permission. . All rights reserved.
PITUITARY TUMOR APOPLEXY at presentation, one was treated preoperatively with L-T4, and the treatment was stopped 2 weeks after surgery. Results The characteristics of the patients included in this study are shown in Table 1. Immunostaining of the resected tissue paralleled the clinical findings in five of eight patients. However, in the remaining three, discrepancies between tumor immunocytochemical staining and the clinical history were evident (patients 1, 5, and 7). This might be caused by the difficulties encountered in instances where necrosis and hemorrhage involved most of the tumor. The clinical history in all three patients was consistent with the presumptive diagnosis of prolactinomas, although the possibility of hyperprolactinemia in association with nonsecreting tumors could not be completely excluded. Preoperatiue endocrine studies Baseline hormone levels obtained before the administration of glucocorticoids showed a wide range of abnormalities in pituitary function. In only one patient was the serum cortisol level appropriately elevated (55.3 fig/dh; 1525.7 nmol/L) for the degree of stress with which they presented. In the remaining seven patients, serum cortisol ranged from 2.3-11.9 (63.5-328.3 nmol/L), with a mean ± SE of 5.9 ± 1.4 (162.8 ± 38 nmol/L). The two patients with the most severe symptoms on admission had the highest cortisol levels (8.0 and 11.9 Mg/dL; 237.3 and 328.3 nmol/L). The remaining five patients had low or low normal cortisol levels (2.3-6.0 /j.g/dL; 63.5-165.4 nmol/L). One patient who presented 19 days after the episode had a clearly low (2.7 Mg/dL; 74.5 nmol/L) morning cortisol level. All seven patients had signs and symptoms of adrenal insufficiency at the time their blood samples were drawn. The administration of cosyntropin (0.25 mg, iv) to three of these seven subjects resulted in normal increments in the serum cortisol level (22, 29, and 39 Mg/dL; 607, 800, and 1076 nmol/L, respectively). Clinical and biochemical evidence for hypothyroidism was
325
noted in three subjects who had low levels of T4 (3.2, 4.1, and 4.8 Mg/dL; 41.2, 52.8, and 61.8 nmol/L, respectively) and free T4 index (3.6, 4.4, and 4.7; normal, 5-12). Serum TSH levels in these patients were in the normal range for euthyroid subjects (0.5-3.5 mU/L). Serum testosterone and gonadotropin levels were low in all four men (Table 2) and consistent with the clinical history of impotence and diminished libido. Serum gonadotropin levels were clearly low in two women and within the normal range in the remaining two female subjects. Serum GH levels were normal (2 and 3 Mg/L) in one patient with acromegaly even though his somatomedin-C levels were evaluated (4.5 and 3.9 U/ml; normal, 0.5-2.0 U/mL). Serum PRL levels were elevated (150 Mg/dL) in only one subject (no. 6) with documented macroprolactinoma. The level was high normal in two subjects (no. 2 and 3) and low in the remaining five subjects (0.5-3 ng/dL). Galactorrhea was noted on examination in three of the five subjects with low serum PRL levels. None of the patients had signs or symptoms to suggest impairment of antidiuretic hormone secretion. Thus, at the time of their initial presentation, seven patients had glucocorticoid deficiency, three had hypothyroidism, and six had hypogonadism (Table 3). There was no correlation between the duration of symptoms and the severity or extent of the neuroophthalmological abnormalities or pituitary dysfunction. Endocrine studies in the immediate postoperative period Glucocorticoid therapy was discontinued 2-4 days after surgical decompression. Consistently low serum cortisol levels (
