Search

To compare bilateral inferior petrosal sinus sampling (IPSS) with high dose dexamethasone (HDD) and CRH testing (using recently proposed stringent response criteria) in the differential diagnosis of ACTH-dependent Cushing's syndrome, we reviewed 53 consecutive cases. The main analysis was limited to 45 cases with confirmed diagnosis: 44 with pituitary dependency, proven by confirmatory histology and/or significant biochemical improvement after pituitary surgery, and 1 with ectopic ACTH syndrome. After HDD (2 mg every 6 h for 48 h), 21 of the 44 pituitary cases met the stringent more than 90% suppression criterion. Twenty-three of the 44 pituitary cases also underwent CRH testing; 16 of 23 met a stringent response criterion of a more than 50% serum cortisol rise. For HDD and CRH testing combined, 8 of 23 fulfilled both stringent criteria, 10 of 23 had discordant results, and 5 of 23 failed to fulfil either of the stringent criteria for pituitary dependency. IPSS was performed in all 44 of the proven pituitary cases; 36 had petrosal/peripheral ACTH ratios of 2.0 or more without CRH stimulation. Thus, in patients with proven pituitary disease, stringent response criteria to HDD and CRH testing were fulfilled by only 48% and 70%, respectively. IPSS, which gave direct evidence of pituitary ACTH secretion in 82% of the cases, is therefore considered necessary in a significant proportion of cases.

Objective: We assessed the current role of bilateral adrenalectomy in the overall management strategy of hypercortisolism., Design: Retrospective review of case notes., Patients: Twenty-six patients (20F/6M); mean age 46 years (range 15-70 years), median duration of follow-up 5.25 years (0.6-19.1 years) who had undergone bilateral adrenalectomy at the Royal Victoria Hospital since 1972. Eighteen had had prior transsphenoidal surgery which did not control the hypercortisolism., Measurements: Morbidity, mortality, incidence of Nelson's syndrome., Results: Surgery was performed through bilateral postero-lateral incisions (20 patients) or a long epigastric incision (six patients). The mean combined weight of the adrenals at surgery was 11.2 g. Twenty patients received subcutaneous heparin and 18 antibiotic prophylaxis peri and post-operatively. There was no operative mortality. Minor complications included one post-operative wound infection and a small pneumothorax requiring drainage. Major complications occurred in two other patients, both with pre-existing invasive pituitary tumours and considered at high risk because of age and general debility. One patient had a massive pulmonary embolus and the other a subphrenic abscess post-operatively. This latter patient, the only mortality, died from an unrelated cause three years post-operatively. Six patients have subsequently undergone pituitary surgery and three have received external pituitary irradiation therapy for expanding tumours., Conclusions: Bilateral adrenalectomy, in experienced hands, is a relatively safe and useful management option in patients with hypercortisolism. Growth of a pituitary adenoma post-operatively is now the most worrying complication.

Objective: We assessed the endocrine outcome after transsphenoidal surgery for Cushing's disease., Design: Five-year (mean) follow-up (range 1 month-12 years) of patients undergoing transsphenoidal surgery for Cushing's disease between 1977 and 1990; review of case notes, current clinical and biochemical assessment including 24-hour urinary free cortisol., Setting: Northern Ireland., Subjects: Forty-one patients (33F: 8M); mean age at diagnosis 39.1 years (9-72 years)., Main Outcome Measures: Measurements of early post-operative 0800 h serum cortisol and 24-hour urinary free cortisol at least 24 hours after withdrawal of oral hydrocortisone therapy. This was followed by low dose dexamethasone testing. Current 24-hour urinary free cortisol measurements. Retrospective definition of cure., Results: Twenty-seven patients were either cured or improved by surgery, 14 were considered definite failures. Of 19 patients cured, eight had unmeasurable early post-operative 0800 h serum cortisol levels while of 15 tested, 13 had complete suppression with dexamethasone and two suppressed normally but to still measurable levels (39 and 60 nmol/l respectively). Seventeen patients in total have subsequently had bilateral adrenalectomy of whom two have developed Nelson's syndrome. Seven of the 41 patients were shown to have definite cyclical cortisol secretion first diagnosed post-operatively in three patients. Hormone deficiency included TSH (5), LH/FSH (1), cortisol (1) and ADH (temporary in 7, permanent in 1). In all, seven patients had some type of permanent hormonal deficiency post-operatively., Conclusions: Transsphenoidal surgery offers a worthwhile cure rate without the necessity of life-long endocrine therapy. Post-operative endocrine assessment must be rigorous so that early further management can be planned in the significant percentage of patients in whom cure is not achieved. Early complete suppression on low dose dexamethasone testing is very suggestive of cure but repeated and long-term monitoring of 24-hour urinary free cortisol is advisable.

The pathogenesis of the hypertension associated with Cushing's syndrome and with acromegaly is poorly understood. We have investigated the possible roles of sodium retention, activation of the renin-angiotensin system and increased sympathetic nervous system activity in untreated patients. In 11 patients with Cushing's disease, seven of whom were hypertensive, total exchangeable sodium was normal despite increased levels of the mineralocorticoid hormones, 11-deoxy-corticosterone and corticosterone. The renin-angiotensin system was also normal. Cardiac sensitivity to the beta-receptor agonist isoprenaline was increased, but this was not due to an increase in beta-adrenoceptor density. Hypertension in Cushing's disease is neither sodium-dependent nor angiotensin II-mediated, but increased cardiac sensitivity to catecholamines, by increasing cardiac output, may contribute to the pathogenesis of hypertension. In nine patients with acromegaly (three of whom were hypertensive) total exchangeable sodium was elevated. Although no correlation between blood pressure and exchangeable sodium was found, hypertension in acromegaly is probably sodium dependent. No evidence was found for a pathogenetic role for either the renin-angiotensin-aldosterone or the sympathetic nervous system.

Human growth hormone releasing hormone (GHRH) fails to stimulate human growth hormone (GH) in hypercortisolism. In order to study whether the responsiveness to GHRH stimulation returns after cure of the hypercortisolism, the GH response to GHRH was examined in 8 patients at least 5 yr after they had undergone bilateral adrenalectomy as their sole treatment for Cushing's disease. None had current evidence of a pituitary macroadenoma. A group of 8 healthy subjects matched for age and sex formed the control group. All patients and subjects received an iv injection of GHRH 1 microgram/kg, after an overnight fast, blood samples were taken before and at 15, 30, 45, 60, 90 and 120 min. There was no statistical difference between the peak GH or area under curve (AUC) response (median, range) in the two groups studied (adrenalectomized peak GH 9.2 (4.6-32.0) vs 16.5 (7.5-63) mU/l, adrenalectomized AUC response 647.5 (344.2-1489.5) vs 1103.5 (339.7-5188.5) mU/l. Patients with Cushing's disease once cured of hypercortisolism, have a GH response to GHRH.

Six consecutive patients with Cushing's disease were treated with the broad spectrum antifungal drug ketoconazole. Urinary cortisol levels rapidly fell to within the normal range in five of the six patients. Acute hypoadrenalism occurred in one patient, and nausea and pyrexia in three. Our experience with hepatotoxicity was different from that reported by others in that reversible hepatotoxicity was demonstrated in three patients within 7 to 12 days of treatment. Further work is required before ketoconazole can be recommended as a standard primary therapy for patients with Cushing's syndrome. Continuing vigilance for both hypoadrenalism and hepatotoxicity is essential in any patient being treated with this drug either for hypercortisolism or for other reasons.

Bilateral inferior petrosal sinus sampling was successfully performed in 12 of 13 consecutive patients with ACTH-dependent Cushing's syndrome. Ten of the patients subsequently had transsphenoidal pituitary microsurgery. Eight patients in whom the inferior petrosal sinus to peripheral vein ACTH level ratio was 1.5 or greater were found to have a pituitary adenoma. One of the remaining two patients who had ratios less than 1.5 had pituitary hyperplasia while the other had no identified abnormality. In five of the patients with pituitary tumour a ratio above 1.5 was present on only one side. Bilateral petrosal sampling is therefore always necessary. Tumour localization within the pituitary was only poorly predicted by either petrosal sinus sampling (four of eight) or computed tomography scanning (three of eight). If petrosal sinus sampling is used early in the differential diagnosis of ACTH-dependent hypercortisolism, then the use of other differential diagnostic tests may not always be necessary.

A controversy still exists in regard to hypothalamic pituitary function long-term after cure of hypercortisolism due to Cushing's disease. In an attempt to resolve this controversy, we have studied 15 patients, treated at least 6 years previously, by bilateral adrenalectomy. None of these patients had had pituitary-directed therapy. The maximum increment response of serum TSH in response to TRH was greater than 5 mU/l in 13 of the 15. Serum PRL response to TRH, GH response to insulin-induced hypoglycaemia, gonadotrophin responses to LHRH and nocturnal PRL secretion were normal in all patients studied. When nocturnal GH secretion was corrected for age, body mass index and menopausal status it was definitely abnormal in only two patients. The mean nocturnal GH secretion did not differ from that measured in a control group of Addisonian patients. The series of patients also did not differ significantly from the Addisonian patients in relation to the pattern of changes in plasma ACTH, over 24 h after an 0800 h oral dose of hydrocortisone. There was a significant rise in plasma ACTH between 2200 h and 0600 h in both groups of patients. The plasma ACTH concentrations were significantly higher in post-adrenalectomy patients. Hypothalamic pituitary function is normal in the long-term in the majority of patients treated by bilateral adrenalectomy for Cushing's disease.

It has been suggested that the first-phase feedback of steroid feedback on ACTH secretion is abnormal in Cushing's disease patients studied after adrenalectomies. We have performed hydrocortisone infusions (100 mg over 2 h) in 15 patients who had previously undergone bilateral adrenalectomy for Cushing's disease. None had had any pituitary-directed therapy and none had a pituitary macroadenoma. Their responses were compared with a group of seven hypoadrenal patients. A significant rise in serum cortisol was observed between 5 and 10 min in both groups. The first significant fall in ACTH occurred between 30 and 45 min in both groups. There was no abnormality of first-phase feedback in the Cushing's group and our results do not therefore support previous speculation that such an abnormality provides evidence for a higher central nervous system (CNS) aetiology of Cushing's disease.

Forty four subjects (23 obese controls, 11 patients with possible Cushing's syndrome, and 10 patients with definite Cushing's syndrome) underwent low dose (0 X 5 mg every six hours for two days) dexamethasone suppression tests during which serum cortisol concentration at 0800 and excretion of urinary free cortisol over 24 hours were measured. Serum cortisol concentration fell to below 60 nmol/1 (2 X 2 micrograms/100 ml) in 31 subjects and remained above 250 nmol/1 (9 X 1 micrograms/100 ml) in the 13 others. Excretion of urinary free cortisol showed a similar response, falling to below 110 nmol (40 micrograms)/24 h in 31 and remaining above 180 nmol (65 micrograms)/24 h in the 13 others. There was complete concordance between the two variables in terms of the pattern of response. Serum cortisol concentration fell to below 60 nmol/1 (2 X 2 micrograms/100 ml) in at least 97% (31 of a possible 32) of subjects without Cushing's syndrome. On the other hand, a serum cortisol concentration of above 250 nmol/1 (9 X 1 micrograms/100 ml) after low dose dexamethasone gave a false positive diagnosis of Cushing's syndrome in at most only one of 13 patients (7 X 7%). Measurement of serum cortisol concentration during the low dose dexamethasone test is simpler than, and as accurate and reliable as, measurements of urinary steroids.

Reported cases of cyclical Cushing's syndrome are rare. Of 14 successive patients with Cushing's syndrome nine collected sequential urine samples for the estimation of cortisol:creatinine ratio. Five had cyclical Cushing's syndrome while two had considerable variation in urinary cortisol excretion without a cyclical pattern being established. Two of the five patients with a cyclical syndrome had paradoxical responses to dexamethasone. In only one patient with a cyclical pattern did the cortisol:creatinine ratio fall after treatment with bromocriptine or cyproheptadine, or both. The high incidence of the cyclical form of Cushing's syndrome has important clinical implications. A high index of suspicion of the syndrome is required in patients with symptoms or signs of Cushing's syndrome but with normal cortisol values, in patients with fluctuating cortisol values, and in patients with anomalous responses to dexamethasone. Because of possible variations in steroidogenesis the results of drug studies in Cushing's syndrome must be interpreted cautiously.

A 23-year-old female presented with severe Cushing's syndrome in the 23rd week of pregnancy. Investigations showed plasma cortisol 770 nmol/l (08.00 h) and 850 nmol/l (23.00 h); plasma ACTH was 10 ng/l (08.00 h) and 27 ng/l (23.00 h); urinary free cortisol excretion was 2460 nmol/24 h. Dexamethasone 2 mg 6-hourly for 48 h suppressed the 08.00 h plasma cortisol only to 680 nmol/l. Abdominal C.T. scan showed a right adrenal adenoma. The patient was treated with metyrapone and a good clinical improvement ensued. Plasma cortisol was reduced to 300-500 nmol/l. Depsite ultrasonographic evidence of normal fetal growth, urinary oestriol excretion was markedly deficient. Prior to the spontaneous onset of labour, there was a marked rise in plasma cortisol despite continuous metyrapone treatment. A normal female infant was born at 37 weeks' gestation. The maternal adrenal adenoma was subsequently removed. The deficiency of oestriol synthesis during the pregnancy may be explained by metyrapone-induced inhibition of C19-hydroxylation.

To establish a rapid test for Cushing's syndrome we measured serum cortisol during and following iv dexamethasone infusion (5 micrograms.kg-1.h-1 for 5 h from 10.00 h) in simple obesity (N = 19) and in Cushing's syndrome (N = 12). We had first established that 5 micrograms.kg-1.h-1 was the lowest dose which consistently lowered serum cortisol in simple obesity. In obesity, serum cortisols ranged from undetectable (less than 30) to 48 nmol/l at 17.00, less than 30 to 37 at 19.00 and less than 30 to 38 at 08.00 h the following day. Serum cortisol at these three times did not show any overlap between simple obesity and Cushing's syndrome. Having established these findings we proceeded to study a group of patients with polycystic ovarian disease. These patients behaved differently. Their values at 17.00 and 19.00 h did not overlap those of Cushing's syndrome. However, at 08.00 h, 5 of the 7 had values within the range seen in Cushing's syndrome with a mean of 290 +/- 99 nmol/l. In conclusion, 17.00 and 19.00 h serum cortisol levels distinguish between Cushing's syndrome and both simple obesity and polycystic ovarian disease. However, in the latter, cortisol suppression is less prolonged than in simple obesity. This finding may be important for our understanding of the pathogenesis of the disease.