Upon-admission adrenal steroidogenesis is adapted to the degree of illness in intensive care unit patients.

Adrenal function was studied in 2 groups of intensive care unit (ICU) patients with varying degrees of illness, as determined by Acute Physiological and Chronic Health Evaluation (APACHE). The 15 seriously ill patients with high APACHE scores (greater than or equal to 25) had elevated Therapeutic Intervention Scores and increased mortality compared to the 15 ill patients (APACHE, less than or equal to 10; 67% vs. 27%). Plasma cortisol, aldosterone, and androstenedione concentrations were increased in the ICU patients compared to those in normal subjects (n = 23), being greater in the seriously ill patients. Plasma dehydroepiandrosterone sulfate (DHEAS) concentrations were low in both groups of ICU patients. The ratios of aldosterone or androstenedione to cortisol were not altered, whereas the DHEAS to cortisol ratios were reduced in the ICU patients. ACTH injection elicited increases in plasma cortisol, aldosterone, and androstenedione concentrations in both groups of ICU patients, and the ratios of aldosterone and androstenedione to cortisol did not change. In the seriously ill patients, plasma DHEAS increased, so that the DHEAS to cortisol ratio did not change, whereas in less ill patients plasma DHEAS did not increase, so that the DHEAS to cortisol ratio was reduced. In this study of patients admitted to an ICU, impairment of adrenal steroid secretion appears to be specific for DHEAS. Although plasma cortisol was elevated in ill patients proportional to the degree of illness, the contribution of the concomitant decrease in DHEAS to this increase is not clear.

Spectrum of serum cortisol response to ACTH in ICU patients. Correlation with degree of illness and mortality.

Recent reports suggest adrenal insufficiency in critically ill patients is common. We found only one case of de novo adrenal insufficiency using admission ACTH injection in 70 selected intensive care unit (ICU) patients. Random serum cortisol levels correlated positively with illness severity in ICU patients using proven methods for assessing illness severity. Those with the highest random serum cortisol levels (greater than 60 micrograms/dl) had the greatest mortality, while those with lower random cortisol levels which stimulated to more than 18 micrograms/dl after ACTH injection had improved outcomes. Based on our results, routine screening for adrenal insufficiency in ICU patients is not warranted. If it is suspected, the cosyntropin test should be performed since low random cortisol levels (even to 5 micrograms/dl) are not diagnostic of adrenal insufficiency.

Thyrotropin-releasing hormone tests in an outpatient clinic.

Measurement of thyroid-stimulating hormone (TSH) levels in various timed blood samples after administration of thyrotropin-releasing hormone (TRH) has become a standard part of the diagnostic armamentarium for suspected thyroid disease. We reviewed the records of 165 patients who received TRH tests in an endocrinology outpatient clinic to assess the diagnostic utility of the test based upon other clinical parameters, and to assess concordance of the 30-minute TSH value after intravenous administration of TRH. Standard (control) values for TRH testing were obtained in 41 normal subjects. Patients were divided into groups of hyperresponders, hyporesponders, and normal responders. The overall concordance rate of the ultimate thyroid diagnosis and the 30-minute TSH value was 97%. With close adherence to clinical indications, the TRH test is of considerable benefit as an outpatient tool in the diagnostic assessment of thyroid disease. Basal and 30-minute TSH values are satisfactory for diagnosis in the majority of patients, thereby rendering unnecessary the discomfort and expense of multiple samples for TSH measurements.

Hypercalcemia in a patient with eosinophilic granuloma.

Persistent hypercalcemia subsequently developed in a 46-year-old black woman with biopsy-proved eosinophilic granuloma. Evaluation of this patient excluded or made highly unlikely any of the known causes of hypercalcemia. As with several other granulomatous diseases, eosinophilic granuloma is likely an occasional cause of hypercalcemia.

Cushing's disease presenting as amenorrhoea with hyperprolactinaemia: report of two cases.

Two patients with Cushing's disease first presented with amenorrhoea, hyperprolactinaemia, and normal tomograms of the sella turcica. Both women underwent trans-sphenoidal removal of a pituitary microadenoma resulting in a decreased serum prolactin concentration in both, and cessation of galactorrhoea and return of menses in one of them. One woman became ACTH deficient while the other has persistent Cushing's disease. A survey of the literature on Cushing's disease and prolactinomas shows the association to be uncommon. A likely mechanism is secretion of both hormones by the microadenomas. These two patients illustrate the need for screening other pituitary hormone function in patients with hyperprolactinaemia.