Departments of Medicine and Psychiatry, The Second Tokyo National Hospital and School of Medicine, Keio University, Tokyo and Institute for Steroid Research and the Clinical Research Center, Montefiore Hospital and Medical Center, Bronx, N. Y. 10467, USA
PLASMA CORTISOL PROFILES IN CUSHING’S SYNDROME By Koichiro Yoshida, Hiroshi Satowa, Akio Sato, Yoichi Ichikawa, Jacob Kream, Joseph Levin, Barnett Zumoff, Leon Hellman and David K. Fukushima1)
ABSTRACT
Plasma cortisol profiles were studied by the frequent sampling method in 5 patients with Cushing’s disease (CD), 7 patients with Cushing’s syndrome due to adrenocortical adenoma (AA), and one patient with bronchogenic carcinoma. Plasma ACTH was measured by radioimmunoassay at 10 min intervals in 2 of the subjects. In CD, there was distinct episodic secretion of cortisol and ACTH; the coefficients of variation about the mean plasma cortisol concentration ranged from 24 to 27 %; plasma ACTH ranged from zero to 455 pg/ml with a mean of 94 pg/ml. In AA, the tumour secreted cortisol at a constant rate with little fluctuation; the coefficients of varia- tion of plasma cortisol concentration ranged from 8 to 14 %; plasma con- centrations of ACTH were always near zero. In the patient with broncho- genic carcinoma, the coefficient of variation of cortisol was 14 %. These results were apparent even in profiles of plasma cortisol concentrations measured over only a 6 h period. It is concluded that characteristics of plasma cortisol and ACTH secretory patterns are helpful in differentiating Cushing’s syndrome of differing aetiology.
Portion of this study was presented at the Vth International Congress of Endocrinology, Hamburg, Fed. Rep. Germany (1976).
1) Reprint requests should be addressed to Dr. David K. Fukushima, Institute for Steroid Research, Montefiore Hospital and Medical Center, Bronx, N. Y. 10467, USA.
The intermittent secretion of cortisol by normal subjects has been well estab- lished (Hellman et al. 1970a; Weitzman et al. 1971; Krieger et al. 1971; Gallagher et al. 1973). In an earlier study carried out in our laboratories, cor- tisol was also found to be secreted episodically in Cushing’s disease (Hellman et al. 1970b). Subsequently, Tourniaire et al. (1971) demonstrated that there was no distinct episodic pattern of cortisol secretion in Cushing’s syndrome due to adrenocortical adenoma. This suggested that the pattern of episodic cortisol secretion might provide a means of differentiating Cushing’s disease from Cushing’s syndrome due to adrenal tumours. Accordingly, we have investi- gated this pattern in a series of patients with Cushing’s syndrome of various aetiologies.
| Subject | Age | Sex | Suppression with 8 mg of dexamethasone | Cortisol production rate (mg/day) | Mean plasma cortisol concentration (ug/100 ml) | Coefficient of variation of plasma cortisol (%) |
|---|---|---|---|---|---|---|
| I. Cushing's disease | ||||||
| 1 | 46 | F | present | 80 | 27.2 *** | 25.0 |
| 2 | 17 | F | present | 58 | 23.0 ** | 27.0 |
| 3 | 55 | F | present | - | 22.3* | 24.2 |
| 4 | 43 | F | absent | 47 | 22.9* | 25.9 |
| (with 3 mg) | ||||||
| 5 | 51 | F | present | - | 19.5* | 26.0 |
| II. Cushing's syndrome due to adrenocortical adenoma | ||||||
| 6 | 33 | F | absent | 55 | 10.6 *** | 13.6 |
| 7 | 51 | M | absent | 95 | 21.3 ** | 7.6 |
| 8 | 23 | M | absent | 99 | 20.3 *** | 7.6 |
| 9 | 47 | F | absent | 101 | 27.3 ** | 11.0 |
| 10 | 51 | F | absent | - | 28.7 ** | 14.3 |
| 11 | 40 | F | absent | - | 21.0* | 8.7 |
| 12 | 29 | F | absent | - | 18.1* | 8.0 |
| III. Cushing's syndrome due to bronchogenic carcinoma | ||||||
| 13 | 57 | M | - | 450 | 65.5 ** | 14.0 |
* Plasma was sampled for 6-10 h.
** Plasma was sampled for 24 h.
MATERIALS AND METHODS
The experimental groups studied are listed in Table 1. Informed consent was obtained from all subjects.
Group I: Untreated patients with clinically active Cushing’s disease. - Diagnosis was made by clinical criteria, in association with laboratory tests, including elevated cortisol production rate and suppression of cortisol secretion by 8 mg per day of dexamethasone but not by 2 mg; laparotomy findings confirmed the diagnosis in every case.
Group II: Patients with Cushing’s syndrome due to adrenal adenoma. - The diagnosis was confirmed in each case by adrenal venography and surgery. [131]] Cholesterol scan- ning of adrenal glands was carried out in some of the cases.
Group III: One patient with Cushing’s syndrome due to ACTH-producing broncho- genic carcinoma. - Diagnosis was based on biopsy.
Blood samples were drawn through an indwelling catheter in a forearm vein at 20 min intervals for 6 to 24 h, as detailed previously (Hellman et al. 1970a). Plasma
40
Subj. [
30-
20
30-
Subj. 2
PLASMA CORTISOL (_g/100ml)
20
30-
Subj. 4
Subj. 3
20-
30-
Subj. 5
20-
10-
0800
1200
1600
2000
2400
0400
0800
CLOCK TIME
cortisol was measured by competitive protein binding (Murphy 1967). Cortisol produc- tion rates were measured by isotope dilution (Fukushima et al. 1976). Plasma ACTH was measured by radioimmunoassay (Gallagher et al. 1973) of samples taken at 10 min intervals in selected cases.
RESULTS
Plasma cortisol profiles in the 5 patients with Cushing’s disease (Group I) are shown in Fig. 1. Cortisol concentrations fluctuated widely in a higher range than in normal subjects (Hellman et al. 1970a). The mean plasma cortisol con- centrations during the study period ranged from 19.5 to 27.2 ug/100 ml. The coefficients of variation were confined to a very narrow range, 24 to 27 % (Table 1). A series of 9 to 10 episodes of cortisol secretion may be defined for the 24 h sampling studies in subjects 1 and 2.
20.
Subj. 6
10
30
Subj. 7
20
PLASMA CORTISOL (g/100ml)
10
30-
Subj. 8
20-
30
Subj. 9
20
30-
Subj. 10
20-
Subj. 11
Subj. 12
10.
0800
1200
1600
2000
2400
0400
0800
CLOCK TIME
SUBJECT 11 ADRENAL ADENOMA
ACTH pg /ml
x-x CORTISOL ug /100 ml
400
40
300
30
200
20
100
10
ACTH
0
2200
2400
0200
0600
0
0400
0800
1000
1200
1400
1600
CLOCK TIME
In the 7 patients with Cushing’s syndrome due to adrenal adenoma (Group II), the tumour apparently secreted steadily, since plasma cortisol did not fluctuate significantly and had no distinct episodic pattern (Fig. 2). Mean plasma cor- tisol concentrations ranged from 10.6 to 28.7 ug/100 ml in these patients, with coefficients of variation ranging from 8 to 14 %.
Subj. 13
PLASMA CORTISOL (µg/100ml )
100
80
60
40
20
0
0800
1200
1600
2000
2400
0400
0800
CLOCK TIME
| Subject | Period of measurement (h) | Mean plasma concentration (ug/100 ml) | Coefficient of variation (%) |
|---|---|---|---|
| I. Cushing's disease | |||
| 1 | 24.00-06.00 | 25.6 | 15.8 |
| 06.00-12.00 | 28.9 | 20.7 | |
| 12.00-18.00 | 23.7 | 29.7 | |
| 18.00-24.00 | 28.2 | 29.9 | |
| 2 | 24.00-06.00 | 6.5 | 29.1 |
| 06.00-12.00 | 5.5 | 20.3 | |
| 12.00-18.00 | 5.1 | 22.0 | |
| 18.00-24.00 | 4.9 | 25.4 | |
| 3 | 10.00-16.00 | 22.3 | 24.2 |
| 4 | 24.00-06.00 | 22.8 | 28.8 |
| 5 | 24.00-06.00 | 18.1 | 25.7 |
| II. Cushing's syndrome due to adrenocortical adenoma | |||
| 6 | 24.00-06.00 | 9.5 | 5.4 |
| 06.00-12.00 | 11.4 | 9.9 | |
| 12.00-18.00 | 10.6 | 12.0 | |
| 18.00-24.00 | 10.3 | 17.8 | |
| 7 | 24.00-06.00 | 21.7 | 9.3 |
| 06.00-12.00 | 20.5 | 4.2 | |
| 12.00-18.00 | 20.7 | 4.8 | |
| 18.00-24.00 | 22.4 | 8.2 | |
| 8 | 24.00-06.00 | 20.7 | 6.7 |
| 06.00-12.00 | 21.0 | 7.6 | |
| 12.00-18.00 | 19.5 | 8.7 | |
| 18.00-24.00 | 19.8 | 6.1 | |
| 9 | 24.00-06.00 | 26.0 | 6.2 |
| 06.00-12.00 | 25.7 | 7.4 | |
| 12.00-18.00 | 20.5 | 4.0 | |
| 18.00-24.00 | 31.2 | 8.5 | |
| 10 | 09.00-15.00 | 28.7 | 14.3 |
| 11 | 10.00-16.00 | 21.0 | 8.7 |
| 12 | 23.00-04.20 | 18.1 | 8.0 |
| III. Cushing's syndrome due to bronchogenic carcinoma | |||
| 13 | 24.00-06.00 | 61.6 | 11.3 |
| 06.00-12.00 | 71.5 | 14.1 | |
| 12.00-18.00 | 64.5 | 12.8 | |
| 18.00-24.00 | 65.6 | 13.7 | |
Fig. 3 shows plasma ACTH values in a patient with Cushing’s disease (sub- ject 4) and a patient with adrenal adenoma (subject 11). In Cushing’s disease, plasma ACTH and cortisol both fluctuated widely and the secretory pattern of ACTH was quite similar to that of cortisol; plasma ACTH ranged from zero to 455 pg/ml. In adrenal adenoma, plasma ACTH showed no significant fluctuation and was always near zero.
In a patient with Cushing’s syndrome due to ACTH secreting bronchogenic carcinoma (Group III), the 24 h mean plasma cortisol concentration was 66 ug/ 100 ml, with a small coefficient of variation, 14 % (Fig. 4).
The coefficients of variation of the mean plasma cortisol concentration in 18 samples taken over 6 h periods chosen arbitrarily are shown in Table 2. In any of the four 6 h periods chosen the coefficient of variation corresponds closely to that calculated over the whole 24 h period. This indicates that ade- quate information to demonstrate the differences in the secretory patterns may be obtained by taking blood samples over the comparatively short period of 6 h.
DISCUSSION
In normal subjects the hypothalamic hormone releasing factors stimulate the pituitary to release ACTH which in turn stimulates cortisol secretion by the adrenals. In Cushing’s disease the hypothalamusry-adrenal axis is func- tioning in a modified form resulting in increased cortisol production by the adrenals. The secretion of cortisol in normal subjects is episodic with a mean of 9 episodes (range 7-13) over a 24 h period (Weitzman et al. 1971). The secretion of ACTH has likewise been demonstrated to be pulsatile (Gallagher et al. 1973; Tanaka et al. 1978). Episodic secretion occurs also in Cushing’s disease but the excursions of the individual episodes are usually greater and the 24 h mean plasma cortisol concentration is higher than in normal subjects (Hellman et al. 1970b). This was the pattern demonstrated in this study in all the patients with Cushing’s disease. It is suggested that the presence of an epi- sodic pattern indicates that the cortisol production is dependent on stimulation of the adrenal by hypothalamustary factors, and where there is excessive production of cortisol the aetiology of the hypercortisolism is Cushing’s disease. On the other hand where there is an absence of episodicity in cortisol secretion, the cortisol is being produced autonomously; such is the case in patients with an adrenal adenoma. The profiles of these patients presented in this study were all relatively flat with no distinct episodes. As a measure of the extent of the fluctuations, the coefficient of variation of the mean plasma cortisol concen- tration over a 24 h period was calculated. This coefficient was 57 % in normal subjects, higher than in both the adenoma and Cushing’s disease patients; this high figure was in part due to the diurnal variation in cortisol secretion. The
group of patients with adenomata were clearly separated from those with Cushing’s disease on the basis of their values for the coefficient of variation: this ranged between 8 and 14.3 % in the former and between 24.2 and 27 % in the latter patients. In Cushing’s syndrome due to an ACTH-secreting bron- chogenic carcinoma, presumably an autonomous functioning lesion, while the absolute fluctuations in plasma cortisol concentration were greater than those seen in adenomata, the coefficient of variation was in the same range and lower than that of the Cushing’s disease patients.
A review of the literature reveals that in most cases the pattern of the pro- file of plasma cortisol concentration is peculiar to the particular diagnosis of the aetiology of Cushing’s syndrome but a number of investigators were un- able to find the consistent differences in the pattern that we report. In the series of patients reported by Sederberg-Olsen et al. (1973) the coefficient of variation of plasma cortisol in 3 out of 4 cases of Cushing’s disease ranged from 20 to 30 %, whereas in the one patient with an adrenal adenoma and 2 with ACTH-secreting bronchogenic carcinoma the range was between 9 and 10 %. Thus these results are essentially in agreement with our findings. The profile of the patient with an adrenal adenoma reported by Tourniaire et al. (1971) is similar to those that we have shown in the adenoma patients. The plasma cortisol concentrations in multiple samples taken over 24 h in 7 patients are depicted graphically by Hagen et al. (1978). The curve of the patient with ACTH secretory bronchogenic carcinoma shown by these investigators is very similar to the one in Fig. 3. The findings in respect of the other patients are also in agreement with our report save in one adenoma case, where there ap- pears to be a diurnal variation in the plasma cortisol concentration. Other investigators have also reported results in some cases differing from our findings (Vetter et al. 1977; Olsen et al. 1978). The reason for these discordant results warrants further investigation.
In patients with nodular adrenal hyperplasia. the results of tests to determine the aetiology of their Cushing’s syndrome often are conflicting. If the hypo- thesis is correct that the condition develops from benign adrenal hyperplasia to autonomous function of nodules, as has been suggested, then one may expect the results of the pituitary dependent tests to vary depending on the stage of the disease. Investigators in our institution have found the 24 h plasma cortisol profile in 2 patients with nodular hyperplasia to resemble that of patients with adrenal hyperplasia and they suggest that such a profile study could help in making the correct diagnosis (Grajower et al., personal communication).
Since it is highly desirable to diagnose the aetiology of Cushing’s syndrome pre-operatively, numerous tests have been applied. Where the results of the dexamethasone suppression, metyrapone and ACTH stimulation tests have given equivocal results adrenal venography has been performed to identify an ade- noma. The latter procedure may result in complications and the use of [13]]] -
19-iodocholesterol to visualize the adrenal glands may be a preferable investi- gation. The scintigrams, however, have sometimes proved unreliable, especially in adrenocortical carcinoma (Anderson & Beierwaltes 1974). The diagnosis of an adrenal adenoma or carcinoma can in most cases be made with confidence if one or two morning samples of plasma show very low ACTH concentrations (Broughton 1975; Rees & Landon 1976). High ACTH concentrations in such samples indicate a diagnosis either of Cushing’s disease or the presence of an ectopic ACTH producing tumour. Usually the plasma ACTH concentration is considerably greater in the latter case but even very high concentrations may be found in Cushing’s disease and this difference can therefore not be relied on to make the diagnosis (Newton et al. 1978). Even after a number of tests the aetiological diagnosis of Cushing’s syndrome may remain in doubt. In those circumstances, it is suggested that the characteristics of plasma cortisol and ACTH secretory patterns may be helpful. Only further experience will prove the extent of the usefulness of such patterns.
ACKNOWLEDGMENTS
The authors are grateful to Morris Kirschenbaum, Ruth Jandorek and Nancy Gallagher for invaluable technical assistance.
This work was supported by Grants from the National Cancer Institute (CA 07304, CA 22795), and the General Clinical Research Centers Branch (RR 53) of the National Institutes of Health.
REFERENCES
Anderson B. G. & Beierwaltes W. H .: Advanc. intern. Med. 19 (1974) 327.
Broughton A .: Amer. J. clin. Path. 64 (1975) 618.
Fukushima D. K., Zumoff B., Bulkin W. & Hellman L .: J. clin. Endocr. 43 (1976) 38.
Gallagher T. F., Yoshida K., Roffwarg H., Fukushima D. K., Weitzman E. D. & Hellman L .: J. clin. Endocr. 36 (1973) 1058.
Hagen C., Kehlet H. & Binder C .: Acta endocr. (Kbh.) 88 (1978) 737.
Hellman L., Nakada F., Curti J., Weitzman E. D., Kream J., Roffwarg H., Ellman S., Fukushima D. K. & Gallagher T. F .: J. clin. Endocr. 30 (1970a) 411.
Hellman L., Weitzman E. D., Roffwarg H., Fukushima D. K., Yoshida K. & Gallagher T. F .: J. clin. Endocr. 30 (1970b) 686.
Krieger D. T., Allen W., Rizzo F. & Krieger H. P .: J. clin. Endocr. 32 (1971) 266.
Murphy B. E. P .: J. clin. Endocr. 27 (1967) 973.
Newton R. W., Semple P. D., Browning M. C. K. & Gunn A .: J. Amer. med. Ass. 240 (1978) 770.
Olsen N. J., Fang V. S. & DeGroot L. J .: Metabolism 27 (1978) 695,
Rees L. H. & Landon J. In: Loraine J. A. and Bell E. T., Eds. Hormone Assays and their Clinical Application, 4th edition. Churchill Livingstone (1976) p. 213.
Sederberg-Olsen P., Binder C., Kehlet H., Neville A. M. & Nielsen L. M .: J. clin. Endocr. 36 (1973) 906.
Tanaka K., Nicholson W. E. & Orth D. N .: J. clin. Endocr. 46 (1978) 883. Tourniaire J., Orgiazzi J., Riviere J. F. & Rousset H .: J. clin. Endocr. 32 (1971) 666. Vetter H., Strass R., Bayer S. M., Beckerhoff R., Armbruster H. & Vetter W .: Clin. Endocr. 6 (1977) 1.
Weitzman E. D., Fukushima D. K., Nogeire C., Roffwarg H., Gallagher T. F. & Hellman L .: J. clin. Endocr. 33 (1971) 14.
Received on August 8th, 1978