Measurement of Urinary Steroid Profile in Patients with Adrenal Tumor As a Screening Method for Carcinoma
SHIGERU MINOWADA, KENJI KINOSHITA*, MAKOTO HARA,
KOICHIRO ISURUGI, TORU UCHIKAWA ** AND TADAO NIIJIMA
Department of Urology, Fuculty of Medicine, University of Tokyo, Hongo, Tokyo,
Department of Urology* and Internal Medicine ** , Tokyo Metropo- litan Komagome Hospital, Hon-Komagome, Tokyo
Abstract
Results of measurement of urinary steroid metabolite profile using gas chromatographic analysis in eight patients with adrenocortical tumors, i.e. 3 adenomas with Cushing’s Syndrome, one adenoma with virilization, one adenoma without clinical manifestations, one carcinoma with Cushing’s syndrome and virilization, one carcinoma with Cushing’s syndrome and feminization, and one carcinoma without endocrinological symptoms, are reported. A unique pattern dominated by 58 and 118-hydroxy steroid metabolites was confirmed in five patients with Cushing’s syndrome consisting of three cases with adenomas and two with carcinomas. Excessive 3, 17a, 21-trihydroxy-58-pregnan-20-one (tetrahydro-11-deoxycortisol, THS) and 45-pregnene-3, 11«, 20a-triol (45- pregnenetriol) values were found in all three carcinomas including a non- functional carcinoma. These findings would strongly suggest the tumor to be a carcinoma, although excessive excretion of THS and 45-pregnenetriol was detected in one patient with a large adenoma associated with virilization. One patient with carcinoma was responsive to ACTH stimulation while the remainder show almost no response to exogenous ACTH. Urinary steroid profiling using gas chromatographic analysis, especially the values for THS and 45-pregnenetriol, appears to be a useful method to use in detecting these steroid metabolic characteristics in patients with adrenocortical carcinoma.
An increasing number of adrenal tumors without endocrine evidence are incidentally found with wide application of CT scans (Prinz et al., 1982 ; Copeland, 1983). Almost all such adrenal tumors, howeve, were proved to be benign legions for which surgery was considered unnecessary retro- spectively. A high prevalance of benign, clinically silent, adrenal adenoma was known from autopsy series in contrast with an
extremely low incidence of adrenal carcinoma. To determine which patients should have surgery is a controversial issue at present. From this viewpoint, endocrinological esti- mations become very important in the dffer- entiation of carcinomas from silent adenomas (Copeland, 1983). Various abnormal patterns of excreted adrenocortical steroids and their metabolites had been observed not only in patients with functional carcinoma but also in those with clinically non-functional carci- noma owing to imbalance of biosynthetic
and metabolic enzymes (Lipsett et al., 1963 ; Kinoshita et al., 1969; Lewinsky et al., 1974). Measurement of the urinary steroid metabolite profile is useful to use in detect- ing characteristic adrenocortical steroid metabolism. We present the urinary neutral steroid profiles of eight patients with adrenal tumors who showed different clinical mani- festations and histological features when gas chromatographic analysis was employed.
Patients and Methods
Subjects
Seven patients proven surgically and one diagnosed by CT scan as having adrenal tumors were included (Table 1). One patient had a clinically silent adenoma incidentally discovered by CT scan (3.0×3.5 cm) but not confirmed by surgery (Case 1). Three patients had Cushing’s syndrome caused by an adrenocortical carcinoma (Case 2~4). One patient had a large histo- logically proved adenoma associated with symptoms of virilization only (Case 5). The other three patients had adrenocortical carcinoma ; one of them had manifestations of femizization and Cushing’s syndrome (Case 6), another had manifestations of virilization and Cushing’s syndrome (Case 7), and the other had no endo- crine signs and symptoms (Case 8). Normal subjects included 14 males aged 17 to 43 years old and 5 females (follicular phase of the men- strual cycle) aged 19 to 43 years old without endocrine or liver diseases. All of the patients and normal subjects were Japanese.
Analytical procedure
An aliquot of 24-hr urine was adjusted to pH 5.0 with acetic acid and acetate buffer. 300
U of beef liver ß-glucuronidase was added to each ml of urine and the urine was incubated at 37°℃ for 48 hr. After cholesteryl butylate was added as an internal standard, the solution was extracted with ethyl acetate and washed with 4 % NaOH and distilled water. The extract was transferred into a screwcapped tube and was reduced to dryness under dry nitrogen. After 500 ml of a stock solution of methoxyamine hydrochloride in pyridine (MOX) was added, the trimethylsilyl ether derivatives were prepared by dissolving in 100 ml of pyridine along with 100 pl of bis-trimethylsilyl-acetamide and 100 ul of trimethylchlorosilane. The mixture was heated at 130℃ overnight (16-hr) and then dried under nitrogen. Immediately thereafter the reaction mixture was dissolved in pyridine and hexane, and a 2-5 pl aliquot of the solution was injected with a splitting system into a glass capillary column. Gas chromatograms were taken on a Shimazu GC-6A equipped with a flame-ionization detector. The column consisted of a 30-m open tubular glass capillary (0.3 mm I.D.) well-coated with OV-101. Temperature programming was employed at 1℃/min from 200°℃. Peak area integration was performed with a computing integrator connected to a gas chromatograph.
Results
Validation of the method
Several methods have been reported for the gas chromatographic separation of urinary neutral steroids using a high-resolution open tubular glass capillary column (Bailey et al., 1974 ; Shackleton and Honour, 1976; Fantl and Gray, 1977 ; Pfaffenberger and Horning, 1977). The method presented here is a modification of these methods and previously
| Case | Age | Sex | Endocrine manifestation | Histological diagnosis | Tumor size or weight |
|---|---|---|---|---|---|
| 1 | 72 | M | non-functioning | (adenoma, probably) | 3.0×3.5 cm by CTscan |
| 2 | 24 | M | Cushing's synd. | adenoma | 60 g |
| 3 | 36 | F | Cushing's synd. | adenoma | 10 g |
| 4 | 56 | F | Cushing's synd. | adenoma | 15 g |
| 5 | 28 | F | varilization | adenoma | 350 g |
| 6 | 20 | M | feminization +Cushing's synd. | carcinoma | 1,160 g |
| 7 | 53 | F | virilization +Cushing's synd. | carcinoma | 1,800 g |
| 8 | 52 | F | non-functioning | carcinoma | 1,600 g |
| Steroid metabolite | Abbreviation | Retention time (min) | Calibration constants | |
|---|---|---|---|---|
| Mean | C.V. (%) | |||
| Androsterone | An | 26. 64 | 0. 70 | 7.5 |
| Etiocholanolone | Et | 27. 52 | 0. 66 | 8.9 |
| Dehydroepiandrosterone | DHEA | 29. 78 | 0. 52 | 10.5 |
| 11-Oxoeticholanolone | O-Et | 32.08 | 0. 43 | 14.8 |
| 11 8-Hydroxyandrosterone | OHA | 36. 21 | 0. 62 | 9.8 |
| 11 8-Hydroxyeticholanolone | OHE | 37. 19 | 0, 61 | 10.1 |
| Pregnanediol | PD | 40.00 | 1. 06 | 7.0 |
| Pregnanetriol | PT | 41.67 | 1.17 | 9.2 |
| Tetrahydro-11-deoxycortisol | THS | 45.49 | 0. 70 | 13.7 |
| Pregnanetriolone | PTL | 47.89 | 0.81 | 9.9 |
| 45-Pregnenetriol | 45-PT | 50. 46 | 0. 94 | 9.7 |
| Tetrahydrocortisone | THE | 51.23 | 0. 35 | 12. 7 |
| Tetrahydrocorticosterone | THB | 53.27 | 0. 39 | 10.1 |
| Tetrahydrocortisol | THF | 54.75 | 0. 64 | 6.5 |
| allo-Tetrahydrocortisol | a-THF | 55.57 | 0. 74 | 10.6 |
| Cortolone | Co | 56.29 | 0. 84 | 9.8 |
| -Cortolone+Cortol | B-Co+Cor | 58.03 | 2. 63/2 | 4.1 |
| Cortol | Cor | 60.51 | 1.21 | 7.2 |
| Cholesteryl butylate | CB | 67.87 | 1.00 | |
reported procedures (Kinoshita et al., 1966). Eighteen neutral steroid metabolites were quantified through the use of an internal reference compound (cholesteryl butylate). Retention times and calibration constants (steroid metabolite/cholesteryl butylate ratios) of these steroid derivatives are shown in Table 2 *. Calibration constants and their coefficient of variation ware calculated by carrying out analyses of authentic standard steroid mixture four times. The corrected recovery rates and their standard deviations for each standard steroid added to an aliquot of urine specimen were also calculated. These results confirmed that the accuracy and reproducibility of the method are satis- factory for clinical application.
Steroid metabolite profiles in normal subjects and patients with adrenal tumors
The gas chromatographic profile of uri- nary steroids for a normal adult male is illustrated in Fig. 1. Mean excretion values and excretion ranges of urinary steroid metabolites in 14 normal males are shown
in Table 3. Excretion of pregnanetriolone was not detectable in any subject. The urinary steroid excretion patterns in 5 normal
* The following trivial names are used in the test for the designation of steroids : 11-oxoetio- cholanolone=3a-hydroxy-5-androstane-11, 17- dione; 118-hydroxyandrosterone = 3a, 118- dihydroxy-5a-androstan-17-one; 11-hydroxy- etiocholanolone=3«, 11-dihydroxy-5-andro- stan-17-one ; pregnanediol=58-pregnane-3«, 20 a-diol ; pregnanetriol=58-pregnane-3a, 17a, 20 a-triol ; tetrahydro-11-deoxycortisol=3a, 17a, 21-trihydroxy-5 -pregnan-20-one ; pregnanetrio- lone=3a, 17a, 20a-trihydroxy-58-pregnan-11- one ; 45-pregnenetriol = 45-pregnene-38, 11, 20 a-triol ; tetrahydrocortisone = 3a, 17a, 21-tri- hydroxy-5 8-pregnane-11, 20-dione ; tetrahydro- corticosterone=3x, 11, 21-trihydroxy-5-preg- nan-20-one; tetrahydrocortisol = 3a, 11, 17, 21-tetrahydroxy-5-pregnan-20-one ; allo-tetra- hydrocortisol=3a, 118, 17a, 21-tetrahydroxy-5 a-pregnan-20-one ; cortolone=3a, 17a, 20a-21- tetrahydroxy-58-pregnan-11-one ; B-cortolone= 3a, 17a, 208-21-tetrahydroxy-53-pregnan-11- one ; B-cortol=58-pregnan-3a, 118, 17a, 20}, 21-pentol ; cortol=58-pregnan-3a, 118, 17a, 20 a, 21-pentol.
females were almost similar, but the values for the 50/5a metabolite ratios in female subjects were significantly higher than the values in male subjects (Table 4). These sex differences mainly stem from distinct liver steroid reductases (Pfaffenberger and Horning, 1977). On the other hand, sulfate conjugated steroids were estimated in 6 normal males using Helix pomatia enzyme- hydrolysis. About 85 % of DHEA and 19 % of androsterone excretion were presented as a sulfate conjugate in urine.
Urinary steroid profiles of a patient with Cushing’s syndrome due to adrenal adenoma (Case 3) and another patient with adrenal carcinoma manifesting virilization and Cushing’s syndrome (Case 7) are illustrated in Fig. 2 and Fig. 3, respectively. Values for each urinary steroid excretion of eight patients with adrenocortical tumor are sum- marized in Table 5. In addition, ratios of Et/An, THF/a-THF, and THE/THF were calculated.
In five patients with Cushing’s syndrome
5
0
€
a
0
0
1
L
I
I
1
I
F
5
₪
HO
1
1
1
U
-
O
m
Q
0
0
5
m
00
4
1
0
1
I
0
0.
I
0
W
un
1
L
W
I
0
w
I
O
0
E
0
4
0
I
1
4
O
1
1
1
1
1
H
1
1
1
20min
40min
60min
(Case 2, 3, 4, 6, 7) the excess excretions of THF and the lower THE/THF ratios re- sulted from hypercortisolism. Besides, the 50/5a steroid metabolite (Et/An, THF/a- THF) ratios were elevated in Cushing’s syndrome in spite of the histological differ- ences between adenoma and carcinoma. In all three carcinoma patients (Case 6, 7, 8) including one without endocrine manifest- ations (Case 8), excessive amounts of THS and 45-pregnenetriol were excreted. This excretion pattern was also demonstrated in one patient with virilization due to an
| Abbreviation | Range (mg/day) | Mean | S. D. |
|---|---|---|---|
| An | 0.80~3.50 | 1.92 | 0. 90 |
| Et | 0.85~4.30 | 1.98 | 1.16 |
| DHEA | 0.01~0.67 | 0. 21 | 0. 19 |
| O-Et | ND~0.89 | 0. 30 | 0. 30 |
| OHA | 0.46~3.15 | 1. 33 | 0. 79 |
| OHE | 0.01~0.90 | 0.19 | 0. 30 |
| PD | 0.14~0.96 | 0. 32 | 0. 22 |
| PT | 0.50~1.49 | 0. 92 | 0. 34 |
| THS | ND~0.55 | 0. 26 | 0. 15 |
| 45-PT | 0.17~0.56 | 0. 36 | 0. 15 |
| THE | 1.24~5.44 | 3. 04 | 1. 42 |
| THB | 0.20~0.71 | 0. 40 | 0. 15 |
| THF | 0.49~2.58 | 1. 50 | 0. 71 |
| a-THF | 0.16~1.19 | 0. 56 | 0. 37 |
| Co | 0.09~1.03 | 0.41 | 0. 37 |
| B-Co+Cor | 0.07~0.93 | 0. 44 | 0. 34 |
| Cor | 0.01~0.50 | 0. 14 | 0. 15 |
ND : not determined
S. D. : standard deviation
| Et/An mean±S.D. | THF/a-THF mean±S.D. | |
|---|---|---|
| Male subjects (n=14) | 1. 07±0. 43 | 3. 39±2.08 |
| Female subjects (n=5) | 1.89±0. 73* | 7.29±2.12 ** |
* P<0.05) Significantly different from male ** P<0.01J subjects.
adenoma (Case 5).
The initial histological diagnosis of Case 7 had been adenoma. Sixteen months later, however, a liver metastasis was found and endocrinological examination showed high excretion values for adrenocortical steroids. Therefore, the clinical diagnosis was changed to adrenal carcinoma. In this case a re- sponse of urinary steroids, especially of cortisol metabolites, was observed on the third day of exogenous administration of ACTH-Z 1 mg for each day (Table 5).
Discussion
Measurements of urinary steroid metabo- lites are important in the diagnosis of adrenocortisol carcinoma which often causes Cushing’r syndrome and virilizatin (Bertagna and Orth, 1981 ; Didolkar et al., 1981). The higher ratios of the 53/5a steroid metabo- lites (Et/An, THF/a-THF) and the lower ratio of THE/THF were found in Cushing’s syndrome both with adenoma and carcinoma. This unique pattern of urinary steroid ex- cretion dominated by 58 and 118-hydroxy steroid metabolites are mainly reflections of decreased activity of liver enzymes involved in 5a-reductase and 118-hydroxy steroid dehydrogenase activity (Phillipou, 1982). The sex differences were also shown in activities of liver 53/5a steroid reductases (Table 4). These hepatic steroid metaboliz- ing enzyme activities might be distinctively programmed at birth by androgens (Gustafs- son and Steinberg, 1974).
Some of the urinary steroid metabolites, such as THS, 45-pregnenetriol and pregn-5- ene-3a, 16a, 20a-triol which result from a relative deficiency of 118-hydroxylase and 3 ß-hydroxysteroid dehydrogenase, were secret- ed excessively in patients with adrenocortisol carcinoma, including those with non-func- tioning carcinoma (Lipsett and Wilson, 1982; Lewinsky et al., 1974 ; Kinoshita et al., 1968 ; Thomas and Steinbeck, 1969). In the three
carcinoma cases with various clinical features the values for THS and 45-pregneneteiol were invariably elevated in all instances. The excessive secretions of these steroid metabolites were not observed in the patients with Cushing’s syndrome due to adenoma. Hence, the excess amounts of THS and 45- pregnenetriol excretion may indicate that the lesion is a carcinoma. On the other hand these findings were observed in one patient with a large adenoma (Case 5) which was diagnosed to be benign by detailed histo-
logical examination and observation of the clinical course for three years after operation without signs of metastasis or reccurrence. This patient manifesting virilization only, seems to be a rare case reported in the literature (Bertagna and Orth, 1981 ; Didolker et al., 1981).
Patients with adrenal carcinoma, in general, showed almost no response to ACTH stimulation (Bertagna and Orth, 1981). One of our patients with a functioning carcinoma responded to exogenous ACTH (Case 7).
THE
a-THE
5,0
An
4-Et
B-Co, Cor
Co
(50).
100()
CB.
A
un
1
0-Et
OHA
OHE
THS
.45-PT
THB
or
-U
20min
40min
60min
3
In this case ACTH stimulation evoked more excessive excretions of THS and cortisol metabolites (THE, THF, OHA etc.) but disclosed no elevation of the PD or PT ex- cretion (Table 5). These findings reveal sufficient enzyme activity of 21-hydroxylase in contrast with deficient activities of 118- hydroxylase and 33-hydroxysteroid dehydro- genase. Although factors which account for
the responsiveness to ACTH stimulation are unknown, it may be related to close re- semblances to adenoma in the histological features. In this Case 7 the initial histo- logical diagnosis had been adenoma, al- though the tumor was extremely large (1800 g weight). Therefore difficulties in histo- logical diagnosis would still occasionally re- main even in cases with advanced carcinoma.
3
=
00
0
0
0
00
-
0
V
0
+ 江
A
i
A
a
à
---- w
A
T
C
0:0
0
I
EI
1
-
0
4
5
C
9-9-
A
E
=
5
0
m
2
00
A
E
OHE
0
4
O
U
0
O
a- THE
F
U
m
1
G
0
m
1
w
C
0
Y
A
0
0
Y
V
7
U
1
Y
y
A
2
1
-20min
40min
60min
| Case | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | |
|---|---|---|---|---|---|---|---|---|---|
| Uninary steroid (mg/day) | (*) | ||||||||
| An | 0. 25 | 0.11 | 0. 34 | 0. 08 | 6. 78 | 2.10 | 5.76 | 9. 69 | 2.08 |
| Et | 0. 20 | 0. 46 | 0. 69 | 0. 19 | 7.11 | 7.61 | 9. 56 | 18.29 | 0. 99 |
| DHEA | 0. 02 | 0. 03 | 0.08 | (-) | 39.71 | 1. 54 | 3. 71 | 5. 02 | 1.14 |
| O-Et | 0. 11 | 0.71 | 0. 31 | 0. 59 | 1.73 | 1.87 | 1. 63 | 2. 51 | 0. 16 |
| OHA | 0. 43 | 0. 55 | 0. 74 | 0. 96 | 2. 55 | 7.98 | 8.57 | 17.72 | 1.61 |
| OHE | 0. 06 | 0.89 | 0. 23 | 0. 17 | 1.99 | 2.09 | 3. 45 | 4.17 | 0. 32 |
| PD | 0. 06 | 0. 52 | 0. 74 | 0. 23 | 4. 66 | 5. 95 | 9.48 | 7.78 | 3.47 |
| PT | 0. 38 | 0. 67 | 0. 55 | 0. 50 | 1.75 | 7.24 | 4. 76 | 6. 66 | 3.57 |
| THS | 0. 08 | 0. 20 | 0. 08 | 0.31 | 4. 32 | 24. 06 | 24.09 | 45.95 | 2. 42 |
| PTL | (-) | (-) | (-) | (-) | 0. 08 | 0. 29 | (-) | (-) | (-) |
| 45-PT | 0.08 | 0. 05 | 0. 11 | 0. 19 | 4.54 | 6.40 | 9. 79 | 9.70 | 8. 67 |
| THE | 0. 76 | 2.04 | 4. 78 | 13.89 | 1. 28 | 17.67 | 14.57 | 14.09 | 1.18 |
| THB | (-) | 0.60 | 0. 32 | 0. 72 | 0. 05 | 0. 60 | 0. 69 | 0. 44 | 0. 42 |
| THF | 0. 98 | 4.03 | 5.18 | 9. 37 | 0. 32 | 12.92 | 3. 62 | 12. 78 | 0. 40 |
| a-THF | 0. 02 | 0. 32 | 1.18 | 2. 24 | 0. 15 | 0. 41 | 0. 77 | 2.04 | 0. 31 |
| Co | 0.47 | 0. 55 | 0. 67 | 2. 30 | 0. 27 | 0. 77 | 0. 78 | 2.14 | 0. 20 |
| -Co+Cor | 0. 26 | 0. 90 | 0. 46 | 3. 36 | 0. 54 | 2. 93 | 1.25 | 2.64 | 0. 15 |
| Cor | 0. 08 | 0. 14 | 0. 19 | 0. 57 | 0. 04 | 0. 35 | 0. 30 | 0. 63 | 0. 54 |
| Et/An ratio | 0. 80 | 4.18 | 2. 03 | 2. 55 | 1. 05 | 3. 62 | 1. 66 | 1.89 | 0. 48 |
| THF/a-THF ratio | 49.00 | 12. 59 | 4. 39 | 4.18 | 2.13 | 31.51 | 4. 70 | 5.92 | 1. 29 |
| THE/THF ratio | 0. 78 | 0. 51 | 0. 92 | 1. 48 | 4.00 | 1. 37 | 1. 26 | 1.18 | 2. 95 |
(*) : 1 mg/day of ACTZ-was administered on the third day.
In these cases urinary steroid metabolic characteristies, especially the high values for THS and 45-pregnenetriol, clearly indicate that the tumor is a carcinoma.
We concluded that measurement of uri- nary steroid profile by gas chromatographic analysis is a good method to use in detect- ing this characteristic stedoid metabolism in patients with adrenal tumor, especially cortical carcinoma. This analytical procedure can be applied to the screening examination for increasing number of incidentally dis- covered adrenal tumors.
References
Bailey, E., M. Fenoughty and J. J. Chapman (1974). Evaluation of a gas-liquid chromato- graphic method for the determination of urinary steroids using high-resolution open-tubular glass capillary columns. J. Chromatogr. 96, 33-46.
Bertagna, C. and D. N. orth (1981). Clinical and laboratory findings and results of therapy in 58 patients with adrenocortical tumors ad- mitted to a single medical center (1951 to 1978). Am. J. med. 71, 855-875.
Didolkar, M. S., R. A. Bescher, E. G. Elias and R. H. Moore (1981). Natural history of adrenal cortical carcinoma : A clinico-patho- logic study of 42 patients. Cancer 47, 2153- 2161.
Copeland, P. M. (1983). The incidentally dis- covered adrenal mass. Ann. Int. Med. 98, 940-945.
Fantl, V. and C. H. Gray (1977). Automated urinary steroid profiles by capillary column gas-liquid chromatography and a computing integrator. Clin. Clim. Acta 79, 237-253.
Gustasson, J .- Å. and A. Steinberg (1974). Irre- versible androgenic programming at birth of microsomal and soluble rat liver enzymes active on 4-androstene-3, 17-dione and 5a- androstane-3a, 173-diol. J. Biol. Chem. 249, 711-718.
Kinoshita, K., K. Isurugi, Y. Kumamoto and H. Takayasu (1966). Gas chromatographic estimation of urinary pregnanetriol, pregnane-
triolone and pregnanetetrol in congenital ad- renal hyperplasia. J. Clin. Endocrinol. Metab. 26, 1219-1226.
Kinoshita, K., K. Isurugi, Y. Matsumoto and H. Takayasu (1969). Gas chromatographic estimation of urinary 45-pregnene-38, 17a, 20 a-triol. Steroids 11, 1-11.
Lewinsy, B. S., K. M. Grigor, T. Symington and A. M. Neville (1974). The clinical and patho- logic features of “nonhormonal” adenocortical tumors. Cancer 33, 778-790.
Lipsett, M. B. and H. Wilson (1962). Adreno- cortical cancer : steroid biosynthesis and metabolism evaluated by urinary metabolities. J. Clin. Endocrinol. Metab. 22, 906-915.
Lipsett, B. M., R. Hertz and G. T. Ross (1963). Clinical and pathophysiologic aspects of ad- renocortical carcinoma. Am. J. Med. 35, 374- 383.
Pfaffenberger, C. D. and E. C. Horning (1977). Sex differences in human urinary steroid metabolic profiles determined by gas chromoto-
graphy. Analyt. Biochem. 80, 329-343.
Phillipou, G. (1982). Investigation of urinary steroid profiles as a diagnostic method in Cushing’s syndrome. Clin. Endocrinol. 16, 433-439.
Prinz, R. A., M. H. Brooks, R. Churchill, J. L. Graner, A. M. Lawrence, E. Palovan and M. Sparagana (1982). Incidental asymptomatic adrenal mass detected by computed tomgraphic scaning. Is operation required ? J. A. M. A. 248, 701-704.
Shackleton, C. H. L. and J. W. Honor (1976). Simultaneous estimation of urinary steroids by semiautomated gas chromatography. Inveti- gation of neo-natal infants and children with abnormal steroid synthesis. Clin. Chim. Acta 69, 267-283.
Thomas, F. J. and A. W. Steinbeck (1969). Semiquantitative estimation of urinary preg- nanetriol, pregnanetriolone and tetrahydro-S in the investigation of adrenocortical function. Acta Endocrinol. 60, 645-656.