Dehydroepiandrosterone Sulfate (DS) Levels, a Rapid Test for Abnormal Adrenal Androgen Secretion
SIGRUN KORTH-SCHUTZ, LENORE S. LEVINE, AND MARIA I. NEW,1 WITH THE TECHNICAL ASSISTANCE OF DIANE M. CHOW
Department of Pediatrics, Division of Pediatric Endocrinology, Cornell University Medical College, New York, New York 10021
ABSTRACT. Dehydroepiandrosterone sulfate (DS) concentration was measured in the sera of pre- mature and full-term infants and in children throughout puberty. Panhypopituitary, Addisonian, and virilized children were also studied. DS decreased slowly during the first weeks of life from a high level in neonates to the low levels observed between one to five years. After five years of age, DS concentrations started to rise. A steeper increase was observed with the onset of puberty, and adult DS concentrations were reached in late puberty. There was no sex dif- ference in DS concentration at any pubertal stage or bone age. Day-to-day variations were small in childhood and during puberty, but were con- siderable in premature infants. DS concentrations
measured at 0900 h were not significantly different from those at 1700 h. There was a positive cor- relation of serum DS concentrations with the ex- cretion of urinary 17-ketosteroids in boys and girls (r = 0.789). Premature infants had DS concentra- tion in or above the late pubertal range. Five panhypopituitary patients and five Addisonian pa- tients had DS concentrations below normal. DS was markedly elevated in patients with congenital adrenal hyperplasia and in one girl with adrenal carcinoma, and was suppressible with dexametha- sone in the former. The ease of measurement and the small amount of blood required make serum DS determination a useful guide for adrenal androgen secretion. (J Clin Endocrinol Metab 42: 1005, 1976)
D EHYDROEPIANDROSTERONE sul- fate (DS) is the most abundant adrenal C19 steroid in the blood. The biologic function of DS is unknown. It serves as a precursor for dehydroepiandrosterone (DHEA), the biological function of which is also not well understood (1).
The ease of measurement of DS and the small quantity of blood necessary for de- termination stimulated this study of DS concentrations in normal infancy, child- hood, and adolescence. DS proved to be
Received June 13, 1975.
Supported by USPHS, NIH Grant HD 00072; by USPHS Division of Research Facilities and Re- sources, Pediatric Clinical Research Center Grant RR 47; by Grant CRBS-278 from the National Founda- tion-March of Dimes; and by Deutsche Forschungs- gemeinschaft.
Presented in part before the Lawson Wilkins Pediatric Endocrine Society Annual Meeting, April, 1975, Denver.
1 Recipient of the Health Research Council of The City of New York Career Scientist Award under contracts I-749 and U-2204.
Reprints: Maria I. New, M.D., The New York Hospital-Cornell Medical Center, 525 East 68th Street, New York, N.Y. 10021.
a good indicator of pubertal development and was useful in the differential diag- nosis of virilization in childhood.
Materials and Methods
DS levels were determined by radioimmuno- assay (RIA), as described by Buster and Abra- ham (2), in diluted serum without solvolysis or extraction. An antibody supplied by Dr. G. E. Abraham, Los Angeles, was used which was raised against DHEA-3-hemisuccinate- HSA and which cross-reacts 100% with DS. It was established that the assay was reliable at lesser dilutions than those described by Buster and Abraham (2). The formula suggested by Jurjens et al. (3) was used to correct for occasional non-specific binding of 3H-DHEA observed in serum diluted 1:50.
Using various dilutions of pooled serum (1:100-1:1000) the amounts of DS measured showed a linear regression with a correlation coefficient of 0.99. The accuracy of the re- covery of DS added to the serum of a hypo- pituitary boy was 101 + 14% in the range of 50 to 1000 pg/sample. The coefficient of varia- tion (CV) within one assay was 4.4%; between 6 assays it was 6.9%. The sensitivity of the standard curve was 25 pg. Thus, in serum
diluted 1:50 1/ug/dl was detectable. Specificity was due to the specific antibody and the high concentration of DS relative to DHEA. Andros- terone sulfate cross-reacted 8% at 50% dis- placement.
As this method measures NaH2-SO3-DHEA, values have to be multiplied with a factor of 0.75 when they are compared with the re- sults of other authors who measure DHEA after solvolysis of DS.
Urinary 17-ketosteroid excretions were meas- ured according to New et al. (4).
For statistical analysis the unpaired t test was used.
Subjects
Serum which was left over from hematocrit or other determinations was used for DS deter- minations in normal subjects. The serum was obtained via heelstick or from venipuncture. Nineteen premature infants 3-37 days old whose birthweight ranged from 980-2000 grams, 33 full-term infants 2-21 days old, and 2 infants 3 months of age were studied. Sera were obtained from 117 normal children and adolescents. Pubic hair development was rated according to Tanner (5). Patients with adrenal pathology (indicated in Table 1) were studied in the Pediatric Clinical Research Center of
the New York Hospital-Cornell Medical Center. Informed consent was obtained for all studies. The physiological studies were carried out as follows:
a) ACTH stimulation test: 40 IU ACTH,2 iv, was infused over 6-8 h; blood was drawn before and after the infusion.
b) Dexamethasone suppression: 2 mg dexa- methasone in four divided doses for 2 days, and 8 mg dexamethasone in four divided doses for 2 days were administered.
Day-to-day variation was determined in 2 normal postpubertal boys (6) over a 26-28 day period, in 5 children with various disorders over a 3-6 day period, and in 7 premature infants over 4-8 days. In 12 children with various disorders, DS was measured at 0900 hours and at 1700 hours on the same day to determine diurnal variation.
Results
During the first days of life, DS concen- trations in the full-term infants were in the postpubertal range and showed a tendency to decrease during the first 3 weeks (Fig. 1), but not to the low level
2 Acthar,® Armour Pharmaceuticals.
| Patient identi- fication number | Sex | Chrono- logic age (yr/mo) | Height age (yr/mo) | Bone age (yr/mo) | Pubic hair* | Urinary excretions | Serum concen- trations DS (mg/dl) | ||
|---|---|---|---|---|---|---|---|---|---|
| p'triol | 17-KS | 17-OH (mg/d/m2) | |||||||
| (mg/d) | |||||||||
| Congenital adrenal hyperplasia, 21-hydroxylase deficiency | |||||||||
| 1 | m | 2 days | 0/1 | 0/1 | I | 0.05 | 6.4 | 1.4 | 700 |
| 2ł | f | 3/3 | 4/0 | 7/10 | II | 7 | 12 | 15 | 113 |
| 3 | m | 3/10 | 5/5 | 6/0 | I | 9 | 9 | 1.5 | 79 |
| 4 | m | 4/6 | 7/0 | 10 | I-II | 13 | 10 | 4.4 | 420 |
| 5 | m | 6/10 | 10/8 | 12-13 | II | 21 | 19 | 2.7 | 285 |
| Congenital adrenal hyperplasia, 30-ol-dehydrogenase deficiency | |||||||||
| 61 | m | 13/1 | 12/0 | 13 | III | 8.4 | 153 | 6.5 | 9,000 |
| Girl with adrenal carcinoma | |||||||||
| 7 | f | 5 | 3/6 | 4/2-5 | II-III | 0.8 | 83 | 5.9 | 1,300 |
* Tanner stages of puberty.
Off hydrocortisone for two days.
# On 10 mg hydrocortisone b.i.d.
400
PH | || ||| IV+V
DEHYDROEPIANDROSTERONE SULFATE, µg/dl
GIRLS . A V .
GIRLS PH Il + ill
BOYS . AVO
BOYS PH (I + III
PH
PH IV+V
300-
250
200
160
140-
100
80
60
40
20
0
2
3
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
CA MONTHS
BONE AGE YEARS
of the prepubertal range. In premature infants (not shown in Fig. 1), DS levels were significantly higher (26-967 ug/dl) than those in the full-term infants (23- 296 µg/dl), whose mean age was 7 days. The mean value for all premature infants at a mean age of 12 days (272 ug/dl) exceeded the mean postpubertal value (248 µg/dl). Venous and capillary serum gave similar results.
There was no striking sex difference in DS concentrations at any age or pubertal stage. A gradual increase in DS concen- tration was observed between 5 and 8 years with a more rapid increase after 8 years, continuing until late puberty. There was a general correlation between the stage of
puberty and the DS concentration, although the levels measured in all pubertal stages overlapped (Figs. 1-3).
When the urinary 17-ketosteroid excretion was compared with the serum DS concen- tration in 50 boys and 46 girls of all ages, they were found to be significantly cor- related with an r = 0.789 (Fig. 4). The correlation was slightly better in the girls (r = 0.883) than in the boys (r = 0.754).
In girls with precocious adrenarche, mean DS concentrations were elevated for bone age but matched the values observed in girls with Tanner stage II pubic hair (Fig. 2).
In boys with simple virilizing congenital adrenal hyperplasia (CAH) due to 21-hy-
(12)
MEAN ± SE (n)
180
(10)
(12)
DEHYDROEPIANDROSTERONE SULFATE, µg/dl
140
100
(10)
60
(12)
20
(8)
PA
£
PA
1-5
5-8
8-12
9-13
BA Yrs
TANNER STAGE
1
11
300
(21)
200
(22)
(12)
(6)
100
(30)
0
PA
TANNER STAGE
1
II
III
IV+V
NORMAL MALES
PATIENTS & VIRILIZING CAH 21 HYDROXYLASE DEFICIENCY
PATIENT & 3 BHSD
PATIENT & VIRILIZING ADRENAL CARCINOMA
8 00
9000
1300
DEHYDROEPIANDROSTERONE SULFATE, ug/dl
MEAN±SD (n)
700
600
500
400
(10)
300
200
(32)
(16)
(6)
(II)
100
(9)
(5)
0
1-14 days
1-5
5-8
8-12
11-14
12-15
15-19 BA Yrs.
TANNER STAGE
I
11
III
IV+V
100000
·-. 21 HYDROXYLASE DEFICIENCY
50000
X-x 3 BHSD
0 ---- o ADRENAL CARCINOMA
DEHYDROEPIANDROSTERONE SULFATE, ug/dl
10000
X
5000
o.
×
1000-
0-
X
500-
100
50
Base line
ACTH
2
3
4
7
30 days
| 2 mg/d | 8 mg/d |
DEXAMETHASONE
droxylase deficiency, the DS concentrations were markedly elevated for bone age and pubertal status (Fig. 3).
The highest DS level was observed in the boy with 30-ol-dehydrogenase defi- ciency (Fig. 3). In one girl with adrenal cortical carcinoma, the DS concentration was very high (Fig. 3). In all but one of the patients with CAH, the DS concentra- tion rose slightly with ACTH, and all demonstrated suppression with dexametha- sone (Fig. 4). In the patient with the adrenal tumor, the increase observed after ACTH continued during dexamethasone ad- ministration.
In patients with decreased adrenal func- tion due to organic or idiopathic hypo- pituitarism, DS concentrations were very low. In contrast, 3 of the boys (patients
13, 14, and 16) with Addison’s disease had DS concentrations only slightly below nor- mal. The levels did not increase, however, after ACTH administration in 2 patients (patients 13 and 14). (Table 2)
The day-to-day variation was only slightly higher (CV = 4.6-12.7% and 27.8%) than the intra- or inter-assay variation (CV = 16-89%) (Table 3). The DS concentra- tions at 1700 hours varied from 31% less to 46% more than the concentrations meas- ured at 0900 hours (Table 4), and the mean morning and evening values were not sig- nificantly different.
Discussion
DS levels have not been previously measured in infants and only sporadically in older children (7-12). During the first
9
400
300
Serum DS, ug / dl
200
100
Males o, r=0.754 Females ., r=0.883
0
0
5
10
15
17-KS, mg/d
three weeks of life in full-term infants, DS concentrations decrease, but not to the very low levels measured in early child- hood (Fig. 1). This slow decrease indicates that the DS concentration does not reflect residual maternal or placental secretion but reflects postnatal secretion. The function of DS in the postnatal period is unknown. The extremely high DS concentrations measured in some of the premature infants may be interpreted as response to stress.
Serum DS levels in early childhood are very low and are in accord with previous reports (7-9). The gradual increase begin- ning at 5 years of age correlates well with the increase of urinary 17-ketosteroids ob-
served by others (13,14). The steep in- crease in DS concentrations after 8 years of age suggests that DS may have a role in puberty. However it is clear from Fig. 1 that DS begins to rise prior to the ap- pearance of pubic hair in both boys and girls. One might speculate that DS has an initiating role in the stimulation of adrenal or gonadal secretion of androgens necessary for the appearance of pubic hair.
A remarkably low level of DS was ob- served in the patients with panhypopituitar- ism, whose levels were lower than those observed in the patients with Addison’s disease (Table 4). This suggests the impor- tance of a pituitary trophic hormone or hor-
| Patient identi- fication number | Sex | Age (yr/mo) | Tanner stage | Diagnosis | Treatment | DS (µg/dl) | |
|---|---|---|---|---|---|---|---|
| Normal range for age | |||||||
| 8 | m | 13 | I | After removal of cranio- pharyngioma | Hydrocortisone, thyroxine, lysine- vasopressin | <1.0 | 49-153 |
| 9 | m | 13/8 | II | After removal of pinealoma | Hydrocortisone thyroxine, lysine- vasopressin | <1.0 | 57-149 |
| 10 | f | 18/10 | IV | Idiopathic | Cyclic estrogens | 2.9 | 171-307 |
| 11 | f | 13/10 | I | panhypo- | hGH | 3.1 | 27-103 |
| 12 | m | 16/3 | I | pituitarism | Hydrocortisone | 1.8 | 49-153 |
| 13 | m | 0/1 | I | Addison's disease | None, ACTH (40 U, 6 h iv) | 17 19 | 35-98 61-290 |
| 14 | m | 6/2 | I | Addison's with Schilder's dis- ease | None, ACTH (40 U, 6 h iv) | 5.8 6.0 | 3-39 61-290 |
| 15 | m | 11/9 | I | Addison's disease | Hydrocortisone | <1.0 | 49-153 |
| 16 | m | 21 | V | Addison's disease | Hydrocortisone | 126 | 172-369 |
| 17 | m | 12/5 | I | Addison's disease | Hydrocortisone | 5-8 | 49-153 |
mones necessary for the secretion of DS. However, the trophic hormones causing the rise in DS and in other adrenal androgens at adrenarche have not been identified.
In the postpubertal Addisonian boy (Table 2) DS may be partially of testicular origin. Testicular secretion of DS has been demon- strated by Saez and Bertrand (9), and Laatikainen, et al. (15), but was not shown by Nieschlag et al. (16).
The lack of sex difference at any stage of puberty suggests that DS does not func- tion as a masculinizing or feminizing hor- mone at puberty. This is supported further by the patient with 30-ol-dehydrogenase deficiency (patient 6, Table 1), who had extremely elevated levels of DS without abnormalities of other androgens and did not demonstrate marked virilization or an advanced bone age.
Significantly elevated DS levels in syn- dromes of adrenal virilization indicate that DS is oversecreted in concert with the other biologically potent adrenal andro-
gens. This may prove useful as a tool in the differential diagnosis of virilized states, without actually demonstrating a biological role for DS. Elevated levels of DS in postnatal serum may be used to diagnose congenital adrenal hyperplasia in the new- born. In infants manifesting salt-wasting, increased DS levels suggest congenital adrenal hyperplasia rather than Addison’s disease, and a markedly elevated DS would suggest 30-ol-dehydrogenase deficiency. The lack of suppressibility of DS with dexamethasone distinguished adrenal tumor from adrenal hyperplasia and is an important use for this measurement.
The main virtue of DS in assessing virilization and adrenal androgen secretion is its relative ease of measurement. Be- cause it is secreted in large amounts and the radioimmunoassay is relatively simple, small amounts of blood for DS measurement can be obtained as a screening procedure even in the new- born.
| JCE & M . 1976 | |
|---|---|
| Vol 42 . No 6 |
| Patient identi- fication number | Sex | Tanner stage | Disorder | Days | Range | Mean (µg/dl) | SD | Coefficient of variation (per cent) |
|---|---|---|---|---|---|---|---|---|
| Children with various disorders | ||||||||
| 18 | m | I | Male pseudo- hermaphrodite | 4 | 3.1-5.4 | 3.9 | 1.1 | 28 |
| 19 | f | II | Precocious puberty | 3 | 45-50 | 47 | 2.2 | 4.6 |
| 20 | f | II | Precocious puberty | 4 | 52-68 | 63 | 7.3 | 12 |
| 21 | f | II | Adrenarche | 4 | 157-185 | 172 | 11.5 | 6.7 |
| 22 | f | IV | Hypertension | 4 | 108-131 | 120 | 9.8 | 8.2 |
| 23 | f | V | Hypertension | 6 | 151-196 | 164 | 17 | 10 |
| Normal postpubertal boys* | ||||||||
| 24 | V | None | 28 | 157-253 | 195 | 25 | 13 | |
| 25 | V | None | 26 | 221-304 | 249 | 26 | 10 | |
| Premature infants Birth weight (g) | ||||||||
| 26 | f | 1,960 | 4 | 67-237 | 165 | 68 | 41 | |
| 27 | m | 1,950 | 4 | 158-326 | 219 | 74 | 34 | |
| 28 | m | 1,330 | 7 | 111-967 | 468 | 368 | 76 | |
| 29 | f | 1,250 | 4 | 112-288 | 191 | 74 | 39 | |
| 30 | m | 1,160 | 7 | 26-448 | 188 | 168 | 89 | |
| 31 | m | 980 | 8 | 84-425 | 169 | 108 | 64 | |
| 32 | f | 780 | 5 | 359-551 | 447 | 71 | 16 | |
* Parks, et al. (6).
| Patient | Sex | Tanner stage | Disease | DS (µg/dl) | ||
|---|---|---|---|---|---|---|
| identi- fication number | AM | PM | Change (percent) | |||
| 33 | m | I | None | 11 | 11 | 0 |
| 18 | m | I | Male pseudohermaphrodite | 5.4 | 6.0 | +11 |
| 34 | m | I | Short stature | 50 | 50 | 0 |
| 35 | f | I | Short stature | 13 | 19 | +46 |
| 36 | m | I | XXY Klinefelter | 80 | 76 | -5 |
| 20 | f | II | Precocious adrenarche | 63 | 68 | +8 |
| 21 | f | II | Precocious adrenarche | 173 | 185 | +7 |
| 37 | f | II | Precocious adrenarche | 86 | 101 | +17 |
| 99 | 99 | 0 | ||||
| 38 | f | II | Clitoromegaly | 52 | 36 | -31 |
| 19 | f | II | Precocious puberty | 50 | 40 | -20 |
| 39 | m | III | Status post-op. orchiopexy | 133 | 144 | +8 |
| 40 | f | V | Polycystic ovaries | 236 | 276 | +17 |
| 237 | 306 | +29 | ||||
The minimal daily variability and the lack of a significant diurnal variation makes DS a stable hormone for measurement at random times.
The significant correlation of urinary 17- ketosteroid excretion and serum DS con- centration suggests that the serum assay may be used in place of the urinary assay, thus avoiding the inconvenience of a 24-hour urine collection.
References
1. Baulieu, E. E., F. Dray, R. Emiliozzi, M. D. Lebeau, P. Mauvais-Jarvis, and P. Robel, An adrenal-secreted “androgen”: Dehydroepiandros- terone sulfate. Its metabolism and a tentative generalization on the metabolism of other con- jugates in man, Recent Prog Horm Res 21: 411, 1972.
2. Buster, J. E., and G. E. Abraham, Radioimmuno- assay for dehydroepiandrosterone sulfate, Anal Lett 5: 543, 1972.
3. Jurjens, H., J. J. Pratt, and G. W. Woldring, Radioimmunoassay of plasma estradiol without extraction and chromatography, J Clin Endo- crinol Metab 40: 19, 1975.
4. New, M. I., B. Miller, and R. E. Peterson, Aldo- sterone excretion in normal children and in children with adrenal hyperplasia, J Clin Endo- crinol Metab 45: 412, 1966.
5. Tanner, J. M., Growth at Adolescence, Black- well Scientific Publ., Oxford, 1962.
6. Parks, G. A., S. Korth-Schutz, R. Penny, R. F. Hilding, K. W. Dumas, S. D. Frasier, and M. I. New, Variation in pituitary-gonadal function in
adolescent male homosexuals and heterosexuals, J Clin Endocrinol Metab 39: 796, 1974.
7. Yamaji, T., and H. Ibayashi, Plasma dehydro- epiandrosterone sulfate in normal and patho- logical conditions, J Clin Endocrinol Metab 29: 273, 1969.
8. Boon, D. A., R. E. Keenan, W. R. Slaunwhite, Jr., and T. Aceto, Jr., Conjugated and uncon- jugated plasma androgens in normal children, Pediatr Res 6: 111, 1972.
9. Saez, J. M., and J. Bertrand, Studies on testicular function in children: Plasma concentrations of testosterone, dehydroepiandrosterone and its sulfate before and after stimulation with human chorionic gonadotrophin, Steroids 12: 749, 1968.
10. Rosenfield, R. L., and W. E. Eberlein, Plasma testosterone and dehydroepiandrosterone sul- fate during puberty, J Pediatr 74: 932, 1968.
11. Migeon, C. J., Adrenal androgens in man, Am J Med 53: 606, 1972.
12. Hopper, B. R., and S. S. C. Yen, Circulating concentrations of dehydroepiandrosterone sulfate during puberty, J Clin Endocrinol Metab 40: 458, 1975.
13. Talbot, N. B., and A. M. Butler, Excretion of 17- ketosteroids by normal and by abnormal children, Am J Dis Child 65: 364, 1943.
14. Hamburger, C., Normal urinary excretion of neutral 17-ketosteroids with special reference to age and sex variations, Acta Endocrinol (Kbh) 1: 19, 1948.
15. Laatikainen, T., E. A. Laitinen, and R. Vihko, Secretion of free and sulfate-conjugated neutral steroids by the human testis. Effect of admin- istration of human chorionic gonadotropin, J Clin Endocrinol Metab 32: 59, 1971.
16. Nieschlag, E., D. L. Loriaux, H. J. Ruder, I. R. Zucker, M. A. Kirschner, and M. B. Lipsett, The secretion of dehydroepiandrosterone sul- phate in man, J Endocrinol 57: 123, 1973.