Serum Cortisol and Dehydroepiandrosterone Sulfate Responses to Adrenocorticotropin Stimulation in Premature Infants

Home , Corticosteroid, Corticosterone, Cortisone, Prednisolone

Pediat. Res. 12: 1057-1061 (1978) Adrenal gland cortisol adrenocorticotropin test dehydroepiandrosterone sulfate corticosteroid premature infant

Serum Cortisol and Dehydroepiandrosterone Sulfate Responses to Adrenocorticotropin Stimulation in Premature Infants

AKIHIKO NOGUCH1'25' AND JOHN W. REYNOLDS Department of Pediatrics, University of Minnesota Medical School, Minneapolis, Minnesota, USA

Summary cortisol and DHAS levels in early life. There are few studies of these steroids in the first months of life (2,6,8,12,13,16,20). The ACTH stimulation tests were performed in 15 premature infants present study was designed to investigate changes in the basal in a serial fashion at the ages of 5-10 days and 27-31 days. Six of levels of these steroids, and their responsiveness to exogenous the 15 had subsequent ACTH stimulation tests at 6-8 weeks ACTH in premature infants studied in a serial fashion. and/or 12-13 weeks. The pre- and post-ACTH serum cortisol and dehydroepiandrosterone sulfate (DHAS) levels were determined by radioimmunoassay. The mean basal levels of cortisol, 55 ng/ml, MATERIALS AND METHODS and DHAS, 4108 ng/ml, were significantly higher (P< 0.05 and P < 0.025, respectively) at 5-10 days than those of 24 ng/rnl and PATIENTS 1858 ng/ml, respectively, at 27-31 days. The mean net change The premature infants studied were patients on the Newborn (A) of cortisol after ACTH at 5-10 days, 95 ng/ml, increased Intensive Care Unit of the University of Minnesota Hospitals. significantly to 148 ng/ml at 27-31 days. However, ADHAS did Fifteen premature infants were studied on at least two occasions not differ significantly between the two periods (1514 ng/ml vs. by blood sampling before and after ACTH administration, and 972 ng/ml). Therefore, ADHAS/Acortisol was lower (P< 0.05) their clinical features are presented in Table 1. All of the infants at 27-31 days than at 5-10 days. No further significant changes were appropriate for gestational age in weight, defined as a birth were observed after 4 weeks of age in the levels of the two steroids. weight between the 10th and 90th percentile of normal weight for There was little correlation of basal levels between cortisol and gestational age using the intrauterine growth chart of Lubchenco DHAS, nor between Acortisol and ADHAS at any age period we et al. (15). This test was performed only when the infants were studied. There were four infants whose mothers had prenatal clinically well or suffering only from periodic apnea of prematurity steroid treatment for the prevention of hyaline membrane disease or occasional seizure episodes. None of the infants were receiving and their values were notdifferent from the other infants. an FiOz more than 25% at the time of the study and all grew satisfactorily in weight and length during the investi ation. ACTH Speculation was administered as Cosyntropin, the synthetic (a1-'5ACTh prep- aration. 0.25 rng in 1 ml normal saline was given iv over 4 rnin at Increased responsiveness of cortisol to ACTH at 1 month of age 5-10 days and at 27-31 days. Six of the 15 infants stayed in age may suggest that there is increasing 3/3-hydroxysteroid dehy- the hospital long enough to have the ACTH administration re- drogenase in the neonatal adrenal gland in this period. The well peated at 6-8 weeks and/or 12-13 weeks (Table 1). On each maintained DHAS responsiveness to ACTH until 3 months of age, occasion 0.5 ml blood was drawn by venipuncture or through however, implies that there is a persistence of fetal cortico- umbilical catheters immediately before and 1 hr after ACTH steroidogenic pathways through the first months of life. injection. The blood was centrifuged after a minimal delay for clotting and the serum was kept frozen until the time of assay. All the samples obtained from any one infant were assayed simulta- At birth, the fetal zone makes up 78% of the adrenal cortex in neously. The study was approved by the Committee on Human premature infants (19). Histochemical (1 1) and biochemical stud- Volunteers of the University of Minnesota and a parent of each ies (5, 18) have shown that 3P-hydroxysteroid dehydrogenase (3P- patient gave full and informed consent before the study was HSD) enzyme activity, which is essential for the production of started. cortisol and other A4-3-ketosteroids,is almost exclusively localized in the permanent zone of the adrenal cortex, and the neonatal PROCEDURE FOR SERUM STEROID ASSAYS adrenal gland as a whole, studied in vitro (21), has lower 3P-HSD activity than does the adult adrenal gland. This enzymatic pattern Cortisol and DHAS were measured by radioimmunoassay. The correlates with the high plasma A5-3P-hydroxysteroids (lo) and radioimmunoassay of cortisol was based on that described by DHAS (8,20) levels in the first weeks of life. During this period, Abraham and Buster (1). We carried out an ether extraction, used cortisol is produced in physiologically adequate amounts, but at cortisol antibody (C-001) from Steranti Research, Ltd., St. Albans, the expense of a high production of physiologically inactive A5- Herts, England, and used saturated (NH4)'S04 solution to separate 3P-hydroxysteroids. By 1 month of age in premature and full term bound from free steroid. The cortisol antibody had the following infants the absolute weight of the adrenal gland decreases to about cross reactivities in our laboratory (cortisol, 100%): cortisone, half of the weight at birth because of the involution of the fetal 2.0%; corticosterone, 1%, 11-deoxycortisol, 15 %; progesterone, zone, which occupies 20% of the total gland at this time (19). In theophylline q 8 x 4 days CPAP and 25% Rx till 6 wk FiO2

1580 None Stable Stable

1020 None X Apneic spells Apneic spells, treated with theophylline CPAP and 25% Rx until 8 wk; FiOs phenobarbital for seizures be- yond discharge

969 None X Apneic spells Apneic spells treated with treated with CPAP and 25% CPAP till 30 Fi02 days; phenobarbital Rx for seizures be- yond discharge 1120 None X X Stable Stable

1620 None X X Stable Stable

1370 Betamethasone, X X Stable Stable 12 mg im 4 hr before

1340 Betamethasone, X X Transient apnea, Stable 12 mg im qd hypotension, X 4 days cultures non- significant

980 None X X Apneic spells Stable treated with CPAP and 25% FiOz

1250 Betamethasone, X Stable Stable 12 mg im qd x 4 days

1170 None X Stable ACTH STIMULATION IN PREMATURE INFANTS 1059

the antibody (S-1502 no. 7) was purchased from Dr. Guy Abra- ham.

STATISTICAL ANALYSIS Two-tailed Student's t-test for paired samples was used to evaluate the significance of differences of values at each age period.

RESULTS $ e zsw g6as In Figure 1 are presented the basal (pre-ACTH stimulation) h 5,s ,$ E ,? -., levels of cortisol and DHAS found in the infants studied. The a$ '5 !2 7, mean and SE of cortisol level at 5-10 days, 55 + 11 ng/ml, was p.5 e, 3 ", gas e, .22@@ significantly higher than at 27-31 days, 24 5 ng/ml (P < 0.05). sx%2d A ggsz + gcsua 8%2 , However, 4 of the 15 patients increased their basal cortisol levels d m 2 3 during this period. The DHAS level also was significantly higher at 5-10 days, 4108 + 948 ng/ml, than at 27-31 days, 1858 + 278 nglml (P < 0.025). Although DHAS further decreased to the mean and SE of 1328 + 138 ng/ml at 6-8 weeks, this decline was not significant. In Figure 2 are shown the net changes (A) of the levels of cortisol and DHAS in response to a standardized ACTH dose. The mean and SE for Acortisol of 95 + 11 ng/ml at 5-10 days significantly increased to 148 + 15 ng/ml at 27-31 days (P < 0.002). In contrast, the mean and SE for ADHAS, 1514 + 360 ng/ml at 5-10 days, was not significantly different from the value of 972 + 142 ng/ml found at 27-31 days (P > 0.05). In addition, the mean for ADHAS was not significantly different at 6-8 weeks of age from the value at 5-10 days. The ratio of ADHAS/Acortisol was significantly lower at 27-31 days than at 5-10 days of age (P < 0.05). There was no further significant change of this ratio in the 6- to 8-week-old group of infants (Fig. 3). The correlations of basal DHAS and cortisol values, and ADHAS and Acortisol values, at the various ages studies were examined. The results were as follows

Basal DHAS vs. Cortisol ADHAS vs. ACortisol

5-10 days r = 0.184 P > 0.5 r = 0.416 p >0.1 27-3 1 days r = 0.149 P > 0.5 r = 0.33 1 P > 0.2 6-8 weeks r= -0.739 P > 0.2 r = 0.74 P> 0.1 There were four infants whose mothers were given prenatal steroids in order to prevent hyaline membrane disease in the pre-

days days weeks weeks Fig. 1. The pre-ACTH levels of cortisol and DHAS. The P values represent the significance of the differences between the 5-10 day and 27-31 day groups by the t-test for paired samples (two-tailedj. NOGUCHI AND REYNOLDS

our study may partly be due to greater clinical illness at this time than at 27-31 days, but the change during this period was less marked than that of DHAS. We found little correlation of cortisol and DHAS concentrations, or Acortisol and ADHAS values in any of the age groups studied. In older children and adults, de Peretti and Forest (8) also found that there was no correlation of responses of these steroid levels to ACTH. It is tempting to consider that the higher mean Acortisol at 27-31 days than at 5-10 days reflects the increase in size of the permanent zone of the adrenal cortex which has been shown to occur by this age (4). It is also known that the ratio of concentrations of plasma cortisone to cortisol is higher in the first week of life (12) and that shortly after injection of ['4C]cortisol in the newborn, the concentration of labeled cortisone in the circulation exceeds that of cortisol (17). Therefore, the Acortisol increase at 1 month of age could be due to a change in the pathways of cortisol metabolism with less conversion to cortisone occurring with increasing age. ADHAS does not decrease significantly in premature infants up to 13 weeks of age in spite of the significant fall in resting DHAS levels during this period. de Peretti and Forest (9) showed that the dehydroepiandrosterone response to long acting ACTH is greater when its basal level is higher from the age of 2 5-10 27-31 6-8 12-13 months to adulthood. We found no such relationship between the days days weeks weeks basal levels of DHAS and the changes in response to ACTH in prematurely born infants from 1 week up to 13 weeks of age. Fig. 2. The changes (A) in steroid values after ACTH administration. It is not known exactly what part of the adrenal cortex contrib- The P values represent the significance of the differences between the 5-10 utes to these higher levels of DHAS in the first months of life. day and 27-31 day groups by the t-test for paired samples (two-tailed). However, Cooke and Taylor (7) observed in vitro, using adrenal glands from previable fetuses, that the highest conversion of pregnenolone to DHAS occurred at the border of the fetal and permanent zones, and also that the permanent zone was capable of carrying out this conversion but at a lower rate. It may be that the production of DHAS from the border zone is maintained for several months after birth. The permanent zone of the adrenal cortex may have a relative underactivity of the 3P-HSD enzyme during the first month of life, explaining the relatively high production of DHAS during this period. Whether the explanation lies with the enzyme activity potential of permanent zone cells, or the persistence of some fetal zone cells at the permanent zone-fetal zone border has not been determined. 5-10 27-31 6-8 12- 13 We also showed that short term prenatal steroid given to the days days weeks weeks mothers to prevent hyaline membrane disease (14) did not seem to affect the adrenal cortical capacity to secrete cortisol and DHAS Fig. 3. The ratio of ADHAS to Acortisol at the various times of testing. at one week or later. This observation is in agreement with the The P value represents the significance of the difference between the 5-10 finding of Ballard et al. (3) that the total glucocorticoid activity in day and 27-31 day groups by the t-test for paired samples (two-tailed). the cord blood after this regimen was in the physiologic stress range, and that suppression of fetal cortisol levels was only tran- sitory. mature newborn (14). The steroid values from these four infants were not significantly different from the rest of the patients. CONCLUSION DISCUSSION In order to investigate the postnatal changes in steroidogenesis in the adrenal cortex, we measured the serum cortisol and DHAS We have demonstrated significant changes in basal levels of levels before and after ACTH stimulation in 15 premature infants cortisol and DHAS in the first month of life by measuring the at different ages in a serial fashion over the first 3 months. In this steroids in samples obtained serially from premature infants. We study we showed that 1) the basal levels of both steroids declined also found that the responsiveness of cortisol and DHAS serum from 1 week to 1 month; 2) cortisol response (A) to ACTH levels to ACTH, expressed by the net increase of concentration increased during the first month, whereas ADHAS did not change after the ACTH administration, changes differently for the two during the study period; 3) there is little correlation either between steroids in the 1- to 4-week period. Reports of serum levels of the basal cortisol and DHAS levels or between Acortisol and cortisol and DHAS in the first months of life are scarce. Our ADHAS; and 4) the short term steroid treatment of the mothers finding of a significant fall in basal DHAS levels over the first for the prevention of hyaline membrane disease had little effect month of life agrees with the recent reports of de Peretti and on steroidogenesis of the newborn. The results were discussed in Forest (8) who, examining more mature infants, have shown a reference to the postnatal involution of the fetal zone of the decline of DHAS concentrations in the first month of life. In adrenal cortex. addition, Turnipseed et al. (20) showed a significantly lower serum DHAS level at 35-73 days than in premature infants 1 day old. REFERENCES AND NOTES The resting level of cortisol during the first month of life did not 1. Abraham, G. E., Buster, J. E., and Teller. R. C.: Radioimmunoassay of plasma change according to the study of Hughes et al. (13). It is possible cortisol. Anal. Lett.. 5: 757 (1972). that the higher levels of cortisol in the infants 5-10 days old in 2. Anders, T.-F., Sachar, E. J.. Kream, J., Roffwarg, H. P., and Hellman, L.: ACTH STIMULATION IN PREMATURE INFANTS 1061

Behavioral state and plasma cortisol response in the human newborn. Pediat- 14. Liggins, G. C., and Howie, R. N.: A controlled trial of antepartum glucocorticoid rics, 46: 4 (1970). treatment for prevention of the respiratory distress syndrome in premature 3. Ballard, P. L., Granberg, P., and Ballard, R. A,: Glucocorticoid levels in maternal infants. Pediatrics, 50: 515 (1972). and cord serum after prenatal betamethazone therapy to prevent respiratory 15. Lubchenco, L. O., Hansman, C., Dressier, M., Boyd, E.: Intrauterine growth as distress syndrome. J. Clin. Invest., 56: 1548 (1975). estimated from liveborn birth-weight data at 24 to 42 weeks of gestation. 4. Benner, M. C.: Studies on the involution of fetal cortex of the adrenal glands. Pediatrics, 32: 793 (1963). Amer. J. Pathol., 16: 787 (1940). 16. Reiter, E. O., Fuldauer, V. G., and Root, A. W.: Secretion of the adrenal 5. Bloch, E., and Benirschke, K.: Steroidogenic capacity of foetal adrenals in vitro. androgen, dehydroepiandrosterone sulfate during normal infancy, childhood, In: A. R. Currie, T. Symington, and J. K. Grant: The Human Adrenal Cortex, and adolescence, in sick infants, and in children with endocrinologic abnor- pp. 589-595 (E. & S. Livingston Ltd., London, 1960). malities. J. Pediat., 90: 766 (1977). 6. Boon, D. A,, Keenan, R. E., Slaunwhite, R., Jr., and Aceto, T., Jr.: Conjugated 17. Seely, J. R.: 4-C14-Cortisol metabolism in the neonate. Amer. J. Dis. Child.. 104: and unconjugated plasma androgens in normal children. Pediat. Res., 6: 11 1 530 (1962). (1972). 18. Shirley, I. M., and Cooke, B. A,: Metabolism of dehydroepiandrosterone by the 7. Cooke, B. A,, and Taylor, P. D.: Site of dehydroepiandrosterone sulfate biosyn- separated zones of the human fetal and newborn adrenal cortex. J. Endocrinol., thesis in the adrenal gland of the previable foetus. J. Endocrinol., 51: 547 44: 41 1 (1969). (1971). 19. Tahka, H.: On the weight and structure of the adrenal glands and the factors 8. de Peretti, E., and Forest, M. G.: Evidence for change in adrenal secretory affecting them, in children of 0-2 years. Acta Pediat. Scand., Suppl. 81 (195 I). pattern: Plasma dehydroepiandrosterone sulfate levels from birth to puberty in 20. Turnipseed, M. R., Bentley, K., and Reynolds, J. W.: Serum dehydroepiandros- basal state and after ACTH stimulation. The Endocrine Society, Chicago, terone sulfate in premature infants and infants with intrauterine growth Illinois, 1977, Abstract 420. retardation. J. Clin. Endocrinol. Metab., 43: 1219 (1976). 9. de Peretti, E., and Forest, M. G.: Unconjugated dehydroepiandrosterone plasma 21. Villee, C. A,, and Loring, J. M.: Synthesis of steroids in the newborn human levels in normal subjects from birth to adolescence in humans: The use of a adrenal in vitro. J. Clin. Endocrinol., 25: 307 (1965). sensitive radioimmunoassay. J. Clin. Endocrinol. Metab., 43: 982 (1976). 22. The authors wish to acknowledge Mrs. Karen Bentley for her technical assistance. 10. Eberlein, W. R.: Steroids and sterols in umbilical cord blood. J. Clin. Endocrinol. 23. This research was supported by NIH Research Grant HD-5001 and by the Metab., 25: 1101 (1965). Minnesota Medical Foundation. 11. Goldman, A. S., Yakovac, W. C., and Bongiovanni, A. M.: Development of 24. The present address of Dr. John W. Reynolds is: Department of Pediatrics, activity of 3P-hydroxysteroid dehydrogenase in human fetal tissues and in two University of Oregon Medical School, Portland, OR 97201. anencephalic newborns. J. Clin. Endocrinol., 26: 14 (1966). 25. Requests for reprints should be addressed to: Akihiko Noguchi, M.D., Newborn 12. Hillman, D. A,, and Giroud, C. J. P.: Plasma cortisone and cortisol levels at birth Division, Department of Pediatrics, University of Cincinnati College of Med- and during the neonatal period. J. Clin. Endocrinol., 25: 243 (1965). icine, 231 Bethesda, Cincinnati, OH 45267 (USA). 13. Hughes, E. R., Seely, J. R., Kelley, V. C., and Ely, R. S.: Corticosteroid levels 26. Received for publication September 16, 1977. before and after corticotropin. Amer. J. Dis. Child., 104: 605 (1962). 27. Accepted for publication January 27, 1978.