0O21-972X/78/4701-0212$02.0O/0 Journal of Clinical Endocrinology and Metabolism Copyright© 1978 by The Endocrine Society

Vol. 47, No. 1 Printed in U.S.A.

CONJUGATED GLUCOCORTICOIDS IN AMNIOTIC FLUID AND FETAL LUNG MATURATION Beverley E. Pearson Murphy with the technical assistance of Mary E. Patchell Reproductive Physiology Unit, Montreal General Hospital and Faculty of Medicine, McGill University, Montreal, Canada ABSTRACT. While recent studies suggest that cortisol plays a role in lung maturation, conflicting reports have recently appeared relating cortisol in amniotic fluid to indices of lung maturation. The variation in the levels and differences in correlation appear to be due to differences in methodology, principally with respect to cross-reactivity in the radioimmunoassays with the conjugates of cortisol and corticosterone. We have found that the glucocorticoid conjugates correlate better with lung maturation (r = 0.79) than does cortisol itself (r = 0.58). This is probably because the conjugates are derived exclusively from the fetal compartment whereas cortisol is also produced by the chorionic membrane. In anencephaly, conjugate levels were very low while cortisol levels were relatively normal. Conjugate levels (mainly sulfates) appear promising as an adjunct to tests of fetal maturation. gated corticoids are derived entirely Many studies in animals and man have led to the conclusion that corfrom the fetal adrenal cortex (8), we tisol can accelerate lung maturity in have recently demonstrated a non-adrenal source of cortisol in amniotic late fetal life (1). However it is still not established whether an influid, viz. its production by the chocrease in this hormone is responsible rionic membrane (9). It therefore for lung maturation in the human fetus seemed possible that in this situation nor whether the therapeutic use of cortisol conjugates might prove to be glucocorticoids might not pose some a more reliable indicator of fetal adlong-term hazard. Recent investigarenocortical function than cortisol tions regarding the relationship of itself and might correlate better with cortisol in amniotic fluid and indices indices of lung maturation. An abof lung maturation have provided a var- stract of this work has appeared (10). iety of results (Table 1 ) . When the methods used are considered MATERIALS AND METHODS . . . -n ., ,„ . , ^ the values fall into 2 groups, those , ... /r,T»\ 1. • L Amniotic fluids were obtained at employing radioimmunoassay (RIA) which . , , : ., i J i • amniocentesis, hysterotomy, or by rupgave higher values and those using _ _ . ' .' ,, . ,. ° . > , ., ° ture of the membranes prior to delivery, radiotransinassay /nn)A (RTA) which gave low- _, . • fc.. . , .. J , _, . . ° They were frozen until analyzed. Unconn er values. Prior extraction of samples . \ , ,.. . ,.v , ' n _. , __. ., 1.. i_ vjugated cortisol (C) and the palmitic/ using . /r./rU • J • J & RIA &gave lower results but this , ,, ... - stearic ratio (P/S), a recognized index step all 3 c n ... , . , r was nevertheless omitted by J ,n .. m. of lung maturation, were determined as groups (2-4). The extraction step se- , ° , ' ,ns v 6 r . , , described previously (7). parates unconjugated cfrom conjugated _.. ,, i , „ , „. • T -m/ i ..• • . Glusulase /T(Endo Lab., Garden City, cortisol. These 2 entities - conjuga- „ „ N , . n ,„ _^ _ , , . ^ , ,.£r • •, , N.Y.) was prepared in 0.1M acetate ted and unconjugated - differ widely , ' „c ^ „,, , , n . ,, . , . i , , . •, i buffer, pH 5.2. These enzymes hydrolize in their physical and biological pro- , . ' r ., , . / ^ ,-L^. J . _. ° . , both glucuronides and sulfates (11). perties yet are rfrequently confused, . , _. ., ,„„. o ^, , , . *.-!.• u Conjugated corticoids (CS) nplus unthe underlying assumption being& that . , „ . ,n i * . . r, . •, r , . . . . f . . . , . conjugated C in 10 yl of amniotic fluid the biologically inactive conjugate is . , , ,. , , - , , .: r. , ^. ? ^i. were estimated together after hydrolya metabolite of and proportional to the . . , ._n . , \ 1 ^ , j. , * . , _ sis with 100 units oGlusulase and extraunconjugated foanu and thus an index of ,n _, , . , . _ , . ? . , >.. ._ ,.n ction into ethyl acetate. The dried J its biological activity. T1 However while ,. . , ., ^, . extracts were assayed for cortisol by present evidence suggests that conju_,_. . . _„ /inN 6& J RTA using horse transcortin (12). CS Submitted March 10, 1978. was calculated as this value minus that 212

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 24 November 2015. at 20:59 For personal use only. No other uses without permission. . All rights reserved.

RAPID COMMUNICATIONS Table 1.

213

REPORTS CONCERNING AMNIOTIC FLUID CORTISOL AND LUNG MATURATION Ave. cortisol Correlation Extr. Method Authors level at term with lung mat. ng/ml

Sivakumaran et al, 1975 (2)

200

RIA

0.37

Fencl & Tulchinsky, 1975 (3)

140 28

RIA

0.83 not done

Aderjan et al., 1977 (4)

70(est.) "40-60% lower"

RIA

not calculated not done

Gewolb et al., 1977 (5)

24

RTA

NS to 0.61

Smith et al, 1977 (6)

26

RTA

NS

RTA 0.55 27 0.55 22 RIA = radioimmunoassay; RTA = radiotransinassay; Extr. = extraction step; mat. = maturation; NS = not significant; est. = estimated; ave. = average. This laboratory, 1977 (7)

for C, as ng cortisol equivalent. To identify the steroids measured, extracts of 7 amniotic fluids at various gestational ages were chromatographed on Sephadex LH-20 columns (13) using methylene chloridermethanol 98:2 before and after hydrolysis. The fractions were then assayed individually by RTA as above. Correlations were assessed using linear regression analysis - Pearson's correlation coefficient r, and by rank correlation - Spearman's coefficient r s (14). RESULTS Specificity Without hydrolysis, the only prominent peak assayed was cortisol as found previously (12). After hydrolysis, the principal peaks detected in the chromatographed extracts corresponded to cortisol and corticosterone, with small amounts of cortisone, 11desoxycortisol and more and less polar compounds. The true corticosterone concentration exceeded that of cortisol, the cortisol/corticosterone ratio being 0.72 ± 0.09; however because of the lower cross-reactivity of corticosterone with horse transcortin (30% that of cortisol) the main steroid measured was cortisol. Cortisol and corticosterone appeared to account for

almost all the increase in the material measured in late gestation. Recovery and precision of CS levels Values given were corrected for recovery which was 80.4 ± 10.7% (S.D.) (n=24). The coefficient of variation within assays at mid-range was ±10% (n=20); that between assays was ±17% (n=20). Relationship with gestational age As seen in Fig. 1 which included 124 samples, CS and C were similar in early pregnancy but after 30 weeks CS rose much more abruptly than C. This late rise in CS corresponded closely to that observed for P/S. At about 25 weeks both CS and C rose transiently; although the amniotic fluids during this period were all obtained from infants of Rh-negative mothers, these infants were normal at delivery. Relationship with lung maturation For 56 samples over 28 weeks, the correlation of CS vs P/S was good; r = 0.76, r s = 0.79. Since C is relatively small compared with CS, this was unchanged by omitting the subtraction of C from the original assay i.e. for C+CS vs P/S, r = 0.75, r s = 0.80. For the same samples, C vs P/S gave r = 0.58, a value similar to that obtained previously (7). When the r values 0.58 and 0.76 were

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 24 November 2015. at 20:59 For personal use only. No other uses without permission. . All rights reserved.

RAPID COMMUNICATIONS

214 (C 4 CS) n: 4

4

12 13

9

3

(P/S) n: 4

4

12 13

9

- - 2 5 3 4 3 1 1 1 4

5

6

6

5

5

Fig. 1. The relationship of cortisol (C), corticoid conjugates (CS) and the palmitic/stearic ratio (P/S) with gestational age. n = number of samples.

3 14 17 10 6 9 5

CORTISOL ng/ml

10

15 20 25 30 35 GESTATIONAL AGE - weeks

JCE&M • 1978 Vol 47 • No 1

are shown in Table 4. The adrenals of all 3 infants were very small and atrophic and showed marked reduction of the fetal zone. Although the unconjugated cortisol values in amniotic fluid were similar to those for normal infants of the same gestational age, the CS values were extremely low and similar to those of infants prior to 20 weeks.

40

DISCUSSION

compared using the Fisher transformation (14), the likelihood that they were drawn from the same population was 0.10>P>0.05 while that for the r s values 0.58 and 0.79 was PO.25, the latter figure an appoximation only since there is no rigorous means of comparing r s values. Table 3 shows the clinical relations for the 56 samples. Only 2 of the 56 were discordant. Of these, one was a false positive, one a false negative by P/S, the former at delivery of a normal infant. Table 3. RELATION BETWEEN CS AND P/S IN 56 PREGNANCIES OVER 28 WKS GESTATION

CS ng/ml >49 32-49

Conjugated glucocorticoids in amniotic fluid and fetal lung maturation.

0O21-972X/78/4701-0212$02.0O/0 Journal of Clinical Endocrinology and Metabolism Copyright© 1978 by The Endocrine Society Vol. 47, No. 1 Printed in U...
346KB Sizes 0 Downloads 0 Views