- Email: [email protected]
Corticotropin-releasing hormone in amniotic fluid during gestation and labor and in relation to fetal lung maturation
Corticotropin-releasing hormone in amniotic fluid during gestation and labor and in relation to fetal lung maturation Timo J. Laatikainen, MD, Ilkka J. Raisanen, MD, and Katariina R. Salminen, BSc Helsinki, Finland Corticotropin-releasing hormone was discovered in the placenta, and its concentration in the maternal plasma was found to increase greatly during the latter half of pregnancy. We studied the concentration of immunoreactive corticotropin-releasing hormone in amniotic fluid in 59 uncomplicated and in 73 complicated pregnancies. The mean { ± SE) value of corticotropin-releasing hormone in amniotic fluid in uncomplicated pregnancies was significantly higher in the third {24.1 ± 3.3 pmoi/L) than in the second {9.1 ± 0.7 pmoi/L) trimester, but no change was found during labor. In groups matched by gestational age, larger mean values of corticotropin-releash1g hormone and cortisol were observed in the group in which the lecithin/sphingomyelin ratio was >2 or the phosphatidylglycerol test was positive than in the group with a lecithin/sphingomyelin ratio <2 or a negative phosphatidylglycerol test result. In samples taken at an interval of 1 to 3 weeks, concomitant increases in corticotropin-releasing hormone and cortisol levels were found with the appearance of phosphatidylglycerol. Concentrations of corticotropin-releasing hormone in amniotic fluid were elevated in patients with diabetes and in women with preeclampsia and intrauterine growth retardation. We conclude that the intrauterine release of corticotropin-releasing hormone increases during the last trimester. This may stimulate the fetal pituitary-adrenal axis and promote fetal maturation. {AM J 0BSTET GYNECOL 1988;159:891·5.)
Key words: Amniotic fluid, corticotropin-releasing hormone, fetal lung maturation, intrauterine growth retardation
Corticotropin-releasing hormone is released into the hypophyseal portal circulation to stimulate the secretion of adrenocorticotropic hormone (ACTH). Its release is affected by stress, circadian variation, and negative feedback.' Demonstration of corticotropinreleasing hormone-like bioactivity in human placenta 2 has given rise to the question of whether placental corticotropin-releasing hormone has any role during pregnancy. Concentrations of corticotropin-releasing hormone greatly increase in maternal plasma during the last trimester of pregnancy,'- 6 then fall rapidly after delivery. 6 These findings suggest• that increased circulating corticotropin-releasing hormone levels during the last trimester may stimulate maternal and possibly also fetal pituitary-adrenal function. Immunoreactive corticotropin-releasing hormone has been detected in umbilical cord blood.'· 5 No data seem to be available on corticotropin-releasing hormone in amniotic fluid. Because its presence in amniotic fluid may reflect fetoplacental corticotropin-releasing From the Departments I and II of Obstetrics and Gynecology, Helsinki University Central Hospital. Supported by a grant from the Academy of Finland. Received for publication January 29, 1988; revised April4, 1988; ru:cepted May 17, 1988. Reprint requests: Timo Laatikainen, MD, Departments of Obstetrics and Gynecology, Helsinki University Central Hospital, Haartmaninkatu 2, SF-00290 Helsinki, Finland_
hormone more closely than hormone in the maternal plasma, we decided to study corticotropin-releasing hormone in amniotic fluid during gestation, in labor, and in various pregnancy complications. Concentrations of corticotropin-releasing hormone were also related to parameters of fetal lung maturation in amniotic fluid and were determined in pregnancies with intrauterine growth retardation (IUGR). Material and methods
In 59 uncomplicated pregnancies, the indication for amniocentesis was determined at weeks 15 to 26 by fetal karyotype and at weeks 34 to 37 by injection of a contrast medium into the amniotic cavity for determination of fetal morphologic characteristics. At weeks 38 to 41, amniotic fluid samples were collected via amnioscope before induction of labor in ll subjects and at the early stage of spontaneous labor in 12 subjects. No fetal abnormality was detected in these cases. Amniotic fluid samples were collected from 73 complicated pregnancies for determination of fetal lung maturity. The main groups of pregnancy complications were insulin-dependent diabetes in 25 patients, pre· eclampsia in 12, diabetes and preeclampsia in 5, and cholestasis of pregnancy in 9. Six women with gestational diabetes, four with Rh incompatibility, four with placenta previa, four with essential hypertension, two 891
892
Laatikainen, Raisanen, and Salminen
October 198~ Am .J Obstet Gynecol
Table I. Concentrations of corticotropin-releasing hormone and cortisol in amniotic fluid in relation to gestational age in uncomplicated pregnancies CRH (pmol/L)
Cortisol (nmoll L)
25 4 7
9.3 ± 0.9 8.0 ± 1.0 16.0 ± 2.0
13.1 ± 1.1 44.2 ± 14 50.0 ± 3.8
11 12
20.1 ± 3.1 29.6 ± 6.5
62.8 ± 8.8 79.5 ± 10.2
Wkof gestation
15-18 19-26 34-37 38-41
No labor Labor
Data are the mean ± SEM. Corticotropin-releasing hormone.
CRH,
with premature rupture of membranes, and two with bronchial asthma were combined in a group of 22 cases with miscellaneous complications. In six pregnancies in the preeclampsia group, the birth weight of the newborn was below the lOth percentile of a large Finnish group (IUGR group); in the remaining 6 birth weight was within the 1Oth to 90th percentiles. In 14 pregnancies, two amniotic fluid samples were collected at intervals of 1 to 3 weeks. Amniotic fluid samples (5 to 10 ml) were collected into polyethylene tubes containing 125 1-11 of a protease inhibitor, aprotinin (Apronin; Medica, Helsinki, Finland), chilled on ice, and centrifuged. The supernatant was stored at - 18° C. For the assay of corticotropin-releasing hormone, synthetic human corticotropin-releasing hormone and tyrosine-corticotropin-releasing hormone for iodination were purchased from Cambridge Biochemicals Ltd., Cambridge, England, and corticotropin-releasing hormone antiserum, developed against synthetic human corticotropin-releasing hormone in rabbits, was purchased from Peninsula Laboratories Europe Ltd., St. Helens, England. 125 I-Tyr-corticotropin-releasing hormone was prepared according to the chloramine-T method. Peptides in the 1 to 2 ml samples of amniotic fluid were extracted with Sep-Pak C18 cartridges from Waters Associates, Milford, Mass., as described previously." The mean recovery of synthetic corticotropinreleasing hormone added to the amniotic fluid samples and carried through the extraction step was 73%. Radioimmunoassay of corticotropin-releasing hormone was carried out as described in detail previously. 6 The intra- and interassay variabilities of corticotropinreleasing hormone determinations in amniotic fluid were 5.2% and 10%, respectively. The concentration of cortisol was determined with a kit purchased from Farmos Diagnostica, Turku, Finland. The cross-reactivity of cortisone in this assay was 0.67%. The intra- and interassay coefficients of varia-
tion of cortisol determinations were 2.8% and 7.1 %, respectively. The lecithin/sphingomyelin ratio and phosphatidylglycerol in amniotic fluid were determined as described in detail elsewhere. 7 The Student t test and analysis of variance were used in statistical analysis of the data.
Results Results of the determinations of corticotropinreleasing hormone and cortisol in uncomplicated pregnancies are summarized in Table I. A significant increase in the concentrations of corticotropin-releasing hormone (p < 0.01) and cortisol (p < 0.001) was found from the second trimester to term. A significant correlation was found between cortisol and corticotropin-releasing hormone values (R = 0.41, p < 0.01). At weeks 38 to 41, the mean concentrations of corticotropin-releasing hormone and cortisol did not differ between the group of 12 women in early spontaneous labor and the group of 11 women who were not in labor (Table I). Table II summarizes results of the determinations of corticotropin-releasing hormone and cortisol at the third trimester in relation to pregnancy complications. The mean gestational age did not differ between the groups. Mean value of corticotropin-releasing hormone was significantly higher in the group of women with diabetes and in the women with preeclampsia and IUGR than in the group of uncomplicated pregnancies. The mean cortisol level was increased in the group of women with preeclampsia and IUGR and was decreased in the group with cholestasis of pregnancy. Concentrations of corticotropin-releasing hormone and cortisol and the parameters of fetal lung maturity in amniotic fluid are related in Table III. Samples taken in uncomplicated and complicated pregnancies at weeks 32 to 38 were included. The groups with a lecithin/sphingomyelin ratio ~2 or >2, and the groups with positive or negative phosphatidylglycerol tests, did not differ in relation to mean gestational age. Mean concentrations of both corticotropin-releasing hormone and cortisol were significantly larger in the group with a lecithin/ sphingomyelin ratio >2 than in the group with a ratio ~2. and also in the group with positive phosphatidylg1ycerol than in the group with negative tests. Follow-up data of the concentrations of corticotropin-releasing hormone and cortisol in amniotic fluid are seen in Fig. l. In 11 of the 14 paired samples of amniotic fluid studied, the result of the phosphatidylglycerol test changed from negative to positive during the follow-up. Paired comparison showed a highly significant increase (p < 0.001) in the concentrations of corticotropin-releasing hormone and cortisol with ad-
Corticotropin-releasing hormone in amniotic fluid 893
Volume 159 Number 4
140 120 100 a. 80 :I 60 -=:
0 E
a: 0
//f/
s0
~ ()
/~
40
240
250
260
270
60
E 50 c 40 0(/) 30
/
20 0 230
70
10 230
280
...--
20 240
Gestation (days)
250
260
270
280
Gestation (days)
Fig. I. Course of concentrations of corticotropin-releasing hormone (CRH) and cortisol in amniotic fluid during third trimester of pregnancy.
Table II. Concentrations of corticotropin-releasing hormone and cortisol in amniotic fluid in uncomplicated and complicated pregnancies at the third trimester of pregnancy Gestation (days)
Group
Uncomplicated Diabetes Diabetes + preeclampsia Preeclampsia IUGR No IUGR Cholestasis of pregnancy Miscellaneous complications
13 25 5 12 6 6 9 22
253 254 251 237 237 237 257 240
± ± ± ± ± ± ± ±
CRH (pmol/ L)
6 2 2 7 9 10 3 7
20 45 62 67 87 47 17 21
± ± ± ± ± ± ± ±
3 6.3* 16 16 27t 13 8.5 2.7
Cortisol ( nmol/ L)
45 ± 49 ± 54± 57 ± 66 ± 43 ± 28 ± 49 ±
3.8 3.0 8.0 6.7 8.9t 5.1 3.5* 3.6
Data are the mean ± SEM. CRH, Corticotropin-releasing hormone.
*p < 0.01
compared with group with uncomplicated pregnancy. tp < 0.05 compared with group with uncomplicated pregnancy.
Table III. Concentrations of corticotropin-releasing hormone and cortisol in amniotic fluid samples at 32 to 38 weeks' gestation in groups with a lecithin/sphingomyelin ratio o;;;2 or negative phosphatidylglycerol test (immature fetal lungs) and in groups with a lecithin/sphingomyelin ratio >2 or a positive phosphatidylglycerol test (mature fetal lungs) Group
N
Gestation (days)
CRH (pmol/L)
Cortisol (nmol/L)
LIS ratio .;;2 >2 PG
25 64
249 ± 3 254 ± 6
19.1 ± 2.9* 47.6 ± 5.3*
37.0 ± 2.4t 48.5 ± 1.9t
41 45
252 ± 2 255 ± 5
26.0 ± 3.8t 47.6 ± 5.3t
38.5 ± 1.7* 52.0 ± 2.5*
+
Data are the mean ± SE. CRH, Corticotropin-releasing hormone; Ll S, lecithin/sphingomyelin; PC, phosphatidylglycerol.
*p < 0.001
compared between groups LIS ,;;; 2 and LS > 2 or PG- and PG +. tp < 0.01 compared between groups LIS,;;; 2 and LS > 2 or PG- and PG+.
vancing gestation and with maturation of fetal lungs (Fig. 1).
Comment Significantly larger concentrations of corticotropinreleasing hormone in amniotic fluid were measured in the third than in the second trimester of pregnancy. A similar increase in corticotropin-releasing hormone
levels was found earlier in the maternal circulation.'-6 The placenta was suggested to be the source of the increased corticotropin-releasing hormone release." Recently, corticotropin-releasing hormone immunoreactivity was demonstrated by immunohistochemical staining in the cytotrophoblast of the placenta, in the decidua, and in the amnion.• Thus the amnion may be a source of corticotropin-releasing hormone
894
Laatikainen, Raisanen, and Salminen
lll amniotic fluid. in addition to the placenta and decidua. Corticotropin-releasing hormone has been shown to stimulate ACTH release by the human fetal pituitary in vitro as early as at week 14 of pregnancy.'' 10 Intravenous injection of a bolus dose (0.0 1 ~J.g/kg) of corticotropin-releasing hormone into a sheep fetus increased fetal corticotropin secretion in vivo, and the response seemed to increase with an increasing dose of corticotropin-releasing hormone. 11 The fetal corticotropin-releasing hormone level was not reported during the experiment. The lowest intravenous dose of corticotropin-releasing hormone that increased corticotropin secretion in human adults, 0.03 ~J.glkg, resulted in a corticotropin-releasing hormone peak level of 45 pmoll L in plasma. 12 This was within the range of corticotropin-releasing hormone values we found in amniotic fluid and within the range (10 to 92 pmol/L) reported earlier in cord plasma.' Thus fetal blood concentrations of corticotropin-releasing hormone may be high enough to stimulate fetal pituitary corticotropin secretion. In the present study, we found a significant correlation between concentrations of corticotropinreleasing hormone and cortisol in amniotic fluid, and the follow-up showed concomitant increases in their levels. This correlation and parallel changes support the hypothesis that corticotropin-releasing hormone originating outside the fetal hypothalamus promotes fetal pituitary-adrenal function during pregnancy. Earlier findings of an increase in the concentration of cortisol in cord plasma 1' and in amniotic fluid 1' suggested a gradual increase in human fetal adrenocortic activity with advancing gestation, but the concentration of ACTH was not found to increase in amniotic fluid during pregnancy. 1' However, ACTH in amniotic fluid may poorly reflect ACTH secretion by the fetal pituitary gland. No evidence of an increased release of corticotropin-releasing hormone or cortisol into the amniotic fluid was found in association with early spontaneous labor in the present study. Production of surfactant in the fetal lungs begins after week 24 of gestation, during the period of a rapid increase in the concentration of corticotropin-releasing hormone in the maternal plasma."·h Our follow-up data showed that at the time of the appearance of phosphatidylglycerol in amniotic fluid, a concomitant increase occurred in the concentrations of corticotropin-releasing hormone and cortisol. Moreover, corticotropin-releasing hormone and cortisol levels were higher in amniotic fluid samples with a lecithin I sphingomyelin ratio > 2 or a positive phosphatidylglycerol test, indicating mature fetal lungs, than in those with a lecithin/sphingomyelin ratio ""2 or a negative phosphatidylglycerol test, indicating immature fetal lungs. Differences in the length of gestation were
October 1988 Am J Obstet Gynecol
excluded in these comparisons, because the i-iroups were matched in relation to gestational age. A corrt'lation between the concentration of cortis(Jl and the parameters of fetal lung maturity in amniotic fluid has been regarded as evidence of the role of the fetal adrenal cortex as an inductor of fetal lung surfactant.11 Ih In the present study, corticotropin-releasing hormone and cortisol showed similar correlations with the parameters of fetal lung maturity in amniotic fluid, suggesting that corticotropin-releasing hormone may be involved in this process. Concentrations of corticotropin-releasing hormone in amniotic fluid were increased in patients with diabetes and in pregnancies complicated by preeclampsia and IUGR. Earlier data on corticotropin-releasing hormone in maternal plasma in pathologic pregnancies are scant. Increased concentrations of corticotropinreleasing hormone in maternal plasma were reported in pregnancy-associated hypertension. 17 Obstetric stress, such as preeclampsia with IUGR, promotes fetal maturation, but the mechanism is not clear. It has been generally thought to operate by stimulating the fetal adrenal cortex to produce corticosteroids. Increased cortisol levels in amniotic fluid have been reported in certain pregnancy complications 1" and may accelerate fetal lung maturation. Because corticotropinreleasing hormone is produced by the placenta and probably also in the decidua, pathologic changes in these tissues, as in preeclampsia, 19 may increase its production. We speculate that corticotropin-releasing hormone might be a "signal substance" from the placenta initiating the process of accelerated fetal maturation in obstetric stress.
REFERENCES I. Vale W, Greer M. Corticotropin-releasing factor. Fed Proc 1985;44: 145-6. 2. Shibasaki T, Odagiri E, Shizume K, Ling N. Corticotropinreleasing factor-like activity in human placental extracts. J Clin Endocrinol Metab 1982;55:384-6. 3. Sasaki A, Liotta AS, Luckey MM,Margioris AN, Suda T, Krieger DT. Immunoreactive corticotropin-releasing factor is present in human maternal plasma during the third trimester of pregnancy. J Clin Endocrinol Metab 1984;59:812-4. 4. Cn>land RS, Wardlaw SL, Stark Rl, Brown LS J r, Frantz AG. High levels of corticotropin-releasing hormone immunoreactivity in maternal and fetal plasma during pregnancy. J Clin Endocrinol Metab 1986;63:ll99-203. 5. Campbell EA, Linton EA, Wolfe CDA, Scraggs PR, Jones MT, Lowry PJ. Plasma corticotropin-releasing hormone concentrations during pregnancy and parturition. J Clin Endocrinol Metab 1987;64:1054-9. 6. Laatikainen T, Virtanen T, Raisanen I, Salminen K. Immunoreactive corticotropin-releasing factor and corticotropin in plasma during pregnancy, labor and puerperium. Neuropeptides 1987;10:343-53. 7. Hallman M, Teramo K. Measurement of the lecithin/ sphingomyelin ratio and phosphatidylglycerol in amniotic fluid: an accurate method for the assessment of
Corticotropin-releasing hormone in amniotic fluid
Volume 159 Number4
8. 9. 10. 11.
12.
13.
fetal lung maturity. Br J Obstet Gynaecol 1981;88:80613. Saijonmaa 0, Laatikainen T, Wahlstrom T. Corticotropinreleasing factor in human placenta: localization, concentration, and release in vitro. Placenta 1988;9 (in press). Ackland J, RatterS, Bourne G, Rees LH. Pro-opiomelanocortin; peptides in the human fetal pituitary. Regul Pept 1983;6:51-61. Blumenfield Z, Jaffe RB. Hypophysiotropic and neuromodulatory regulation of adrenocorticotropin in the human fetal pituitary gland. J Clin Invest 1986;78:288-94. Rose JC, Hargrave BY, Meis PJ, LaFave M, Torpe B. Corticotropin-releasing factor-induced adrenocorticotropic hormone release in the sheep fetus: blockade by cortisol. AMJ 0BSTET GYNECOL 1985;151:ll28-33. Watson SJ, Lopez JF, Young EA, Vale W, Rivier J, Akil H. Effect oflow dose ovine corticotropin-release hormone in humans: endocrine relationships and ~-endorphin/~ lipotropin responses. J Clin Endocrinol Metab 1987; 66:10-5. Smith ID, Shearman RP. Fetal plasma steroids in relation
14. 15. 16. 17.
18. 19.
to parturition. I. The effect of gestational age upon umbilical plasma corticosteroid levels following vaginal delivery. J Obstet Gynaecol Br Commonw 1974;81:11-5. de Fencl MM, Tulchinsky D. Total cortisol in amniotic fluid and fetal lung maturation. N EnglJ Med 1975;292: 133-6. Tuimala R, Kauppila A, Haapalahti J. ACTH levels in amniotic fluid during pregnancy. Br J Obstet Gynaecol 1976;83:853-6. Kresch MJ, Gross J. The biochemistry of fetal lung de-· velopment. Clin Perinatol1987;14:481-507. Wolfe CDA, Campbell EA. Lintone EA, et al. Plasma corticotropin-releasing factor (CRF) in abnormal pregnancy states. Presented at the first European Congress of Endocrinology, Copenhagen, Denmark, 1987. Gewolb IH, Hobbins JC, Tan SY. Amniotic fluid cortisol in high-risk human pregnancies. Obstet Gynecol 1977; 49:466-70. Boyd PA. Placenta and umbilical cord. In: JW Keeling, ed. Fetal and neonatal pathology. New York: SpringerVerlag, 1987:45-67.
Supression of newborn natural killer cell activity by prostaglandin E2 Peter 0. Milch, PhD, William Salvatore, BS, Benjamin Luft, MD, and David A. Baker, MD Stony Brook, New York The effect of prostaglandin E2 on natural killer cell activity of cord blood was examined. Natural killer cell activity, determined by chromium 51 release, was significantly reduced after prostaglandin E2 (1 IJ.g/ml) treatment. Prostaglandin E2 has been found to enhance the cellular spread of herpesvirus. Thus prostaglandins may enhance viral infections indirectly by suppressing natural killer cell activity. (AM J 0BsTET GVNECOL 1988;159:895-6.)
Key words: Herpesvirus, prostaglandin, natural killer cell activity
The effect of prostaglandin E2 (PGE) and several frequently used drugs on natural killer cell activity of cord blood was examined in vitro. PGE has been found to enhance the cell-to-cell spread of herpesvirus in vitro. 1 Ibuprofen, a prostaglandin synthetase inhibitor, has been shown to be ineffective against herpesvirus in vivo!
Material and methods Cord blood was obtained at the time of delivery. All deliveries were at term (36 to 42 weeks) and without From the Departmmts of Obstetrics and Gynecology and Medicine, State University of New York at Stony Brook. Supported by a grant from the RGK Foundation. Received for publication November 14, 1987; accepted May 3, 1988. Reprint requests: David A. Baker, MD, Department of Obstetrics and Gynecology, HSC, T-9, State University of New York at Stony Brook, Stony Brook, NY 11794.
complications. The protocol was previously approved by the Investigational Review Board. Lymphocytes were isolated with the Ficoli-Hypaque method, washed with RPMI-1640 medium containing 10% fetal bovine serum, and counted with a Picoscale cell counter. Natural killer cell activity was determined with a chromium 51-release assay with K 562 cells used as targets. The last 2 hours of the incubation were in the presence or absence (control) of PGE (1 !J.g/ml), ibuprofen (50 IJ.g/ml), progesterone (0.5 !J.g/ml), or acyclovir (0.5 IJ.g/ml). (PGE and Ibuprofen were kindly supplied by Upjohn Co.; acyclovir was kindly supplied by Burroughs Wellcome).
Results and comment As shown in Table I, a 2-hour incubation with PGE, caused a statistically significant reduction in natural killer cell activity of cord blood (26.9% ± 3.8% control
895
Key words: Amniotic fluid, corticotropin-releasing hormone, fetal lung maturation, intrauterine growth retardation
Corticotropin-releasing hormone is released into the hypophyseal portal circulation to stimulate the secretion of adrenocorticotropic hormone (ACTH). Its release is affected by stress, circadian variation, and negative feedback.' Demonstration of corticotropinreleasing hormone-like bioactivity in human placenta 2 has given rise to the question of whether placental corticotropin-releasing hormone has any role during pregnancy. Concentrations of corticotropin-releasing hormone greatly increase in maternal plasma during the last trimester of pregnancy,'- 6 then fall rapidly after delivery. 6 These findings suggest• that increased circulating corticotropin-releasing hormone levels during the last trimester may stimulate maternal and possibly also fetal pituitary-adrenal function. Immunoreactive corticotropin-releasing hormone has been detected in umbilical cord blood.'· 5 No data seem to be available on corticotropin-releasing hormone in amniotic fluid. Because its presence in amniotic fluid may reflect fetoplacental corticotropin-releasing From the Departments I and II of Obstetrics and Gynecology, Helsinki University Central Hospital. Supported by a grant from the Academy of Finland. Received for publication January 29, 1988; revised April4, 1988; ru:cepted May 17, 1988. Reprint requests: Timo Laatikainen, MD, Departments of Obstetrics and Gynecology, Helsinki University Central Hospital, Haartmaninkatu 2, SF-00290 Helsinki, Finland_
hormone more closely than hormone in the maternal plasma, we decided to study corticotropin-releasing hormone in amniotic fluid during gestation, in labor, and in various pregnancy complications. Concentrations of corticotropin-releasing hormone were also related to parameters of fetal lung maturation in amniotic fluid and were determined in pregnancies with intrauterine growth retardation (IUGR). Material and methods
In 59 uncomplicated pregnancies, the indication for amniocentesis was determined at weeks 15 to 26 by fetal karyotype and at weeks 34 to 37 by injection of a contrast medium into the amniotic cavity for determination of fetal morphologic characteristics. At weeks 38 to 41, amniotic fluid samples were collected via amnioscope before induction of labor in ll subjects and at the early stage of spontaneous labor in 12 subjects. No fetal abnormality was detected in these cases. Amniotic fluid samples were collected from 73 complicated pregnancies for determination of fetal lung maturity. The main groups of pregnancy complications were insulin-dependent diabetes in 25 patients, pre· eclampsia in 12, diabetes and preeclampsia in 5, and cholestasis of pregnancy in 9. Six women with gestational diabetes, four with Rh incompatibility, four with placenta previa, four with essential hypertension, two 891
892
Laatikainen, Raisanen, and Salminen
October 198~ Am .J Obstet Gynecol
Table I. Concentrations of corticotropin-releasing hormone and cortisol in amniotic fluid in relation to gestational age in uncomplicated pregnancies CRH (pmol/L)
Cortisol (nmoll L)
25 4 7
9.3 ± 0.9 8.0 ± 1.0 16.0 ± 2.0
13.1 ± 1.1 44.2 ± 14 50.0 ± 3.8
11 12
20.1 ± 3.1 29.6 ± 6.5
62.8 ± 8.8 79.5 ± 10.2
Wkof gestation
15-18 19-26 34-37 38-41
No labor Labor
Data are the mean ± SEM. Corticotropin-releasing hormone.
CRH,
with premature rupture of membranes, and two with bronchial asthma were combined in a group of 22 cases with miscellaneous complications. In six pregnancies in the preeclampsia group, the birth weight of the newborn was below the lOth percentile of a large Finnish group (IUGR group); in the remaining 6 birth weight was within the 1Oth to 90th percentiles. In 14 pregnancies, two amniotic fluid samples were collected at intervals of 1 to 3 weeks. Amniotic fluid samples (5 to 10 ml) were collected into polyethylene tubes containing 125 1-11 of a protease inhibitor, aprotinin (Apronin; Medica, Helsinki, Finland), chilled on ice, and centrifuged. The supernatant was stored at - 18° C. For the assay of corticotropin-releasing hormone, synthetic human corticotropin-releasing hormone and tyrosine-corticotropin-releasing hormone for iodination were purchased from Cambridge Biochemicals Ltd., Cambridge, England, and corticotropin-releasing hormone antiserum, developed against synthetic human corticotropin-releasing hormone in rabbits, was purchased from Peninsula Laboratories Europe Ltd., St. Helens, England. 125 I-Tyr-corticotropin-releasing hormone was prepared according to the chloramine-T method. Peptides in the 1 to 2 ml samples of amniotic fluid were extracted with Sep-Pak C18 cartridges from Waters Associates, Milford, Mass., as described previously." The mean recovery of synthetic corticotropinreleasing hormone added to the amniotic fluid samples and carried through the extraction step was 73%. Radioimmunoassay of corticotropin-releasing hormone was carried out as described in detail previously. 6 The intra- and interassay variabilities of corticotropinreleasing hormone determinations in amniotic fluid were 5.2% and 10%, respectively. The concentration of cortisol was determined with a kit purchased from Farmos Diagnostica, Turku, Finland. The cross-reactivity of cortisone in this assay was 0.67%. The intra- and interassay coefficients of varia-
tion of cortisol determinations were 2.8% and 7.1 %, respectively. The lecithin/sphingomyelin ratio and phosphatidylglycerol in amniotic fluid were determined as described in detail elsewhere. 7 The Student t test and analysis of variance were used in statistical analysis of the data.
Results Results of the determinations of corticotropinreleasing hormone and cortisol in uncomplicated pregnancies are summarized in Table I. A significant increase in the concentrations of corticotropin-releasing hormone (p < 0.01) and cortisol (p < 0.001) was found from the second trimester to term. A significant correlation was found between cortisol and corticotropin-releasing hormone values (R = 0.41, p < 0.01). At weeks 38 to 41, the mean concentrations of corticotropin-releasing hormone and cortisol did not differ between the group of 12 women in early spontaneous labor and the group of 11 women who were not in labor (Table I). Table II summarizes results of the determinations of corticotropin-releasing hormone and cortisol at the third trimester in relation to pregnancy complications. The mean gestational age did not differ between the groups. Mean value of corticotropin-releasing hormone was significantly higher in the group of women with diabetes and in the women with preeclampsia and IUGR than in the group of uncomplicated pregnancies. The mean cortisol level was increased in the group of women with preeclampsia and IUGR and was decreased in the group with cholestasis of pregnancy. Concentrations of corticotropin-releasing hormone and cortisol and the parameters of fetal lung maturity in amniotic fluid are related in Table III. Samples taken in uncomplicated and complicated pregnancies at weeks 32 to 38 were included. The groups with a lecithin/sphingomyelin ratio ~2 or >2, and the groups with positive or negative phosphatidylglycerol tests, did not differ in relation to mean gestational age. Mean concentrations of both corticotropin-releasing hormone and cortisol were significantly larger in the group with a lecithin/ sphingomyelin ratio >2 than in the group with a ratio ~2. and also in the group with positive phosphatidylg1ycerol than in the group with negative tests. Follow-up data of the concentrations of corticotropin-releasing hormone and cortisol in amniotic fluid are seen in Fig. l. In 11 of the 14 paired samples of amniotic fluid studied, the result of the phosphatidylglycerol test changed from negative to positive during the follow-up. Paired comparison showed a highly significant increase (p < 0.001) in the concentrations of corticotropin-releasing hormone and cortisol with ad-
Corticotropin-releasing hormone in amniotic fluid 893
Volume 159 Number 4
140 120 100 a. 80 :I 60 -=:
0 E
a: 0
//f/
s0
~ ()
/~
40
240
250
260
270
60
E 50 c 40 0(/) 30
/
20 0 230
70
10 230
280
...--
20 240
Gestation (days)
250
260
270
280
Gestation (days)
Fig. I. Course of concentrations of corticotropin-releasing hormone (CRH) and cortisol in amniotic fluid during third trimester of pregnancy.
Table II. Concentrations of corticotropin-releasing hormone and cortisol in amniotic fluid in uncomplicated and complicated pregnancies at the third trimester of pregnancy Gestation (days)
Group
Uncomplicated Diabetes Diabetes + preeclampsia Preeclampsia IUGR No IUGR Cholestasis of pregnancy Miscellaneous complications
13 25 5 12 6 6 9 22
253 254 251 237 237 237 257 240
± ± ± ± ± ± ± ±
CRH (pmol/ L)
6 2 2 7 9 10 3 7
20 45 62 67 87 47 17 21
± ± ± ± ± ± ± ±
3 6.3* 16 16 27t 13 8.5 2.7
Cortisol ( nmol/ L)
45 ± 49 ± 54± 57 ± 66 ± 43 ± 28 ± 49 ±
3.8 3.0 8.0 6.7 8.9t 5.1 3.5* 3.6
Data are the mean ± SEM. CRH, Corticotropin-releasing hormone.
*p < 0.01
compared with group with uncomplicated pregnancy. tp < 0.05 compared with group with uncomplicated pregnancy.
Table III. Concentrations of corticotropin-releasing hormone and cortisol in amniotic fluid samples at 32 to 38 weeks' gestation in groups with a lecithin/sphingomyelin ratio o;;;2 or negative phosphatidylglycerol test (immature fetal lungs) and in groups with a lecithin/sphingomyelin ratio >2 or a positive phosphatidylglycerol test (mature fetal lungs) Group
N
Gestation (days)
CRH (pmol/L)
Cortisol (nmol/L)
LIS ratio .;;2 >2 PG
25 64
249 ± 3 254 ± 6
19.1 ± 2.9* 47.6 ± 5.3*
37.0 ± 2.4t 48.5 ± 1.9t
41 45
252 ± 2 255 ± 5
26.0 ± 3.8t 47.6 ± 5.3t
38.5 ± 1.7* 52.0 ± 2.5*
+
Data are the mean ± SE. CRH, Corticotropin-releasing hormone; Ll S, lecithin/sphingomyelin; PC, phosphatidylglycerol.
*p < 0.001
compared between groups LIS ,;;; 2 and LS > 2 or PG- and PG +. tp < 0.01 compared between groups LIS,;;; 2 and LS > 2 or PG- and PG+.
vancing gestation and with maturation of fetal lungs (Fig. 1).
Comment Significantly larger concentrations of corticotropinreleasing hormone in amniotic fluid were measured in the third than in the second trimester of pregnancy. A similar increase in corticotropin-releasing hormone
levels was found earlier in the maternal circulation.'-6 The placenta was suggested to be the source of the increased corticotropin-releasing hormone release." Recently, corticotropin-releasing hormone immunoreactivity was demonstrated by immunohistochemical staining in the cytotrophoblast of the placenta, in the decidua, and in the amnion.• Thus the amnion may be a source of corticotropin-releasing hormone
894
Laatikainen, Raisanen, and Salminen
lll amniotic fluid. in addition to the placenta and decidua. Corticotropin-releasing hormone has been shown to stimulate ACTH release by the human fetal pituitary in vitro as early as at week 14 of pregnancy.'' 10 Intravenous injection of a bolus dose (0.0 1 ~J.g/kg) of corticotropin-releasing hormone into a sheep fetus increased fetal corticotropin secretion in vivo, and the response seemed to increase with an increasing dose of corticotropin-releasing hormone. 11 The fetal corticotropin-releasing hormone level was not reported during the experiment. The lowest intravenous dose of corticotropin-releasing hormone that increased corticotropin secretion in human adults, 0.03 ~J.glkg, resulted in a corticotropin-releasing hormone peak level of 45 pmoll L in plasma. 12 This was within the range of corticotropin-releasing hormone values we found in amniotic fluid and within the range (10 to 92 pmol/L) reported earlier in cord plasma.' Thus fetal blood concentrations of corticotropin-releasing hormone may be high enough to stimulate fetal pituitary corticotropin secretion. In the present study, we found a significant correlation between concentrations of corticotropinreleasing hormone and cortisol in amniotic fluid, and the follow-up showed concomitant increases in their levels. This correlation and parallel changes support the hypothesis that corticotropin-releasing hormone originating outside the fetal hypothalamus promotes fetal pituitary-adrenal function during pregnancy. Earlier findings of an increase in the concentration of cortisol in cord plasma 1' and in amniotic fluid 1' suggested a gradual increase in human fetal adrenocortic activity with advancing gestation, but the concentration of ACTH was not found to increase in amniotic fluid during pregnancy. 1' However, ACTH in amniotic fluid may poorly reflect ACTH secretion by the fetal pituitary gland. No evidence of an increased release of corticotropin-releasing hormone or cortisol into the amniotic fluid was found in association with early spontaneous labor in the present study. Production of surfactant in the fetal lungs begins after week 24 of gestation, during the period of a rapid increase in the concentration of corticotropin-releasing hormone in the maternal plasma."·h Our follow-up data showed that at the time of the appearance of phosphatidylglycerol in amniotic fluid, a concomitant increase occurred in the concentrations of corticotropin-releasing hormone and cortisol. Moreover, corticotropin-releasing hormone and cortisol levels were higher in amniotic fluid samples with a lecithin I sphingomyelin ratio > 2 or a positive phosphatidylglycerol test, indicating mature fetal lungs, than in those with a lecithin/sphingomyelin ratio ""2 or a negative phosphatidylglycerol test, indicating immature fetal lungs. Differences in the length of gestation were
October 1988 Am J Obstet Gynecol
excluded in these comparisons, because the i-iroups were matched in relation to gestational age. A corrt'lation between the concentration of cortis(Jl and the parameters of fetal lung maturity in amniotic fluid has been regarded as evidence of the role of the fetal adrenal cortex as an inductor of fetal lung surfactant.11 Ih In the present study, corticotropin-releasing hormone and cortisol showed similar correlations with the parameters of fetal lung maturity in amniotic fluid, suggesting that corticotropin-releasing hormone may be involved in this process. Concentrations of corticotropin-releasing hormone in amniotic fluid were increased in patients with diabetes and in pregnancies complicated by preeclampsia and IUGR. Earlier data on corticotropin-releasing hormone in maternal plasma in pathologic pregnancies are scant. Increased concentrations of corticotropinreleasing hormone in maternal plasma were reported in pregnancy-associated hypertension. 17 Obstetric stress, such as preeclampsia with IUGR, promotes fetal maturation, but the mechanism is not clear. It has been generally thought to operate by stimulating the fetal adrenal cortex to produce corticosteroids. Increased cortisol levels in amniotic fluid have been reported in certain pregnancy complications 1" and may accelerate fetal lung maturation. Because corticotropinreleasing hormone is produced by the placenta and probably also in the decidua, pathologic changes in these tissues, as in preeclampsia, 19 may increase its production. We speculate that corticotropin-releasing hormone might be a "signal substance" from the placenta initiating the process of accelerated fetal maturation in obstetric stress.
REFERENCES I. Vale W, Greer M. Corticotropin-releasing factor. Fed Proc 1985;44: 145-6. 2. Shibasaki T, Odagiri E, Shizume K, Ling N. Corticotropinreleasing factor-like activity in human placental extracts. J Clin Endocrinol Metab 1982;55:384-6. 3. Sasaki A, Liotta AS, Luckey MM,Margioris AN, Suda T, Krieger DT. Immunoreactive corticotropin-releasing factor is present in human maternal plasma during the third trimester of pregnancy. J Clin Endocrinol Metab 1984;59:812-4. 4. Cn>land RS, Wardlaw SL, Stark Rl, Brown LS J r, Frantz AG. High levels of corticotropin-releasing hormone immunoreactivity in maternal and fetal plasma during pregnancy. J Clin Endocrinol Metab 1986;63:ll99-203. 5. Campbell EA, Linton EA, Wolfe CDA, Scraggs PR, Jones MT, Lowry PJ. Plasma corticotropin-releasing hormone concentrations during pregnancy and parturition. J Clin Endocrinol Metab 1987;64:1054-9. 6. Laatikainen T, Virtanen T, Raisanen I, Salminen K. Immunoreactive corticotropin-releasing factor and corticotropin in plasma during pregnancy, labor and puerperium. Neuropeptides 1987;10:343-53. 7. Hallman M, Teramo K. Measurement of the lecithin/ sphingomyelin ratio and phosphatidylglycerol in amniotic fluid: an accurate method for the assessment of
Corticotropin-releasing hormone in amniotic fluid
Volume 159 Number4
8. 9. 10. 11.
12.
13.
fetal lung maturity. Br J Obstet Gynaecol 1981;88:80613. Saijonmaa 0, Laatikainen T, Wahlstrom T. Corticotropinreleasing factor in human placenta: localization, concentration, and release in vitro. Placenta 1988;9 (in press). Ackland J, RatterS, Bourne G, Rees LH. Pro-opiomelanocortin; peptides in the human fetal pituitary. Regul Pept 1983;6:51-61. Blumenfield Z, Jaffe RB. Hypophysiotropic and neuromodulatory regulation of adrenocorticotropin in the human fetal pituitary gland. J Clin Invest 1986;78:288-94. Rose JC, Hargrave BY, Meis PJ, LaFave M, Torpe B. Corticotropin-releasing factor-induced adrenocorticotropic hormone release in the sheep fetus: blockade by cortisol. AMJ 0BSTET GYNECOL 1985;151:ll28-33. Watson SJ, Lopez JF, Young EA, Vale W, Rivier J, Akil H. Effect oflow dose ovine corticotropin-release hormone in humans: endocrine relationships and ~-endorphin/~ lipotropin responses. J Clin Endocrinol Metab 1987; 66:10-5. Smith ID, Shearman RP. Fetal plasma steroids in relation
14. 15. 16. 17.
18. 19.
to parturition. I. The effect of gestational age upon umbilical plasma corticosteroid levels following vaginal delivery. J Obstet Gynaecol Br Commonw 1974;81:11-5. de Fencl MM, Tulchinsky D. Total cortisol in amniotic fluid and fetal lung maturation. N EnglJ Med 1975;292: 133-6. Tuimala R, Kauppila A, Haapalahti J. ACTH levels in amniotic fluid during pregnancy. Br J Obstet Gynaecol 1976;83:853-6. Kresch MJ, Gross J. The biochemistry of fetal lung de-· velopment. Clin Perinatol1987;14:481-507. Wolfe CDA, Campbell EA. Lintone EA, et al. Plasma corticotropin-releasing factor (CRF) in abnormal pregnancy states. Presented at the first European Congress of Endocrinology, Copenhagen, Denmark, 1987. Gewolb IH, Hobbins JC, Tan SY. Amniotic fluid cortisol in high-risk human pregnancies. Obstet Gynecol 1977; 49:466-70. Boyd PA. Placenta and umbilical cord. In: JW Keeling, ed. Fetal and neonatal pathology. New York: SpringerVerlag, 1987:45-67.
Supression of newborn natural killer cell activity by prostaglandin E2 Peter 0. Milch, PhD, William Salvatore, BS, Benjamin Luft, MD, and David A. Baker, MD Stony Brook, New York The effect of prostaglandin E2 on natural killer cell activity of cord blood was examined. Natural killer cell activity, determined by chromium 51 release, was significantly reduced after prostaglandin E2 (1 IJ.g/ml) treatment. Prostaglandin E2 has been found to enhance the cellular spread of herpesvirus. Thus prostaglandins may enhance viral infections indirectly by suppressing natural killer cell activity. (AM J 0BsTET GVNECOL 1988;159:895-6.)
Key words: Herpesvirus, prostaglandin, natural killer cell activity
The effect of prostaglandin E2 (PGE) and several frequently used drugs on natural killer cell activity of cord blood was examined in vitro. PGE has been found to enhance the cell-to-cell spread of herpesvirus in vitro. 1 Ibuprofen, a prostaglandin synthetase inhibitor, has been shown to be ineffective against herpesvirus in vivo!
Material and methods Cord blood was obtained at the time of delivery. All deliveries were at term (36 to 42 weeks) and without From the Departmmts of Obstetrics and Gynecology and Medicine, State University of New York at Stony Brook. Supported by a grant from the RGK Foundation. Received for publication November 14, 1987; accepted May 3, 1988. Reprint requests: David A. Baker, MD, Department of Obstetrics and Gynecology, HSC, T-9, State University of New York at Stony Brook, Stony Brook, NY 11794.
complications. The protocol was previously approved by the Investigational Review Board. Lymphocytes were isolated with the Ficoli-Hypaque method, washed with RPMI-1640 medium containing 10% fetal bovine serum, and counted with a Picoscale cell counter. Natural killer cell activity was determined with a chromium 51-release assay with K 562 cells used as targets. The last 2 hours of the incubation were in the presence or absence (control) of PGE (1 !J.g/ml), ibuprofen (50 IJ.g/ml), progesterone (0.5 !J.g/ml), or acyclovir (0.5 IJ.g/ml). (PGE and Ibuprofen were kindly supplied by Upjohn Co.; acyclovir was kindly supplied by Burroughs Wellcome).
Results and comment As shown in Table I, a 2-hour incubation with PGE, caused a statistically significant reduction in natural killer cell activity of cord blood (26.9% ± 3.8% control
895