- Email: [email protected]
β-Endorphin concentrations in fetal blood during the second half of pregnancy
~- Endorphin
concentrations in fetal blood during the second half of pregnancy Nebojsa Radunovic, MD, PhD,b Charles J. Lockwood, MD: Manuel Alvarez, MD: Danica Nastic, PhD/ and Richard L. Berkowitz, MD" New York, New York, and Belgrade, Yugoslavia OBJECTIVE: Endogenous opiates may playa role in both fetal physiologic functions and the adaptation to intrauterine stress. However, our understanding of this role is hampered by an absence of data on circulating levels of these substances during fetal life. STUDY DESIGN: We measured serum J3-endorphin values with a radioimmunoassay in 81 paired fetal and maternal blood samples and 24 neonatal cord specimens. The former samples were uneventfully obtained from uncomplicated pregnancies between 18 and 39 weeks of gestation at the time of cordocentesis for prenatal diagnosis. RESULTS: Mean fetal J3-endorphin concentrations were significantly lower than J3-endorphin values from neonates (90.5 pg/ml [± 59.4] vs 228.4 pg/ml [± 166.2]; p < 0.001), but significantly higher than mean maternal values (70.5 pg/ml (±48.8]; p < 0.02). Although fetal J3-endorphin levels decreased between 18 and 28 weeks' gestation, the correlation between fetal J3-endorphin values and gestational age was not significant (r = - 0.193; P = 0.07). However, fetal J3-endorphin concentrations were significantly correlated with maternal values (Spearman's rank r = 0.47; P < 0.001). CONCLUSION: These findings suggest that delivery or fetal adaptation to an extrauterine environment is associated with significant increases in J3-endorphin release. Moreover, although the fetal pituitary may be the primary source of circulating fetal J3-endorphin, a maternal or placental contribution cannot be excluded. Our data identify a physiologic range for fetal J3-endorphin concentrations. (AM J OSSTET GVNECOL 1992;167:740-4.)
Key words: Fetal
~-endorphin,
cordocentesis
Endogenous opiates are likely to playa role in diverse fetal physiologic processes, including the regulation of blood pressure, respiratory activity, and intestinal motility.I.2 In addition, these compounds may also participate in fetal adaptation to stress.' One such opiate, ~-endorphin, has been well characterized in adults" However, our current knowledge of the ontogeny of ~-endorphin synthesis, secretion, and regulation in humans has been derived from immunohistochemistry studies of the fetal pituitary obtained after elective terminations. 6 or from analyses of neonatal cord blood samples. 7 .9 These results may therefore have been influenced by the termination or delivery process itself, and as such are not necessarily representative of a physiologic intrauterine milieu. An understanding of in
utero ~-endorphin homeostasis requires knowledge of circulating levels of fetal ~-endorphin. Unfortunately, there are no data available on ~-en dorphin concentrations in fetal blood before delivery. However, the development of percutaneous umbilical blood sampling as a safe technique for obtaining fetal blood 10 has provided access to the fetal circulation in a relatively undisturbed state. The purpose of our study was to measure ~-endorphin concentrations in fetal blood during the second half of uncomplicated pregnancy to establish "physiologic" ranges for fetal serum ~-endorphin concentrations. In addition, we sought to establish possible correlations between fetal ~-endor phin levels, gestational age, and maternal ~-endorphin concentrations.
From the Department of Obstetrics, Gynecology and Reproductive Science, Mount Sinai School of Medicine: and the Clinic of Obstetrics and Gynecology, University of Belgrade School of Medicine.' Supported by a Fulbright Foundation Fellowship (N.R.), KennedyDannreuther Fellowship, Revson Foundation Fellowship (CIL.), and the American Association of Obstetricians and Gynecologists Foundation (CIL.). Received for publication July 24, 1991; revised February 1, 1992; accepted February 28,1992. Reprint requests: Charles]. Lockwood, MD, Department of Obstetrics, Gynecology and Reproductive Sciences, Mount Sinai School of Medicine, 1 Gustave L. Levy Pl., New York, NY 10029-6574. 611137557
A cross-sectional analysis was undertaken of paired fetal and maternal serum ~-endorphin concentrations obtained between 18 and 39 weeks' gestation. Mothers were free of medical and obstetric complications, denied tobacco or illicit drug use, and took no medication other than iron and vitamin supplements. All pregnancies were uncomplicated, singleton gestations that resulted in healthy term infants whose growth was appropriate for gestational age. Only samples obtained
740
Material and methods
Fetal and maternal j3-endorphin levels 741
Volume 167 Number 3
within 30 to 120 seconds through a single needle insertion into the maternal abdomen and umbilical vein were included. In addition, samples were obtained only from those pregnancies in which the fetus was found to have umbilical pH values" and hematocrits within normal limits for gestational age 12 and in which the other relevant laboratory studies obtained (karyotype, factor VIII levels, etc.) were normal. Eighty-one paired maternal and fetal samples were available for study. Fetal gestational age was determined ante natally from reliable menstrual data confirmed by ultrasonographic scan <20 weeks' gestation. For comparison with maternal and fetal serum l3-endorphin concentrations, venous blood was also obtained from 26 nonpregnant female volunteers in the proliferative phase of their menstrual cycle and from 24 neonates 12 hours after spontaneous vaginal delivery, respectively. All specimens were obtained at the Clinic of Obstetrics and Gynecology of the University of Belgrade between May 1, 1989, and May 1, 1991. The majority of percutaneous umbilical blood sampling procedures were performed for fetal karyotype assessment in which results were needed rapidly because of advancing gestational age (n = 46). In addition, percutaneous umbilical blood sampling were carried out to rule out inherited coagulopathies (hemophilia A and B, von Willebrand disease) (n = 7), severe combined immunodeficiency disease (n = 3), fetal blood group analysis in Rh-isoimmunized mothers (n = 17), and blood gas analyses for persistent nonreactive nonstress tests (n = 8). This study was approved by the local ethics committee. In all cases a maternal peripheral blood sample was obtained before the percutaneous umbilical blood sampling procedure, both of which were performed between 8:30 AM and 12:30 PM with a 20-gauge needle and local anesthesia without maternal sedation or fetal paralysis according to previously described techniques. 1O The vessel sampled was identified ultrasonographically with a convex probe 3.75 MHz (Toshiba Model SSA 90). The umbilical vein was differentiated from the artery by assessing the direction of flow after the injection of 0.5 ml of normal saline solution. The absence of maternal blood or amniotic fluid contamination was confirmed by Kleihauer-Betke analysis and assay for coagulation factor V. A 0.5 ml aliquot of fetal serum was used for our study and stored at - 70° C until assayed. Quantitative l3-endorphin determinations were performed by l3-endorphin radioimmunoassay (Allegro, Nichols Institute, San Juan Capistrano, Calif.) according to the manufacturer's specifications. The intraassay and interassay coefficients of variation were <4.5% and <9%, respectively, with a threshold of detection of 14 pg/ml. This assay has a 16% cross-reactivity to human
13-lipotropin and <0.03% cross-reactivity to 0:- and 13melanocyte-stimulating hormone and corticotropin. All statistical analyses were performed with a statistical software package (Statgraphics, version 3.0, STSC, Rockville, Md.) and included Spearman's rank correlation, multiple linear regressions, and paired and unpaired Student's t tests. "Box-and-whisker" plots were used to demonstrate the interquartile range (box), median value (horizontal line) and absolute range (vertical line) for l3-endorphin levels. Values are presented as mean ± standard deviation. A p value of <0.05 was considered significant.
Results Fetal l3-endorphin values across gestation are presented in Fig. 1. A decrease in fetall3-endorphin levels occurred between 18 and 28 weeks, but no significant correlation was found between fetal l3-endorphin values and gestational age (r = -0.193; P = 0.07). Mean fetal l3-endorphin levels were significantly lower than l3-endorphin values obtained from neonates (90.5 pg/ml [±59.5] vs 228.4 pg/ml [± 166.1]; p < 0.001). The overall mean fetal serum l3-endorphin concentration was significantly higher than the overall mean maternal (3-endorphin value (Table I). However, when assessing fetal and maternal (3-endorphin levels in increments of 5 weeks' gestation fetal and maternal values were found to differ significantly only in the 18- to 22week interval. In addition, a correlation was found between maternal and fetal l3-endorphin levels (Spearman's rank correlation r = 0.47; P < 0.001). Maternal (3-endorphin concentrations across the second half of pregnancy and nonpregnant control values are presented in Fig. 2. Maternal l3-endorphin values displayed a significant though minimal correlation with gestational age (r = 0.216; P = 0.049). Mean concentrations of serum (3-endorphin were significantly reduced in the overall pregnant patient population compared with nonpregnant controls (70.5 pg/ml [± 48.8] vs 114.6 pg/ml [± 59.7]; p < 0.001), but not during the last 4 weeks of pregnancy (94.8 [± 56.6] vs 114.6 pg/ml [±59.7]; p = 0.27). There were no significant differences in age between the pregnant patients and the nonpregnant controls (34.9 yrs [± 5.7] vs 33.4 yrs [± 8.8]; p = 0.46); both groups had a median parity of one.
Comment There is no information available on physiologic patterns of (3-endorphin release into the fetal circulation during intrauterine life in humans. The availability of percutaneous umbilical blood sampling has provided an opportunity to obtain such data, but it is unclear whether this invasive procedure can truly assess a physiologic fetal state. This conundrum is particularly rel-
742
Radunovic et al.
September 1992 Am J Obstet Gynecol
500 n=24
--E
CI
400
C
300
Co
.c
...
n=13
Co 0
'0 C
Q)
200
n=21
I
n=28
«S
Q)
.c
100
0 18 • 22
~ ~
$
23 • 28
35 • 40
29 • 34
n:19
neonatal controls
weeks Fig. 1. Box-and-whisker plot of serum l3-endorphin throughout gestation in 81 fetuses and 12 hours after delivery in 24 neonates. Gestational ages are grouped in 6-week intervals from 18 to 40 weeks. Although values declined between 18 and 28 weeks, no correlation was found with gestational age.
Table I. Serum l3-endorphin levels (mean ± SD) in unstressed fetal and maternal matched pairs from 18 to 40 weeks' gestation Matched pairs Gestational wk
n
(18-40)
(81)
18-22 23-28 29-34 35-40
13 21 28 19
Maternal
(70.5 ± 48.8)
56.4 62.2 75.4 94.8
± ± ± ±
35.4 51.5 61.8 56.6
evant to the assessment of fetal opiate homeostasis because any significant stress is likely to promote the release of these substances. 13 To minimize pertubations in fetal opiate homeostasis we examined l3-endorphin concentrations obtained from uncomplicated pregnancies that resulted in the delivery of healthy infants and in which the fetal blood sample was obtained without difficulty via a single needle insertion. Fetal l3-endorphin values obtained from these uncomplicated procedures were significantly lower than neonatal values. Thus delivery or fetal adaptation to an extrauterine environment appears to be associated with significant elevations in j3-endorphin release, as had been hypothesized by others. I'. 15 The precise source(s) of l3-endorphin in the fetal circulation remains unknown. Goland et al. 16 contend that fetal j3-endorphin is not of maternal origin because they could not demonstrate a correlation between maternal and neonatal l3-endorphin levels. Although the human placenta can synthesize l3-endorphin,I7 Wardlaw et alY demonstrated higher umbilical artery levels compared
I
Fetal
(90.5 ± 59.4)
105.9 86.4 89.7 85.5
± 70.3 ± 55.7 ± 67.1
± 44.8
Paired t test p Value «0.02)
<0.05 0.17 0.42 0.58
= = =
with umbilical venous (3-endorphin levels in neonates, a finding not consistent with a major placental contribution to fetalj3-endorphin levels. In contrast, Goebelsmann et al. 18 found higher umbilical venous levels compared with umbilical artery j3-endorphin values in unstressed neonates. 18 Unfortunately, these studies relied on neonatal observations which, as noted previously, may not reflect physiologic intrauterine conditions. We noted modestly higher j3-endorphin levels in fetal blood compared with maternal blood, which suggests that the fetal pituitary is the source of the preponderance of circulating fetal j3-endorphin. However, the correlation between maternal and fetal j3-endorphin concentrations observed in our study may reflect maternal or placental contribution to fetal l3-endorphin levels. Petrucha et al. 3 observed that j3-endorphin concentrations in amniotic fluid were significantly higher in specimens obtained from the second trimester compared with the third trimester.' Although j3-endorphin values in the amniotic fluid may have little relevance to
Fetal and maternal [3-endorphin levels 743
Volume 167 Number 3
400
.!§
DI
300
C.
n=19 C
or:
...C.0
200
'0 C
,
a:I
Q)
.c
n=28
n=13
Q)
100
0
n=26
~ 18 -
22
th n=21
23 _ 28
nonpregnant controls 29 - 34
35 _ 40
weeks Fig. 2. Box-and-whisker plot of serum [3-endorphin levels throughout gestation in 81 pregnant women and 26 nonpregnant female controls. Gestational ages are grouped in 6-week intervals from 18 to 40 weeks. Maternal [3-endorphin values correlated minimally with gestational age (r = 0.216; p = 0.049).
fetal blood values, the observations of Petrucha et a1. are consistent with our findings of higher fetal ~-en dorphin concentrations at <24 weeks' gestation. Moreover, we found that fetal ~-endorphin levels were significantly elevated in comparison with maternal values only between 18 and 22 weeks' gestation. There are conflicting reports on the pattern of maternal ~-endorphin secretion during pregnancy. Goebelsmann et a1. 18 found a significant decrease in plasma ~-endorphin concentrations during pregnancy, with a nadir in the second trimester. 18 Genazzani et a1. 19 also reported a significant reduction in maternal plasma ~ endorphin levels between 9 and 12 weeks' gestation but noted a subsequent small rise to maximum values at 36 to 37 weeks. In contrast, others have found a modest rise in maternal ~-endorphin levels throughout pregnancy, reaching statistical significance in the third trimester. 2o • 21 This wide variability in published results may be explained in part by selection biases, the number of subjects studied, and the varying accuracy of the assays used. In addition, this observed variability may reflect enormous individual variation among circulating ~-endorphin levels. We noted that concentrations of maternal ~-endorphin are increased during the second half of pregnancy but were significantly lower than nonpregnant values. This decrease is consistent with previous reports. 18.22 Our study provides data on the physiologic range of fetal serum ~-endorphin concentrations and provides insights into the regulation of~-endorphin homeostasis in utero. Because increased concentrations of umbilical ~-endorphin are a biochemical marker of intrapartum
hypoxia,13 fetal serum ~-endorphin values that exceed these ranges may be an indicator of fetal stress. Potential indications for such measurements include persistent non reassuring testing in a premature fetus and suspected intrauterine growth retardation. REFERENCES 1. Espinoza M, Riquelme R, Germain AM, Tevah J, Parer JT, Llanos AJ. Role of endogenous opioids in the cardiovascular responses to asphyxia in fetal sheep. Am J Physiol 1989;256:RI063-8. 2. Jansen AH, Ioffe S, Chernick V. Influence of naloxone on fetal breathing and the respiratory response to hypercapnia. Respir Physiol 1989;78(2):187-96. 3. Petrucha RA, Goebelsman U, Hunt TT, Haase HR, Lobo RA. Amniotic fluid [3-endorphin and [3-lipotropin concentrations during the second and third trimesters. AM J OBSTET GVNECOL 1983;146:644-51. 4. Kreiger DT, Martin JB. Brain peptides. New Engl J Med 1981 ;304:876-85. 5. Facchinetti F, Storchi AR, Petraglia F, Garuti G, Genazzani AR. Ontogeny of pituitary [3-endorphin and related peptides in the human embryo and fetus. AM J OBSTET GyNECOL 1987;156:735-9. 6. Surico N, Lanzani A, Crivello T, Debbi C, Porcelli A. Maternal and embryonal/fetal beta-endorphin concentrations during the first trimester of pregnancy. Eur J Obstet Gynecol Reprod Bioi 1989;31:207-11. 7. Kimball CD, Chang CM, Huang SM, Houck JC. Immunoreactive endorphin peptides and prolactin in umbilical vein and maternal blood. AM J OBSTET GVNECOL 1981; 140: 157-64. 8. Shaaban M, Hung TT, Hoffman DI, Lobo RA, Goebelsmann U. [3-endorphin and [3-lipotropin concentrations in umbilical cord blood. AM J OBSTET GVNECOL 1982;144:560-8. 9. Ramanthan S, Puig MM, Turndorf H. Plasma beta-endorphin levels in the umbilical cord blood of preterm human neonates. Bioi Neonate 1989;56:117-20. 10. Daffos F, Cappela-Pavlosky M, Forestier F. Fetal blood
744
II.
12.
13.
14. 15. 16.
Radunovic et al.
sampling via umbilical cord using a needle guided by ultrasound. Prenat Diagn 1983;3:271-7. Soothill PW, Nicolaides KH, Rodeck CH, Campbell S. Effect of gestational age on fetal and intervillous blood gas and acid base values in human pregnancy. Fetal Ther 1986; 1:168-75. Forestier F, Daffos F, Galacteroso P, Bradakjian], Rainaut M, Beuzard Y. Hematological values of 163 normal fetuses between 18 and 30 weeks of gestation. Pediatr Res 1986;20:342-6. Wardlaw SL, Stark RI, Baxi L, Frantz AG. Plasma betaendorphin and beta-lipotropin in the human fetus at delivery: correlation with arterial pH and O 2 • ] Clin Endocrinol Metab 1979;49:888-91. Facchinetti A, Bagnoli F, Bracci R, Genazzani AR. Plasma opioids in the first hours of life. Pediatr Res 1982;16: 95-8. Facchinetti F, Lanzani A, Genazzani AR. Fetal intermediate lobe is stimulated by parturition. AM] OBSTET GyNECOL 1989;161:1267-70. Goland RS, Wardlaw SL, Stark RI, Frantz AG. Human plasma beta-endorphin during pregnancy, labor and delivery.] Clin Endocrinol Metab 1981;52:74-8.
September 1992 Am J Obstet Gynecol
17. Odagiri E, Sherrell B], Mount CD, Nicholson WE, Orth DN. Human placental immunoreactive corticotropin, lipotropin and beta-endotrophin: evidence for a common precursor. Proc Nat! Acad Sci USA 1979;76:2027-31. 18. Goebelsmann U, Abboud TK, Hoffman DI, Hung TT. Beta-endorphin in pregnancy. Eur] Obstet Gynecol Reprod Bioi 1984;17:77-89. 19. Genazzani AR, Facchinetti F, Parrini D. Beta-lipotropin and beta-endotrophin plasma levels during pregnancy. Clin Endocrinol (Oxf) 1981;14:409-18. 20. Newnhan]P, Tomlin S, Ratter S], Bourne GL, Rees LH. Endogenous opioid peptides in pregnancy. Br ] Obstet GynaecoI1983;90:535-8. 21. Browning A]F, Butt WR, Lynch SS, Shakespear RA. Maternal plasma concentrations of beta-lipotropin, beta-endorphin, and t-lipotropin throughout pregnancy. Br] Obstet Gynaecol 1983;90;1147-51. 22. Hoffman Dl, Abboud TK, Haase HR, Hung TT, Goebelsmann U. Plasma J3-endorphin concentrations prior to and during pregnancy, in labor, and after delivery. AM] OBSTET GYNECOL 1984;150:492-6.
concentrations in fetal blood during the second half of pregnancy Nebojsa Radunovic, MD, PhD,b Charles J. Lockwood, MD: Manuel Alvarez, MD: Danica Nastic, PhD/ and Richard L. Berkowitz, MD" New York, New York, and Belgrade, Yugoslavia OBJECTIVE: Endogenous opiates may playa role in both fetal physiologic functions and the adaptation to intrauterine stress. However, our understanding of this role is hampered by an absence of data on circulating levels of these substances during fetal life. STUDY DESIGN: We measured serum J3-endorphin values with a radioimmunoassay in 81 paired fetal and maternal blood samples and 24 neonatal cord specimens. The former samples were uneventfully obtained from uncomplicated pregnancies between 18 and 39 weeks of gestation at the time of cordocentesis for prenatal diagnosis. RESULTS: Mean fetal J3-endorphin concentrations were significantly lower than J3-endorphin values from neonates (90.5 pg/ml [± 59.4] vs 228.4 pg/ml [± 166.2]; p < 0.001), but significantly higher than mean maternal values (70.5 pg/ml (±48.8]; p < 0.02). Although fetal J3-endorphin levels decreased between 18 and 28 weeks' gestation, the correlation between fetal J3-endorphin values and gestational age was not significant (r = - 0.193; P = 0.07). However, fetal J3-endorphin concentrations were significantly correlated with maternal values (Spearman's rank r = 0.47; P < 0.001). CONCLUSION: These findings suggest that delivery or fetal adaptation to an extrauterine environment is associated with significant increases in J3-endorphin release. Moreover, although the fetal pituitary may be the primary source of circulating fetal J3-endorphin, a maternal or placental contribution cannot be excluded. Our data identify a physiologic range for fetal J3-endorphin concentrations. (AM J OSSTET GVNECOL 1992;167:740-4.)
Key words: Fetal
~-endorphin,
cordocentesis
Endogenous opiates are likely to playa role in diverse fetal physiologic processes, including the regulation of blood pressure, respiratory activity, and intestinal motility.I.2 In addition, these compounds may also participate in fetal adaptation to stress.' One such opiate, ~-endorphin, has been well characterized in adults" However, our current knowledge of the ontogeny of ~-endorphin synthesis, secretion, and regulation in humans has been derived from immunohistochemistry studies of the fetal pituitary obtained after elective terminations. 6 or from analyses of neonatal cord blood samples. 7 .9 These results may therefore have been influenced by the termination or delivery process itself, and as such are not necessarily representative of a physiologic intrauterine milieu. An understanding of in
utero ~-endorphin homeostasis requires knowledge of circulating levels of fetal ~-endorphin. Unfortunately, there are no data available on ~-en dorphin concentrations in fetal blood before delivery. However, the development of percutaneous umbilical blood sampling as a safe technique for obtaining fetal blood 10 has provided access to the fetal circulation in a relatively undisturbed state. The purpose of our study was to measure ~-endorphin concentrations in fetal blood during the second half of uncomplicated pregnancy to establish "physiologic" ranges for fetal serum ~-endorphin concentrations. In addition, we sought to establish possible correlations between fetal ~-endor phin levels, gestational age, and maternal ~-endorphin concentrations.
From the Department of Obstetrics, Gynecology and Reproductive Science, Mount Sinai School of Medicine: and the Clinic of Obstetrics and Gynecology, University of Belgrade School of Medicine.' Supported by a Fulbright Foundation Fellowship (N.R.), KennedyDannreuther Fellowship, Revson Foundation Fellowship (CIL.), and the American Association of Obstetricians and Gynecologists Foundation (CIL.). Received for publication July 24, 1991; revised February 1, 1992; accepted February 28,1992. Reprint requests: Charles]. Lockwood, MD, Department of Obstetrics, Gynecology and Reproductive Sciences, Mount Sinai School of Medicine, 1 Gustave L. Levy Pl., New York, NY 10029-6574. 611137557
A cross-sectional analysis was undertaken of paired fetal and maternal serum ~-endorphin concentrations obtained between 18 and 39 weeks' gestation. Mothers were free of medical and obstetric complications, denied tobacco or illicit drug use, and took no medication other than iron and vitamin supplements. All pregnancies were uncomplicated, singleton gestations that resulted in healthy term infants whose growth was appropriate for gestational age. Only samples obtained
740
Material and methods
Fetal and maternal j3-endorphin levels 741
Volume 167 Number 3
within 30 to 120 seconds through a single needle insertion into the maternal abdomen and umbilical vein were included. In addition, samples were obtained only from those pregnancies in which the fetus was found to have umbilical pH values" and hematocrits within normal limits for gestational age 12 and in which the other relevant laboratory studies obtained (karyotype, factor VIII levels, etc.) were normal. Eighty-one paired maternal and fetal samples were available for study. Fetal gestational age was determined ante natally from reliable menstrual data confirmed by ultrasonographic scan <20 weeks' gestation. For comparison with maternal and fetal serum l3-endorphin concentrations, venous blood was also obtained from 26 nonpregnant female volunteers in the proliferative phase of their menstrual cycle and from 24 neonates 12 hours after spontaneous vaginal delivery, respectively. All specimens were obtained at the Clinic of Obstetrics and Gynecology of the University of Belgrade between May 1, 1989, and May 1, 1991. The majority of percutaneous umbilical blood sampling procedures were performed for fetal karyotype assessment in which results were needed rapidly because of advancing gestational age (n = 46). In addition, percutaneous umbilical blood sampling were carried out to rule out inherited coagulopathies (hemophilia A and B, von Willebrand disease) (n = 7), severe combined immunodeficiency disease (n = 3), fetal blood group analysis in Rh-isoimmunized mothers (n = 17), and blood gas analyses for persistent nonreactive nonstress tests (n = 8). This study was approved by the local ethics committee. In all cases a maternal peripheral blood sample was obtained before the percutaneous umbilical blood sampling procedure, both of which were performed between 8:30 AM and 12:30 PM with a 20-gauge needle and local anesthesia without maternal sedation or fetal paralysis according to previously described techniques. 1O The vessel sampled was identified ultrasonographically with a convex probe 3.75 MHz (Toshiba Model SSA 90). The umbilical vein was differentiated from the artery by assessing the direction of flow after the injection of 0.5 ml of normal saline solution. The absence of maternal blood or amniotic fluid contamination was confirmed by Kleihauer-Betke analysis and assay for coagulation factor V. A 0.5 ml aliquot of fetal serum was used for our study and stored at - 70° C until assayed. Quantitative l3-endorphin determinations were performed by l3-endorphin radioimmunoassay (Allegro, Nichols Institute, San Juan Capistrano, Calif.) according to the manufacturer's specifications. The intraassay and interassay coefficients of variation were <4.5% and <9%, respectively, with a threshold of detection of 14 pg/ml. This assay has a 16% cross-reactivity to human
13-lipotropin and <0.03% cross-reactivity to 0:- and 13melanocyte-stimulating hormone and corticotropin. All statistical analyses were performed with a statistical software package (Statgraphics, version 3.0, STSC, Rockville, Md.) and included Spearman's rank correlation, multiple linear regressions, and paired and unpaired Student's t tests. "Box-and-whisker" plots were used to demonstrate the interquartile range (box), median value (horizontal line) and absolute range (vertical line) for l3-endorphin levels. Values are presented as mean ± standard deviation. A p value of <0.05 was considered significant.
Results Fetal l3-endorphin values across gestation are presented in Fig. 1. A decrease in fetall3-endorphin levels occurred between 18 and 28 weeks, but no significant correlation was found between fetal l3-endorphin values and gestational age (r = -0.193; P = 0.07). Mean fetal l3-endorphin levels were significantly lower than l3-endorphin values obtained from neonates (90.5 pg/ml [±59.5] vs 228.4 pg/ml [± 166.1]; p < 0.001). The overall mean fetal serum l3-endorphin concentration was significantly higher than the overall mean maternal (3-endorphin value (Table I). However, when assessing fetal and maternal (3-endorphin levels in increments of 5 weeks' gestation fetal and maternal values were found to differ significantly only in the 18- to 22week interval. In addition, a correlation was found between maternal and fetal l3-endorphin levels (Spearman's rank correlation r = 0.47; P < 0.001). Maternal (3-endorphin concentrations across the second half of pregnancy and nonpregnant control values are presented in Fig. 2. Maternal l3-endorphin values displayed a significant though minimal correlation with gestational age (r = 0.216; P = 0.049). Mean concentrations of serum (3-endorphin were significantly reduced in the overall pregnant patient population compared with nonpregnant controls (70.5 pg/ml [± 48.8] vs 114.6 pg/ml [± 59.7]; p < 0.001), but not during the last 4 weeks of pregnancy (94.8 [± 56.6] vs 114.6 pg/ml [±59.7]; p = 0.27). There were no significant differences in age between the pregnant patients and the nonpregnant controls (34.9 yrs [± 5.7] vs 33.4 yrs [± 8.8]; p = 0.46); both groups had a median parity of one.
Comment There is no information available on physiologic patterns of (3-endorphin release into the fetal circulation during intrauterine life in humans. The availability of percutaneous umbilical blood sampling has provided an opportunity to obtain such data, but it is unclear whether this invasive procedure can truly assess a physiologic fetal state. This conundrum is particularly rel-
742
Radunovic et al.
September 1992 Am J Obstet Gynecol
500 n=24
--E
CI
400
C
300
Co
.c
...
n=13
Co 0
'0 C
Q)
200
n=21
I
n=28
«S
Q)
.c
100
0 18 • 22
~ ~
$
23 • 28
35 • 40
29 • 34
n:19
neonatal controls
weeks Fig. 1. Box-and-whisker plot of serum l3-endorphin throughout gestation in 81 fetuses and 12 hours after delivery in 24 neonates. Gestational ages are grouped in 6-week intervals from 18 to 40 weeks. Although values declined between 18 and 28 weeks, no correlation was found with gestational age.
Table I. Serum l3-endorphin levels (mean ± SD) in unstressed fetal and maternal matched pairs from 18 to 40 weeks' gestation Matched pairs Gestational wk
n
(18-40)
(81)
18-22 23-28 29-34 35-40
13 21 28 19
Maternal
(70.5 ± 48.8)
56.4 62.2 75.4 94.8
± ± ± ±
35.4 51.5 61.8 56.6
evant to the assessment of fetal opiate homeostasis because any significant stress is likely to promote the release of these substances. 13 To minimize pertubations in fetal opiate homeostasis we examined l3-endorphin concentrations obtained from uncomplicated pregnancies that resulted in the delivery of healthy infants and in which the fetal blood sample was obtained without difficulty via a single needle insertion. Fetal l3-endorphin values obtained from these uncomplicated procedures were significantly lower than neonatal values. Thus delivery or fetal adaptation to an extrauterine environment appears to be associated with significant elevations in j3-endorphin release, as had been hypothesized by others. I'. 15 The precise source(s) of l3-endorphin in the fetal circulation remains unknown. Goland et al. 16 contend that fetal j3-endorphin is not of maternal origin because they could not demonstrate a correlation between maternal and neonatal l3-endorphin levels. Although the human placenta can synthesize l3-endorphin,I7 Wardlaw et alY demonstrated higher umbilical artery levels compared
I
Fetal
(90.5 ± 59.4)
105.9 86.4 89.7 85.5
± 70.3 ± 55.7 ± 67.1
± 44.8
Paired t test p Value «0.02)
<0.05 0.17 0.42 0.58
= = =
with umbilical venous (3-endorphin levels in neonates, a finding not consistent with a major placental contribution to fetalj3-endorphin levels. In contrast, Goebelsmann et al. 18 found higher umbilical venous levels compared with umbilical artery j3-endorphin values in unstressed neonates. 18 Unfortunately, these studies relied on neonatal observations which, as noted previously, may not reflect physiologic intrauterine conditions. We noted modestly higher j3-endorphin levels in fetal blood compared with maternal blood, which suggests that the fetal pituitary is the source of the preponderance of circulating fetal j3-endorphin. However, the correlation between maternal and fetal j3-endorphin concentrations observed in our study may reflect maternal or placental contribution to fetal l3-endorphin levels. Petrucha et al. 3 observed that j3-endorphin concentrations in amniotic fluid were significantly higher in specimens obtained from the second trimester compared with the third trimester.' Although j3-endorphin values in the amniotic fluid may have little relevance to
Fetal and maternal [3-endorphin levels 743
Volume 167 Number 3
400
.!§
DI
300
C.
n=19 C
or:
...C.0
200
'0 C
,
a:I
Q)
.c
n=28
n=13
Q)
100
0
n=26
~ 18 -
22
th n=21
23 _ 28
nonpregnant controls 29 - 34
35 _ 40
weeks Fig. 2. Box-and-whisker plot of serum [3-endorphin levels throughout gestation in 81 pregnant women and 26 nonpregnant female controls. Gestational ages are grouped in 6-week intervals from 18 to 40 weeks. Maternal [3-endorphin values correlated minimally with gestational age (r = 0.216; p = 0.049).
fetal blood values, the observations of Petrucha et a1. are consistent with our findings of higher fetal ~-en dorphin concentrations at <24 weeks' gestation. Moreover, we found that fetal ~-endorphin levels were significantly elevated in comparison with maternal values only between 18 and 22 weeks' gestation. There are conflicting reports on the pattern of maternal ~-endorphin secretion during pregnancy. Goebelsmann et a1. 18 found a significant decrease in plasma ~-endorphin concentrations during pregnancy, with a nadir in the second trimester. 18 Genazzani et a1. 19 also reported a significant reduction in maternal plasma ~ endorphin levels between 9 and 12 weeks' gestation but noted a subsequent small rise to maximum values at 36 to 37 weeks. In contrast, others have found a modest rise in maternal ~-endorphin levels throughout pregnancy, reaching statistical significance in the third trimester. 2o • 21 This wide variability in published results may be explained in part by selection biases, the number of subjects studied, and the varying accuracy of the assays used. In addition, this observed variability may reflect enormous individual variation among circulating ~-endorphin levels. We noted that concentrations of maternal ~-endorphin are increased during the second half of pregnancy but were significantly lower than nonpregnant values. This decrease is consistent with previous reports. 18.22 Our study provides data on the physiologic range of fetal serum ~-endorphin concentrations and provides insights into the regulation of~-endorphin homeostasis in utero. Because increased concentrations of umbilical ~-endorphin are a biochemical marker of intrapartum
hypoxia,13 fetal serum ~-endorphin values that exceed these ranges may be an indicator of fetal stress. Potential indications for such measurements include persistent non reassuring testing in a premature fetus and suspected intrauterine growth retardation. REFERENCES 1. Espinoza M, Riquelme R, Germain AM, Tevah J, Parer JT, Llanos AJ. Role of endogenous opioids in the cardiovascular responses to asphyxia in fetal sheep. Am J Physiol 1989;256:RI063-8. 2. Jansen AH, Ioffe S, Chernick V. Influence of naloxone on fetal breathing and the respiratory response to hypercapnia. Respir Physiol 1989;78(2):187-96. 3. Petrucha RA, Goebelsman U, Hunt TT, Haase HR, Lobo RA. Amniotic fluid [3-endorphin and [3-lipotropin concentrations during the second and third trimesters. AM J OBSTET GVNECOL 1983;146:644-51. 4. Kreiger DT, Martin JB. Brain peptides. New Engl J Med 1981 ;304:876-85. 5. Facchinetti F, Storchi AR, Petraglia F, Garuti G, Genazzani AR. Ontogeny of pituitary [3-endorphin and related peptides in the human embryo and fetus. AM J OBSTET GyNECOL 1987;156:735-9. 6. Surico N, Lanzani A, Crivello T, Debbi C, Porcelli A. Maternal and embryonal/fetal beta-endorphin concentrations during the first trimester of pregnancy. Eur J Obstet Gynecol Reprod Bioi 1989;31:207-11. 7. Kimball CD, Chang CM, Huang SM, Houck JC. Immunoreactive endorphin peptides and prolactin in umbilical vein and maternal blood. AM J OBSTET GVNECOL 1981; 140: 157-64. 8. Shaaban M, Hung TT, Hoffman DI, Lobo RA, Goebelsmann U. [3-endorphin and [3-lipotropin concentrations in umbilical cord blood. AM J OBSTET GVNECOL 1982;144:560-8. 9. Ramanthan S, Puig MM, Turndorf H. Plasma beta-endorphin levels in the umbilical cord blood of preterm human neonates. Bioi Neonate 1989;56:117-20. 10. Daffos F, Cappela-Pavlosky M, Forestier F. Fetal blood
744
II.
12.
13.
14. 15. 16.
Radunovic et al.
sampling via umbilical cord using a needle guided by ultrasound. Prenat Diagn 1983;3:271-7. Soothill PW, Nicolaides KH, Rodeck CH, Campbell S. Effect of gestational age on fetal and intervillous blood gas and acid base values in human pregnancy. Fetal Ther 1986; 1:168-75. Forestier F, Daffos F, Galacteroso P, Bradakjian], Rainaut M, Beuzard Y. Hematological values of 163 normal fetuses between 18 and 30 weeks of gestation. Pediatr Res 1986;20:342-6. Wardlaw SL, Stark RI, Baxi L, Frantz AG. Plasma betaendorphin and beta-lipotropin in the human fetus at delivery: correlation with arterial pH and O 2 • ] Clin Endocrinol Metab 1979;49:888-91. Facchinetti A, Bagnoli F, Bracci R, Genazzani AR. Plasma opioids in the first hours of life. Pediatr Res 1982;16: 95-8. Facchinetti F, Lanzani A, Genazzani AR. Fetal intermediate lobe is stimulated by parturition. AM] OBSTET GyNECOL 1989;161:1267-70. Goland RS, Wardlaw SL, Stark RI, Frantz AG. Human plasma beta-endorphin during pregnancy, labor and delivery.] Clin Endocrinol Metab 1981;52:74-8.
September 1992 Am J Obstet Gynecol
17. Odagiri E, Sherrell B], Mount CD, Nicholson WE, Orth DN. Human placental immunoreactive corticotropin, lipotropin and beta-endotrophin: evidence for a common precursor. Proc Nat! Acad Sci USA 1979;76:2027-31. 18. Goebelsmann U, Abboud TK, Hoffman DI, Hung TT. Beta-endorphin in pregnancy. Eur] Obstet Gynecol Reprod Bioi 1984;17:77-89. 19. Genazzani AR, Facchinetti F, Parrini D. Beta-lipotropin and beta-endotrophin plasma levels during pregnancy. Clin Endocrinol (Oxf) 1981;14:409-18. 20. Newnhan]P, Tomlin S, Ratter S], Bourne GL, Rees LH. Endogenous opioid peptides in pregnancy. Br ] Obstet GynaecoI1983;90:535-8. 21. Browning A]F, Butt WR, Lynch SS, Shakespear RA. Maternal plasma concentrations of beta-lipotropin, beta-endorphin, and t-lipotropin throughout pregnancy. Br] Obstet Gynaecol 1983;90;1147-51. 22. Hoffman Dl, Abboud TK, Haase HR, Hung TT, Goebelsmann U. Plasma J3-endorphin concentrations prior to and during pregnancy, in labor, and after delivery. AM] OBSTET GYNECOL 1984;150:492-6.