Adrenomedullin is a novel peptide, widely distributed in human tissues and circulating in plasma, that influences control of systemic and local circulation and humoral secretion.1,2 It has been reported that cultured vascular smooth-muscle cells contain adrenomedullin mRNA and that endothelial cells produce adrenomedullin, which acts on specific receptors, eliciting long-lasting vasodilation through synthesis of cyclic AMP.3 Immunocytochemistry studies showed that adrenomedullin is distributed widely and modulates hormonal secretion.4 Adrenomedullin inhibits the secretion of ACTH, aldosterone, and insulin, and its secretion is stimulated by thyroid hormones, progesterone, and dexamethasone.5
We reported high concentrations of adrenomedullin in maternal and umbilical cord plasma, and in amniotic fluid, in term human pregnancies. The immunoreactive peptide was also in the placenta and fetal membranes,6 suggesting potential influence of the peptide during gestation. Epithelial cells of fetal membranes express adrenomedullin mRNA, and immunoreactive adrenomedullin was detected in amniotic fluid midgestation,7 suggesting that placental adrenomedullin might regulate hormonal and peptide secretion by placenta and fetal membranes. Recent studies found that adrenomedullin might modulate vascular tone as a paracrine regulator through the inhibition or stimulation of vasoactive agents, such as nitric oxide8 and endothelin,9 both of which are produced by human placenta and fetal membranes and involved in the control of myometrial contractility.10,11 We hypothesize that adrenomedullin, directly through the cyclic AMP or indirectly through the regulation of nitric oxide and endothelin or other modulators (eg, corticotrophin releasing hormone, and prostaglandins) could also affect uterine contractility.
To determine whether placental adrenomedullin secretion was increased in association with preterm labor, we measured its concentration in amniotic fluid. Because adrenomedullin acts as a local mediator in an autocrine-paracrine manner,12 amniotic fluid adrenomedullin, more than maternal plasma, might better indicate its production from placental tissues.
Material and Methods
Enrolled in the study were 61 women admitted to the 2nd Institute of Obstetrics and Gynecology, “La Sapienza” University, Rome, Italy, between March 1996 and December 1997, with indications for amniocentesis, between 27 and 35 weeks' gestation. After exclusion of women with multiple pregnancies, fetal anomalies, diabetes, preeclampsia, and fetal growth restriction, 27 with presumed preterm labor and nine with premature rupture of membranes (PROM) had amniocentesis for fetal lung-maturity assessment. We also studied 25 pregnant women without uterine contractions at the same gestational ages (range 26–35 weeks' gestation) who had amniocentesis to exclude fetal infection by cytomegalovirus (polymerase chain reaction technique) suspected on the basis of maternal serology, according to a protocol at our institute.13 Eighteen of the women had prenatal diagnoses and neonatal outcomes that were negative for infection. They were used as controls.
Preterm labor was defined as regular uterine contractions, at least two every 10 minutes over a period of at least 60 minutes, and cervical dilatation at least 2 cm. Women with cervical dilatation greater than 4 cm were not included because betamethasone was given immediately after admission and amniocentesis was not done. None of the subjects had clinical evidence of infection at the time of sampling, and all amniocentesis were done within 24 hours from admission (mean 11 ± 6 hours).
Women with PROM who were included did not have preterm labor at the time of sampling, although they all received prophylactic oral ritrodine, 10 mg twice daily (Miolene, Lusofarmaco S.p.a., Milan, Italy), but no antibiotic. Women with preterm labor received intravenous ritrodine infusion (800–1200 mg daily) at the time of admission. On the basis of fetal lung maturity tests, 20 women received betamethasone for stimulation of fetal lung maturity, but amniocentesis was done before administration of glucocorticoids.
Women with preterm labor were categorized according to their response to tocolysis; 17 responded and contractions stopped for at least 7 days, whereas ten had dilation progress beyond 5 cm and delivered within a few days (range 1–6 days). One women delivered at ± days after sampling, by cesarean without labor; she was included in the responder group because contractions stopped after tocolysis until cesarean was done. Clinical and demographic characteristics are shown in Table 1. This study was approved by the local ethics committee and informed consent was obtained from all participants.
Amniotic fluid samples were collected by transab-dominal amniocentesis and immediately centrifuged at 1800 g for 15 minutes at 4C. The supernatants were divided into aliquots and stored at −80C until assayed. Amniotic fluid adrenomedullin concentration was measured by radioimmunoassay after extraction and purification using a commercial kit (Phoenix Pharmaceuticals Inc, Mountain View, CA) with rabbit polyclonal antibody raised against human adrenomedullin 1–52.14 The antibody cross-reacts 100% with human adrenomedullin but not with rat adrenomedullin, amylin, calcitonin gene-related peptide, endothelin-1, or α-atrial natriuretic peptide. The intra- and interassay coefficients of variance were 5.1% and 12.0%, respectively.
All tissues were evaluated carefully at the time of delivery for histologic evidence of chorionamnionitis, defined as the presence of infiltration of the chorioamnion with polymorphonuclear white blood cells. Analysis of the data was done both including and excluding women who had histologic evidence of infection.
Data were expressed as mean values ± standard deviation (SD). Statistical analysis was done with determination of Spearman rank-order correlation, and comparison between groups was done by pairwise multiple-comparison test (Student-Newman-Keuls test) because data were normally distributed. Fisher exact test was used to compare proportions between groups. Statistical significance was set at P < .05.
In amniotic fluid from women with preterm labor who responded to tocolysis concentrations of adrenomedullin (102.3 ± 26.8 pg/mL) were similar to those from pregnant women at the same gestational age with uncomplicated pregnancies (101.2 ± 28.1 pg/mL). Premature rupture of membranes and premature labor unresponsive to tocolysis were associated with significantly higher (P < .01) amniotic fluid adrenomedullin concentrations (177.0 ± 22.5 pg/mL and 182.7 ± 22.0 pg/mL, respectively). In women with preterm labor unresponsive to tocolysis, the mean interval between sampling and delivery was significantly shorter than that in controls (3 ± 2 days versus 38 ± 6 days; P < .01) and in women responsive to tocolysis (15 ± 5 days; P < .05). However, the correlation between adrenomedullin levels and time of delivery after sampling was not significant (correlation coefficient 0.153). In PROM subjects, the mean interval (13 ± 6 days) did not differ from women who responded; however, the percentage of women who delivered within 7 days was significantly higher (56% versus 6%), due to elective cesarean at 34 weeks in women with positive fetal lung-maturity tests. Cervical dilatation at enrollment did not differ significantly between responsive and unresponsive groups, whereas in women with PROM, cervical dilatation was less (Table 1). None of the neonates presented with clinical infection at birth; however, placental histology showed evidence of chorionamnionitis in six cases (Table 1). When those women were excluded (three in preterm labor unresponsive and three in PROM), there was still a significant difference (P < .01) in adrenomedullin concentrations (unresponsive 158.3 ± 14.1 pg/mL; PROM 163.2 ± 17.3 pg/mL) between those groups and uncomplicated pregnancies, or women who responded.
The present study found that adrenomedullin is associated with preterm delivery, suggesting that it might affect regulation of mechanisms of human parturition. A preliminary study reported that adrenomedullin exerted an inhibitory effect on contractile responses of rat uterine muscle to galanin through an increase in cyclic AMP.15 However, we found increased adrenomedullin in women with refractory preterm labor and with PROM. It is possible that in women who did not respond to tocolysis adrenomedullin increased in amniotic fluid to compensate for increased synthesis and release of other uterotonin substances, such as prostaglandins from the placenta and fetal membranes, which in turn could affect adrenomedullin release from fetoplacental tissues. We did not find a significant correlation between adrenomedullin concentrations and time of delivery after sampling. Cervical dilatation at sampling in women with refractory preterm labor and those who responded to therapy was similar. High adrenomedullin concentrations were also detected in women with PROM who were not in labor, indicating that increased levels of adrenomedullin were not associated directly with uterine contractility. Bacterial intraamniotic contamination was reported to be the most common cause of preterm labor and was always present in PROM microbial invasions of amniotic cavities.16 Interleukins were found in amniotic fluid of women with preterm labor and intramniotic infections and were secreted by human decidua in response to bacterial products.17 Bacterial endotoxins and cytokines induce expression of adrenomedullin; in culture adrenomedullin secretion is stimulated by cytokines, epidermal growth factor, and bacterial lipopolysaccharides.18 Adrenomedullin is increased in vivo in women with sepsis.19 We hypothesize that in women who did not respond to tocolysis and in women with PROM, adrenomedullin levels increased in association with bacterial contamination of amniotic fluid or fetuses. Those women might have had subclinical infections responsible for macrophage activation and cytokine production. Although we did not culture amniotic fluid, the incidence of histologic evidence of chorionamnionitis was significantly higher in placental tissues collected from women with PROM and preterm delivery unresponsive to tocolysis than in controls and women responsive to tocolysis. We also found that amniotic fluid adrenomedullin levels were higher in women with histologic chorionamnionitis, although the levels did not affect significantly the mean concentrations in the groups. Increased concentrations of adrenomedullin in amniotic fluid of those women might derive from enhanced synthesis from placenta or fetal membranes stimulated by bacterial products. Alternatively, the increased concentration of adrenomedullin in amniotic fluid samples from nonresponding women could be associated with modification of hormonal secretion during parturition. Adrenomedullin is an important mediator of hormonal secretion and might influence regulation of endocrine secretion by placenta and fetal tissues. Rat adrenomedullin inhibits basal and corticotrophin-releasing hormone–stimulated ACTH release from rat anterior pituitary cells, suggesting that adrenomedullin might regulate hypothalamuspituitary-adrenal feedback.4 Hormonal factors, such as steroid hormones, affect regulation of adrenomedullin secretion in vascular tissue.5 Those findings are of interest because of the effect of corticotrophin-releasing hormone and ACTH on human term and preterm parturition.20 It was reported that corticotrophin-releasing hormone concentration and the ratio of cortisol to cortisol-binding globulin was higher in women with true preterm labor than in women with false preterm labor. Increased corticotrophin-releasing hormone and availability of biologically active cortisol might stimulate adrenomedullin secretion from placental tissues, accounting for higher levels of it in amniotic fluid of pregnant women unresponsive to tocolysis who delivered shortly after sampling.
1. Kitamura K, Kangawa K, Kawamoto M, Ichiki Y, Nakamura S, Matsuo H, et al. Adrenomedullin: A novel hypotensive peptide isolated from human pheochromocytoma. Biochem Biophys Res Commun 1993;192:553–60.
2. Ichiki Y, Kitamura K, Kangawa K, Kamoto M, Matsuo H, Eto T. Distribution and characterization of immunoreactive adrenomedullin in human tissue and plasma. FEBS Lett 1994;338:6–10.
3. Ishizaka Y, Ishizaka Y, Tanaka M, Kitamura K, Kangawa K, Minamino N, et al. Adrenomedullin stimulates cyclic AMP formation in rat vascular smooth muscle cells. Biochem Biophys Res Commun 1994;200:642–6.
4. Samson WK, Murphy T, Schell DA. A novel vasoactive peptide, adrenomedullin, inhibits pituitary adrenocorticotropin release. Endocrinology 1995;136:2349–52.
5. Minamino N, Shoji H, Sugo S, Kangawa K, Matsuo H. Adrenocortical steroids, thyroid hormones and retinoic acid augment the production of adrenomedullin in vascular smooth muscle cells. Biochem Biophys Res Commun 1995;211:686–93.
6. Marinoni E, Di Iorio R, Letizia C, Villaccio B, Scucchi L, Cosmi EV. Immunoreactive adrenomedullin in human fetoplacental tissues. Am J Obstet Gynecol 1998;179:784–7.
7. Macri CJ, Martinez A, Moody TW, Gray KD, Miller MJ, Gallagher M, et al. Detection of adrenomedullin, a hypotensive peptide, in amniotic fluid and fetal membranes. Am J Obstet Gynecol 1996; 175:906–11.
8. Miura K, Ebara T, Okumura M, Matsuura T, Kim S, Yukimura T, et al. Attenuation of adrenomedullin-induced renal vasodilatation by NG
-nitro L-arginine but not glibenclamide. Br J Pharmacol 1995;115:917–24.
9. Kohno M, Kano H, Horio T, Yokokawa K, Yasunari K, Takeda T. Inhibition of endothelin production by adrenomedullin in vascular smooth muscle cells. Hypertension 1995;25:1185–90.
10. Izumi H, Yallampalli C, Garfield RE. Gestational changes in L-arginine-induced relaxation of pregnant rat and human myometrial smooth muscle. Am J Obstet Gynecol 1993;169:1327–37.
11. Word RA, Kamm KE, Casey ML. Contractile effect of prostaglandins, oxytocin, and endothelin-1 in human myometrium in vitro: Refractoriness of myometrial tissue of pregnant women to prostaglandins E2
and F2 alpha
. J Clin Endocrinol Metab 1992;75:1027–32.
12. Seguchi H, Mishimura N, Kobayashi S, Kumazawa J, Kanaide H. Autocrine regulation of the renal arterial tone by adrenomedullin. Biochem Biophys Res Commun 1995;215:619–25.
13. Grose C, Meehan T, Weiner CP. Prenatal diagnosis of congenital cytomegalovirus infection by virus isolation after amniocentesis. Pediatr Infect Dis J 1992;11:605–7.
14. Letizia C, Cerci S, Centanni M, De Toma G, Subioli S, Scuro L, et al. Circulating levels of adrenomedullin in patients with Addison's disease before and after corticosteroid treatment. Clin Endocrinol 1998;48:20–3.
15. Upton PD, Austin C, Taylor GM, Nandha A, Clark AJ, Ghatei MA, et al. Expression of adrenomedullin (ADM) and its binding sites in the rat uterus: Increased number of binding sites and ADM messenger ribonucleic acid in 20-day pregnant rats compared with nonpregnant rats. Endocrinology 1997;138:2508–14.
16. Romero R, Avila C, Brekus CA, Morotti R. The role of systemic and intrauterine infection in preterm parturition. Ann N Y Acad Sci 1991;622:355–75.
17. Romero R, Brody DT, Oyarzun E, Mazor M, Wu YK, Hobbins JC, et al. Infection and labor. III. Interleukin-1: A signal for the onset of parturition. Am J Obstet Gynecol 1989;160:1117–23.
18. Sugo S, Minamino N, Shoji H, Kangawa K, Kitamura K, Eto T, et al. Interleukin-1, tumor necrosis factor and lipopolysaccharide additively stimulate production of adrenomedullin in vascular smooth muscle cells. Biochem Biophys Res Commun 1995;207:25–32.
19. Hirata Y, Mitaka C, Sato K, Nagura T, Tsunoda Y, Amaha K, et al. Increased circulating adrenomedullin, a novel vasodilatory peptide, in sepsis. J Clin Endocrinol Metab 1996;81:1449–53.
20. Challis JRG, Gibb W. Control of parturition. Prenat Neonatal Med 1996;1:283–91.