OBJECTIVE: To test electric field stimulation on human placental vessels.
METHODS: The effects of electric field stimulation on placental vessels were examined in an isometric myograph.
RESULTS: Electric field stimulation induced contractions in human placental blood vessels in vitro under isometric conditions when bubbling carbogen through the organ bath. After reaching half-maximal contractions, the vessel rings showed spontaneous relaxation. Pretreatment with verapamil (10−6 mol/L) or nickel (Ni2+) (2 mmol/L) inhibited the contractions to a magnitude of 63.81% ± 7.69% and 88.36% ± 12.17% (mean ± standard error of the mean), respectively. In calcium (Ca2+)-free medium after combined cyclopiazonic acid (10−5 mol/L) and Ni2+ treatment, it was not possible to elicit contractions with electric field stimulation. Bubbling through physiologic in utero hypoxic gases enhanced the contractile responses of the human placental vessel rings to electric field stimulation. The spontaneous relaxation of the veins was not altered, but those of the arteries were reduced to zero. Testing the same gases on mesenteric arteries of rats had an opposite effect concerning contractility. Sodium nitrite decreased the contractions of the placental vessel rings, but the efficacy was decreased by the in utero gases.
CONCLUSION: Electric field stimulation has a direct, non-neurogenic contractile effect on isolated placental vessels, which mainly depends on the influx of extracellular Ca2+ and on a mechanism independent of intracellular Ca2+ concentration elevation. Physiologic hypoxia has a stimulatory effect on the contractility of human placental vessels, therefore in utero gases should be used instead of carbogen gas; and electric field stimulation is a suitable method for the investigation of the direct effects of pharmacologic agents on human placental vessels.
Electric field stimulation is a new experimental possibility for investigations of the direct placental vascular effects of different pharmacologic agents given during pregnancy and delivery.
Department of Pharmacodynamics and Biopharmacy, University of Szeged, Szeged, Hungary.
Address reprint requests to: George Falkay, DSc, University of Szeged, Department of Pharmacodynamics and Biopharmacy, Eötvösu. 6., SzegedH-6720, Hungary; E-mail: firstname.lastname@example.org.
This study was financially supported by Hungarian Research Grant FKFP 0174/2001.
Received May 14, 2002. Received in revised form July 4, 2002. Accepted August 1, 2002.
The human placenta has a dual fundamental role: It connects the embryonal, and later the fetal circulation to the maternal circulation, and it isolates the conceptus from the maternal organism. Fetal circulation in the trophoblast starts in the very early period of pregnancy: It can be detected from days 21–23 of gestation.1 Sufficient placental circulation is indispensable for the healthy development of the fetus, for an uncomplicated delivery, and for the tolerance of stress from birth.
An acute insufficiency of the placental circulation results in hypoxia, acute in utero distress, preterm birth, spontaneous abortion, or in utero death. A chronic insufficiency of the placental circulation can cause intra-uterine growth restriction,2 dysmaturity, and low birth weight. A severe placental insufficiency can cause fetal myocardial cell damage3 and coronary heart disease in adulthood.4 Clinical evidence has linked an intrauterine compromise, such as a prolonged period of insufficient placental circulation during the last trimester, to a poor neurological outcome in the newborn, which may manifest its consequences only at the ages of 5–8.5 At the present time, the mechanisms responsible for the regulation of placental blood flow are poorly understood. Moreover, detailed knowledge is needed of the direct effects of all the drugs that may be given during pregnancy and during delivery on the human placental vasculature. There have been several reports on the direct placental vascular effects of different chemical substances, including endothelin-1, prostaglandin F2-alfa, sodium nitroprusside,6 histamine,7 ouabain, serotonin,8,9 and ketanserin.10
Electric field stimulation is widely used in physiologic and pharmacologic research, for example, studies of the effects of calcitonin gene–related peptide in a rat cremaster muscle model11 and investigations of the contractility of the pregnant rat uterus.12,13 A number of reports have been published on investigations of the physiologic and pharmacologic behavior of different vessels by means of electric field stimulation, including the iridial arteries of the rat,14 the small mesenteric veins of the rabbit,15 the dog mesenteric artery and the rabbit ear artery,16,17 and the rabbit mesenteric artery and aorta.18 We have found no data on the application of electric field stimulation to study human placental blood vessels in the literature.
The aim of this work was to develop an electric field stimulation model suitable for investigations of the contractility of placental veins and arteries, and of the direct effects of different pharmacologic agents on the placental blood vessels. Most investigations of electric field stimulation are undertaken to stimulate sympathetic/adrenergic nerve terminals and to evoke local release of transmitters.19 A valid reason for determining the effects of electric field stimulation is that the placental circulation appears to be under very little or no functional sympathetic control20,21; for example, there is no or only a very poor response of the vessels of this circulation to exogenous noradrenaline.22 Another reason for using electric field stimulation is to provide a means of constricting these vessels in a receptor-independent fashion prior to addition of a dilatator/constrictor agonist.
MATERIALS AND METHODS
Experimentation on human placentas was approved by the Institutional Review Board (permission No. 909/ 1998 of the Ethics Committee of the University of Szeged). Placentas were obtained from the delivery room of the Department of Obstetrics and Gynecology, University of Szeged, immediately after birth; they were transferred in 500 mL icy Krebs-Henseleit buffer, and the experiments were begun 10–30 minutes after birth. All the placentas were selected randomly for study inclusion from term pregnancies of healthy, white mothers that ended with uncomplicated deliveries. The ages of the women from whom the placentas were obtained ranged between 19 and 28 years, with an average of 25.3 years. The gestational age at delivery was between 37 and 40 weeks, with an average of 38.1 weeks. The vast majority (80%) of the mothers were primiparas. The numbers of placentas included in each series of experiments are indicated by “n” in each figure caption. At least eight placentas were used for each series of experiments, and sometimes 12. We obtained four vessel rings from each placenta. After the umbilical cord had been cut off, thin polyethylene cannulae were led into the vein (with larger diameter) and the two arteries (with smaller diameters) in the stub, to separate the veins and arteries on the fetal surface of the placenta. The vessels were prepared for in vitro measurement according to the method outlined by Angus and Wright.23 Rings 1–1.3 mm in diameter24,25 were dissected from the identified veins and arteries just before their heading toward the stem villi. The precise length of the rings (4 mm) was achieved by use of a fixed, double-bladed scalpel. The loose connective tissue was carefully removed under a cool fiber optic light source and binocular dissection microscope (10×). The rings were taken distally from the site of introduction of the cannulae, the endothelium therefore remaining intact.
Rat mesenteric arteries were dissected from 200–220-g Sprague-Dawley rats through laparotomy after the animals were killed. Rat studies were allowed by permission No. I.-74-8/2002 of the Animal Ethics Committee of the University of Szeged.
The rings were mounted diametrically (as ring preparations) between two platinum electrodes in an organ bath containing 10 mL Krebs-Henseleit buffer (in mmol/L: 118 NaCl, 5 KCl, 2 CaCl2, 0.5 MgSO4, 1 KH2SO4, 25 NaHCO3, 10 glucose, pH = 7.4). The organ bath was maintained at 37C. Carbogen gas (95% O2 = 5% CO2) or gas simulating the in utero conditions (Messer, Szeged, Hungary) (for veins: pO2= 38 mm Hg, pCO2= 43 mm Hg; for arteries: pO2= 22 mm Hg, pCO2= 48 mm Hg)26 was bubbled through it. After mounting, the rings were equilibrated for 90 minutes before the experiment, and the buffer was changed every 10 minutes. The passive force was set at approximately 3.75g and 3.25g for veins and arteries, respectively. The optimal degree of stretch was ascertained by determining a contraction versus passive force curve in response to an electric field stimulation stimulus with a stimulating potential of 30 V, a period of 4 seconds, and a pulse width of 80 milliseconds. Using the same technique, the optimal passive force for rat mesenteric arteries turned out to be approximately 2.2g. These passive forces were similar to those used in previous electric field stimulation studies on other mammalian vascular smooth muscle preparations.27
Contractions were elicited by a digital, programmable stimulator (ST-02, Experimetria, London, United Kingdom). The force of the vessel rings was measured with a gauge transducer (SG-02, Experimetria U.K.), and recorded by an ISOSYS Data Acquisition System (Experimetria U.K.).
Measured or calculated points were plotted, and curves were fitted to these points with GraphPad Prism 2.01 software (San Diego, CA). Data were statistically analyzed with SPSS for Windows 9.0 (SPSS, Chicago, IL). Two- or three-way repeated measures analysis of variance was used to evaluate the significance levels of differences in all the figures except Figure 3. In Figure 3, two-way analysis of variance was used. Probability values lower than .05 were considered significant.
The optimal period time (interval between two stimuli) was determined by decreasing the applied period time until the individual contractions fused to produce a smooth contractile response. Twenty and 30 V were used as stimulating potentials (both supramaximal). The pulse width (the duration of a single stimulus) values used to elicit half-maximal contractions were determined by applying stimuli of different durations (pulse width: 25, 50, 100, and 200 milliseconds), and the elicited contraction responses were registered.
The effects of sodium nitrite (NaNO2) (3 × 10−7−10−3 mol/L) were studied. NaNO2 was administered in a cumulative way. The first dose was added at half-maximal contraction and the next doses every 5 minutes. All pharmacologic compounds were purchased from Sigma Aldrich, Budapest, Hungary. The length of the study enrollment period was 11.5 months (348 days).
Electric field stimulation induced fast and reproducible contractions in the human placental blood vessel rings. The optimal period time was found to be 5 seconds. For both tested stimulating potentials (20 V and 30 V), the strength of the contractions was plotted against the logarithm of the pulse width, and a sigmoidal curve was fitted. The contraction versus pulse width curve was analyzed, and the pulse width location at half-maximal force was calculated with GraphPad Prism 2.01 software. With stimulation at 20 V, the pulse width location at half-maximal force was greater than 500 milliseconds for both veins and arteries. These values were so extremely high that the curve was considered unsaturable (Figure 1). With stimulation at 30 V, the optimal pulse width (the pulse width location at half-maximum force) for veins and arteries was 100 milliseconds and 119 milliseconds, respectively (Figure 2). The contractions were not changed by tetrodotoxin (10−6 mol/L). Pre-treatment with verapamil (10−6 mol/L), which is a blocker of voltage-operated (L-type) calcium (Ca2+) channels or nickel (Ni2+) (2 mmol/L) (nonselective blocker of cation channels), inhibited the contractions to a magnitude of 63.81% ± 7.69% and 88.36% ± 12.17% (mean ± standard error of the mean), respectively. Combined verapamil and Ni2+ treatment inhibited the contractions to a similar magnitude as Ni2+ treatment by itself. In Ca2+-free medium after combined cyclopiazonic acid (10−5 mol/L) (depletes Ca2+ through inhibition of sarcoplasmic reticulum Ca2+-adenosinetriphos-phatase) and Ni2+ treatment, it was not possible to elicit contractions with electric field stimulation (Figure 3).
With stimulation at 20 V, the contractions of the veins were significantly stronger than those of the arteries at all applied pulse widths (P < .05). With stimulation at 30 V, the contractions of the veins were stronger at all applied pulse widths, but none of the differences were significant (P > .05).
After reaching half-maximal contraction, the placental vessel rings exhibited a time-dependent spontaneous relaxation, despite continuous stimulation. The arteries displayed a slightly greater relaxation, but the difference between the relaxation of the veins and the arteries was not significant (P > .05). The proportion of the spontaneous relaxation as a function of time after the half-maximal contraction was reached could be described by two (veins and arteries) third-degree polynomial functions (y = A + Bx + Cx2 + Dx3) (Figure 4) (second-degree polynomial functions are also appropriate). According to these functions, the proportion of the spontaneous relaxation could be determined at any moment within 40 minutes after the half-maximal contraction.
Bubbling through physiologic in utero hypoxic gases significantly enhanced the contractile responses of both the human placental vein and artery rings to electric field stimulation at all applied pulse widths (Figure 2), though the optimal pulse widths were not changed significantly (99 milliseconds and 105 milliseconds for veins and arteries, respectively). The spontaneous relaxation of the veins was not altered, but those of the arteries were reduced to zero (Figure 4).
Electric field stimulation (with the same stimulating parameters as used for placental vessel rings) also induced fast and reproducible contractions on the rat mesenteric arterial rings. After reaching half-maximal contraction, the rat mesenteric rings also exhibited a time-dependent spontaneous relaxation, despite continuous stimulation. The proportion of the spontaneous relaxation as a function of time after the half-maximal contraction was reached could be described by another third-degree polynomial function. Contrary to the placental vessel rings, in utero gases significantly blunted the forces of the contractions at all applied doses (Figure 5), but they did not alter the spontaneous relaxation of the rat mesenterial arterial rings (Figure 6).
The inhibitory or stimulatory effects caused by any drug itself, besides the spontaneous relaxations, could be calculated by correcting the measured contractions by the third-degree polynomial functions. The correction procedure was as follows: 1) multiplication of the actual contractions by the proportions of the spontaneous relaxation measured 5, 10, 15, 20, 25, 30, 35, and 40 minutes after addition of the first dose. The results of these multiplications were the actual spontaneous relaxations, which could have been measured without addition of the drugs; 2) subtraction of the actual spontaneous relaxation from the actual measured relaxation, leaving the actual relaxation caused by the drug itself; and 3) division of the actual drug-caused relaxation by the actual contraction, resulting in the proportions (percentages) of the drug-caused relaxation, which were plotted against the logarithms of the concentrations (semilogarithmic dose–response curve).
NaNO2 antagonized the contractions of the placental vessel rings in a significant and dose-dependent manner, but the efficacy of NaNO2 was significantly decreased by the in utero gases (Figure 7).
The electric field stimulation method has the advantage that another pharmacologic agent is not needed to elicit contractions, and interactions between the contractile agent and the examined pharmacons can therefore be avoided. In this study, we examined the effects of electric field stimulation on human placental blood vessels obtained from uncomplicated term pregnancies. Small-diameter resistance vessels, compared with conduit vessels, probably contribute more to the hemodynamics of placental bed perfusion, though the question of the location and the diameter of resistance vessels is, in general, unsolved, and the feed arteries can be as active in flow control as the microvasculature.28 According to recent data, the vessel rings obtained by us can be considered both as conduit and resistance vessels. In addition, evidence that larger or smaller segments contribute to a lesser or greater degree to changes in the blood flow of human organs (including the placenta) is not available.28 The optimal parameters of electric field stimulation were then determined. It has been proven that in oxygenated Krebs-Henseleit buffer, electric field stimulation does not generate substances that change the contractile state of the smooth muscle.29 To cause less harm to the tissue sample, mild stimulation was desirable. The lower the frequency (the longer the period time) of the stimulation, the less stressful it is to the tissue.
To investigate both the dilatator and the constrictor effects of the pharmacons, half-maximal contractions were needed. The pulse width location at half-maximal force of these contraction versus pulse width curves were considered to be the optimal pulse widths. With stimulation at 20 V, the contraction versus pulse width curve proved to be unsaturable, and it was therefore impossible to determine the optimal pulse widths. With stimulation at 30 V, the contraction versus pulse width curve was saturable; accordingly, 30 V was chosen as the optimal stimulating potential in our further experiments, despite the fact that the higher stimulating potential might be more harmful to the tissue. The assessed optimal pulse width parameters conform to the theory that the use of a short pulse width (0.7–5 milliseconds) is thought to selectively stimulate nerves but not smooth muscle, which requires a much longer pulse width (60–133 milliseconds) for direct excitation.30 Though with the high pulse widths and voltages applied, responses to electric field stimulation were not likely due to stimulation of nerves, the non-neurogenic nature of the contractions was checked functionally by blocking the nerve action potentials conduction with tetrodotoxin (10−6 mol/L). These results therefore support previous studies, which suggested an absolute lack of sympathetic innervation in the placental circulation.20,21 The involvement of extracellular and/or intracellular Ca2+ and a possible subsequent opening of voltage-operated Ca2+ channels was assessed by conducting experiments in Ca2+-free medium and with blockers of both voltage-operated Ca2+ channels and the intracellular Ca2+ stores. The results of these experiments suggested that the direct, non-neurogenic contractile effect of electric field stimulation on isolated human placental blood vessel rings mainly depends on the influx of extracellular Ca2+ via voltage-operated Ca2+ channels, partly on the mobilization of intracellular Ca2+ stores, and on a mechanism independent of intracellular Ca2+ concentration elevation. Because of the approximately 25% difference between the inhibitory effects of verapamil and Ni2+, this mechanism independent of intracellular Ca2+concentration elevation is likely to be the influx of other extracellular cations besides Ca2+. In our experiments, veins gave stronger contractile responses than arteries. These functional results confirmed the morphological properties of the walls of the placental veins and arteries. Force-producing smooth muscle cells lie within the media,31 and the smooth muscle–containing tunica media is thicker in placental veins than in arteries.32 The clinical relevance of the differences between the contractions of the veins and the arteries in the testing of future drugs is difficult to predict, because the response of the vessels to drug actions depends not only on the muscle layer thickness but also on the densities of different receptors, the activities of second messenger systems, etc. These parameters differ for every pharmacon. Otherwise, in general it might be stated that drugs that can alter the diameter either of the veins or the arteries may change the placental blood flow. Increased blood flow may be beneficial in preeclampsia and intrauterine growth restriction.
The parameters needed for half-maximal contractions of the veins and arteries were different, but the difference was not significant. The spontaneous relaxation after the half-maximal contractions were reached did not display a significant difference either.
Because the placenta is a hypoxic organ, the effects of in utero gases on human placental blood vessel rings were examined. In addition, as a comparison, the effects of the same gases were also investigated on rat mesenteric arterial rings. Interestingly, the results of these experiments suggested that, contrary to other, nonplacental vessels, in utero physiologic hypoxic circumstances have a stimulatory/enhancing effect on the contractility of human placental vessels. These results also suggest that during electric field stimulation studies on human placental blood vessels in utero, gases should be used instead of carbogen gas.
To test the pharmacologic capability of our model, the effects of a nitrovasodilator (NaNO2) was examined.
NaNO2 is a well-known nitric oxide donor. As might be expected, it antagonized the electric field stimulation contractions of the placental vessel rings in a significant and dose-dependent way. The efficacy of NaNO2 was significantly decreased by the in utero gases, leading us to the conclusion that the contractions of human placental blood vessel rings are better under physiologically hypoxic in utero conditions. The relaxant effect of NaNO2 was more pronounced on placental veins than on arteries at all applied doses, but none of the differences reached the level of significance. This was in agreement with data from the literature, that placental vessels obtained from normotensive pregnancies are sensitive to the relaxant effect of nitrovasodilatators (glyceryl trinitrate, sodium nitroprusside, and S-nitroso-N-acetylpenicillamine).33 These results also confirm the benefit of using in utero gases during human placental blood vessel electric field stimulation studies.
From our present findings, it may be concluded that we have successfully applied electric field stimulation for the study of human placental vessels, which is therefore a new experimental possibility for investigations of the direct placental vascular effects of different pharmacologic agents.
1. Steven DH. Comparative placentation. Essays in structure and function. London: Academic Press, 1975:142.
2. Manning FA, Hill LM, Platt LD. Qualitative amniotic fluid volume determination by ultrasound: Antepartum detection of intrauterine growth retardation. Am J Obstet Gynecol 1981;139:254–8.
3. Makikallio K, Voulteenaho O, Jouppila P, Rasanen J. Association of severe placental insufficiency and systemic venous pressure rise in the fetus with increased neonatal cardiac troponin T levels. Am J Obstet Gynecol 2000;183:726–31.
4. Hall EF, McKeigue PM. Essays in prevention. Fetal origins of coronary heart disease. Evid Cardiovasc Med 1999;3:87–9.
5. Mallard EC, Rees S, Stringer M, Cock ML, Harding R. Effects of chronic placental insufficiency in fetal sheep. Pediatr Res 1998;43:262–70.
6. Clausen HV, Jorgensen JC, Ottesen B. Stem villous arteries from the placentas of heavy smokers: Functional and mechanical properties. Am J Obstet Gynecol 1999;180:476–82.
7. Bertrand C, St-Louis J. Reactivities to serotonin and histamine in umbilical and placental vessels during the third trimester after normotensive pregnancies and pregnancies complicated by preeclampsia. Am J Obstet Gynecol 1999; 180:650–9.
8. Sanchez-Ferrer CF, Ponte A, Casado MA, Rodriguez-Manaz L, Pareja A, Gonzalez R, et al. Endothelial modulation of the vascular sodium pump. J Cardiovasc Pharmacol 1993;22:99–101.
9. Okatani Y, Watanabe K, Nakano Y, Sagara Y. Relaxant effect of nitric oxide and prostacyclin on serotonin-induced vasocontraction of human umbilical artery. Acta Obstet Gynecol Scand 1996;75:108–12.
10. Marin J, Revirego J, Fernandez-Alfonso MS. Ability of ketanserin to block different receptors in human placental vessels. J Pharm Pharmacol 1990;42:217–20.
11. Allen DM, Chen LE, Seaber AV, Urbaniak JR. Calcitonin gene-related peptide and reperfusion injury. J Orthop Res 1997;15:243–8.
12. Gaspar R, Marki A, Zupko I, Falkay G. Evidence of non-synaptic regulation of postpartum uterine contractility in the rat. Life Sci 1998;62:1119–24.
13. Gaspar R, Foldesi I, Havass J, Marki A, Falkay G. Characterization of late-pregnant rat uterine contraction via the contractility ratio in vitro significance of alpha1-adrenoceptors. Life Sci 2001;68:1119–29.
14. Hirst GD, Edwards FR, Gould DJ, Sandow SL, Hill CE. Electrical properties of iridial arteries of the rat. Am J Physiol 1997;273:2465–72.
15. Marijic J, Madden JA, Kampine JP, Bosnjak ZJ. The effect of halothane on norepinephrine responsiveness in rabbit small mesenteric veins. Anesthesiology 1990;73:479–84.
16. Sun FY, Zhang AZ. Dynorphin receptor in the blood vessel. Neuropeptides 1985;5:595–8.
17. Bao WL, Zhen FC, Sun FY, Zhang AZ. Effects of sigma and phencyclidine receptor ligands on electric field stimulated rabbit ear artery constriction in vitro. Zhongguo Yao Li Xue Bao 1994;15:320–2.
18. Li JY, Kuriyama H. Comparison of actions of endothelium-derived nitric oxide and sodium nitroprusside on mechanical responses evoked in aorta and mesenteric artery of the rabbit. Gen Pharmacol 1993;24:377–85.
19. Angus JA, Broughton A, Mulvany MJ. Role of alpha-adrenoceptors in constrictor responses of rat, guinea-pig and rabbit small arteries to neural activation. J Physiol 1988;403:495–510.
20. Khong TY, Tee JH, Kelly AJ. Absence of innervation of the uteroplacental arteries in normal and abnormal human pregnancies. Gynecol Obstet Invest 1997;43:89–93.
21. Buttery LD, McCarthy A, Springall DR, Sullivan MH, Elder MG, Michel T, et al. Endothelial nitric oxide synthase in the human placenta: Regional distribution and proposed regulatory role at the feto-maternal interface. Placenta 1994;15:257–65.
22. Manyonda IT, Slater DM, Fenske C, Hole D, Choy MY, Wilson C. A role for noradrenaline in pre-eclampsia: Towards a hypothesis for the pathophysiology. Br J Obstet Gynaecol 1998;105:641–8.
23. Angus JA, Wright CE. Techniques to study the pharma-codynamics of isolated large and small blood vessels. J Pharmacol Toxicol Methods 2000;44:395–407.
24. Omar HA, Figueroa R, Omar RA, Tejani N, Wolin MS. Hydrogen peroxide and reoxygenation cause prostaglandin-mediated contraction of human placental arteries and veins. Am J Obstet Gynecol 1992;167:201–7.
25. Omar HA, Ramirez R, Gibson M. Properties of a progesterone-induced relaxation in human placental arteries and veins. J Clin Endocrinol Metab 1995;80:370–3.
26. Longo LD. Placental gas exchange. In: Knobil E, Neill JD, eds. Encyclopedia of reproduction, volume 3. London: Academic Press, 1998:864.
27. Ehrreich SJ, Clopper EL. Physiologic and pharmacologic responses of mammalian vascular smooth muscle during electric field stimulation. Experientia 1970;26:1229–31.
28. Christensen KL, Mulvany MJ. Location of resistance vessels. J Vasc Res 2001;38:1–12.
29. Jongejan RC, de Jongste JC, Raatgeep RC, Bonta IL, Kerrebijn KF. Electrically stimulated Krebs-Henseleit buffer does not relax precontracted human bronchi in vitro. Agents Actions 1989;26:75–6.
30. Tomita T. Electrical properties of mammalian smooth muscle. In: Bulbring E, Brading AF, Jones AW, Tomita T, eds. Smooth muscle. London: Edward Arnold, 1970: 197–243.
31. Mulvany MJ. Determinants of vascular hemodynamic characteristics. Hypertension 1984;6:III-13–III-18.
32. Tanaka C, Kuwabara Y, Sakai T. Structural identification and characterization of arteries and veins in the placental stem villi. Anat Embryol 1999;199:407–18.
© 2003 The American College of Obstetricians and Gynecologists
33. Gonzalez C, Cruz MA, Gallardo V, Miguel P, Carrasco G. Relative potency of nitrovasodilatators on human placental vessels from normal and preeclamptic pregnancies. Gynaecol Obstet Invest 1997;43:219–24.