We present a detailed method to study human placental physiology in vivo at term. The method combines blood sampling from the incoming and outgoing vessels on the maternal and fetal sides of the placenta with ultrasound measurements of volume blood flow and placental tissue sampling.
The human placenta is highly inaccessible for research while still in utero. The current understanding of human placental physiology in vivo is therefore largely based on animal studies, despite the high diversity among species in placental anatomy, hemodynamics and duration of the pregnancy. The vast majority of human placenta studies are ex vivo perfusion studies or in vitro trophoblast studies. Although in vitro studies and animal models are essential, extrapolation of the results from such studies to the human placenta in vivo is uncertain. We aimed to study human placenta physiology in vivo at term, and present a detailed protocol of the method. Exploiting the intraabdominal access to the uterine vein just before the uterine incision during planned cesarean section, we collect blood samples from the incoming and outgoing vessels on the maternal and fetal sides of the placenta. When combining concentration measurements from blood samples with volume blood flow measurements, we are able to quantify placental and fetal uptake and release of any compound. Furthermore, placental tissue samples from the same mother-fetus pairs can provide measurements of transporter density and activity and other aspects of placental functions in vivo. Through this integrative use of the 4-vessel sampling method we are able to test some of the current concepts of placental nutrient transfer and metabolism in vivo, both in normal and pathological pregnancies. Furthermore, this method enables the identification of substances secreted by the placenta to the maternal circulation, which could be an important contribution to the search for biomarkers of placenta dysfunction.
According to the National Institutes of Health, USA, the placenta is the least understood organ in the human body 1,2,3. It is difficult to access and study the human placenta in vivo without imposing unethical risks on the ongoing pregnancy. Studies of placental function in the human are therefore largely based on in vitro and ex vivo models. The majority of previous in vivo studies of placental transport and metabolism have been performed in animals 4,5,6. However, as placental structure and functions vary considerably between species, extrapolation of results from animals to humans must be done with caution. Only a few smaller human in vivo studies have investigated placental and fetal uptake and transport under normal physiological conditions, and none have explored the integrated transfer of several compounds 7,8,9,10,11,12,13. These fundamental studies illustrate that in vivo studies of the human placenta are feasible, and that they may serve several purposes. First, current concepts of placental functions mainly derived from in vitro, ex vivo and animal studies may be tested in a human setting and thus provide novel and more specific insight into the human placenta. Second, properties of the dysfunctional placenta associated with aberrant fetal growth, preeclampsia, maternal diabetes, metabolic syndrome and other maternal metabolic disturbances may be better characterized. Third, human in vivo studies provide an opportunity to develop diagnostic and predictive tools of placental function.
On this background we aimed to establish a comprehensive collection of physiological data to investigate human placental function in vivo. During a planned cesarean section, we exploit the intraabdominal access to the uterine vein to collect blood samples from the incoming and outgoing vessels on the maternal and fetal sides of the placenta (the 4-vessel sampling method). These samples are used to calculate the paired arteriovenous concentration differences of nutrients and other substances 14. In addition, we measure volume blood flow on both sides of the placenta by ultrasound. Consequently, placental and fetal uptake of any compound may be quantified. Further, it is possible to determine substances released by the placenta to the maternal and fetal circulations 15,16,17. When combined with clinical parameters of mother and child, and analyses of placental and other relevant tissues, this method has the exciting potential to integrate many aspects of placental functions in vivo in the same mother-fetus pairs.
The study was approved by the data protection officials at Oslo University Hospital and the Regional Committee for Medical and Health Research Ethics, Southern Norway 2419/2011. All participants signed a written informed consent at inclusion.
1. Preparations
NOTE: A timeline for the procedures is outlined in Figure 1.
Figure 1: Flowchart Describing the Timing and the Personnel Involved in the 4-vessel Sampling Procedure.
One color represents one person. Detailed description of the method is given in the protocol. Please click here to view a larger version of this figure.
2. Maternal Characteristics
3. Ultrasound
4. 4-vessel Blood Sampling
NOTE: The timeline for the procedures is outlined in Figure 1 and an overview of the samples is illustrated in Figure 2.
Figure 2: Schematic Illustration of the Placental Vasculature and the Sampling Sites.
In the 4-vessel sampling method blood samples are drawn from the uterine vein, the radial artery (as a proxy for the uterine artery) and the umbilical arteries and vein. Blood flow in the uterine artery and the umbilical vein is measured by ultrasound. Tissue samples from the placenta are collected. Illustration: Øystein H. Horgmo, University of Oslo. Please click here to view a larger version of this figure.
5. Collection of Placental Tissue
6. Neonatal Characteristics
7. Calculations
The 4-vessel sampling method is applicable in clinical practice and we have successfully obtained blood samples from 209 mother/infant-pairs. In 128 of these we also achieved to measure volume blood flow. Complete 4-vessel sampling and good quality flow measurements of both maternal and fetal vessels were obtained in 70 mother-fetus pairs (Figure 3). In addition, we have so far collected blood and placenta samples from 30 preeclamptic patients. We have previously published articles on human placental transfer of nutrients, as well as placental release of vasoactive factors, demonstrating two applications of the method 14,15,16.
Example of how the 4-vessel method is used to study placental transfer
There are significant arteriovenous glucose differences on both sides of the placenta demonstrating an in vivo uteroplacental and fetal uptake of glucose (Table 1). The placental transfer of glucose is dependent on the maternal-fetal glucose gradient, and thereby on the maternal glucose levels. However, we have previously demonstrated that this gradient, and thus the glucose transfer, is significantly influenced also by fetal insulin levels and glucose consumption. This is an example of how this method illustrates important maternal-fetal interplay 14.
Vessel | Glucose mmol/L | p-value* |
Radial artery | 4.49 [4.22, 4.84] | |
Uterine vein | 4.23 [3.94, 4.53] | |
Umbilical vein | 3.78 [3.52, 4.06] | |
Umbilical artery | 3.24 [2.95, 3.56] | |
Paired differences | ||
Radial artery – uterine vein | 0.29 [0.13, 0.41] | <0.001 |
Umbilical artery – umbilical vein | 0.54 [0.29, 0.76] | <0.001 |
Radial artery – umbilical artery | 1.25 [1.03, 1.51] | <0.001 |
Table 1: Median [Q1, Q3] Concentrations and Arteriovenous Differences of Glucose
* Wilcoxon Signed-Rank test
The fetal glucose uptake from (placental release to) the umbilical circulation is believed to be dependent not only on the maternal-fetal gradient, but on placental blood flow 5. Likewise, it may be relevant to study the fetal glucose uptake as a function of placenta weight or birth weight. In n= 128 we found a median [Q1, Q3] total umbilical venous flow of 196.2 [158.3, 232.2] mL/min, and calculated a median [Q1, Q3] fetal glucose uptake from (placental release to) the umbilical circulation of 0.10 [0.05, 0.15] mmol/min. Normalized for birthweight this equals 0.03 [0.02, 0.04] (mmol/min)/kg. The placenta releases 0.16 [0.10, 0.26] (mmol/min) per kg placenta.
Example of how the 4-vessel method is used to study placental uptake
Animal studies suggest that glutamic acid is important both in interconversion of amino acids in the placenta and the fetal liver, and as an oxidative fuel in other metabolic pathways 27. Using the placenta 4-vessel sampling method we studied the uteroplacental and umbilical arteriovenous differences of glutamic acid in humans (Table 2). We found a placental uptake (thus a fetal release) of glutamic acid from the umbilical circulation. Further we found a placental uptake of glutamic acid from the maternal circulation. This placental uptake from both circulations is an example of how the 4-vessel method can be used to demonstrate in vivo in the human that placental metabolism of nutrients is a part of the regulation of the transplacental transfer.
Vessel | Glutamic acid µmol/L | p-value* |
Radial artery | 61.5 [51.0, 77.7] | |
Uterine vein | 51.0 [36.3, 65.0] | |
Umbilical vein | 39.3 [24.7, 52.8] | |
Umbilical artery | 44.7 [33.1, 59.3] | |
Paired differences | ||
Radial artery- uterine vein | 10.4 [1.6, 21.2] | <0.001 |
Umbilical artery –umbilical vein | -8.7 [-16.0, 0.2] | <0.001 |
Table 2: Median [Q1, Q3] Concentrations and Arteriovenous Differences of Glutamic Acid
* Wilcoxon Signed-Rank test
Example of how the 4-vessel method is used to study placental release
It is established that the placenta secretes progesterone and in order to validate our 4-vessel method on the maternal side of the placenta, we measured the in vivo release of progesterone at term 28. We found a significant placental release of progesterone to the maternal circulation (Table 3). The observed arteriovenous difference demonstrates how the placental 4-vessel sampling method can be used to detect substances released by the placenta, and is highly relevant when studying pathological pregnancies.
Vessel | Progesterone nmol/L | p-value* |
Radial artery | 678 [514, 971] | |
Uterine vein | 1852 [1059, 2786] | |
Paired differences | ||
Radial artery- uterine vein | -1187 [-1855, -404] | p<0.001 |
Table 3: Median [Q1, Q3] concentrations and uteroplacental arteriovenous difference of progesterone
* Wilcoxon Signed-Rank test
Figure 3: Flowcharts and illustration of included and lost participants.
A. Shows the inclusion of the participants, demonstrating that participants were lost mainly due to start of labor prior to caesarean section or lack of sufficient personnel to conduct the study. B. Of the 179 women with normal pregnancies (blue) complete 4-vessel blood samples were obtained in 162 (91%) (incomplete fetal blood samples: black, incomplete maternal blood samples: grey). Fifty-one (28%) participants were not included for ultrasound measurements due to logistical limitations. Of the 128 participants (72%) subjected to ultrasound, blood flow measurements at the fetal side of the placenta were obtained in all participants (light green), whereas complete blood flow measurements at both the maternal and fetal side were obtained in 77 (60%) (dark green). Illustration: Øystein H. Horgmo, University of Oslo. Please click here to view a larger version of this figure.
The placenta 4-vessel sampling method is relevant for three main purposes. First, it can be used to study how specific substances are taken up by the placenta on the maternal side and possibly transferred to the umbilical circulation and the fetus, as demonstrated by our glucose and amino acid studies. Second, the method is highly relevant to study substances produced by the placenta and released to the maternal or the fetal circulations, as demonstrated by the progesterone results. Third, it may be useful to study how the fetus in vivo eliminates waste products during the rapid growth and tissue remodeling.
Critical steps and logistical issues of the 4-vessel method
Placenta 4-vessel sampling has previously been used to determine the placental transfer of macronutrients in vivo in normal physiologic pregnancies 8,10,12, but with a limited number of study subjects. The limited use of the 4-vessel sampling is probably due to the demanding logistics of the procedure. A successful coordination between the patient, the researchers and the personnel at the departments of fetal-medicine, obstetrics and anesthesiology is essential in order to employ this method. We believe it has been a great advantage that several of the main investigators are obstetricians, knowing clinical routines and key personnel. Thereby the study procedures have been implemented alongside the daily clinical practice. By coordinating and securing every step of the procedure, we have obtained samples from over 200 mother/newborn-pairs. Furthermore, our strategy has been to keep the sampling procedure on few hands because there are technical challenges in the procedure that are decisive for the successful outcome of the sampling, and too many samplers can introduce unnecessary sources of variation of the data. Consequently, we recommend that all sonographic examinations should be performed by the same examiner using the same ultrasound machine. The equipment should be chosen with care, as the resolution of the vessel wall is crucial. The uterine arteries in third trimester are especially technically challenging to measure because the uterine size and content makes it difficult to obtain both the perpendicular insonation angle used for the diameter measurements, and the low insonation angle for flow velocity measurements at the exact same site. Moreover, successful blood sampling demands identification of appropriate sampling site of the uterine vein and gentle aspiration. On the umbilical side the fragility of the fetal erythrocytes requires particular attention to the aspiration force. It is our experience that delayed delivery or minor stress to the infant was associated with early constriction of the umbilical artery resulting in reduced arterial sample volume.
Ethical issues of the 4-vessel method
The "placenta 4-vessel sampling" method is an invasive and demanding procedure. Inclusion and exclusion criteria should therefore be well defined according to the research question to avoid unnecessary sampling procedures. Patients should only be approached for inclusion after the decision about delivery mode has been made, to ensure that indications for cesarean delivery are not influenced by participation in the research project. Although the procedure requires little extra time at the operation theater, it requires the presence of more personnel making it mandatory to limit the inconvenience and disturbance caused by the sampling. We have used the radial artery as a proxy for the uterine artery as the insertion of an arterial line is less invasive and ensures simultaneous sampling from the artery and vein. Some groups use arterialized blood which is an even less invasive procedure 13. However, apart from one incidence of a local hematoma in relation to the arterial line resulting in temporary paresthesia of the hand, we have not experienced adverse effects during sampling from any of the 4 sites. In particular, we have not observed any bleeding from the punctured uterine vein.
Methodological/ Analytical issues of the 4-vessel method
It is important to address several methodological issues in the interpretation of the results from a placenta 4-vessel study. First, if the aim is to calculate the mass of a substance taken up, or released, by the placenta it is important to consider the volume of blood passing. It should be kept in mind that the uterine vein not only drains the placenta, but also drains the uterine muscle, and that the uterine vasculature in various degrees anastomoses with the ovarian and vaginal vasculature. Next, it is important to consider that the exchange of water across the placenta may influence the concentrations measured, and thereby affect the arteriovenous concentration differences calculated. Ideally this is best addressed by adjusting the concentration differences for the water lost or gained in each mother-fetus pair. This may be achieved by measuring a substance which is not taken up or released by the placenta or the uterus. The hemoglobin concentrations or calculated percentages of erythrocytes (hematocrit) may serve as correction factors for water exchange. Further, when interpreting uptake or release of compounds on the maternal side of the placenta it may be of interest to obtain arteriovenous differences in other tissues for comparison. We have therefore included a blood sample from the antecubital vein to characterize specific features of the placenta by comparing the arteriovenous differences over the placenta with those of the capillary bed of the forearm. We found this comparison particularly interesting when we tested placental release of sFlt-1 and placental growth factor because systemic endothelial cells could be a potential source of these compounds 14. Depending on the research question it might be of importance to relate the arteriovenous differences to the weight of the placenta to calculate the placental efficiency, i.e. in terms of (mmol/L)/kg or (mmol/min)/kg placenta.
Limitations and strengths of the 4-vessel method
Although placental physiology is less affected by cesarean section than by the stress of vaginal delivery, there are several limitations to this method. Vaginal delivery is recommended in most cases of common pregnancy complications (like preeclampsia, diabetes, obesity and moderate fetal macrosomia), which may limit and bias the recruitment. Even when optimizing every step of the method it is hard to obtain complete measurements and samples in each patient because of the technical difficulties in the blood sampling procedure and ultrasound volume flow measurements (Figure 3). In addition, although the ultrasound measurements are conducted as close in time to the surgery as possible, they are inherently conducted prior to the spinal anesthesia and the blood sampling. From this follows that maternal cardiac output (CO) may change and possibly affect uteroplacental (and even feto-placental) blood flow. The possible change in CO caused by spinal anesthesiacan be compensated by phenylephrine which was used in the current study. Preliminary data from a subset of our participants (n= 23) show no significant alteration in CO before spinal anesthesia and at the time of sampling (unpublished data). Using the 4-vessel sampling method in humans, as opposed to in animals, limits both the possibility to introduce a time variable and to manipulate the blood content 5,29,30. From these considerations, it follows that the 4-vessel sampling method is cross-sectional and largely observational by nature, and the data obtained must be analyzed accordingly. On the other hand, the 4-vessel sampling method provides a unique possibility to study human placental physiology and pathophysiology in vivo, with all the interacting factors at play, a situation that can never be reproduced in vitro. It offers an excellent opportunity to test hypotheses that have emerged from animal or other experimental studies. Likewise, it can generate new hypotheses that need to be tested mechanistically in vitro and in animal studies.
Potential applications of the 4-vessel method
In pathological pregnancies, the maternal and fetal arteriovenous concentration differences have, so far, been studied separately, and allowed for testing of some of the existing hypotheses in vivo15,16,31. The 4-vessel sampling method offers the appealing opportunity to study the maternal, placental and fetal unit together rather than as separate entities in pathologic pregnancies and may shed new light on both old and new questions within the scope of maternal-fetal interaction. The 4-vessel sampling method may be applied to a wide range of research questions in both normal and pathological pregnancies depending on the further analyses of the samples. The choice of vacutainers, the volume of blood, the range of placenta and other tissue samples should be decided according to the research question. Burton et al. have recently published an excellent paper describing procedures to ensure good quality samples of placental tissue, and to allow merging of different biobanks in order to address certain puzzles that need a large amount of samples to be solved 20. The 4-vessel samples may be analyzed to study the specific release of exosomes to maternal circulation, the transfer of medication, metabolites and toxins. Big scale omics (metabolomics, proteomics and lipidomics) analyzes has the potential to identify substances and metabolites in maternal plasma that are secreted by the placenta. Thereby the establishment of the 4-vessel sampling method may identify placenta derived factors in the maternal circulation and possible tease out biomarkers of placental function. Combined with techniques to separate the maternal facing microvillous and fetal facing basal plasma membranes of the syncytiotrophoblast, transfer of a substance may be studied together with the activity and location of relevant transporter proteins 21. Further, mechanisms regulating nutrient transfer in vivo may be elucidated by analyzing the levels of nutrients and micronutrients in the four vessels and relate the transfer of nutrients to measurements of the nutrient and energy sensing system in the placenta32. The combination of 4-vessel sampling with the use of stable isotopes may represent a powerful approach that would provide more dynamic and mechanistic information, as demonstrated for amino acid transfer 33. Glucose infusion prior to delivery is another possible approach. Placental transfer could be related to maternal metabolic variables like BMI, glucose and plasma lipid profiles and to fetal outcomes like birthweight and body composition 18. Together, these approaches will possibly illuminate the integrative role of the placenta, being situated in the center of the interplay between maternal and fetal conditions and needs.
The authors have nothing to disclose.
First and foremost, we sincerely thank the mothers who participated in this project. Next, we acknowledge all the personnel that assisted and facilitated the sampling procedure, the Anesthesiologist, the nurse anesthesiologist and the surgical nurses. The project would not have been possible without funding from the South-Eastern Norway Regional Health Authority and Norwegian Advisory Unit on Women’s Health, Oslo University and local funding provided by Oslo University Hospital.
Maternal body composition | |||
Impedance scale | Tanita | or similar | |
Ultrasound measurements | |||
Sequoia 512 ultrasound machine | Acuson | equipped with a curved transducer with colour and pulsewave Doppler (frequency bandwidth 2-6 MHz) | |
Blood samples | |||
Arerial cannula | BD Medical | 682245 | or similar |
20cc Eccentric Luer Tip Syringe without Needle | Termo | SS-20ES | or similar. 3 needed. |
10cc Eccentric Luer Tip Syringe without Needle | Termo | SS-10ES | or similar. 9 needed. |
5cc 6% Luer Syringe without Needle | Termo | SS-05S1 | or similar. 2 needed. |
Arterial blood gas syringe | Radiometer Medical | or similar. 4 needed. | |
Sterile winged needle connected to flexible tubing, 21 gauge | Greiner Bio-One | 450081 | (intended for single use).3 needed. |
Vacutainer tube 6 mL EDTA | Greiner Bio-One | 456043 | or similar. Label with sample site. 10 needed. |
Vacutainer tube 5 ml LH Lithium Heparin Separator | Greiner Bio-One | 456305 | or similar. Label with sample site. 5 needed. |
Vacutainer tube 6 mL Serum Clot Activator | Greiner Bio-One | 456089 | or similar. Label with sample site. 5 needed. |
Vacutainer tube 3 ml 9NC Coagulation sodium citrate 3,2% | Greiner Bio-One | 454334 | or similar. Label with sample site. 5 needed. |
Cryogenic vials, 2.0 mL | Corning | 430488 | or similar. Label with sample site, serum/type of plasma and ID. 90 needed. |
Marked trays to transport the syringes | to transport the blood samples in the operation theatre | ||
Rocker | for gentle mixing of the samples | ||
Ice | in styrofoam box | ||
Liquid nitrogen | in appropriate container | ||
Placenta samples | |||
Metal tray | |||
Ice | in styrofoam box | ||
Calibrated scale | |||
Metal ruler | |||
1 M Phosphate buffered saline | Sigma | D1408 | or similar. Dilute 10 M to 1M before use |
RNA stabilization solution | Sigma | R0901-500ML | or similar |
Optimal Cutting Temperature (O.C.T.) compound | vwr | 361603E | or similar |
Cryogenic vials, 2.0 mL | Corning | 430488 | or similar. Label with sample site. content and ID. 10 needed. |
Centrifuge tubes, conical bottom 50 mL | Greiner Bio-One | 227,285 | or similar. Label with "RNA later", sample site and ID. 2 needed. |
Liquid nitrogen | in appropriate container | ||
Fetal body composition | |||
Calibrated scale | |||
Measuring tape |