Autopsy revision to the MRI stillbirth proposal

            I revised the autopsy protocol portion for the MRI Stillbirth study and posted it. 

The MRI Stillbirth study has three aims directly referring to three parts of the methods. Part 1 is the in vivo MRI of the umbilical cord. Part 2 is the close questioning of the parents about immediate events around the time of the stillbirth and Part 3 is gathering correlation from clinical risk factors, the autopsy and the placenta. The next method section that needs expanding is the placenta examination. I want to get beyond the cataloging of lesions to the calculation of the reserve of the placenta to survive either decreased fetal blood flow, decreased maternal blood flow or decreased maternal oxygenation. 

            To achieve this goal, I need to do some preliminary work. A first problem is estimating the volume of the placenta, as any histological measure of function will need to be multiplied by that volume. A crude estimate is to use either the placental weight or the surface area times an estimate of the thickness. Another approach is that the MRI could give an in vivo estimate of placenta volume. The MRI would measure the volume under normal (or possibly reduced) maternal inflow. Early in my career, I developed a simple approach to measuring placental volume.  I bought a second-hand meat slicer to cut even 1 cm placenta slices that could be photographed and traced to yield a volume. To my surprise, it worked. I probably would not take that kind of risk today with sharp rotating electric blades. The problem with simply putting the placenta into a fluid to measure displacement is that the intervillous space can absorb some of the fluid, and the volume of interest is the combined placenta substance and the intervillous space, or at least that portion of the intervillous space that does not collapse after delivery of the placenta.

            The quantity of capillary syncytial membranes (cap-syn membrane) is the most important element in determining the reserve for oxygen exchange. Mayhew’s group demonstrated oxygen diffusion varies by the inverse third power of the barrier thickness between the fetal and maternal red cells¹. Therefore, oxygen exchange depends almost entirely on the area of cap-syn membrane, analogous to alveolar capillary exchange in the lung. The problem is how to measure the surface area of these membranes over the whole placenta. One long standing approach first used by Harold Fox was to count the number of membranes per terminal villus on the microscope slide. This approach has two problems. The first is finding the total volume of terminal villi in a placenta which will be considered under sampling below. The second is that the estimate of the linear extent of cap-syn membranes along a villus likely depends on the extent of capillary distention which for whatever reason (perhaps cord clamping time) varies remarkably among placental samples that we see in practice. We might get around this by distending all placentas via the umbilical vein to the same pressure of saline, but this presents problems. Instead, I want to do a preliminary study of the relationship of the villous surface expansion to the capillary diameters by experimentally infusing a few chorionic veins at selected pressures. With luck the relationship will be simple and there will be a formula that will give the linear extent of cap syn membrane using the mean capillary width, and the villous area on the microscope slide.

            Maturation of the placenta, that is the increase in terminal villi with syncytial knots and cap-syn membranes appears directly related to maternal blood flow. That blood flow is not uniform because of the placentone effect, which is the perfusion volume of one spiral artery². In the very mature placenta, this effect may be visible on a single slide with two ends showing small villi with increased syncytial knots, and a central zone of less mature appearing villi. This pattern is a reflection of the spiral artery inflow as a fountain coming in through a basal orifice and flowing down the periphery, which was shown in Dr. Ramsey’s radiographic primate studies³. The phenomenon at the extreme can be seen grossly by ultrasound in the grade 4 placenta. A few random samples for histology may not consistently capture the effect of this difference in regional flow. In order to estimate the total volume of terminal villi for gas exchange, a different sampling approach may be needed.

            To understand the placentone concept, consider: The law of continuity (a variant of conservation of mass in a flowing system) requires that the velocity of flow times the perpendicular cross-sectional area must remain constant. This is why a nozzle increases the velocity or fluid from a down pointing faucet narrows. This concept gets complicated in the intervillous space, but as the blood in the fountain from the spiral artery comes back down from the fetal surface, it covers a wider intervillous volume and must slow down, which would improve the oxygen extraction. The immature villi that actively extract nutrients tend to be in the center of spiral artery inflow and the mature villi with oxygen extraction will be at the periphery of the fountain. The problem is how to account for the effect of the placentone on sampling and still have a reasonable number of samples to determine the anatomic capacity to extract oxygen. 

            I have considered sampling not in the usual vertical manner but taking horizontal sections in sequence. Not only will this put the sections in context of the spiral artery entrance, but it will give a better idea of the size of the placentone effect, a value that may be a proxy for maternal flow. Finally, stem villi do not correlate precisely with the spiral artery flow, so an approach that samples stem villi at the top, and spiral arteries and basal veins on the bottom can map a stem villous to placentone relationship. The method would first fix an area of placenta large enough to completely encompass one placentone. Then all corners would be color inked to maintain orientation of the horizontal sections to each other. This approach, which is tedious, would need to be evaluated to see if it really is a more accurate way to get a functional picture of the placenta. 

            A final complication to measuring total functional capacity of the placenta in a stillbirth is the effect of fetal death on the fetal vessels that mimics cessation of flow from local in vivo occlusions. One difference is that thrombi by definition do not form after death, but mapping loss of villi from fetal vascular malperfusion will require tracing vascular trees and correlating changes with other measures of postmortem retention. 

            The premortem fetal blood flow may be estimated from the umbilical vein diameter and thickness as in general the diameter of the vein will reflect flow and the thickness of the media pressure. As with capillary beds, we need to distend the vein at different pressure in vitro to see the relationship of wall thickness to diameter under pressure. The postpartum umbilical vein diameter and wall thickness also need to be correlated with in vivo Doppler studies. 

            A final preliminary study needs to look further at a study of CD15 staining to identify immature villi that was positive in stillbirth⁴. With acute abruption, the staining was similar to controls. Few details of the IUFD category were provided. Cases of infants with hydrops and with parvo infection were positive as well. This may be a very helpful marker of villous dysmaturity, but it needs more preliminary work with autopsy correlation. 

            Unlike the autopsy portion of methods, the placental examination is going to explore new approaches. My plan is to write a separate protocol for these that uses “waste tissue”, that is portions of placenta and cord not needed for diagnosis. That unfunded proposal should receive accelerated IRB approval, and provide some preliminary data by the time the MRI stillbirth protocol is submitted. The MRI Stillbirth study is a pilot, but I want to have better formed placental hypotheses to be further tested in the context of the study. 

1.         Mayhew TM, Jackson MR, Haas JD. Microscopical morphology of the human placenta and its effect on oxygen diffusion: a morphometric model. Placenta 1986;7:121-31.

2.         Schuhmann RA. Placentone structure of the human placenta. Biblthca anat 1982;22:26-57.

3.         Ramsey EM, Donner MW. Placental vasculature and circulation. Philadelphia: W. B. Saunders Company Ltd; 1980.

4.         Seidmann L, Suhan T, Kamyshanskiy Y, Nevmerzhitskaya A, Gerein V, Kirkpatrick CJ. CD15 – a new marker of pathological villous immaturity of the term placenta. Placenta 2014;35:925-31.