The goal of this study was to determine whether maternal anemia affects placental and fetal oxygenation throughout gestation, as assessed by PAI. To summarize, maternal iron restriction caused: (1) progressively severe maternal anemia throughout pregnancy and severe anemia and growth restriction in fetus by GD21; (2) no effects on maternal, placental and fetal oxygenation throughout pregnancy when dams were exposed to FiO2 21%; (3) increased placental oxygenation (particularly on the fetal face) when dams were exposed to FiO2 100% on GD21; (4) minimal differences in tissue oxygenation when dams underwent an acute hypoxic episode. These findings may suggest reduced oxygen extraction and utilization by IDA fetuses in late gestation.
Outside of pregnancy, oxygen delivery to most tissues exceeds metabolic demands (20), and therefore the reduced oxygen carrying capacity associated with mild anemia can be buffered by increased tissue oxygen extraction (21, 22). However, there is a limit to this reserve; in cases of severe anemia, oxygen extraction exceeds supply and results in impaired tissue oxygenation (23) (24, 25). During gestation, physiological reserve is diminished, in part due to increased metabolic demands imposed by growth and development of the fetus and placenta, as well as a reliance on the maternal circulation for oxygen and nutrient supply. We therefore expected sO2 levels to decrease in IDA groups as maternal tissue, placentas, and fetuses may already extract a large portion of delivered oxygen. Contrary to our hypothesis, we found that tissue sO2 were largely unaffected by anemia at all gestational days. In fact, sO2 were increased in the fetal face of the placenta in IDA at GD21 with exposure to FiO2 of 100%, suggesting overall reduced oxygen extraction. This was surprising, because GD21 not only reflects the period during which maternal and fetal anemia are most severe (i.e. low resources), but the last 3 days of gestation also represents a period during which fetal weight increases 3-4-fold (i.e. high metabolic demand) (26).
This seemingly paradoxical finding could be explained in the context of fetal resource allocation in the wake of a suboptimal uterine environment. We and others have shown that IDA results in maladaptive changes in the placenta; despite being larger, IDA placentas have proportionally smaller labyrinth zones (26) with reduced vascularization (27). The placentas are therefore less efficient, and unable to sustain optimal growth and development, particularly as anemia worsens and metabolic demands of the fetus increase throughout pregnancy (26). When faced with this suboptimal in utero environment, the fetus may alter its developmental trajectories in an effort to reduce energy expenditure, and allocate the limited resources available to essential processes that ensure survival (26, 28). In this way, the lower growth rate of the IDA fetus is sustainable despite a less-efficient placenta and the relative hypoxemia induced by maternal anemia, and the transient provision of excess oxygen (i.e., exposure to FiO2 100%) is not used by the fetus. This could explain why sO2 levels in the fetal face of the placenta remain elevated in the IDA group, but fetal values are unchanged. Similarly, the acute exposure to hypoxia had comparable effects on sO2 levels in the IDA group as controls despite overall reduced fetal oxygen delivery (due by anemia) (28) because of the proportionally smaller IDA fetus.
In addition to measuring differences between compartments (i.e. fetus, placenta), we also show that the newest generation of PAI could clearly measure differences in sO2 between the maternal and fetal faces of the placenta at GD13 in rats. Notably, these differences could not be detected using 2D imaging, but only using the sensitive 3D imaging, which measures oxygenation over the whole region rather than a single plane. This is perhaps not surprising because placental vascularization is not homogenous, and this central planar image is unlikely to reflect oxygenation of the entire placenta. This finding is consistent with a report by Arthuis et al. that showed 2D PAI could not detect placental regional differences in sO2 at GD14 in rats (14), whereas Yamaleyeva et al. could using 3D PAI at GD14 in mice (13); an important caveat here is that GD14 represents a comparatively later developmental timepoint compared to rats, since gestational length is 4 days shorter in mice (term = GD18). Despite the increased sensitivity of the 3D imaging approach, we also show that a single plane (as assessed using 2D imaging) is sensitive to changes in inspired oxygen, and can be measured in real time. Therefore, PAI could be used to measure oxygenation noninvasively in the context of acute fetal or maternal distress.
Although PAI has been used in the clinical setting, it has mostly been adopted as a research tool to measure blood flow, oxygenation, or metabolite clearance in peripheral tissues (e.g., tumour, muscle) and the vasculature (29–31). Feto-placental imaging by PAI could be useful for monitoring fetal well-being. However, currently depth of penetration remains a limitation for its use in obstetrics, since imaging tissues with a depth beyond 3.5 cm is not feasible. However, PAI may be a practical choice in many cases, particularly in the third trimester of pregnancy, when the placenta is pressed closer to the surface (32, 33). It is important to remember that PAI is a relatively new technology, with commercially available lasers only recently entering the market. The development of powerful lasers which can quickly switch between wavelengths, in this case to excite oxygenated and deoxygenated Hb; as well as ultrasound transducers sensitive enough to measure these signals is currently underway, with these advancements PAI's applications in obstetrics could be more widespread. there are other imaging modalities that could be used to measure oxygen saturation and delivery (e.g., MRI (34)), PAI offers advantages, including real-time imaging, portability, and cost-effectiveness. Most importantly, due to the ubiquity of sonography use at the bedside, PAI could be easily integrated as a standard of care, particularly with the ongoing developments of 3D photoacoustic tomography which will further improve resolution.
In conclusion, photoacoustic imaging offers a method to measure fetal, placental, and maternal oxygenation noninvasively. To our surprise these results show that IDA results in little or no differences in tissue oxygenation as measured with the current system, which may reflect fetal and placental adaptations to ensure fetal survival in a suboptimal gestational environment. This technology has the potential to determine the role of placental and fetal hypoxia in several different disease models, regardless of fetal or maternal hemoglobin.