Bridging cotyledon histopathology and perfusion in healthy primate pregnancy

Healthy and diseased placentae alike often display some degree of histopathology. However, quantitative techniques to characterize common placental lesions and their relationship to local maternal hemodynamics in healthy pregnancy are currently limited.

Seven pregnant macaques with healthy fetoplacental biometrics were recruited for study. Placentae were imaged by MRI at three time points across mid-to late-gestation to quantify volume, blood flow, and perfusion of maternal blood across pregnancy. Near term, we collected placental cotyledons, imaged hematoxylin/eosin-stained slides, then annotated major histopathological features (sub-tissues, intervillous gaps, fibrin deposition, villous agglutination, inflammatory agglutination, and stromal mineralization) within each cotyledon. Histopathological lesions were assessed in relation to each other and MRI-based perfusion metrics, in a cotyledon-specific manner. Parallel analyses were performed to investigate both basic (Spearman correlation) and animal variance-conditioned (dimensionality-reduction) relationships.

Dimensionality reduction analysis revealed maternal vascular malperfusion-associated histopathology as the main contributor to dataset variance and was positively associated with increased maternal blood perfusion across gestation. Cotyledons with increased stromal mineralization demonstrated low blood perfusion across pregnancy, alongside changes potentially indicative of maternal compensation. Additionally, histopathology associated with healthy placental function demonstrated low cotyledon blood flow at all timepoints.

We revealed several relationships connecting histopathology and perfusion in the healthy pregnancy, such as maternal vascular malperfusion-associated lesions increasing with increased maternal blood perfusion - all accomplished at the smallest functional unit of the placenta. This methodological framework embeds pathologist-refined morphological expertise into a quantitative, spatially resolved format that can better ground unsupervised computational approaches to placental analysis.