Assessing within‐subject rates of change of placental MRI diffusion metrics in normal pregnancy

• Published in: Magnetic resonance in medicine Vol. 90; no. 3; pp. 1137 - 1150
• Main Authors: Cromb, Daniel, Slator, Paddy J., De La Fuente, Miguel, Price, Anthony N., Rutherford, Mary, Egloff, Alexia, Counsell, Serena J., Hutter, Jana
• Format: Journal Article
• Language: English
• Published: United States Wiley Subscription Services, Inc 01.09.2023; John Wiley and Sons Inc
• Subjects: Benchmarking; Compartments; Covariance; Cross-Sectional Studies; diffusion imaging; Diffusion Magnetic Resonance Imaging - methods; Diffusion rate; Female; Gestation; Humans; longitudinal imaging; Magnetic resonance imaging; Magnetic Resonance Imaging - methods; Medical imaging; Microstructure; Motion; Placenta; Placenta - diagnostic imaging; placental MRI; Preclinical and Clinical Imaging; Pregnancy; T2$$ {\mathrm{T}}_2^{\ast } $$ relaxometry; Trends; placental MRI; relaxometry; diffusion imaging; longitudinal imaging
• DOI: 10.1002/mrm.29665
• ISSN: 0740-3194; 1522-2594

Purpose Studying placental development informs when development is abnormal. Most placental MRI studies are cross‐sectional and do not study the extent of individual variability throughout pregnancy. We aimed to explore how diffusion MRI measures of placental function and microstructure vary in individual healthy pregnancies throughout gestation. Methods Seventy‐nine pregnant, low‐risk participants (17 scanned twice and 62 scanned once) were included. T2‐weighted anatomical imaging and a combined multi‐echo spin‐echo diffusion‐weighted sequence were acquired at 3 T. Combined diffusion–relaxometry models were performed using both a T2*$$ {\mathrm{T}}_2^{\ast } $$‐ADC and a bicompartmental T2*$$ {\mathrm{T}}_2^{\ast } $$‐intravoxel‐incoherent‐motion (T2*IVIM$$ {\mathrm{T}}_2^{\ast}\;\mathrm{IVIM} $$) model fit. Results There was a significant decline in placental T2*$$ {\mathrm{T}}_2^{\ast } $$ and ADC (both P < 0.01) over gestation. These declines are consistent in individuals for T2*$$ {\mathrm{T}}_2^{\ast } $$ (covariance = −0.47), but not ADC (covariance = −1.04). The T2*IVIM$$ {\mathrm{T}}_2^{\ast}\;\mathrm{IVIM} $$ model identified a consistent decline in individuals over gestation in T2*$$ {\mathrm{T}}_2^{\ast } $$ from both the perfusing and diffusing placental compartments, but not in ADC values from either. The placental perfusing compartment fraction increased over gestation (P = 0.0017), but this increase was not consistent in individuals (covariance = 2.57). Conclusion Whole placental T2*$$ {\mathrm{T}}_2^{\ast } $$ and ADC values decrease over gestation, although only T2*$$ {\mathrm{T}}_2^{\ast } $$ values showed consistent trends within subjects. There was minimal individual variation in rates of change of T2*$$ {\mathrm{T}}_2^{\ast } $$ values from perfusing and diffusing placental compartments, whereas trends in ADC values from these compartments were less consistent. These findings probably relate to the increased complexity of the bicompartmental T2*IVIM$$ {\mathrm{T}}_2^{\ast}\;\mathrm{IVIM} $$ model, and differences in how different placental regions evolve at a microstructural level. These placental MRI metrics from low‐risk pregnancies provide a useful benchmark for clinical cohorts.