Abstract

Objective Human fetal growth restriction (FGR) is associated with cardiac dysfunction. However, it remains unclear whether alterations in the fetal heart in human pregnancy affected by FGR are a consequence of chronic fetal hypoxia. In this study, we used novel in‐vivo ultrasound imaging modalities in pregnant sheep to evaluate fetal cardiac responses to progressive hypoxia in late gestation. Methods This was a prospective study of data collected between July 2019 and March 2022 from 42 Welsh mountain ewes, each carrying a single fetus. Seventeen ewes were exposed to hypoxia in isobaric chambers (10% oxygen (O2)) from 105 days of gestational age (dGA; with term being approximately 147 days) for 11 ± 1 days (H1, n = 5), 17 ± 1 days (H2, n = 6) or 32 ± 2 days (H3, n = 6). An ultrasound system with vendor‐specific software was used to determine fetal left ventricular (LV) and right ventricular (RV) geometry and function. Outcomes were compared with the remaining 25 gestational age‐matched ovine normoxic control pregnancies at 117 ± 1 (N1, n = 13), 124 ± 1 (N2, n = 5) and 133 ± 3 (N3, n = 7) dGA. Results Compared to controls with expected fetal RV dominance in late gestation, hypoxic fetal sheep showed LV dominance with increasing duration of hypoxia, including statistically significant increases in mean ± SD LV/RV end‐diastolic area ratio (N1, 1.1 ± 0.2 vs H1, 1.3 ± 0.1; N2, 1.2 ± 0.4 vs H2, 1.7 ± 0.5; N3, 0.9 ± 0.1 vs H3, 1.4 ± 0.2), LV sphericity index (N3, 0.44 ± 0.04 vs H3, 0.54 ± 0.11) and LV/RV cardiac output ratio (N1, 0.55 ± 0.18 vs H1, 1.00 ± 0.31; N2, 0.80 ± 0.10 vs H2, 1.62 ± 0.59; N3, 0.74 ± 0.23 vs H3, 1.50 ± 0.58). The mean ± SD LV myocardial performance index was significantly greater in the hypoxic groups, signifying global myocardial dysfunction (N1, 0.45 ± 0.07 vs H1, 0.64 ± 0.07; N2, 0.40 ± 0.05 vs H2, 0.66 ± 0.07; N3, 0.36 ± 0.07 vs H3, 0.60 ± 0.10). While LV apical radial strain and LV apical systolic rotation were initially increased after 17 days of hypoxia, these indices were significantly reduced after 32 days of hypoxia (median LV apical radial strain: N1, 27% (interquartile range (IQR), 26–28%) vs H1, 24% (IQR, 21–26%); N2, 24% (IQR, 22–25%) vs H2, 70% (IQR, 66–79%); N3, 51% (IQR, 48–54%) vs H3, 29% (IQR, 27–32%); mean ± SD LV apical systolic rotation: N1, 7 ± 5° vs H1, 5 ± 3°; N2, 9 ± 1° vs H2, 14 ± 2°; N3, 15 ± 2° vs H3, 7 ± 2°). Hypoxic fetuses showed biventricular hypertrophy and evidence of biventricular diastolic dysfunction, with significant LV impairment presenting after 11 days of hypoxia (mean ± SD LV isovolumetric relaxation time (IVRT′) normalized by cardiac cycle (cc) length: N1, 0.10 ± 0.02 ms vs H1, 0.15 ± 0.02 ms; N2, 0.09 ± 0.01 ms vs H2, 0.14 ± 0.02 ms; N3, 0.08 ± 0.03 ms vs H3, 0.12 ± 0.02 ms), preceding RV impairment after 32 days of chronic hypoxia (mean ± SD RV‐IVRT′ normalized by cc length: N3, 0.08 ± 0.03 ms vs H3, 0.14 ± 0.03 ms). Conclusions Progressive fetal hypoxia in sheep leads to profound changes in fetal cardiac structure and function, resulting from a switch to LV dominance triggered by fetal brain sparing. The findings also indicate that fetal cardiac compensatory reserves become exhausted with progressive hypoxia. © 2025 The Author(s). Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.