T2*-weighted placental magnetic resonance imaging: a biomarker of placental dysfunction in small-for-gestational-age pregnancies

BACKGROUND: The antenatal identification of placental dysfunction in small-for-gestational-age fetuses with normal fetal Doppler flows remains an obstetrical challenge. In a significant fraction of such pregnancies, placental dysfunction is revealed by clinical manifestations such as preeclampsia, preterm delivery, or severe small-for-gestational-age at birth or by abnormal findings in the postnatal placental histologic examination. Therefore, new methods to identify placental function directly in pregnancy at the time of small-for-gestational-age diagnosis is highly needed. T2*-weighted placental magnetic resonance imaging is sensitive to changes in placental morphology and oxygenation and is thereby related to placental function. Previous studies have demonstrated that pregnancies complicated by low birthweight and preeclampsia are characterized by low placental T2* values. However, the specific performance of placental T2* in the prediction of placenta-related outcomes in small-for-gestational-age pregnancies with normal fetal Doppler flows remains to be explored. OBJECTIVE: In small-for-gestational-age pregnancies with normal fetal Doppler flows, we aimed to evaluate T2*-weighted placental magnetic resonance imaging as an antenatal biomarker of placental dysfunction. In addition, we aimed to investigate the correlation between placental T2* and Doppler flow measurements of fetal and uterine arteries at the time of magnetic resonance imaging. STUDY DESIGN: In this prospective cohort study, the inclusion criterion was suspected small-for-gestational-age (ultrasound estimated fetal weight Z-score ≤−2.0 [2.3rd centile]) with normal fetal Doppler flows (middle cerebral artery pulsatility index Z-score > −2.0 and umbilical artery pulsatility index Z-score <2.0). The T2*-weighted placental magnetic resonance imaging scan was performed at inclusion in a 1.5 T system. The outcomes was placental dysfunction at birth defined by low birthweight (Z-score ≤−2.0), preeclampsia, preterm delivery (gestational age<37 weeks), or abnormal placental histologic examination such as placental vascular malperfusion according to the Amsterdam Consensus Statement. RESULTS: We included 92 pregnancies at 26+5 to 39+6 weeks gestation. The median time interval between the magnetic resonance imaging scan and birth was 4.6 weeks (interquartile range, 2.7–7.8 weeks). At birth, 55% (51/92) of pregnancies revealed at least 1 sign of placental dysfunction; 49% (40/81) had abnormal placental histologic examination, 29% (27/92) were born with low birthweight, 13% (12/92) were delivered preterm, and 7% (6/92) had preeclampsia. When adjusted for gestational age at magnetic resonance imaging, the placental T2* Z-score was a significant predictor of abnormal placental histologic examination (area under the curve, 0.73; P=.001), small-for-gestational-age at birth (area under the curve, 0.63; P=.030), preeclampsia (area under the curve, 0.88; P=.005), and preterm delivery (area under the curve, 0.81; P=.001). The placental T2* was reduced in pregnancies with a combination of clinical manifestations and abnormal placental histologic examination (T2* Z-score=−1.52±1.35 [mean±standard deviation]; P=.0001) and in clinically uneventful pregnancies with abnormal placental histologic examination (T2* Z-score=−0.79±0.97; P=.045). At the time of magnetic resonance imaging, the placental T2* Z-score showed a significant linear correlation with the uterine artery pulsatility index Z-scores (r=−0.24; P=.016) and the middle cerebral artery pulsatility index Z-scores (r=0.29; P=.017) but not with the umbilical artery pulsatility index Z-scores (r=0.18; P=.17) and the cerebroplacental ratio (r=0.03; P=.77). CONCLUSION: This study indicates that placental dysfunction is frequent in small-for-gestational-age fetuses with normal fetal Doppler flows. In this cohort, T2*-weighted placental magnetic resonance imaging is a sensitive biomarker of placental dysfunction regardless of the clinical manifestations. This finding supports a paradigm shift in the conception of placental dysfunction that may cover a wide spectrum of clinical and subclinical manifestations.