Abnormal Doppler and perinatal outcomes according to the placental lesions of maternal and fetal vascular malperfusion in preterm fetal growth restriction

Statistical analysis

Continuous variables were compared by Student’s t-test and summarized as mean and standard deviation. Categorical variables were tested by the Chi-square test or Fisher’s exact test and summarized as frequency and percentage. We calculated odds ratio (OR) using a binary logistic regression model. Crude OR was calculated for the univariable model, and adjusted OR was calculated for the multivariable model after controlling for GAD. OR and 95 % confidence interval (CI) were expressed visually using forest plots. p-values less than 0.05 were considered statistically significant. All statistical analyses were conducted using SAS software, version 9.4 (SAS Institute Inc., Cary, NC, USA), and R software, version 4.4.0 (R Project for Statistical Computing).

Baseline characteristics of the study population

Among the study population, the average GAD and neonatal birth weight were 32 weeks (±4 weeks) and 1,299 g (±575 g), respectively. The smallest baby was 230 g at 24+5 weeks, and the most premature baby was born at 23+0 weeks with a birth weight of 380 g. In our study population, preeclampsia was comorbid in 48.4 % of cases (225/465). With regard to placental findings, lesions of the MVM or FVM were observed in 90.3 % (419/495) and 44.8 % (208/495) of patients, respectively. Among cases with available Doppler results, abnormal UA Doppler was found in 32.8 % (131/400) and abnormal MCA Doppler was found in 70.4 % (267/379).

Villous and vascular lesions of MVM or FVM in the placenta

We compared maternal characteristics, Doppler findings, and perinatal outcomes according to villous and vascular lesions of MVM or FVM in the placenta (Table 1). Both villous and vascular (both p<0.001) lesions of MVM were associated with higher rates of preeclampsia. Interestingly, villous lesions of FVM were associated with a significantly lower rate of preeclampsia (32.6 vs. 52.7 %, p<0.001), whereas vascular lesions of FVM showed a trend toward lower prevalence (37.7 vs. 50.3 %, p=0.054), although this trend did not reach significance. Villous lesions of MVM tended to be associated with advanced maternal age (p=0.057), whereas vascular lesions of MVM were associated with higher pre-pregnancy BMI (p=0.002).

Both villous and vascular lesions of MVM were significantly associated with abnormal UA and abnormal MCA Doppler (p<0.05 for all). Villous and vascular lesions of FVM were also associated with abnormal MCA Doppler (both p<0.05) but not with UA Doppler.

For perinatal outcomes, both villous and vascular lesions of MVM and FVM were associated with lower neonatal birth weight (p<0.01 for all). Villous and vascular lesions of MVM were also related with higher rate of Cesarean section.

Of note, neither villous nor vascular lesions of MVM were associated with neonatal mortality in our study population. However, both villous (15.8 vs. 8.2 %, p=0.024) and vascular (20.3 vs. 8.1 %, p=0.002) lesions of FVM were significantly related to neonatal mortality.

Specific key findings of MVM or FVM in the placenta

Next, we compared maternal characteristics, Doppler findings, and perinatal outcomes according to the specific key findings of MVM and FVM in the placenta, which were considered clinically significant [4], 18]. As shown in Table 2, distal villous hypoplasia, acute atherosis of basal plate arteries and/or decidual arterioles, fetal thrombotic vasculopathy, and intramural fibrin deposition in large fetal vessels were analyzed.

Distal villous hypoplasia (53.8 vs. 27.4 %, p=0.017) and acute atherosis of basal plate arteries and/or decidual arterioles (38.5 vs. 24.2 %, p=0.01) were associated with abnormal UA Doppler. Similarly, these lesions were significantly associated with abnormal MCA Doppler (84.6 vs. 56.6 %, p=0.038 for distal villous hypoplasia and 69.2 vs. 52.8 %, p=0.017 for acute atherosis of basal plate arteries and/or decidual arterioles). Fetal thrombotic vasculopathy and intramural fibrin deposition of large fetal vessels were associated with abnormal MCA Doppler (p<0.05 for each) but not with UA Doppler.

For perinatal outcomes, four of the specific villous and vascular lesions of MVM or FVM were associated with lower neonatal birth weight (p<0.05 for all). However, only intramural fibrin deposition in large fetal vessels was significantly related to neonatal mortality (22.9 vs. 8.4 %, p=0.004).

Risk of abnormal Doppler and perinatal outcomes according to MVM or FVM of the placenta

We calculated OR by crude and adjusted models (adjusted by gestational age at delivery and preeclampsia) and visualized the results as forest plots. For perinatal outcomes, Cesarean section and neonatal mortality were selected as outcome variables.

Villous MVM was associated with abnormal UA and MCA Doppler (Figure 1). However, this association was attenuated after adjustment for GAD and/or preeclampsia. A similar trend was observed for vascular MVM lesions. However, for villous and vascular changes of FVM, the association with abnormal MCA Doppler remained significant even after adjustment for GAD. The associations between villous and vascular changes of MVM and Cesarean section were attenuated after adjustment for GAD and/or preeclampsia.

Figure 1:

Forest plot of risk of abnormal Doppler and perinatal outcomes according to placental findings of maternal vascular malperfusion (MVM) and fetal vascular malperfusion (FVM). Odds ratio (OR) with 95 % confidence interval (CI) of the crude model in univariable analysis and of the adjusted model in multivariable analysis after controlling for GAD and preeclampsia. UA, umbilical artery; MCA, middle cerebral artery.

Regarding four specific key findings of MVM or FVM in the placenta, distal villous hypoplasia was highly associated with abnormal UA Doppler (crude OR 5.01, 95 % CI 1.27–19.68) (Figure 2). Acute atherosis of basal plate arteries and/or decidual arterioles was associated with Cesarean section (crude OR 2.35, 95 % CI 1.27–4.33). Intramural fibrin deposition was associated with abnormal MCA Doppler (crude OR 2.49, 95 % CI 1.01–6.13) and neonatal mortality (crude OR 3.25, 95 % CI 1.52–6.92). As expected, adjustment for GAD and/or preeclampsia resulted in attenuation of most associations.

Figure 2:

Forest plot of risk of abnormal Doppler and perinatal outcomes according to key findings of maternal and fetal vascular malperfusion in the placenta. Odds ratio (OR) with 95 % confidence interval (CI) of the crude model in univariable analysis and of the adjusted model in multivariable analysis after controlling for GAD and preeclampsia. UA, umbilical artery; MCA, middle cerebral artery.

Study summary

Placental MVM lesions (villous or vascular) were observed in the majority of FGR neonates delivered preterm and were significantly related to abnormal UA and MCA Doppler findings. Meanwhile, FVM lesions were observed in about half of our study population and were significantly related to abnormal MCA but not abnormal UA Doppler. Among specific key findings of MVM or FVM, distal villous hypoplasia was highly associated with abnormal UA Doppler (crude OR 5.01), and intramural fibrin deposition of large fetal vessels was significantly associated with abnormal MCA Doppler (crude OR 2.49). Both villous and vascular lesions of MVM and FVM were associated with earlier gestational age at delivery and lower neonatal birth weight. FVM lesions but not MVM lesions of the placenta were associated with neonatal mortality. Specifically, intramural fibrin deposition in large fetal vessels was most closely associated with neonatal mortality with a crude OR of 3.25. However, most of these associations were attenuated when adjusted for GAD and/or preeclampsia, emphasizing the potential mediating role of prematurity and maternal hypertensive disorders in these outcomes.

Associations between placental pathology and Doppler

Our observations that placental findings of MVM are closely related to abnormal UA and MCA Doppler are in line with those of previous studies [11], [12], [13], [14, 19], 20]. Our study expands existing knowledge by demonstrating associations with specific lesions. Distal villous hypoplasia, a high-grade MVM lesion, was significantly associated with abnormal UA Doppler, suggesting that villous maldevelopment can substantially increase fetoplacental vascular resistance [21], 22], leading to absent or reversed end-diastolic flow. Intramural fibrin deposition in large fetal vessels, a type of high-grade FVM lesion, was associated with abnormal MCA Doppler, possibly reflecting thrombo-occlusive phenomena compromising cerebral perfusion [4], 23], 24]. These mechanistic interpretations align with pathophysiologic models of fetoplacental circulation and reinforce the diagnostic utility of Doppler surveillance in FGR.

Our results are in agreement with those of a recent study by Spinillo et al. including 250 early-onset FGR neonates [25], which also observed distinct associations between abnormal UA Doppler and severe MVM lesions [14]. However, our study demonstrated a higher prevalence of MVM (90 vs. 42.8 %), likely reflecting differences in study population, as our patients were delivered at earlier gestational ages with lower birth weights, representing a more severe FGR phenotype. Consistent with other studies [26], [27], [28], our data support the hypothesis that MVM lesions are more prevalent and clinically relevant in early-onset compared to late-onset FGR. We observed similar results, as abnormal MCA Doppler was related to both villous and vascular changes of MVM and FVM and, more specifically, to intramural fibrin deposition in large fetal vessels [13].

Associations between placental pathology and perinatal outcomes

Consistent with previous studies [29], [30], [31], we found that villous or vascular MVM and FVM lesions were significantly related to earlier gestational age at delivery and lower birth weight. Notably, neonatal mortality itself was associated with FVM but not MVM lesions of the placenta, suggesting that FVM may more directly compromise fetal status or reflect more immediate fetal compromise. Of note, intramural fibrin deposition in large fetal vessels, a high-grade type of FVM lesion, was significantly related to neonatal mortality although the significance decreased when adjusted for gestational age at delivery and preeclampsia, suggesting that perinatal outcomes are partially mediated by these clinical factors.

Associations between placental pathology and preeclampsia

We confirmed that preeclampsia is associated with villous and vascular lesions of MVM including distal villous hypoplasia and acute atherosis of basal plate arteries and/or decidual arterioles. In contrast, FVM was not associated with a higher rate of preeclampsia with FGR. Instead, fetal thrombotic vasculopathy was less frequent in cases of preeclampsia with FGR, suggesting that other mechanisms, such as umbilical cord abnormalities or fetal thrombotic conditions, contribute to FVM independent of maternal hypertensive disease.

Distal villous hypoplasia

Although the actual prevalence of distal villous hypoplasia was low (2.8 %), it was significantly associated with earlier gestational age at delivery and lower neonatal birth weight. Pathophysiologically, distal villous hypoplasia represents villous maldevelopment under chronic hypoxia, leading to increased vascular resistance in the fetal vessels, further causing absent or reversed end-diastolic UA Doppler [22], 32]. Due to increased vascular resistance, the normal villous branching is disrupted, resulting in the formation of narrow and elongated villous structures that restrict transplacental gas transfer [11], 21]. Although the crude OR for abnormal UA Doppler was strong, this was lost after adjustment for GAD and/or preeclampsia, highlighting the influence of gestational age and preeclampsia.

Intramural fibrin deposition

Intramural fibrin deposition in large fetal vessels, observed in 10.3 % of cases, was the only specific lesion consistently associated with neonatal mortality. Its association with abnormal MCA Doppler (crude OR 2.49) suggests a link to altered cerebral hemodynamics and possibly brain injury [13], 23], 31]. Pathophysiologically, such lesions may reflect increased venous pressure from partial cord occlusion, leading to endothelial damage and thrombus formation in large fetal vessels [4], 24]. Although adjusted models weakened the association, the crude effects remain clinically meaningful.

Strengths and limitations

A major strength of this study is the use of centralized and standardized histopathologic evaluation using Amsterdam consensus definitions [18] performed by a single pathologist. A large cohort of preterm FGR neonates was followed for more than a decade, allowing robust subgroup analyses and adjustments for relevant clinical covariates. However, some limitations must be noted. First, the absence of a non-FGR control group restricts comparisons to normal pregnancies. Second, only liveborn preterm FGR cases were included, excluding stillbirths due to incomplete antenatal and outcome data. As a result, the impacts of severe placental lesions on in utero fetal death may have been underestimated. Third, the wide range of gestational age at delivery introduces heterogeneity, although this was partially addressed by subgroup analysis, stratified by early-onset (<32 weeks) and late-onset (≥32 weeks) FGR based on crude and adjusted OR (Supplementary Figures 1 and 2). The results of subgroup analysis suggest gestational age-dependent expression of lesion severity or a higher threshold for clinical decompensation in late gestation [26], [27], [28]. The lack of significant associations in late preterm FGR aligns with the results of prior studies, indicating milder placental pathology and better tolerance to hypoxia later in gestation [13], 20].