Fetal Cardiac Remodeling and Subclinical Myocardial Dysfunction in Pregnancies Complicated by Placenta Accreta Spectrum

Abstract

Objectives

To investigate the pattern of fetal cardiac remodeling and dysfunction associated with placenta accreta spectrum (PAS), a disorder characterized by abnormal placental angioarchitecture, and to test the hypothesis that PAS imposes a distinct hemodynamic burden on the fetal heart.

Methods

This prospective case–control study included 50 singleton pregnancies diagnosed with PAS and 50 gestational age‐matched uncomplicated pregnancies as controls. Comprehensive fetal echocardiography was performed to assess structural parameters—including right and left ventricular free wall thickness, interventricular septal thickness, and the cardiothoracic ratio—and functional parameters including the modified myocardial performance index (Mod‐MPI), isovolumetric contraction time, isovolumetric relaxation time, and ejection time. Standard fetoplacental Doppler indices were also evaluated.

Results

Standard Doppler parameters, including the umbilical artery pulsatility index, did not differ between groups (p = .240). However, significant cardiac alterations were observed in the PAS group. Compared with controls, fetuses in the PAS group demonstrated thicker right ventricular free wall (4.59 ± 0.73 versus 4.21 ± 0.68 mm; p = .008) and interventricular septum (4.72 ± 0.47 versus 4.27 ± 0.83 mm; p = .001), as well as a higher cardiothoracic ratio (0.56 ± 0.06 versus 0.49 ± 0.04; p < .001). Functionally, the Mod‐MPI was significantly increased in the PAS group (0.53 ± 0.07 versus 0.46 ± 0.08; p < .001), primarily driven by prolongation of the isovolumetric relaxation time (53.3 ± 7.74 versus 45.9 ± 10.35 ms; p = .001), consistent with impaired diastolic function.

Conclusion

PAS is associated with a distinct fetal cardiac phenotype characterized by eccentric hypertrophy and diastolic dysfunction, compatible with a chronic volume‐overload state. This remodeling occurs despite normal standard fetoplacental Doppler indices, suggesting that the low‐resistance vascular architecture of PAS imposes a unique hemodynamic load on the fetal heart. Fetal echocardiography appears to be a sensitive tool for detecting this subclinical cardiac involvement, which may have implications for fetal programming and long‐term cardiovascular health.

Abbreviations

CTR

cardiothoracic ratio: ratio between cardiac size and thoracic size used to assess cardiomegaly

ET

ejection time: duration of ventricular blood ejection into the aorta

IVCT

isovolumetric contraction time: the interval between mitral valve closure and aortic valve opening, reflecting early systolic myocardial function

IVRT

isovolumetric relaxation time: the interval between aortic valve closure and mitral valve opening, reflecting myocardial relaxation and diastolic function

IVS

interventricular septal thickness: thickness of the septum separating the right and left ventricles

LVFW

left ventricular free wall thickness: measurement of the left ventricular myocardial wall

Mod‐MPI

modified myocardial performance index: a Doppler‐derived parameter reflecting global cardiac function combining systolic and diastolic time intervals

PAS

placenta accreta spectrum

RVFW

right ventricular free wall thickness: reflects myocardial wall thickness and may indicate ventricular hypertrophy when increased

The placenta and the fetal heart develop in a coordinated and interdependent manner, forming a fundamental basis for a healthy pregnancy. The placenta functions as the primary site of nutrient and gas exchange and exerts a major influence on fetal hemodynamics. ¹ , ² In late gestation, it receives nearly 40% of the combined fetal cardiac output and provides a low‐resistance vascular pathway that is essential for normal cardiac growth and function. ³ The vascular impedance of the placental bed directly affects the afterload of the fetal right ventricle, whereas umbilical venous return determines the preload of the left ventricle. Accordingly, deviations from normal placental development may alter cardiac loading conditions and lead to adaptive or maladaptive remodeling of the fetal heart. ² , ⁴

Fetal cardiac remodeling refers to structural and functional adaptations of the fetal heart that occur in response to sustained alterations in loading conditions and has been documented across a range of obstetric disorders. Conditions characterized by increased placental vascular resistance, such as fetal growth restriction (FGR) and preeclampsia, typically create a pressure‐overload environment that promotes a more globular ventricular shape and concentric hypertrophy. ⁵ , ⁶ These structural changes are frequently accompanied by subclinical myocardial dysfunction, which can be identified using sensitive echocardiographic parameters, particularly the modified myocardial performance index (Mod‐MPI), a validated indicator of global systolic and diastolic function. ⁷ An increased Mod‐MPI, most commonly associated with prolongation of the isovolumetric relaxation time (IVRT), reflects early impairment in myocardial relaxation and increased ventricular stiffness. ⁸

Placenta accreta spectrum (PAS) refers to a group of abnormal placental implantation disorders associated with defective decidualization at the site of a previous uterine scar, resulting in abnormal adherence of chorionic villi to the myometrium. ⁹ The pathophysiology of PAS is characterized by marked neovascularization and the formation of tortuous, dilated vascular channels and lacunae within the placenta and at the utero‐placental interface. ¹⁰ Unlike the high‐resistance circulation observed in FGR, this abnormal vascular architecture may create a low‐resistance, high‐flow system with substantial arteriovenous shunting. Such a hemodynamic milieu could impose a chronic volume‐overload state on the fetal heart, a condition known to promote eccentric hypertrophy and ventricular dilation. ⁵

Although abnormal placentation in the form of placenta previa has been associated with fetal cardiac dysfunction, ¹¹ the effects of the distinctive vascular abnormalities seen in PAS on fetal cardiac structure and function remain poorly defined. It is therefore hypothesized that the abnormal vascular architecture in PAS creates a low‐resistance, high‐flow placental circulation that may impose a chronic volume‐overload state on the fetal heart, leading to detectable cardiac remodeling and subclinical myocardial dysfunction. Clarifying these potential adaptations is important, as fetal cardiac programming may have long‐term consequences for cardiovascular health after birth. Therefore, the aim of this study was to test the hypothesis that pregnancies complicated by PAS exhibit detectable fetal cardiac remodeling and subclinical myocardial dysfunction compared with uncomplicated pregnancies. To evaluate this hypothesis, we assessed a comprehensive set of fetal echocardiographic parameters, including right ventricular free wall thickness (RVFW), left ventricular free wall thickness (LVFW), interventricular septal thickness (IVS), the cardiothoracic ratio (CTR), and the Mod‐MPI.

Materials and Methods

This prospective case–control study was conducted between January 2025 and July 2025 in the perinatology outpatient clinics and inpatient units of Başakşehir Çam and Sakura City Hospital, a tertiary referral center. This study was conducted and reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines for observational studies. The study protocol was approved by the Ethics Committee of Başakşehir Çam and Sakura City Hospital Scientific Research Ethics Committee No. 1 IRB number: 2024‐KAEK‐11/213 on 06/11/2024. and written informed consent was obtained from all participants before enrollment. All procedures were carried out in accordance with the principles of the Declaration of Helsinki.

The study population consisted of 2 groups. The case group included 50 singleton pregnancies in which PAS was identified antenatally based on characteristic transvaginal and transabdominal ultrasound findings, with definitive diagnosis established intraoperatively at delivery and confirmed when available by histopathological examination. PAS was defined by the presence of a placenta located in the lower uterine segment completely covering the internal cervical os, together with at least 3 of the following sonographic markers, as previously described ¹² , ¹³ :

• Loss of the clear zone (absence of the hypoechoic myometrium–bladder interface).

• Myometrial thickness less than 1 mm.

• Multiple, irregular placental lacunae with turbulent flow.

• Increased vascularity at the bladder–placental interface (subplacental hypervascularity).

• Sonographic evidence of placental tissue extending into the uterine serosa, bladder, or parametrium

The control group consisted of 50 prospectively enrolled, uncomplicated singleton pregnancies matched to the PAS group by gestational age. The control group consisted of pregnancies with normally located placentas on antenatal ultrasound, with no sonographic evidence of placental abnormalities and no intraoperative evidence of abnormal placentation at delivery. Exclusion criteria for both groups included multiple gestation, pregnancies with known fetal chromosomal or major structural anomalies, maternal pre‐existing chronic diseases (such as hypertension or diabetes mellitus), and inadequate visualization during ultrasonographic examination.

Delivery management was performed according to institutional protocols. In our institution, PAS cases were generally managed with a combined anesthesia approach, in which regional (spinal) anesthesia was maintained until fetal delivery, after which general anesthesia was initiated when required for the continuation of surgery. All deliveries were performed in an operating room with a multidisciplinary team.

All ultrasonographic examinations were performed by experienced maternal–fetal medicine specialists (B.B., H.P., K.K.B., V.A., and F.E.) using the Arietta 850 ultrasound system (Hitachi Medical Corporation, Tokyo, Japan), equipped with a 3.5‐MHz transducer and advanced Doppler capabilities, with settings optimized for fetal echocardiographic assessment. Measurement reproducibility was evaluated through inter‐observer and intra‐observer reliability analyses, and intraclass correlation coefficients (ICC) exceeded 0.90, indicating excellent reliability for clinical use (ICC <0.50: poor; 0.50–0.75: moderate; 0.75–0.90: good; >0.90: excellent).

A detailed fetal echocardiographic evaluation was performed in accordance with standard guidelines to assess cardiac parameters that may be affected by PAS. ¹⁴ The examination included assessment of cardiac anatomy, ventricular function, and ductus arteriosus blood flow, as well as the systematic evaluation of the presence and severity of tricuspid regurgitation. RVFW, LVFW, and IVS were measured in the lateral 4‐chamber view. An IVS >5 mm was considered the diagnostic threshold for hypertrophic cardiomyopathy. Cardiac and thoracic circumferences were obtained using the trace method in an apical 4‐chamber view, and the CTR was calculated as cardiac area/thoracic area (mm²/mm²). A value <0.35 was accepted as the normal reference range for the CTR. ¹⁵

To evaluate both systolic and diastolic function, the Mod‐MPI, as described by Hernandez‐Andrade et al, was used. ¹⁶ For this measurement, an apical 4‐chamber view was obtained, and a pulsed Doppler sample volume was placed immediately below the aortic valve and above the mitral valve. After visualizing at least 3 consecutive uniform Doppler waveforms, the tracing with clearly identifiable valve clicks, stable baseline, and absence of fetal movement or breathing artifacts was selected for measurement. The Mod‐MPI was calculated using isovolumetric contraction time (IVCT), IVRT, and ejection time (ET) according to the formula (IVCT + IVRT)/ET (Figure 1). ¹⁷ This index provides an objective and comprehensive assessment of global cardiac performance. In the reference study of 557 pregnant women by Hernandez‐Andrade et al, mean Mod‐MPI values were 0.35 in mid‐pregnancy and 0.37 in late pregnancy, reflecting a slight increase with advancing gestation. ¹⁶

Figure 1.

To assess the Mod‐MPI, an apical 4‐chamber view of the heart was acquired, and a pulse Doppler sample gate was placed immediately beneath the aorta and above the mitral valve. Doppler parameters obtained from this pulse wave Doppler method, including isovolumetric contraction time (IVCT), isovolumetric relaxation time (IVRT), and ejection time (ET), were used to calculate Mod‐MPI using the formula (IVCT+IVRT)/ET.

Doppler parameters, including the pulsatility index (PI) of the umbilical artery, middle cerebral artery, and bilateral uterine arteries, were obtained using standardized techniques and in accordance with international guidelines. ¹⁸ Measurements were performed during periods without fetal movement or breathing, using a sampling volume positioned within the vessel lumen and an insonation angle <30°. For each vessel, at least 3 consecutive, uniform waveforms were recorded, and Doppler indices were calculated automatically by the system.

All echocardiographic and Doppler measurements were obtained at the time of admission after the diagnosis of PAS. Demographic characteristics, perinatal outcomes, Doppler indices, and detailed cardiac parameters—including structural measurements, Mod‐MPI, and tricuspid regurgitation—were compared between the PAS and normal groups.

Statistical Analyses

All statistical analyses were conducted using IBM SPSS Version 24.0 (Chicago, USA). Initially, the distribution of continuous data was examined for normality using histogram plots, skewness and kurtosis values and performing Kolmogorov–Smirnov and Shapiro–Wilk tests. While continuous variables that were normally distributed were expressed as mean ± standard deviation and compared using student's t‐test, variables that were not normally distributed were expressed as median (minimum–maximum) and compared using the Mann–Whitney U test. Categorical variables were presented as percentage and count and were compared using the chi‐square test. A p‐value of <.05 was considered statistically significant.

Before initiating the study, a power analysis was performed using G*Power Version 3.1.9.6 to calculate the required sample size. And the sample size was estimated based on the study published by Dogru et al which yielded a large effect size. When we assumed a large effect size (Cohen's d = 0.92) for a student's t‐test with a power of 0.90, alpha level of 0.05, the estimated total sample size was 74 patients (37 per group).

Results

A total of 100 pregnant women were recruited for the study, and the study population was subsequently divided into 2 groups, as PAS and the control group. The PAS group consisted of 50 pregnant women in whom PAS was identified antenatally, whereas the control group comprised 50 women with uncomplicated singleton and without PAS pregnancies.

Demographic characteristics of the patients were presented in Table 1. Demographic features of the study population showed largely homogeneous distribution. The mean maternal age of patients in the PAS and control groups was 33.2 (±4.55) and 29.4 (±4.23), respectively with a statistically significant difference between them (p = .001). In addition, no statistically difference was observed in terms of BMI among the groups (28.6 ± 3.85, 29.2 ± 4.21, p = .447). Since history of cesarean section numbers are higher in PAS group, statistically significant differences were observed in terms of parity (p = .001), gravida (p = .001) and number of previous cesarean sections (p = .001). The mean gestational age at diagnosis in PAS and control groups was 33.1 (30.3–33.5) weeks and 32 (30.2–33.4) weeks and comparison of this parameter revealed no significant difference (0.549). Since the control group consisted patients who did not require blood products (erythrocyte suspension, fresh frozen plasma), a statistically significant difference was found between PAS and control groups (p = .001).

Table 1.

The Demographic Characteristics and the Perinatal Outcomes of the Study Groups

	PAS Group (n = 50)	Control Group (n = 50)	Total (n = 100)	p Values a
Age (years) b	33.2 (±4.55)	29.4 (±4.23)	31.3 (±4.77)	.001
Gravida (n) c	3 (2–9)	2 (1–6)	3 (1–9)	.001
Parity (n) c	2 (1–7)	1 (0–4)	2 (0–7)	.001
Number of previous CS c	2 (0–4)	0 (0–2)	1 (0–4)	.001
BMI (kg/m2) b	28.6 (±3.85)	29.2 (4.21)	28.0 (±4.03)	.447
GA at diagnosis c	33.1 (30.3–33.5)	32 (30.2–33.4)	32.4 (±30.2–33.5)	.054
GA at delivery (weeks) c	34.5 (33.8–36.7)	39.1 (33.4–40.3)	36.6 (33.8–40.3)	.001
Birth weight (g) b	2565 (±328)	3330 (±325)	2948 (±503)	.001
ES transfusion c	2 (0–7)	—	2 (0–7)	.001
Fresh frozen plasma c	1 (0–4)	—	1 (0–4)	.001
APGAR 1. minute c	7 (3–9)	8 (7–9)	7 (3–9)	.001
APGAR 5. minute c	8 (6–9)	9 (8–10)	8 (6–10)	.001
Gender d
Female	27 (54%)	24 (48%)	51 (49%)	.548
Male	23 (46%)	26 (52%)	49 (49%)

Note: Bold values indicate statistically significant differences between groups (p < 0.005).

PAS, placenta accreta spectrum; BMI, body mass index; GA, gestational age; ES, erythrocyte suspension; C/S, cesarean section.

^a Level of significance p < .05.

^b Continuous variables distributed normally are expressed as mean and standard deviation and are compared using student's t‐test.

^c Data that were not normally distributed are expressed as median (minimum–maximum) and were compared using Mann–Whitney U test.

^d Categorical variables were presented as percentage (%) and count (n) and were compared using the chi‐square test.

Since patients in the PAS group were delivered at an earlier gestational age, the median gestational age at delivery was significantly lower in the PAS group (34.5 weeks) compared to the control group (39.1 weeks, p = .001). The mean birth weight in the PAS and control groups was 2565 ± 328 g and 3330 ± 325g, respectively, and significantly lower in the PAS group compared to controls (p = .001). Consistent with these findings, APGAR scores of neonates were significantly lower in the PAS group compared with the control group (p = .001).

Comparisons of fetal Doppler analysis and cardiac parameters were shown in Table 2 and revealed no statistically significant differences between the groups in terms of pulsatility indexes of umbilical artery, middle cerebral artery, and mean uterine artery (p = .240, p = .457, p = .373). Among fetal cardiac structures, IVS (4.72 ± 0.47, 4.27 ± 0.83, p = .001) and RVFW (4.59 ± 0.73, 4.21 ± 0.68, p = .008) were found to be significantly lower in the PAS group compared with the control group, whereas LVFW was comparable between study groups (p = .053). Moreover, the CTR (trace method) in the PAS and control groups was 0.56 (±0.06) and 0.49 (±0.04), respectively, and it was significantly higher in the PAS group compared with the control group (p = .001). Similarly, mean CTR (area method) among study groups was 0.36 (±0.05) and 0.29 (±0.04), respectively, and a statistically significant difference was observed between study groups in terms of this parameter (p = .001). Both these 2 cardiac measurement parameters were found to be significantly higher in the PAS group compared with the control groups.

Table 2.

Comparison of Fetal Doppler and Cardiac Function parameters Between the PAS and Control Groups

	PAS Group (n = 50)	Control Group (n = 50)	Total (n = 100)	p Values a
UA‐PI b	0.91 (±0.17)	0.87 (±0.17)	0.89 (±0.17)	.240
MCA‐PI b	1.80 (±0.17)	1.85 (±0.45)	1.82 (±0.34)	.457
Mean uterine artery PI b	0.84 (±0.13)	0.87 (±0.15)	0.85 (±0.14)	.373
Mod MPI b	0.53 (±0.07)	0.46 (±0.08)	0.49 (±0.08)	.001
IVCT (ms) b	43.3 (±12.74)	48.5 (±9.08)	47.4 (±11.06)	.336
IVRT (ms) b	53.3 (±7.74)	45.9 (±10.35)	49.6 (±9.82)	.001
ET (ms) b	180.6 (±26.58)	178.8 (±15.07)	179.7 (±21.51)	.692
RVFW (mm) b	4.59 (±0.73)	4.21 (±0.68)	4.40 (±0.73)	.008
LVFW (mm) b	4.72 (±0.94)	4.40 (±0.66)	4.56 (±0.83)	.053
IS (mm) b	4.72 (±0.47)	4.27 (±0.83)	4.49 (±0.71)	.001
CTR area (mm2/mm2)b	0.36 (±0.05)	0.29 (±0.04)	0.33 (±0.06)	.001
CTR trace (mm/mm) b	0.56 (±0.06)	0.49 (±0.04)	0.53 (±0.06)	.001

Note: Bold values indicate statistically significant differences between groups (p < 0.005).

PAS, placenta accreta spectrum; UA‐PI, pulsatility index of umbilical artery; MCA‐PI, pulsatility index of middle cerebral artery; MPI, modified myocardial performance index; IVCT, isovolumetric contraction time; IVRT, isovolumetric relaxation time; ET, ejection time; RVFW, right ventricle free wall; LVFW, left ventricle free wall; interventricular septum; CTR, cardiothoracic ratio.

^a Level of significance p < .05.

^b Continuous variables distributed normally are expressed as mean and standard deviation and are compared using student's t test.

Among cardiac functions parameters, no significant difference was observed in terms of IVCT and ET between study groups (p = .336, p = .692); however, the IVRT was significantly higher in the PAS group (53.3 ± 7.74 versus 45.9 ± 10.35, p = .001). In addition, the mean Mod‐MPI in PAS and control groups was 0.53 (±0.07), 0.46 (±0.08), respectively, and it was significantly higher in PAS group compared with control group (0.001).

An exploratory subanalysis was performed within the PAS group according to the number of prior cesarean deliveries. The PAS cohort was stratified based on the number of previous cesarean sections, and fetal cardiac parameters were compared between these subgroups. The pregnant women among participants with a diagnosis of PAS were divided into 2 groups based on the number of previous cesarean sections: ≥3 prior CS, <3 prior CS groups. Fetal Doppler and cardiac function parameters for patients in the ≥3 prior CS, <3 prior CS groups are shown in Table 3. Comparison of these parameters indicated that there was no significant effect of the number of previous cesarean sections on the fetal Doppler and cardiac function parameters, including Mod‐MPI (p > .05). No significant relationship was observed among the number of previous cesarean sections and fetal cardiac functions and Doppler parameters.

Table 3.

Comparison of Fetal Doppler and Cardiac Function Parameters According to Number of Previous Cesarean Section in PAS Group

	≥3 Previous CS Group (n = 50)	<3 Previous CS Group (n = 50)	Total (n = 100)	p Values a
UA‐PI b	0.92 (±0.16)	0.90 (±0.18)	0.91 (±0.17)	.806
MCA‐PI b	1.83 (±0.22)	1.77 (±0.12)	1.80 (±0.17)	.457
Mean uterine artery PI b	0.83 (±0.09)	0.85 (±0.15)	0.84 (±0.13)	.133
Mod MPI b	0.52 (±0.06)	0.54 (±0.07)	0.53 (±0.07)	.072
IVCT (ms) b	44.6 (±10.45)	47.7 (±14.32)	46.3 (±12.74)	.219
IVRT (ms) b	50.6 (±7.23)	55.4 (±7.59)	53.3 (±7.74)	.056
ET (ms) b	188.7 (±19.47)	174.2 (±29.85)	180.6 (±26.58)	.089
RVFW (mm) b	4.46 (±0.58)	4.69 (±0.83)	4.59 (±0.73)	.946
LVFW (mm) b	4.45 (±0.90)	4.93 (±0.94)	4.72 (±0.94)	.227
IS (mm) b	4.55 (±0.40)	4.85 (±0.48)	4.72 (±0.47)	.128
CTR (mm2/mm2) b	0.35 (±0.05)	0.36 (0.05)	0.36 (±0.05)	.875
CTR (mm/mm) b	0.56 (±0.05)	0.56 (±0.06)	0.56 (±0.06)	.362

PAS, placenta accreta spectrum; UA‐PI, pulsatility index of umbilical artery; MCA‐PI, pulsatility index of middle cerebral artery; Mod MPI, modified myocardial performance index; IVCT, isovolumetric contraction time; IVRT, isovolumetric relaxation time; ET, ejection time; RVFW, right ventricle free wall; LVFW, left ventricle free wall; interventricular septum; CTR, cardiothoracic ratio.

^a Level of significance p < .05.

^b Continuous variables distributed normally are expressed as mean and standard deviation and are compared using student's t test.

Correlation analyses performed in the PAS group between the maternal and neonatal outcomes, requirement of blood products, and fetal Mod‐MPI are presented in Table 4. There was no statistically significant correlation between the fetal Mod‐MPI and gestational age at delivery, birth weight, transfusion of blood products (erythrocyte suspension, fresh frozen plasma) and Apgar scores of neonates (p > .005).

Table 4.

Correlation Analyses Between Fetal Mod MPI Levels and Clinical Parameters in PAS Group

Mod MPI
	r	p a
Birth weight	0.248	.083
Number of previous CS	−0.170	.238
GA at delivery	0.215	.134
Birth weight	0.248	.083
ES transfusion	0.228	.111
Fresh frozen plasma	−0.243	.089
APGAR 1 minute	0.063	.662
APGAR 5 minute	0.088	.545

PAS, placenta accreta spectrum; Mod MPI, modified myocardial performance index; ES, erythrocyte suspension; C/S, cesarean section; GA, gestational age; r, Pearson's correlation.

^a Level of significance p < .05.

As a result of our ROC analysis of fetal Mod MPI for prediction of blood‐product requirement in patients diagnosed with PAS, we did not achieve a discriminative ability (AUC = 0.612; 95% CI, 0.440–0.785) (Figure 2).

Figure 2.

The receiver operating characteristic (ROC) curve analysis of Mod MPI level for prediction of blood‐product requirement in patients diagnosed with PAS. AUC, area under curve; PAS, placenta accreta spectrum; Mod MPI, modified myocardial performance index.

Box‐plot distributions of fetal Mod‐MPI levels in the PAS and control groups are shown in Figure 3.

Figure 3.

Box‐plot distributions of fetal Mod‐MPI levels in the PAS and control groups. PAS, placenta accreta spectrum; Mod MPI, modified myocardial performance index. Box‐plot distributions of fetal Mod‐MPI levels in preeclampsia and control groups. The median is shown by the central line, boxes represent the 25th to 75th percentiles, and whiskers indicate the range (excluding outliers).

Results of our comparisons revealed that there was a significant effect of abnormal invasion of placenta on the fetal cardiac structural and functional parameters Mod‐MPI.

Discussion

This study provides novel evidence that PAS is associated with a distinct pattern of fetal cardiac remodeling characterized by biventricular hypertrophy, cardiomegaly, and subclinical diastolic dysfunction. Our principal finding is that fetuses in pregnancies complicated by PAS demonstrate significantly increased ventricular wall thickness, a larger CTR, and a higher Mod‐MPI, despite having normal conventional fetoplacental Doppler indices. This dissociation suggests that the unique vascular architecture of PAS imposes a specific hemodynamic load on the fetal heart, one that differs fundamentally from the high‐resistance placental insufficiency observed in other major obstetric disorders.

The pattern of cardiac remodeling observed in this study—namely the marked increase in RVFW and IVS together with a significantly enlarged CTR—represents the classical phenotype of eccentric hypertrophy and ventricular dilation, which develops in response to chronic volume overload. ⁵ This contrasts sharply with the concentric hypertrophy and more globular ventricular configuration typically associated with the pressure‐overload states seen in FGR and preeclampsia. ⁶ Our findings support the hypothesis that the pathophysiological environment in PAS creates a low‐resistance, high‐flow circulatory shunt. Defective decidualization and excessive trophoblast invasion in PAS give rise to extensive, tortuous vascular lacunae and arteriovenous malformations at the utero‐placental interface. ¹⁰ , ¹⁹ This abnormal vascular network functions in effect as a large arteriovenous fistula, reducing placental vascular resistance and returning an increased volume of blood to the fetal heart. The resulting augmentation in venous return chronically elevates cardiac preload, leading to ventricular dilation and hypertrophy through Frank–Starling–mediated adaptive mechanisms designed to accommodate the increased circulatory volume. ⁴

The normal umbilical artery pulsatility index (UA‐PI) observed in our PAS cohort is a key component of this hemodynamic profile. In FGR, an elevated UA‐PI reflects increased downstream placental resistance; however, the normal UA‐PI in our study effectively excludes a high‐resistance form of placental insufficiency. Instead, it supports the interpretation that the hemodynamic disturbance in PAS arises from excessive rather than restricted blood flow. This observation challenges the reliance on umbilical artery Doppler as a sole indicator of placental health in PAS and underscores the need for a more nuanced hemodynamic assessment in these pregnancies.

Beyond the structural changes, our findings also shed light on the functional consequences of this volume‐overload state. The significantly elevated Mod‐MPI, a robust indicator of global myocardial performance, ⁷ was predominantly driven by prolongation of IVRT, a sensitive marker of diastolic dysfunction. This suggests that, although the fetal heart undergoes structural adaptation to accommodate the increased volume load, this adaptation is accompanied by impaired myocardial relaxation. The underlying mechanisms are likely multifactorial. Chronic ventricular dilation and wall stress may activate molecular signaling pathways such as the PI3K–Akt cascade, promoting physiologic cardiomyocyte growth and eccentric hypertrophy. ²⁰ However, persistent overstretching can also induce alterations in the extracellular matrix and modify the expression of structural proteins such as titin, which regulate passive myocardial stiffness. ²¹ The prolonged IVRT observed in our cohort likely reflects a combination of delayed myocardial relaxation—possibly related to impaired sarcoplasmic reticular calcium reuptake—and increased passive stiffness of the hypertrophied ventricular walls, processes that are closely linked to myocardial energetic demands. ¹⁷

It is noteworthy that the increase in LVFW did not reach statistical significance, although a similar trend was observed. This finding may reflect the distinctive characteristics of the fetal circulation, in which the right ventricle is the dominant chamber and accounts for approximately two‐thirds of the combined cardiac output. ³ Because the placental circulation is primarily supplied via the umbilical arteries originating from the fetal internal iliac arteries, the hemodynamic load generated by the placental shunt is predominantly transmitted to the right ventricle. Accordingly, the right ventricle is expected to experience the greatest preload burden and to demonstrate the most pronounced hypertrophic response.

The identification of subclinical cardiac dysfunction in fetuses with PAS has important clinical implications within the framework of fetal programming and the Developmental Origins of Health and Disease (DOHaD) concept. ²² The intrauterine environment can shape long‐term cardiovascular risk, and the cardiac remodeling observed in our study likely represents more than a transient compensatory response; rather, it reflects a modification of the developmental trajectory of the fetal heart. Although eccentric hypertrophy may constitute an appropriate adaptation to volume overload in utero, it may become maladaptive after birth, when the circulation transitions from a low‐resistance placental system to a high‐resistance systemic environment. This abrupt shift imposes a pressure load on a heart remodeled primarily for volume accommodation, potentially predisposing affected neonates to transient myocardial dysfunction, challenges in postnatal circulatory adaptation, and possibly an increased vulnerability to later‐life cardiovascular morbidity.

Our study did not demonstrate a correlation between Mod‐MPI and immediate adverse perinatal outcomes. This finding is likely related to the active clinical management of PAS, in which iatrogenic preterm delivery is frequently undertaken to reduce the risk of severe maternal hemorrhage, thereby interrupting the natural course of fetal cardiovascular adaptation. Nevertheless, the absence of this association does not diminish the relevance of the cardiac findings. Instead, it indicates that Mod‐MPI may not predict the timing of delivery but rather reflects underlying fetal cardiovascular strain. In this context, fetal echocardiography—particularly the assessment of Mod‐MPI and IVRT—may serve as a valuable tool for long‐term risk stratification rather than short‐term obstetric decision‐making. Prospective longitudinal studies are needed to follow infants from PAS pregnancies, determine whether these cardiac alterations persist after birth, and clarify their implications for long‐term cardiovascular health.

In the present study, we performed a subanalysis within the PAS group according to the number of prior cesarean deliveries; however, no significant differences in fetal cardiac parameters were observed between the subgroups.

The number of prior cesarean deliveries was used only as an exploratory stratification variable and does not necessarily reflect the clinical severity of PAS. Stratification according to FIGO clinical classification would provide a more accurate assessment of disease severity and should be considered in future studies.

However, pregnancies complicated by PAS were delivered at an earlier gestational age, which may partially contribute to some of the observed neonatal differences. Therefore, the potential influence of prematurity should be taken into account when interpreting these findings. Lower APGAR scores were observed in the PAS group compared with controls. This finding may be related to the earlier gestational age at delivery commonly observed in PAS pregnancies. When interpreting the lower APGAR scores observed in the PAS group, it should be noted that, in our institution, PAS cases were typically managed with a combined anesthesia protocol in which regional anesthesia was maintained until fetal delivery and general anesthesia was initiated only after delivery. Therefore, direct fetal exposure to maternal general anesthesia before birth was limited.

The principal strengths of this study include its prospective case–control design, the use of rigorous and standardized diagnostic criteria for PAS, a comprehensive echocardiographic protocol performed by experienced operators, and an adequately powered sample size. By concentrating on a well‐defined cohort, we were able to isolate the specific cardiac effects associated with PAS.

However, several limitations should be acknowledged. First, as a single‐center study, the generalizability of the findings may be limited. Second, although our hypothesis of a volume‐overload state is strongly supported by the observed pattern of cardiac remodeling and the established pathophysiology of PAS, it is based on indirect evidence. Definitive confirmation would require direct measurements of fetal cardiac output and placental blood flow, which are technically challenging to obtain. Third, the absence of postnatal echocardiographic follow‐up precludes conclusions regarding the persistence and clinical relevance of these findings after birth. It should also be acknowledged that the absence of a comparison group consisting of pregnancies with placenta previa without PAS. Inclusion of such a group could help distinguish whether the observed fetal cardiac remodeling is related primarily to abnormal placental invasion or to placental location alone. Future studies including normal placentation, placenta previa without PAS, and PAS may help clarify this potential spectrum of fetal cardiovascular adaptation. A limitation of this study is that PAS cases were not stratified according to clinical severity (eg, FIGO grading). It is possible that the degree of placental invasion may influence the magnitude of fetal cardiac remodeling. Future studies evaluating fetal cardiac function across different PAS severity grades may help clarify this potential relationship. Furthermore, maternal smoking and other environmental exposures were not systematically recorded and therefore could not be evaluated as potential confounding factors. Finally, although groups were matched for gestational age, unavoidable differences in maternal characteristics (such as age and parity) may represent potential confounders; nevertheless, the cardiac alterations identified are most plausibly attributable to the direct hemodynamic consequences of PAS.

Conclusion

In conclusion, this study provides a detailed characterization of a distinct fetal cardiac phenotype in pregnancies complicated by PAS. Our findings suggest that PAS induces a chronic volume‐overload state that leads to significant cardiac remodeling, including eccentric hypertrophy, ventricular dilation, and subclinical diastolic dysfunction. These results enhance our understanding of the complex interaction between placental pathology and fetal cardiovascular development and demonstrate that the fetal heart is an active organ affected by the unique hemodynamic environment created by PAS. This study highlights the potential value of advanced fetal echocardiography in the surveillance of PAS pregnancies and establishes a foundation for future longitudinal research to clarify the long‐term cardiovascular consequences for children born from pregnancies affected by this severe placental disorder.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.