Investigating the Impact of Prenatal Alcohol Exposure on Fetal Cardiac Development Using Photoacoustic Tomography

Abstract

Prenatal alcohol exposure (PAE) is a leading cause of developmental abnormalities, yet its effects on fetal cardiac development remain understudied. We employed real-time, label-free multispectral photoacoustic tomography (PAT) to noninvasively assess cardiac development in mouse fetuses exposed to chronic alcohol. Using a custom-built PAT system, fetal hearts were imaged from E12-E16 in alcohol-exposed (3 g/kg ethanol via oral gavage, n=9) and control (n=7) CD-1 mice. PAT enabled quantitative measurements of cardiac morphology, oxygen saturation (sO₂), and heart rate. Alcohol-exposed fetuses exhibited consistently lower sO₂ and greater heart rate variability, particularly at later gestational stages. While structural growth progressed in both groups, functional impairments became more pronounced with alcohol exposure. These findings suggest PAE alters fetal cardiovascular regulation despite normal anatomical development. This study highlights the utility of PAT as a high-resolution, noninvasive tool for monitoring fetal cardiac health and supports its potential application in developmental biology and prenatal diagnostics.

1. Introduction

In the United States, approximately 50% of women of reproductive age report consuming alcohol, with 1 in 10 reporting alcohol consumption within the past 30 days and 1 in 22 engaging in binge drinking [1,2]. These statistics are particularly concerning given that nearly half of all pregnancies go unrecognized until the fifth or sixth week of gestation, resulting in unintentional fetal exposure to alcohol during critical periods of organogenesis [3]. Exposure to alcohol during these early stages of embryonic development has been linked to a wide spectrum of adverse outcomes, collectively referred to as fetal alcohol spectrum disorders (FASD) [4]. These outcomes include structural anomalies, neurobehavioral deficits, and long-term cognitive and motor impairments [5]. Early diagnosis and intervention are essential for mitigating the impact of FASD, as timely support can improve developmental outcomes and reduce healthcare costs [6].

The central nervous system (CNS) has long been regarded as the primary target of alcohol’s teratogenic effects. Numerous studies have documented alcohol-induced neurodevelopmental abnormalities, including disruptions in neuronal migration, altered synaptogenesis, and impaired vascular development in the fetal brain [7]. Beyond the brain, increasing evidence shows that prenatal alcohol exposure (PAE) also adversely affects multiple fetal organ systems, including the liver, kidneys, gastrointestinal tract, endocrine, and immune systems [8,9]. Among these, the cardiovascular system has gained growing attention as an important yet underexplored target, warranting closer investigation into how alcohol alters fetal heart development. Congenital heart defects (CHDs) are the most common type of birth defect, affecting nearly 1% of live births globally [10]. These defects can range from simple structural anomalies to complex malformations that require surgical correction shortly after birth. While genetic and environmental factors contribute to CHD etiology, prenatal exposure to toxic substances such as alcohol has emerged as a potentially modifiable risk factor [11]. Clinical and epidemiological studies have reported associations between maternal alcohol use and increased rates of ventricular septal defects, conotruncal anomalies, and other cardiac malformations in newborns [10]. Moreover, autopsy-based and echocardiographic studies in animal models have shown that alcohol interferes with cardiac morphogenesis by altering cell proliferation, migration, and differentiation within the developing myocardium [12]. In addition to structural defects, functional impairments in fetal cardiovascular physiology are also suspected to result from PAE. Several preclinical models have demonstrated changes in heart rate variability, cardiac output, cardiac structural abnormalities, including changes in extracellular matrix (ECM) composition and myocardial metabolism in alcohol-exposed fetuses and neonates [13,14]. These abnormalities may stem from both direct effects of ethanol and its metabolites on cardiomyocytes and indirect effects mediated through altered placental function or disrupted fetal hemodynamics. Despite these findings, a comprehensive, longitudinal assessment of fetal cardiac development in vivo under chronic PAE conditions remains lacking. Advancing our understanding of these developmental disruptions requires imaging technologies that can non-invasively and dynamically assess fetal cardiovascular function throughout gestation.

Various imaging modalities have been employed to study fetal cardiovascular development, including magnetic resonance angiography (MRA) [15,16], ultrasound Doppler imaging [17,18], and optical coherence tomography (OCT) [19–21]. While these approaches have provided useful insights into blood flow and vascular anatomy, they are limited in their ability to comprehensively assess dynamic cardiac physiology across embryonic stages. Micro-MRA offers high-resolution anatomical imaging but requires contrast agents and is not suited for real-time or repeated longitudinal studies due to cost and accessibility constraints. Ultrasound Doppler imaging is commonly used to assess fetal hemodynamics but lacks the spatial resolution necessary to resolve fine-scale vascular and cardiac structures. OCT provides exceptional resolution for superficial vasculature but has a restricted penetration depth of only a few hundred micrometers, limiting its ability to image deeper fetal structures such as the heart. Moreover, OCT often requires invasive procedures to access embryos in utero, which may interfere with normal maternal–fetal physiology and reduce the accuracy of functional assessments. These limitations collectively hinder the ability to conduct high-resolution, longitudinal, and non-invasive evaluations of fetal cardiac development under pathological conditions such as PAE.

Photoacoustic tomography (PAT) offers a promising solution by combining optical absorption contrast—particularly from hemoglobin—with ultrasonic spatial resolution and depth penetration [22–24]. PAT can capture images of vasculature and heart sections in adult mice, achieving a spatial resolution of at least 100 μm and an imaging depth of up to 4 cm [24]. Additionally, PAT integrated with photon transport modeling facilitates quantitative assessments of physiological parameters such as oxy- and deoxy-hemoglobin concentrations, oxygen saturation, and heart rates [25]. These functional measurements are essential for tracking fetal vascular development, evaluating hemodynamic alterations, and examining the cardiovascular impacts of maternal alcohol exposure [26]. Importantly, PAT achieves these capabilities without the need for exogenous contrast agents, making it well-suited for repeated, real-time imaging of fetal development in small animal models [27].

Our group has previously demonstrated the feasibility of using a custom multispectral PAT system to image cerebral vasculature in mouse fetuses under PAE conditions [28]. Building on this experience, we now apply PAT to investigate fetal heart development across embryonic stages E12 to E16. Using a well-established mouse model of chronic alcohol exposure, we performed in vivo imaging to assess key parameters such as oxygen saturation, vessel density, and heart rate. These measurements enabled us to characterize structural and functional deviations in fetal cardiac physiology associated with PAE. This study addresses a critical gap in FASD research by providing a longitudinal view of fetal cardiac development under chronic alcohol exposure and demonstrating the utility of PAT as a label-free, high-resolution imaging platform for developmental research.

2. Materials and Methods

2.1. PAT Imaging System

A real-time multispectral photoacoustic tomography (PAT) system was used for in vivo fetal imaging, similar to prior studies [28]. The system consisted of a portable OPO laser (680–900 nm, 20 Hz repetition rate; OPOtek Inc. Phocus Mobile, Carlsbad, California) coupled to a custom fiber bundle (CeramOptec GmbH, Germany) for wide-field illumination. A 256-element linear transducer array (Japan Probe Co Ltd, Japan) of 4 MHz center frequency, 80% bandwidth acquired the photoacoustic (PA) signals, which were digitized by a high-speed DAQ system system (PhotoSound Technologies Inc., Houston, Texas) featuring a sampling rate of 40MSPS, a resolution of 12-bit, along with an adjustable gain ranging from 40 to 91 dB. The collected signal is transmitted to a computer via a USB 3.0 interface in real-time. Pregnant mice were anesthetized with isoflurane in oxygen (4% isoflurane for induction and 1–2.5% for maintenance throughout the entire imaging session) and placed supine on a custom-built animal holder that maintained body positioning and physiological stability. The abdominal area was gently shaved to improve acoustic and optical transmission. A temperature-controlled water bath (37°C) and respiratory monitoring system ensured homeostasis during imaging. A schematic of the setup and the custom animal holder is shown in Fig. 1.

Fig. 1.

Schematic of the photoacoustic imaging system and animal setup used for in vivo fetal cardiac imaging. (a) Overview of the custom-built multispectral PAT system setup. (b) Photograph of an anesthetized pregnant mouse positioned in the imaging chamber during PAT image acquisition. PC: Personal Computer; DAQ: Data Acquisition System.

To acquire comprehensive coverage of the fetal region, the transducer array was affixed to a motorized linear stage (Newmark Systems) that enabled automated vertical scanning of the maternal abdomen in 2 mm increments. At each step, photoacoustic signals were recorded over a 10-second interval at three different wavelengths (720 nm, 760 nm, and 810 nm) to allow for functional imaging. Depending on the abdominal size and fetal positioning, the complete scan duration ranged between 4 and 6 minutes. The collected raw data were subsequently reconstructed into cross-sectional photoacoustic images using a delay-and-sum algorithm. To minimize motion artifacts during imaging, animals were placed under isoflurane anesthesia and secured using a custom-designed stable animal holder to reduce gross movement. Maternal respiratory motion was corrected during image reconstruction to prevent displacement artifacts caused by breathing. In addition, fetal motion was identified by observing disruptions in the photoacoustic signal over time. When such fetal movements were detected, targeted corrections were applied to align the signal temporally and spatially. This multi-tiered motion management approach—combining physical stabilization, maternal motion correction, and fetal signal tracking—ensured high-quality reconstructions suitable for quantitative analysis.

2.2. Multispectral Imaging and sO₂ Analysis

Multispectral PAT enables functional imaging of ethanol-induced changes in fetal oxygen saturation, a critical factor in cardiac development. To obtain functional images, photoacoustic signals were captured at three wavelengths: 720, 760, and 810 nm. After recovering the absorption images at each wavelength, functional parameters, including oxy- and deoxy-hemoglobin (HbO, HbR) and sO₂, were quantitatively calculated. The PAT data were first processed using the delay-and-sum reconstruction method, and the spectral results were unmixed using least-squares spectral fitting to determine the distribution of HbO and HbR. Finally, the sO₂ map was computed. The least-squares method is described as follows:

$$\text{P}\left({\lambda }_i,\text{x},\text{y}\right)=\Phi \left(\lambda \right)\left({\epsilon }_{\text{HbR}}\left({\lambda }_i\right){\text{C}}_{\text{HbR}}\left(\text{x},\text{y}\right)+{\epsilon }_{\text{HbO}}\left({\lambda }_i\right){\text{C}}_{\text{HbO}}\left(\text{x},\text{y}\right)\right)$$ (1)

where P(λ_i, x, y) is the reconstructed PAT image at a specific wavelength λ_i, Φ(λ) is the local optical fluence, ε_HbR(λ_i) and ε_HbO(λ_i) are molar extinction coefficients of HbR and HbO, respectively. C_HbR (x, y) and C_HbO (x, y) are molar concentrations of HbR and HbO, respectively.

By solving the linear equations of multiple wavelengths, C_HbR (x, y) and C_HbO (x, y) can be determined:

$$\left[\begin{array}{c}{\text{C}}_{\text{HbR}}\left(\text{x},\text{y}\right)\\ {\text{C}}_{\text{HbO}}\left(\text{x},\text{y}\right)\end{array}\right]={\left({\epsilon }^{\text{T}}\epsilon \right)}^{-1}{\epsilon }^{\text{T}}\text{P}$$ (2)

where

$$\text{P=}\left[\begin{array}{c}\text{P}\left({\lambda }_1,\text{x},\text{y}\right)\\ ⋮\\ \text{P}\left({\lambda }_{\text{N}},\text{x},\text{y}\right)\end{array}\right];\epsilon =\left[\begin{array}{cc}{\epsilon }_{\text{HbR}}\left({\lambda }_1\right)& {\epsilon }_{\text{HbO}}\left({\lambda }_1\right)\\ ⋮& ⋮\\ {\epsilon }_{\text{HbR}}\left({\lambda }_{\text{N}}\right)& {\epsilon }_{\text{HbO}}\left({\lambda }_{\text{N}}\right)\end{array}\right]$$

2.3. Animal Preparation

All experimental protocols were approved by the Institutional Animal Care and Use Committee (IACUC) of the University of South Florida and adhered to the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Female CD-1 mice (6 weeks old, Charles River Laboratories, Wilmington, MA) were used for timed breeding. To synchronize estrous cycles, females were group-housed (5 per cage) for 10–14 days and exposed to soiled bedding collected from male cages. After synchronization, each female was paired with an individual male overnight. Mating was confirmed the following morning by the presence of a vaginal plug, designated as embryonic day 0.5 (E0.5), and body weight was recorded to track pregnancy progression.

Dams were randomly assigned to one of two experimental groups: (1) alcohol-exposed or (2) control. The alcohol group received 3 g/kg ethanol, prepared from 95% ACS-grade ethanol (Acros Organics, Fisher Scientific), diluted in distilled water to yield a 40% v/v solution. This was administered via oral gavage every 12 hours between E10 and E20 to maintain blood alcohol levels reported in the literature (~117 mg/dL) [29]. Control dams received an isocaloric maltose-dextrin solution on the same schedule.

All animals were housed under a 12-hour light/dark cycle in a temperature- and humidity-controlled facility, with ad libitum access to food and water. Longitudinal photoacoustic imaging was conducted at E12, E14, and E16 to monitor fetal cardiac development. Data from fetuses within a litter were averaged and treated as a single biological replicate.

2.4. Image Processing and Quantification

Photoacoustic data were post-processed to extract cardiac parameters. The fetal heart region was manually identified based on clearly visible heartbeats (Supplementary Materials S1), where dynamic regions exhibiting rhythmic signal fluctuations corresponding to cardiac pulsation were distinguishable due to periodic changes in photoacoustic intensity across sequential image frames. Manual segmentation was guided by anatomical context and verified across adjacent slices for consistency. Ambiguous cases were independently reviewed by two observers to ensure accuracy and minimize subjectivity. Frangi filtering was applied to enhance vessel-like structures. Binary vessel masks were generated to calculate vessel density (vessel pixel area/ROI area). Heart rate was extracted by analyzing temporal oscillations in PA signal intensity in the ROI. sO₂ maps were computed pixel-wise within the vessel mask.

2.5. Statistical Analysis

Data are reported as mean ± standard error of the mean (SEM). For each group (alcohol-exposed and control), alcohol n=9 and control n=7. Data from fetuses within a litter were averaged and treated as a single unit of analysis. Group comparisons at each timepoint were assessed using two-tailed unpaired Student’s t-tests, with significance defined as p < 0.05.

3. Results

3.1. In Vivo Visualization of Fetal Heart Morphology

Representative 2D cross-sectional PAT images of the fetal heart are shown in Fig. 2 for embryonic days E12 (a), E14 (b), and E16 (c). As development progresses, cardiac anatomical features become increasingly distinguishable, with improved boundary contrast and chamber definition. At E16, the spatial separation between the left atrium (LA), right atrium (RA), left ventricle (LV), right ventricle (RV), and aorta (Ao) was evident after image processing using vessel enhancement filters (Fig. 2d). This allowed reliable anatomical segmentation of major heart structures.

Fig. 2.

In vivo photoacoustic imaging of fetal cardiac development. Representative grayscale photoacoustic images of the fetal heart captured at (a) E12, (b) E14, and (c) E16, showing increased anatomical definition with gestational age. (d) Cardiac chambers identified after image processing, including the left atrium (LA), left ventricle (LV), aorta (Ao), right atrium (RA), and right ventricle (RV). (e) Quantification of fetal heart growth over time, showing increasing heart size between E12 and E16.

Quantitative measurements of heart size were obtained by computing the average area of the segmented heart ROI across different gestational days. A clear increasing trend in heart dimension was observed over time (Fig. 2e), reflecting normal physiological growth and cardiac remodeling during embryogenesis. The growth curve supports the feasibility of PAT for longitudinal tracking of fetal heart development noninvasively.

3.2. Mapping of Fetal Cardiac Oxygen Saturation

To assess the impact of prenatal alcohol exposure on fetal oxygenation, multispectral PAT was used to compute sO₂ within the fetal heart. A cross-sectional image of a pregnant mouse at E16 (Fig. 3a) was used to localize the heart region of interest (ROI), which was further magnified (Fig. 3b) and processed with vessel enhancement (Fig. 3c). The resulting sO₂ maps (Figs. 3d–3e) illustrate clear differences between alcohol-exposed and control fetuses. In the alcohol group, sO₂ levels were generally reduced, particularly at later gestational stages.

Fig. 3.

Assessment of fetal cardiac oxygenation using multispectral photoacoustic tomography. (a) Cross-sectional raw PAT image of a pregnant mouse at E14, with the heart ROI outlined for focused analysis. (b) Grayscale image of the reconstructed fetal heart region. (c) Frangi-filtered image highlighting vascular structures within the ROI. (d–e) Corresponding oxygen saturation (sO₂) maps of fetal hearts from alcohol-exposed and control groups, respectively. (f) Quantitative comparison of average sO₂ levels across gestational days E12, E14, and E16 between groups (alcohol: n=9, control: n=7). Error bars represent SEM.

Average sO₂ values were calculated across E12, E14, and E16 for both groups (Fig. 3f). Control fetuses exhibited stable oxygenation across time, while alcohol-exposed fetuses showed a trend toward reduced sO₂, most noticeably at E16. While variation existed across gestational stages, alcohol-exposed fetuses consistently exhibited lower oxygen saturation compared to controls, with the difference becoming more pronounced at later stages of development.

The ability to resolve such group-specific oxygenation differences through label-free, multispectral PAT highlights its utility for assessing functional fetal cardiovascular status. These findings underscore the sensitivity of PAT for detecting hemodynamic alterations induced by environmental insults such as alcohol.

3.3. Heart Rate Analysis

In addition to structural and oxygenation assessments, fetal heart rate was quantified using temporal fluctuations in photoacoustic signal intensity within the heart ROI. This noninvasive measurement enabled dynamic evaluation of cardiac function during development under both control and alcohol-exposed conditions.

Figure 4 presents a comparison of mean fetal heart rates at gestational days E12, E14, and E16 for both groups. In the control group, heart rate remained relatively stable, with a slight decrease by E16. In contrast, alcohol-exposed fetuses demonstrated greater variability in heart rate across developmental stages. At E12, heart rate in the alcohol group was slightly elevated compared to controls, but this difference was reversed by E16, where the alcohol group exhibited lower average heart rate.

Fig. 4.

Comparison of fetal heart rates between alcohol-exposed and control groups across gestational days. Bar plots illustrate average heart rate (bpm) at E12, E14, and E16 for both groups. Data represent mean ± SEM. Control: n = 7; Alcohol: n = 9.

The inclusion of heart rate analysis complements the structural and oxygenation measurements, offering a more comprehensive assessment of functional cardiac development under prenatal alcohol exposure. Together, these metrics demonstrate the feasibility of using photoacoustic tomography to monitor multiple physiological parameters relevant to fetal health in a label-free and noninvasive manner.

3.4. Photoacoustic Trends Across Developmental Timepoints

The combined evaluation of structural growth, oxygenation, and heart rate across embryonic days E12 to E16 revealed progressive alterations in fetal cardiac development associated with prenatal alcohol exposure. In early gestation, the differences between control and alcohol-exposed groups were minimal; however, by E14 and E16, alcohol-exposed fetuses began to exhibit more pronounced impairments, particularly in terms of oxygen saturation and heart rate.

This progressive divergence suggests that alcohol’s effects may accumulate over time, interfering with key developmental milestones during mid-to-late gestation when rapid cardiovascular growth and functional maturation occur. Structural imaging indicated clear and consistent heart growth in both groups, but functional parameters like sO₂ and heart rate were more variable in the alcohol group, potentially indicating subtle disruptions in fetal hemodynamics.

Together, these findings underscore the value of photoacoustic tomography as a sensitive and noninvasive tool for detecting and tracking alcohol-induced disruptions in fetal cardiac development. By integrating structural and functional information over time, PAT provides a comprehensive picture of how environmental insults, such as alcohol exposure, shape cardiovascular health during embryogenesis.

4. Discussion and Conclusions

This study provides new insights into the effects of PAE on fetal heart development by utilizing real-time PAT to evaluate structural and functional cardiac parameters in utero. PAT enabled the longitudinal tracking of fetal heart growth, oxygen saturation, and heart rate from E12 to E16 in both alcohol-exposed and control groups. Our findings revealed developmental deviations in cardiac function and oxygenation associated with chronic alcohol exposure, demonstrating the potential of PAT to assess subtle, yet progressive, impairments in fetal cardiovascular health.

The ability to image fetal cardiac structures in vivo at different gestational stages allowed us to detect anatomical growth patterns and monitor cardiac chamber differentiation. Quantitative analyses showed clear evidence of heart enlargement with advancing gestation, consistent with expected physiological growth. Importantly, while structural development appeared largely intact in alcohol-exposed fetuses, functional disruptions were more pronounced. Multispectral imaging revealed a consistent trend of reduced sO₂ in alcohol-exposed fetuses compared to controls, particularly at E16, a timepoint associated with rapid cardiovascular remodeling. This reduction in oxygenation may reflect altered placental exchange or impaired fetal hemodynamics. In parallel, heart rate analysis showed variability in alcohol-exposed fetuses, with lower mean values compared to controls at later gestational stages. These subtle yet consistent changes suggest that PAE interferes with normal cardiovascular regulation and oxygen delivery, even in the absence of gross structural anomalies. Collectively, the longitudinal patterns observed in this study suggest a cumulative impact of alcohol exposure over time. Early-stage differences were modest but became more pronounced by E14 and E16, implying that PAE may disrupt critical windows of cardiovascular maturation. This temporal dimension is especially important for understanding how teratogenic insults evolve and manifest across gestation.

This work also highlights the strengths of PAT as a noninvasive and label-free imaging platform that combines optical contrast with acoustic resolution. Unlike conventional techniques such as MRI or Doppler ultrasound, PAT offers superior spatial resolution, depth penetration, and the ability to assess hemoglobin oxygenation without contrast agents. These features make it particularly suited for dynamic, longitudinal studies of small animal models. While the current study was limited to a 3 g/kg ethanol exposure model and a small sample size, the trends reported here support the feasibility and value of PAT for developmental teratology research. Future studies with expanded timepoints, dose ranges, and functional endpoints (e.g., cardiac output or mitochondrial activity) will further refine our understanding of alcohol’s impact on fetal cardiac health.

In conclusion, we demonstrate that PAT is a powerful imaging modality for detecting early-stage functional cardiac impairments induced by prenatal alcohol exposure. The observed alterations in heart rate and oxygen saturation provide a window into the physiological consequences of in utero ethanol exposure and underscore the need for continued research into noninvasive diagnostics for fetal cardiovascular health.

Supplementary Material

Video S2

Video S2: This video shows PAT imaging of a pregnant mouse from the alcohol group at embryonic day 16 (E16). As in Video S1, the heartbeat is clearly visible for blood vessel structure and oxygenation.

Video S1

Video S1: This video captures PAT imaging of a pregnant mouse from the control group at embryonic day 14 (E14). The heartbeat is clearly visible for blood vessel structure and oxygenation.

Additional supporting information is available in the online version of this article at the publisher’s website or from the author.

Abbreviations

PAE

Prenatal Alcohol Exposure

PAT

Photoacoustic Tomography

FASD

Fetal Alcohol Spectrum Disorders

CHD

Congenital Heart Defect

ROI

Region of Interest

MRA

Magnetic Resonance Angiography

OCT

Optical Coherence Tomography

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.