The time from gestation through the first year of life is a critical period during which the brain matures most rapidly (Tierney and Nelson, 2009), and insults during this critical period have been associated with enduring developmental consequences (Andersen, 2003; Matthews et al., 2018; Raschle et al., 2012). Subcortical neurons begin forming as early as 10 weeks of gestation, and white matter myelination, axonal synaptogenesis, dendritic arborization, and formation of functional connections occur during the second and third trimesters (Adams-Chapman, 2009; Dufford et al., 2021; Ouyang et al., 2019; Stiles and Jernigan, 2010). The timing of substance exposure may thus yield different neurodevelopmental signatures, with additional consequences if the exposure is associated with preterm birth (Adams-Chapman, 2009).

Various neurotransmitter systems play fundamental roles in guiding neuronal proliferation, migration, synaptogenesis, and circuit refinement during critical periods of fetal brain development. During gestation, substances may cross the placenta through diffusion and transport mechanisms, affecting key processes and leading to alterations in these neurotransmitter systems, potentially altering neural organogenesis, growth, and/or function, depending on target receptors (Eléfant et al., 2020). Disruptions to these systems can therefore have cascading effects on neural architecture. Specifically, in the developing fetus, substance exposures may be associated with decreased dopamine synthesis and release (cocaine, methamphetamine, and opioids, Boggess and Risher, 2022; Little et al., 2021; Tomášková et al., 2020),a compensatory up-regulation of dopamine D1 and D2 receptor density (cocaine and opioids, Boggess and Risher, 2022; Little et al., 2021), altered dopamine release and serotonin turnover (nicotine, Little et al., 2021), and disrupted glutamatergic and GABAergic neuron development and signaling (alcohol, cannabis, and nicotine, Little et al., 2021). For example, GABA and glutamate not only act as neurotransmitters but also as trophic factors that regulate early neuronal differentiation and cortical patterning (Egbenya et al., 2021; Ji et al., 2024). Substance exposures are also linked to changes in glucocorticoid receptor expression, inflammatory cytokine production, and HPA axis functioning (Frank et al., 2011; Franks et al., 2019; Salisbury et al., 2009). Additionally, substance exposure may be linked to physiological alterations within the pregnant individual, including alterations to placental vasculature and physiology (Ortigosa et al., 2012), that impact the delivery of oxygen and nutrients to the fetus and further affect neurodevelopment (Rees and Harding, 2004). This brief overview necessarily simplifies a highly complex and dynamic set of neurodevelopmental processes, yet highlights the potential mechanisms through which prenatal substance impacts the developing fetus. The neurobiology of prenatal substance exposure is further described in Ortigosa et al. (2012) and Ross et al. (2015).

Understanding the shared and unique neurochemical pathways of substances of abuse is essential to interpreting their effects on the developing brain. While each substance of abuse exhibits distinct pharmacologic profiles, there are shared mechanisms between neurochemical pathways (US Department of Health and Human Services, 2016). Depressants, such as alcohol, interact with neurotransmitter systems in the brain, including GABA, glutamate, and other systems to produce a depressant effect (Heilig and Egli, 2006). Opioids bind to opioid receptors in the brain, leading to dopamine release from the nucleus accumbens (Wang, 2019). Relatedly, stimulants including cocaine and methamphetamine increase the amount of dopamine and norepinephrine in the brain’s reward circuitry (Koob, 1992). Although nicotine acts on nicotinic acetylcholine receptors and cannabis interacts with cannabinoid receptors, these substances also share the ultimate effect of activating the dopamine system throughout the brain (Chayasirisobhon, 2021; Miller and Picciotto, 2016). While the targets of substances of abuse differ, altered neurotransmitter function within reward circuitry across substance exposures may contribute to overlapping substance exposure profiles.

The differential neurodevelopmental effects of prenatal substance exposure have been studied using cellular and animal models, which have detailed the pharmacologic actions, neurotransmission effects, implicated regions, and behavioral effects of various substance exposures (Eiden et al., 2023). Animal studies have reported concentration-dependent effects of substance exposures, regions impacted across substance exposures (including the basal ganglia and reward network) and concentration-dependent cytotoxicity in offspring exposed to substances prenatally (Eiden et al., 2023; Kuhn et al., 2019; Lovinger and Alvarez, 2017; Ross et al., 2015). This foundational work provides potential biological mechanisms to explain observed relations between prenatal substance exposure and neurodevelopmental outcomes in humans, using mechanistic experimental models that are not feasible in humans (Ross et al., 2015). To translate findings from animal models to clinical populations, magnetic resonance imaging (MRI) may be used as a non-invasive, high-resolution method for studying early human brain development (Ouyang et al., 2019).

Although several reviews have examined the use of MRI to study individuals exposed to substances in utero (Donald et al., 2015a; Irner, 2012; Sanjari Moghaddam et al., 2021), these often focus on brain changes observed later in childhood and adolescence, overlooking the earliest manifestations of brain disruption following prenatal substance exposure. Two reviews have focused specifically on the neonatal and infant periods, when the brain is rapidly developing and may be particularly vulnerable to disruption (Dufford et al., 2021; Pulli et al., 2018). Another extended the scope to include functional neuroimaging and electroencephalography studies from infancy to early adulthood, with an emphasis on non-alcohol substance exposure (Morie et al., 2019). These works synthesize structural and functional MRI findings in infants with a wide range of prenatal exposures, including alcohol, nicotine, illicit substances, pharmaceuticals, maternal obesity, and inflammatory conditions. While they highlight alterations in brain volume, microstructure, and functional connectivity, they offer limited insight into how these changes relate to early developmental outcomes.

Building upon these prior reviews, we provide a targeted overview of the effects of prenatal substance exposure on brain development using infant MRI research, focusing on studies performed during the neonatal and infant periods and examining how neuroimaging findings relate to immediate developmental outcomes a dimension that has received limited attention in prior work. Unlike previous reviews, which either emphasize older developmental stages (Donald et al., 2015a; Sanjari Moghaddam et al., 2021) or include a broader selection of neuroimaging modalities (e.g., EEG and fNIRS in Morie et al., 2019), our review is the first to examine findings from structural, functional, and diffusion MRI modalities specifically in infants during the earliest postnatal stages. We excluded other functional modalities such as EEG and fNIRS to maintain a consistent focus on MRI-based methodologies, which provide both high spatial resolution and multi-modal anatomical and functional insights into early brain development (Dubois et al., 2021). In narrowing our scope, we bridge a critical gap in the literature and offer a foundation for identifying neurobiological markers that could inform intervention efforts in the earliest postnatal stages.

The literature search for this review occurred on October 12th, 2024, covering papers that were published between January 1st, 2000 and October 12th, 2024. The following inclusion criteria were used to select studies: (a) be empirical and published in a scholarly, peer-reviewed journal in English; (b) include human infants from gestation to 1 year old; (c) include a group exposed to one or more of the 6 most commonly used substances of abuse during gestation (SAMHSA, 2020), including alcohol, nicotine, opioids, cocaine, methamphetamine, or cannabis; and (d) utilize an MRI modality. Treatment-related studies were included if they met inclusion criteria a-d and reported relationships between substance exposure and brain MRI findings in alignment with our goal of summarizing MRI findings associated with prenatal substance exposure. While we were interested in reported relationships between MRI signatures and developmental outcomes, developmental assessment was not an inclusion criterion as we were interested in substance use’s effects on the brain directly as well as immediate developmental outcomes. Exclusion criteria included case studies, review articles, non-English articles, articles that did not meet all inclusion criteria (age, prenatal substance exposure, and MRI), and non-human studies.

The literature search was conducted across Academic Search Premier, ERIC, MedLine, PsycArticles, PsycINFO, and PubMed by the first author (LS), with the initial search We searched using the following keywords: (“neonat*” OR “newborn” OR “infant” OR “prem*” OR “bab*”) AND (“substance” OR “drug” OR “alcohol”) AND (“MRI” OR “magnetic resonance imaging” OR “magnetic” OR “resonance” OR “imaging”). Filters were applied to limit results to peer-reviewed, empirical studies published in English, excluding case studies and review articles. The initial search yielded a total of 3,544 articles.

Following duplicate removal, the titles and abstracts of the 3,356 articles were screened for eligibility by the first author (LS), leading to the removal of 3,330 articles based on the inclusion and exclusion criteria. The remaining 26 manuscripts were assessed by one of EGR or LGS, with additional review by LS, with 4 additional articles excluded upon in-depth review. The manuscript references of the remaining 22 articles were reviewed by EGR, LGS, and/or LS, and 3 additional articles were identified for inclusion based on bibliography review. The search process led to 25 articles deemed appropriate for inclusion in this review. During manuscript preparation, all included articles were approved by EWR, LGS, and LS, with no discrepancies in article inclusion decisions between authors.

Relevant information for each study included the primary substance exposure of interest, sample sizes of the exposed and control groups, sample age range, MRI techniques and parameters, MRI regions of interest (ROIs), developmental measures, and the main reported outcomes.

The 7 studies that investigated prenatal alcohol exposure are listed in Table 1. The number of exposed infants from each study ranged from 11 to 50, all assessed within the first 7 weeks of life. All data were collected in Cape Town, South Africa as part of a large child health study. Although cohort overlap could not be precisely determined, substantial duplication across studies is likely. Collectively, data on alcohol use during pregnancy was associated with disrupted white matter maturation in cortical regions such as the superior longitudinal fasciculus and cerebellar regions (Donald et al., 2024; Taylor et al., 2015), reduced subcortical, corpus callosum, and basal ganglia volumes (Donald et al., 2015b; Jacobson et al., 2017; Warton et al., 2021), and elevated resting state functional connectivity (rsFC) in somatosensory, motor, occipital, brainstem, and subcortical networks in alcohol-exposed infants in the first weeks of life (Donald et al., 2016; Roos et al., 2021). Brain structure and microstructure changes were associated with abnormal neonatal behaviors at birth (Donald et al., 2015c) and were linked to lower intelligence at 12 months (Warton et al., 2021), with some evidence suggesting that choline supplementation may partially mitigate these effects (Warton et al., 2021).

Seven studies examined prenatal opioid exposure (see Table 2), including a total of 174 exposed infants across separate cohorts. Two of these studies reported brain imaging findings in methadone-exposed infants. All data was collected prior to 9 weeks of life. None of the included studies examined behavioral outcomes associated with brain imaging findings. Overall, these findings suggest that opioid exposure is associated with heterogenous white matter alterations across cortical regions (Monnelly et al., 2018; Walhovd et al., 2012), disrupted subcortical volumes (Merhar et al., 2021; Yuan et al., 2014), and abnormal functional connectivity in reward-related brain networks (Jiang et al., 2022; Liu et al., 2022; Radhakrishnan et al., 2021), with possible attenuation in the context of medication for opioid use disorder (MOUD) therapy (Liu et al., 2022). Importantly, no literature to our knowledge has compared infant neuroimaging findings in opioid exposure with developmental assessments.

Four studies examined the effects of prenatal methamphetamine exposure on early brain development, with study cohorts of methamphetamine exposed infants ranging from 11–36 infants (see Table 3). Three studies included overlapping cohorts. Initial imaging was conducted within the first 9 weeks of life, with some longitudinal data extending to 16 weeks of life (Chang et al., 2016). Studies included both term (>37 weeks GA) and preterm (<37 weeks GA) infants. One study compared methamphetamine exposure (with or without tobacco exposure) to tobacco exposure; this study was included in the methamphetamine category rather than the polysubstance category given the focus on methamphetamine over and above the use of tobacco, rather than on tobacco alone. Overall, prenatal methamphetamine exposure is associated with smaller subcortical volumes in the basal ganglia and hippocampus (Warton et al., 2018a) and disrupted white matter microstructure in commissural, association, and projection regions (Chang et al., 2016; Warton et al., 2020; Warton et al., 2018b). Exposure was associated with delayed active muscle tone development at birth which normalized in early infancy (Chang et al., 2016). One longitudinal study reported sex-specific findings, including persistently disrupted microstructure in the anterior corona radiata in methamphetamine exposed females, and disrupted microstructure in the superior and posterior corona radiata in methamphetamine exposed males that normalized by 3 months (Chang et al., 2016).

Two studies within the same cohort ranging from 73 to 88 exposed infants (with likely overlap between study cohorts) examined relationships between prenatal cocaine exposure and brain development in infants less than 6 weeks of age (see Table 4). Together, these findings suggest that prenatal cocaine exposure is associated with both decreases in cortical gray matter volumes (Grewen et al., 2014) and altered functional connectivity between the thalamus and cortical regions, with early evidence of cognitive and motor developmental delays in exposed infants (Salzwedel et al., 2016). Altered thalamo-cortical connectivity was associated with lower cognitive, fine motor, and composite motor scores at 3 months in exposed infants, suggesting a relationship between neonatal functional connectivity and development in exposed infants (Salzwedel et al., 2016).

Two studies examined the effects of prenatal nicotine exposure on brain development in a total of 93 preterm infants, all born before 32 weeks GA. No included studies assessed behavioral outcomes in relation to imaging findings (see Table 5). Overall, these studies indicate that prenatal nicotine exposure is linked to smaller frontal and cerebellar brain volumes (Ekblad et al., 2010) and diffuse white matter abnormalities (Mahabee-Gittens et al., 2023) in preterm infants. However, there is a lack of data on term-born infants, non-tobacco nicotine exposures, and associations with developmental outcomes.

Only 1 study reported associations between cannabis exposure and brain development in a sample of 43 exposed term (> 36 weeks GA) infants (see Table 6). This work revealed cannabis-associated reductions in caudate and insular connectivity with the cerebellum, occipital regions, and fusiform regions in exposed infants (Grewen et al., 2015). No work to our knowledge has examined relationships between brain imaging findings, such as those reported in Grewen et al. (2015), and developmental outcomes in infancy.

Two studies within the same cohort of term and preterm infants examined the effects of polysubstance exposure on infant brain development, with study cohorts ranging from 73–75 exposed infants with likely overlap between study samples (see Table 7). Together, these findings suggest that polysubstance exposure during gestation contributes to widespread disruptions in functional connectivity (Salzwedel et al., 2020; Salzwedel et al., 2015) and may increase risk for early developmental delays, particularly in language and motor domains (Salzwedel et al., 2020).

Current evidence highlights substance-specific differences in early brain development between infants with and without prenatal substance exposure. This review examined neuroimaging differences associated with prenatal exposure and explored how these imaging signatures relate to developmental outcomes, where such associations were reported. Across studies, we identified substance-related associations with brain structure and function, including volumetric alterations, microstructural differences, and variations in rsfMRI. These findings spanned cortical and subcortical gray and white matter regions, including fronto-limbic and reward networks, motor regions including the cerebellum and brainstem, hippocampus, thalamus, basal ganglia, and white matter relay tracts.

Research points to differences between substance-exposed infants and non-exposed infants in regional volumes measured using structural MRI, microstructural markers of brain maturity as measured by diffusion MRI techniques, and regional and global connectivity patterns studied using rsfMRI techniques. Figure 1 summarizes regions implicated by neuroimaging data across substance exposures.

Evidence from the included studies suggests both overlapping and distinct patterns of brain alterations associated with prenatal exposure to different substances. Several regions appear consistently vulnerable across substances, while others show exposure-specific associations (Figure 1).

Across substances, including opioids, alcohol, methamphetamine, cannabis, and polysubstance use, exposed infants consistently exhibit alterations in structures within the reward and limbic systems, including the amygdala, nucleus accumbens, hippocampus, basal ganglia, and insula that are associated with these exposures. Given the widespread effects of prenatal substance exposures on these regions, reward and limbic differences may represent a shared neural phenotype of prenatal substance exposure. The involvement of these networks in infants exposed to substances parallels alterations within the reward system reported among adolescents with substance use disorders (Salmanzadeh et al., 2020; Squeglia and Gray, 2016). Early disruptions in these circuits may compromise affective processing and reward sensitivity, possibly mediated by perturbations in dopamine signaling due to substance exposure (Sesack and Grace, 2010). Ultimately, reward circuitry changes evident in infancy could contribute to heightened vulnerability for later-life substance use (Dodge et al., 2019), although this risk likely emerges from a complex interplay of genetic predisposition, neurobiological consequences of exposure, and broader environmental and social determinants (Abu and Roy, 2021; Nygaard et al., 2017; Oei, 2018).

Further, differences in primary sensory and motor relay centers, such as the thalamus, internal and external capsules, and corticospinal tract, were reported in infants with prenatal exposure to opioids, alcohol, and methamphetamine. These findings are consistent with adult studies linking substance use with alterations in similar relay structures (Huang et al., 2018; Pando-Naude et al., 2021) and suggest a vulnerability of sensorimotor pathways, potentially driven by disruptions in excitatory and inhibitory (glutamate and GABAergic) signaling which are affected across substance exposures and are critical to early thalamocortical and corticospinal development (Luján et al., 2005; Nishimaru and Kakizaki, 2009; Amitai, 2001). Given the role of these tracts in motor coordination and sensorimotor integration, disruptions may underlie the motor impairments observed in substance-exposed infants and children (Salzwedel et al., 2016; Willford et al., 2010).

Additionally, cortical disruptions, particularly within the frontal and prefrontal cortex were observed in infants exposed to opioids, cocaine, cannabis, nicotine, or polysubstance use (Ekblad et al., 2010; Grewen et al., 2014, 2015; Radhakrishnan et al., 2021; Salzwedel et al., 2020; Salzwedel et al., 2015, 2016). These regions, particularly within the prefrontal cortex, have frequently been implicated in adults and adolescents with substance use disorders and have been associated with cognitive and executive function impairments (Goldstein and Volkow, 2011; Squeglia et al., 2009). Dysfunction in these regions may contribute to the cognitive findings observed in infancy (Donald et al., 2015b; Salzwedel et al., 2020; Salzwedel et al., 2016). This may be driven by disruptions to synaptic pruning, myelination, or neurochemical systems such as dopamine and GABA, which shape early cortical development (Ojeda and Ávila, 2019).

Substance-specific patterns also emerged. For example, cerebellar differences were most consistently reported in infants exposed to alcohol, nicotine, and cannabis, suggesting selective vulnerability of cerebellar networks to these substances. This substance-specific pattern may be related to the high concentration of GABA (target for alcohol), cholinergic and nicotinic (targets for nicotine), and cannabinoid (target for cannabis) receptors in the cerebellum (Hauser et al., 2003; Hsiao et al., 1999; Takahashi and Linden, 2000), although future work in large samples is needed to further assess the effects of additional substance exposures on cerebellar development. Given the cerebellum’s role in sensorimotor coordination and emerging links to cognitive function, its alteration may underlie both motor and attentional deficits observed in exposed populations (Konczak and Timmann, 2007; Lyu et al., 2025).

While not the primary focus of this review, regions implicated in human infant neuroimaging studies are well-supported by preclinical literature on prenatal substance exposure (Ross et al., 2015). Animal studies have found that prenatal stimulants may affect dopaminergic system development (Thompson et al., 2009), while prenatal cocaine exposure may alter basal ganglia, hippocampus, amygdala, and cortical development (Dow-Edwards et al., 1990). Alcohol exposure in utero has been linked to abnormalities in the neocortex, hippocampus, cerebellum, and other regions (Guerri, 2002). Additionally, murine models have implicated the hippocampus and basal ganglia (opioids), hippocampus and cortex (nicotine), and the cerebellum, thalamus, hypothalamus, and hippocampus (cannabis) in substance-related developmental disruption (Benevenuto et al., 2022; Byrnes and Vassoler, 2018; Roy et al., 2002). This overlap between human neuroimaging and preclinical data provides a valuable foundation for future mechanistic studies.

In sum, the current literature suggests a core set of regions, including the limbic and reward systems, white matter relay tracts, and the frontal cortex, that are commonly affected across exposure types, possibly reflecting shared neurodevelopmental mechanisms (e.g., dysregulation of dopamine, glutamate, and GABA systems). At the same time, certain regions such as the cerebellum appear to show more substance-specific vulnerability, underscoring the importance of both common and unique pathways through which prenatal exposures impact early brain development. Nonetheless, results are preliminary, within small samples with significant cohort overlap between studies, so additional substance-specific effects and effects across substance exposures may be revealed with future work. Implicated regions aligned with both adolescent studies and animal models of prenatal substance exposure. Further whole-brain analyses examining effects of prenatal substance exposure may reveal additional relationships across the brain.

Prenatal substance exposure has been linked to a range of developmental outcomes in infancy, but the existing neuroimaging literature provides limited insight into the neurobiological mechanisms of these associations. Across alcohol and methamphetamine exposures, neuroimaging findings were associated with impairments in central and peripheral neurological function in the neonatal period. In substance exposed infants, neuroimaging differences relative to healthy controls were associated with poorer cognitive and fine motor skills (cocaine exposure) and worse cognitive and language skills (polysubstance exposure) on the BSID-III at 3 months of age. These associations provide early insight into how substance-related alterations in brain structure and function may translate into long-term developmental challenges. Nonetheless, while a number of included studies reported abnormal developmental findings at birth in substance exposed infants, one study reported normalization of these findings by 3–4 months of age, highlighting the need for longitudinal follow-up of developmental findings (Chang et al., 2016). Moreover, by design, few included studies evaluated development beyond the neonatal period, making it challenging to contextualize the long-term functional relevance of neuroimaging findings.

Thus, it remains unclear whether these outcomes reflect a delay in developmental progression, permanent deficits in specific neural systems, or a combination of both mechanisms. Such ambiguity underscores the need for longitudinal studies to track developmental trajectories over time in exposed infants. Importantly, protective factors, including supportive caregiving, parental mental health, and enriched home environments, have been shown to moderate adverse outcomes in adolescence and may serve as critical buffers against early brain vulnerability (Bada et al., 2012; Motz et al., 2011). Future neuroimaging studies should examine how these protective factors influence the relationship between prenatal substance exposure and brain development.

While not the main focus of this work, two studies explored the impact of medications for substance use. One study found normalized regional brain volumes in the thalamus, basal ganglia, and corpus callosum in pregnant people with alcohol exposure treated with high-dose choline supplementation during pregnancy, which related to better recognition memory at 12 months, compared with non-treated alcohol exposed infants (Warton et al., 2021). Another study found a reduction in opioid-associated changes in rsFC in infants exposed to methadone therapy relative to infants exposed to opioids without treatment (Liu et al., 2022; Radhakrishnan et al., 2021). These studies provide promising evidence for treatment-related effects on infant brain development, albeit in small sample sizes. Future work should continue to evaluate treatment-related effects of medications for substance use in pregnancy on infant brain and developmental outcomes.

One overarching limitation is the small sample sizes across studies, with exposed infant cohorts ranging from 10 to 88 participants. Additionally, most studies excluded occasional users but did not consistently address critical variables such as dosage or timing of substance exposure which are likely to influence outcomes and clinical translation of findings. A tightly orchestrated sequence of neurodevelopmental events unfolds throughout gestation, with varied developmental trajectories across brain circuits and regions, so the timing of substance exposures across trimesters may have varying impacts on brain development and merits further examination with sufficiently powered analyses (Andescavage et al., 2017; Xu et al., 2022). Only 2 studies within the same cohort (Warton et al., 2020; Warton et al., 2018b) examined dose effects, and none evaluated timing of exposure. Polysubstance use presents an additional methodological challenge. Although some studies focused on or included data from polysubstance-exposed groups, most did not control for the influence of multiple exposures. Therefore, observed neural differences may reflect synergistic effects or confounding interactions between commonly co-ingested substances such as alcohol, nicotine, and cannabis. Furthermore, although some studies linked neuroimaging signatures to developmental outcomes, these associations were typically based on cross-sectional assessments conducted at the time of imaging. Few studies explored infant developmental outcomes beyond the first weeks of life. No included studies assessed whether imaging findings were associated with clinical diagnoses, such as FASD or NOWS, which could help contextualize these neuroimaging findings. Similarly, no studies included comparisons between neuroimaging findings and physiological measures such as inflammatory markers or hormonal changes, which are known to be disrupted in fetal substance exposure (Frank et al., 2011; Franks et al., 2019; Salisbury et al., 2009).

Further, the existing body of literature is limited in its control of confounding factors and co-occurring influences that could contribute to observed differences between substance-exposed infants and healthy control infants. These factors include maternal psychosocial health, use of prescription medications, physical health status, and socioeconomic conditions, which may independently impact fetal brain development and interact with substance exposure (Mravčík et al., 2020). Although studies variably excluded or controlled for preterm birth, prematurity itself is linked to altered neurodevelopment and is more common among substance-exposed neonates (Bada et al., 2005; Ortinau and Neil, 2015; Umer et al., 2023). One study found that prematurity mediated nearly one-third of the effect of tobacco exposure on brain development, highlighting its significant influence (Mahabee-Gittens et al., 2023).

Sex differences also remain underexplored. Only one study in this review reported sex-specific findings (Chang et al., 2016), despite established sex-related differences in brain development and susceptibility to prenatal insults, including substance exposure (Dipietro and Voegtline, 2015; Hines, 2010; Kaczkurkin et al., 2019; Terasaki et al., 2016). Future studies should prioritize evaluating sex-specific effects to more precisely identify vulnerable subgroups.

Finally, the predominantly cross-sectional nature of included studies limits the ability to determine whether observed neural differences represent long-term changes or delays that might resolve with early intervention. Longitudinal imaging studies will be crucial to understanding whether and how brain alterations evolve across development. While infant neuroimaging findings mirror those reported in older children and adolescents, including disruptions in cortical, cerebellar, and fronto-limbic structures, as well as in the basal ganglia, corpus callosum, amygdala, and hippocampus, postnatal experiences and social environments continue to shape neural development over time (Derauf et al., 2009; Willford et al., 2016). These dynamic interactions emphasize the need for long-term follow-up studies.

Certain limitations of this review should be noted. First, despite efforts to develop a comprehensive search strategy, including reviewing the bibliographies of the studies in this review, it is possible that some relevant studies may not have been included due to the limitations of the selected databases and search terms. Variability in how studies are indexed and the terminologies used by the authors may have contributed to the omission of pertinent literature. Additionally, to include all existing literature on imaging findings in infants with substance exposure, diffusion-weighted, structural, and functional MRI are included in this review, making direct comparisons between these studies challenging. Moreover, since the initial database search was conducted by a single author, this may introduce a limitation regarding potential selection bias.

Within modalities, different statistical approaches and techniques for delineating regions of interest were applied. For example, among the DTI analyses, tract-based spatial statistics were leveraged in two studies; four studies utilized probabilistic tractography to delineate regions of interest and one study used atlas-segmented regions of interest. Similarly, brain volumes could not be directly compared across the existing structural data, which applied T1−/T2-weighted MRI, T1-weighted MRI, T2-weighted MRI, or proton density-weighted MRI to estimate brain volumes. These methodological inconsistencies may account for some differences observed across studies.

Additionally, several studies originated from the same research groups or used overlapping samples, potentially biasing findings. Some infants were scanned beyond the neonatal window, introducing postnatal environmental influences that could confound results. Furthermore, the focus on infancy may limit the generalizability of findings to later developmental stages, although existing reviews in older populations complement this work (Irner, 2012; Sanjari Moghaddam et al., 2021).

Despite these limitations, the findings of this review point to meaningful differences in brain development associated with prenatal substance exposure, particularly in the context of polysubstance use, which warrants further study given its high prevalence (Jarlenski and Krans, 2021). Future research should continue to evaluate the distinct and combined effects of individual substances on brain development while accounting for moderating variables such as sex, gestational age, birth weight, maternal mental health, and socioeconomic status (Dufford et al., 2021; Pulli et al., 2018).

Additional investigation into affected neural circuits, especially those involved in reward processing, sensory-motor integration, cognition, and white matter connectivity, is warranted. Researchers should leverage modern diffusion-weighted models and quantitative structural techniques to further investigate affected brain regions. Longitudinal studies are especially needed to clarify how early neural alterations relate to evolving developmental trajectories and functional outcomes.

Further, examining the impact of the timing and dosage of prenatal substance exposure will be critical for disentangling its specific neurodevelopmental effects. Understanding whether pharmacological or psychosocial interventions can mitigate adverse outcomes is essential to inform clinical care and public health policy. Large, diverse, and nationally representative samples will be necessary to contextualize substance-related effects within broader social and environmental risk factors.