Historically, studies addressing the safety profile of antidepressant use in pregnancy faced significant methodological challenges, most notably the small sample sizes and inadequate adjustment for critical confounders such as maternal mental health status and family psychiatric history. Previous meta-analyses conducted nearly a decade ago suggested a modest increase in risk for children’s neurodevelopmental disorders following in utero exposure to antidepressants. However, these findings lacked robust control for parental health variables, thereby obscuring the potential interplay between genetic predisposition and environmental or pharmacological influences.

The current meta-analysis, led by a multidisciplinary team from the University of Hong Kong, employed rigorous statistical adjustments to control for maternal mental health disorders and other potential confounders, thereby isolating the effect of antidepressant exposure per se. After controlling for these factors, the initially observed small elevations in ASD and ADHD risk essentially vanished. Crucially, a similar increase in risk was also seen in offspring from fathers who were prescribed antidepressants during the mother’s pregnancy, as well as mothers who used antidepressants prior to but not during pregnancy. These findings strongly implicate shared genetic and environmental factors rather than medication exposure as the underlying contributors to observed associations.

This nuanced conclusion is of profound clinical relevance, underscoring the need for balanced, individualized risk-benefit assessments when considering antidepressant treatment for pregnant women. The study’s lead author, Dr. Wing-Chung Chang, emphasizes that untreated moderate to severe depression itself carries substantial risks, including maternal relapse, which could adversely impact fetal and child development. Consequently, discontinuing antidepressants based solely on concerns about neurodevelopmental risks may inadvertently cause greater harm, underscoring the importance of maintaining maternal mental health continuity.

Investigators pooled data from over 600,000 pregnant women with documented antidepressant exposure compared against nearly 25 million pregnancies without such exposure. In unadjusted analyses, maternal antidepressant use was associated with a 35% increased risk of ADHD and a 69% increased risk of autism in offspring, echoing earlier reports. However, advanced multivariate models accounting for confounders diminished these risk estimates substantially, often rendering them statistically insignificant. This attenuation of risk highlights the critical role of confounding variables such as maternal psychiatric diagnoses, socioeconomic status, and other familial factors that may predispose children to neurodevelopmental challenges independent of pharmacological exposure.

Intriguingly, when analyses focused exclusively on mothers diagnosed with mental health disorders, the commonly prescribed selective serotonin reuptake inhibitors (SSRIs)—which represent the frontline pharmacotherapy for depression—were found not to increase risk. By contrast, two tricyclic antidepressants, amitriptyline and nortriptyline, remained associated with a slight elevation in ADHD and autism risk. Given these medications are typically reserved for treatment-resistant or more severe depressive episodes, the researchers caution that this association may reflect the severity and chronicity of parental mental illness rather than a direct pharmacological teratogenic effect.

Another important finding was the absence of a significant dose-response relationship across the pooled studies. Both low and high dosages of antidepressants showed no differential impact on the risk of neurodevelopmental disorders, suggesting that even higher exposure levels are unlikely to pose increased danger in this domain.

From a mechanistic perspective, the study points to the complex genetic architecture shared between parents and offspring and acknowledges the multifactorial influences of the familial environment, parental caregiving behaviors, and socioeconomic conditions. Chronic parental mental illness may alter the home milieu through factors like elevated family stress, disrupted routines, or variations in parental responsiveness, all of which can influence neural development trajectories in children. This integrative, biopsychosocial framework helps to contextualize the epidemiological findings and guides holistic public health strategies.

The authors also candidly recognize the limitations inherent in the reviewed studies, including inadequate data on socioeconomic status, lifestyle factors, and perinatal variables such as low birth weight. These gaps emphasize the need for future research utilizing richer, longitudinal datasets to unravel the complex interplay of genetics, environment, medication exposure, and other perinatal influences on neurodevelopment.

Expert commentators unaffiliated with the study, such as Lisa Vitte and colleagues from the University of Rouen Normandy, have hailed this research as a significant advance in the field, affirming the protective importance of antidepressants for maternal mental health without compromising fetal neurodevelopment. Their commentary underscores the public health importance of dispelling misinformation that can lead to unwarranted medication discontinuation during pregnancy.

In conclusion, this meta-analysis marks a pivotal step in refining clinical guidance regarding antidepressant use in expectant mothers by dispelling fears of a direct causal link to neurodevelopmental disorders in children. It advocates a balanced clinical dialogue that respects the intricate genetic and environmental factors influencing mental health and neurodevelopment. As depression prevalence remains significant among pregnant populations, preserving maternal well-being with evidence-based therapy remains paramount for optimizing outcomes for both mothers and their children.

Subject of Research: People
Article Title: Maternal and paternal antidepressant use before and during pregnancy and offspring risk of neurodevelopmental disorders: a systematic review and meta-analysis
News Publication Date: 14-May-2026
Web References: http://dx.doi.org/10.1016/S2215-0366(26)00089-1
Keywords: Antidepressants, Pregnancy, Autism, Attention-deficit/hyperactivity disorder, Neurodevelopmental disorders, Maternal mental health, Meta-analysis

Valproic acid (VPA), an anticonvulsant and mood-stabilizing drug, has long been implicated in teratogenic effects when administered during pregnancy, with previous research linking prenatal VPA exposure to increased risk of autism-like behaviors in animal models. However, this new research broadens that narrative by investigating VPA’s impact within the context of gestational diabetes mellitus (GDM), a metabolic complication characterized by impaired glucose metabolism during pregnancy. The confluence of these two risk factors creates a unique prenatal environment that significantly influences brain development.

The researchers employed a murine model to simulate the combined prenatal conditions of VPA exposure and gestational diabetes, meticulously analyzing behavioral phenotypes and neural tissue characteristics in offspring. Behavioral assays revealed that pups subjected to both prenatal VPA and GDM manifested pronounced autistic-like behaviors, including deficits in social interaction and repetitive behaviors, compared to control groups or those exposed to either insult alone. This synergistic exacerbation underscores the multifactorial nature of neurodevelopmental disorders and highlights the necessity of considering comorbid maternal conditions in assessing developmental risk.

At the cellular level, the team focused on the oligodendroglial lineage, key players in the myelination process critical for proper neural circuit function. Oligodendrocytes produce myelin sheaths that insulate neuronal axons, facilitating rapid electrical impulse transmission essential for cognition and behavior. Intriguingly, offspring from the combined VPA and GDM group exhibited modest but notable disruption in oligodendrocyte maturation and consequent myelination deficits, suggesting that myelin integrity may be a vital substrate affected by prenatal insults contributing to ASD-like phenotypes.

The study’s neuropathological analyses utilized advanced immunohistochemical techniques to quantify oligodendroglial markers, revealing a subtle but statistically significant decrease in the expression of key myelination proteins such as myelin basic protein (MBP) and proteolipid protein (PLP). These disruptions present a plausible mechanistic link between environmental-metabolic perturbations in utero and the emergence of neural dysfunction observed behaviorally. The data evoke an emerging paradigm where white matter pathologies, often overshadowed by gray matter considerations, play an essential role in neurodevelopmental disorders.

Further molecular investigations examined the balance of oligodendrocyte progenitor cells (OPCs) versus mature oligodendrocytes, illustrating that prenatal exposure to VPA on the backdrop of GDM interferes with the differentiation process. This impaired maturation diminishes myelination capacity, potentially contributing to anomalous neural connectivity patterns typical in ASD brains. The research suggests an epigenetic or metabolic disruption cascade triggered by maternal conditions that perturbs normal glial development, a hypothesis requiring future mechanistic exploration.

Of particular importance is the revelation that GDM alone did not provoke substantial behavioral or myelination abnormalities; it was the combination with VPA that induced the pronounced effects. This synergy directs attention to the amplifying role of metabolic dysfunction in enhancing the vulnerability of the developing fetal brain to environmental toxins. Such findings call for the integration of maternal metabolic health screening in risk assessments for ASD and possibly tailoring pharmacological interventions accordingly.

This study also raises critical concerns regarding the clinical use of VPA during pregnancy, especially in women with underlying metabolic disorders. While VPA remains a vital therapeutic agent for certain neurological conditions, its teratogenic potential may be heightened in metabolically compromised pregnancies. The nuanced findings advocate for personalized medicine approaches during pregnancy, balancing maternal health needs with fetal neurodevelopmental risk mitigation.

From an epidemiological perspective, the increasing prevalence of GDM worldwide, parallel with rising ASD diagnosis rates, underscores the public health relevance of these findings. The intersection between metabolic disease and neurodevelopmental disorders could reflect a larger, systemic interplay between maternal physiology and fetal brain programming, warranting more comprehensive prenatal care guidelines and targeted interventions to reduce ASD incidence.

Moreover, this research offers a compelling case for the role of white matter abnormalities and oligodendroglial dysfunction in ASD pathology. Historically, autism research has prioritized neuronal and synaptic anomalies; however, this study enriches the discourse by implicating myelination deficiencies as a subtle but impactful contributor to ASD behaviors. It suggests that therapies aiming to promote oligodendrocyte maturation and myelin repair might hold promise as future treatment strategies.

In addition, the subtlety of the myelination deficits observed raises important methodological considerations about the sensitivity of current diagnostic tools and animal models in capturing the full scope of ASD-related neuropathology. Slight disruptions in oligodendrocyte maturation may escape detection in conventional assays yet significantly alter neural circuit function, emphasizing the need for refined investigative techniques focusing on glial biology.

Lastly, the research team calls for longitudinal studies to track the lasting impact of prenatal VPA and GDM exposure on cognitive and social behaviors, as well as myelination patterns across developmental milestones. Such studies are essential to elucidate the trajectory of neurodevelopmental disturbances and to identify critical windows for intervention.

In conclusion, the integration of metabolic and pharmacological prenatal insults epitomized by this study provides an innovative framework for understanding ASD etiology. By demonstrating that prenatal VPA exposure within the context of gestational diabetes exacerbates autistic-like behaviors while subtly impairing oligodendroglial maturation and myelin integrity, this research pioneers new directions in both clinical awareness and mechanistic neurobiology. As the quest to decipher autism’s origins continues, these insights reaffirm the importance of maternal health and its profound influence on the developing brain.

Subject of Research:
Prenatal impact of valproic acid exposure combined with gestational diabetes on offspring neurodevelopment and behavior.

Article Title:
Prenatal valproic acid on the basis of gestational diabetes also induces autistic behavior and disrupts myelination and oligodendroglial maturation slightly in offspring.

Article References:
Li, M., Qiao, Z., Li, J. et al. Prenatal valproic acid on the basis of gestational diabetes also induces autistic behavior and disrupts myelination and oligodendroglial maturation slightly in offspring. Transl Psychiatry 15, 271 (2025). https://doi.org/10.1038/s41398-025-03450-z

Image Credits: AI Generated

DOI: https://doi.org/10.1038/s41398-025-03450-z