How Melatonin and Artificial Light May Affect Mother and Baby Health During Pregnancy

The main questions to be answered in this review were as follows: (i) What is the role of melatonin in maternal health and fetal development during pregnancy? (ii) How does exposure to artificial light during pregnancy influence maternal and fetal melatonin levels? (iii) What changes in maternal–fetal health may occur due to reduced melatonin levels? (iv) Is there an association between exposure to artificial light and adverse maternal–fetal health outcomes? (v) What strategies can minimize the adverse effects of artificial light on maternal–fetal health?

The literature search on the impacts of exposure to artificial light during pregnancy was conducted in the PubMed/MEDLINE and Google Scholar databases in two stages: (i) direct search for articles in the selected databases and (ii) indirect search in the reference lists of the included studies and websites, using the snowballing strategy.

The following search terms (“blue light” OR “ALAN” OR “melatonin”) AND (“pregnancy”) were applied to Title/Abstract, with the following filters: (i) clinical trials, randomized trials, cohort studies, case–control studies, and case reports; (ii) publications between 2015 and 2025; (iii) studies involving human subjects and nonhuman animal models; (iv) female population; (v) languages: English, Portuguese, or Spanish.

The following were defined as inclusion criteria: (i) original studies; (ii) studies that investigated exposure to artificial light during pregnancy in relation to melatonin levels and the repercussions for maternal health and fetal development, or that examined melatonin variation in the context of pregnancy; (iii) study designs such as clinical trials, randomized trials, cohort studies, case–control studies, and case reports; (iv) published between 2015 and 2025; (v) conducted with humans or animal models (mice and rabbits).

The population of interest comprises pregnant women and their fetuses, including studies on exposure to artificial light or melatonin variation related to maternal–fetal outcomes. The outcomes analyzed include melatonin levels, physiological changes, hormonal effects, circadian rhythm homeostasis, sleep quality, and maternal or perinatal complications, such as gestational disorders, preterm birth, and fetal size.

Studies that did not address exposure to artificial light during pregnancy or melatonin variation in the gestational context, investigations in animal populations other than mice or rabbits, studies irrelevant to the research question, review articles, conference abstracts, and letters to the editor were excluded. Additionally, information was extracted from studies that proposed strategies to mitigate potential adverse effects of artificial light exposure during pregnancy.

Information from the selected articles was extracted and organized into tables, separately for studies in humans and animals (mice and rabbits). An independent reviewer (TAC) extracted the essential data, organized into descriptive levels, as detailed below:

For humans, the table columns included:

For animals (mice and rabbits), the columns included:

Any discrepancies in the extracted data were discussed and resolved jointly with two additional reviewers (MCC and GCO). No formal assessment of risk of bias was performed, given the methodological heterogeneity of the included studies.

The search strategy yielded 115 studies (PubMed/Medline: 27; Google Scholar: 88). After removing 21 duplicates and 5 review articles, 73 studies were excluded for having inappropriate topics, as identified by reading the titles and abstracts. Sixteen articles were evaluated in full text, of which 4 were excluded for failing to meet the eligibility criteria. Additionally, 13 further articles were identified through indirect searches of the reference lists of the included studies and on websites, using the snowballing strategy. Thus, 25 studies were included.

The information extracted from the 25 included studies was organized into tables, separately for humans and animals (mice and rabbits). The animal evidence covers pinealectomy, constant light, and dim ALAN, but not the chronic photoperiod shifting (CPS) model repeated shifts of the light:dark cycle that mimic shift work and jet lag in pregnant rodents. The pinealectomy, including constant light, and dim ALAN, and CPS, has emerged as one of the most translational experimental models of gestational chronodisruption because it reproduces the repeated phase advances and delays experienced by rotating shift workers and individuals with chronic jet lag. In this model, pregnant rodents undergo recurrent reversals of the light–dark cycle every 3–4 days throughout gestation. CPS induces persistent fetal programming effects involving metabolic dysfunction, adiposity, insulin resistance, adrenal clock dysregulation, altered steroidogenesis, renal dysfunction, hypertension, and behavioral abnormalities in adult offspring, thereby providing compelling evidence that maternal circadian misalignment alone can program long‐term offspring health [11, 12].

For humans, the following were recorded: source (author and year), number of participants, population, gestational trimester, study type, analyses performed, sample characteristics, duration, effect of the ALAN intervention or exposure (+/−), and primary outcomes. For animals, the following were included: source, number of participants and population, study design, analyses performed, sample characteristics, duration, clinical aspects observed, effect (+/−), and final outcomes. Among the 25 studies, 19 were conducted in humans (Supporting table S1 and Figure 1) and 6 in animal models, including mice and rabbits (Supporting table S2 and Figure 2).

The included studies (2019–2025) evaluated the effects of ALAN exposure and melatonin disruption on reproductive, maternal, and neonatal health across different populations, mainly from China, Europe, Brazil, and the USA. Sample sizes ranged from 9 to 81,820 participants, including healthy pregnant women, women with pregnancy complications, night‐shift workers, and neonatal cohorts.

Of the 19 human studies included in this review, 7 (7/19) analyzed healthy pregnant women, 4 (4/19) investigated nulliparous pregnant women, 2 (2/19) evaluated women in the first trimester (2/19) of pregnancy using blue light‐blocking glasses, 2 (2/19) directly analyzed the risk of preterm birth, 2 (2/19) analyzed fetal size, 2 (2/19) included pregnant women and/or postpartum women with depression and/or anxiety, and 2 (2/19) focused on pregnant women with gestational diabetes. In addition, one study (1/19) investigated pregnant women working night shifts, 1 (1/19) evaluated pregnant women undergoing elective cesarean section, 1 (1/19) studied hospitalized pregnant women with healthy fetuses, 1 (1/19) evaluated gestational fertility, 1 (1/19) analyzed pregnant women and newborns regarding the diversity and composition of the gut microbiota, 1 (1/19) investigated pregnant women with gestational weight gain, 1 (1/19) analyzed pregnant women with fetuses with congenital heart defects, and 1 (1/19) included pregnant women with hypothyroidism.

The mean maternal age in the studies ranged approximately from 27 to 33 years, with a predominance of pregnant women without serious comorbidities, who were nonsmokers, abstained from alcohol, and had singleton pregnancies. In large‐scale Chinese studies, the mean age ranged from 27 to 32 years, with a majority of Han ethnicity, in addition to details on parity, education level, and specific obstetric conditions, lending greater methodological robustness to the samples.

Regarding study design, 6 (6/19) studies were prospective cohort studies, 4 (4/19) were retrospective cohort studies, 4 (4/19) were randomized clinical trials, 2 (2/19) observational case series/cross‐sectional studies, 2 (2/19) retrospective case–control studies, and 1 (1/19) cross‐sectional ecological study. Regarding the gestational period, 7 (7/19) studies included exclusively the third trimester, 4 (4/19) followed from the first to the third trimester, 2 (2/19) covered from the first to the third trimester and the postpartum period, 2 (2/19) evaluated from the pre‐pregnancy period through the postpartum period, and 1 (1/19) study covered the second to third trimesters, 1 (1/19) the first to second trimesters, 1 (1/19) the pre‐pregnancy period to the third trimester, and 1 (1/19) did not specify the period (aggregated population data by census tract).

Regarding the analyses performed, 12 (12/19) studies assessed exposure to ALAN, with eleven (12/19) using satellite imagery. 5 (5/19) studies analyzed melatonin and/or 6‐Ω levels, with 3 (3/19) using urinary parameters, 1 (1/19) using serum and placental assays, and 2 (2/19) using saliva samples. 5 (5/19) studies assessed sleep using actigraphy, scales, and/or sleep diaries. 2 (2/19) studies assessed mood/depression/anxiety using specific scales. One (1/19) analyzed fetal size via ultrasound and anthropometric outcomes at birth. One (1/19) study assessed low birth weight, with adjustments for green space coverage (NDVI). 1 (1/19) article investigated fertility using structured questionnaires and interviews at four time points (first, second, third trimester, and predelivery), with blood and urine collection. 2 (2/19) studies analyzed gestational diabetes, with diagnosis performed via an oral glucose tolerance test (OGTT). 1 (1/19) analyzed maternal and neonatal gut microbiota via 16S rRNA sequencing of stool samples. 1 (1/19) investigated gestational weight gain and excessive weight gain, using logistic and linear regression with analyses stratified by pre‐pregnancy BMI. 1 (1/19) assessed congenital heart defects confirmed by cardiac examinations and procedures, and 1 (1/19) study assessed hypothyroidism through laboratory tests of thyroid function. 3 (3/19) articles conducted interventions aimed at reducing the consequences of exposure to ALAN.

Regarding the effects of exposure to ALAN and the interventions tested, 14 studies (14/19) reported negative effects, 3 (3/19) observed positive effects, 1 (1/19) showed no significant effect, and 1 (1/19) study did not address the relationship with ALAN.

Regarding studies that reported positive effects or no effect, 4 (4/19) evaluated interventions to mitigate the effects of ALAN: 2 (2/19) used blue‐light‐blocking glasses, showing an improvement in the circadian melatonin profile or no significant effect on sleep, and 1 (1/19) applied a biodynamic lighting system, promoting an increase in nighttime melatonin. In addition, 1 (1/19) reported improved mood in pregnant women with peripartum depression through a combined sleep and light intervention.

Regarding negative effects on maternal–fetal outcomes, five studies (5/19) evaluated changes in melatonin: 3 (3/19) identified reduced melatonin secretion associated with ALAN, with a negative impact on maternal sleep, increased risk of preterm birth, and neonatal complications, while 2 (2/19) studies showed increased serum or placental melatonin in pregnant women, with a peak observed in the third trimester.

In addition to assessing metabolic, obstetric/fetal, psychological, and other biological outcomes associated with exposure to ALAN: 2 (2/19) on the risk of preterm birth, one (1/19) on increased fetal abdominal circumference and macrosomia, 1 (1/19) on low birth weight, 1 (1/19) on reduced fertility, 2 (2/19) to the increased risk of gestational diabetes mellitus, 1 (1/19) to changes in maternal gut microbiota, 1 (1/19) to an elevated risk of gestational depression and anxiety, 1 (1/19) to unhealthy weight gain during pregnancy, 1 (1/19) to an increased risk of fetal congenital heart disease, and 1 (1/19) to an increased risk of hypothyroidism, especially in pregnant women who were overweight prior to pregnancy.

According to Supporting Table S1, [13–39] small clinical studies demonstrated that pregnancy is associated with increased melatonin production, particularly during the third trimester, with the placenta acting as an important source of melatonin. Reduced nocturnal melatonin levels associated with nighttime light exposure or circadian disruption were linked to poorer sleep quality, increased risk of preterm birth, assisted delivery, lower Apgar scores, and breastfeeding difficulties [4, 13, 14].

Interventional studies suggest that preserving natural light‐dark cycles may improve maternal circadian regulation. Biodynamic lighting increased nocturnal melatonin production during hospitalization, while blue‐light‐blocking glasses improved melatonin profiles in pregnant women, although effects on sleep outcomes remain inconsistent [15–17]. Light‐based and sleep interventions also showed potential benefits for mood regulation in women with peripartum depression [18].

Large observational cohorts identified associations between ALAN exposure and multiple adverse outcomes. Higher ALAN exposure during pregnancy was associated with increased risk of preterm birth, altered fetal growth, macrosomia, low birth weight, gestational diabetes mellitus, gestational hypothyroidism, excessive gestational weight gain, maternal anxiety and depression symptoms, altered gut microbiota, and congenital heart defects [7, 19–27]. Additionally, pre‐pregnancy exposure to outdoor ALAN was associated with reduced fertility, suggesting possible effects on reproductive function before conception [28].

We chose to include studies with nonhuman animals due to ethical and methodological limitations in studies with human pregnant women. Studies in nonhuman animals allow for more invasive interventions, such as pinealectomy, and detailed assessments of fetal development, especially regarding neurodevelopment, which would not be possible in humans.

Of the 6 (6/6) studies with nonhuman animals included, 4 (4/6) analyzed pregnant and/or lactating Wistar rats, 1 (1/6) evaluated nonpregnant and pregnant Wistar rats, and 1 (1/6) studied pregnant Hi‐Plus rabbits. Among the studies with Wistar rats, 3 (3/5) evaluated the offspring during the postnatal period, and all 5 (5/5) kept adult females on a 12:12 h light/dark cycle. In the study with Hi‐Plus rabbits, nulliparous adult females were monitored during pregnancy and their offspring during the postnatal period.

All 6 (6/6) studies were experimental. 3 (3/6) evaluated pregnant and/or lactating rats with pinealectomy and/or melatonin replacement; 1 (1/6) evaluated pregnant rats exposed to constant light (24 h/0 h) and nonpregnant rats for pineal gland analysis; 1 (1/6) evaluated pregnant rats exposed to dim artificial light; and 1 (1/6) evaluated pregnant Hi‐Plus rabbits administered melatonin during the first and second weeks of gestation.

Regarding the effects of exposure to ALAN and/or dim light, four studies (4/6) reported negative effects, and 2 (2/6) studies did not address the relationship with ALAN.

Three studies (3/6) evaluated pregnant and/or lactating rats that underwent pinealectomy and/or melatonin replacement; of these, 1 (1/3) demonstrated that melatonin deficiency delayed the physical growth, neurobehavioral development, and cognitive development of the offspring, 1 (1/3) impaired maternal metabolic adaptation and pancreatic β‐cell function, and 1 (1/3) altered energy metabolism, adiposity, and thermoregulation in the offspring effects that were partially reversed by melatonin replacement. One study (1/6) evaluated pregnant rats exposed to constant 24 h/0 h light, showing reduced maternal serum melatonin, morphological changes in the pineal gland, increased IL‐6, and early termination of pregnancy. Another study (1/6) investigated pregnant rats exposed to dim artificial light (< 2 lx) during pregnancy, showing disruption of the development of circadian rhythms of hormones and metabolites in the offspring, suggesting an endocrine‐disrupting effect. In the study with Hi‐Plus rabbits (1/6), melatonin administration during the first half of pregnancy increased follicle count, fetal and embryonic weight, placental efficiency, and the expression of antioxidant and regulatory genes, in addition to improving estradiol and progesterone levels, with effects dependent on the sensitive window of application.

According to Supporting Table S2, experimental animal studies provide mechanistic evidence that maternal melatonin is essential for pregnancy adaptation, placental function, and offspring development. In rodent models, maternal melatonin deficiency caused by pinealectomy or exposure to constant light resulted in impaired offspring growth, delayed neurobehavioral and cognitive development, altered glucose metabolism, impaired pancreatic remodeling, reduced insulin secretion, and disrupted energy homeostasis. These effects were largely prevented or reversed by melatonin replacement, demonstrating the protective role of maternal melatonin during gestation and lactation [29–31].

Exposure to ALAN in pregnant rats reduced melatonin secretion, induced pineal gland dysfunction, increased inflammatory responses (particularly IL‐6), and was associated with pregnancy loss [32]. Even low‐intensity nighttime light exposure altered the development of offspring circadian, hormonal, and metabolic rhythms, indicating that gestational ALAN may act as an endocrine disruptor affecting fetal programming [33].

Melatonin supplementation studies further demonstrated beneficial effects on reproductive and placental outcomes. In rabbits, melatonin administration during early pregnancy improved ovarian and placental function, increased fetal growth, reduced embryonic loss, and enhanced antioxidant gene expression, with the second week of gestation identified as a critical period for placental effects [34].

Data sharing is not applicable to this article as no new data were created or analyzed in this study.