Impact of GLP-1 and GLP-1/GIP receptor agonists before, during, and after pregnancy: a narrative review

GLP-1RAs and GLP-1RA/GIP-RAs represent a significant advance in the management of obesity and T2DM in women of reproductive age. Preconception metabolic optimization may reduce gestational complications such as GDM, hypertensive disorders of pregnancy, preterm birth, and cesarean delivery; however, the risk of excessive weight regain following drug discontinuation during pregnancy must be considered. Although evidence regarding fetal safety remains limited, observational studies have not demonstrated a significant increase in congenital malformations or adverse maternal outcomes following inadvertent exposure during early pregnancy, in contrast to unfavorable findings reported in animal models. Nevertheless, these reassuring findings must be interpreted with caution, since their apparent safety may reflect substantial methodological biases in the evaluated studies and not a lack of risk. In the postpartum and lactation period, again, the absence of robust clinical data and the potential excretion of these agents in the breast milk support their contraindication in this period, despite the potential metabolic benefits of postpartum weight loss in women with obesity and T2DM.

This narrative review was conducted through a comprehensive literature search in PubMed, Scopus, Web of Science, Cochrane Library, BVS (Virtual Health Library), and LILACS databases. The search strategy was developed using the PubMed Advanced Search Builder. No publication date restrictions were applied, and the studies ultimately included in this review were published between 2003 and 2025. Medical Subject Headings (MeSH) terms and related keywords were used, including: glucagon-like peptide-1 receptor agonists; incretins; dual GLP-1/GIP receptor agonists; multi-receptor agonists; pregnancy; fertility; infertility; obesity; polycystic ovary syndrome; menstruation; women’s health; reproductive health; ovulation; menstrual cycle; weight loss; gestational weight gain; pharmacovigilance; time-to-pregnancy; congenital abnormalities; embryonic and fetal development; pregnancy complications; gestational diabetes; stillbirth; perinatal death; maternal-fetal exchange; teratogenesis; long-term adverse effects; lactation; breastfeeding; child development; pregnancy outcomes; postpartum period; and birth weight. Articles published in English or Portuguese were considered eligible. We specified the eligibility criteria and the types of studies considered for inclusion, such as clinical trials, systematic and narrative reviews, meta-analyses, clinical guidelines, regulatory reports, animal and ex vivo studies, case reports and case series, and also studies with observational, cohort, case-control and cross-sectional designs, relevant to the review objectives. Conference abstracts were excluded. Preprints were exceptionally considered when addressing topics for which peer-reviewed evidence remained scarce or unavailable, particularly in emerging areas of research. Studies were selected based on their relevance to the effects GLP-1RAs on female reproductive health, fertility, pregnancy, lactation, and offspring outcomes. Publications unrelated to these topics, duplicate records, and studies lacking sufficient methodological or clinical information were excluded. The literature search and study selection were independently performed by three authors (VMS, ACMR, LKP). Titles, abstracts, and full texts were screened for relevance, and any disagreements regarding study inclusion were resolved through discussion and consensus among the authors. Additionally, the reference lists of all selected articles were reviewed to identify further relevant publications not captured during the initial database search. These additional studies were included when they provided clinically relevant information or addressed topics insufficiently covered by the primary search strategy. Receptor agonists were developed to mimic the biological effects of endogenous GLP-1 and GIP by activating their signaling pathways. Analogues are structurally modified agonists with enhanced stability, enzymatic resistance, and prolonged half-lives [ 20 ]. Initially indicated for glycemic control in individuals with T2DM, incretin-based therapies have also been extensively studied and widely used for the treatment of obesity, owing to their favorable metabolic effects, including increased satiety, delayed gastric emptying, fullness sensation, reduced food intake, and consequent weight loss [ 11 , 21 , 22 ]. GLP-1RAs promote weight loss and regulate blood glucose through multiple mechanisms, including stimulation of insulin secretion, glucagon modulation, and exerting central effects on metabolism and appetite. Furthermore, they reduce endogenous glucose production by suppressing glucagon release, which is typically elevated in T2DM [ 11 , 21 ]. In the early 1990 s, exendin-4 was identified in the saliva of the Gila monster ( Heloderma suspectum ). This peptide shares approximately 53% sequence homology with human GLP-1, exhibits similar biological activity, and is resistant to dipeptidyl peptidase-4 (DPP-4) degradation, resulting in a half-life of approximately 2.4 h. This discovery led to the development of exenatide, the first GLP-1RA approved for clinical use in 2005 [ 21 , 23 ]. Subsequently, several GLP-1RAs were developed with differing durations of action and metabolic effects. Liraglutide was introduced in 2010, followed by lixisenatide (2013), dulaglutide (2014), semaglutide (2017), and, more recently, tirzepatide (2022). Tirzepatide is a dual GLP-1RA/GIP-RA, structurally derived from both incretin hormones, which enhances metabolic effects, resulting in superior glycemic control and greater weight reduction [ 24 – 27 ]. Structurally, exenatide and lixisenatide are exendin-4-derived GLP-1RAs (non-human analogues) with 50–53% amino acids sequence homology with human GLP-1, whereas liraglutide, dulaglutide, and semaglutide are human GLP-1 analogues with 90–97% sequence similarity and prolonged duration of action [ 28 ]. Tirzepatide differs from conventional GLP-1RAs as a synthetic dual GLP-1RA/GIP-RA. It shares approximately 67% sequence homology with GIP and 50% with GLP-1, activating the GIPR with a potency comparable to the endogenous ligand while showing lower activity at the GLP-1R [ 27 ]. GIPR and GLP-1R are expressed in several tissues of the gut-gonadal axis, including the gastrointestinal tract, pituitary gland, ovaries, and endometrium. The gut-gonadal axis is a bidirectional communication pathway between gastrointestinal and reproductive systems, mediated by GIP and GLP-1. GIPR and GLP-1R are expressed in female reproductive tissues, including the pituitary gland, ovaries, and uterus. In rodents, GLP-1 stimulates kisspeptin expression in the hypothalamus and increases LH secretion during the estrous cycle, raising the number of mature follicles, while GIP increases pituitary secretion of LH and FSH [ 29 ]. In the study by Khan et al. (2022) [ 29 ], GIPR-knockout (GIPR -/-) and GLP-1R-knockout (GLP-1R -/-) female mices exhibited impaired estrous cycle progression, completing fewer cycles during the observation period and a drastic reduction in reproductive success (50% reduction of fertile females for GIPR -/- mice and 83% for GLP-1R -/-). These findings suggest a critical role for incretin hormone receptors in the regulation of female reproductive function. In animal models, exenatide was detected in the uterus, decidua, and fetal membranes of pregnant mice with systemic inflammation comparable to that observed in metabolic disorders such as T2DM and obesity, whereas no fetal tissue exposure was observed in healthy pregnant animals. In contrast, in rats, mice, and rabbits without systemic inflammation, exposure to other GLP-1RAs (except exenatide) was associated with reduced fetal weight, growth retardation, delayed or irregular ossification, skeletal variants, and decreased embryonic survival. These findings were mainly attributed to decreased food intake and low maternal weight gain [ 30 , 31 ]. Interestingly, endometriosis, a condition associated with subfertility and characterized by the presence of ectopic endometrial tissue, is largely explained by the retrograde menstruation theory (Sampson’s theory), in which displaced endometrial cells are transported into the peritoneal cavity. In this context, peritoneal fluid macrophages play a crucial role in clearing this ectopic tissue [ 32 ]. Investigating this immune mechanism, Krasnyi et al. (2022) [ 32 ] observed significantly lower GLP-1 levels in the peritoneal fluid of women with endometriosis, which correlated with decreased CD86 expression in pro-inflammatory macrophages. Since CD86 is a key marker of macrophage activation required for this clearance process, its reduction suggests that local GLP-1 deficiency may compromise the peritoneal immune response, thereby facilitating the survival of endometrial fragments and the development of the disease. Regarding endometrial receptivity, an experimental study by Artunc-Ulkumen et al. (2015) [ 33 ] observed that exenatide improved endometrial quality and receptivity in rats with DM by reducing inflammation and fibrosis within the endometrial tissue. The use of GLP-1RAs by women of reproductive age has increased substantially in recent years, given their efficacy in treating obesity and DM. Since an elevated BMI is associated with reduced response to ovulation-inducing agents, GLP-1RAs emerge as a potential adjuvant therapy in the treatment of obesity-related infertility [ 34 ]. Polycystic ovary syndrome (PCOS), characterized by subfertility, irregular menstrual cycles, and hirsutism, represents the most common metabolic disorder found in women of reproductive age and is frequently associated with other pathological conditions, such as T2DM, dyslipidemia, and metabolic syndrome [ 35 ]. Most pharmacological treatments for PCOS primarily target these metabolic disturbances. Among different pharmacological strategies, GLP-1RAs have demonstrated significant efficacy in promoting weight reduction in women with obesity and PCOS. The potential benefit of using GLP-1RA in women with PCOS may not be solely associated with weight or hyperandrogenism reduction. In animal models of PCOS, activation of the GLP-1R has demonstrated an effect in promoting cell cycle regulation and follicular maturation. Moreover, high doses of GLP-1RA have been associated with granulosa cell degeneration and altered expression of steroidogenic enzymes. The effect of GLP-1RA on LH through GnRH stimulation raises a theoretical concern about their use in PCOS, which could potentially worsen hyperandrogenism and ovulatory dysfunction, especially in women without obesity. To date, there are no reports of this effect in humans [ 36 , 37 ]. A randomized controlled trial (RCT) comparing the use of exenatide with metformin for 12 weeks in 176 women with PCOS showed a significant increase in the number of spontaneous pregnancies (43.60% vs. 18.70%, p < 0.05) in those who used the GLP-1RA [ 38 ]. A recent meta-analysis that included RCTs evaluating 397 women with PCOS comparing the use of GLP-1 agonists (liraglutide, exenatide, or semaglutide) with metformin or placebo, showed that GLP-1RA had a higher impact on reducing body weight and improving insulin sensitivity [ 39 ]. However, it should be mentioned that these studies were not classified as having high-quality evidence. GLP-1RAs emerge as a potential therapeutic option for PCOS and related comorbidities, although further evidence is required due to methodological limitations of the existing studies. Regarding the preconception safety of these agents, a systematic review by Gunabalasingam et al. (2025) [ 40 ], identified a marked scarcity of RCTs addressing women with T2DM during the preconception and postpartum periods, as most studies focused on pregnancy. Beyond this lack of studies, pharmacological interventions primarily comprised metformin and insulin, with no mention of GLP-1RAs-based therapies. The dearth of specific evidence regarding their use in the preconception period limits safe and well-informed clinical decision-making, thereby compromising optimal glycemic control in women of reproductive age. Consequently, the use of GLP-1RAs in this setting remains challenging, as insufficient data preclude a balanced assessment of their risks and benefits in this phase. Recent studies have also pointed to potential drug-drug interactions between oral hormonal contraceptives (OHC) and GLP-1RAs or dual GLP-1RA/GIP-RA, mainly attributed to delayed gastric emptying. Prolonged gastric transit time may delay the delivery of gastric contents into the intestine, where most OHC are absorbed, thereby altering their pharmacokinetic parameters. According to Min et al. (2025) [ 41 ], the impact of GLP-1RAs (semaglutide, exenatide, liraglutide, and dulaglutide) on OHC pharmacokinetics was minimal and not clinically relevant. Exenatide was the only agent associated with changes in maximum plasma concentration (Cmax), an effect attributed to the methodological design of the study by Kothare et al. (2012) [ 42 ], in which the OHC was administered 30 min after exenatide injection, coinciding with peak delay in gastric emptying. Conversely, a phase 1 clinical trial ( NCT04172987 ) [ 43 ] showed that tirzepatide had a more pronounced effect on OHC absorption compared with other incretin-based therapies. Reductions in Cmax and area under the curve (AUC) values of ethinylestradiol by 59% and 21%, respectively, and of norelgestromin by 55% and 22%, respectively, were observed, along with an increase in Tmax, with OHC taking between 2.5 and 4.5 h longer to reach peak plasma concentrations. These findings occurred during maximal gastric emptying delay following a single 5 mg dose of tirzepatide. Importantly, the study did not account for tachyphylaxis associated with repeated dosing, a phenomenon in which the delay in gastric emptying diminishes over time with continued treatment. This differential effect on OHC absorption is likely related to tirzepatide’s unique dual agonism of GLP-1R and GIPR, resulting in a greater delay in gastric emptying compared with conventional GLP-1RAs. Consequently, for patients using tirzepatide, the use of an additional contraceptive method, such as barrier contraception, or switching to non-oral contraceptive methods (e.g., intrauterine devices or hormonal implants), is recommended for four weeks following treatment initiation or dose escalation [ 44 , 45 ]. (Table 1 ). Table 1 Impact of GLP-1 and GLP-1/GIP receptor agonists in the preconception period Focus Area Author Study Design/Population Main Outcomes Limitations Reproductive Function Khan et al. [ 33 ] (2022) Experimental. 2 separate groups of mice (GLP-1 -/- and GIPR -/-), Each of these groups was then compared to age-matched wild-type control mice Incretin receptors are essential for the gut-gonadal axis; female mice GIPR -/-) or GLP-1R -/- exhibited significantly disrupted estrous cycles; GLP-1R -/- females showed up to an 83% reduction in reproductive success, while GIPR -/- showed an average reduction of 50%; The average number of litters produced was also significantly lower in GIPR -/- (2 to 3 litters) and GLP-1R -/- (1 litter) groups compared to controls (4 to 5 litters). Murine knockout model; human physiological translation requires further study. Endometriosis Krasnyi et al. [ 34 ] (2022) Case-control study comparing 54 women with endometriosis and 30 controls with uterine myoma. GLP-1 levels are significantly lower in the peritoneal fluid of endometriosis patients and positively correlate with CD86 expression in peritoneal macrophages of women with endometriosis. This suggests that reduced GLP-1 impairs the T-lymphocyte activation required for the immunological clearance of ectopic tissue. Cross-sectional design limits causal inference between factors and endometriosis development Endometrial Cancer Li et al. [ 35 ] (2024) In vitro study evaluating GLP-1R knockdown and exendin-4 treatment in human endometrial cancer cells. GLP-1R is significantly up-regulated in endometrial cancer, and its activation by exendin-4 facilitated proliferative and metastatic potentials while protecting cells from ferroptosis. Preclinical in vitro models; clinical implications for systemic incretin therapies require prospective validation Kanda et al. [ 36 ] (2018) Clinical tissue analysis (endometrial cancer microarray) and in vitro experimental study (Ishikawa cells treated with liraglutide). higher GLP-1R expression correlates with positive estrogen and progesterone receptor status and improved progression-free survival, with in vitro liraglutide inhibiting cell growth and inducing autophagy. Retrospective study without normal tissue immunohistochemical control; in vitro results reflect pharmacological activation only. Zhang et al. [ 37 ] (2015) In vitro (Ishikawa cell line) and in vivo (nude mice xenografts) experimental study Exendin-4 demonstrated antitumor effects, primarily by promoting apoptosis, which resulted in a significant reduction in tumor volume and weight in the in vivo model. Solely preclinical data; the antineoplastic efficacy and safety of liraglutide for human endometrial cancer lack clinical trial validation. Endometrial Receptivity Artunc-Ulkumen et al. [ 38 ] (2015) Experimental (Streptozotocin-induced diabetic rat model) Exenatide treatment significantly reduced fibrosis (endometrial and ovarian), oxidative stress, and inflammatory markers Chemically induced T1DM-like model; may not fully replicate human T2DM or PCOS metabolic complexity PCOS Management Han, Li, and He [ 40 ] (2019) Systematic review and meta-analysis of 8 RCTs ( n  = 375 patients) GLP-1RAs were significantly more effective than metformin in reducing body weight, BMI, abdominal girth, and improving insulin resistance (HOMA-IR). Small sample sizes, short follow-up durations, and significant clinical heterogeneity among the included studies. Interactions with OHC Min et al. [ 44 ] (2025) Comprehensive narrative review analyzing clinical studies and trial data (including Phase 1 trial NCT04172987 ) [ 46 ] on GLP-1RAs and dual GLP-1/GIPR Minimal impact on OHC pharmacokinetics; exenatide was the sole agent altering Cmax, largely due to administration at peak gastric stasis, while tirzepatide ( NCT04172987 data) delayed Tmax and reduced Cmax of ethinylestradiol and norelgestromin, requiring additional barrier contraception though the effect diminishes over time due to tachyphylaxis Heterogeneous designs (e.g., OHC timing); Phase 1 data in healthy women may not replicate the complex metabolic and hormonal baseline of target populations. Impact of GLP-1 and GLP-1/GIP receptor agonists in the preconception period Incretin receptors are essential for the gut-gonadal axis; female mice GIPR -/-) or GLP-1R -/- exhibited significantly disrupted estrous cycles; GLP-1R -/- females showed up to an 83% reduction in reproductive success, while GIPR -/- showed an average reduction of 50%; The average number of litters produced was also significantly lower in GIPR -/- (2 to 3 litters) and GLP-1R -/- (1 litter) groups compared to controls (4 to 5 litters).

Obesity and diabetes mellitus (DM) are important public health problems with a significant impact on women’s reproductive health. Between August 2021 and August 2023, the estimated prevalence of obesity among American women aged 20 years or older was 41.4% [ 1 ]. In Brazil, data from the Food and Nutritional Surveillance System (SISVAN) showed that in 2018, among 840,243 women followed in primary care, 48.6% were overweight in the pregestational period [ 2 ]. Obesity causes hormonal changes that impair the hypothalamic-pituitary-ovarian axis, resulting in menstrual disorders such as cycle irregularity, oligo-ovulation, and anovulation. In addition, obesity negatively affects pregnancy by hindering endometrial embryo implantation, contributing to delayed conception, and increasing rates of spontaneous abortion [ 3 ]. Beyond the effects on fertility, maternal obesity may also compromise fetal development. An analysis of 32 studies, encompassing more than 4 million births, demonstrated that obesity was associated with a higher occurrence of fetal congenital heart malformations, which increased in parallel with maternal body mass index (BMI) [ 4 ]. Maternal obesity is also associated with a higher incidence of large for gestational age and macrosomic infants, which may result in delivery complications as well as short- and long-term health impairment for the child [ 5 ]. Infants born to obese mothers are more likely to require admission to neonatal intensive care units and present higher rates of stillbirth and preterm births [ 6 ]. Obesity also poses risks to maternal health and is associated with increased risk of gestational diabetes mellitus (GDM) and hypertensive disorders of pregnancy, such as preeclampsia. These conditions are associated with higher rates of obstetric complications, including the need for cesarean delivery and higher risk of postpartum hemorrhage [ 6 ]. Similarly, DM is highly prevalent among women of reproductive age. In 2024, approximately 118.5 million live births occurred worldwide, with 19.7% being affected by some kind of dysglycemia, with about 23 million hyperglycemia-exposed children globally. Most cases were attributed to GDM (79.2%), defined as any degree of glucose intolerance with onset or first recognition during pregnancy [ 7 ]. Additionally, 11% of cases are pregestational diabetes mellitus (PGDM), corresponding to the presence of type 1 diabetes mellitus (T1DM) or type 2 diabetes mellitus (T2DM) before pregnancy. The remaining 9.9% corresponded to other types of DM first detected during pregnancy [ 8 ]. Pregnancy in women with DM is associated with a higher risk of maternal and fetal complications. Maternal hyperglycemia during organogenesis promotes molecular and epigenetic changes, such as increased expression of glucose transporters (GLUT-1 and GLUT-2), increased production of free radicals, and reduced expression of the PAX3 gene, which may lead to congenital malformations. The most frequent anomalies are congenital heart defects, neural tube defects, renal and urinary tract anomalies, as well as cleft lip with or without cleft palate [ 9 ]. Furthermore, maternal DM is associated with a higher risk of hypertensive disorders of pregnancy, preeclampsia, preterm birth, cesarean delivery, and the worsening of preexisting DM-related chronic complications, such as retinopathy and nephropathy [ 10 ]. The growing number of women of reproductive age with obesity and DM has driven the use of incretin-mimetic medications to treat both conditions, reinforcing the need for in-depth research on their potential impacts on fertility, as well as the safety of their use during pregnancy and breastfeeding. Insulin and glucagon effects are pivotal in glycemic homeostasis, as well as the actions of incretin hormones such as glucose-dependent insulinotropic polypeptide (GIP) and glucagon-like peptide-1 (GLP-1). Produced by the gastrointestinal tract in response to food intake (GLP-1 by L cells in the ileum and GIP by K cells in the duodenum), these hormones potentiate glucose-dependent insulin secretion by binding to their respective receptors on pancreatic β cells, characterizing the so-called incretin effect [ 11 – 15 ]. Both GIP and GLP-1 act across multiple physiological systems. GLP-1 delays gastric emptying, regulates glucose absorption, reduces postprandial glycemic excursions, improves skeletal muscle glucose uptake, activates brown adipose tissue, increasing energy expenditure, and has beneficial effects on cardiovascular health [ 11 , 12 , 15 – 17 ]. Conversely, GIP promotes lipid uptake and storage in white adipose tissue via activation of lipoprotein lipase (LPL), favoring subcutaneous adipose tissue (SAT) expansion. This adipose tissue expansion is beneficial, since SAT serves as the primary body lipid reservoir, limiting circulating triglycerides and thereby preventing ectopic lipid deposition (“lipid spillover”) in organs such as liver and skeletal muscle. Regarding their effects on glucagon regulation, GLP-1 suppresses its secretion during hyperglycemia, whereas GIP stimulates its release during hypoglycemia. However, glucagon response to GIP is not fully understood, as it may also increase glucagon levels during hyperglycemia, which is not observed in healthy individuals [ 18 , 19 ]. The use of incretin mimetic therapy throughout the reproductive life of women with obesity or T2DM has the potential of bringing benefits to women’s health and improving pregnancy outcomes. Although the evidence to indicate or avoid the use of these medications is not yet conclusive, there is a growing number of studies that have assessed their effects on female reproductive health. In this context, this narrative review aims to appraise the available evidence regarding the use of GLP-1RA and GIP-RA-based therapies in women’s reproductive health, spanning preconception, pregnancy, and postpartum periods, with focus on maternal and fetal safety.

A cohort study conducted with more than 50,000 pregnant women with T2DM in 6 countries demonstrated that the use of GLP-1RAs during the gestational period did not increase the risk of major congenital malformations in the offspring beyond the risk previously associated with DM, compared with pregnant women who used insulin, dipeptidyl peptidase 4 inhibitors (DPP4i), sodium-glucose cotransporter 2 inhibitors (SGLT2i), sulfonylureas, and dual GLP-1RA/GIP-RAs [ 30 ]. Literature reviews by Varughese, O’Mahony, and Varadhan (2025) [ 46 ], as well as by Price and Nankervis (2025) [ 47 ], highlight the deleterious effects that metabolic diseases such as obesity exert on pregnancy. Accordingly, medical societies recommend weight loss during the preconception period to mitigate adverse maternal and fetal outcomes, which has contributed to the increasing use of GLP-1RAs. Although no fetal malformations have been reported following inadvertent exposure to these agents, animal studies suggest that the use of GLP-1RAs during pregnancy may result in harmful effects on the offspring, primarily due to their impact on maternal nutrition. There is some evidence, from small studies, for the occurrence of adverse offspring effects of GLP-1RA [ 31 ]. Semaglutide exposure in animals and human offspring was linked to visceral abnormalities and major skeletal malformations. During lactation, substantial variability in milk transfer was observed among different GLP-1RAs, with liraglutide reaching approximately 50% of maternal plasma concentrations and also being associated with fetal growth restriction [ 31 ]. Therefore, effective contraception is recommended during GLP-1RAs therapy, along with drug discontinuation and a washout period of at least five half-lives length (approximately 4–6 weeks), preconceptionally. Thus, preconception counseling is essential for appropriate glycemic and weight management. Guidance on discontinuation of GLP-1RAs, nutritional and physical activity counseling, and proper transition to alternative pharmacological therapies for T2DM should be offered to avoid excessive weight gain. Studies on the reproductive safety regarding exposure to GLP-1RA or GLP-1RA/GIP-RAs before and during pregnancy are still scarce; therefore, the balance between potential risks and benefits concerning glycemic control and weight management in women of reproductive age remains unanswered. Consequently, it is essential to conduct additional studies evaluating and monitoring larger samples of patients. Due to the increasing prevalence of T2DM among women of reproductive age, the inadvertent use of non-insulin antidiabetic drugs during pregnancy has also increased, including GLP-1RAs and GLP-1RA/GIP-RAs [ 48 ]. In this context, it became relevant to investigate the potential impact of these agents on pregnancy outcomes. Accordingly, an ex vivo placental perfusion study with exendin-4 has detected extremely low concentrations of this drug on the fetal side. However, the ex vivo nature of the study precluded definitive conclusions regarding the risk of fetal exposure in vivo [ 49 ]. A case was reported involving a woman with T2DM who continued the use of liraglutide throughout pregnancy. After delivery, at 39 weeks of gestation, liraglutide concentrations in umbilical cord were below the limit of detection and similar to values observed in a control sample from an unexposed pregnant woman. In contrast, the drug remained detectable in maternal serum at higher concentrations than in the control sample, suggesting low placental transfer [ 50 ]. Another case involved a woman with obesity and PCOS who experienced two pregnancies while using exenatide. In the first instance, she inadvertently used the medication from the 13th to the 17th week of gestation. She gave birth at 37 weeks of gestation to a girl with an atrial septal defect. Two years later, she was using exenatide again and discovered she was pregnant at 13 weeks of gestation. This pregnancy progressed without complications [ 51 ]. A retrospective study evaluating pregnant women with T2DM (≥ 18 years of age) between June 2005 and February 2023, using the TriNetX database was performed. Two cohorts were formed: one with women exposed to GLP-1RAs up to one year before and up to one month after first-trimester pregnancy diagnosis, and another formed by women without exposure to GLP-1RAs during the same period. Participants were followed during the entire gestational period and the immediate postpartum. No significant increase in all-cause maternal mortality, secondary maternal outcomes (gestational hypertension, preeclampsia, and eclampsia), or fetal cardiac and renal anomalies were found in the exposed group [ 52 ]. The largest study conducted to date evaluating the use of GLP1-RAs in the periconceptional period comprised a cohort of 51,826 pregnancies in women with T2DM treated with various antidiabetic medications from six countries. The use of GLP-1RAs was not associated with a significant increase in the risk of major congenital malformations, compared with insulin, the gold standard for DM treatment during pregnancy [ 30 ]. Another study using data from the European Network of Teratology Information Services (ENTIS) evaluated 168 pregnant women exposed to GLP-1RAs during the first trimester, compared with 156 women living with DM and treated with other antidiabetic agents and with 163 women without DM with overweight/obesity. After adjustment, no significant associations were observed between GLP-1RAs exposure and an increased risk of major congenital malformations, pregnancy loss, or preterm birth [ 53 ]. One of the most important pharmacovigilance studies on the use of GLP-1RAs during pregnancy used data from the FDA (Food and Drug Administration) Adverse Event Reporting System (FAERS) to identify potential adverse drug reactions (ADRs). A progressive increase in ADRs reports associated with GLP-1RA use has been observed since 2012, with a disproportionate association with events such as spontaneous abortion, preeclampsia, and gastrointestinal disorders. However, many exposed women were of advanced age and had T2DM or obesity, which represented important confounding factors. These findings reinforce the need for caution in the use of GLP-1RAs during pregnancy and support current guidelines recommending insulin as the first-line treatment for T2DM during pregnancy [ 48 ]. Another study demonstrated that exendin-4 exerts relevant immunomodulatory effects in animal models with maternal systemic inflammation induced by lipopolysaccharide. Exendin-4 reduced the rate of preterm births and significantly improved neonatal survival. These beneficial effects were associated with the attenuation of the inflammatory response, including macrophage polarization toward the M2 phenotype, increased Interleukin-10 producing neutrophils, and normalization of neonatal cytokines profile. However, the benefits were less evident when inflammation was predominantly intra-amniotic, suggesting that the efficacy of GLP-1RAs may depend on the involved compartment [ 54 ]. Another retrospective cohort study that also used the TriNetX database aimed to determine whether the use of these incretin-based agents could influence the course of subsequent pregnancies. It was found that women previously exposed to GLP-1RAs had a significantly lower incidence of GDM, as well as lower rates of hypertensive disorders of pregnancy, preterm birth, and cesarean delivery compared with those without prior exposure. (Figure 1 ) These findings suggest that the metabolic benefits induced by GLP-1RAs treatment may persist after drug discontinuation, reinforcing the hypothesis that adequate glycemic control, weight loss, and reduction of systemic inflammation prior to conception may positively impact pregnancy outcomes [ 55 ]. A retrospective cohort study based on electronic health record data from a large academic health system evaluated singleton pregnancies occurring between 2016 and 2025 to investigate the impact of GLP-1RAs discontinuation during the preconception period or early pregnancy. The results demonstrated that prior exposure to GLP-1RAs, followed by discontinuation before or during early gestation, was associated with greater gestational weight gain and a higher risk of excessive weight gain, as well as an increased incidence of GDM, hypertensive disorders of pregnancy, and preterm birth. Although a slight increase in birth weight was observed in pregnancies with GLP-1RA exposure, no increased risk of large- or small-for-gestational-age neonates or cesarean delivery was identified. These findings suggest that discontinuation of GLP-1RAs may potentially attenuate the benefits achieved before conception [ 56 ]. Therefore, it should be noted that the safety of withdrawing these drugs during the periconceptional period is still a matter of debate. It is of note that the risk of weight regain can be related to individual weight-trajectory patterns and difficulty in maintaining healthy dietary habits acquired during the treatment [ 57 ]. (Figure 1 ) Studies conducted with non-pregnant populations showed that up to two-thirds of the weight lost can be regained within a period of 12 months. However, weight regain after discontinuation of these medications follows a progressive, not abrupt pattern, with gradual and only partial reversal of metabolic effects [ 58 , 59 ]. Although most studies involving the inadvertent or intentional use of GLP-1RAs during pregnancy have not shown a significant increase in adverse outcomes for the mother and fetus, insulin is the treatment of choice for hyperglycemia during pregnancy, as advocated by most medical societies, such as American Diabetes Association (ADA) and the Brazilian Diabetes Society (SBD) [ 60 , 61 ] (Table 2 ). Table 2 Maternal-fetal outcomes after GLP-1RAs exposition during pregnancy Study Study Type/Population Main Outcomes Limitations Hiles et al. (2003) [ 50 ] Experimental ex vivo study/Term human placentas Minimal placental transfer, with no evidence of significant fetal accumulation Ex vivo model; no maternal–fetal clinical evaluation; no assessment of prolonged exposure Ivanisevic et al. (2018) [ 51 ] Case report/Pregnant woman with T2DM Healthy term newborn; liraglutide detectable in maternal serum and undetectable in fetal circulation Single case ( n  = 1); single measurement; no serial pharmacokinetic assessment; no long-term follow-up Doğan et al. (2023) [ 52 ] Case report/Woman with obesity, PCOS, and T2DM 1 st pregnancy: atrial septal defect with spontaneous closure; 2nd pregnancy: no complications; normal child follow-up Single case ( n  = 1); no control group; inability to establish causality Hanif et al. (2025) [ 53 ] Retrospective cohort (TriNetX)/3,652 pregnant women with T2DM No significant increase in maternal mortality, hypertensive disorders of pregnancy, or cardiac and renal malformations Reliance on electronic health records; possible exposure misclassification; lack of long-term follow-up data Cesta et al. (2023) [30*] Multinational cohort/51,826 pregnancies with pregestational T2DM No increased risk of major or cardiac congenital malformations compared to insulin Observational design; potential residual confounding; relatively small number of exposed cases; lack of long-term outcomes Dao et al. (2023) [ 54 ] Prospective multicenter cohort/168 exposed pregnancies No significant association with major malformations, pregnancy loss, or prematurity Limited sample size; lack of systematic HbA1c data; possible selection bias; early medication discontinuation Zhou et al. (2025) [ 49 ] Retrospective observational study of spontaneous reports/354 cases Signals of disproportionality for miscarriage and preeclampsia; progressive increase in reports over time Subject to under/overreporting; no incidence calculation; multiple confounders; cannot establish causality Garcia-Flores V et al., (…) (2018) Experimental study in a pregnant rat model. Inflammation was induced by LPS administration via intraperitoneal and intra-amniotic routes. Exendin-4 reduced the rate of preterm birth; improved neonatal survival in the systemic inflammation model; normalized neonatal cytokine profiles; promoted macrophage polarization toward the M2 phenotype and increased anti-inflammatory neutrophils; effects were less pronounced in the intra-amniotic inflammation model. Animal-only model; dosing and routes of administration may not reflect clinical use; short-term evaluation with neonatal follow-up limited to 15 days. Imbroane et al. (2025) [ 55 ] Retrospective cohort/Women with preconception exposure Lower incidence of gestational diabetes, hypertensive disorders, and prematurity Possible coding errors; lack of detailed BMI and glycemic control data; inability to establish causality Maya et al. (2025) [ 56 ] Retrospective cohort/Singleton pregnancies Greater gestational weight gain; higher risk of GDM, hypertensive disorders, and preterm birth; no increase in LGA or SGA Retrospective design; confounding by indication; reliance on electronic records; limited data on adherence and exact timing of exposure Maternal-fetal outcomes after GLP-1RAs exposition during pregnancy Fig. 1 Maternal and fetal implications of GLP-1RAs and GLP-1/GIPRAs therapy . GLP-1RA: Glucagon-like peptide-1 receptor agonist; GDM: Gestational diabetes mellitus. The visual elements of this illustration were generated with the assistance of generative AI (Gemini) and subsequently reviewed, validated, and edited by the authors Maternal and fetal implications of GLP-1RAs and GLP-1/GIPRAs therapy . GLP-1RA: Glucagon-like peptide-1 receptor agonist; GDM: Gestational diabetes mellitus. The visual elements of this illustration were generated with the assistance of generative AI (Gemini) and subsequently reviewed, validated, and edited by the authors Maternal weight reduction during the puerperal period may confer long-term metabolic benefits. A 2021 meta-analysis demonstrated that postpartum weight retention constitutes an important risk factor for maternal obesity, metabolic syndrome, and cardiovascular disease, reinforcing the relevance of weight loss during this period. Although the literature prioritizes interventions based on lifestyle modification, structural, behavioral, and psychosocial barriers inherent to this period significantly contribute to lower adherence and, consequently, reduced effectiveness of these strategies in clinical practice. In this context, such findings support the relevance of investigating complementary therapeutic approaches for weight management, tailored to the particularities of the puerperium [ 62 ]. Among emerging pharmacological approaches, GLP-1RAs and GIP-RAs have progressively gained attention as a possible strategy. Despite the potential role of these medications in promoting postpartum weight loss and improving maternal metabolic profile, evidence regarding their safety during lactation period, as well as their effects on maternal and offspring health, remains limited [ 63 ]. Preclinical studies suggest that, in lactating rodents, injectable semaglutide was excreted into breast milk, and its potential risks to the infant cannot be disregarded. In its oral formulation, the medication contains sodium salcaprozate, a compound that may be transferred to milk and potentially accumulate in the offspring. Regarding tirzepatide, manufacturer data suggest low or negligible maternal plasma levels of the drug and that, when administered subcutaneously, it presents undetectable concentrations in milk, resulting in minimal infant exposure. However, although this scenario appears favorable, the scarcity of robust human safety data implies in the contraindication of GLP-1RAs and GIP-RAs during lactation by major regulatory authorities [ 64 – 66 ]. The safety of the semaglutide, lixisenatide, and liraglutide during pregnancy and lactation was further evaluated in a 2023 review that analyzed 26 studies published up to mid-year. The authors reported relevant adverse outcomes in animal offspring, including reduced fetal and postnatal growth, delayed ossification, and skeletal variations, supporting recommendations against the initiation of these agents in the postpartum period. It was also observed, in animal models, that exposure to GLP-1RAs during lactation resulted in detectable concentrations in maternal plasma and milk. Among the analyzed drugs, liraglutide exhibited the highest milk transfer (approximately 50%), being associated with impaired neonatal growth. No human safety data was identified. In this study, the prevailing hypothesis is that, despite their high molecular weight, GLP-1RAs may be transferred into breast milk at low levels; and although probable degradation in the infant’s gastrointestinal tract is expected, the occurrence of biological effects similar to those observed in preclinical studies cannot be ruled out [ 31 ]. In contrast to the presented data, the study conducted by Diab et al. (2024) [ 67 ] prospectively evaluated infants of mothers between six months and two years postpartum, who were self-administering subcutaneous semaglutide at variable doses (between 0.25 and 1 mg/week) for a minimum period of three weeks. The investigation included serial collection of breast milk following drug administration, with subsequent laboratory analysis for drug detection, combined with questionnaires assessing offspring clinical parameters, including growth and developmental milestones. The results demonstrated undetectable drug concentrations in the milk, associated with relative infant doses below adopted thresholds. Regarding clinical parameters, weight and height measurements remained compatible with expectations for age groups, without the presence of significant adverse effects. In a similar context, a preprint study of Thompson et al. (2025) [ 68 ] followed infants of mothers between seven and eleven months postpartum who were using subcutaneous self-administered tirzepatide at doses ranging from 2.5 to 5 mg per week. In this case, breast milk samples were collected after a 168-hour administration interval. Undetectable levels of the drug were found in all analyzed samples, although small peaks of drug concentration were observed [ 68 ]. Nevertheless, the reduced sample sizes and short follow-up periods of existing cohorts, frequently limited to a few weeks postpartum, preclude robust inferences regarding long-term outcomes, restricting the evaluation of risks and benefits for both the mother and the newborn. Therefore, larger-scale clinical trials and well-designed observational studies are necessary to elucidate the metabolic, perinatal, and clinical effects of GLP-1RAs and GLP-1RA/GIP-RAs during breastfeeding and post-gestational period (Table  3 ). Table 3 Use of GLP-1RAs postpartum: implications for puerperium and lactation Study Study type/Population Main outcomes Limitations Müller TD, et al. [ 29 ] (2023) Systematic review. Included 39 records ( n  = 9 theoretical; n  = 7 human; n  = 23 animal), comprising 76 human offspring. Reduced neonatal growth No evidence available in humans; exposure to analogue limited to animal models; small sample size Makama M, et al. [ 59 ] (2021) Narrative review. Included postpartum women (up to 5 years), including specific subgroups and those with risk factors. Reduction in body weight and postpartum weight retention based on behavioral changes Scarcity of specific evidence regarding the use of GLP-1 analogues Diab, et al. [ 62 ] (2024) Prospective observational clinical study. Mothers and infants between 6 months and 2 years postpartum were included, with self-administration of semaglutide for 3 weeks or longer ( n  = 8). Semaglutide was undetectable in breast milk; absence of clinically relevant adverse effects in infants; preserved infant growth and development. Small sample size; no analysis of milk production/composition; no long-term follow-up of exposed infants; analysis restricted to mature or involuting milk production phases Diab, et al. [ 63 ] (2025) Prospective observational clinical study. Mothers of infants between 7 and 11 months postpartum with self-administration of tirzepatide were included ( n  = 5). Tirzepatide was undetectable in breast milk. Small sample size; information gaps; use of low doses. Use of GLP-1RAs postpartum: implications for puerperium and lactation Systematic review. Included 39 records ( n  = 9 theoretical; n  = 7 human; n  = 23 animal), comprising 76 human offspring. Prospective observational clinical study. Mothers and infants between 6 months and 2 years postpartum were included, with self-administration of semaglutide for 3 weeks or longer ( n  = 8). Although this review provides a comprehensive and updated synthesis of the use of GLP-1 and GIP-RAs across the reproductive cycle, some limitations must be considered. First, the narrative nature of the study entails a qualitative selection of the literature, which may introduce bias compared with systematic reviews. Furthermore, the scarcity of RCTs in pregnant women rely predominantly on pharmacovigilance data and retrospective cohorts. As the clinical use of incretin-based therapies continues to expand among women of reproductive age, the rising number of inadvertent gestational exposures will naturally provide a broader basis of real-world evidence. This growing volume of clinical data is expected to clarify current methodological gaps and further refine our understanding of the maternal-fetal safety profile of these medications.