Abstract
Objective
This study aimed to systematically characterize the features and drug distribution of neonatal adverse drug reactions using data from the FAERS, with a focus on exposure-specific patterns and stratified analyses by sex and clinical outcomes.
Methods
Reports submitted to FAERS from the first quarter of 2004 to the fourth quarter of 2024 were utilized. Patients experienced the ADR at an age of no more than 28 days. Four quantitative disproportionality analysis methods, including ROR, PRR, BCPNN, and MGPS, were used to detect signals of adverse drug events.
Results
A total of 15,456 neonatal cases exposed to the target drugs were included, yielding 60,611 adverse event reports, 95.45% of which were classified as serious. The median time to onset of ADRs was 264 days for intrauterine drug exposure, compared to 1–3 days for extrauterine exposure. The most affected SOCs were injury and procedural complications (19.53%), congenital disorders (15.96%), and pregnancy/perinatal conditions (8.65%). Transplacental exposure accounted for the highest proportion (52.47%), followed by intravenous (9.34%), oral (6.77%), breastfeeding (1.80%), intramuscular (1.48%), and inhalation (1.29%). The top maternal exposure drugs included venlafaxine, sertraline, quetiapine, lamotrigine, and levetiracetam. For intravenous use, levetiracetam, zidovudine, indomethacin, ibuprofen, and vancomycin were most common. Female neonates had higher risks of microcephaly, ventricular septal defect, and growth restriction, while male neonates were more prone to hypospadias, cryptorchidism, and agitation. Serious AEs were mainly linked to maternal drug exposure during pregnancy.
Conclusion
These results showed that the use of antidepressants, antiepileptics, and antivirals during pregnancy represents a significant risk factor for neonatal adverse reactions, particularly congenital malformations. Consequently, it is imperative to implement precise prevention strategies tailored to specific exposure stages and to advocate for the establishment of an international pharmacovigilance network for neonates.
Keywords: Neonates, Adverse event report, Drug exposure routes, Pharmacovigilance, FAERS
Introduction
According to data from the World Health Organization (WHO), over 2.3 million children died within the first month of life in 2022, accounting for 47% of all deaths among children under five years of age. The leading causes of neonatal mortality include preterm birth, intrapartum complications, congenital anomalies, and infections, with approximately 240,000 newborns dying from congenital disorders 1 . Due to ethical and technical constraints, neonates are often excluded from clinical drug trials, resulting in their pharmacotherapy being largely based on extrapolation from adult or older pediatric dosing regimens as well as empirical use 2 . Consequently, off-label drug use occurs in up to 90% of neonates, which is a significant risk factor for adverse drug reactions (ADRs) in this population 3 4 5 . This issue is particularly critical given the immature hepatic and renal function in neonates and their markedly different pharmacokinetic profiles compared to other populations. These factors contribute to distinct patterns of ADRs in neonates, posing major challenges for their identification and monitoring 6 7 . Therefore, leveraging real-world data to systematically mine and evaluate ADRs in neonates is of great clinical and public health significance for ensuring medication safety in this vulnerable population.
Previous studies have conducted preliminary investigations into ADRs in pediatric populations, revealing distinct patterns in terms of severity, system organ class (SOC) distribution, and medication use profiles 8 9 10 11 . A cross-sectional study comparing ADR characteristics between adults and children as well as among pediatric subgroups reported that neonates exhibited the highest proportion of serious ADRs (96.2%), followed by infants (79.8%), young children (67.9%), older children (59.5%), and adolescents (52.7%) 12 . However, most existing research tends to focus on the broad category of "children" and lacks stratified analyses targeting the high-risk neonate population with unique pharmacological responses.
A prospective cohort study was carried out at the Hospital Clínico San Carlos in Madrid (Spain), which adds that 17% neonates experienced an ADR, more than one-third of ADRs needed specific treatment, and gastrointestinal problems were the most common ADR 13 . Another study based on data from a French pharmacovigilance database specifically analyzed the distribution of drugs and affected organ systems in neonates with ADRs resulting from direct drug exposure, explicitly excluding maternal-mediated (transplacental or breastfeeding-related) exposure 14 .
Most current studies on ADRs in neonates have not adequately accounted for the various routes of drug exposure unique to this population. Neonates may be exposed to medications through three primary pathways: (1) transplacental transfer during the fetal period; (2) direct administration after birth via intravenous, oral, or other routes; and (3) indirect exposure through breastfeeding. The risk of ADR occurrence, the organ systems affected, and the spectrum of implicated drugs may differ significantly depending on the route of exposure.
Furthermore, drug exposure during early and late pregnancy has fundamentally distinct effects on the embryo and fetus. The early gestational period is critical for organ formation, during which drug exposure primarily poses a risk of teratogenic effects, leading to structural birth defects. For instance, a European study based on 2.1 million live births 15 found that first-trimester exposure to selective serotonin reuptake inhibitors (SSRIs) was significantly associated with an increased risk of various congenital heart defects. In contrast, exposure during late pregnancy, particularly in the second and third trimesters, is mainly linked to functional impairments or adaptation issues in neonates. Similarly, evidence from a Danish nationwide cohort study 16 indicated that antibiotic exposure during mid-to-late pregnancy was associated with an elevated risk of low birth weight. Therefore, the timing of drug exposure is critical in determining its impact on the fetus.
The study aims to characterize all reported ADRs associated with multiple drug exposure pathways in neonates based on data from the FDA Adverse Event Reporting System (FAERS). Furthermore, the study will perform stratified analyses according to neonatal sex, reported severity of ADRs, and mortality outcomes. These findings are expected to provide a critical foundation for the development of a scientifically sound risk assessment framework for drug use during the perinatal period.
Methods
Data source
FAERS is a publicly available database that collects AE and medication error-associated reports submitted by healthcare professionals, consumers, and drug manufacturers 17 . The FAERS database is updated quarterly and contains Individual Case Safety Reports (ICSRs), which include multiple fields such as basic patient information, drug names, routes of administration, descriptions of adverse events, sources of reports, and clinical outcomes.
In a broad sense, ADR refers to unintended responses to a drug when used at normal doses. In the study, an ADR case was defined as an individual FAERS report meeting the following criteria:
The patient population was restricted to neonates (identified in the DEMO table by an AGE value of ‘0–28’).
The report included one and only one “Primary Suspect” drug (listed in the DRUG table with a ROLE_COD of ‘PS’).
The report contained at least one adverse event (coded by a Preferred Term [PT] in the REAC table using the Medical Dictionary for Regulatory Activities, MedDRA v27.1).
Target drug screening for the neonatal population
The study population consisted of neonates (0–28 days) within the FAERS database. For each case, the unique “Primary Suspect (PS)” drug was designated as the “Target Drug.” All database drug names were then standardized to generic names using the WHO Drug Dictionary (September 2024 version), and the screening was conducted following this standardization.
Study design
In this study, FAERS data from the first quarter of 2004 to the fourth quarter of 2024 were utilized. The original ASCII-format datasets were downloaded for data mining and statistical analysis. Reports of adverse drug reactions in neonates (defined as age≤28 days) were identified, and data deduplication and report exclusion were performed in accordance with the FDA’s official guidance 18 . Adverse event terms were coded using the MedDRA v27.1. Reports related to neonatal drug exposure were extracted and analyzed to investigate the risk characteristics and potential safety signals associated with medication use during the perinatal and early postnatal periods.
Ethical compliance
This research is a secondary analysis of the de-identified FAERS database, which is publicly accessible. Since the data do not contain personal identifiers and the analysis does not involve direct human subject interaction, this study was exempt from ethical approval in accordance with the Declaration of Helsinki and other internationally recognized ethical guidelines.
Statistical analysis
This study evaluated the signal strength of various adverse events to neonates through the application of commonly utilized disproportional analysis techniques, which included the reporting advantage ratio (ROR), the proportional reporting ratio (PRR), the Bayesian Confidence Propagation Neural Network (BCPNN), and the Multi-Item Gamma Poisson Shrinker (MGPS) 19 20 21 , with the particular formulas and screening criteria employed for this analysis outlined in Table 1 . Data processing and analysis were conducted using SAS 9.4 Vision.
Table 1 Disproportionality analysis.
| Target Adverse Event | Other Adverse Events | Total | |
|---|---|---|---|
| Suspected Drug | a | b | a+b |
| Other Drugs | c | d | c+d |
| Total | a+c | b+d | N=a+b+c+d |
a: Number of ADR cases attributable to the suspected drug; b: Number of ADR cases attributable to other drugs; c: Number of other ADR cases attributable to the suspected drug; d: Number of other ADR cases attributable to other drugs.
Results
Baseline characteristics of ADR reports
Between Q1 2004 and Q4 2024, a total of 15,456 neonatal reports were identified, with 60,611 adverse event occurrences recorded. As shown in Table 2 , the male-to-female ratio was approximately 1:1.3. The median body weight was 2.87 kg (IQR: 2.08–3.38 kg). Of all reported events, 95.45% were classified as serious, and 11.08% resulted in fatal outcomes. The majority of reports originated from Germany (29.34%) and the United States (20.57%), with 81.49% submitted by healthcare professionals. The median time from drug exposure to the onset of adverse events was 231 days (IQR: 1.00–271.00 days), while the time to death among fatal cases was 2 days (IQR: 0.00–13.00 days). Analysis by exposure route revealed significant differences in the median time-to-onset of adverse drug reactions in neonates, with the following specific results: placental transmission (264.00 days, IQR: 231.00–276.00 days); routes including breast milk, intravenous, and others (1.00 day, IQR: 0.00–12.00 days); intramuscular and oral administration (2.00 days, IQR: 0.00–9.00 days); and inhalation (3.00 days, IQR: 1.00–8.00 days).
Table 2 The baseline characteristics of neonate FAERS ICSRs.
| Characteristics | Number (%) |
|---|---|
| Gender | |
| Female (%) | 5763(37.29) |
| Male (%) | 7549(48.84) |
| Not Specified (%) | 2144(13.87) |
| Age | |
| Median (Q1, Q3) | 0.00(0.00,0.02) |
| Serious report | |
| Serious (%) | 14753(95.45) |
| Non-Serious (%) | 703(4.55) |
| Reporter | |
| Physician (%) | 5835(37.75) |
| Other health professional (%) | 3324(21.51) |
| Consumer (%) | 2706(17.51) |
| Pharmacist (%) | 2290(14.82) |
| Lawyer (%) | 197(1.27) |
| Not Specified (%) | 1104(7.14) |
| Patient’s Continent | |
| Europe (%) | 28145(46.40) |
| Unknown (%) | 19628(32.43) |
| North America (%) | 8648(14.33) |
| Asia (%) | 3039(5.02) |
| South America (%) | 538(0.96) |
| Africa (%) | 379(0.61) |
| Oceania (%) | 234(0.46) |
| Outcome a | |
| Life-Threatening (%) | 1314(8.50) |
| Hospitalization – Initial or Prolonged (%) | 4665(30.18) |
| Disability (%) | 323(2.09) |
| Death (%) | 1712(11.08) |
| Congenital Anomaly (%) | 4298(27.81) |
| Required Intervention to Prevent Permanent Impairment/Damage (%) | 161(1.04) |
| Other (%) | 9701(62.77) |
| Adverse event occurrence time – Date of administration (days) | |
| Median (Q1, Q3) | 231.00(1.00,271.00) |
| Inhalation (Q1, Q3) | 3.00(1.00,8.00) |
| Intramuscular (Q1, Q3) | 2.00(0.00,9.00) |
| Intravenous (Q1, Q3) | 1.00(0.00,3.00) |
| Oral (Q1, Q3) | 2.00(0.00,12.00) |
| Other (Q1, Q3) | 1.00(0.00,6.00) |
| Transmammary (Q1, Q3) | 1.00(0.00,12.00) |
| Transplacental (Q1, Q3) | 264.00(231.00,276.00) |
| Weight (kg) | |
| Median (Q1, Q3) | 2.87(2.08,3.38) |
a: The outcome proportion is defined as the number of occurrences of a specific outcome divided by the total number of patients.
The composition of adverse events
The top 50 PT signals were categorized according to the SOC classification as shown in Fig. 1 . The most frequently reported SOC was “Injury, poisoning and procedural complications.” The PTs linked to this SOC largely consisted of fetal exposure during pregnancy (ROR>1). Despite this PT not indicating a specific ADR, outcome analysis revealed its association with serious medical events and congenital anomalies as shown in Fig. 2 . The second most frequent SOC was “Congenital, familial and genetic disorders,” accounting for 15.96% of the signals. Representative conditions included ventricular septal defect, patent ductus arteriosus, dysmorphic features, hypospadias, and cryptorchidism. Further analysis of exposure pathways delineated the origins of signals in this SOC as shown in Fig. 3 : 50% stemmed from intrauterine (placental) exposure, 20% from drugs transmitted through breastfeeding, and a proportion—equivalent to that of intrauterine exposure—originated from direct extrauterine exposure. Additionally, the SOC “Pregnancy, puerperium and perinatal conditions” constituted 8.65% of the reports, including preterm birth, small for gestational age, macrosomia, and fetal growth restriction. A notable finding was the relatively high proportion (3.88%) of all PT events associated with “off-label use” in the neonatal population. Off-label use refers to the administration of a medication outside the specifications of its officially approved labeling, necessitating further evaluation of the potential association between such usage and adverse events. As shown in Table 3 , the most frequently implicated medications in off-label use reports were propranolol, levetiracetam, ibuprofen, valganciclovir, and azithromycin.
Fig. 1.
The top 50 most frequent PTs. X-axis: represents the ROR value. Triangles: indicate that both the ROR value and its confidence interval exceed the upper limit of the axis scale. Squares: represent the point estimate of the ROR, with the ends of the line denoting the upper and lower bounds of the confidence interval. Red vertical line: corresponding to ROR=1, serving as the reference line. This study focuses on the neonatal population. ROR>1: indicates a positive association of the PT with the neonatal population, suggesting a potential safety signal. ROR=1: suggests no significant association of the PT with the neonatal population. ROR<1: implies that the reporting proportion of the PT is lower in the neonatal population than in other populations.
Fig. 2.
Outcomes by PTs. X-axis: the outcome. The sum of proportions exceeds 100% because individual patients can contribute to more than one outcome category. Y-axis: PT adverse events. The color bar (right): case numbers.
Fig. 3.
The distribution of administration by SOCs. X-axis: drug exposure routes. Y-axis: the SOCs. The color bar (right): case numbers.
Table 3 The top 10 drugs of “off-label-use”.
| Drug Name | Number (%) | Drug Name | Number (%) |
|---|---|---|---|
| Propranolol | 84(11.95%) | Paracetamol | 14(1.99%) |
| Levetiracetam | 70(9.96%) | Sildenafil | 14(1.99%) |
| Ibuprofen | 51(7.25%) | Amiodarone | 13(1.85%) |
| Valganciclovir | 21(2.29%) | Amphotericin B | 13(1.85%) |
| Azithromycin | 16(2.28%) | Nitric Oxide | 13(1.85%) |
The outcome of adverse events
Fig. 2 shows the patient outcomes associated with different PTs for adverse events. It is important to note that the MedDRA terminology system is designed for precise medical descriptions and includes a large number of PTs with similar or related meanings. As shown in Table 2 , among the various patient outcomes, the highest proportions are “hospitalization” (30.18%) and “congenital malformations” (27.81%), excluding “other serious.” A further analysis of Fig. 2 reveals that the top five PTs leading to hospitalization are as follows: fetal exposure during pregnancy, neonatal drug withdrawal syndrome, preterm infants, and atrial septal defect. The top five PTs leading to congenital malformations are: fetal exposure during pregnancy, atrial septal defect, patent ductus arteriosus, preterm infants, and small for gestational age infants. These results highlight a new and significant warning: in addition to previously concerned adverse events such as restricted fetal growth and development, congenital malformations related to the cardiovascular system also pose a significant medication risk and should be given high attention during clinical monitoring and assessment.
Distribution of drug exposure
The statistical data for the primary PTs corresponding to the top 20 drugs are shown in Table 4 . The majority were medications acting on the nervous system (34.27%) and the sensory organs (26.49%). The 10 most frequently reported drugs as the primary suspected agents were, in descending order: venlafaxine, sertraline, lamotrigine, quetiapine, citalopram, valproic acid, levetiracetam, methadone, fluoxetine, and escitalopram. To characterize the temporal pattern of ADRs following intrauterine versus extrauterine drug exposures, we analyzed the median time to ADR onset across different administration routes in Table 2 . Placental drug transfer, representing intrauterine exposure, was associated with a markedly prolonged median ADR latency of 264 days. In contrast, extrauterine exposures demonstrated substantially shorter median times (1–3 days). These results clearly demonstrate distinct temporal dynamics in ADR manifestation between intrauterine and extrauterine drug exposures, reflecting the unique pharmacokinetic and pharmacodynamic profiles of drug transfer during gestation. The reported outcomes were primarily classified as other serious medically important events and congenital anomalies, as shown in Fig. 4 . The primary drugs associated with congenital malformations were citalopram, lamotrigine, levetiracetam, escitalopram, and metoprolol. These findings suggest that neonatal adverse events are predominantly associated with utero exposure to maternal medications.
Table 4 Primary PTs of the top 20 drugs.
| Drug (Number of occurrences) | Top 5 PTs | Number (%) |
|---|---|---|
| Venlafaxine (2909) | Fetal exposure during pregnancy | 428(14.71%) |
| Atrial septal defect | 153(5.26%) | |
| Small-for-dates baby | 131(4.50%) | |
| Respiratory disorder neonatal | 125(4.30%) | |
| Selective eating disorder | 100(3.44%) | |
| Sertraline (2725) | Fetal exposure during pregnancy | 367(13.47%) |
| Atrial septal defect | 121(4.44%) | |
| Small-for-dates baby | 103(3.78%) | |
| Respiratory disorder neonatal | 96(3.52%) | |
| Agitation neonatal | 69(2.53%) | |
| Lamotrigine (2515) | Fetal exposure during pregnancy | 412(16.39%) |
| Atrial septal defect | 136(5.41%) | |
| Respiratory disorder neonatal | 79(3.14%) | |
| Selective eating disorder | 68(2.70%) | |
| Small-for-dates baby | 57(2.27%) | |
| Quetiapine (2412) | Fetal exposure during pregnancy | 378(15.67%) |
| Small-for-dates baby | 120(4.98%) | |
| Respiratory disorder neonatal | 114(4.73%) | |
| Atrial septal defect | 109(4.52%) | |
| Drug withdrawal syndrome neonatal | 66(2.74%) | |
| Citalopram(2218) | Fetal exposure during pregnancy | 398(17.94%) |
| Small-for-dates baby | 151(6.81%) | |
| Respiratory disorder neonatal | 108(4.87%) | |
| Atrial septal defect | 108(4.87%) | |
| Ventricular septal defect | 81(3.65%) | |
| Valproic acid (1873) | Exposure during pregnancy | 94(5.02%) |
| Fetal anticonvulsant syndrome | 92(4.91%) | |
| Dysmorphism | 65(3.47%) | |
| Fetal exposure during pregnancy | 62(3.31%) | |
| Developmental delay | 46(2.46%) | |
| Levetiracetam (1553) | Fetal exposure during pregnancy | 233(14.91%) |
| Small-for-dates baby | 85(5.44%) | |
| Off label use | 69(4.41%) | |
| Drug ineffective | 69(4.41%) | |
| Atrial septal defect | 52(3.33%) | |
| Methadone (1404) | Drug withdrawal syndrome neonatal | 519(36.97%) |
| Fetal exposure during pregnancy | 441(31.41%) | |
| Premature baby | 120(8.55%) | |
| Maternal exposure during pregnancy | 65(4.63%) | |
| Maternal drugs affecting fetus | 24(1.71%) | |
| Fluoxetine (1343) | Fetal exposure during pregnancy | 121(9.00%) |
| Exposure during pregnancy | 65(4.84%) | |
| Premature baby | 44(3.27%) | |
| Small-for-dates baby | 38(2.83%) | |
| Drug withdrawal syndrome neonatal | 35(2.61%) | |
| Escitalopram (1168) | Fetal exposure during pregnancy | 250(21.40%) |
| Small-for-dates baby | 118(10.10%) | |
| Atrial septal defect | 71(6.08%) | |
| Premature baby | 39(3.33%) | |
| Drug withdrawal syndrome neonatal | 39(3.33%) | |
| Paroxetine (1071) | Fetal exposure during pregnancy | 131(12.23%) |
| Atrial septal defect | 43(4.01%) | |
| Drug withdrawal syndrome neonatal | 24(2.24%) | |
| Patent ductus arteriosus | 21(1.96%) | |
| Maternal drugs affecting fetus | 17(1.59%) | |
| Olanzapine (858) | Fetal exposure during pregnancy | 93(10.84%) |
| Exposure during pregnancy | 46(5.36%) | |
| Selective eating disorder | 32(3.73%) | |
| Small-for-dates baby | 26(3.03%) | |
| Atrial septal defect | 26(3.03%) | |
| Metoprolol (799) | Fetal exposure during pregnancy | 150(18.77%) |
| Small-for-dates baby | 86(10.76%) | |
| Atrial septal defect | 57(7.13%) | |
| Hypoglycemia neonatal | 31(3.88%) | |
| Hypospadias | 24(3.00%) | |
| Indometacin (730) | Neonatal disorder | 31(4.24%) |
| Gastrointestinal perforation | 15(2.05%) | |
| Drug ineffective | 15(2.05%) | |
| Oliguria | 14(1.92%) | |
| Renal failure | 12(1.64%) | |
| Aripiprazole (702) | Fetal exposure during pregnancy | 174(24.78%) |
| Premature baby | 21(2.99%) | |
| Hypospadias | 18(2.56%) | |
| Small-for-dates baby | 15(2.14%) | |
| Large-for-dates baby | 13(1.85%) | |
| Nitric oxide (695) | Patent ductus arteriosus | 36(5.18%) |
| Intraventricular hemorrhage neonatal | 30(4.32%) | |
| Neonatal disorder | 27(3.88%) | |
| Oxygen saturation decreased | 23(3.33%) | |
| Ibuprofen (619) | Off-label use | 40(6.46%) |
| Intestinal perforation | 35(5.65%) | |
| Pulmonary hypertension | 15(2.42%) | |
| Product use issue | 15(2.42%) | |
| Oliguria | 13(2.10%) | |
| Mirtazapine (465) | Fetal exposure during pregnancy | 81(17.42%) |
| Atrial septal defect | 25(5.38%) | |
| Small-for-dates baby | 22(4.73%) | |
| Drug withdrawal syndrome neonatal | 19(4.09%) | |
| Patent ductus arteriosus | 17(3.66%) |
Fig. 4.
Outcome of the top 50 reported drugs. X-axis: primary suspected drug. Y-axis: the outcome. The color bar (right): case numbers.
Stratified analysis
We conducted a comparative analysis based on route of administration, sex, and severity of reported adverse events. Among these factors, the route of administration emerged as a key determinant of neonatal drug exposure patterns and associated adverse outcomes, as shown in Fig. 5 . Transplacental (i. e., intrauterine exposure) accounted for the highest proportion of cases (52.47%), followed by intravenous injection (9.34%), oral administration (6.77%), transmammary (1.80%), intramuscular injection (1.48%), and inhalation (1.29%). Notably, venlafaxine, sertraline, quetiapine, lamotrigine, and levetiracetam consistently ranked among the top five drugs involved in both intrauterine and lactational exposures.
Fig. 5.
The administration of adverse events. X-axis: drug exposure routes. Y-axis: case numbers.
A sex-difference analysis was conducted to ascertain the association patterns between certain adverse events and gender. A complete case analysis was performed, excluding the 14% of cases with missing gender data. As shown in Fig. 6 , conditions such as microcephaly, ventricular septal defect, and small-for-dates baby were reported more frequently among female neonates, suggesting a higher susceptibility to craniofacial and cardiac developmental abnormalities as well as intrauterine growth restriction. In contrast, male neonates demonstrated a greater risk of urogenital malformations, including hypospadias and cryptorchidism. Additionally, the incidence of neonatal irritability was also higher in males, which may be attributed to sex-related differences in hormonal regulation or neurodevelopmental sensitivity.
Fig. 6.
Volcano plot comparing male and female groups. X-axis: fold change (Log 2 ROR). Larger absolute values indicate a greater magnitude of difference between groups. Y-axis: p -value, with p -value<0.05 considered statistically significant. The horizontal gray dashed line: points above it represent p <0.05.
Analysis of the severity of reported adverse events revealed notable differences across exposure types and clinical conditions as shown in Fig. 7 . Drug exposures via breastfeeding were predominantly associated with non-serious reports, whereas prenatal drug use, atrial septal defects, and neonatal withdrawal syndrome were more frequently linked to serious adverse event reports. These findings suggest that in utero exposure during critical periods of fetal development carries a substantially higher risk of severe outcomes compared to lactational exposure. This may be attributed to the greater vulnerability of organogenesis and central nervous system development during gestation, as well as the higher systemic drug concentrations typically associated with transplacental transfer.
Fig. 7.
Volcano plot comparing non-serious and serious groups. X-axis: fold change (Log 2 ROR). Larger absolute values indicate a greater magnitude of difference between groups. Y-axis: p -value, with p -value<0.05 considered statistically significant. The horizontal gray dashed line: points above it represent p <0.05.
Discussion
Due to the challenges in conducting clinical trials in the neonatal population, many drug labels lack specific information regarding neonatal use. Therefore, computer-based risk signal detection for neonatal drug safety assessment is particularly critical. This study analyzed reports of ADRs in neonates resulting from both maternal and direct drug exposure, utilizing data from the FAERS database spanning the first quarter of 2004 to the fourth quarter of 2024. Among the target population of 15,456 neonatal patients, a total of 60,611 adverse events were reported. Strikingly, 95.45% of these events were classified as serious, and 11.08% of reports documented a fatal outcome. These findings underscore the urgent need to address perinatal medication safety, particularly in the highly sensitive neonatal population. Furthermore, a marked increase in neonatal ADR reports has been observed since 2014. This trend may be associated with multiple factors, including shifts in prescribing patterns, incidence rates of adverse events, the number of newly approved drugs, advancements in industry-funded initiatives, enhanced public awareness, and evolving regulatory policies 22 23 24 25 . The study further analyzed the geographic distribution bias of neonatal adverse drug reaction reports by continent. The vast majority of reports originated from Europe (42.64%) and North America (14.09%), which highlights significant global disparities in neonatal pharmacovigilance capacity. The high reporting rates from developed regions reflect their well-established monitoring networks and standardized reporting protocols, providing actionable models for regions with less mature surveillance systems to reference and adapt.
Given that the majority of neonatal ADRs occur within the NICU, 81.49% of reports were submitted by healthcare professionals, resulting in generally more comprehensive and standardized documentation. Additionally, literature reviews indicate that pediatric ADRs predominantly occur within the first few years of life 11 26 27 28 29 , which aligns with our findings. Moreover, the study identified orders of magnitude differences in reaction times between in utero (264.00 days) and extrauterine (1.00–3.00 days) drug exposure in neonates, reflecting distinct mechanisms of “delayed developmental effects” versus “acute toxic reactions.” This underscores the need for route-specific monitoring windows in neonatal drug safety assessment.
The results of this study indicate that SOC categories involved in neonatal ADRs are predominantly concentrated in congenital disorders and perinatal conditions, accounting for approximately 43.25% of all cases. Current research data suggest that 65%–75% of birth defects are of unknown origin and are generally thought to arise from multifactorial interactions between genetic predispositions and environmental influences. The most common identifiable genetic etiologies are single-gene disorders (15%–20%) and chromosomal abnormalities (5%). Approximately 10% of birth defects are linked to environmental exposures, including pharmaceuticals, maternal conditions, infectious pathogens, electromagnetic radiation, and environmental contaminants 30 31 . It is imperative to recognize that pharmaceutical exposures hold a unique status among these etiologies due to their potential for prevention through careful medication management before and during pregnancy.
Drugs associated with congenital disorders and perinatal conditions were mainly those targeting the nervous system, representing 34.27%. Among different routes of administration, transplacental exposure accounted for the highest proportion. Notably, antidepressants were the most frequently implicated drugs, regardless of whether the exposure occurred in utero or postnatally. This finding differs markedly from the ADR profiles observed in children and adolescents. Previous studies have shown that ADRs in pediatric populations are primarily associated with “General disorders and administration site conditions” and “Nervous system disorders,” with anti-infective agents being the most commonly implicated drug class 12 .
According to published literature, depression during pregnancy is relatively prevalent, with an overall incidence rate as high as 39.0%. Approximately 9.02% of pregnant women experience moderate depression, and the incidence of postpartum depression ranges between 10% and 15% 32 33 . Most antidepressants exert their therapeutic effects by modulating the levels of serotonin, norepinephrine, and dopamine in the body. From a pharmacokinetic perspective, selective serotonin reuptake inhibitors (SSRIs) and serotonin-norepinephrine reuptake inhibitors (SNRIs) can cross the placental barrier and be excreted into breast milk, thereby directly affecting neonatal development 34 . The timing of drug exposure is a key determinant for fetal and neonatal outcomes, necessitating a differential risk assessment between first-trimester teratogenicity and late-pregnancy neonatal adaptation problems. A cohort study conducted in Canada further demonstrated that infants born to women who used serotonergic antidepressants during early pregnancy were at an increased risk of developing structural abnormalities in the cardiovascular, musculoskeletal, and respiratory systems 35 . Therefore, in the neonatal population, special attention should be given to maternal exposure to antidepressants during early pregnancy.
In neonatal ADRs, intravenous drug exposure accounts for 9.34%, ranking only behind transplacental and breast milk-mediated exposures. The most frequently implicated intravenous medications include levetiracetam, zidovudine, ibuprofen, and indomethacin. Notably, among anti-infective agents, antiretroviral drugs used for HIV treatment pose the highest risk. Zidovudine, a first-line antiretroviral agent for the prevention of mother-to-child transmission of HIV 36 , is primarily associated with adverse effects involving the hematologic and gastrointestinal systems 37 . Additionally, ibuprofen and indomethacin are commonly used to treat patent ductus arteriosus in preterm infants; however, both are administered off-label in this population. These drugs may cause serious adverse effects, including neonatal renal impairment, necrotizing enterocolitis, and gastrointestinal perforation 38 . Off-label drug use is widely recognized as a major contributor to ADRs in pediatric clinical practice, and such use is even more prevalent among neonates. Therefore, the clinical use of these medications in neonates should be carried out with caution, ensuring rational prescribing practices and enhanced pharmacovigilance.
There are certain limitations to this study. Although the database offers advantages such as a large sample size and broad coverage, several limitations remain. First, as a spontaneous reporting system, FAERS is subject to underreporting, incomplete information, and challenges in establishing definitive causal relationships. Second, the absence of key clinical variables such as birth weight, gestational age, and specific dosage limits the ability to conduct an in-depth analysis of ADR risk factors.
Conclusions
The study presents a distinctive overview of ADRs associated with neonatal drug exposure. Maternal use of antidepressants during pregnancy was identified as a prominent risk factor for neonatal ADRs. Moreover, particular attention should be paid to ADRs resulting from direct neonatal exposure to antiretroviral agents, antiepileptic drugs, and medications used for patent ductus arteriosus closure, as the majority of these drugs are administered off-label in this vulnerable population. Given these findings, predictive models for placental permeability of antidepressants and fetal risk assessment tools can be developed to optimize risk management of medication use during pregnancy. Furthermore, this study highlights the urgent need to address the systemic challenges neonates face within global drug safety monitoring systems. We advocate for the creation of an international pharmacovigilance network specifically for neonates. Its efforts should be directed towards standardizing reporting requirements, establishing robust mechanisms for the transnational exchange of anonymized data, and implementing advanced signal detection methodologies to uncover subtle risks in this distinct patient group.
Acknowledgement
We extend our sincere gratitude to all the participants involved in this study.
Footnotes
Conflict of Interest The authors declare that they have no conflict of interest.
Contributorsʼ Statement Zhuqing Yang: Data curation, Formal analysis, Investigation, Methodology, Writing – original draft, Writing – review & editing. Meng Hou: Formal analysis, Validation, Visualization, Writing – original draft, Writing – review & editing. Tingting Li: Conceptualization, Supervision, Validation, Writing – original draft, Writing – review & editing.
References
- 1.Newborn mortality. WHO. Available online:https://www.who.int/zh/news-room/fact-sheets/detail/newborn-mortality [Google Scholar]
- 2.Campion S, Inselman A, Hayes B et al. The benefits, limitations and opportunities of preclinical models for neonatal drug development. Dis Model Mech. 2022;15 doi: 10.1242/dmm.049065. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Al-Turkait A, Szatkowski L, Choonara I, Ojha S. Drug utilisation in neonatal units in England and Wales: a national cohort study. Eur J Clin Pharmacol. 2022;78:669–677. doi: 10.1007/s00228-021-03267-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Gade C, Trolle S, Mørk M L et al. Massive presence of off-label medicines in Danish neonatal departments: A nationwide survey using national hospital purchase data. Pharmacol Res Perspect. 2023;11:e01037. doi: 10.1002/prp2.1037. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Stark A, Smith P B, Hornik C P et al. Medication use in the neonatal intensive care unit and changes from 2010 to 2018. J Pediatr. 2022;240:66–710000. doi: 10.1016/j.jpeds.2021.08.075. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Alghamdi A A, Keers R N, Sutherland A, Ashcroft D M. Prevalence and nature of medication errors and preventable adverse drug events in paediatric and neonatal intensive care settings: A systematic review. Drug Saf. 2019;42:1423–1436. doi: 10.1007/s40264-019-00856-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Hawcutt D B, O’Connor O, Turner M A. Adverse drug reactions in neonates: could we be documenting more? Expert Rev Clin Pharmacol. 2014;7:807–820. doi: 10.1586/17512433.2014.956090. [DOI] [PubMed] [Google Scholar]
- 8.Brameli A, Yuan I H, Phillips E J, Stone C A. Pediatric drug-induced anaphylaxis reports in the FDA Adverse Event Reporting System (FAERS) J Allergy Clin Immunol Pract. 2024;12:2506–25090. doi: 10.1016/j.jaip.2024.05.042. [DOI] [PubMed] [Google Scholar]
- 9.Rani N. Pattern of adverse drug reactions among pregnant women and pediatric patients in a tertiary care hospital. Curr Drug Saf. 2023;18:190–195. doi: 10.2174/1574886317666220404105653. [DOI] [PubMed] [Google Scholar]
- 10.Joyner L M, Alicea L A, Goldman J L et al. Methods for Detecting Pediatric Adverse Drug Reactions From the Electronic Medical Record. J Clin Pharmacol. 2021;61:1479–1484. doi: 10.1002/jcph.1916. [DOI] [PubMed] [Google Scholar]
- 11.Leporini C, De Sarro C, Palleria C et al. Pediatric drug safety surveillance: A 10-year analysis of adverse drug reaction reporting data in Calabria, Southern Italy. Drug Saf. 2022;45:1381–1402. doi: 10.1007/s40264-022-01232-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Phan M, Cheng C, Dang V et al. Characterization of pediatric reports in the US Food and Drug Administration Adverse Event Reporting System from 2010–2020: A cross-sectional study. Ther Innov Regul Sci. 2023;57:1062–1073. doi: 10.1007/s43441-023-00542-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Belén Rivas A, Arruza L, Pacheco E et al. Adverse drug reactions in neonates: a prospective study. Arch Dis Child. 2016;101:371–376. doi: 10.1136/archdischild-2015-309396. [DOI] [PubMed] [Google Scholar]
- 14.Kaguelidou F, Beau-Salinas F, Jonville-Bera A P, Jacqz-Aigrain E. Neonatal adverse drug reactions: an analysis of reports to the French pharmacovigilance database. Br J Clin Pharmacol. 2016;82:1058–1068. doi: 10.1111/bcp.13034. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Wemakor A, Casson K, Garne E et al. Selective serotonin reuptake inhibitor antidepressant use in first trimester pregnancy and risk of specific congenital anomalies: a European register-based study. Eur J Epidemiol. 2015;30:1187–1198. doi: 10.1007/s10654-015-0065-y. [DOI] [PubMed] [Google Scholar]
- 16.Tomar N, Uldbjerg C S, Bech B H et al. Prenatal antibiotic exposure and birth weight. Pediatr Obes. 2022;17:e12831. doi: 10.1111/ijpo.12831. [DOI] [PubMed] [Google Scholar]
- 17.Harpaz R, DuMochel W, Shah N H. Big data and adverse drug reaction detection. Clin Pharmacol Ther. 2015;99:268–270. doi: 10.1002/cpt.302. [DOI] [PubMed] [Google Scholar]
- 18.Sakaeda T, Tamon A, Kadoyama K, Okuno Y. Data mining of the public version of the FDA Adverse Event Reporting System. Int J Med Sci. 2013;10:796–803. doi: 10.7150/ijms.6048. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Bate A, Evans SJ W. Quantitative signal detection using spontaneous ADR reporting. Pharmacoepidemiol Drug Saf. 2009;18:427–436. doi: 10.1002/pds.1742. [DOI] [PubMed] [Google Scholar]
- 20.van Puijenbroek E P, Bate A, Leufkens HG M et al. A comparison of measures of disproportionality for signal detection in spontaneous reporting systems for adverse drug reactions. Pharmacoepidemiol Drug Saf. 2002;11:3–10. doi: 10.1002/pds.668. [DOI] [PubMed] [Google Scholar]
- 21.Evans SJ W, Waller P C, Davis S. Use of proportional reporting ratios (PRRs) for signal generation from spontaneous adverse drug reaction reports. Pharmacoepidemiol Drug Saf. 2001;10:483–486. doi: 10.1002/pds.677. [DOI] [PubMed] [Google Scholar]
- 22.Harinstein L, Kalra D, Kortepeter C M et al. Evaluation of postmarketing reports from industry-sponsored programs in drug safety surveillance. Drug Saf. 2018;42:649–655. doi: 10.1007/s40264-018-0759-7. [DOI] [PubMed] [Google Scholar]
- 23.Marwitz K, Jones S C, Kortepeter C M et al. An Evaluation of postmarketing reports with an outcome of death in the US FDA Adverse Event Reporting System. Drug Saf. 2020;43:457–465. doi: 10.1007/s40264-020-00908-5. [DOI] [PubMed] [Google Scholar]
- 24.Muñoz M A, Dal Pan G J. The impact of litigation-associated reports on signal identification in the US FDA’s Adverse Event Reporting System. Drug Saf. 2019;42:1199–1201. doi: 10.1007/s40264-019-00834-1. [DOI] [PubMed] [Google Scholar]
- 25.New Molecular Entity (NME) Drug and New Biologic Approvals. US Food and Drug Administration. Available online:https://www.fda.gov/drugs/nda-and-bla-approvals/new-molecular-entity-nme-drug-and-new-biologic-approvals
- 26.Cliff-Eribo K O, Sammons H, Choonara I. Systematic review of paediatric studies of adverse drug reactions from pharmacovigilance databases. Expert Opin Drug Saf. 2016;15:1321–1328. doi: 10.1080/14740338.2016.1221921. [DOI] [PubMed] [Google Scholar]
- 27.Dernoncourt A, Liabeuf S, Bennis Y et al. Fetal and neonatal adverse drug reactions associated with biologics taken during pregnancy by women with autoimmune diseases: Insights from an analysis of the World Health Organization Pharmacovigilance Database (VigiBase®) BioDrugs. 2022;37:73–87. doi: 10.1007/s40259-022-00564-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Allegaert K, van den Anker J. Dose-related adverse drug events in neonates: Recognition and assessment. J Clin Pharmacol. 2021;61:S-152–S160. doi: 10.1002/jcph.1827. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Yalcin N, van den Anker J, Samiee-Zafarghandy S, Allegaert K. Drug related adverse event assessment in neonates in clinical trials and clinical care. Expert Rev Clin Pharmacol. 2024;17:803–816. doi: 10.1080/17512433.2024.2390927. [DOI] [PubMed] [Google Scholar]
- 30.Harris B S, Bishop K C, Kemeny H R et al. Risk factors for birth defects. Obstet Gynecol Surv. 2017;72:123–135. doi: 10.1097/OGX.0000000000000405. [DOI] [PubMed] [Google Scholar]
- 31.Corsello G, Giuffre M. Congenital malformations. J Matern Fetal Neonatal Med. 2012;25:25–29. doi: 10.3109/14767058.2012.664943. [DOI] [PubMed] [Google Scholar]
- 32.O’Hara M W, McCabe J E. Postpartum depression: Current status and future directions. Annu Rev Clin Psychol. 2013;9:379–407. doi: 10.1146/annurev-clinpsy-050212-185612. [DOI] [PubMed] [Google Scholar]
- 33.Miller ES M, Metz T, Moore Simas T et al. Treatment and management of mental health conditions during pregnancy and postpartum: ACOG Clinical Practice. Guideline No. 5. Obstet Gynecol. 2023;141:1262–1288. doi: 10.1097/AOG.0000000000005202. [DOI] [PubMed] [Google Scholar]
- 34.Perić M, Bečeheli I, Čičin-Šain L et al. Serotonin system in the human placenta – the knowns and unknowns. Front Endocrinol (Lausanne) 2022;13:1.061317E6. doi: 10.3389/fendo.2022.1061317. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Horackova H, Karahoda R, Cerveny L et al. Effect of selected antidepressants on placental homeostasis of serotonin: Maternal and fetal perspectives. Pharmaceutics. 2021;13:1306. doi: 10.3390/pharmaceutics13081306. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Mandelbrot L, Berrebi A, Rouzioux C et al. Reproductive options for people living with HIV: 2013 guidelines from the F. Gynecol Obstet Fertil. 2014;42:543–550. doi: 10.1016/j.gyobfe.2014.05.019. [DOI] [PubMed] [Google Scholar]
- 37.Hundscheid T, Onland W, Kooi EM W et al. Expectant management or early ibuprofen for patent ductus arteriosus. N Engl J Med. 2023;388:980–990. doi: 10.1056/NEJMoa2207418. [DOI] [PubMed] [Google Scholar]
- 38.Mitra S, Florez I D, Tamayo M E et al. Association of placebo, indomethacin, ibuprofen, and acetaminophen with closure of hemodynamically significant patent ductus arteriosus in preterm infants. JAMA. 2018;319:1221–1238. doi: 10.1001/jama.2018.1896. [DOI] [PMC free article] [PubMed] [Google Scholar]