Abstract
OBJECTIVE
To investigate whether maternal diabetes in pregnancy was associated with altered neonatal global IgG repertoire and early-life infections in offspring.
RESEARCH DESIGN AND METHODS
This study included 2,702 mother-infant pairs enrolled at birth and followed longitudinally at the Boston Medical Center. Maternal diabetes and infant infections were extracted from electronic medical records. Cord blood IgG antibodies against a wide range of microbes were quantified using Phage ImmunoPrecipitation Sequencing.
RESULTS
Overall, 327 infants (12.1%) were born to mothers with gestational diabetes mellitus (GDM) and 138 (5.1%) to mothers with pregestational diabetes mellitus (PDM). Of these, 416 infants (15.4%) and 1,425 infants (52.7%) had at least one infection in the neonatal period and the first 6 months of life, respectively. Compared with no diabetes, both maternal GDM (risk ratio [RR] 1.20, 95% CI 1.09–1.32) and PDM (RR 1.28, 95% CI 1.12–1.47) were significantly associated with an elevated risk of infections in infants during the first 6 months. These associations were particularly pronounced among infants born preterm, delivered via cesarean section, or with lower IgG repertoire diversity. Additionally, PDM was associated with a lower newborn’s global IgG repertoire diversity, compared with no diabetes, with the effect more marked among infants whose mothers had prepregnancy overweight or obesity.
CONCLUSIONS
This study provides strong evidence of an increased infection risk in the infants of mothers with diabetes and a reduced IgG repertoire diversity in those of PDM mothers. Lower IgG diversity exacerbated the diabetes-infection link. These findings suggest that maternal metabolic conditions may impact an infant’s passive immunity and susceptibility to infections.
Graphical Abstract
Introduction
The prevalence of diabetes in pregnancy has risen globally alongside the continuously rising number of people living with diabetes and obesity (1). Maternal diabetes in pregnancy has adverse impacts on both mother’s and offspring’s health (2). Diabetes has been linked to compromised function of the immune system, which, in turn, can lead to high risk of infections and a worse outcome from infections (3,4). Given that neonates rely on maternally derived antibodies for protection against infections (5,6), maternal compromised immune function may put her baby into a high risk of infections. In fact, a few studies reported that infants born to mothers with diabetes were disproportionally affected by infections (7,8). However, the data are limited, particularly regarding the impact of maternal diabetes on placental IgG antibody transfer.
After birth, newborns are suddenly exposed to a variety of pathogens that are prevalent in the extrauterine environment at a time when the immune system of neonates is immature and relatively incompetent (9). As a result, newborns and infants younger than 6 months of age are disproportionally vulnerable to infectious diseases (10,11). Maternal antibodies that are transmitted to the fetus across the placenta and in breast milk have been recognized to provide the newborn with passive immunity against infectious agents during the first months of life (5,6). Indeed, maternal immunizations during pregnancy have been confirmed to provide protection for infants from infections (12). However, it is unknown whether maternal diabetes may have an impact on neonatal antibody repertoire and which, in turn, was associated with early life infections. Further research into this field should improve our understanding of immunity in early life and will inform the development of optimal prevention strategies.
Using the Boston Birth Cohort (BBC), a well-established racially and ethnically diverse birth cohort, we aimed to examine whether maternal diabetes in pregnancy was associated with the risk of offspring’s infections in the first 6 months of life. We also quantified cord blood IgG antibodies against a large array of infectious agents, including viral, bacterial, fungal, parasitic, and toxin proteins using a Phage ImmunoPrecipitation Sequencing (PhIP-Seq) technology. Finally, using this wealth of IgG antibody data, we explored whether maternal diabetes in pregnancy can alter the neonatal global IgG repertoire and how the alteration links to the risk of infections.
Research Design and Methods
Study Population
The study participants came from the Children Heath Study of the BBC, an ongoing prospective longitudinal birth cohort study, which recruits mother-infant pairs at birth at Boston Medical Center (Boston, MA) as described previously (13). By 2023, 3,416 mother-infant pairs had been enrolled and followed from birth up to the maximum of 21 years. Of these mother-infant pairs, 140 pairs were excluded because they had not entered the database yet (n = 3) and had no documented clinical encounters (n = 137). In addition, 572 mother-infant pairs of which the infants were born in 2002 or earlier were excluded because the electronic medical record (EMR) was not available during this period. Additionally, two mother-infant pairs who were infected with HIV were also excluded. Finally, this study included 2,702 mother-infant pairs. Cord blood IgG antibody profiles were quantified in a subset (n = 971). An exploratory analysis was performed in this subset. The characteristics of this subset are similar to the total study participants (Supplementary Table 1). The number of study participants and the attrition is shown in Supplementary Fig. 1. The study protocol was approved by the Boston University Medical Center (Boston, MA) and Johns Hopkins Bloomberg School of Public Health (Baltimore, MD) Institutional Review Boards. Written informed consent was obtained from all of the study mothers.
Determination of Diabetes in Pregnancy
Pregestational diabetes mellitus (PDM) and gestational diabetes mellitus (GDM) were defined based on physician diagnosis, laboratory testing, and glucose-lowering medicine prescription, as detailed previously (14).
Ascertainment of Infections in the First 6 Months of Life
Infant infections in the first 6 months of life were defined based on physician diagnoses. ICD-9 and ICD-10 diagnosis codes were abstracted from the EMR. The presence of infection was established with ICD codes that are listed in any infection category in Supplementary Table 2. The total number of infections in the first 6 months of life was counted by different infection categories within 30 days or the same categories over 30 days to avoid overcounting the number of infections. For example, if an infant had more than one diagnosis in the same infection categories within 30 days, one infection was counted.
Quantification of IgG Antibodies
Umbilical cord blood was collected at birth. Plasma was stored in a −80°C freezer. Plasma pathogen-specific IgG antibodies were quantified using the PhIP-Seq technology in Johns Hopkins Medical Center (Baltimore, MD). Two peptide libraries (VirScan and ToxScan) were used to quantify specific IgG antibodies against a large array of viral, bacterial, fungal, parasitic, and toxin proteins (15). The detailed laboratory procedure has been published previously (16,17). A brief description of the quantification of IgG antibody is provided in the Supplementary Material.
Definition of Maternal and Perinatal Characteristics
At enrollment, each mother was assessed for prepregnancy weight, height, race and ethnicity, education, smoking, and parity by a questionnaire interview. Prepregnancy BMI was calculated as prepregnancy weight in kilograms divided by height in meters squared and further dichotomized into two groups: no overweight or obesity (non-OWO, BMI <25 kg/m2) and overweight or obesity (OWO, BMI ≥25 kg/m2) (18). Race and ethnicity was based on maternal response to fixed categories in the questionnaire and regrouped as non-Hispanic Black versus other. Education attainment was grouped into high school or less versus college or above. Maternal parity was classified as primiparous versus multiparous. Maternal smoking during pregnancy was classified into two groups: nonsmoker versus smoker (19). Birth complications were abstracted from EMR. The mode of birth was categorized into cesarean section (C-section) or vaginal birth. Gestational age at birth was assessed based on both the first day of the last menstrual period and early prenatal ultrasonographic results, as described previously (19), further grouped into term birth (≥37 weeks) and preterm birth (<37 weeks). Low birth weight was defined as birth weight <2,500 g (20). Maternal blood glucose markers during pregnancy were abstracted from the EMR. Optimal glycemic control was defined as average random blood glucose <120 mg/dL and/or average fasting blood glucose less <95 mg/dL and/or A1C <6%. Information about breastfeeding, day care attendance, and living condition was gathered in postnatal follow-up visits. The correlation matrix of these covariates with maternal diabetes in pregnancy and infant infection is presented in Supplementary Fig. 2.
Statistical Analysis
At first, to reduce false positives, we filtered out any peptides that were not enriched in at least 10 (1%) of the samples. The diversity of IgG repertoire was defined at three levels: peptide, species, and protein. At each level, we summarized the total number of epitopes hit, reactive species, and proteins for each study participant. The definition of species took into account potential cross-reactivity, which was addressed using an approach comparable to that described by Xu et al. (16).
Poisson regression with a robust variance estimator was applied to analyze the associations of maternal diabetes, IgG antibody repertoire diversity, and blood glucose markers with infant infections. A multinomial logistic regression model was used to examine the associations of maternal diabetes with categories of infant infection episodes (without infection, one infection episode, at least two infection episodes). To identify whether GDM or PDM were associated with IgG level and IgG repertoire diversity, multivariable linear regression models were used with adjusting for relevant covariates. IgG level was logarithmically transformed before being used to build model.
In addition, we also explored the synergistic effects between maternal diabetes and prepregnancy OWO, preterm birth, and C-section on infant infections and IgG antibody repertoire diversity. Effect modification by an effect modifier (prepregnancy OWO, preterm birth, and C-section) was examined using stratified analysis. The interaction between maternal diabetes and effect modifier was tested by including maternal diabetes, modifier, and their cross-product term in the regression model.
We performed all analyses using RStudio 2025.05.0.496 software (PBC, Boston, MA), and all statistical tests were two sided. We calculated the false discovery rate using the Benjamini-Hochberg method to correct for multiple testing, and a corrected P < 0.05 was considered significant.
Results
Characteristics of Study Population
Of 2,702 total study infants, 327 were born to mothers with GDM, and 138 were born to mothers with PDM. Mean maternal age at delivery was 28.8 (SD 6.5) years. Mothers with PDM were oldest, followed by mothers with GDM and then mothers without diabetes (Table 1). Maternal prepregnancy OWO, multiparous, C-section, preterm birth, and low birth weight were more prevalent in mothers with diabetes in pregnancy. Infants born to mothers with diabetes were also more likely to have higher rates as well as total numbers of infections.
Table 1.
Characteristics of total study population
| Maternal diabetes during pregnancy | |||||
|---|---|---|---|---|---|
| Characteristics | Overall (N = 2,702) | No diabetes (n = 2,237) | GDM (n = 327) | PDM (n = 138) | P value |
| Maternal characteristics | |||||
| Age, mean (SD) (years) | 28.6 (6.5) | 28.2 (6.4) | 29.9 (6.4) | 31.7 (6.5) | <0.001 |
| Race and ethnicity | 0.387 | ||||
| Non-Hispanic Black | 1,522 (56.3) | 1,258 (56.2) | 179 (54.7) | 85 (61.6) | |
| Other | 1,180 (43.7) | 979 (43.8) | 148 (45.3) | 53 (38.4) | |
| Education attainment | 0.357 | ||||
| High school or less | 1,705 (63.1) | 1,423 (63.6) | 202 (61.8) | 80 (58.0) | |
| College or above | 997 (36.9) | 814 (36.4) | 125 (38.2) | 58 (42.0) | |
| Smoking status | 0.691 | ||||
| Nonsmoker | 2,211 (81.8) | 1,837 (82.1) | 263 (80.4) | 111 (80.4) | |
| Smoker | 491 (18.2) | 400 (17.9) | 64 (19.6) | 27 (19.6) | |
| Parity | |||||
| Primiparous | 1,178 (43.6) | 1,011 (45.2) | 118 (36.1) | 49 (35.5) | 0.001 |
| Multiparous | 1,524 (56.4) | 1,226 (54.8) | 209 (63.9) | 89 (64.5) | |
| Prepregnancy OWO | 1,411 (52.2) | 1,079 (48.2) | 219 (67.0) | 113 (81.9) | <0.001 |
| Mode of birth | <0.001 | ||||
| Vaginal birth | 1,713 (63.4) | 1,517 (67.8) | 139 (42.5) | 57 (41.3) | |
| C-section | 989 (36.6) | 720 (32.2) | 188 (57.5) | 81 (58.7) | |
| Child’s characteristics | |||||
| Boys | 1,370 (50.7) | 1,130 (50.5) | 169 (51.7) | 71 (51.4) | 0.910 |
| Preterm birth | 769 (28.5) | 558 (24.9) | 149 (45.6) | 62 (44.9) | <0.001 |
| Low birth weight | 746 (27.6) | 576 (25.7) | 130 (39.8) | 40 (29.0) | <0.001 |
| Breastfeeding status | 0.381 | ||||
| Formula only | 595 (22.0) | 486 (21.7) | 75 (22.9) | 34 (24.6) | |
| Breastfeed exclusively | 604 (22.4) | 513 (22.9) | 65 (19.9) | 26 (18.8) | |
| Both | 1,119 (41.4) | 914 (40.9) | 140 (42.8) | 65 (47.1) | |
| Missing | 384 (14.2) | 324 (14.5) | 47 (14.4) | 13 (9.4) | |
| Infection in the first 6 months | 1,425 (52.7) | 1,140 (51.0) | 196 (59.9) | 89 (64.5) | <0.001 |
| Infection in the first month | 416 (15.4) | 300 (13.4) | 85 (26.0) | 31 (22.5) | <0.001 |
| No. of infections in the first 6 months | 0.001 | ||||
| 0 | 1,277 (47.3) | 1,097 (49.0) | 131 (40.1) | 49 (35.5) | |
| 1 | 769 (28.5) | 611 (27.3) | 114 (34.9) | 44 (31.9) | |
| ≥2 | 656 (24.3) | 529 (23.6) | 82 (25.1) | 45 (32.6) | |
| Day care in first 6 months | 0.175 | ||||
| No | 1,765 (65.3) | 1,445 (64.6) | 225 (68.8) | 95 (68.8) | |
| Yes | 556 (20.6) | 472 (21.1) | 54 (16.5) | 30 (21.7) | |
| Missing | 381 (14.1) | 320 (14.3) | 48 (14.7) | 13 (9.4) | |
| No. of persons in house | 0.668 | ||||
| <5 | 1,440 (53.3) | 1,189 (53.2) | 175 (53.5) | 76 (55.1) | |
| ≥5 | 814 (30.1) | 675 (30.2) | 94 (28.7) | 45 (32.6) | |
| Missing | 448 (16.6) | 373 (16.7) | 58 (17.7) | 17 (12.3) | |
Data are presented as n (%) unless indicated otherwise as mean (SD).
Maternal Diabetes in Pregnancy and Infant Infections
Overall, 416 (15.4%) and 1,425 infants (52.7%) had at least one infection episode in the neonatal period (within 1 month) and in the first 6 months of life, respectively. The most frequent infections included upper respiratory infections (34.6%), mycoses (21.5%), lower respiratory infection (8.5%), and otitis media (3.0%). Maternal diabetes was associated with an increased risk of infant infections in the first 6 months of life. Infants born to mothers who had GDM and PDM had a 1.21-fold (95% CI 1.10–1.33) and 1.33-fold (95% CI 1.16–1.51) increased risk of infections, respectively, compared with those born to mothers without diabetes after adjustment for social demographic confounders (Table 2, model 1). The associations persisted after additional adjustments for maternal prepregnancy OWO and season of birth (model 2), as well as daycare attendance and living condition (model 3). The associations persisted from neonatal to early infancy (1–6 months). Furthermore, we also observed a significant association between maternal diabetes and infection burden. The relative risk rate for having at least two infection episodes versus without infection was 1.38 (95% CI 1.02–1.87) and 2.02 (95% CI 1.31–3.13) for infants born to mothers with GDM and PDM, respectively, compared with those born to mothers without diabetes (Supplementary Table 3).
Table 2.
Associations between maternal diabetes status in pregnancy and the risk of child infections in the first 6 months of life
| Model 1 | Model 2 | Model 3 | ||||||
|---|---|---|---|---|---|---|---|---|
| Diabetes status | No. | Case, n (%) | RR | 95% CI | RR | 95% CI | RR | 95% CI |
| Aged 0–6 months | ||||||||
| No diabetes | 2,237 | 1,140 (51.0) | 1.00 | 1.00 | 1.00 | |||
| GDM | 327 | 196 (59.9) | 1.21 | 1.10–1.33 | 1.20 | 1.09–1.32 | 1.20 | 1.09–1.32 |
| PDM | 138 | 89 (64.5) | 1.33 | 1.16–1.51 | 1.30 | 1.13–1.49 | 1.28 | 1.12–1.47 |
| Aged 0–1 month | ||||||||
| No diabetes | 2,237 | 300 (13.4) | 1.00 | 1.00 | 1.00 | |||
| GDM | 327 | 85 (26.0) | 2.01 | 1.62–2.48 | 1.99 | 1.60–2.47 | 2.00 | 1.61–2.48 |
| PDM | 138 | 31 (22.5) | 1.76 | 1.27–2.44 | 1.72 | 1.23–2.41 | 1.71 | 1.22–2.39 |
| Aged 1–6 months | ||||||||
| No diabetes | 2,237 | 1,014 (45.3) | 1.00 | 1.00 | 1.00 | |||
| GDM | 327 | 155 (47.4) | 1.08 | 0.95–1.21 | 1.06 | 0.94–1.20 | 1.07 | 0.95–1.20 |
| PDM | 138 | 79 (57.2) | 1.33 | 1.14–1.56 | 1.30 | 1.11–1.52 | 1.28 | 1.09–1.50 |
Model 1 adjusted for maternal age, race and ethnicity, education, smoking status during pregnancy, parity, and child’s sex. Model 2 adjusted for covariates in model 1 plus maternal prepregnancy OWO and season of birth. Model 3 adjusted for covariates in model 2 plus daycare attendance and living condition.
To further examine whether preterm births, low birth weight, C-section, breastfeeding, or maternal glycemic control in pregnancy mediated the diabetes-infection association, we additionally included these variables in the models. As shown in Supplementary Table 4, the effect sizes were slightly reduced but remained significant.
Maternal Glycemic Control in Pregnancy and Infant Infections
In subgroups with available information for maternal blood glucose markers, we found that hyperglycemia was associated with increased infant infections. A per 5 mg/dL increase in average random blood glucose and average fasting blood glucose were associated with 1.03-fold (95% CI 1.02–1.03) and 1.03-fold (95% CI 1.01–1.05) increased risk of infection, respectively (Supplementary Table 5). Furthermore, unideal glycemic control was associated with 1.15-fold (95% CI 1.00–1.32) increased risk of infection compared with optimal control (Supplementary Table 6).
Synergistic Effects Between Maternal Diabetes and Prepregnancy OWO, Preterm Birth, or Mode of Birth on Infant Infections
As shown in Supplementary Table 7, although both PDM (risk ratio [RR] 1.43, 95% CI 1.07–1.91) and OWO (RR 1.12, 95% CI 1.03–1.21) were individually associated with infant infections, no synergistic effect was observed (RR 1.38, 95% CI 1.18–1.61). In contrast, we observed synergistic effects between maternal PDM and either preterm birth or C-section. Compared with infants born at term and to mothers without diabetes, those born preterm and to PDM mothers had 1.37-fold (95% CI 1.14–1.66) increased risk of infections (Supplementary Table 8). Similarly, infants born to PDM mother and via C-section had 1.33-fold (95% CI 1.11–1.59) greater risk of infections compared with their peers without those two conditions (Supplementary Table 9).
Effect Modification of Maternal OWO, Preterm Birth, and Mode of Birth
PDM was associated with increased risk of infant infection regardless maternal prepregnancy OWO status and mode of birth (Supplementary Tables 7 and 9). Meanwhile, the association between GDM and infections was only significant among infants born to non-OWO mothers (Supplementary Table 7) or among infants born via C-section (Supplementary Table 9). The interaction between GDM and OWO was significant (P = 0.002), while the interaction between GDM and mode of birth was marginal (P = 0.068). GDM and PDM were both associated with increased risk of infection regardless preterm birth (Supplementary Table 8).
Maternal Diabetes in Pregnancy and Neonatal Global IgG Antibody Repertoire Diversity
Newborns born to mothers with PDM had a lower diversity of global repertoire to pathogens across all three levels: species, peptide, and protein (Fig. 1). Compared with no diabetes, PDM was associated with a 4.07 (β = −4.07, 95% CI −7.74 to −0.39) decrease in the total number of seropositive species, a 179.5 (β = −179.5; 95% CI −351.5 to −7.6) decrease in the total number of peptide hits, and a 87.1 (β = −87.1, 95% CI −166.7 to −7.43) decrease in the total number of reactive proteins, after adjustment for maternal age, race and ethnicity, education, smoking status during pregnancy, parity, and neonatal sex. The associations reduced slightly after removing the mothers who had poorly controlled blood glucose or no available information about blood glucose markers (Supplementary Table 10) and with additional adjustment for maternal glycemia control status (Supplementary Table 11). The associations were consistent even when VirScan and ToxScan libraries were analyzed separately (Fig. 1). However, GDM was not associated with IgG repertoire diversity. Furthermore, maternal glycemic control status was not significantly associated with IgG repertoire diversity when stratified by type of diabetes (Supplementary Table 12).
Figure 1.
Associations between maternal diabetes in pregnancy and all three-tiered IgG antibody repertoire diversity (no diabetes was the reference group). PDM mothers included 56 with type 2 and 1 with type 1 diabetes. Model was adjusted for maternal age, race and ethnicity, education, smoking status during pregnancy, parity, and child’s sex.
Maternal Diabetes in Pregnancy and Pathogen Species Targeted by Passive Immunity
To examine whether maternal diabetes was associated with specific species targeted by passive immunity, we analyzed all species with a 10–90% seroprevalence in our cohort. Maternal GDM was not associated with IgG antibody reactivity to any viral species (Supplementary Fig. 3A) but was positively associated with two bacterial species (Coccidioides posadasii and Streptococcus agalactiae) (Supplementary Fig. 3C). Maternal PDM was inversely associated with IgG reactivity to 2 viral species (human adenovirus D and Cosavirus A) and one bacterial species (Porphyromonas gingivalis) (Supplementary Fig. 3B and D) but positively associated with only one viral species (human parvovirus B19) (Supplementary Fig. 3B). However, none of these associations passed the multiple testing correction.
Maternal Diabetes in Pregnancy and Levels of IgG Antibodies
To explore whether maternal diabetes affected the strength of IgG antibodies in cord blood, we analyzed the peptide with maximal value for all peptides across that protein. Maternal GDM was positively associated with 28 peptides in the ToxScan library and 90 peptides in the VirScan library and inversely associated with 14 peptides in the ToxScan library and 46 peptides in the VirScan library (Supplementary Fig. 4A and C). Maternal PDM was positively associated with 17 peptides in the ToxScan library and 57 peptides in the VirScan library but inversely associated with 38 peptides in the ToxScan library and 97 peptides in the VirScan library (Supplementary Fig. 4B and D). However, only one association between a peptide from enterovirus C and PDM remained significant after false discovery rate multiple testing correction (Supplementary Fig. 4B).
Synergistic Effect Between Diabetes and Prepregnancy OWO, Preterm Birth, or C-Section on IgG Repertoire
Although maternal prepregnancy OWO alone was not significantly associated with the diversity of IgG repertoire, OWO enhanced the effect of PDM on IgG diversity across all three levels (species, peptide, and protein). Compared with newborns born to mothers without OWO and diabetes, those born to mothers with both OWO and PDM had 4.46 (SE 2.12), 210.1 (SE 99.1), and 103.1 (SE 45.9) decreases in the total number of reactive species, epitopes, and proteins, respectively (Supplementary Table 13). A similar synergistic effect between diabetes and C-section was also observed. Compared with newborns born vaginally to mothers without diabetes, those born to PDM mothers and via C-section had 6.1 (SE 2.6), 292.3 (SE 122.4), and 139.2 (SE 56.7) decreases in the total number of reactive species, epitopes, and proteins, respectively (Supplementary Table 14). In contrast, we did not observe a synergistic effect between maternal diabetes and preterm birth on IgG repertoire (Supplementary Table 15).
IgG Repertoire Diversity and Infant Infections
In a subset (n = 926) with information available for both IgG repertoire and infection phenotype, infants with a top tertile of the total number of species targeted by maternal antibody exhibited a 5% reduction in the risk of infection (RR 0.95, 95% CI 0.83–1.10) compared with infants with the lowest tertiles, but this difference was not statistically significant (Supplementary Table 16). When stratified by maternal diabetes status, the top tertile of species diversity was significantly associated with a reduced risk of infection (RR 0.26, 95% CI 0.08–0.92) compared with lowest tertile of diversity among infants born to PDM mothers. The interaction of PDM and the top tertile of species diversity was marginal (P = 0.057) (Supplementary Table 16).
Synergistic Effect Between Diabetes and IgG Antibody Diversity on Infant Infections
Compared with infants born to mothers without diabetes and with the top tertile of species diversity, those born to mothers with diabetes and with the lowest tertile of species diversity had 1.35 times (95% CI 1.08–1.70) increase in the risk of infections in the first 6 months of life. The association remained significant (RR 1.29, 95% CI 1.02–1.63) after further adjusting for maternal prepregnancy OWO, daycare attendance, and living condition (Fig. 2).
Figure 2.
Synergistic effect between maternal diabetes and total number of targeted species by passive immunity on infant infections. Model was adjusted for maternal age, race and ethnicity, education, smoking status during pregnancy, parity, prepregnancy OWO, child’s sex, daycare, living condition, and season of birth. T, tertile.
Sensitivity Analysis
To reduce the influence of extreme hyperglycemia, sensitivity analysis was performed after removing 210 mothers with unideal glycemic control in pregnancy or without available information for blood glucose markers. The association remained significant although the magnitude of association was slightly reduced (Supplementary Table 17). Moreover, to reduce the potential influence of cross-reactive antibody, we conducted a sensitivity analysis after removing those epitope hits that may be caused by cross-reactivity using the Xu et al. (16) strategy. The association between maternal diabetes and IgG antibody diversity did not change materially (Supplementary Fig. 5).
Conclusions
To our knowledge, this is the first prospective birth cohort study to investigate the impact of both maternal GDM and PDM on the risk of infant infections. We documented a link between exposure to diabetes in utero and an increased risk of infections in the first 6 months of life regardless the type of diabetes. The link persisted from the neonatal period to the first 6 months of age and was independent of maternal OWO. We evidenced that hyperglycemia was an important contributing factor to the infections. We identified a subgroup of infants of PDM mothers, who were born preterm or via C-section, were more susceptible to early life infections. In addition, we demonstrated an impaired IgG repertoire diversity in newborns of PDM mothers. Finally, we revealed that infants born to mothers with diabetes and coupled with lower IgG repertoire diversity were at an increased risk of infections relative to their peers with the opposite conditions. Our findings highlight that maternal diabetes in pregnancy may pose a risk for infants. GDM and PDM had a distinct impact on offspring passive immunity. Diabetes and impaired IgG repertoire act synergistically to increase infection risk.
In agreement with previous studies that have established a link between maternal diabetes in pregnancy and neonatal infection (7,21,22), our data showed that PDM and GDM were both associated with higher risk of infant infections, although their onset and progression differed considerably. Our study further extends previous findings, indicating that the adverse impacts of maternal diabetes on their baby’s vulnerability to infections persist from neonatal to at least the first 6 months of life. Importantly, our data further pinpointed that the associations were exacerbated by preterm birth, C-section, and lower IgG repertoire diversity. Our findings raise the possibility that optimal management of pregnant women in both the preconception and perinatal period may reduce the risk of infant infections.
Interestingly, our data revealed a divergent IgG repertoire pattern in cord blood from PDM and GDM. Newborns born to PDM mothers had a markedly reduced IgG repertoire diversity, but infants of GDM mothers were indistinguishable from those of mothers without diabetes. In support of our findings, one study reported lower concentrations of total IgG and subclass IgG in cord blood from PDM, but higher concentrations from GDM (23). It is well known that cord blood IgG antibodies reflect maternal immunologic memory. The transfer of maternal IgG antibodies is an active process and mediated by Fc-receptor (FcRn) expressed in syncytiotrophoblasts (24). A previous study reported a lower FcRn expression in PDM, but a higher FcRn expression in GDM (23). In addition, IgG glycan pattern influences the affinity of IgG for FcRn (25), while diabetes has been linked to changes in the glycosylation patterns of IgG (26). Furthermore, placental transfer of maternal IgG antibodies also depends on maternal IgG levels (24). Individuals with diabetes display blunted IgG responses to infection (27) and had lower serum IgG concentrations compared with individuals without diabetes (28), but the antibody response of GDM mothers was comparable to mothers without diabetes (29). Taken together, our findings emphasize that the characteristics specific to PDM and GDM may drive the divergent impact on infant passive immunity.
Furthermore, we explored whether maternal glycemic status plays a role in infant infections. We found that hyperglycemia was associated with an increased risk. When restricted to infants of mothers with diabetes, poorly controlled glycemic status also led to an increased risk. This finding is consistent with a previous study that reported a higher rate of infections and higher C-reactive protein levels in neonates born to mothers with poorly controlled GDM than that in those born to mothers with optimal controlled GDM (7). In the meantime, our data also showed that the diabetes-infection association was beyond maternal glycemic status, suggesting additional risk factors for infant infections besides maternal hyperglycemia. In support of our findings, studies showed that the inflammation and autophagy in the placenta (8) and T-cell dysfunction (7) were responsible to the link between maternal GDM and increased neonatal infection. Nevertheless, our data reiterate the potential importance of appropriate diabetes management in pregnancy as highlighted by the American Diabetes Association (30).
Notably, we did not observe significantly consistent associations between maternal glycemic status and IgG repertoire diversity for infants of PDM and GDM mothers. Literature about the relationship between maternal glycemic status and neonatal circulating IgG levels is controversial. Some studies showed a positive association (31), others showed a negative relationship (7,29), or no relationship (27). This divergence further confirmed that other mechanisms besides hyperglycemia also drive placental IgG transfer. The absence of a definitive passive immunity profile in the infants of mothers with diabetes confirms the need for extended investigation in this field.
Strengths and Limitations
The major strength of this study is a large array of pathogen-specific IgG antibodies, including antibacterial, antiviral, antifungal, and antiparasitic IgG detected by the high-throughput PhIP-Seq technology. In addition, we were able to differentiate the effects of PDM and GDM.
Major limitations of our approach include that VirScan and ToxScan cannot detect IgG against capsular polysaccharide and the protective capacity of IgG antibodies. We did not assess infant circulating IgG repertoire in the first 6 months, so we cannot conclude whether the duration of the presence of maternally derived antibody is also responsible to the risk of infections. Additionally, early-life infections were defined based on ICD codes, which limits our ability to analyze the infections by specific pathogen or its categories. Finally, incomplete prescription information for GDM mothers limits our ability to further explore the impact of severity of GDM on neonatal IgG repertoire diversity.
Our findings have important clinical and public health implications. Our findings raise the possibility that appropriate screening and management of maternal diabetes across the entire gestational period, and reducing rate of preterm birth and C-section may reduce the risk of infections in early life and achieve long-lasting benefits. Infection prevention strategies are critical for infants born to mothers with diabetes and these infants should be treated promptly if they develop an infection. In addition, our study provides new insights into the role of maternal diabetes on neonatal passive immunity as well as strong evidence of distinct IgG transfer patterns of GDM and PDM. Given that maternally derived antibody is a critical component of newborn immunity against pathogens, it warrants further investigation.
In conclusion, in this prospective birth cohort study, we demonstrated a strong association between maternal diabetes in pregnancy and infant infections in early life. Preterm birth, C-section, and lower IgG repertoire diversity enhanced the association. We also revealed a divergent IgG repertoire feature from newborns born to mothers with PDM or GDM. These findings offer new insights into how maternal metabolic conditions may influence their baby’s passive immunity and susceptibility to infections.
This article contains supplementary material online at https://doi.org/10.2337/figshare.30871508.
Article Information
Acknowledgments. The authors thank the study participants, the nursing staff at Labor and Delivery of the Boston Medical Center, and the field team for their contributions to the Boston Birth Cohort. Linda Rosen, MSEE, and the Boston Medical Center Clinical Data Warehouse assisted in obtaining relevant clinical information; she was compensated for her time.
P.A.F.-G.’s contributions were made as part of her official duties as a National Institutes of Health federal employee, are in compliance with agency policy requirements, and are considered works of the U.S. Government. However, the findings and conclusions presented in this paper are those of the authors and do not necessarily reflect the views of the National Institutes of Health or the U.S. Department of Health and Human Services. This information or content and conclusions are those of the authors and should not be construed as the official position or policy of, nor should any endorsements be inferred by, the funding agencies.
Duality of Interest. H.B.L. is a founder of Infinity Bio, a provider of antibody reactome profiling services. I.R. is a paid consultant for Infinity Bio. No other potential conflicts of interest relevant to this article were reported.
Author Contributions. G.W. and X.W. designed and conceptualized the study and drafted and revised the manuscript. I.R. and H.B.L. contributed to data analysis. I.R., H.B.L., M.J.G., P.A.F.-G., X.H., H.J., C.P., W.G.A., and X.W. interpreted the results and reviewed and revised the manuscript. C.P. and W.G.A. contributed to the acquisition of data. All authors contributed to the discussion, reviewed/edited the manuscript, and approved the final manuscript. G.W. and X.W. are the guarantors of this work and, as such, had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Handling Editors. The journal editors responsible for overseeing the review of the manuscript were John B. Buse and David Simmons.
Funding Statement
The Clinical Data Warehouse service is supported by Boston University’s Clinical and Translational Institute and the National Institutes of Health Clinical and Translational Science Award (grant U54-TR001012). The Boston Birth Cohort (the parent study) is supported in part by National Institutes of Health grants from Eunice Kennedy Shriver National Institute of Child Health and Human Development (2R01HD041702, R01HD098232, and R21HD116039), the National Institute of Environmental Health Sciences (R01ES031272, R01ES031521, and U01 ES034983), and by the Health Resources and Services Administration (HRSA) of the U.S. Department of Health and Human Services cooperative agreement (UT7MC45949). This study is also supported in part by American Diabetes Association grant (7-24-ICTSWH-13). P.A.F.-G. is supported by the Intramural Research Program of the National Institutes of Health.
Supporting information
References
- 1. Shah NS, Wang MC, Freaney PM, et al. Trends in gestational diabetes at first live birth by race and ethnicity in the US, 2011-2019. JAMA 2021;326:660–669 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Godfrey KM, Reynolds RM, Prescott SL, et al. Influence of maternal obesity on the long-term health of offspring. Lancet Diabetes Endocrinol 2017;5:53–64 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Yefet E, Bejerano A, Iskander R, Zilberman Kimhi T, Nachum Z. The association between gestational diabetes mellitus and infections in pregnancy-systematic review and meta-analysis. Microorganisms 2023;11:1956. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Holt RIG, Cockram CS, Ma RCW, Luk AOY. Diabetes and infection: review of the epidemiology, mechanisms and principles of treatment. Diabetologia 2024;67:1168–1180 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Hobbs JR, Davis JA. Serum gamma-G-globulin levels and gestational age in premature babies. Lancet 1967;1:757–759 [DOI] [PubMed] [Google Scholar]
- 6. Pou C, Nkulikiyimfura D, Henckel E, et al. The repertoire of maternal anti-viral antibodies in human newborns. Nat Med 2019;25:591–596 [DOI] [PubMed] [Google Scholar]
- 7. Wang B-B, Xue M. Early neonatal complications in pregnant women with gestational diabetes mellitus and the effects of glycemic control on neonatal infection. World J Diabetes 2023;14:1393–1402 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Li Y-X, Long D-L, Liu J, et al. Gestational diabetes mellitus in women increased the risk of neonatal infection via inflammation and autophagy in the placenta. Medicine (Baltimore) 2020;99:e22152. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Dowling DJ, Levy O. Ontogeny of early life immunity. Trends Immunol 2014;35:299–310 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Li Y, Wang X, Blau DM, et al.; Respiratory Virus Global Epidemiology Network; RESCEU investigators . Global, regional, and national disease burden estimates of acute lower respiratory infections due to respiratory syncytial virus in children younger than 5 years in 2019: a systematic analysis. Lancet 2022;399:2047–2064 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Wang X, Li Y, Deloria-Knoll M, et al.; Respiratory Virus Global Epidemiology Network . Global burden of acute lower respiratory infection associated with human parainfluenza virus in children younger than 5 years for 2018: a systematic review and meta-analysis. Lancet Glob Health 2021;9:e33–e43 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Switzer C, D’Heilly C, Macina D. Immunological and clinical benefits of maternal immunization against pertussis: a systematic review. Infect Dis Ther 2019;8:499–541 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Pearson C, Bartell T, Wang G, et al. Boston Birth Cohort profile: rationale and study design. Precis Nutr 2022;1:e00011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Wang G, Buckley JP, Bartell TR, Hong X, Pearson C, Wang X. Gestational diabetes mellitus, postpartum lipidomic signatures, and subsequent risk of type 2 diabetes: a lipidome-wide association study. Diabetes Care 2023;46:1223–1230 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Hong X, Frischmeyer-Guerrerio P, Wang G, et al. Integrating exposures and multi-omics in the Boston Birth Cohort to elucidate immune development across the life course: rationale and study design. Precis Nutr 2025;4:e00097. [PMC free article] [PubMed] [Google Scholar]
- 16. Xu GJ, Kula T, Xu Q, et al. Viral immunology. Comprehensive serological profiling of human populations using a synthetic human virome. Science 2015;348:aaa0698. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Angkeow JW, Monaco DR, Chen A, et al. Phage display of environmental protein toxins and virulence factors reveals the prevalence, persistence, and genetics of antibody responses. Immunity 2022;55:1051–1066.e4 e1054 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. National Institute of Diabetes and Digestive and Kidney Diseases . Definition and facts for adult overweight and obesity. Accessed 10 April 2025. Available from https://www.niddk.nih.gov/health-information/weight-management/adult-overweight-obesity/definition-facts
- 19. Wang G, Divall S, Radovick S, et al. Preterm birth and random plasma insulin levels at birth and in early childhood. JAMA 2014;311:587–596 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. World Health Organization . Low birth weight. Accessed 10 April 2025. Available from https://www.who.int/data/nutrition/nlis/info/low-birth-weight
- 21. Feleke BE, Feleke TE, Adane WG, et al. Maternal and newborn effects of gestational diabetes mellitus: a prospective cohort study. Prim Care Diabetes 2022;16:89–95 [DOI] [PubMed] [Google Scholar]
- 22. Lee H-Y, Hsu Y-L, Lee W-Y, et al. Maternal infections, antibiotics, steroid use, and diabetes mellitus increase risk of early-onset sepsis in preterm neonates: a nationwide population-based study. Pathogens 2025;14:89–93 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. de Souza EG, Hara CCP, Fagundes DLG, et al. Maternal-foetal diabetes modifies neonatal fc receptor expression on human leucocytes. Scand J Immunol 2016;84:237–244 [DOI] [PubMed] [Google Scholar]
- 24. Palmeira P, Quinello C, Silveira-Lessa AL, Zago CA, Carneiro-Sampaio M. IgG placental transfer in healthy and pathological pregnancies. Clin Dev Immunol 2012;2012:985646. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Jennewein MF, Alter G. The immunoregulatory roles of antibody glycosylation. Trends Immunol 2017;38:358–372 [DOI] [PubMed] [Google Scholar]
- 26. Lemmers RFH, Vilaj M, Urda D, et al. IgG glycan patterns are associated with type 2 diabetes in independent European populations. Biochim Biophys Acta Gen Subj 2017;1861:2240–2249 [DOI] [PubMed] [Google Scholar]
- 27. Farnsworth CW, Shehatou CT, Maynard R, et al. A humoral immune defect distinguishes the response to Staphylococcus aureus infections in mice with obesity and type 2 diabetes from that in mice with type 1 diabetes. Infect Immun 2015;83:2264–2274 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Guo X, Meng G, Liu F, et al. Serum levels of immunoglobulins in an adult population and their relationship with type 2 diabetes. Diabetes Res Clin Pract 2016;115:76–82 [DOI] [PubMed] [Google Scholar]
- 29. de Morais Oliveira-Scussel AC, Ferreira PTM, de Sousa Resende R, et al. Association of gestational diabetes mellitus and negative modulation of the specific humoral and cellular immune response against Toxoplasma gondii. Front Immunol 2022;13:925762. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. American Diabetes Association Professional Practice Committee. 15. Management of diabetes in pregnancy: Standards of Care in Diabetes—2025. Diabetes Care 2025;48(Suppl. 1):S306–S320 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. França EL, Calderon IdMP, Vieira EL, Morceli G, Honorio-França AC. Transfer of maternal immunity to newborns of diabetic mothers. Clin Dev Immunol 2012;2012:928187 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.