Abstract
Background:
Gestational diabetes mellitus (GDM) may exacerbate hyperbilirubinemia-induced auditory neurotoxicity via oxidative stress, whereas genetic susceptibility (e.g., GJB2 variants) heightens cochlear vulnerability. However, the tripartite interplay among maternal metabolic dysregulation, neonatal hyperbilirubinemia and genetic predisposition remains poorly characterized.
Objective:
This study aimed to investigate the synergistic interaction between GDM and neonatal hyperbilirubinemia in auditory dysfunction, focusing on the modifying role of deafness-associated genetic variants (GJB2, SLC26A4, and mtDNA).
Methods:
In this prospective cohort study (our hospital, 2023–2025), 350 neonates were classified by an exposure-based 2 × 2 factorial design into four groups: GDM-exposed hyperbilirubinemia (n = 98), non-GDM hyperbilirubinemia (n = 185), GDM-only (n = 42) and healthy controls (n = 25). Genetic screening was performed via Sanger sequencing. Serial audiological assessments (automated auditory brainstem response) and brainstem auditory evoked potential [BAEP] and biochemical monitoring (bilirubin-to-albumin [B/A], malondialdehyde [MDA] and superoxide dismutase) were conducted at birth, 72 h, 1 month, and 3 months. Cox proportional hazards models, multivariable logistic regression and bootstrapped causal mediation analysis were used to evaluate risk factors and oxidative stress-mediated pathways.
Results:
GDM-exposed hyperbilirubinemia neonates exhibited a 2.3-fold higher incidence of severe hearing loss compared with those with hyperbilirubinemia alone (34.7% vs. 15.1%, P < 0.001). GJB2 variant carriers showed significantly greater BAEP threshold elevation (Δ = 9.2 ± 3.1 dB vs. 3.1 ± 1.8 dB, P = 0.002) than their counterparts. The B/A ratio was a GDM-dependent prognostic marker, with B/A > 5.5 predicting hearing loss only in the GDM subgroup (odds ratio = 4.17, 95% confidence interval [CI]: 2.33–7.45). MDA accounted for 38% of the total effect of GDM–hyperbilirubinemia on auditory impairment (indirect effect β = 0.28, 95% CI: 0.19–0.39).
Conclusion:
GDM significantly modifies hyperbilirubinemia-induced auditory toxicity through oxidative stress and genetic interaction, particularly in GJB2 carriers. Integrated screening of maternal glycemic status, neonatal bilirubin dynamics, and deafness-associated genes is warranted.
Keywords: gestational diabetes mellitus, neonatal hyperbilirubinemia, sensorineural hearing loss, connexin 26, oxidative stress
KEY MESSAGES
-
(1)
Neonates exposed to both maternal gestational diabetes mellitus and hyperbilirubinemia showed a higher incidence of severe hearing impairment than those with hyperbilirubinemia alone, consistent with a positive joint effect on the absolute-risk scale.
-
(2)
Oxidative stress, indexed by malondialdehyde, significantly mediated the association between maternal hyperglycemia and cochlear injury.
-
(3)
Carriers of pathogenic Gap Junction Beta 2 gene variants exhibited heightened susceptibility, supporting a gene–environment interaction model.
-
(4)
The bilirubin-to-albumin ratio outperformed total serum bilirubin as a prognostic marker in diabetes-exposed neonates, supporting integrated maternal, biochemical and genetic screening for personalized risk stratification.
INTRODUCTION
Gestational diabetes mellitus (GDM), a condition affecting 7%–25% of pregnancies globally, has emerged as a critical determinant of neonatal morbidity, with hyperbilirubinemia being one of its most prevalent complications.[1,2] Neonatal hyperbilirubinemia is traditionally managed as a transient condition, but accumulating evidence highlights its neurotoxic potential, particularly in the auditory system, where unconjugated bilirubin disrupts synaptic transmission and induces oxidative damage to cochlear hair cells.[3,4] Notably, recent studies suggest that maternal metabolic dysregulation in GDM exacerbates bilirubin toxicity through inflammatory and oxidative pathways,[5,6] yet the mechanistic interplay between these factors remains poorly understood.
The pathophysiology of GDM-associated hyperbilirubinemia involves a triad of placental inflammation, altered bilirubin–albumin binding kinetics and genetic susceptibility. Placental dysfunction in GDM activates the Toll-like receptor 4/myeloid differentiation primary response 88/nuclear factor kB pathway, elevating pro-inflammatory cytokines (e.g., interleukin-6 and interleukin-8) and oxidative stress markers such as malondialdehyde (MDA).[7,8] These changes impair the neonatal blood–brain barrier and reduce albumin’s bilirubin-binding capacity, facilitating neurotoxin deposition in auditory nuclei.[9] Simultaneously, genetic variants in prelingual deafness-associated genes (GJB2, SLC26A4, and mtDNA) may heighten cochlear vulnerability to bilirubin. For instance, GJB2 mutations disrupt connexin 26 gap junctions, which are critical for potassium recycling in the stria vascularis, thereby amplifying oxidative damage under hyperbilirubinemia conditions.[10,11] Current clinical guidelines fail to incorporate maternal glycemic status, neonatal bilirubin fluctuations, and genetic profiles for risk stratification, representing a critical gap given that 28%–35% of severe neonatal hearing loss cases occur in GDM-exposed infants with hyperbilirubinemia.[12,13]
Emerging data highlight the limitations of universal bilirubin thresholds. The bilirubin-to-albumin (B/A) ratio, a marker of unbound bilirubin, demonstrates context-dependent prognostic value, with GDM-exposed neonates exhibiting heightened sensitivity to B/A elevations.[14] Furthermore, extended phototherapy, though effective in reducing total serum bilirubin (TSB), paradoxically increases oxidative stress through light-induced lipid peroxidation, potentially worsening auditory outcomes.[15] These observations were aligned with cohort studies reporting subclinical hearing deficits (brainstem auditory evoked potential [BAEP] 20–35 dBnHL) in 25%–30% of GDM–hyperbilirubinemia cases,[16] suggesting early neurotoxicity even at bilirubin levels below standard intervention thresholds. Therefore, this study aimed to investigate (1) the synergistic interaction between GDM-associated metabolic dysregulation and hyperbilirubinemia in auditory toxicity, (2) the modifying role of genetic variants in GJB2 and related deafness genes, and (3) the oxidative stress mechanisms bridging maternal–fetal environments to cochlear injury. By employing longitudinal audiological phenotyping, genetic screening, and causal mediation analysis, we aimed to redefine risk prediction models for neonatal hearing loss in this vulnerable population.
MATERIALS AND METHODS
Study Design
This single-center prospective cohort study was conducted in the Department of Neonatology of Yuhuan Branch, The First Affiliated Hospital of Wenzhou Medical University, from January 2023 to January 2025. Neonates were classified using an exposure-based 2 × 2 factorial design into four groups: (1) GDM-exposed hyperbilirubinemia (n = 98), (2) non-GDM hyperbilirubinemia (n = 185), (3) GDM-only (n = 42), and (4) healthy controls (n = 25). All participants underwent serial evaluations of audiological and biochemical parameters at multiple time points: at birth, 72 h postnatally and at 1 and 3 months of age. The study protocol received approval from the Ethics Committee of our hospital (approval code: 2023[58]. Written informed consent was obtained from parents or legal guardians following a comprehensive explanation of the implications of genetic testing. Data anonymization protocols were implemented to protect participant confidentiality, and access to genetic information was restricted to authorized investigators. Sample size calculations were performed using G*Power 3.1 (Heinrich Heine Universität Düsseldorf, Düsseldorf, Germany) with α = 0.05 (two-sided), power = 0.80 and expected odds ratio (OR) = 2.5 for the primary comparison (GDM-exposed hyperbilirubinemia vs. non-GDM hyperbilirubinemia).[17] The required sample was 280 neonates; accounting for a 25% attrition rate, the planned enrolment was set at 350. The achieved post-hoc power for the primary comparison (n1 = 98, n2 = 185) was approximately 85%.
Inclusion Criteria
The inclusion criteria were as follows: (1) gestational age ≥37 weeks, birth weight between 2500 and 4000 g; singleton. (2) Group-specific TSB criteria. Hyperbilirubinemia groups (GDM + hyper and non-GDM hyper): peak TSB meeting hour-specific AAP phototherapy threshold. GDM-only and healthy control groups: peak TSB below the AAP phototherapy threshold.[12] (3) Maternal inclusion required a confirmed GDM diagnosis per the International Association of Diabetes and Pregnancy Study Groups criteria based on a 75-g oral glucose tolerance test performed at 24–28 weeks of gestation.[18] (4) Availability of maternal glycated hemoglobin (HbA1c) and neonatal genetic screening (GJB2, SLC26A4, and mtDNA). All enrolled neonates passed otoacoustic emission screening within 24 h of birth.
Exclusion Criteria
Neonates were excluded for (1) congenital infections (e.g., cytomegalovirus and toxoplasmosis), (2) chromosomal abnormalities or (3) major malformations affecting auditory pathways. Additional exclusions included (4) severe perinatal asphyxia (5-min Apgar score ≤3),[19] (5) hemolytic disorders (ABO/Rh incompatibility or G6PD deficiency confirmed by enzyme assay), (6) incomplete audiological, or (7) biochemical follow-up data. Maternal exclusions comprised (1) pre-existing diabetes (type 1 or 2), (2) chronic renal/hepatic insufficiency, (3) autoimmune disorders, or (4) use of ototoxic medications during pregnancy. These criteria ensured homogeneity in evaluating GDM–hyperbilirubinemia interactions while minimizing confounding genetic and environmental factors.
Data Collection
Data were systematically collected using standardized case report forms, encompassing the following domains: (1) neonatal bilirubin kinetics: quantified via hourly transcutaneous bilirubinometry (JM-105, Dräger) during phototherapy and confirmed with TSB measurements (Vitros 5600, Ortho Clinical Diagnostics). (2) Genetic data: Variant classification adhered to the 2015 American College of Medical Genetics and Genomics/Association for Molecular Pathology (ACMG/AMP) guidelines.[20] Specific mutations in GJB2 (c.35delG and c.235delC), SLC26A4 (c.919-2A>G and c.2168A>G) and mtDNA (m.1555A>G and m.1494C>T) were analyzed via Sanger sequencing of DNA extracted from dried blood spots. (3) Audiological profiles: Serial audiological assessments were conducted at predefined intervals. These included automated auditory brainstem response (AABR) results (classified as pass or refer) at 72 h postnatally, BAEP thresholds (measured in dBnHL) at 1 and 3 months of age and interaural latency difference (ILD) of wave V in response to click stimuli ranging from 70 dBnHL to 30 dBnHL. (4) Oxidative stress biomarkers (MDA and superoxide dismutase [SOD] were analyzed using high-performance liquid chromatography (HPLC; Agilent 1260) at matched timepoints. (5) Biochemical parameters: TSB, HbA1c, plasma albumin and the B/A ratio were quantified at the peak of hyperbilirubinemia (72 ± 12 h postnatally) using enzymatic assays performed on the Roche Cobas 8000 analyzer. (6) Clinical covariates: Relevant clinical variables included gestational age, birth weight and duration of phototherapy.
Follow-up Protocol
The primary outcome was hearing outcomes, which were categorized as severe impairment (BAEP >35 dBnHL + AABR failure),[21] subclinical deficit (BAEP 20–35 dBnHL with normal AABR)[22] or at risk (ABR wave I–III latency prolongation >2.8 ms).[23] The follow-up protocol for neonates at risk of auditory impairment adopted a tiered surveillance schedule to optimize early detection and longitudinal monitoring. (1) During the initial phase (0–1 month), genetic screening (GJB2, SLC26A4, and mtDNA variants) and AABR testing were performed on day 3 to establish baseline risk profiles. For infants exhibiting rebound jaundice, TSB and albumin levels were reassessed at week 2 to guide phototherapy adjustments. (2) At 1 month, diagnostic BAEP testing with tympanometry was conducted to evaluate cochlear and middle ear function, identifying subclinical abnormalities such as interaural latency disparities (>0.25 ms) or elevated thresholds (>20 dBnHL). (3) The extended monitoring phase (3–12 months) included repeat BAEP at 3 months using frequency-specific stimuli (500–4000 Hz) to map threshold progression, particularly in carriers with variant-dependent vulnerabilities. (4) Quarterly evaluations were mandated for carriers demonstrating subclinical impairment (BAEP 20–35 dBnHL), focusing on the neurodevelopmental integration of auditory pathways and bilirubin-induced synaptic remodeling. The primary outcome was hearing impairment at 3 months, defined as BAEP >20 dBnHL in either ear. The tiered follow-up strategy and threshold progression criteria are summarized in [Figure 1].
Figure 1.
Threshold progression map for hearing impairment.
Statistical Analysis
Statistical analyses were conducted using R version 4.3.1 (R Foundation for Statistical Computing, Vienna, Austria), using the survival, lme4 and mediation packages. Two-sided P < 0.05 was considered statistically significant.
Baseline comparisons: Continuous variables were assessed for normality using the Shapiro–Wilk test. Normally distributed continuous variables were compared across the four groups using one-way ANOVA and are presented as mean ± standard deviation; non-normally distributed variables were compared using the Kruskal–Wallis test and are presented as median (interquartile range). Categorical variables were compared using chi-square tests or Fisher’s exact tests as appropriate and are presented as n (%). F-statistics or χ2 values are reported alongside P values in all tables.
Univariate and multivariable analyses: Univariate logistic regression was first performed for each candidate variable to identify risk factors for hearing impairment (defined as BAEP >20 dBnHL). Variables with P < 0.1 in univariate analysis, or deemed clinically relevant based on prior literature, were entered into a multivariable logistic regression model. Adjusted ORs with 95% confidence intervals (CIs) were calculated.
Survival analysis: Time to hearing impairment was analyzed using Cox proportional hazards regression. The proportional hazards assumption was tested using Schoenfeld residuals. Unadjusted and adjusted hazard ratios (HRs) were reported. The adjusted model included the following covariates: maternal HbA1c, GDM treatment modality, gestational age, birth weight, peak TSB, and phototherapy duration. Covariates were selected a priori based on (i) clinical relevance, (ii) statistical significance at P < 0.10 in univariate analysis, and (iii) prior literature on perinatal hearing outcomes.
Mediation analysis: Bootstrapped causal mediation analysis was performed using the mediation package in R,[24] with MDA and SOD specified as separate mediators in two parallel models. The mediator equation and the outcome equation were adjusted for the following: maternal HbA1c, GDM treatment modality, gestational age, birth weight, peak TSB, and phototherapy duration. E-value sensitivity analysis for unmeasured confounding was conducted.
Subgroup analysis: To evaluate the influence of genetic susceptibility, we conducted stratified analyses comparing hearing impairment incidence between carriers and noncarriers of pathogenic variants in GJB2, SLC26A4, and mtDNA within each exposure group. Chi-square or Fisher’s exact tests were used for between-group comparisons.
Longitudinal analysis: Repeated measurements of MDA, SOD, and bilirubin (birth, 72 h, 1 month, and 3 months) were analyzed using linear mixed-effects models (lme4 package) with subject-specific random intercepts. Fixed effects included exposure group, time and group × time interaction, adjusted for gestational age and birth weight.
Interaction analysis: The multiplicative interaction between GDM and hyperbilirubinemia was assessed via a product term (GDM × hyperbilirubinemia) in the multivariable logistic regression, accompanied by a likelihood-ratio test. The additive interaction was quantified by the relative excess risk.
RESULTS
Enrolment Flow
Among 3212 neonates screened between January 2023 and January 2025, 2812 were ineligible (primary reasons: congenital infections, major malformations affecting auditory pathways, hemolytic disorders, pre-existing maternal diabetes, and other inclusion/exclusion criteria). Of the 400 eligible neonates enrolled, 50 (12.5%) were subsequently excluded because of incomplete follow-up (n = 28, 7.0%), parental withdrawal (n = 12, 3.0%) or technical data loss (n = 10, 2.5%), leaving 350 neonates for final analysis (GDM-exposed hyperbilirubinemia, n = 98; non-GDM hyperbilirubinemia, n = 185; GDM-only, n = 42; and healthy controls, n = 25).
Comparison of Baseline Characteristics
The data showed that neonatal factors, including gestational age, birth weight, and sex distribution, were similar across all groups (χ2 = 0.62, P > 0.05), indicating a well-matched cohort for comparison. However, significant differences were evident in maternal GDM status and neonatal bilirubin levels (F = 412.6, P < 0.001) and the B/A ratio (F = 385.2, P < 0.001), consistent with the group definitions. Additionally, genetic variants (GJB2, SLC26A4, and mtDNA) were evenly distributed (GJB2: χ2 = 0.39, P = 0.945; SLC26A4: χ2 = 0.13, P = 0.987; mtDNA: χ2 = 0.20, P = 0.978), suggesting no genetic predisposition bias. This balanced baseline allows for a clear evaluation of the specific effects of GDM and hyperbilirubinemia on hearing outcomes [Table 1].
Table 1.
Baseline Characteristics between the Four Groups
| Variable | GDM-Exposed Hyperbilirubinemia (n = 98) | Non-GDM Hyperbilirubinemia (n = 185) | GDM-Only (n = 42) | Healthy Controls (n = 25) | P |
|---|---|---|---|---|---|
| Gestational age (weeks) | 38.5 ± 1.0 | 38.6 ± 1.1 | 38.4 ± 0.9 | 38.7 ± 1.0 | 0.782 |
| Birth weight (g) | 3250 ± 300 | 3280 ± 310 | 3200 ± 290 | 3300 ± 280 | 0.654 |
| Male, n (%) | 52 (53.1%) | 98 (53.0%) | 22 (52.4%) | 13 (52.0%) | 0.892 |
| GJB2 variants, n (%) | 10 (10.2%) | 20 (10.8%) | 4 (9.5%) | 2 (8.0%) | 0.945 |
| SLC26A4 variants, n (%) | 5 (5.1%) | 10 (5.4%) | 2 (4.8%) | 1 (4.0%) | 0.987 |
| mtDNA mutations, n (%) | 3 (3.1%) | 6 (3.2%) | 1 (2.4%) | 1 (4.0%) | 0.978 |
| AABR refer at 72 h, n (%) | 20 (20.4%) | 20 (10.8%) | 3 (7.1%) | 1 (4.0%) | 0.002 |
| BAEP threshold at 1M (dBnHL) | 25.3 ± 5.2 | 20.1 ± 4.8 | 18.5 ± 3.9 | 17.2 ± 3.5 | <0.001 |
| BAEP threshold at 3M (dBnHL) | 23.8 ± 4.9 | 18.9 ± 4.2 | 17.8 ± 3.6 | 16.5 ± 3.2 | <0.001 |
| ILD wave V at 3Ms (ms) | 0.28 ± 0.12 | 0.15 ± 0.08 | 0.12 ± 0.07 | 0.10 ± 0.06 | <0.001 |
| TSB at peak (mg/dL) | 16.2 ± 1.5 | 15.8 ± 1.4 | 9.5 ± 0.9 | 8.8 ± 0.8 | <0.001 |
| Plasma albumin (g/dL) | 3.5 ± 0.3 | 3.6 ± 0.3 | 3.5 ± 0.2 | 3.6 ± 0.3 | 0.456 |
| B/A ratio | 0.42 ± 0.05 | 0.40 ± 0.04 | 0.25 ± 0.03 | 0.23 ± 0.02 | <0.001 |
| Phototherapy duration (h) | 48.5 ± 12.3 | 42.1 ± 10.5 | 0.0 ± 0.0 | 0.0 ± 0.0 | <0.001 |
Note: GDM, gestational diabetes mellitus; TSB, total serum bilirubin; B/A, bilirubin-to-albumin ratio (milligram of bilirubin per gram of albumin); AABR, automated auditory brainstem response; BAEP, brainstem auditory evoked potential; ILD, interaural latency difference. Continuous variables are mean ± SD; categorical are n (%). Given the small absolute number of carriers, particularly for mtDNA mutations (n = 11), 95% CIs for proportions are wide and meaningful between-group imbalance cannot be excluded.
Hearing Impairment Incidence
The GDM-exposed hyperbilirubinemia group showed a significantly higher rate of severe hearing impairment (34.7%) compared with the non-GDM hyperbilirubinemia (15.1%), GDM-only (11.9%), and healthy control (4.0%) groups (χ2 = 24.8, P < 0.001). The any impairment rate was markedly elevated in the GDM-exposed group (63.3%, χ2 = 38.5, P < 0.001) [Table 2].
Table 2.
Rate of Hearing Impairment between the Four Groups n (%)
| Hearing Outcome | GDM-Exposed Hyperbilirubinemia (n = 98) | Non-GDM Hyperbilirubinemia (n = 185) | GDM-only (n = 42) | Healthy Controls (n = 25) | P |
|---|---|---|---|---|---|
| Severe impairment | 34 (34.7%) | 28 (15.1%) | 5 (11.9%) | 1 (4.0%) | <0.001 |
| Subclinical deficit/at risk | 28 (28.6%) | 35 (18.9%) | 7 (16.7%) | 2 (8.0%) | 0.032 |
| Any impairment | 62 (63.3%) | 63 (34.1%) | 12 (28.6%) | 3 (12.0%) | <0.001 |
Note: GDM, gestational diabetes mellitus. Pairwise comparisons were performed using chi-square or Fisher’s exact tests with Bonferroni correction (six comparisons per outcome). Severe impairment: GDM + hyper vs. non-GDM hyper, P_adj = 0.002; GDM + hyper vs. GDM-only, P_adj = 0.065; GDM + hyper vs. controls, P_adj = 0.032; non-GDM hyper vs. GDM-only, P_adj > 0.99; non-GDM hyper vs. controls, P_adj > 0.99; GDM-only vs. controls, P_adj > 0.99. Any impairment: GDM + hyper vs. non-GDM hyper, P_adj < 0.001; GDM + hyper vs. GDM-only, P_adj = 0.002; GDM + hyper vs. controls, P_adj < 0.001; non-GDM hyper vs. GDM-only, P_adj > 0.99; non-GDM hyper vs. controls, P_adj = 0.273; GDM-only vs. controls, P_adj > 0.99. Subclinical deficit/at-risk: no pairwise contrast survived correction (all P_adj > 0.05).
GJB2 Carrier-Specific Threshold Elevation
Among the 36 neonates carrying a pathogenic GJB2 variant, the mean BAEP threshold elevation at 3 months relative to the healthy control mean was 9.2 ± 3.1 dB, compared with 3.1 ± 1.8 dB in the 314 noncarriers (independent-samples t test, t = 4.12, P = 0.002). This carrier-specific elevation was most pronounced within the GDM-exposed hyperbilirubinemia subgroup.
Univariate and Multivariable Logistic Regression
Univariate logistic regression identified GDM exposure, peak TSB (per 1 mg/dL), B/A ratio (per 0.1 unit), phototherapy duration (per 24 h), GJB2 variant carriage, SLC26A4 variant carriage, mtDNA mutation carriage, MDA at 72 h (per 1 nmol/mg), SOD at 72 h (per 1 U/mg) and maternal HbA1c (per 1%) as significant predictors of hearing impairment (all P < 0.05). Gestational age and birth weight were not significant and were therefore not retained. After adjustment in multivariable logistic regression, GDM exposure (aOR = 2.41, 95% CI: 1.38–4.21, P = 0.002), B/A ratio per 0.1 unit (aOR = 2.23, 95% CI: 1.47–3.39, P < 0.001), GJB2 variant carriage (aOR = 2.87, 95% CI: 1.49–5.52, P = 0.002), SLC26A4 variant carriage (aOR = 2.19, 95% CI: 1.01–4.75, P = 0.047), MDA at 72 h (aOR = 1.35, 95% CI: 1.11–1.64, P = 0.003), SOD at 72 h (aOR = 0.78, 95% CI: 0.62–0.98, P = 0.033), and maternal HbA1c (aOR = 1.43, 95% CI: 1.01–2.03, P = 0.044) remained independent predictors of hearing impairment. Peak TSB, phototherapy duration and mtDNA mutation carriage were no longer significant after adjustment, suggesting that their univariate associations were mostly influenced by correlated covariates, particularly B/A ratio and GDM status [Table 3].
Table 3.
Univariate and Multivariable Logistic Regression Analyses for Hearing Impairment
| Variable | Univariate |
Multivariable |
||
|---|---|---|---|---|
| OR (95% CI) | P | aOR (95% CI) | P | |
| GDM exposure (yes vs. no) | 2.85 (1.72–4.73) | <0.001 | 2.41 (1.38–4.21) | 0.002 |
| Peak TSB (per 1 mg/dL) | 1.34 (1.12–1.60) | 0.001 | 1.18 (0.96–1.45) | 0.112 |
| B/A ratio (per 0.1 unit) | 2.76 (1.89–4.03) | <0.001 | 2.23 (1.47–3.39) | <0.001 |
| Phototherapy duration (per 24 h) | 1.28 (1.09–1.50) | 0.003 | 1.15 (0.95–1.39) | 0.148 |
| GJB2 variant (carrier vs. non) | 3.12 (1.68–5.79) | <0.001 | 2.87 (1.49–5.52) | 0.002 |
| SLC26A4 variant (carrier vs. non) | 2.45 (1.18–5.09) | 0.016 | 2.19 (1.01–4.75) | 0.047 |
| mtDNA mutation (carrier vs. non) | 2.91 (1.15–7.36) | 0.024 | 2.53 (0.94–6.81) | 0.066 |
| MDA at 72 h (per 1 nmol/mg) | 1.52 (1.28–1.81) | <0.001 | 1.35 (1.11–1.64) | 0.003 |
| SOD at 72 h (per 1 U/mg) | 0.71 (0.58–0.87) | 0.001 | 0.78 (0.62–0.98) | 0.033 |
| Maternal HbA1c (per 1%) | 1.68 (1.22–2.31) | 0.002 | 1.43 (1.01–2.03) | 0.044 |
| Gestational age (per week) | 0.85 (0.67–1.08) | 0.182 | — | — |
| Birth weight (per 100 g) | 0.94 (0.84–1.05) | 0.263 | — | — |
Note: GDM, gestational diabetes mellitus; TSB, total serum bilirubin; B/A, bilirubin-to-albumin ratio (mg bilirubin per g albumin); carrier vs. non, carrier vs. noncarrier.
Cox Proportional Hazards Regression
As part of sensitivity analysis with interval-censored ascertainment, Cox regression demonstrated a significant dose–response relationship across exposure groups. Compared with healthy controls, GDM-exposed hyperbilirubinemia neonates had the highest risk of hearing impairment (adjusted HR = 3.89, 95% CI: 1.88–8.05), followed by the non-GDM hyperbilirubinemia group (adjusted HR = 2.58, 95% CI: 1.52–4.38). The GDM-only group showed a non-significant trend (adjusted HR = 1.52, 95% CI: 0.79–2.92). The proportional hazards assumption was met for all covariates (Schoenfeld residual test, all P > 0.05) [Table 4].
Table 4.
Cox Proportional Hazards Regression for Time to Hearing Impairment
| Group | Events, n (%) | Unadjusted HR (95% CI) | P | Adjusted HR (95% CI)† | P |
|---|---|---|---|---|---|
| Healthy controls (ref) | 3 (12.0%) | 1.00 | — | 1.00 | — |
| GDM-only | 12 (28.6%) | 1.67 (0.89–3.14) | 0.112 | 1.52 (0.79–2.92) | 0.209 |
| Non-GDM hyper | 63 (34.1%) | 2.91 (1.76–4.82) | <0.001 | 2.58 (1.52–4.38) | <0.001 |
| GDM + hyper | 62 (63.3%) | 4.32 (2.15–8.67) | <0.001 | 3.89 (1.88–8.05) | <0.001 |
Note: GDM, gestational diabetes mellitus; hyper, hyperbilirubinemia; †Adjusted for maternal HbA1c, GDM treatment modality, gestational age, birth weight, peak TSB and phototherapy duration.
Interaction Between GDM and Hyperbilirubinemia on Multiplicative and Additive Scales
We assessed the interaction between GDM and hyperbilirubinemia on both multiplicative and additive scales. On the multiplicative scale, the GDM × hyperbilirubinemia product term in multivariable logistic regression was not statistically significant (OR = 0.99, 95% CI: 0.65–1.52; likelihood-ratio test P = 0.96). By contrast, additive interaction analysis revealed a positive supra-additive effect, with RERI = 1.05 (95% CI: 0.15–2.42), AP = 0.20 (95% CI: 0.03–0.38) and synergy index S = 1.33 (95% CI: 1.05–1.68). These findings indicate that GDM and hyperbilirubinemia do not interact multiplicatively, but their joint effect exceeds the sum of their individual effects on the absolute-risk scale, consistent with additive (public-health) synergy rather than statistical effect-multiplication.
Mediation Analysis
Bootstrapped causal mediation analysis revealed that MDA significantly mediated the effect of GDM–hyperbilirubinemia on hearing impairment, accounting for 37.8% of the total effect (indirect effect β = 0.28, 95% CI: 0.19–0.39, P < 0.001). The direct effect remained significant (β = 0.46, 95% CI: 0.28–0.65), indicating that oxidative stress partially but substantially explains the GDM–hyperbilirubinemia–hearing impairment pathway [Table 5].
Table 5.
Bootstrapped Causal Mediation Analysis: MDA as Mediator
| Effect | Estimate (β) | 95% CI | P |
|---|---|---|---|
| Total effect | 0.74 | 0.52–0.96 | <0.001 |
| Direct effect | 0.46 | 0.28–0.65 | <0.001 |
| Indirect effect (via MDA) | 0.28 | 0.19–0.39 | <0.001 |
| Proportion mediated | 37.8% | 24.1%–51.6% | <0.001 |
Note: Adjusted for maternal HbA1c, GDM treatment modality, gestational age, birth weight, peak TSB, and phototherapy duration.
B/A Ratio as a GDM-Dependent Prognostic Marker
Stratified logistic regression using a B/A ratio cutoff of 5.5 (milligrams of bilirubin per gram of albumin)[25] revealed group-dependent prognostic value. In the GDM-exposed neonates (n = 140; pooled GDM-exposed hyperbilirubinemia and GDM-only), B/A > 5.5 was strongly associated with hearing impairment at 3 months (OR = 4.17, 95% CI: 2.33–7.45; P < 0.001). By contrast, the same threshold was not significantly associated with hearing impairment among non-GDM neonates (OR = 1.18, 95% CI: 0.62–2.24; P = 0.612). The interaction between B/A category (> 5.5 vs. ≤ 5.5) and GDM status was statistically significant (P < 0.05), supporting the use of the B/A ratio as a GDM-dependent indicator rather than a universal prognostic indicator.
Deafness gene carriers (GJB2/SLC26A4/mtDNA variants, n = 65) exhibited elevated hearing impairment rates across all subgroups. Among GJB2 carriers (n = 36), the total hearing impairment rate was 55.6% (20/36), with 33.3% (12/36) developing manifested hearing impairment. SLC26A4 carriers (n = 18) showed a total hearing impairment rate of 61.1% (11/18), and mtDNA mutation carriers (n = 11) had the highest total rate at 72.7% (8/11). In each genetic subgroup, the GDM-exposed hyperbilirubinemia group consistently demonstrated the highest impairment rates (GJB2: 80.0%; SLC26A4: 80.0%; mtDNA: 100.0%), as detailed in Table 6.
Table 6.
Subgroup Analysis of Hearing Impairment Incidence
| Subgroup Variable | Category, n (%) | GDM-exposed Hyperbilirubinemia (n = 98) | Non-GDM Hyperbilirubinemia (n= 185) | GDM-only (n = 42) | Healthy Controls (n = 25) | P |
|---|---|---|---|---|---|---|
| Genetic variant type | ||||||
| GJB2 variants | Any impairment | 8 (80.0%) | 10 (50.0%) | 2 (50.0%) | 0 (0.0%) | 0.023 |
| Severe impairment | 6 (60.0%) | 5 (25.0%) | 1 (25.0%) | 0 (0.0%) | 0.045 | |
| Subclinical deficit | 2 (20.0%) | 5 (25.0%) | 1 (25.0%) | 0 (0.0%) | 0.678 | |
| SLC26A4 variants | Any impairment | 4 (80.0%) | 6 (60.0%) | 1 (50.0%) | 0 (0.0%) | 0.112 |
| Severe impairment | 3 (60.0%) | 3 (30.0%) | 0 (0.0%) | 0 (0.0%) | 0.156 | |
| Subclinical deficit | 1 (20.0%) | 3 (30.0%) | 1 (50.0%) | 0 (0.0%) | 0.543 | |
| mtDNA mutations | Any impairment | 3 (100.0%) | 4 (66.7%) | 1 (100.0%) | 0 (0.0%) | 0.098 |
| Severe impairment | 3 (100.0%) | 2 (33.3%) | 0 (0.0%) | 0 (0.0%) | 0.034 | |
| Subclinical deficit | 0 (0.0%) | 2 (33.3%) | 1 (100.0%) | 0 (0.0%) | 0.123 |
Note: GDM, gestational diabetes mellitus. The mtDNA-exposed subgroup (n = 3 in the GDM-exposed cell) is descriptive only; statistical inference is unreliable at this sample size.
DISCUSSION
This prospective birth cohort study provides compelling evidence for a positive joint effect, with additive synergy among GDM, neonatal hyperbilirubinemia and genetic predisposition in the pathogenesis of auditory dysfunction in neonates. Our findings demonstrated that neonates exposed to GDM and hyperbilirubinemia exhibited a 2.3-fold higher incidence of severe hearing loss compared with those with hyperbilirubinemia alone (34.7% vs. 15.1%, P < 0.001). Furthermore, genetic susceptibility, particularly in carriers of GJB2 variants, significantly amplified this risk, with 55.6% of GJB2 carriers developing hearing impairment. The B/A ratio emerged as a critical GDM-dependent prognostic marker, with a B/A ratio >5.5 predicting hearing loss in GDM-exposed neonates (OR, 4.17; 95% CI, 2.33–7.45). These results underscore the importance of integrated screening for maternal glycemic status, neonatal bilirubin dynamics and deafness-associated genetic variants to identify neonates at high risk for auditory dysfunction.
The observed interaction between GDM and hyperbilirubinemia in inducing auditory neurotoxicity can be attributed to oxidative stress pathways, which are exacerbated by maternal metabolic dysregulation. GDM is known to induce a state of chronic oxidative stress in the fetus due to hyperglycemia and dyslipidemia, leading to increased production of reactive oxygen species (ROS) and impaired antioxidant defenses.[26,27] Hyperbilirubinemia, typically managed with phototherapy, can induce oxidative stress, particularly when bilirubin levels exceed the binding capacity of albumin, resulting in free bilirubin that can cross the blood–brain barrier and accumulate in the auditory brainstem nuclei.[28] The combination of these two conditions likely creates a “perfect storm” for oxidative damage to the cochlea and auditory pathways, especially in neonates with genetic variants that increase cochlear vulnerability.
Our study identified GJB2 variants as a key modifier of this interaction. GJB2 encodes connexin 26, a gap junction protein crucial for potassium recycling in the cochlea, and mutations in this gene are the most common cause of prelingual non-syndromic hearing loss.[29] The presence of GJB2 variants likely disrupts cochlear homeostasis, making the auditory system highly susceptible to bilirubin-induced oxidative damage. This is supported by our finding that GJB2 carriers exhibited an approximately threefold greater BAEP threshold elevation than non-carriers (9.2 ± 3.1 dB vs. 3.1 ± 1.8 dB; t = 4.12, P = 0.002). Carriers of the SLC26A4 and mtDNA variants showed a rapid deterioration in hearing, further emphasizing the role of genetic predisposition in this context.
Previous studies have established hyperbilirubinemia as a risk factor for neonatal hearing loss, particularly in the context of kernicterus.[30] However, the role of GDM in potentiating this risk has been insufficiently explored. A recent meta-analysis by Chen et al. found that GDM is associated with a 1.5-fold increased risk of neonatal hearing impairment, but this analysis did not account for the interaction with hyperbilirubinemia or genetic factors.[31] Our study extended these findings by demonstrating that the combination of GDM and hyperbilirubinemia significantly amplifies the risk beyond what is observed with either condition alone. Additionally, although genetic variants in GJB2 and SLC26A4 are well-known causes of congenital hearing loss, their interaction with environmental factors such as hyperbilirubinemia has not been thoroughly investigated.[32] Our results were aligned with the gene–environment interaction model proposed by Smith et al., which suggests that genetic susceptibility can modify the effects of perinatal insults on auditory development.[33]
The findings of this study have significant clinical implications for the management of neonates at risk for hearing loss. Current risk stratification models for neonatal hyperbilirubinemia primarily focus on TSB levels and phototherapy thresholds.[12] However, our results suggest that the B/A ratio is a sensitive predictor of auditory toxicity in GDM-exposed neonates, particularly when it exceeds 5.5. This highlights the need to incorporate maternal glycemic status and neonatal albumin levels into risk assessments for hyperbilirubinemia-induced hearing loss. The high prevalence of subclinical hearing deficits (28.6%) in the GDM-exposed hyperbilirubinemia group underscored the importance of early and comprehensive audiological evaluations, including BAEP and AABR, to detect severe and subtle auditory impairments. Given that subclinical deficits may progress over time, longitudinal monitoring is essential for prompt intervention. Our study also advocates for the incorporation of genetic screening into routine neonatal care, particularly for deafness-associated genes such as GJB2, SLC26A4, and mtDNA. The early identification of genetic carriers may provide personalized treatment strategies, such as aggressive phototherapy or antioxidant therapies, to mitigate the risk of hearing loss in high-risk neonates.
LIMITATIONS
Despite the strengths of this study, including its prospective design and comprehensive audiological and genetic assessments, several limitations must be acknowledged. First, the sample size, although adequate for detecting significant differences in the primary outcomes, may have limited the power to detect minor effects in subgroup analyses, particularly for rare genetic variants such as mtDNA mutations. Second, the study was conducted at a single center, which may limit the generalizability of the findings to other populations with different genetic backgrounds or healthcare practices. Third, although we adjusted for several maternal and neonatal confounders, unmeasured variables such as maternal medication use or neonatal sepsis could have influenced the results. Fourth, the follow-up period was limited to 3 months, and longer-term studies are needed to assess the persistence or progression of hearing deficits identified in this cohort. Finally, although interaction between GDM and hyperbilirubinemia was formally tested on both multiplicative and additive scales, the two analyses yielded discordant results: the multiplicative interaction was not statistically significant (likelihood-ratio test P = 0.96), whereas the additive interaction was positive and significant (RERI = 1.05, 95% CI: 0.15–2.42; synergy index = 1.33, 95% CI: 1.05–1.68). This scale-dependent pattern indicates that the joint effect of GDM and hyperbilirubinemia exceeds the sum of their individual effects on the absolute-risk (public-health) scale, but is not amplified on the odds-ratio scale. References to “synergy” in this study should therefore be interpreted in the additive sense rather than as evidence of statistical effect-multiplication. Confirmation in larger, multicenter cohorts with a greater number of events is warranted to characterize this asymmetric interaction structure more precisely, and to test gene–environment interactions that the present sample was underpowered to evaluate formally.
CONCLUSION
This study establishes GDM as a critical modifier of hyperbilirubinemia-induced auditory toxicity, mediated by oxidative stress pathways and exacerbated by genetic susceptibility, particularly in GJB2 carriers. The B/A ratio emerged as a GDM-dependent prognostic marker, highlighting the need for integrated screening of maternal glycemic status, neonatal bilirubin dynamics and deafness-associated genes. These findings advocate for a paradigm shift in the management of neonatal hyperbilirubinemia, emphasizing the importance of personalized risk assessment and early intervention in high-risk populations. Future research should focus on validating these findings in large, multicenter cohorts and exploring targeted therapeutic strategies to mitigate the auditory risks associated with GDM and hyperbilirubinemia.
Availability of Data and Materials
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.
Author Contributions
LiQin Ruan and XiaoXiao Su made substantial contributions to conception, design and drafting the manuscript; LingJingYan acquisition of data, analysis and interpretation of data; Zong Zheng made contributions to reviewing and editing the manuscript. All authors read and approved the final manuscript. All authors have participated sufficiently in the work to take public responsibility for appropriate portions of the content and agreed to be accountable for all aspects of the work in ensuring that questions related to it.
Ethics Approval and Consent to Participate
Written informed consent was obtained from the parents or legal guardians of all participating neonates prior to enrollment. The consent process included a comprehensive explanation of the study procedures, serial audiological assessments, and the implications of genetic testing for deafness-associated variants (GJB2, SLC26A4, and mtDNA). The study protocol was approved by the Ethics Committee of the First Affiliated Hospital of Wenzhou Medical University, Yuhuan Branch (approval code: 2023[58]. Participation was voluntary, and parents could withdraw their child at any time without affecting clinical care.
Conflicts of Interest
The authors declare that they have no conflicts of interest or competing interests.
Acknowledgment
The authors thank the patients who participated in this study.
Funding Statement
This research was supported and funded by the Zhejiang Medical and Health Science and Technology Plan Project (2023XY089).
REFERENCES
- 1.McIntyre HD, Catalano P, Zhang C, Desoye G, Mathiesen ER, Damm P. Gestational diabetes mellitus. Nat Rev Dis Primers. 2019;5:47. doi: 10.1038/s41572-019-0098-8. [DOI] [PubMed] [Google Scholar]
- 2.Wang H, Li N, Chivese T, et al. IDF diabetes atlas: estimation of global and regional gestational diabetes mellitus prevalence for 2021 by International Association of Diabetes in Pregnancy Study Group’s Criteria. Diabetes Res Clin Pract. 2022;183:109050. doi: 10.1016/j.diabres.2021.109050. [DOI] [PubMed] [Google Scholar]
- 3.Shapiro SM, Popelka GR. Auditory impairment in infants with kernicterus. Semin Perinatol. 2011;35:121–6. doi: 10.1053/j.semperi.2011.02.011. [DOI] [PubMed] [Google Scholar]
- 4.Watchko JF, Tiribelli C. Bilirubin-induced neurologic damage—mechanisms and management. N Engl J Med. 2013;369:2021–30. doi: 10.1056/NEJMra1308124. [DOI] [PubMed] [Google Scholar]
- 5.González JC, Corujo-Santana C, Borkoski-Barreiro SA, Ramos-Macías A. Neonatal hyperbilirubinemia as a risk factor for hearing loss. Curr Pediatr Res. 2017;21:460–4. [Google Scholar]
- 6.Watchko JF. Bilirubin-induced neurotoxicity in the preterm neonate. Clin Perinatol. 2016;43:297–311. doi: 10.1016/j.clp.2016.01.007. [DOI] [PubMed] [Google Scholar]
- 7.Zhang SF, Li Y, Zhou J. TLR4/NF-κB-mediated inflammation and autophagy in GDM placentas. Placenta. 2024;145:12–21. [Google Scholar]
- 8.Liu T, Zheng W, Wang L, Wang L, Zhang Y. TLR4/NF-κB Signaling pathway participates in the protective effects of apocynin on gestational diabetes mellitus induced placental oxidative stress and inflammation. Reprod Sci. 2020;27:722–730. doi: 10.1007/s43032-019-00078-5. [DOI] [PubMed] [Google Scholar]
- 9.Ahlsson F, Diderholm B, Jonsson B. Association between neonatal hyperbilirubinemia and diabetes mellitus during pregnancy. Acta Paediatr. 2010;99:186–9. [Google Scholar]
- 10.Kelsell DP, Dunlop J, Stevens HP. Connexin 26 mutations in hereditary non-syndromic sensorineural deafness. Nature. 1997;387:80–3. doi: 10.1038/387080a0. [DOI] [PubMed] [Google Scholar]
- 11.Wangemann P. K+ cycling and the endocochlear potential. Hear Res. 2002;165:1–9. doi: 10.1016/s0378-5955(02)00279-4. [DOI] [PubMed] [Google Scholar]
- 12.Kemper AR, Newman TB, Slaughter JL, et al. Clinical practice guideline revision: management of hyperbilirubinemia in the newborn infant 35 or more weeks of gestation. Pediatrics. 2022;150:e2022058859. doi: 10.1542/peds.2022-058859. [DOI] [PubMed] [Google Scholar]
- 13.Gu H, Zhong XM, Wang YQ, Yang JG, Cai Y. Brainstem auditory evoked potential combined with high resolution cranial base CT can optimize the diagnosis of auditory nerve injury. Chin J Traumatol. 2022;25:156–60. doi: 10.1016/j.cjtee.2021.12.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Zhou T, Xu X, Liu J. Subclinical auditory deficits in GDM-exposed neonates. Front Pediatr. 2023;11:1098765. [Google Scholar]
- 15.Hansen TWR, Wong RJ, Stevenson DK. Molecular physiology and pathophysiology of bilirubin handling by the blood, liver, intestine, and brain in the newborn. Physiol Rev. 2023;103:2075–143. doi: 10.1152/physrev.00004.2019. [DOI] [PubMed] [Google Scholar]
- 16.Bhutani VK, Stark AR, Lazzeroni LC. Predischarge screening for severe neonatal hyperbilirubinemia. Semin Fetal Neonatal Med. 2023;28:101433. [Google Scholar]
- 17.Aggarwal K, Ravi R. Effect of gestational diabetes mellitus on newborn hearing: a systematic review. Ann Otol Rhinol Laryngol. 2025;134:87–92. doi: 10.1177/00034894241287014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.International Association of Diabetes and Pregnancy Study Groups Consensus Panel. Metzger BE, Gabbe SG, Persson B, et al. International association of diabetes and pregnancy study groups recommendations on the diagnosis and classification of hyperglycemia in pregnancy. Diabetes Care. 2010;33:676–82. doi: 10.2337/dc09-1848. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.American Academy of Pediatrics Committee on Fetus and Newborn. American College of Obstetricians and Gynecologists Committee on Obstetric Practice. The Apgar Score. Pediatrics. 2015;136:819–22. [Google Scholar]
- 20.Richards S, Aziz N, Bale S, et al. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med. 2015;17:405–424. doi: 10.1038/gim.2015.30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Olusanya BO, Davis AC, Hoffman HJ. Hearing loss grades and the International Classification of Functioning. Disability and Health. Bull World Health Organ. 2019;97:725–728. doi: 10.2471/BLT.19.230367. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Berg AL, Prieve BA, Serpanos YC, Wheaton MA. Hearing screening in a well-infant nursery: profile of automated ABR-fail/OAE-pass. Pediatrics. 2011;127:269–275. doi: 10.1542/peds.2010-0676. [DOI] [PubMed] [Google Scholar]
- 23.Sininger YS, Hunter LL, Hayes D, Roush PA, Uhler KM. Evaluation of speed and accuracy of next-generation auditory steady state response and auditory brainstem response audiometry in children with normal hearing and hearing loss. Ear Hear. 2018;39:1207–1223. doi: 10.1097/AUD.0000000000000580. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Imai K, Keele L, Tingley D. A general approach to causal mediation analysis. Psychol Methods. 2010;15:309–334. doi: 10.1037/a0020761. [DOI] [PubMed] [Google Scholar]
- 25.Iskander I, Gamaleldin R, El Houchi S, et al. Serum bilirubin and bilirubin/albumin ratio as predictors of bilirubin encephalopathy. Pediatrics. 2014;134:e1330–9. doi: 10.1542/peds.2013-1764. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Lappas M, Hiden U, Desoye G, et al. The role of oxidative stress in the pathophysiology of gestational diabetes mellitus. Antioxid Redox Signal. 2011;15:3061–3100. doi: 10.1089/ars.2010.3765. [DOI] [PubMed] [Google Scholar]
- 27.Coughlan MT, Vervaart PP, Permezel M, Georgiou HM, Rice GE. Altered placental oxidative stress status in gestational diabetes mellitus. Placenta. 2004;25:78–84. doi: 10.1016/S0143-4004(03)00183-8. [DOI] [PubMed] [Google Scholar]
- 28.Brites D, Fernandes A. Bilirubin-induced neural impairment: a special focus on myelination, age-related windows of susceptibility and associated co-morbidities. Semin Fetal Neonatal Med. 2015;20:14–9. doi: 10.1016/j.siny.2014.12.002. [DOI] [PubMed] [Google Scholar]
- 29.del Castillo FJ, del Castillo I. DFNB1 non-syndromic hearing impairment: mutations of GJB2 and del (GJB6-D13S1830) and their genotype-phenotype correlations. Acta Otolaryngol. 2017;137:1175–1179. [Google Scholar]
- 30.Shapiro SM. Bilirubin toxicity in the developing nervous system. Pediatr Neurol. 2003;29:410–421. doi: 10.1016/j.pediatrneurol.2003.09.011. [DOI] [PubMed] [Google Scholar]
- 31.Li X, Bu W, Hu X, Han T, Xuan Y. Gestational diabetes mellitus and the hearing of newborns: A nested case-control study in tropical province of China. Int J Pediatr Otorhinolaryngol. 2024;184:112056. doi: 10.1016/j.ijporl.2024.112056. [DOI] [PubMed] [Google Scholar]
- 32.Morton CC, Nance WE. Newborn hearing screening—a silent revolution. N Engl J Med. 2006;354:2151–2164. doi: 10.1056/NEJMra050700. [DOI] [PubMed] [Google Scholar]
- 33.Smith RJH, Bale JF, Jr, White KR. Sensorineural hearing loss in children. Lancet. 2005;365:879–890. doi: 10.1016/S0140-6736(05)71047-3. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.