To editor: Congenital heart diseases (CHDs) are the most prevalent congenital anomaly worldwide. The outcomes of CHDs are influenced not only by anatomical complexity and clinical severity but also by maternal–fetal biological interactions. Although genetic and environmental determinants of CHDs are well-recognized, maternal–fetal immuno-hematological dynamics, particularly ABO blood group and Rhesus (Rh) antigen discordance, remain underexplored despite their established relevance in hemolytic disease of the newborn.1 The ABO blood group system is determined by three alleles at the ABO locus on chromosome 9q34. The A and B alleles encode glycosyltransferases that modify the H antigen by adding N-acetylgalactosamine (GalNAc) or galactose, respectively, whereas the O allele yields a truncated, non-functional protein.2 ABO discordance can attenuate Rh alloimmunization by promoting rapid clearance of incompatible fetal erythrocytes and influencing placental immune modulation.3 Emerging evidence further suggests that glycosyltransferase-mediated ABO pathways may intersect with developmental signaling cascades involved in cardiovascular morphogenesis.4 However, large-scale studies evaluating whether analogous immune-hematological mechanisms influence CHD severity are lacking. We conducted an ethics-approved retrospective cohort study at a high-volume, free-of-cost tertiary pediatric cardiac center in North India. The study was approved by the Institutional Ethics Committee of Sri Sathya Sai Sanjeevani Research Foundation, Palwal, Haryana, India (PS00002/IEC/5/2018) and was conducted in accordance with the guidelines laid down by the Declaration of Helsinki. Informed consent has been obtained from the participants, and from the parents or legal guardian in case of children. The study included 900 singleton children with CHD (aged 0–12 years) who underwent surgical correction between March 2022 and July 2025, along with their biological mothers of Indian ethnicity. Dyads with recent blood/platelet transfusion, syndromic features, extra-cardiac anomalies, or chronic disorders based on clinical history were excluded from the study. Maternal–child ABO and Rh blood groups, sociodemographic characteristics, and clinical data were collected on REDCap using a validated in-house questionnaire. CHD severity was classified according to the Risk Adjustment in Congenital Heart Surgery (RACHS) score into simple (RACHS 1), moderate (RACHS 2), and complex (RACHS ≥ 3) categories, along with cyanosis status.5 ABO and Rh discordance were defined using the standard compatibility matrices. Comparative analyses and ordinal regression models were performed using SPSS version 22.0 (IBM Corp., Armonk, NY, USA). Variables with a univariate association threshold of P ≤ 0.20 were included in multivariate models following a purposeful covariate selection strategy aimed at retaining biologically and clinically relevant confounders that may not demonstrate strong univariate significance.6 Sensitivity analyses using a strict P-value threshold (≤ 0.05) included age-stratified regression models (0–1 year, > 1–5 years, > 5–12 years) and stratification by cyanotic versus acyanotic CHD to assess potential referral and survivor bias. Blood group B predominated among mothers (35.4% (319/900)) and was similarly predominant in the children (35.1% (316/900)) (Supplementary Table 1, https://links.lww.com/MFM/A121). Maternal ABO distribution differed significantly across RACHS severity categories (P = 0.034), whereas Rh distribution did not (P = 0.224). ABO discordance was observed in 28.6% (257/900) and Rh discordance in 4.9% (44/900) of dyads. Simple (mild) CHDs accounted for only 7.1% (64/900) of the cohort, reflecting the tertiary referral nature of the study center, where children with moderate and complex lesions requiring surgical intervention are preferentially referred. Sociodemographic analyses revealed that lower maternal age at conception correlated with increased CHD complexity (P = 0.009), and lower socioeconomic status was disproportionately represented among complex CHD cases (P 5–12 years) represent a selected survivor population, introducing potential survivor and referral bias. Additionally, referral timing to tertiary centers may differ by age and disease severity. Consequently, these findings should be regarded as hypothesis-generating rather than confirmatory. Previous studies primarily evaluated ABO and Rh groups as risk factors for CHD occurrence, with inconsistent results. Lev et al.7 reported an excess of blood group B among African American children with CHDs (P < 0.02), without significant discordance effects. Furthermore, although limited by its sample size, a study by Brendemoen8 noted an increased risk of ventricular septal defects in children with blood group A. A later study in a large Chinese cohort reported lower CHD risk in blood group A compared to O (OR = 0.82).9 However, a recent study found that maternal group A was associated with an increased CHD risk in patients with Down syndrome (OR = 6.56), while ABO discordance itself was not significant (P = 0.539).10 Importantly, most previous studies focused on disease incidence rather than severity, and none systematically examined maternal–child ABO discordance. Our findings extend the available evidence by demonstrating a potential severity gradient linked to immuno-hematological factors, adjusted for key potential confounders (Supplementary Table 4, https://links.lww.com/MFM/A123). Several biological mechanisms may plausibly explain these associations. ABO gene products influence cell-surface carbohydrate structures, modulating endothelial interactions, coagulation pathways, and immune signaling, which may indirectly affect embryonic cardiovascular development. The proximity of the ABO locus to NOTCH1 and EHMT1 on chromosome 9q34 provides a potential genetic link, as both genes are crucial for semilunar valve formation and outflow tract development, and structural variants or local linkage disequilibrium could influence CHD susceptibility.4 In parallel, glycosylation-dependent pathways are critical in cardiac morphogenesis. Hedgehog (Hh) signaling requires GalNAc-transferase 1-mediated O-linked glycosylation of Sonic Hedgehog (Shh) for appropriate activation. ABO-derived glycosyltransferases may compete with these enzymes, potentially altering Shh activity and consequently modulating CHD severity.9 This study’s strengths include a large, clinically well-characterized cohort, standardized RACHS severity classification, and robust statistical adjustment. Key limitations are its single-center retrospective design, potential referral and survivor bias, and lack of genetic or mechanistic biomarkers to support causal inference. Furthermore, data on maternal environmental exposures and medication dosages during pregnancy were not uniformly available, introducing the possibility of residual confounding. Longitudinal survival analyses were also not performed and may have further clarified age-related patterns. Additionally, multiple subgroup and age-stratified analyses increase the risk of type I error; formal correction for multiple comparisons was not applied; therefore, this study’s findings should be interpreted as exploratory. In conclusion, our findings indicate that maternal–child ABO discordance is associated with CHD severity in children, with potential protective effects observed for specific blood groups. These findings suggest a possible immunogenetic contribution to disease modulation and warrant validation using independent, prospective, multicentric cohorts, alongside mechanistic investigations integrating immunology, genomics, and placental biology. Until such evidence is available, ABO or Rh incompatibility should not be used for individual risk stratification, prognostic counseling, or to guide clinical decision-making. Acknowledgements The authors acknowledge the contribution of the research and IT teams for assistance with data acquisition and management, respectively. Funding This study was supported by intramural funding. Author Contributions S.A.: conceptualization, methodology, data curation, formal analysis, visualization, validation, writing-original draft; A.T.: sample acquisition, clinical validation; A.S.: sample acquisition, clinical Validation; P.K.: conceptualization, validation, supervision, writing-review and editing. S.A. and P.K. take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. Conflicts of Interest The authors declare no conflicts of interest. Data Availability The datasets contain confidential patient information and are available from the corresponding author upon reasonable request.
Ahamad et al. (Wed,) studied this question.