Neurodevelopmental outcomes among children with congenital gastrointestinal anomalies using Korean National Health Insurance claims data

This study evaluated NDI in children with CGIA using data from a large national cohort. Children with CGIA exhibited higher NDI rates than healthy controls; however, most children with CGIA did not experience severe NDI. EA, JIA, ARM, and CMC were significantly associated with NDI. Among the risk factors, GF regarding height and weight correlated with NDI. Clinical factors such as the duration of oxygen support, mechanical ventilation, PN, and the number of surgeries significantly influenced NDI.

In this study, the CGIA score was linked to NDI at 3 years of age. Previous studies reported variable neurodevelopmental outcomes across different age groups. Bevilacqua et al. conducted a short-term study at 6 and 12 months of age and observed normal neurodevelopmental outcomes in children with major congenital anomalies⁵. In contrast, Stalwijk et al. reported NDI in children with noncardiac anomalies at 12 and 24 months of age⁷. Roorda et al. conducted a systematic review and meta-analysis of cognitive and motor impairments in patients with non-cardiac congenital malformations up to 24 months of age¹⁶, demonstrating a significant association between CGIA and NDI, partially corroborating our findings.

EA is a severe form of CGIA. In this study, children with EA had the highest NDI among the CGIA group. Several studies have reported the neurodevelopmental outcomes in children with EA^17,18,19,20. Our findings are consistent with those of Mawlana et al.¹⁷, who reported NDI in 24% of children aged 12–36 months with EA. Notably, ARM is associated with NDI, although ARM-specific studies are scarce. Van den Hondel et al., in their prospective long-term follow-up study, reported that non-syndromic children with ARM were at risk of growth impairment. Motor development was slightly impaired at all assessed ages (0.5, 1, 2, and 5)²¹, consistent with our study findings. In contrast, Miyake et al. reported that patients with ARM were more prone to underperforming in junior high school compared with healthy controls and were at a greater risk of NDI²². This tendency was observed only in those in junior high school, but not in kindergarten and lower grades. Moreover, CMC was associated with NDI. These data are consistent with those of Cowap et al., who reported that patients with CMC demonstrated NDI at a preschool age²³. Duodenal atresia is often associated with syndromes. However, since our study excluded children with chromosomal anomalies, non-syndromic DA alone might not show an association with syndromes. Nevertheless, cautious interpretation is necessary due to the small sample size.

The prevalence of NDI was higher among children with CGIA than healthy controls, with varying rates depending on the CGIA type. This increased prevalence is attributed to the complexity of the disease, which is supported by the following observations: (1) CGIA occurs early in development²⁴; and (2) it is closely associated with other organs such as the trachea, lung, bile tract, and genitourinary tract²⁵. Furthermore, CGIA are associated with a high risk of concomitant malformations. Multi-organ malformations, including vertebral anomalies, anal atresia, cardiac defects, tracheoesophageal fistulas, renal anomalies, and limb abnormalities, may accompany CGIA^25,26. This intricate association may influence the course, complications, and treatment of CGIA.

Risk factors influencing neurodevelopmental outcomes in children with CGIA are likely multifactorial^5,7. This study revealed a significantly elevated risk of NDI among children with CGIA, even after excluding infants with low birth weight and chromosomal/brain anomalies.

Growth failure in height and weight was correlated with NDI. Previous studies have demonstrated a link between head growth and neurodevelopmental outcomes^27,28,29,30. Additionally, several studies have linked increased height, weight, and neurodevelopmental outcomes^31,32. For patients with CGIA undergoing surgery within the first year after birth, growth patterns may reflect nutritional challenges arising from surgery and its complications.

In this study, CHD showed no association with NDI in CGIA cases. The etiology of NDI involves multifactorial interplay, including cardiac conditions, genetics, perioperative brain injuries like hypoxia, general anesthesia, cardiopulmonary bypass surgery, and postoperative events^33,34. Children with complex CHD are particularly vulnerable to cerebral ischemia. Since most patients with CHD in this study had septal defects, NDI may not be significantly linked to CHD at age 3 years (Supplementary Table S1).

Here, we investigated the influence of clinical factors on NDI, an indicator of disease severity. In this study, clinical factors were found to be significant in children with JIA, ARM, and total CGIA. Disease severity might be affected more by postnatal conditions because JIA does not feature a relatively high risk of syndromic or associated congenital anomalies compared with other CGIA²⁵. Moreover, there might be no association with NDI in children with EA, DA, or CMC, but the results should be interpreted cautiously given the limited sample size. PN duration reflects complications that affect neurodevelopmental outcomes^17,35,36,37. PN duration correlated with NDI in children with JIA. JIA typically require surgical correction and PN support during the neonatal period. Short bowel syndrome (SBS), characterized by severe malabsorption due to bowel loss in patients with JIA, often leads to PN dependence^36,37. The severity of SBS depends on the length of the remaining small intestine, presence of the ileocecal valve, and PN duration, leading to GF and adversely affects cognition and learning in preschool-aged children^36,37,38.

In children with CGIA, NDI severity is influenced by oxygen support duration, mechanical ventilation use, and the number of surgeries. Duration of oxygen support and mechanical ventilation serve as clinical markers that reflect the complexity of the underlying condition. Patients with CGIA frequently undergo multiple surgeries and interventions under general anesthesia^13,22, exposing them to various stressors during rapid growth and brain development. During this period, neuronal migration and cortical folding reach maturity, whereas myelination, dendritic branching, and axonal arborization progress continuously^39,40,41. Furthermore, patients with CGIA encounter stress factors, such as intubation, mechanical ventilation, and medications, which may cause vital instability in blood pressure, perfusion, and oxygenation during surgery³⁶.

The major strength of our study lies in its robust estimates of NDI in children with CGIA without known high-risk factors, utilizing nationwide live-born data from the NHIS database. This approach enhances the generalizability of our results and minimizes bias. Overall, our study contributes valuable insights into the literature on children with CGIA and serves as a reference guide for counseling and public health planning.

However, this study had several limitations, due to its retrospective design. Firstly, excluding children with CGIA due to missing developmental assessments may introduce selection bias, potentially overlooking those with severe NDI. Accurately representing the overall population becomes challenging due to the varied types and severity levels within each CGIA, compounded by the small sample size. Second, NDI was defined based on the results of developmental assessments using a questionnaire completed by the parents. However, these developmental assessments are simple and validated tests for screening, with good discrimination power in general. Third, we lacked data on gestational age, mode of delivery, height, and HC at birth because this information was not provided by the NHI and was consequently unavailable in the database. Fourth, detailed individual medical information regarding disease subtypes, including anatomical variations, sepsis, reoperation, and surgical complications, was not available. Factors influencing the difficulty of surgical treatment and prognosis were excluded. Despite these limitations, this study is significant because it was conducted using currently available data. However, owing to the constraints of the study methodology, direct application in clinical practice is challenging. Therefore, well-defined follow-up studies should be conducted in additional prospective cohorts.

In conclusion, children diagnosed with CGIA are at a high risk of adverse NDI. Although our study identified specific risk factors for NDI within certain CGIA subgroups, including growth failure and clinical factors, most children with CGIA do not develop severe NDI. These findings emphasize the importance of cautious and proactive monitoring of neurodevelopmental milestones in this population to ensure timely interventions and optimize outcomes rather than causing unnecessary concern among families. Systemic long-term follow-up is essential for the early detection of NDI and timely and tailored interventions.