Longitudinal Analysis of Growth and Neurodevelopmental Outcomes in Very Low Birth Weight Infants With Congenital Anomalies Over Three Years

Abstract

Background

Very low birth weight infants (VLBWIs) are vulnerable to growth restrictions and neurodevelopmental impairments. Congenital anomalies further complicate these risks; however, their long-term effects remain unclear. This study examined the impact of congenital anomalies on the growth and neurodevelopment of VLBWIs.

Methods

This prospective cohort study analyzed data from the Korean Neonatal Network (2013–2017). A total of 172 VLBWIs with congenital anomalies were matched by gestational age to 516 without anomalies at 18–24 months corrected age, and 136 were matched to 408 at 3 years of age. Growth was assessed using WHO standards, and neurodevelopment was evaluated using the Bayley Scales of Infant Development (II/III) and Korean Developmental Screening Test. Logistic regression analyses were used to identify factors associated with adverse outcomes, with statistical significance set at P < 0.05.

Results

VLBWIs with congenital anomalies had significantly lower weight, height, and head circumference z-scores at both time points. Growth restriction persisted, and neurodevelopmental delays, particularly in motor function, were more prevalent. Infants with multiple congenital anomalies had the highest risk of severe growth restriction and developmental impairment.

Conclusion

Congenital anomalies pose significant challenges to the growth and neurodevelopment of VLBWIs. Early and individualized interventions, structured neurodevelopmental follow-up, and multidisciplinary care are essential for improving long-term outcomes.

Graphical Abstract

INTRODUCTION

Infants with congenital anomalies often experience delayed physical growth and developmental milestones, requiring individualized care and interventions to address their unique growth and developmental needs.1,2 Preterm infants, particularly those with low gestational ages and birth weights, are at an increased risk of delayed growth and significant neurodevelopmental challenges.3,4,5 These infants often experience more frequent rehospitalizations and require early intervention services.6 Among them, very low birth weight infants (VLBWIs) represent a particularly vulnerable subgroup that accounts for a significant proportion of the neonatal population in need of specialized medical attention.7 Advances in neonatal intensive care have improved survival rates of these infants; however, questions remain regarding their long-term functional outcomes. The association between congenital anomalies and developmental outcomes in VLBWIs is complex and involves many factors.8 Congenital anomalies may further amplify the developmental risks already associated with preterm birth in these infants.9,10 The coexistence of premature birth and congenital anomalies may have a compounded impact, potentially increasing the risk of brain injury and neurodevelopmental impairment due to factors such as chronic hypoxia, metabolic instability, nutritional deficits, frequent surgeries, and prolonged exposure to stressful environments.11,12

Previously, we conducted a nationwide cohort study of VLBWIs with congenital anomalies and analyzed their short-term neonatal outcomes, including mortality and major morbidities, during their stay in the neonatal intensive care unit (NICU).13 That study demonstrated that congenital anomalies were associated with significantly higher in-hospital mortality and increased risks of bronchopulmonary dysplasia (BPD) and severe intraventricular hemorrhage (IVH) compared with observations of VLBWIs without anomalies. However, there is an important need to extend these findings and investigate how congenital anomalies influence long-term outcomes such as growth trajectories and neurodevelopment.

Building on previous findings, this study aimed to evaluate the long-term impact of congenital anomalies on post-discharge growth and neurodevelopment in VLBWIs. Specifically, we examined differences in growth parameters and key neurodevelopmental outcomes such as motor and mental developmental delays and overall neurodevelopmental impairment between infants with and without congenital anomalies at two critical time points: 18–24 months corrected age (CA) and 3 years of age. By analyzing outcomes longitudinally, we sought to clarify the extent to which these comorbid conditions affect developmental trajectories and to provide insights that may inform targeted follow-up and intervention strategies.

METHODS

Study design

Since 2013, the Korean Neonatal Network (KNN) has been a collaborative healthcare initiative, currently involving 78 NICUs.14,15 KNN's main focus is the healthcare of VLBWIs born weighing less than 1,500 g. In this study, conducted between January 2013 and December 2017, 10,387 VLBWIs from the KNN, a national neonatal registry for VLBWIs in Korea, were selected for analysis. The exclusion criteria were as follows: infants with inborn errors of metabolism (n = 3), preterm birth at 22 weeks of gestation (n = 27), congenital infections (n = 128), and infants who did not survive until discharge (n = 1,373). After excluding cases of post-discharge mortality (n = 80) and missing data on growth and development at 18–24 months CA (n = 3,189) and 3 years of age (n = 4,777), the final group of participants was selected for comparison. Using 1:3 frequency matching for gestational age, 172 VLBWIs with congenital anomalies were compared to 516 without anomalies at 18–24 months CA, and 136 VLBWIs with anomalies were compared to 408 without anomalies at 3 years of age (Fig. 1).

Fig. 1. Flowchart of study population selection from the Korean Neonatal Network (2013–2017). Very low birth weight infants with and without congenital anomalies were selected after excluding those with inborn errors of metabolism, congenital infections, gestational age < 22 weeks, death before or after discharge, and missing follow-up data. A total of 172 VLBWIs with anomalies and 516 without were included at 18–24 months corrected age, and 136 with anomalies and 408 without at 3 years of age. Frequency matching by gestational age (1:3) was applied.

VLBWI = very low birth weight infant.

Classification of congenital abnormalities

In the present study, congenital anomalies were categorized using a unique classification system developed by the KNN. This system, while referencing the International Classification of Diseases and European Surveillance of Congenital Anomalies, has its own distinct structure. Within the KNN classification, congenital anomalies are grouped into eight subcategories according to the affected organ system (Supplementary Table 1). Each case was classified into more than one subgroup if multiple organ systems were involved. For example, an infant with both an imperforate anus and a ventricular septal defect would be classified under the digestive system and congenital heart defects categories, respectively. In this study, the presence of multiple congenital anomalies (MCA) was defined as the involvement of two or more subgroups.

Data collection / milestone analysis

In our study of VLBWIs with congenital anomalies, we conducted a comprehensive evaluation of both growth and developmental outcomes at two key stages: 18–24 months CA and 3 years of age. For growth assessment, we evaluated weight, height, and head circumference, and growth restriction was defined in accordance with the World Health Organization's Child Growth Standards. Specifically, growth restriction was defined as a z-score below −1.28, corresponding to less than the 10th percentile for any of the three parameters (weight, height, or head circumference).16 Developmental outcomes were assessed using multiple standardized tools. Cognitive and language development were evaluated using the mental development domain of the Bayley Scales of Infant Development, Second Edition (BSID II); the cognitive and language domains of the Bayley Scales of Infant and Toddler Development, Third Edition (BSID III); and the cognition, language, and self-help domains of the Korean Developmental Screening Test for Infants and Children (K-DST). Motor development was assessed using the psychomotor development domain of the BSID II; motor domain of the BSID III; and gross motor and fine motor domains of the K-DST.

Mental developmental delay was defined as a Mental Development Index score below 70 on BSID II, a cognitive or language domain score below 70 on BSID III, or a result of “further evaluation needed” in any cognitive, language, or self-help domain of the K-DST. Motor developmental delay was defined as a Psychomotor Development Index score below 70 on the BSID II, a motor domain score below 70 on the BSID III, or a result of “further evaluation needed” in either the gross or fine motor domain of the K-DST. Cerebral palsy, defined as a Gross Motor Function Classification System level 2 or higher, was also considered a marker of motor delay. Neurodevelopmental impairment was defined as the presence of any significant developmental disability, including mental or motor delays as defined above.17

To account for the use of different developmental assessment tools, we prioritized BSID scores when both BSID and K-DST results were available, ensuring consistency and diagnostic validity. In each case, developmental delay was determined based on a single tool: BSID if available, or K-DST if BSID was not performed. We did not combine results from different tools when defining delay. At 18–24 months CA, among infants with congenital anomalies, 33.7% were assessed with BSID II, 30.8% with BSID III, and 35.5% with K-DST only; among those without anomalies, 34.5%, 26.4%, and 39.1% underwent BSID II, BSID III, and K-DST, respectively. At 3 years of age, BSID assessment became less frequent, whereas K-DST assessments became more common. Although the proportions of assessment tools differed somewhat from those at 18–24 months CA, the distribution between infants with and without congenital anomalies remained relatively balanced.

Statistical analysis

Growth and developmental outcomes were assessed at 18–24 months CA and at 3 years of age. Continuous variables are reported as mean ± standard deviation, and categorical variables as percentages with 95% confidence intervals (CIs). Given the substantial difference in sample sizes between the groups, 1:3 frequency matching was performed based on gestational age to ensure comparability. Subsequent comparisons between the groups were conducted using t tests for continuous variables and χ² tests for categorical variables, as appropriate. Logistic regression analyses were conducted to evaluate the association between congenital anomalies and each outcome, adjusting for small for gestational age (SGA), Apgar score at 5 minutes, maternal chorioamnionitis, pregnancy-induced hypertension, BPD (≥ moderate), and periventricular leukomalacia (PVL). Results were presented as odds ratios (ORs) with 95% CIs, and statistical significance was defined as P < 0.05. Statistical analyses were performed using SPSS Statistics version 29 (IBM Corp., Armonk, NY, USA).

Ethics statement

The data registry was approved by the Institutional Review Board (IRB) of Samsung Medical Center (2013-03-002) and the IRBs of all 78 hospitals participating in the Korean Neonatal Network (KNN). Written consent was obtained from the parents of the infants during their enrollment in the KNN. Data availability was subject to the Act on Bioethics and Safety [Law No. 1518, Article 18: Provision of Personal Information]. Sharing or accessing the data is possible only through the KNN data committee (http://knn.or.kr) and after approval from the Korea Disease Control and Prevention Agency. This study was approved by the KNN Data Management Committee and Kyung Hee University Hospital at Gangdong IRB (Number 2023-12-037).

RESULTS

Prevalence of congenital anomalies

We analyzed 10,387 VLBWIs registered in the KNN between January 2013 and December 2017. Among them, 492 (4.7%) had at least one major congenital anomaly. The digestive system was the most frequently affected (n = 160, 32.5%), followed by congenital heart defects (n = 152, 30.9%), chromosomal anomalies (n = 46, 9.3%), and other anomalies (n = 37, 7.5%). Genitourinary tract (n = 32, 6.5%), central nervous system (n = 27, 5.5%), respiratory system (n = 19, 3.9%), and undefined anomalies (n = 19, 3.9%) were also observed. Ventricular septal defect was the most common congenital heart defect (11.4%). The most common digestive system anomalies were esophageal atresia (6.1%), cleft palate and/or lip (5.7%), and imperforate anus (5.7%). Among chromosomal anomalies, trisomy 18 (3.3%) and trisomy 21 (2.4%) were the most frequent (Supplementary Table 1).

Perinatal and clinical characteristics

After excluding infants with inborn errors of metabolism, congenital infections, and those who did not survive until discharge, 8,776 VLBWIs were eligible for analysis, including 236 infants with congenital anomalies and 8,540 without congenital anomalies. Compared with observations in those without congenital anomalies, infants with congenital anomalies were born at a significantly higher gestational age (30.6 ± 3.2 weeks vs. 29.3 ± 2.8 weeks) but were more likely to be SGA (45.8% vs. 26.7%). They also had lower Apgar scores at 5 min, and maternal pregnancy-induced hypertension was more frequent in this group (14.8% vs. 10.9%), as was moderate to severe BPD (45.3% vs. 27.7%) and PVL (11.9% vs. 6.1%). By contrast, birth weight, sex distribution, multiple gestation rate, and the incidence of other major morbidities, including severe IVH (≥ grade III), necrotizing enterocolitis (≥ stage II), and sepsis, were not significantly different between the groups (Table 1). These baseline differences were generally consistent after matching, with SGA, lower Apgar scores, higher rates of BPD and PVL remaining significantly more prevalent among infants with congenital anomalies at both 18–24 months CA and 3 years of age (Table 2, Supplementary Table 2).

Table 1. Perinatal characteristics and major morbidities of very low birth weight infants with and without congenital anomalies (before matching).

Parameters	With congenital anomalies (n = 236)	Without congenital anomalies (n = 8,540)	P value
Gestational age, wk	30.6 ± 3.2	29.3 ± 2.8	< 0.001
Birth weight, g	1,153.9 ± 247.7	1,131.0 ± 257.5	0.177
Male	44.9 (38.70–51.29)	49.9 (48.9–51.0)	0.145
Multiple gestation	31.8 (26.2–38.0)	35.5 (34.5–36.5)	0.270
Small for gestational age	45.8 (39.5–52.1)	26.7 (25.8–27.7)	< 0.001
Apgar score at 1 min	4.8 ± 2.0	4.9 ± 2.0	0.207
Apgar score at 5 min	6.9 ± 1.8	7.1 ± 1.7	0.048
Antenatal steroid therapy	78.0 (72.3–82.8)	77.0 (76.1–77.9)	0.782
Maternal chorioamnionitis	22.5 (17.6–28.2)	32.0 (31.0–33.0)	0.003
Maternal GDM	7.2 (4.6–11.2)	8.5 (7.9–9.1)	0.569
Maternal PIH	14.8 (10.9–19.9)	20.6 (19.8–21.5)	0.037
Cesarean section	80.1 (74.5–84.7)	79.1 (78.3–80.0)	0.784
Intraventricular hemorrhage ≥ grade III	7.6 (4.9–11.7)	5.5 (5.0–5.6)	0.195
Bronchopulmonary dysplasia ≥ moderate	45.3 (39.1–51.7)	27.7 (26.8–28.7)	< 0.001
Periventricular leukomalacia	11.9 (8.3–16.6)	7.4 (6.8–8.0)	0.014
Necrotizing enterocolitis ≥ stage II	5.5 (3.3–9.2)	4.8 (4.4–5.3)	0.722
Sepsis	22.0 (17.2–27.8)	19.0 (18.1–19.8)	0.270

Values are expressed as mean ± standard deviation or percentage (95% confidence interval).

GDM = gestational diabetes mellitus, PIH = pregnancy-induced hypertension.

Table 2. Baseline characteristics of matched very low birth weight infants with and without congenital anomalies at 18–24 months corrected age.

Parameters	With congenital anomalies (n = 172)	Without congenital anomalies (n = 516)	P value
Gestational age, wk	30.5 ± 3.3	30.2 ± 2.9	0.346
Male	43.0 (35.9–50.5)	45.3 (41.1–49.7)	0.658
Multiple gestation	32.0 (25.5–39.3)	34.3 (30.3–38.5)	0.642
Small for gestational age	45.9 (38.7–53.4)	29.8 (26.1–33.9)	< 0.001
Apgar score at 1 min	5.1 ± 2.1	4.8 ± 2.0	0.035
Apgar score at 5 min	6.9 ± 1.8	7.2 ± 1.7	0.012
Antenatal steroid therapy	78.5 (71.8–84.0)	77.3 (73.5–80.7)	0.833
Maternal chorioamnionitis	28.5 (22.3–35.6)	42.2 (38.1–46.6)	0.002
Maternal GDM	8.7 (5.4–13.9)	6.4 (4.6–8.8)	0.388
Maternal PIH	11.6 (7.7–17.3)	24.8 (21.3–28.7)	< 0.001
Cesarean section	78.5 (71.8–84.0)	80.6 (77.0–83.8)	0.620
Intraventricular hemorrhage ≥ grade III	8.7 (5.4–13.9)	4.7 (3.1–6.8)	0.071
Bronchopulmonary dysplasia ≥ moderate	45.9 (38.7–53.4)	25.2 (21.6–29.1)	< 0.001
Periventricular leukomalacia	11.6 (7.7–17.3)	6.2 (4.4–8.6)	0.030
Necrotizing enterocolitis ≥ stage II	5.8 (3.2–10.4)	3.9 (2.5–5.9)	0.389
Sepsis	19.8 (14.5–26.4)	13.6 (10.9–16.8)	0.065

Values are expressed as mean ± standard deviation or percentage (95% confidence interval).

GDM = gestational diabetes mellitus, PIH = pregnancy-induced hypertension.

Growth and developmental outcomes at 18–24 months CA and 3 years of age

This study analyzed the long-term growth and developmental outcomes of VLBWIs with and without congenital anomalies at two follow-up points: 18–24 months CA and 3 years of age. After frequency matching by gestational age, 172 VLBWIs with congenital anomalies were compared with 516 VLBWIs without congenital anomalies at 18–24 months CA, and 136 VLBWIs with congenital anomalies were compared with 408 VLBWIs without congenital anomalies at 3 years of age (Tables 3 and 4). At both time points, VLBWIs with congenital anomalies consistently had lower growth parameters than those of VLBWIs without anomalies. At 18–24 months CA, weight, height, and head circumference z-scores were significantly low in the congenital anomaly group, with a high prevalence of growth restriction (59.2% vs. 41.1%). Similar trends were observed at 3 years of age, with significantly low weight, height, and head circumference z-scores and a high prevalence of growth restriction (75.8% vs. 56.0%). At each time point, infants with congenital anomalies had consistently lower mean z-scores than those of infants without anomalies. Both groups demonstrated some degree of postnatal catch-up growth between birth and 18–24 months CA, however, the growth disparity persisted and, in some cases, widened by 3 years of age. A higher proportion of infants with congenital anomalies fell below the 10th percentile, demonstrating a greater burden of severe growth restriction persisting through 3 years of age.

Table 3. Comparison of growth and developmental outcomes between very low birth weight infants with and without congenital anomalies at 18–24 months corrected age.

Parameters	With congenital anomalies (n = 172)	Without congenital anomalies (n = 516)	P value
Birth weight	1,151.5 ± 252.1	1,174.3 ± 244.0	0.300
Z-scores of weight	−1.2 ± 1.5	−0.5 ± 1.1	< 0.001
Z-scores of height	−1.0 ± 1.5	−0.4 ± 1.3	< 0.001
Z-scores of head circumference	−1.0 ± 1.6	−0.4 ± 1.3	< 0.001
Weight (z-score < −1.28)	44.5 (37.1–52.2)	21.3 (18.0–25.1)	< 0.001
Height (z-score < −1.28)	37.1 (30.0–44.8)	21.4 (18.0–25.2)	< 0.001
Head circumference (z-score < −1.28)	41.7 (33.6–50.2)	21.4 (17.7–25.7)	< 0.001
Growth restriction	59.2 (51.5–66.4)	36.4 (32.3–40.6)	< 0.001
Mental developmental delay	28.4 (21.4–36.5)	18.9 (15.4–23.1)	0.027
Motor developmental delay	35.5 (28.0–43.8)	16.5 (13.1–20.4)	< 0.001
Neurodevelopmental impairment	39.9 (32.1–48.2)	24.3 (20.3–28.8)	< 0.001
Cerebral palsy	9.8 (5.5–14.7)	4.6 (2.8–6.7)	0.026
Hearing impairment	12.0 (6.8–18.0)	2.3 (0.9–4.0)	< 0.001
Rehabilitative support	53.3 (45.6–60.9)	33.0 (29.0–37.2)	< 0.001
Language support	13.1 (8.3–18.5)	6.4 (4.4–8.6)	0.010

Values are expressed as mean ± standard deviation or percentage (95% confidence interval).

Table 4. Comparison of growth and developmental outcomes between very low birth weight infants with and without congenital anomalies at 3 years of age.

Parameters	With congenital anomalies (n = 136)	Without congenital anomalies (n = 408)	P value
Birth weight	1,128.5 ± 260.6	1,121.8 ± 265.8	0.801
Z-scores of weight	−1.9 ± 1.5	−1.2 ± 1.4	< 0.001
Z-scores of height	−1.6 ± 1.3	−1.0 ± 1.1	< 0.001
Z-scores of head circumference	−1.3 ± 1.5	−0.8 ± 1.1	0.013
Weight (z-score < −1.28)	64.7 (55.0–73.3)	46.0 (40.2–51.9)	< 0.001
Height (z-score < −1.28)	58.1 (48.3–67.4)	39.9 (34.2–45.9)	< 0.001
Head circumference (z-score < −1.28)	45.7 (34.5–57.2)	29.8 (23.8–36.7)	0.009
Growth restriction	75.8 (66.6–83.1)	56.0 (50.1–61.8)	< 0.001
Mental developmental delay	35.7 (25.1–47.9)	22.0 (16.6–28.4)	0.019
Motor developmental delay	35.7 (25.1–47.9)	19.1 (14.1–25.3)	0.003
Neurodevelopmental impairment	47.6 (35.9–59.6)	27.6 (21.7–34.4)	0.001
Cerebral palsy	12.6 (7.1–18.9)	7.6 (5.1–10.4)	0.128
Hearing impairment	9.4 (4.7–15.1)	1.6 (0.3–3.2)	< 0.001
Rehabilitative support	45.7 (37.2–54.3)	23.2 (19.1–27.6)	< 0.001
Language support	31.8 (24.0–40.3)	19.3 (15.5–23.5)	0.005

Values are expressed as mean ± standard deviation or percentage (95% confidence interval).

Developmental outcomes also demonstrated significant differences between the groups. At 18–24 months CA, the congenital anomaly group had high rates of mental developmental delay (28.4% vs. 18.9%) and motor developmental delay (35.5% vs. 16.5%), as well as an increased prevalence of neurodevelopmental impairment (39.9% vs. 24.3%). The incidence rates of cerebral palsy (9.8% vs. 4.6%) and hearing impairment (12.0% vs. 2.3%) were also high in infants with congenital anomalies. These disparities persisted at 3 years of age, with continued increases in mental developmental delay (35.7% vs. 22.0%), motor developmental delay (35.7% vs. 19.1%), and neurodevelopmental impairment (47.6% vs. 27.6%) in the congenital anomaly group. Although the difference in cerebral palsy prevalence was not statistically significant at 3 years of age (12.6% vs. 7.6%, P = 0.128), the rate of hearing impairment (9.4% vs. 1.6%) remained high. Additionally, at both time points, significantly more infants with congenital anomalies received rehabilitation and language support.

Relation between congenital anomalies and long-term outcomes

Multivariate logistic regression analyses were conducted to examine the association between congenital anomalies and growth and developmental outcomes at 18–24 months CA and 3 years of age, adjusting for perinatal and neonatal factors. At 18–24 months CA, congenital anomalies were significantly associated with increased risks of weight, height, and head circumference z-scores below −1.28 (adjusted ORs [AORs], 2.55, 1.79, and 2.15, respectively), as well as overall growth restriction (AOR, 2.12). Motor developmental delay (AOR, 2.34) and neurodevelopmental impairment (AOR, 2.17) were also significantly more common in VLBWIs with congenital anomalies; however, their association with mental developmental delay was not statistically significant (AOR, 1.39). At 3 years of age, congenital anomalies remained significantly associated with weight and height z-scores below −1.28 (AOR, 1.99 and 1.64, respectively) and neurodevelopmental impairment (AOR, 1.86). At this time point, low head circumference z-scores and mental developmental delay were not significantly associated with congenital anomalies, despite elevated ORs (Table 5).

Table 5. Adjusted odds of growth and neurodevelopmental outcomes associated with congenital anomalies in very low birth weight infants.

Outcomes		Corrected age 18–24 mon	3 yr of age
Growth outcomes
	Weight (z-score < −1.28)	2.55 (1.67–3.91)	1.99 (1.27–3.11)
	Height (z-score < −1.28)	1.79 (1.17–2.74)	1.64 (1.05–2.57)
	Head circumference (z-score < −1.28)	2.15 (1.35–3.41)	1.64 (0.96–2.80)
	Growth restriction	2.12 (1.42–3.15)	2.22 (1.36–3.60)
Developmental outcomes
	Mental developmental delay	1.39 (0.84–2.29)	1.53 (0.85–2.75)
	Motor developmental delay	2.34 (1.44–3.82)	1.82 (1.00–3.31)
	Neurodevelopmental impairment	1.77 (1.12–2.81)	1.86 (1.07–3.25)

Values are presented as adjusted odds ratios with 95% confidence intervals.

Multivariate logistic regression analyses were adjusted for small for gestational age, Apgar score at 5 minutes, maternal chorioamnionitis, pregnancy-induced hypertension, moderate to severe bronchopulmonary dysplasia, and periventricular leukomalacia.

Impact of single vs. MCA on growth and neurodevelopment

The risks of growth restriction and neurodevelopmental impairment were compared between VLBWIs with isolated single or MCA. At 18–24 months CA, both groups had significantly higher risks for these outcomes compared with those of infants without congenital anomalies, with the highest risks observed in those with MCA (growth restriction: AOR, 3.91; 95% CI, 2.27–6.73; neurodevelopmental impairment: AOR, 4.27; 95% CI, 2.33–7.84). At 3 years of age, this pattern persisted, with VLBWIs with MCA continuing to have the highest risks of growth restriction (AOR, 2.82; 95% CI, 1.26–6.32) and neurodevelopmental impairment (AOR, 4.79; 95% CI, 2.31–9.95). Although VLBWIs with isolated single anomalies also had increased risk at both time points, the magnitude was lower than that in the MCA group (Table 6).

Table 6. Comparison of growth restriction and neurodevelopmental impairment between very low birth weight infants with single and multiple congenital anomalies.

Subgroups	Corrected age 18–24 mon		3 yr of age
	Growth restriction	Neurodevelopmental impairment	Growth restriction	Neurodevelopmental impairment
Isolated single anomaly	2.06 (1.36–3.11)	1.68 (1.05–2.70)	2.15 (1.31–3.54)	1.76 (1.01–3.15)
Multiple congenital anomalies	3.91 (2.27–6.73)	4.27 (2.33–7.84)	2.82 (1.26–6.32)	4.79 (2.31–9.95)

Values are presented as odds ratio with 95% confidence interval.

Multivariate logistic regression analyses, adjusted for small for gestational age, Apgar score at 5 minutes, maternal chorioamnionitis, maternal pregnancy-induced hypertension, moderate to severe bronchopulmonary dysplasia, and periventricular leukomalacia were performed to compare the risks of growth restriction and neurodevelopmental impairment between very low birth weight infants with single and multiple congenital anomalies.

DISCUSSION

This study confirmed that congenital anomalies significantly affect both the growth and neurodevelopment of VLBWIs. Infants with congenital anomalies had consistently low weight, height, and head circumference z-scores at 18–24 months CA and 3 years of age. The prevalence of growth restriction remained high, indicating that the effect of congenital anomalies extends beyond the neonatal period. Neurodevelopmental outcomes were also poor, with high rates of motor developmental delay and overall neurodevelopmental impairment persisting into early childhood. Infants with MCA had the highest risk, suggesting that the presence of multiple structural abnormalities lead to more severe complications.

Growth restriction is more common in infants with congenital anomalies and persists into early childhood. This finding aligns with studies showing that congenital anomalies contribute to poor postnatal growth, particularly in preterm infants, who already face challenges with weight gain and linear growth owing to immaturity and medical complications.6 Among the subgroups, infants with digestive system anomalies were particularly vulnerable, likely because of feeding difficulties, malabsorption, and the need for multiple surgical interventions.13 Gastrointestinal anomalies, such as esophageal and intestinal atresia, are associated with prolonged hospitalization and dependence on parenteral nutrition, further exacerbating growth deficits.18 Although neonatal care has improved the survival of VLBWIs, our findings suggest that current nutritional interventions may not fully address the effects of congenital anomalies. Given these challenges, individualized nutritional strategies, including early enteral feeding, close monitoring of growth trajectories, and specialized dietary interventions, may be necessary to improve long-term growth in this high-risk population.

Infants with congenital anomalies had significantly high rates of neurodevelopmental delays at both 18–24 months CA and 3 years of age, which is consistent with the results of prior studies.12 Several factors likely contributed to these outcomes. First, congenital anomalies often require prolonged hospitalization and medical intervention, reducing opportunities for early neurodevelopmental stimulation.19 Second, conditions, such as congenital heart defects and chromosomal abnormalities, are frequently associated with brain injury resulting from chronic hypoxia and metabolic dysregulation.20 Finally, congenital anomalies were associated with high rates of PVL and BPD, both of which are well-documented risk factors for adverse neurodevelopmental outcomes in preterm infants.19 Particularly, congenital heart defects have been linked to an increased risk of motor and cognitive impairment, possibly due to fluctuations in cerebral perfusion and disruptions in brain maturation.20 Chromosomal disorders often involve structural brain abnormalities and developmental delays that affect neurodevelopment. Given these risks, early intervention programs, including physical and speech therapy, as well as early developmental support, should be integrated into routine follow-up care. Evidence suggests that early rehabilitative support can significantly enhance cognitive, motor, and language skills, particularly in high-risk infants. Implementing a structured neurodevelopmental follow-up program, ideally within the first year of life, improves outcomes in similar high-risk neonatal populations.12,21,22,23 The decline in z-scores for weight, height, and head circumference at 3 years of age compared to 18–24 months CA, regardless of congenital anomaly status, may reflect the time-limited nature of postnatal catch-up growth. Multiple studies have shown that catch-up growth in preterm infants predominantly occurs within the first two years of life, with growth velocity slowing thereafter. For instance, Raaijmakers et al.24 reported that extremely low birth weight infants exhibited meaningful improvements in anthropometric z-scores up to 24 months of age, followed by a deceleration in growth and continued challenges in reaching target height by adolescence. Similarly, Lim et al.25 found that although initial improvements were observed, 35.2% of VLBWIs still exhibited growth failure at 36 months, highlighting the persistence of suboptimal growth outcomes beyond infancy. Additionally, the observed decline in z-scores may partly reflect differences in age-based assessment methods: in the KNN dataset, growth outcomes at 18–24 months are evaluated using CA, whereas 3-year data are based on chronological age, potentially amplifying the apparent decline in standardized growth parameters.

For multivariable analyses, adjustment variables were selected not only based on baseline differences between groups but also because they are well-established risk factors for poor growth and neurodevelopmental outcomes in preterm infants. SGA infants are more likely to experience both growth restriction and cognitive delays, as demonstrated by higher rates of underweight and stunting in early childhood, as well as significantly lower cognitive scores compared to infants appropriate for gestational age in meta-analyses.26,27 A low 5-minute Apgar score is associated with increased risks of neurological morbidities, as evidenced by higher rates of neurological hospitalizations among children with scores below 7 compared to those with higher scores.28 Maternal chorioamnionitis has been linked to an increased risk of neurodevelopmental disorders such as cerebral palsy, epilepsy, and intellectual disabilities, likely due to fetal inflammatory response and brain injury triggered by intrauterine inflammation.29,30 Similarly, PIH is associated with restricted fetal growth and elevated risks of neurodevelopmental impairment, particularly in extremely preterm or SGA infants, in whom the effects of poor placental function are most pronounced.31 BPD is one of the most consistently reported predictors of neurodevelopmental impairment and postnatal growth delay; more severe forms of BPD correlate with poorer outcomes, potentially due to disrupted metabolic regulation, insulin-like growth factor 1 dysfunction, and the need for prolonged respiratory support.32,33,34 Finally, PVL—particularly when coexisting with IVH—has been strongly linked to motor and cognitive impairments, as well as suboptimal physical growth, especially in infants with comorbid BPD.35,36 Even after adjusting for these clinically significant covariates, congenital anomalies remained independently associated with poor growth outcomes, particularly lower z-scores for weight and height, and with adverse neurodevelopmental outcomes, including motor delays and overall neurodevelopmental impairment at both 18–24 months CA and 3 years of age. These findings reinforce the specific contribution of congenital anomalies to long-term developmental risks, beyond what can be explained by other perinatal or neonatal complications.

Interestingly, after adjusting for various neonatal and perinatal risk factors, the association between congenital anomalies and mental developmental delay in VLBWIs was no longer statistically significant at either 18–24 months CA or 3 years of age. This finding contrasts with the persistent significant differences observed in motor developmental outcomes and overall neurodevelopmental impairments. This result suggests that the environment and social factors surrounding a child after discharge may have played a significant role. Factors such as socioeconomic status, parental education level, and the quality of caregiving can greatly influence cognitive and language development, potentially offsetting some of the early disadvantages associated with congenital anomalies. Previous studies have demonstrated that a supportive home environment and effective early intervention programs significantly promote cognitive resilience and language acquisition, even among high-risk infants with early brain injuries or structural anomalies.37,38 Furthermore, inherent neuroplasticity, the brain's ability to reorganize and adapt structurally and functionally in response to early injuries, might also explain the limited long-term cognitive impact of congenital anomalies. Studies using advanced neuroimaging techniques in preterm infants have shown significant adaptive neural reorganization following early brain injuries, which may preferentially support cognitive development over motor functions. The brain's reorganization tends to favor cognitive functions, as cognitive pathways are more adaptable and capable of recruiting alternative neural circuits to preserve function.39,40 Motor pathways, on the other hand, are more rigidly organized and therefore show less functional recovery, which may help explain the ongoing motor impairments.41 Additionally, the developmental tools used in this study mainly assess broad cognitive and language abilities, which may overlook subtle or specific deficits linked to particular congenital anomalies. Future studies should explore how psychosocial factors and brain plasticity contribute to cognitive outcomes in this high-risk group.

Infants with MCA have the most severe impairments in both growth and neurodevelopment. Compared with infants with isolated anomalies, infants with MCA had nearly twice the odds of growth restriction and neurodevelopmental impairment, suggesting that the presence of multiple structural abnormalities leads to more severe complications.42,43 MCA is a major determinant of poor outcomes, particularly in cases involving congenital heart defects or gastrointestinal anomalies, which impose significant metabolic and nutritional challenges.44,45 Frequent surgeries, prolonged hospitalization, and complex medical management in patients with MCA further delay recovery and developmental progression. These findings suggest that infants with MCA require specialized follow-up programs with an emphasis on early nutritional support, developmental therapy, and close medical supervision to address the unique needs of this high-risk group.

These findings indicate the need for a multidisciplinary approach in the care of VLBWIs with congenital anomalies. Given the high prevalence of growth restriction and neurodevelopmental impairment, early intervention, including individualized nutritional support and comprehensive neurodevelopmental assessments, should be a key focus.46 Advances in fetal medicine and neonatal surgery have improved survival rates; however, additional efforts are needed to enhance long-term functional outcomes and quality of life. Future studies should explore targeted therapeutic strategies to address the effects of congenital anomalies on postnatal growth and neurodevelopment. Long-term follow-up studies extending beyond early childhood are necessary to assess the effects of congenital anomalies on cognitive development, educational attainment, and overall wellbeing.47,48,49 These efforts are important for guiding evidence-based interventions for this vulnerable population.

This study has several limitations that should be considered when interpreting the findings. First, as a prospective analysis using registry data, differences in clinical practices across the participating NICUs may have influenced growth and neurodevelopmental outcomes. Variability in treatment approaches, nutritional support, and follow-up care could have contributed to inconsistencies in the reported results. Second, the classification of congenital anomalies followed the KNN system, which differs from internationally recognized frameworks, such as the International Classification of Diseases and European Surveillance of Congenital Anomalies. This discrepancy may have led to challenges in categorizing certain anomalies, particularly those affecting multiple organ systems, thereby limiting direct comparisons with other studies. Third, although a longitudinal design tracking the same cohort across both time points (18–24 months CA and 3 years of age) would have provided additional insights, the number of infants with complete follow-up at both stages was limited. A substantial number of infants had follow-up data at only one of the two time points, making it impractical to construct a consistent longitudinal cohort. Restricting the analysis to those with complete data would have significantly reduced the sample size and weakened statistical power, particularly for subgroup comparisons. To retain analytic strength and maximize available data, we treated each follow-up point as an independent cohort. While this design limits the ability to assess within-subject trajectories, it provides a broader and more generalizable perspective on growth and neurodevelopment at each stage. Fourth, developmental assessments were performed using different tools (BSID II, BSID III, and K-DST), which have variations in structure, scoring, and administration. Although BSID scores were prioritized when available, differences in tool characteristics could have influenced developmental outcome classification. Finally, missing follow-up data and small subgroup sample sizes may have led to underestimation of adverse outcomes and limited detailed comparisons by anomaly type. Future studies with standardized classification systems, larger cohorts, and longer follow-up are needed to better understand the long-term impact of congenital anomalies in VLBWIs.

This study provides important evidence of the long-term impact of congenital anomalies on the growth and neurodevelopment of VLBWIs. Infants with congenital anomalies consistently exhibited poorer growth outcomes and a higher prevalence of neurodevelopmental impairments at 18–24 months CA and 3 years of age. Patients with MCA face the greatest challenges, as the presence of multiple structural abnormalities increases the risk of adverse outcomes. Although advances in neonatal care have improved survival rates, the findings of this study emphasize the need for tailored interventions to address the unique challenges faced by these infants. Nutritional strategies should be optimized to support postnatal growth, and comprehensive neurodevelopmental follow-ups should be prioritized to ensure early identification and management of developmental delays. Given the persistent impact of congenital anomalies in early childhood, long-term monitoring and individualized support programs are crucial to improving functional outcomes and quality of life.

SUPPLEMENTARY MATERIALS

Supplementary Table 1

Classification of major congenital anomalies based on the Korean Neonatal Network system

Supplementary Table 2

Baseline characteristics of matched very low birth weight infants with and without congenital anomalies at 3 years of age