European Journal of Cardiovascular Medicine

Preterm birth is a major global health issue and is closely associated with increased risk of mortality and long-term neurodevelopmental impairment. As survival of very preterm and very low birth weight infants improves, attention has shifted from short-term survival to the early identification of brain injury and prediction of later developmental outcomes. Sequential cranial ultrasound examinations, often complemented by advanced imaging, have been shown to contribute substantially to early prognostication of neurodevelopmental outcome in preterm infants, underlining the central role of bedside neuroimaging in their follow-up [1].

Parenchymatous brain injury, particularly germinal matrix–intraventricular hemorrhage (IVH) and periventricular leukomalacia (PVL), represents the most common pattern of brain damage in premature infants and is strongly implicated in subsequent motor and cognitive disabilities [2]. These lesions typically occur in the vulnerable periventricular white matter and germinal matrix of the developing brain and can lead to a spectrum of outcomes ranging from subtle cognitive deficits to severe cerebral palsy. Given their frequency and prognostic significance, systematic detection and grading of such lesions in the neonatal period is essential in any comprehensive preterm follow-up program.

Neurosonography has evolved beyond a simple screening tool to an integral component of neuroprotection strategies in extremely preterm infants. Recent advances in technique and interpretation have positioned cranial ultrasound not only as a diagnostic modality but also as a means of risk stratification and early intervention planning in this high-risk group [3]. At the same time, its portability, safety and relatively low cost make it particularly attractive for use in resource-limited settings, where more advanced neuroimaging such as MRI may not be routinely available.

Evidence from systematic reviews and meta-analyses indicates that germinal matrix–intraventricular hemorrhage is strongly associated with adverse neurodevelopmental outcomes, with increasing severity of hemorrhage conferring progressively higher risks of impairment [4]. Similarly, studies in preterm children with PVL have demonstrated that the degree of white matter injury correlates closely with the severity of motor, cognitive and neuropsychological deficits, reinforcing the concept of a dose–response relationship between lesion burden and outcome [5]. Long-term follow-up cohorts of infants born extremely preterm further show that neonatal cranial ultrasound abnormalities are associated with a broad range of neurodevelopmental problems, including cerebral palsy, cognitive delay and behavioral difficulties extending into late childhood [6].

Despite this growing body of evidence, most data arise from high-income settings, and there remains a need for context-specific information from resource-constrained environments where cranial ultrasound is often the primary, and sometimes only, neuroimaging modality available in the neonatal period. Moreover, early outcome assessment around one year corrected age is particularly relevant for initiating timely early intervention. In this context, the present prospective cohort study aims to evaluate the correlation between cranial ultrasound findings and neurodevelopmental outcomes at one year corrected age in preterm neonates admitted to the neonatal intensive care unit of the Indian Institute of Medical Science and Research, Badnapur, Jalna.

Objectives           

Primary Objective

• To assess whether cranial ultrasound findings in preterm neonates are associated with neurodevelopmental status at one-year corrected age.

Secondary Objectives

1. To describe the frequency and pattern of cranial ultrasound abnormalities (including intraventricular hemorrhage, periventricular leukomalacia, ventriculomegaly and other lesions) among preterm neonates admitted to the neonatal intensive care unit.
2. To describe the neurodevelopmental profile of these preterm infants at one year corrected age using standardized developmental assessment and neurological examination.
3. To explore the relationship between the type and severity of specific cranial ultrasound abnormalities and the risk of adverse neurodevelopmental outcomes (such as developmental delay or cerebral palsy) at one-year corrected age.
4. To explore whether abnormal cranial ultrasound findings help to identify preterm infants who may benefit from closer follow-up and early intervention services.

Study design and setting

This was a hospital-based prospective cohort study conducted in the Neonatal Intensive Care Unit (NICU) and high-risk infant follow-up clinic of the Department of Paediatrics, Indian Institute of Medical Science and Research, Badnapur, Jalna, Maharashtra, India. Preterm neonates were enrolled over a period of six months, and each infant was followed until one-year corrected age.

Study population               

All consecutive preterm neonates admitted to the NICU during the study period were screened for eligibility. Eligible infants were enrolled after obtaining written informed consent from a parent or legal guardian.

Inclusion criteria

Preterm neonates were included if they met all of the following criteria:

1. Gestational age <37 completed weeks (as per best obstetric estimate and/or early ultrasound when available).

2. Admission to the NICU within 72 hours of birth.

3. Underwent at least one cranial ultrasound examination during the NICU stay according to the unit protocol.

4. Clinically stable at discharge and planned for follow-up at the high-risk clinic.

5. Parents/guardians provided written informed consent and agreed to return for neurodevelopmental assessment at one year corrected age.

Exclusion criteria

Neonates were excluded if any of the following were present:

1. Major congenital malformations or chromosomal syndromes involving the central nervous system.

2. Suspected or confirmed congenital infections known to affect the brain (e.g. TORCH, CMV) at the time of enrollment.

3. Inborn errors of metabolism diagnosed during the neonatal period.

4. Severe congenital sensory deficits (e.g. anophthalmia, major external ear malformations) identified at birth.

5. Infants who died before completion of initial cranial ultrasound or before hospital discharge.

6. Infants who were lost to follow-up or did not attend the scheduled neurodevelopmental assessment at one year corrected age.

Enrollment and baseline data collection

Eligible preterm neonates were identified soon after NICU admission. After consent, baseline information was recorded using a structured proforma, including maternal details (age, obstetric history, antenatal complications), perinatal factors (mode of delivery, resuscitation at birth, Apgar scores), and neonatal characteristics (gestational age, birth weight, sex, multiple pregnancy, need for respiratory support, sepsis, and other major morbidities).

Cranial ultrasound protocol

Cranial ultrasound was performed using a standard ultrasound machine with a high-frequency (typically 5–8 MHz) sector or convex probe. Examinations were carried out through the anterior fontanelle by a radiologist or neonatologist trained in neonatal neurosonography.

The unit protocol included:

• An initial cranial ultrasound between days 3 and 7 of life (or at admission if later).

• A follow-up scan between days 10 and 14 of life.

• A further scan near term-equivalent age or prior to discharge, whichever occurred earlier.

Additional scans were performed if clinically indicated (e.g., sudden deterioration, seizures, rapidly increasing head circumference).

Standard coronal and sagittal views were obtained. Findings were documented and categorized as:

• Normal

• Germinal matrix–intraventricular hemorrhage (IVH) Grade I–II

• IVH Grade III–IV

• Periventricular leukomalacia (PVL)

• Ventriculomegaly/post-hemorrhagic ventricular dilatation

• Other lesions (e.g. congenital malformations, parenchymal cysts)

IVH was graded according to standard classification (Papile), and PVL was diagnosed based on increased periventricular echogenicity with or without cyst formation. For analysis, cranial ultrasound was also dichotomized as normal versus abnormal (any of the lesion categories).

Neurodevelopmental assessment at one year

Neurodevelopmental evaluation was scheduled at one-year corrected age (12 ± 2 weeks) in the high-risk infant follow-up clinic. Assessments were performed by a trained pediatricians and/or clinical psychologist who were not involved in the neonatal cranial ultrasound reporting.

The evaluation included:

1. Standardized developmental assessment using the Bayley Scales of Infant and Toddler Development, Third Edition (Bayley-III), with particular emphasis on the Cognitive Composite score.

2. Neurological examination to assess muscle tone, posture, reflexes and motor milestones, and to diagnose cerebral palsy where present.

3. Screening for visual and hearing impairment based on clinical examination and available audiological/ophthalmological reports.

For the purpose of this study, adverse neurodevelopmental outcome at one-year corrected age was defined as the presence of any one of the following:

• Bayley-III Cognitive Composite score <85, and/or

• Diagnosis of cerebral palsy, and/or

• Significant visual or hearing impairment requiring intervention, and/or

• Clinical diagnosis of global developmental delay.

Follow-up and outcome ascertainment

Parents were counselled regarding the importance of follow-up at discharge and provided with appointment dates. Reminder phone calls were made where feasible. Infants who did not attend the scheduled visit were contacted and offered a rescheduled appointment when possible. Only children who underwent neurodevelopmental assessment at the specified time window were included in the final outcome analysis.

Data management and statistical analysis

Data were recorded on predesigned forms and subsequently entered into a password-protected electronic database. Continuous variables (e.g. birth weight, gestational age, Bayley scores) will be summarized as mean ± standard deviation or median (interquartile range), depending on distribution. Categorical variables (e.g. presence/absence and type of cranial ultrasound abnormality, adverse outcome) will be presented as frequencies and percentages.

The association between cranial ultrasound findings (normal vs abnormal, and specific lesion categories) and adverse neurodevelopmental outcome at one year corrected age will be examined using appropriate comparative tests (such as chi-square test or Fisher’s exact test for categorical variables and t-test or ANOVA for continuous variables). Simple measures of diagnostic performance (such as sensitivity, specificity, and predictive values) may be calculated for abnormal cranial ultrasound in relation to adverse outcome. No complex multivariable modelling is planned, in keeping with the clinical and exploratory nature of this single-centre cohort.

Ethical considerations

The study protocol was reviewed and approved by the Institutional Ethics Committee of the Indian Institute of Medical Science and Research, Badnapur, Jalna. Written informed consent was obtained from the parents or legal guardians of all participating infants. Confidentiality of patient information was maintained throughout the study, and all procedures were conducted in accordance with institutional and national ethical guidelines for human research.

1. Study population

During the 6-month study period, a total of 152 preterm neonates (<37 weeks) were admitted to the NICU. Of these, 6 infants had major congenital anomalies or suspected congenital infections involving the central nervous system, 4 died before completion of the initial cranial ultrasound examination, and parents of 2 infants declined consent. The remaining 140 preterm neonates fulfilled the eligibility criteria and were enrolled in the study; all survived to discharge and attended the scheduled neurodevelopmental assessment at one-year corrected age (Figure 1).

The mean gestational age of the cohort was 30.2 ± 2.8 weeks, and the mean birth weight was 1,320 ± 380 g. Extremely and very preterm infants formed the majority of the population: 25 (17.9%) were born at 24–27 weeks, 50 (35.7%) at 28–30 weeks, 45 (32.1%) at 31–33 weeks and 20 (14.3%) at 34–36 weeks of gestation. There were 76 (54.3%) male and 64 (45.7%) female infants. Cesarean delivery accounted for just over half of births (72/140, 51.4%), and multiple gestation was present in 32 (22.9%) infants. Common neonatal morbidities included respiratory distress syndrome in 81 (57.9%) infants, culture-proven sepsis in 38 (27.1%), bronchopulmonary dysplasia in 15 (10.7%) and necrotizing enterocolitis in 9 (6.4%). A total of 62 infants (44.3%) required mechanical ventilation during their NICU stay. Baseline characteristics of the study population are summarized in Table 1.

Table 1. Baseline characteristics of the study population (n = 140)

2. Cranial ultrasound findings

Cranial ultrasound was performed in all 140 enrolled preterm infants according to the unit protocol. Overall, 78 infants (55.7%) had a normal cranial ultrasound, while 62 infants (44.3%) demonstrated at least one abnormality. Among the abnormal scans, the most frequent lesions were periventricular leukomalacia (PVL) in 16 infants (11.4% of the cohort) and Grade I–II intraventricular hemorrhage (IVH) in 15 infants (10.7%). Ventriculomegaly/post-hemorrhagic ventricular dilatation was seen in 13 infants (9.3%), other lesions (including congenital malformations and parenchymal cysts) in 10 infants (7.1%), and Grade III–IV IVH in 8 infants (5.7%) (Table 2).

Severe lesions (Grade III–IV IVH, PVL and ventriculomegaly) tended to occur more frequently in the lowest gestational age groups, particularly among infants born before 30 weeks, whereas normal cranial ultrasound findings were more common in those born at later gestational ages.

Flow Diagram of the study Participants:

Table 2. Pattern of cranial ultrasound findings in the study cohort (n = 140)

IVH = intraventricular hemorrhage; PVL = periventricular leukomalacia.
¹Includes congenital malformations, parenchymal cysts and other less frequent abnormalities.

Figure 2. Distribution of cranial ultrasound categories among preterm infants in the cohort. Bar chart showing the proportion of infants with normal cranial ultrasound, Grade I–II IVH, Grade III–IV IVH, PVL, ventriculomegaly and other lesions (expressed as percentage of the total sample, n = 140).

3. Neurodevelopmental outcomes at one year

At one year corrected age, neurodevelopmental assessment was completed for all 140 infants. The mean Bayley-III Cognitive Composite score for the cohort was 89.1 ± 13.5. A total of 53 infants (37.9%) had a Cognitive score <85.

Overall, 64 infants (45.7%) met criteria for the composite adverse neurodevelopmental outcome (Bayley-III Cognitive score <85 and/or cerebral palsy and/or significant visual/hearing impairment and/or global developmental delay). Among these, 18 infants (12.9% of the total cohort) were diagnosed with cerebral palsy, and 7 infants (5.0%) had significant visual and/or hearing impairment requiring intervention. Many infants with cerebral palsy or sensory impairment also had low cognitive scores, so these categories were not mutually exclusive. The distribution of neurodevelopmental outcomes is summarized in Table 3.

Table 3. Neurodevelopmental outcomes at one-year corrected age (n = 140)

¹Composite adverse outcome defined as Bayley-III Cognitive Composite score <85 and/or cerebral palsy and/or significant visual or hearing impairment and/or global developmental delay at one-year corrected age (categories not mutually exclusive).

4. Association between cranial ultrasound findings and neurodevelopmental outcome

Infants with abnormal cranial ultrasound had a markedly higher burden of adverse neurodevelopmental outcome at one year corrected age compared with those with normal scans. Among the 78 infants with a normal cranial ultrasound, 18 (23.1%) had an adverse outcome, whereas 46 of 62 infants (74.2%) with an abnormal scan met criteria for adverse outcome. This difference was statistically significant (χ² = 36.4, df = 1, p < 0.001), corresponding to an odds ratio of approximately 9.6 for adverse outcome in infants with abnormal versus normal cranial ultrasound (Table 4). When abnormal cranial ultrasound was considered as a simple screening test for adverse outcome, the sensitivity was 71.9%, specificity 78.9%, positive predictive value 74.2% and negative predictive value 76.9%.

Mean Bayley-III Cognitive scores also differed substantially between the two groups. Infants with normal cranial ultrasound had higher cognitive scores (98.5 ± 8.1) than those with abnormal findings (79.1 ± 12.1), and this difference was highly significant on independent samples t-test (t = 10.84, df ≈ 102, p < 0.001). The distribution of scores is illustrated in Figure 3, which demonstrates a clear downward shift and greater spread of cognitive performance in the abnormal cranial ultrasound group.

Table 4. Comparison of neurodevelopmental outcomes at one year between infants with normal and abnormal cranial ultrasound

*cUS = cranial ultrasound.
*Independent samples t-test for continuous variable; chi-square test for categorical variable.
†Adverse outcome defined as Bayley-III Cognitive score <85 and/or cerebral palsy and/or significant visual or hearing impairment and/or global developmental delay at one-year corrected age.

Figure 3. Box-and-whisker plot showing the distribution of Bayley-III Cognitive Composite scores at one year corrected age in infants with normal versus abnormal cranial ultrasound. The plot demonstrates higher median scores and a narrower interquartile range in the normal cUS group, with a pronounced downward shift and wider dispersion of scores in the abnormal cUS group.

5. Neurodevelopmental outcomes by specific cranial ultrasound lesion

When infants were classified according to specific cranial ultrasound lesion type, a clear gradient in neurodevelopmental risk was observed. The lowest proportion of adverse outcome was seen in infants with normal scans (18/78, 23.1%), followed by those with Grade I–II IVH (7/15, 46.7%) and “other” lesions (6/10, 60.0%). In contrast, almost all infants with PVL (15/16, 93.8%) and all those with Grade III–IV IVH (8/8, 100%) experienced adverse neurodevelopmental outcomes at one year. Infants with ventriculomegaly also had a high rate of adverse outcome (10/13, 76.9%). The overall difference in adverse outcome across the six cranial ultrasound categories was statistically significant (global χ² = 46.4, df = 5, p < 0.001) (Table 5).

A similar pattern was reflected in Bayley-III Cognitive Composite scores. Mean scores were highest in infants with normal cranial ultrasound (98.5 ± 8.1) and progressively lower in those with Grade I–II IVH (92.1 ± 8.8), ventriculomegaly (79.3 ± 8.5), other lesions (81.6 ± 8.4), Grade III–IV IVH (73.6 ± 7.4) and PVL (68.1 ± 8.7). One-way ANOVA demonstrated a highly significant difference in cognitive scores across lesion categories (F(5, 134) = 51.9, p < 0.001), consistent with a dose–response relationship between the severity of ultrasound-detected brain injury and the degree of neurodevelopmental impairment.

Table 5. Neurodevelopmental outcomes at one year corrected age according to specific cranial ultrasound lesion

IVH = intraventricular hemorrhage; PVL = periventricular leukomalacia.
¹Adverse outcome defined as Bayley-III Cognitive Composite score <85 and/or cerebral palsy and/or significant visual or hearing impairment and/or global developmental delay at one year corrected age.

In this prospective cohort of preterm infants from a tertiary centre in rural Maharashtra, nearly half of the infants (45.7%) had an adverse neurodevelopmental outcome at one year corrected age, and abnormal cranial ultrasound was associated with almost a ten-fold increase in the odds of adverse outcome compared with normal scans. Campbell et al. reported similar patterns in a large cohort of infants born extremely preterm, showing that moderate to severe neonatal cranial ultrasound abnormalities were strongly associated with adverse neurodevelopmental outcomes at 10 years of age, including cerebral palsy, lower IQ and behavioural problems, with major impairment in roughly 30–40% of those with significant lesions versus about 10–15% with normal scans [6]. Our early follow-up data at one year therefore appear numerically consistent with the longer-term burden described in that study, albeit in a different setting and with a broader gestational age range. From a broader perspective, our findings fit within the range of neurodevelopmental outcomes described in narrative and systematic reviews. Allen summarised that approximately 15–25% of very preterm infants develop major neurosensory disabilities, while an additional 30–40% exhibit milder cognitive, learning or behavioural problems at school age [7]. In our cohort, 12.9% had cerebral palsy and 5.0% had significant visual or hearing impairment, while 37.9% had Bayley-III Cognitive scores below 85 and 45.7% met criteria for a composite adverse outcome. These figures fall toward the higher side of the ranges summarised by Allen, which is not unexpected given our single-centre population, higher proportion of very preterm infants and the resource-limited context in which the study was conducted [7]. The strength of the association between cranial ultrasound lesions and developmental outcome in our data is in line with earlier work comparing ultrasound with basic clinical predictors. Vollmer et al. found that both gestational age and neonatal cranial ultrasound abnormalities were independent predictors of long-term outcome in very preterm infants, with moderate to severe ultrasound abnormalities associated with a three- to five-fold increase in odds of adverse outcome compared with normal scans [8]. In our study, the adverse outcome rate increased from 23.1% among infants with normal ultrasound to 74.2% among those with any abnormality, corresponding to an odds ratio of roughly 9.6. Although our effect size is somewhat larger than that reported by Vollmer et al., the direction and magnitude are broadly compatible, particularly when considering differences in gestational age range, follow-up timing and outcome definitions [8]. The gradient of risk we observed across IVH severity categories is also consistent with the literature. Roland and Hill reviewed the management and outcomes of germinal matrix–intraventricular haemorrhage in premature newborns and highlighted that severe (Grade III–IV) IVH is frequently associated with high rates of cerebral palsy (often 40–80%) and cognitive impairment, whereas lower grade haemorrhages carry a more modest but still important risk of subsequent deficits [9]. In our cohort, 46.7% of infants with Grade I–II IVH and 100% of those with Grade III–IV IVH experienced adverse neurodevelopmental outcomes by one year, and mean Bayley-III Cognitive scores were approximately 20–30 points lower in the high-grade group than in those with normal scans. Although our follow-up period is shorter than in many of the series summarised by Roland and Hill, these numbers align with the concept of a dose–response relationship between IVH severity and adverse outcome [9]. Periventricular leukomalacia and broader leukomalacia patterns were also strongly associated with poor outcomes in our study. Song et al. examined the long-term outcomes of leukomalacia in both preterm and term infants and reported that leukomalacia was associated with cerebral palsy rates in the order of 50–70% and cognitive impairment in 40–60%, with cystic and more extensive lesions portending particularly poor outcomes [10]. In our cohort, 93.8% of infants with PVL met criteria for adverse outcome at one year, and their mean cognitive scores were the lowest among all lesion categories. While our follow-up is at an earlier age than in the study by Song et al., the severity of impairment associated with PVL appears directionally similar and may well translate into substantial long-term burden if not mitigated by timely intervention [10]. The relative performance of ultrasound compared with MRI is also relevant to interpreting our findings. Burkitt et al. compared cranial ultrasound and MRI in extremely preterm infants and found that MRI detected additional subtle white matter abnormalities that were not always apparent on ultrasound, yet moderate to severe lesions visualised on ultrasound still showed a strong correlation with neurological outcomes at one and three years [11]. In that cohort, the presence of significant white matter injury on either modality was associated with substantially higher rates of cerebral palsy and developmental delay compared with infants with normal imaging [11]. Our results, derived purely from ultrasound without routine MRI, therefore support the view that, especially in resource-limited settings, a carefully performed and interpreted neurosonogram can still provide clinically meaningful prognostic information. Ventriculomegaly in our cohort carried an intermediate but clinically important risk, with 76.9% of affected infants showing adverse outcomes at one year. Pappas et al. evaluated extremely premature neonates with ventriculomegaly in the absence of overt periventricular–intraventricular haemorrhage and reported that these infants had significantly higher rates of neurodevelopmental and behavioural impairment at 18–22 months than those without ventriculomegaly, with composite impairment rates approaching 30–40% versus about 20% in controls [12]. Our higher adverse outcome proportion likely reflects a broader definition of ventriculomegaly, differences in lesion severity and the inclusion of additional risk factors typical of a rural Indian NICU population, but the direction of effect is similar—ventriculomegaly appears to be a meaningful risk marker even in the absence of extensive haemorrhage [12]. Interpretation of our Bayley-III data requires caution. We observed a mean Cognitive Composite score of 89.1 ± 13.5, with 37.9% of infants scoring below 85. Spencer-Smith et al. showed that Bayley-III tends to yield somewhat higher scores than Bayley-II and that a score just below 85 at toddler age may underestimate later difficulties; children with Bayley-III Cognitive scores in the low 80s had increased risk of school-age cognitive and language deficits compared with those scoring 90 or above [13]. Against that background, the nearly 40% of our cohort with scores below 85 probably represents a clinically important at-risk group rather than a mild deviation of limited significance [13]. Regional data from India offer some useful points of comparison. Nagaraj et al. studied high-risk neonates in a NICU setting and reported neurosonogram abnormalities in roughly one third of infants, with short-term adverse neurological outcomes in about 20–30% depending on lesion type and severity [14]. In contrast, we found ultrasound abnormalities in 44.3% and a composite adverse outcome rate of 45.7%. Differences in case mix (a higher proportion of very preterm infants in our cohort), follow-up timing (one year corrected age versus earlier neonatal outcomes) and outcome definition likely explain much of this discrepancy, but they also highlight that our results may represent a more severely affected subset of preterm infants within the broader Indian context [14]. Agut et al. proposed a refined ultrasound-based classification of preterm white matter injury and showed that higher grades of injury were associated with major motor or cognitive impairment in 40–70% of infants, while lower grades had more variable but still elevated risk [15]. The stepwise decline in cognitive scores and rising adverse outcome rates across our normal, mild IVH, severe IVH, PVL and ventriculomegaly groups parallels the graded associations described in that classification [15]. Our findings also have implications for follow-up and intervention. While we did not formally test an intervention, the high proportion of infants with early developmental risk underscores the importance of linking preterm infants with abnormal ultrasound findings to structured early intervention services. Leib et al. demonstrated that early intervention and stimulation programmes for preterm infants could reduce the proportion of children with cognitive delay at school age by approximately 15–20 percentage points compared with standard care [16]. Given that nearly half of our cohort met criteria for adverse outcome at one year, the potential absolute benefit of similar interventions in our setting could be substantial, particularly if targeted toward infants with high-risk ultrasound patterns such as PVL, high-grade IVH or ventriculomegaly [16]. At the same time, several limitations must be acknowledged. This was a single-centre study with a modest sample size, and the reliance on ultrasound alone, without systematic MRI, means that more subtle diffuse or cortical injuries may have been missed, as suggested by Burkitt et al. [11] and Agut et al. [15]. Our follow-up was limited to one year corrected age, whereas many cognitive, behavioural and learning problems only become fully evident at school age, as shown in the long-term cohorts described by Campbell et al. and Spencer-Smith et al. [6,13]. Differences in perinatal care, infection burden and access to rehabilitation services between our setting and those in high-income countries described by Allen, Pappas and others also limit direct numerical comparisons [7,12]. Nonetheless, within these constraints, our results are broadly consistent in both direction and approximate magnitude with existing literature and add context-specific data on the prognostic value of cranial ultrasound in preterm infants cared for in a rural Indian NICU [6–16].

In this single-centre prospective cohort of preterm infants, nearly half had an adverse neurodevelopmental outcome at one year, and abnormal cranial ultrasound findings were strongly associated with both lower Bayley-III Cognitive scores and higher rates of cerebral palsy and sensory impairment. Severe lesions—particularly Grade III–IV intraventricular haemorrhage, periventricular leukomalacia and ventriculomegaly—were almost uniformly linked with poor outcome, while a normal scan was associated with largely favourable development but did not completely exclude impairment.

These findings confirm that routine cranial ultrasound, when systematically performed and interpreted, is a valuable and pragmatic tool for early risk stratification in preterm infants, especially in resource-limited settings where access to MRI is restricted. Integrating ultrasound-based risk profiles with structured neurodevelopmental follow-up and early intervention programmes should be a priority to reduce the long-term burden of disability among preterm survivors in similar contexts.

1.       Brouwer, M. J., van Kooij, B. J., van Haastert, I. C., Koopman-Esseboom, C., Groenendaal, F., de Vries, L. S., & Benders, M. J. (2014). Sequential cranial ultrasound and cerebellar diffusion weighted imaging contribute to the early prognosis of neurodevelopmental outcome in preterm infants. PloS one, 9(10), e109556.

2.       Barria, M., & Flández, A. (2012). Parenchymatous brain injury in premature infants: intraventricular hemorrhage and periventricular leukomalacia. Brain Injury in Preterm Infants. Neonatal Care, 87-102.

3.       Tang, L., Li, Q., Xiao, F., Gao, Y., Zhang, P., Cheng, G., ... & Zhou, W. (2024). Neurosonography: shaping the future of neuroprotection strategies in extremely preterm infants. Heliyon, 10(11).

4.       Zhou, M., Wang, S., Zhang, T., Duan, S., & Wang, H. (2024). Neurodevelopmental outcomes in preterm or low birth weight infants with germinal matrix-intraventricular hemorrhage: a meta-analysis. Pediatric Research, 95(3), 625-633.

5.       Choi, J. Y., Rha, D. W., & Park, E. S. (2016). The effects of the severity of periventricular leukomalacia on the neuropsychological outcomes of preterm children. Journal of child neurology, 31(5), 603-612.

6.       Campbell, H., Check, J., Kuban, K. C., Leviton, A., Joseph, R. M., Frazier, J. A., ... & O’Shea, T. M. (2021). Neonatal cranial ultrasound findings among infants born extremely preterm: associations with neurodevelopmental outcomes at 10 years of age. The Journal of pediatrics, 237, 197-205.

7.       Allen, M. C. (2008). Neurodevelopmental outcomes of preterm infants. Current opinion in neurology, 21(2), 123-128.

8.       Vollmer, B., Roth, S., Baudin, J., Stewart, A. L., Neville, B. G., & Wyatt, J. S. (2003). Predictors of long-term outcome in very preterm infants: gestational age versus neonatal cranial ultrasound. Pediatrics, 112(5), 1108-1114.

9.       Roland, E. H., & Hill, A. (2003). Germinal matrix–intraventricular hemorrhage in the premature newborn: management and outcome. Neurologic clinics, 21(4), 833-851.

10.    Song, J., Yue, Y., Sun, H., Cheng, P., Xu, F., Li, B., ... & Zhu, C. (2023). Clinical characteristics and long-term neurodevelopmental outcomes of leukomalacia in preterm infants and term infants: a cohort study. Journal of Neurodevelopmental Disorders, 15(1), 24.

11.    Burkitt, K., Kang, O., Jyoti, R., Mohamed, A. L., & Chaudhari, T. (2019). Comparison of cranial ultrasound and MRI for detecting BRAIN injury in extremely preterm infants and correlation with neurological outcomes at 1 and 3 years. European Journal of Pediatrics, 178(7), 1053-1061.

12.    Pappas, A., Adams-Chapman, I., Shankaran, S., McDonald, S. A., Stoll, B. J., Laptook, A. R., ... & Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network. (2018). Neurodevelopmental and behavioral outcomes in extremely premature neonates with ventriculomegaly in the absence of periventricular-intraventricular hemorrhage. JAMA pediatrics, 172(1), 32-42.

13.    Spencer-Smith, M. M., Spittle, A. J., Lee, K. J., Doyle, L. W., & Anderson, P. J. (2015). Bayley-III cognitive and language scales in preterm children. Pediatrics, 135(5), e1258-e1265.

14.    Nagaraj, N., Berwal, P. K., Srinivas, A., Sehra, R., Swami, S., Jeevaji, P., ... & Berwal, A. (2016). A study of neurosonogram abnormalities, clinical correlation with neurosonogram findings, and immediate outcome of high-risk neonates in Neonatal Intensive Care Unit. Journal of pediatric neurosciences, 11(3), 200-205.

15.    Agut, T., Alarcon, A., Cabañas, F., Bartocci, M., Martinez-Biarge, M., & Horsch, S. (2020). Preterm white matter injury: ultrasound diagnosis and classification. Pediatric Research, 87(Suppl 1), 37-49.

16.  Leib, S. A., Benfield, D. G., & Guidubaldi, J. (1980). Effects of early intervention and stimulation on the preterm infant. Pediatrics, 66(1), 83-90.