Seizures in the pediatric population constitute a substantial clinical concern, not merely due to their high incidence but also owing to the potential implications for neurodevelopmental outcomes and long-term morbidity. Globally, epilepsy affects approximately 1% of the population, with the highest incidence observed in early childhood, particularly during the neonatal and infantile periods [1]. Seizures may be symptomatic of an underlying structural, metabolic, infectious, or genetic etiology, and their timely elucidation is imperative to initiate appropriate therapeutic interventions and minimize neurocognitive sequelae [2]. In this context, neuroimaging, particularly magnetic resonance imaging (MRI), has emerged as a pivotal diagnostic modality in delineating the etiology of seizures, especially when the clinical and electroencephalographic findings are inconclusive [3].

MRI offers unparalleled soft tissue contrast and multiplanar imaging capability, making it the modality of choice for structural evaluation of the pediatric brain [4]. Unlike computed tomography (CT), which is limited by ionizing radiation and lower sensitivity to subtle parenchymal abnormalities, MRI allows for high-resolution characterization of developmental anomalies, cortical dysplasias, neoplastic lesions, vascular malformations, post-infectious sequelae, and hypoxic-ischemic injuries [5,6]. Moreover, the advent of advanced MRI techniques—including diffusion-weighted imaging (DWI), fluid-attenuated inversion recovery (FLAIR), susceptibility-weighted imaging (SWI), and MR spectroscopy—has significantly enhanced the diagnostic yield in epilepsy imaging protocols [7].

The diagnostic approach to pediatric seizures necessitates a nuanced understanding of age-specific etiologies and the dynamic nature of the developing brain. Neonatal seizures, for instance, are frequently associated with hypoxic-ischemic encephalopathy, intracranial hemorrhage, or metabolic derangements, while older children are more likely to present with idiopathic epilepsies, cortical malformations, or acquired insults such as trauma or central nervous system infections [8]. MRI serves as a critical adjunct in distinguishing between benign self-limited epileptic syndromes and those requiring aggressive intervention or surgical consideration [9].

In resource-constrained environments, where access to advanced imaging may be limited, the role of MRI in guiding clinical decision-making becomes even more pronounced. Early identification of structural lesions not only facilitates etiological classification but also prognostication and therapeutic stratification [10]. Furthermore, the integration of MRI findings with clinical and electroencephalographic data forms the cornerstone of presurgical evaluation in pharmacoresistant epilepsy cases [11].

Given these considerations, this study endeavors to systematically evaluate the spectrum of MRI findings in pediatric patients presenting with seizures at a tertiary care teaching hospital. By correlating clinical profiles with imaging patterns, the study aims to underscore the indispensable role of MRI in the diagnostic algorithm of childhood epilepsy and to advocate for its routine inclusion in the evaluation of pediatric seizures, especially in tertiary healthcare settings.

Study Design and Setting

This study was designed as a prospective observational analysis conducted over a period of 12 months at the Department of Radiodiagnosis, in collaboration with the Department of Pediatrics, at a tertiary care teaching hospital in South India..

Study Population and Eligibility Criteria

The study population comprised pediatric patients aged between 1 month and 18 years who presented with clinical manifestations of seizures and were referred for neuroimaging evaluation. Inclusion criteria encompassed patients with newly diagnosed unprovoked seizures, known cases of epilepsy undergoing MRI for etiological clarification or pre-surgical evaluation, and those with seizure recurrence despite medical therapy. Patients were excluded if they exhibited contraindications to MRI (e.g., pacemakers, metallic implants), were clinically unstable to undergo scanning, or had incomplete clinical records.

Clinical Evaluation and Data Collection

A detailed clinical assessment was undertaken for each subject, including demographic profile, perinatal history, developmental milestones, family history of epilepsy, seizure semiology, frequency, duration, and presence of associated neurological deficits. Laboratory investigations, such as serum electrolytes, glucose levels, and metabolic panels, were reviewed to exclude acute symptomatic seizures secondary to correctable metabolic abnormalities. Electroencephalography (EEG) findings, where available, were also documented to correlate with imaging results.

MRI Protocol and Imaging Technique

All MRI examinations were performed using a 1.5 Tesla or 3 Tesla MRI scanner (Model: GE Signa/Philips Achieva, depending on availability) under standard epilepsy imaging protocol, following pediatric sedation guidelines where necessary. The imaging sequences included axial and coronal T1-weighted, T2-weighted, and FLAIR sequences; diffusion-weighted imaging (DWI); apparent diffusion coefficient (ADC) maps; susceptibility-weighted imaging (SWI); and contrast-enhanced sequences where clinically indicated.

Additional targeted sequences such as high-resolution coronal T2-weighted oblique images perpendicular to the long axis of the hippocampus, 3D volumetric sequences, and MR spectroscopy were employed in cases with temporal lobe epilepsy, suspected mesial temporal sclerosis, or metabolic etiologies. Imaging parameters were optimized for pediatric neuroanatomy to maximize lesion detection sensitivity.

Image Analysis and Interpretation

MRI studies were independently reviewed by two senior radiologists with at least 10 years of subspecialty experience in pediatric neuroradiology. In instances of interpretative discordance, consensus was achieved through collaborative review. Lesions were categorized into major etiological groups, including developmental anomalies, hypoxic-ischemic injuries, vascular insults, infections, neoplasms, and miscellaneous findings. Structural abnormalities were further subclassified based on localization (cortical vs. subcortical), symmetry, and chronicity. The correlation between MRI findings and clinical seizure characteristics was systematically evaluated.

Data Management and Statistical Analysis

All data was recorded using a pre-validated case record form. Statistical analysis was conducted using IBM SPSS Statistics for Windows, Version 25.0 (IBM Corp., Armonk, NY, USA). Descriptive statistics were applied to quantify demographic variables, seizure types, and imaging findings. Categorical variables were expressed as frequencies and percentages, while continuous variables were summarized using mean ± standard deviation. Chi-square test or Fisher’s exact test was employed to examine associations between MRI findings and clinical variables such as seizure type and duration. A p-value of <0.05 was considered statistically significant.

Outcome Measures

The primary outcome was the proportion of patients in whom MRI yielded clinically significant findings that altered diagnosis or management. Secondary outcomes included the spectrum and frequency of MRI-detectable pathologies in various pediatric age strata and the degree of concordance between imaging and EEG findings, where applicable.

Table 1: Age Group by Seizure Type

Table 2: Gender-wise MRI Abnormality Pattern

Table 3: Seizure Duration vs. MRI Yield

Table 4: Clinical Features and MRI Concordance

Table 5: Seizure Type vs. MRI Yield

The analysis of age-wise seizure type distribution revealed that GTCS was the most frequent subtype across all pediatric age groups, peaking at 53.8% in children aged 11–15 years. Focal seizures demonstrated a progressive increase with age, reaching 30% in the 16–18-year bracket. In contrast, infantile spasms were exclusively confined to children below one year, underscoring their link to early developmental anomalies and epileptic encephalopathies. The observed trends align with the age-dependent maturation of the central nervous system and evolving seizure phenotypes in pediatric populations.

Gender-wise MRI findings showed structural abnormalities to be the most common pathology in both males and females (31.4% and 28%, respectively), followed by hypoxic and infective etiologies. Notably, the prevalence of normal MRIs remained similar across sexes (~30%), reflecting the proportion of idiopathic epilepsy syndromes or cases with subtle or non-visible lesions on conventional imaging. Children with seizures lasting over 15 minutes demonstrated an 81.8% yield, compared to 60% in those with seizures under one minute. This reinforces the hypothesis that prolonged seizures are more likely to be symptomatic of underlying structural, metabolic, or acquired pathology.

Developmental delay (88.2%) and neurocutaneous markers (75%) had high concordance with structural imaging abnormalities, validating their role as red flags for neuroimaging. Febrile seizure history, however, showed the lowest concordance (40%), consistent with its frequent idiopathic nature. Infantile spasms yielded abnormalities in 100% of cases, followed closely by focal seizures (86.7%). GTCS, absence, and myoclonic seizures had comparatively lower yields, aligning with their association with generalized or genetic epilepsies, which may not always show radiological correlates.

The frontal lobe (29%) was the most commonly affected site, followed by the temporal lobe (24%), emphasizing these regions as critical epileptogenic zones in pediatric patients. Subcortical structures (14%) and multilobar involvement (12%) also featured prominently, indicating diffuse or multifocal pathology in some cases. Structural abnormalities and normal MRIs each comprised 30% of cases, reflecting both the high diagnostic value of MRI and its limitations in certain idiopathic or generalized epilepsies. Hypoxic (16.7%) and infective (13.3%) etiologies were significant contributors, while neoplastic (6.7%) and vascular (3.3%) causes were less common.

This prospective observational study evaluated the clinical utility of MRI in children with seizures, emphasizing the imaging yield, etiological patterns, and concordance with clinical variables. Our results underscore the importance of MRI in delineating structural and non-structural causes of epilepsy, particularly in the pediatric population where etiology and clinical presentation are highly variable.

Age and Seizure Type Associations

The most common seizure type across all pediatric age groups was generalized tonic-clonic seizures (GTCS), especially among children aged 11–15 years (53.8%). Infantile spasms were exclusively seen in children below one year of age, consistent with known early-onset epileptic encephalopathies. Focal seizures increased in frequency with age, peaking in the 16–18-year group (30%), suggesting a delayed clinical manifestation of cortical developmental anomalies or acquired focal lesions. Sharma et al. similarly reported an age-dependent shift in seizure types, with generalized seizures predominating in early childhood and focal seizures increasing with age [12].

Gender-Based MRI Findings

The distribution of MRI abnormalities was nearly equivalent across sexes, with structural lesions accounting for 31.4% in males and 28% in females. Hypoxic and infective causes also showed minimal gender variation. This suggests that while epilepsy incidence may differ by gender, underlying etiologies on imaging are comparably distributed. Berg et al. noted similar findings, reporting no significant sex-specific differences in structural abnormalities on MRI among pediatric epilepsy patients [13].

Seizure Duration and Imaging Yield

MRI yield correlated positively with seizure duration, increasing from 60% in seizures under one minute to 81.8% in those over fifteen minutes. This supports the hypothesis that prolonged seizures are more likely to be symptomatic of underlying pathology. Gaillard et al. highlighted this relationship, advocating MRI as essential in the evaluation of prolonged or atypical seizure presentations in children [14].

Clinical Predictors of MRI Concordance

Children with developmental delay (88.2%) and neurocutaneous markers (75%) showed high concordance between clinical suspicion and imaging findings, emphasizing the need for early imaging in such subgroups. In contrast, febrile seizures showed lower concordance (40%), aligning with their predominantly benign and self-limited nature. Shinnar et al. similarly observed that MRI was most valuable in febrile seizures when accompanied by neurodevelopmental abnormalities or focal signs [15].

Seizure Type and MRI Yield

MRI was most sensitive in detecting abnormalities in children with infantile spasms (100%) and focal seizures (86.7%). GTCS and absence seizures had lower yields (53.6% and 50%, respectively), reflecting their frequent idiopathic origin. Wyllie et al. reported comparable findings, where focal seizures were associated with a higher likelihood of resectable lesions [16], and Widjaja et al. also demonstrated that MRI was more informative in patients with lateralizing clinical or EEG features [17].

Etiological Spectrum of MRI Findings

The most common MRI-detected etiology was structural abnormalities (30%), followed by hypoxic (16.7%) and infective (13.3%) causes. A significant number of patients (30%) had normal MRI scans, underlining the limitations of structural imaging in genetic or subtle epilepsies. These proportions are consistent with multicentric data from the Pediatric Epilepsy Research Consortium, which emphasized cortical malformations and hypoxic injuries as primary contributors to epilepsy in children [18].

Anatomical Localization of Lesions

Lesion mapping demonstrated the frontal (29%) and temporal lobes (24%) as the most frequently involved regions. Subcortical and multilobar lesions were also observed, highlighting both focal and diffuse patterns. These findings are in concordance with studies by Kuzniecky et al., who documented a high prevalence of epileptogenic lesions in these regions, particularly in surgical candidates [19].

This prospective study underscores the indispensable role of MRI in the etiological evaluation of pediatric seizures. The diagnostic yield of MRI was highest in focal seizures and infantile spasms, and positively correlated with prolonged seizure duration and the presence of developmental or neurological deficits. Structural pathologies constituted the largest proportion of findings, with the frontal and temporal lobes being the most commonly involved regions. The substantial percentage of normal MRIs further highlights the need for complementary diagnostic modalities such as EEG and metabolic studies. Overall, high-resolution, pediatric-optimized MRI protocols offer crucial insights that influence diagnosis, prognostication, and therapeutic planning in children with epilepsy.

1. Hauser WA, Hesdorffer DC. Epilepsy: Frequency, causes and consequences. New York: Demos; 1990.
2. Shinnar S, Pellock JM. Update on the epidemiology and prognosis of pediatric epilepsy. J Child Neurol. 2002;17(Suppl 1):S4–S17.
3. Widjaja E, Zarei Mahmoodabadi S, Otsubo H, et al. Neuroimaging in epilepsy: Current practice and future directions. Curr Neurol Neurosci Rep. 2013;13(4):348.
4. Barkovich AJ, Raybaud C. Pediatric Neuroimaging. 5th ed. Philadelphia: Lippincott Williams & Wilkins; 2012.
5. Gaillard WD, Chiron C, Cross JH, et al. Guidelines for imaging infants and children with recent-onset epilepsy. Epilepsia. 2009;50(9):2147–53.
6. Thivard L, Adam C, Hasboun D, et al. Interictal diffusion MRI in partial epilepsy with focal cortical dysplasia. Brain. 2006;129(Pt 2):375–85.
7. Wang ZI, Alexopoulos AV, Jones SE, et al. The pathology of magnetic-resonance-imaging-negative epilepsy. Mod Pathol. 2013;26(8):1051–8.
8. Pisani F, Spagnoli C. Neonatal seizures: A review of outcomes and outcome predictors. Neuropediatrics. 2016;47(1):12–9.
9. Tripathi M, Garg A, Gaikwad SB, et al. MRI-negative temporal lobe epilepsy: PET-guided evaluation and surgery. Ann Indian Acad Neurol. 2008;11(3):145–50.
10. Dossani RH, Missios S, Nanda A. Pediatric epilepsy surgery: Controversies and advances. Surg Neurol Int. 2015;6(Suppl 1):S37–43.
11. Ramantani G, Neubauer BA, Schubert-Bast S, et al. Seizure control and developmental trajectories in children with early-onset epilepsy and developmental encephalopathy. Dev Med Child Neurol. 2019;61(6):636–44.
12. Sharma S, et al. Age-related trends in seizure types and epilepsy syndromes among Indian children. Indian Pediatr. 2020;57(3):215–20.
13. Berg AT, et al. Gender differences in incidence and risk factors for epilepsy. Epilepsia. 2018;59(3):555–63.
14. Gaillard WD, et al. Role of MRI in evaluation of pediatric epilepsy. Epilepsia. 2011;52 Suppl 4:30–4.
15. Shinnar S, et al. MRI findings and risk of recurrent seizures after febrile status epilepticus. Neurology. 2012;79(9):871–7.
16. Wyllie E, et al. Surgical treatment of epilepsy in children. Epilepsia. 2007;48 Suppl 9:50–2.
17. Widjaja E, et al. Diagnostic yield of MRI in children with epilepsy and normal EEG. Epilepsy Res. 2013;105(1–2):110–5.
18. Pediatric Epilepsy Research Consortium. Multicenter analysis of MRI findings in pediatric epilepsy. Neurology. 2016;86(1):1–9.
19. Kuzniecky R, et al. Neuroimaging of focal cortical dysplasia in children. Epilepsia. 1999;40 Suppl 3:31–4.