Comparative diagnostic accuracy of ultrasound and magnetic resonance imaging for spinal dysraphism in infants: insights from a single-center study.

Abstract

Background: Spinal dysraphism refers to congenital abnormalities of the spine and spinal cord resulting from defective development of ectodermal, mesodermal, and neuroectodermal tissues. It is broadly classified into open (aperta) and closed (occulta) forms, presenting with conditions such as meningocele, myelomeningocele, and lipomyelomeningocele. Our study aimed to assess the diagnostic accuracy of ultrasound (US) in detecting spinal dysraphism in infants, with magnetic resonance imaging (MRI) as the reference standard.

Methodology: This comparative analytical study was conducted over 12 months across two tertiary medical institutions. Based on a calculated sample size derived from expected sensitivity (82%) and specificity (95%), 106 infants were included, with 80% test power, 10% margin of error, and 95% confidence level.

Results: Among the 106 infants assessed, US demonstrated a sensitivity of 86.54%, specificity of 81.48%, positive predictive value (PPV) of 81.2%, and negative predictive value (NPV) of 86.96%, yielding an overall diagnostic accuracy of 83.96%. Of the total cases, 58 infants were diagnosed with open spinal dysraphism (including myelomeningocele and myelocele), while 20 infants had closed forms (such as tethered cord, dermal sinus tracts, and others). These findings highlight the reliability of US as an initial diagnostic tool for both types of spinal dysraphism.

Conclusion: Ultrasound is a practical, noninvasive screening tool for identifying spinal dysraphism in infants, with an overall high diagnostic accuracy. However, it has a diagnostic discordance rate, requiring MRI confirmation in patients with unusual presentations or high clinical suspicion. However, MRI is still necessary for a precise diagnosis and discrimination between open and closed types of spinal dysraphism.