Comparative Diagnostic Performance of Early and Term Magnetic Resonance Imaging in Preterm Infants: A Diagnostic Test Accuracy Systematic Review and Bayesian Meta-Analysis | AMiner
Comparative Diagnostic Performance of Early and Term Magnetic Resonance Imaging in Preterm Infants: A Diagnostic Test Accuracy Systematic Review and Bayesian Meta-Analysis
INTRODUCTION:Early identification of neurodisability and delay after preterm birth is critical for timely intervention. Term-equivalent magnetic resonance imaging (MRI) (36-44 weeks) is widely used for prognostication, and earlier MRI (<36 weeks) may offer additional value. METHODS:To provide a comparative diagnostic test accuracy synthesis of early versus term-equivalent MRI for motor and non-motor neurodevelopmental outcomes, incorporating advanced imaging modalities including volumetric and diffusion measures, PubMed, Embase, Cochrane Central, Scopus, and Web of Science were searched to January 22, 2026. PRISMA-DTA guidelines were followed with PROSPERO registration (CRD42024528207). Studies enrolling preterm infants (<37 weeks) undergoing both early and term-equivalent MRI with neurodevelopment assessment between 12 and 36 months were included. Two reviewers extracted data and assessed bias risk. Meta-analysis used a Bayesian bivariate random-effects model pooling diagnostic metrics for the presence of lesion, intraventricular haemorrhage, white matter injury, and cerebellar haemorrhage. RESULTS:Of 3,249 records screened, 30 studies met inclusion. Term-equivalent MRI lesion presence (diagnostic odds ratio [DOR] = 14.17 vs. 2.67; sensitivity 84% vs. 76%; specificity 73% vs. 45%) and white matter injury (DOR = 5.92 vs. 2.20; sensitivity 65% vs. 52% and specificity 76% vs. 67%) demonstrated superior predictivity than early MRI. White matter injury at term best predicted motor outcomes (DOR = 18.43). Volumetric and diffusion measures improved predictive accuracy, especially for cognitive outcomes. Heterogeneity in MRI protocols, biomarker definitions, and outcomes limited comparability, and the modest study number reduced precision. CONCLUSION:Term-equivalent MRI offers higher predictivity than early MRI. Standardised scoring and multimodal integration are needed to optimise early risk stratification in preterm populations.