Feasibility and Safety of Contrast-Enhanced Ultrasound of the Neonatal Brain: A Prospective Study Using MRI as the Reference Standard.

Abstract

BACKGROUND. MRI is the reference standard for neonatal brain imaging, but it is expensive, time-consuming, potentially limited by availability and accessibility, and contraindicated in some patients. Transfontanelle neonatal head ultrasound is an excellent alternative but may be less sensitive and specific than MRI. Contrast-enhanced ultrasound (CEUS) has the potential to improve the capabilities of ultrasound. OBJECTIVE. The purpose of this study is to prospectively evaluate the feasibility, safety, and diagnostic performance of transfontanelle neonatal brain CEUS, with MRI used as the reference standard. METHODS. Neonates in the institutional neonatal ICU who were undergoing MRI as part of their clinical care were prospectively recruited to undergo portable brain ultrasound and CEUS for research purposes. Brain ultrasound and CEUS were performed portably without moving the patient from the isolette or crib in the neonatal ICU. Adverse events were recorded. Two radiologists independently evaluated ultrasound and CEUS images for abnormalities and then reached consensus regarding discrepancies. A separate radiologist reviewed MRI examinations. Sensitivity, specificity, and interreader agreement were evaluated, with MRI used as the reference. Qualitative post hoc image review was performed. RESULTS. Twenty-six neonates (nine boys and 17 girls; mean [± SD] age, 15.2 ± 14.0 days) were included. No significant alteration in patient vital signs or adverse reaction to the ultrasound contrast agent (UCA) occurred. The mean duration of the examination was significantly shorter for combined ultrasound and CEUS than for MRI (21.1 ± 4.7 vs 74.2 ± 34.8 minutes; p < .001). Interrater agreement for any abnormality was almost perfect for both ultrasound and CEUS (κ = 0.92 and 0.85, respectively). Sensitivity for any abnormality was 86.7% for ultrasound and 93.3% for CEUS; specificity was 100.0% for both. CEUS had sensitivity of 87.5% for acute or subacute ischemia and 100.0% for chronic ischemia; its specificity was 100.0% for acute or subacute ischemia and chronic ischemia. For both ultrasound and CEUS, sensitivity for subdural and intraparenchymal hemorrhage was poor (22.2-50.0%). On CEUS but not on MRI, post hoc review showed a case of postischemic hyperperfusion, which was confirmed by subsequently performed contrast-enhanced CT. CONCLUSION. The use of portable brain CEUS in neonates is feasible, safe, and more rapid than MRI. CLINICAL IMPACT. The potential diagnostic utility of brain neonatal CEUS relative to conventional ultrasound, particularly for ischemia, warrants further investigation.